American College of Toxicology

45th Annual Meeting

Poster Abstracts

JW Marriott Austin, Austin, Texas

The abstracts within this PDF were reviewed and approved by the ACT Program Committee. Every effort has been made to reproduce the abstracts as submitted. The author(s) of the abstracts are solely responsible for the data and content provided. The American College of Toxicology does not claim to validate, promote, or share in the representation of the content provided by the submitting author.

General Toxicology

P101 - International Travel Grant Recipient

Adenopus Breviflorus Fruit Extract Exhibits Antioxidant and Cytotoxic Properties in a Cellular Model of Lung Cancer

Tolulope Oyedeji ¹ , Aminat Erinfolami ¹ , Saheed Benson ² , and Amos Fatokun ²

¹ University of Lagos, Lagos, Nigeria

² Liverpool John Moore University, Liverpool, United Kingdom

Lung cancer is a leading cause of mortality. Exposure to environmental carcinogens and cigarette smoke may increase the chances of developing lung cancer. Adenopus breviflorus fruit is used traditionally to treat various ailments, including cancer. However, its use in treating lung cancer has not been scientifically established. We therefore investigated the potential of A. breviflorus fruit extract to elicit antioxidant and cytotoxic effects (two anticancer mechanisms) in the transformed MRC5-SV2 cell line – a model of lung cancer. Antioxidant and peroxidation-inhibitory effects were assessed against the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical and iron-induced lipid peroxidation (LPO). The cytotoxic effect of the extract was assessed using the 3-[4, 5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay, while reactive oxygen species (ROS) were quantified using the 2′,7′-dichlorofluorescein diacetate (DCFDA) dye. The extract significantly inhibited DPPH radical and LPO in a concentration-dependent manner, with IC50 values of 351 µg/mL and 425 µg/mL, respectively. Furthermore, it was cytotoxic, with a cell viability IC50 of 80.37 µg/mL (48 h). Interestingly, at 100, 200 and 500 µg/mL, it did not significantly (p > 0.05) induce ROS, but significantly inhibited hydrogen peroxide (H2O2) -induced ROS in a concentration-dependent manner, with a fold change value for fluorescence intensity (relative to the negative control) of 1.31±0.04 for 100 µg/ml, compared to 8.07±3.21 for H2O2 (200µM) alone. Our findings show that the aqueous fruit extract of A. breviflorus has antioxidant, ROS-inhibiting, and cytotoxic properties in a lung cancer model, thus validating its chemotherapeutic use in folkloric medicine and identifying it as a potential source of anti-cancer compounds.

P102 - International Travel Grant Recipient

Chemoprotective Role of Gallic Acid on Steroidogenesis, Folliculogenesis, and Follicular Atresia in the Ovary of Rats Exposed to Cyclophosphamide

Adedayo Adebisi ¹ , and Ebenezer Farombi ²

¹ Lagos State University, Lagos, Nigeria

² University of Ibadan, Ibadan, Oyo, Nigeria

Cyclophosphamide, a chemotherapeutic drug used in cancer treatment, has been associated with reproductive dysfunction and infertility in female cancer patients. Gallic acid (GA) is a natural phenolic compound that mediates anti-inflammation, and anti-cancer activities. This study investigated the protective effects of GA against cyclophosphamide-induced ovarian insufficiency in rats. Female Wistar rats (thirty-two) (210 ± 3 g) were assigned into four groups of eight rats each and treated with GA (20 mg/kg b.w.), cyclophosphamide (80 mg/kg b.w.), and a combination of both, orally for 28 days. After treatment, ovarian tissues were analyzed for immunological markers. Serum level of anti-mullerian hormone was assessed using ELISA kits; protein expression of anti-inflammatory genes (interleukin-10, nuclear factor erythroid-2 related factor 2, heme oxygenase 1), steroidogenic genes (steroidogenic acute regulatory protein, CYP 11A1, CYP 17A1, CYP 19A1), folliculogenesis (follicle stimulating hormone receptor, Inhibin B, and prostaglandin synthase-2), and follicular atresia (SDF-1, CXCR-4) were assessed in the ovary using immunohistochemistry technique. The level of expression was quantified using ImageJ (NIH) and data were analyzed using GraphPad Prism 8.0. GA elevated serum anti-mullerian hormone levels, upregulated the expression of anti-inflammatory and steroidogenic genes and promoted folliculogenesis and increased follicular number in the ovary of animals co-treated with GA and cyclophosphamide. These findings suggest that GA may prevent reproductive dysfunction caused by cyclophosphamide through its anti-inflammatory mechanism, thus presenting an opportunity for further investigation into its potency in the management of cyclophosphamide-induced ovarian failure.

P103 - International Travel Grant Recipient

-141C Insertion/Deletion Polymorphism of Dopamine Receptor D2 Is Associated with Acute Risperidone-Induced Changes in Body Metabolic Indices among Nigerians

Holiness Olasore ¹ , and Abdur-Rasheed Awesu ²

¹ University of Lagos, Surulere, Lagos, Nigeria

² Federal Neuropsychiatric Hospital, Yaba, Lagos, Nigeria

The metabolic side effects of atypical antipsychotics such as risperidone may be due to their transient blockade of the dopamine receptor D2 (DRD2). We therefore examined the relationship between -141C insertion/deletion (Ins/Del) polymorphism of the DRD2 gene and acute risperidone-induced changes in metabolic indices. We recruited 234 newly diagnosed patients with psychotic disorders (129 males and 105 females) from the Federal Neuropsychiatric Hospital, Yaba, Lagos, Nigeria. Body weight, fasting blood glucose (FBG), plasma triglycerides (TG), total cholesterol (TChol), low-density lipoprotein cholesterol (LDLChol), and high-density lipoprotein cholesterol (LDLChol) were all determined at baseline and the end of six weeks of administration of risperidone (2 mg twice daily). Genotyping was done by the polymerase chain reaction-restriction fragment length polymorphism method, while the relationship between allele/genotype and metabolic indices was statistically determined. The allele frequencies (n) of Ins and Del were 0.39 (183) and 0.61 (285) while the genotype frequencies (n) of the Ins/Ins, Ins/Del, and Del/Del were 0.18 (41), 0.43 (101), and 0.39 (92) respectively. The population however was not in Hardy-Weinberg equilibrium (χ2 = 2.055, df = 1, p< 0.01). The mean changes in weight, FBG, TG, TChol, and LDLChol were significantly (p< 0.05) higher in the Del/Del group compared to other genotype groups. However, there was no significant difference in the mean change in HDLChol across all genotype groups. The DRD2 -141C Del allele is associated with higher acute risperidone-induced metabolic indices except for HDLChol among Nigerians. This finding may be useful in designing a patient-tailored antipsychotic dose regimen.

P104 - North American Graduate Fellowship Recipient

Liver-Specific Deletion of RNA-Binding Proteins, ZFP36L1 and ZFP36L2, Attenuates Acetaminophen-Induced Acute Liver Injury

Rahul Kumar ¹ , Sonika Patial ² , and Yogesh Saini ¹

¹ North Carolina State University, Raleigh, North Carolina, United States

² National Institute of Environmental Health Sciences, Durham, United States

Zinc Finger Protein 36 Like 1 (ZFP36L1) and Zinc Finger Protein 36 Like 2 (ZFP36L2) are mRNA destabilizing proteins involved in post-transcriptional control of gene expression. They are enriched in the liver; however, their function in drug-induced liver injury remains poorly understood. Here, we investigated the role of ZFP36L1 and ZFP36L2 in acetaminophen-induced acute liver injury. Liver-specific ZFP36L1 and ZFP36L2 double knockout (L1/L2dKO; AlbCre+/Zfp36l1flox/flox/Zfp36l2flox/flox) and flox-only control (L1/L2FLX; AlbCre-/Zfp36l1flox/flox/Zfp36l2flox/flox) adult male mice were fasted overnight and were intraperitoneally challenged with acetaminophen (APAP) dissolved in normal saline (300mg/Kg body weight). The hepatic levels of glutathione (GSH) and cytochrome P4502E1 (CYP2E1) were measured in unchallenged overnight fasted mice. Liver function enzymes and liver histology were assessed in APAP-challenged and saline-treated mice at 24h post-challenge. Liver GSH levels were significantly higher in unchallenged L1/L2dKO compared to L1/L2FLX mice. Interestingly, both mRNA and protein levels of CYP2E1, an enzyme known to bioactivate APAP, were significantly downregulated in unchallenged L1/L2dKO compared to the L1/L2FLX group with consistently low expression in centrilobular hepatocytes. Consequently, the serum levels of liver function enzymes ALT and AST were significantly lower in APAP-challenged L1/L2dKO mice than in APAP-challenged L1/L2FLX mice. Furthermore, the APAP challenge resulted in significantly less hepatic necrosis in L1/L2dKO compared to the L1/L2FLX group. Overall, the combined deletion of ZFP36L1 and ZFP36L2 attenuates APAP-induced acute liver injury via downregulation of CYP2E1 and enrichment of hepatic antioxidant defense, which suggests a pathogenic role of ZFP36L1 and ZFP36L2 in APAP-induced acute liver injury.

P105 - North American Travel Grant Recipient

Resolution Kinetics of Mixed Allergen-Induced Allergic Airway Inflammation in Mice

Rekha K C ¹ , Yogesh Saini ¹ , and Ishita Choudhary ¹

¹ North Carolina State University, Raleigh, North Carolina, United States

Allergic asthma is characterized by chronic airway inflammation, which when fails to resolve leads to a progressive decline of lung function. It is crucial to understand the kinetics of resolution of airway inflammation in order to delineate the specific cellular and molecular pathways involved in it. Accordingly, we intranasally challenged 8-week-old C57BL/6 mice with mixed allergens (MA) or normal saline (NS), three times a week for four weeks and examined the endpoints at 16h, 3d, 7d, 14d, 21d, and 42d post-last challenge. At 16h time-point, MA-challenged mice exhibited all the hallmark features of allergic asthma, including eosinophilic/lymphocytic cell infiltration, production of Th2 cytokines/chemokines, and mucous cell metaplasia (MCM). At 3d time-point and onwards, the inflammatory mediators including IL-4, IL-13, Eotaxin-1, Eotaxin-2, RANTES, IL-17, MIP-1α, MIP-1β, TARC, and MDC decreased in bronchoalveolar lavage fluid (BALF). Eosinophilic and lymphocytic infiltration significantly reduced at 7d timepoint and onwards in MA-challenged mice. The mRNA expression of the specialized pro-resolving mediators (SPMs) receptor, Formyl Peptide Receptor 2 (FPR2) decreased in asthmatic lungs and later increased gradually at 14d timepoint and onwards suggesting the role of SPMs in mediating resolution of chronic inflammation. MA-induced MCM, which was highest at 16h, ameliorated 21d timepoint onwards. By 42d timepoint, almost all the MA-induced lung pathologies were resolved. Overall, this study demonstrates the resolution kinetics of chronic allergic airway inflammation in MA-induced allergic asthma.

P106 - North American Travel Grant Recipient

Variable DNA Methylation Regions Are Responsible for Conserved Patterns of Sperm Epigenome Response to Chemical Stressors

Olatunbosun Arowolo ¹ , and Alexander Suvorov ¹

¹ University of Massachusetts, Amherst, Massachusetts, United States

Evidence demonstrates that exposure to a broad range of chemicals and other stressors produces similar changes in sperm DNA methylation. These changes suggest a systemic response determined by unique characteristics of DNA regions responsive to reprograming. The stochastic epigenetic variation hypothesis predicts that sperm epigenome contains naturally variable methylation regions (VMRs) associated with developmental genes and that sperm methylation changes in response to stressors occur due to the increased methylation variation of these VMRs. The goal of this study is to test these predictions. We used a range of bioinformatic approaches and reduced representation bisulfite sequencing (RRBS) data obtained from mouse and rat studies testing the effects of chemical exposures (phthalates, 2,2’,4,4’-tetrabromodiphenyl ether, and cadmium) on sperm methylome. First, we identified VMRs in control animals and analyzed biological categories associated with these VMRs. Next, we analyzed overlaps between VMRs and regions undergoing changes in DNA methylation (DMRs) in response to exposures. Lastly, we developed a model that explains how increased variation in VMRs averaged across many cells or organisms results in a shift of mean methylation values. Results showed that VMRs overlap significantly with stress-dependent DMRs and are enriched with similar biological categories, such as embryonic development, neurodevelopment, and metabolism. Increasing methylation variation in VMRs is responsible for the formation of stress-dependent DMRs due to asymmetrical increase in variation of hypo- and hyper-methylated regions. Overall, our data provide for the first time evolutionary and molecular mechanistic explanations of the conserved patterns of DNA methylation change in response to chemical exposures.

P107 - North American Travel Grant Recipient

The Impact of Diet-Induced Obesity on the Hepatic Transcriptomic Response to Dimethylbenz(a)anthracene Exposure in Female Mice

Imaobong Inyang ¹ , M. Estefanía González-Alvarez ¹ , Emma Johanns ¹ , Aileen Keating ¹ , and Kelsey Timme ¹

¹ Iowa State University, Ames, Iowa, United States

Obesity disrupts ovarian chemical biotransformation, potentially exacerbating the impact of toxicant exposures; however, limited knowledge exists on the impact of obesity on hepatic xenobiotic biotransformation. This study investigated the hypothesis that Dimethylbenz(a)anthracene (DMBA) exposure would alter hepatic biotransformation mRNA and that obesity would influence this response. Six-week-old C57BL/6J lean and obese female mice were fed either a chow or a high-fat high-sucrose (HFHS) diet for 13 weeks until the obese group reached ∼30% higher body weight than the lean group. Subsequently, mice were intraperitoneally injected with either corn oil or DMBA (1 mg/kg) for seven days before euthanasia. Body weight was increased (P < 0.05) in the obese group fed the HFHS diet. Hepatic RNA abundance was investigated using qPCR and several metabolism genes were altered differently by DMBA exposure in the livers of lean and obese mice. The abundance levels of Cyp2e1, Adh (P < 0.05), and Cyb5r3 (P < 0.1) were decreased in lean but were increased (P < 0.05) in obese-exposed relative to control mice. DMBA exposure increased (P < 0.1) the abundance of Mgst2, Gad2, and Hsd17b1 (P < 0.05) in lean mice but these mRNAs were decreased (P < 0.05) in obese mice. Transcription factors Hes5 and Irf4 were predicted as regulators of some genes in the livers of lean and obese mice respectively, due to DMBA exposure. In summary, our findings provide insight into the impact of obesity on hepatic metabolism during chemical exposure and the role of physiological status in toxicity assessment.

P108 - North American Travel Grant Recipient

Tristetraprolin (TTP) Protects Against the Ozone-Induced Acute Lung Injury and Inflammation in Mice

Richa Lamichhane ¹ , Yogesh Saini ¹ , Ishita Chaudhary ¹ , Sonika Patial ² , Dhruthi Singamsetty ¹ , and Thao Vo ¹

¹ North Carolina State University, Raleigh, North Carolina, United States

² National Institute of Environmental Health Sciences, Durham, North Carolina, United States

Tristetraprolin (TTP), a translational product of the Zfp36 gene, is an anti-inflammatory protein that mediates mRNA decay, especially of transcripts encoding inflammatory cytokines, including those implicated in acute lung injury (ALI). TTP modulates various pathological outcomes in diverse inflammatory diseases, however, its role in ozone-induced ALI has never been tested. Accordingly, we hypothesized that the loss of TTP would exacerbate ozone-induced ALI and that the systemic overexpression of TTP levels would protect against ALI. TTP-deficient (TTPKO), airway epithelium cell-specific TTP-deficient (TTPEpiKO), myeloid cell-specific TTP-deficient (TTPMyeKO), TTP-overexpressed (TTPΔARE) adult male and female mice and their respective control littermates were exposed to single 3-hour long duration to either ozone or FA. The endpoints were accessed 21-24 hours post-ozone or FA exposure. As compared to the TTP-sufficient mice, the TTP-deficient mice exhibited significant increases in ozone-induced ALI characterized by neutrophilic infiltration, cytokine/chemokine production, and lung pathology. The severity of these outcomes was relatively less pronounced in O3-exposed TTPEpiKO or TTPMyeKO mice compared to O3-exposed TTPKO mice. Notably, O3-exposed TTPEpiKO or TTPMyeKO mice partially contributed to the exaggerated inflammatory response observed in O3-exposed TTPKO mice. Conversely, the TTPΔARE mice manifested protection against O3-induced ALI via downregulation of inflammatory cytokines/chemokines production, reduced neutrophilic infiltration, and mitigated lung pathology. Overall, these findings suggest that enhancing TTP expression could serve as a potential therapeutic strategy for simultaneously targeting multiple inflammatory cytokines in various inflammatory diseases.

P109 - North American Travel Grant Recipient

Platelets Mediate Neutrophil Infiltration and Exacerbate Liver Injury and Endothelial Cell Damage after Severe Acetaminophen Overdose

Olamide Adelusi ¹ , Caylie McKimens ¹ , Anup Ramachandran ¹ , and Hartmut Jaeschke ¹

¹ University of Kansas Medical Center, Kansas City, Kansas, United States

Acetaminophen (APAP) overdose, the leading cause of acute liver failure in the US, is accompanied by hepatocyte necrosis and liver sinusoidal endothelial cell (LSEC) damage, with hepatic neutrophil and platelet deposition. We have also shown that the thrombopoietin (the master platelet regulator) agonist, JNJ-26366821, protects against APAP-induced liver injury. However, the exact role played by platelets in the pathophysiology is not fully understood. To investigate this, we depleted platelets in C57Bl/6J mice by injecting 2mg/kg anti-CD41 antibody or IgG control 12 and two hours before a moderate (300 mg/kg) or severe (600 mg/kg) APAP overdose. Platelet depletion did not affect liver injury as measured by plasma ALT levels and hepatic areas of necrosis 24 hours after a moderate APAP overdose but reduced hepatic neutrophil infiltration. After a severe overdose, platelet depletion caused a significant reduction in hepatic neutrophil infiltration, but contrary to the moderate overdose, this reduction in neutrophil infiltration was accompanied by a corresponding reduction in liver injury in the platelet depleted mice. These results corroborate previous findings which showed that neutrophils exacerbate liver injury only after a severe APAP overdose but not a moderate one. Furthermore, severe APAP overdose induced LSEC injury 24 hours after APAP which was significantly reduced in platelet deficient mice as measured by intrahepatic hemorrhage and increased expression of von Willebrand factor on LSECs. Thus, platelets contribute to endothelial cell injury and mediate neutrophil recruitment which aggravates liver injury after a severe APAP overdose, underscoring the role of platelets in the pathophysiology.

P110 - North American Travel Grant Recipient

The Biological Activities of Azadirachta Indica and Vernonia Amygdalina Plant Extract on Human Epithelial Colorectal Adenocarcinoma Cells

Amirah Yusuf ¹ , Shodimu-Emmanuel Olufemi ¹ , Ademola Oyagbemi ² , and Momoh Yakubu ¹

¹ Texas Southern University, Houston, Texas, United States

² University of Ibadan, Ibadan, Nigeria

The second leading cause of cancer-related death worldwide is colorectal cancer (CRC). Recent studies have revealed that herbal-based medicinal products possess tumor-suppressing properties. Extracts from plants such as Azadirachta indica (AI) and Vernonia amygdalina (VA) have been shown to possess anti-cancer effects. However, there is limited information on the combined therapeutic abilities of both plants in mitigating cancer. Our aim is to investigate the combined effect of Azadirachta indica and Vernonia amygdalina on the HT-29 cancer cell line. Cell cytotoxicity and anti-proliferative properties of AI and VA were evaluated using MTT assays. Qualitative and quantitative phytochemical screenings were conducted on both plant extracts. Cell morphology, protein, DNA, and RNA were analyzed. Additionally, western blot and quantitative real-time PCR (qRT-PCR) were performed to determine the apoptosis-related genes and mRNA expression levels of migration. Results showed that AI and VA reduced the viability of the HT-29 cells when used individually and in combination. Phytochemical screening indicated the presence of tannin, flavonoid, saponin, and terpenoid in both plants. There was an alteration in cell morphology and in the expression of apoptosis-related genes. Furthermore, both AI and VA synergistically inhibited AKT/PKB expression and its downstream genes. Our findings provide an important insight into the use of combined herbal medicinal products as an alternative strategy for the effective treatment and management of colorectal cancer.

P111 - North American Travel Grant Recipient

Seasonal Comparison of Fine Particulate Matter (PM2.5) Concentration, Composition, and Oxidative Potential in Tennessee

Voke Tonia Aminone ¹ , Emma Lee ¹ , Gao Rong Lin ¹ , and Courtney Roper ¹

¹ University of Mississippi, Oxford, Mississippi, United States

Fine particulate matter (PM2.5) is a health-relevant air pollutant that varies in concentration and composition based on locations, sources, and seasons. Black carbon (BC) is a component of PM2.5 that is not regulated but has been associated with adverse cardiovascular effects. Exposure to PM2.5 and its components cause oxidative stress known to worsen chronic illnesses as it induces inflammation. Oxidative potential (OP) is a health-based measure that determines the intrinsic ability of PM2.5 to cause oxidative stress. Our research aimed to identify seasonal differences in PM2.5, BC, and OP at 10 locations in Tennessee. PM2.5 collected on filters was provided by the Tennessee Department of Environment and Conservation (TDEC), and BC was analyzed at 370nm.PM2.5 was extracted from filters, and oxidative potential was measured using the dithiothreitol (DTT) assay. Our preliminary data from Fall 2014 showed varying concentrations across months and locations for PM2.5 (2.0 to 21.5μg/m3) and BC (0.01 to 4.9μg/m3). Oxidative potential during the fall ranged from 0 to 1.29 pmol/min/m3. In the winter of 2015, PM2.5 ranged from 1.5 to 20.7μg/m3 while BC ranged from 0 to 5.4 μg/m3. We observed no significant differences in PM2.5 concentrations for fall and winter but significant differences in BC concentrations between locations were observed, such as between Kingsport and 8 other locations. PM2.5, BC, and OP analysis for additional seasons is underway. Our research intends to highlight the importance of seasonal variations, chemical composition, location, and sources on the intrinsic oxidative potential of these air pollutants.

P112

Withdrawn

P113

Withdrawn

P114

Withdrawn

P115

Prevalence of Common Infectious Agents in Nonhuman Primates

Rajeev Dhawan ¹ , Theresa Albers ¹ , Connor Balzer ¹ , William Clifford ¹ , Ken Henderson ¹ , and William Shek ¹

¹ Charles River Laboratories, Wilmington, Massachusetts, United States

Non-human primates (NHPs) are important nonrodent species for safety testing of new drugs and understanding infectious diseases in humans. Routine health monitoring of SPF NHP colonies is performed by various methodologies including serology and PCR testing. Therefore, knowledge about prevalence of common pathogens in colonies is helpful in ascertaining reliable and reproducible research results. In this study, screening data of common pathogens generated in our laboratory starting 2006 was summarized with more than 1.2 million NHP samples from around the world. Routine serological monitoring of common viruses included simian T-cell leukemia virus (STLV), simian retrovirus (SRV), simian immunodeficiency virus (SIV), Herpesvirus simiae (B-virus), filovirus (FV), and measles virus (MV). As expected MV had highest antibody prevalence with 67% seropositivity which is consistent with routine vaccination of NHP colonies. B-virus was prevalent with seropositivity of 1.02%, followed by STLV and SRV at 0.25% and 0.11%, respectively. It is noteworthy that since 2006, the seroprevalence of these viruses has trended downward. NHPs were commonly screened for Campylobacter spp., Shigella spp., Salmonella spp., and Yersinia pseudotuberculosis, with prevalence levels by cultural isolation of 7.56%, 2.03%, 0%, and 1.02%, respectively. Prevalence data for these viruses and bacteria by PCR was also summarized. Finally, recent parasite prevalence for both Trypanosoma cruzi and Plasmodium spp. was just over 5%. In summary, surveillance of imported NHPs in quarantine as well as routine monitoring of breeding and research colonies should include these prevalent agents to avoid the confounding effects of adventitious infections and loss of valuable research resources.

P116

Transcriptomic Profiling to Understand Retinal Function Deficits in Male vs. Female Wistar Rats

Chang-Ning Liu ¹ , Seo-Kyoung Hwang ¹ , Darien Capunitan ¹ , Sophia Lee ¹ , Tom Lanz ¹ , Elias Oziolor ¹ , and Nagappan Mathialagan ¹

¹ Pfizer, Groton, Connecticut, United States

Our previous work demonstrated that 13%–19% incidence of retinal functional deficits in naïve adult male Wistar Han (WH) rats that does not present in females. In the current study, among the 11 males evaluated with electroretinography (ERG), 2 (18%) had no responses, and 3 others had smaller responses, but fundus imaging failed to detect any abnormalities in these animals’ retinas. Six additional male and 11 female WH rats in the study had normal ERG responses. Following retinal functional, vascular and morphology assessment by ERG and fundus imaging, RNA was extracted from eyecups and whole transcriptome sequencing (RNA-Seq) was performed. A total of 52 (p < 0.05, adjusted) differentially expressed genes (DEGs) in male retinas and 36 (p < 0.05) DEGs in female retinas were identified. Highly expressed genes in male retina (Eif2s3y, Uty, Kdm5d, Ddx, Uba1y) and female (Pbdc1, Kdm6a, Kdm5c, Eif2s3, Fam104b) were localized in sex chromosomes. Comparison of individual transcriptome profiles of male retina showed a higher expression of 8 genes (Iqcc, Trpm2, Ppm1n, Fam53b, Pde2a, Pipp2, Plkl, Hacd3) in normal ERG samples, but not in non- or low-ERG response males. Pathway analysis of the data revealed an increased gluconeogenesis, glycolysis and glucose metabolism in samples from non-ERG response male WH rats. Our transcriptomic data provide new insight into the possible genetic bases for sex-related retinal functional deficit in commonly used toxicology species Wistar Han rats.

P117

Hyperglycemia-Induced Loss of Blood Brain Barrier Integrity: An In Vitro Model Representing Secondary Neuron Damage

Ayse Basak Engin ¹ , Atilla Engin ² , and Evren Doruk Engin ³

¹ Gazi University, Faculty of Pharmacy, Ankara, Ankara, Turkey

² Gazi University, Faculty of Medicine, Ankara, Turkey

³ Ankara University, Biotechnology Institute, Ankara, Turkey

Due to traumatic brain injury approximately ten million deaths are recorded annually. Half of the surviving patients remain with neurological disabilities because of the secondary neuronal injury. Patients who died have a significantly higher blood glucose levels than survivors within the first three days of hospital admission. It is claimed that the blood-brain barrier (BBB) permeability increases in non-injured region of brain tissue and results in unpredicted neuron damage. Aim of this study is to investigate whether the post-traumatic hyperglycemia and associated activation of complement cascade in the non-injured neurons may cause to the development of neuronal injury. BBB model is created with the triple cell culture which consists of endothelial cell, basal membrane-pericytes and neurons. Integrity of BBB was confirmed with the occludin, claudin-5, ZO-1 expressions, and “Trans-Endothelial Electrical Resistance” values. A successful insulin resistance model was established in both compartments of BBB by the high glucose and insulin transportation from luminal to the abluminal compartment. At the end of the first 48 hours, supernatants were collected from the abluminal compartment and neuronal nitrite nitrate, glutamate, methyl glyoxal (MGO), C1q, C3a and CD59 levels, oxidative stress coefficient and mitochondrial metabolic activity (MTT) were measured. While glutamate toxicity and oxidative stress increasing 3-fold, GLUT 3, IRS-1, C1q, CD59, MGO and MTT decreased significantly. Hyperglycemia-induced loss of BBB integrity as well as neuronal energy deficiency and complement imbalance may be responsible from the secondary neuronal damage in the undamaged region of the brain. *Supported by TUBITAK, 214S112 and 112S437.

P118

Toxicity Beyond the Test Article—Database of Excipient Tolerability in Species Used for Nonclinical Studies

Gregory Sondag ¹ , and Donald Stump ¹

¹ Charles River, Ashland, Ohio, United States

Careful consideration is given to how excipients can affect test article bioavailability and pharmacodynamic effects in a nonclinical study. It is also essential to assess the impact of excipients on animal health, independent of test article action. Gad et al., 2016 documented vehicles that have been commonly used in nonclinical studies and how well the vehicles were tolerated. However, certain excipients have little information on the range of tolerable dose volumes or use in different animal models. This highlights the necessity for a vehicle database that can be referenced for the most contemporaneous safety information. A site-specific database of vehicles, including how well the vehicles were tolerated, was developed for Charles River Laboratories Ashland in commonly used animal models (Beagle dogs, Cynomolgus monkeys, rabbits). This database includes each vehicle, dose volume (mL/kg), route of administration, duration of dosing, and outcomes. Because many vehicles contain multiple components, the database can be searched for each individual component and multiple -component vehicles. This information was compiled on the internal IACUC website. The utility of this list allows immediate access to all scientists to investigate whether a selected vehicle has a toxicity concern. This compilation is a valuable resource to identify potential concerns of less commonly used excipients prior to study initiation. The database has had a positive impact on animal welfare by minimizing adverse events due to vehicle toxicity.

P119

Retrospective Study of Survivability and Histopathology Findings in Cannulated Rodent

Zhimei Wang ¹ , Sue McPherson ¹ , and Lei Guo ¹

¹ WuXi AppTec (Suzhou) Co., Ltd., Suzhou, Jiangsu, China (People’s Republic)

A retrospective study was performed to determine the survivability and operation related pathology findings in control intravenous (IV) (femoral vein) cannulated rodents. IV cannulation is a technique in which a cannula is placed inside a vein to provide venous access, which is typically used for long duration administration and/or alleviating the irritation problem of dose formulations. Data were collected from 281 females and 306 females, aged 6 to 10 weeks, from 32 studies (mostly 5 to 28 days duration, ranged from 0.16 to 2 hours infusion) evaluated at the facility between 2019 and 2023. The survival rate was 97% (male) and 98% (female). For early terminated animals, the cause of death was attributed to the primary procedure-related macroscopic/microscopic findings of the single nodule (correlating microscopically with suppurative inflammation) in the injection site or major secondary embolic pneumonia in the lung/bridging hepatocellular necrosis in the liver/pyelonephritis or nephropathy in the kidney/suppurative inflammation in the urinary bladder/endocarditis and thrombosis in the heart. For schedule terminated animals, the most notable microscopic changes included intravascular thrombosis, intimal fibrosis, mixed cell infiltration of inflammatory cells and/or mineralization. The secondary procedure-related microscopic findings resulting from directly expanding or indirectly spreading through the blood/lymph circulation from the primary thrombosis/suppurative inflammation consist direct expanding caused suppurative inflammation; indirectly spreading caused secondary findings in multiple organs; compensatory responses; stress-related non-specific responses. Such data base will help both the toxicologist and pathologist differentiate the study results so that the test article effects can be distinguished from procedural related findings.

P120

An Examination of the Effects of Body Temperature on QT Interval in Non-Naïve Telemetered Göttingen Minipigs®

Miri Pannu ¹ , Ted Baird ² , Emily Griffith ² , Rebecca Smith ² , and Joel Baublits ³

¹ Altasciences Preclinical Columbia, Auxvasse, Missouri, United States

² Altasciences, Auxvasse, Missouri, United States

³ Cardiovascular Analytics, Newbury Park, California, United States

Changes in body temperature (BT) have been shown to affect the QT/QTcH interval. Given the known inverse relationship in other species (e.g. dog and human), the study objective was to propose a correction formula for miniature swine. Eight male telemetered Göttingen Minipigs® were used to determine the relationship between QT/QTcH interval and BT. The study was conducted in two phases. In Phase 1, the animals received a singular intramuscular (IM) injection of Zoletil (tiletamine and zolazepam for injection) (9 mg/kg), due to known effects on heart rate and BT. In Phase 2, the same animals were anesthetized, and manual cooling and heating procedures were applied to further investigate the relationship between BT and QT/QTcH interval. After administration of Zoletil, significant (p< 0.05) increases in the QT/QTcH interval and significant (p< 0.05) decreases in BT were observed. Peak difference for QT/QTcH occurred at 2 hours postdose with values returning to normal by approximately 4 hours postdose. Peak difference for BT occurred at 2 hours postdose with values returning to normal by approximately 4 hours postdose. Given the known inverse relationship between BT and QT/QTcH, these changes were to be expected. The relationship between BT and the QTcH interval was established in Göttingen Minipigs® in both conscious and anesthetized states. This study provided evidence of an inverse, apparent linear relationship between the two parameters. A QTcH formula with a correction factor for BT changes (QTcHcT) was derived for both conscious and anesthetized states.

P121

T2-MRI Mapping and Fluidic Neurofilament Light (NfL) as a Minimally Invasive Correlate of Central Nervous System (CNS) Toxicity in a Cuprizone Model: A Biomarker Study

Syed Imam ¹ , Lindsay Gill ² , Kimberly Kelly ² , Natalya Sadovova ¹ , Jennifer Pierson ³ , Ruth Roberts ⁴ , Michael Aschner ⁵ , William Slikker ⁶ , Tucker Patterson ¹ , and Serguei Liachenko ¹

¹ US FDA/NCTR, Jefferson, Arkansas, United States

² CDC/NIOSH, Morgantown, West Virginia, United States

³ HESI, Washington, District of Columbia, United States

⁴ ApconiX, Macclesfield, England, United Kingdom

⁵ Albert Einstein College of Medicine, Bronx, New York, United States

⁶ Retired (US FDA/NCTR), Jefferson, Arkansas, United States

Neurotoxicity has been linked to exposure to a number of drugs and chemicals. However, efficient, predictive, and minimally-invasive methods to detect neuroanatomical effects are not routinely used in non-clinical assessments. We have shown significant T2-MRI changes in a rat model of trimethyltin neurotoxicity that correlates with CNS neurotoxicity and pathology as visualized by Flouro Jade C staining. Our primary objective was to identify possible T2-MRI relaxation changes that predict myelin specific neurotoxicity resulting from exposure to the known neurotoxic agent, cuprizone, and correlate T2-MRI relaxation with fluidic NfL and neuropathological endpoints. Adult-male rats (3-months old) were exposed to a daily dose of cuprizone (600 mg/kg p.o., daily) or corn oil for 4 weeks. Prior to treatment, animals underwent MRI scans to establish baseline. Additional MRI scans were performed weekly after beginning exposure to cuprizone. At the end of 4 weeks, a final MRI was completed, and animals were sacrificed for collection of fluids and tissue samples. Significant increase in T2-MRI relaxation was observed in the deep cerebellar nuclei (DCN) in cuprizone-treated rats compared to controls. CNS pathology of post-mortem brain also shows damage to the DCN, a possible correlation with MRI changes. Additionally, both plasma and serum NfL were significantly increased after exposure to cuprizone. Our data demonstrate that MRI-based endpoints and fluidic NfL levels may represent robust minimally invasive neurotoxicity biomarkers in a myelin-specific neurotoxicity model. Disclaimer. This presentation reflects the views of the authors and does not necessarily reflect those of the Food and Drug Administration.

P122

Intravenous Sampling and Administration Using Instech Vascular Access Buttons™ in Rodents: Pathology Findings Related to Indwelling Vascular Catheter

John MacMaster ¹ , Julie Forget ¹ , Norbert Makori ¹ , and Shunji Nakatsuji ¹

¹ Altasciences Preclinical Seattle, Everett, Washington, United States

Repeated reliable access to the vascular space in rodents can be challenging in preclinical modeling. While vascular access catheterization has been extensively used in the past, Vascular Access Buttons (VAB™) are quickly becoming a more common method of external access for rodents. This approach minimizes potential trauma to the tail vein, increases the success rate of repeated intravenous administrations, and allows for continuous and intermittent infusion designs. While Vascular Access Buttons (VAB™) have become a more conventional method for vascular access catheterization, there have been some histopathological observations related to the indwelling vascular catheter and at the exteriorization site of the VAB™ that might be a confounding factor in the histopathological and hematological evaluations in the study. Major pathological changes related to VAB™ included mild to moderate ulceration and/or necrosis of the skin and mild chronic granulomatous inflammation (including thread-like foreign bodies) at the VAB™ surgical site (skin); and moderate thrombus formation, minimal perivascular hemorrhage, and fibroplasia at the catheter-implanted administration site (vena cava). Additionally, inflammatory changes along with thrombus and alveolar epithelial hyperplasia in the lungs and in the perivascular/sinusoids of the liver were noted in likely associated with administration site changes (thromboembolism). Even though there were histopathological challenges associated with the use of VAB™s, findings were adequately documented and assessed to be associated with the administration method. In addition, the overall number of affected animals was limited and did not adversely impact study outcomes, therefore demonstrating that this method of administration is suitable for preclinical safety assessments.

P123

Lipid Nanoparticles: A Comprehensive Assessment of Liver Enzymes, Clinical Presentation, and Immunopathology Markers in Nonhuman Primates

Gabriela Campoy ¹ , Narine Lalayeva ² , Julie Forget ² , and Norbert Makori ²

¹ Altasciences, Snohomish, Washington, United States

² Altasciences Preclinical Seattle, Everett, Washington, United States

Lipid nanoparticles (LNPs) represent a significant advancement in drug delivery technology as carriers for non-viral targeted therapies. Their ability to encapsulate and transport molecules across cell membranes, protecting them from degradation, has opened new avenues in gene therapies and editing. Evaluating safety and efficacy of these new products is paramount, and while the Food and Drug Administration (FDA) provides regulatory guidance, a paucity of published data makes it challenging to design pre-clinical studies. Despite reports of transient increases in liver enzymes, including activation of complement, there is a lack of published data detailing the range and peak of these increases, their correlation with clinical presentation, onset of adverse events, and associated histopathologic and immune activation findings. Identifying potential toxicities is crucial for effective monitoring and risk mitigation; however, there is no regulatory guidance available to recommend specific monitoring schedules for these parameters. Based on data review from studies conducted from 2021 to 2024 using LNPs in cynomolgus macaques, significant increases in alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in the hundreds to thousands of units per liter were observed. Notable changes in neutrophil and lymphocyte distribution, as well as inflammatory markers, were most noticeable 24 to 72 hours post dose and trended towards baseline within two weeks. Histopathologic findings include hepatocellular hypertrophy, hepatocellular microvesicular vacuolation, and mononuclear cell infiltration. This review aims to bridge these gaps in knowledge and contribute to the safe and effective use of LNPs in gene editing and therapy in a preclinical model closely resembling humans.

P124

DICTrank: The Largest Reference Dataset for Drug-Induced Cardiotoxicity (DICT) Risk Annotation of 1318 Human Drugs Based on FDA Labeling

Dongying Li ¹ , Yanyan Qu ¹ , Ting Li ¹ , Zhichao Liu ² , and Weida Tong ¹

¹ US FDA, National Center for Toxicological Research, Jefferson, Arkansas, United States

² Boehringer Ingelheim Pharmaceuticals, Ridgefield, Connecticut, United States

Drug-induced cardiotoxicity (DICT) is a major factor contributing to failed drug development and drug withdrawal. Presently, cardiotoxicity annotation for drugs primarily focuses on specific clinical manifestations or mechanisms, such as QT prolongation and effects on cardiomyocytes. Considering the wide range of adverse effects of drugs to the heart, comprehensive evaluation of DICT risk for various drugs and diverse therapeutic categories is urgently needed. To that end, this study set out to develop an extensive reference list to classify drugs by their DICT risk to facilitate the discovery of effective diagnostic biomarkers and the development of New Approach Methods (NAMs) for early DICT identification. We studied over 1300 human prescription drugs listed in FDA drug labeling documents extracted from FDALabel, with a specific focus on those with a single active ingredient via parenteral or oral administration. Based on adverse cardiac events described in labeling documents and cardiotoxicity severity, we ranked the drugs to four DICT-concern levels in the DICTrank dataset, which included 341 Most-, 528 Less-, and 343 No-DICT-Concern drugs, as well as 106 drugs of Ambiguous-DICT-Concern due to insufficient evidence. We incorporated a variety of drug classes in DICTrank and identified those enriched with drugs of Most-DICT-Concern, including beta blockers, sex hormones and genital system modulators, and anti-inflammatory products. In conclusion, DICTrank is to date the most extensive drug list of comprehensive annotation of DICT, and it could enhance the development and utilization of NAMs and AI models in drug safety evaluation for drug development and other applications.

P125

Withdrawn

P126

Utility of Immunodeficient SRG Rat in Identifying Immunogenicity-Related Hepatoxicity with a LNP-Encapsulated mRNA Encoding a Secreted Monovalent Antibody

Elena Shevtsova ¹ , Laura Cregar ¹ , Sydney Mukaratirwa ¹ , Shan Naidu ¹ , Harkewal Singh ¹ , Jessica Ortiz ¹ , Ryan Liese ¹ , Susan Goody ¹ , and Binu Philip ¹

¹ Moderna, Cambridge, Massachusetts, United States

mRNA-XXX, a lipid nanoparticle (LNP)-encapsulated messenger mRNA-based therapeutic encoding a secreted monovalent antibody fused to human serum albumin, was evaluated for repeat-dose toxicity in Sprague Dawley (SD) rats and cynomolgus monkeys (NHPs). mRNA-XXX was administered via intravenous infusion at 0.5, 1.5, or 5 mg/kg/dose once weekly for four weeks, followed by a four-week recovery period. All doses were clinically well tolerated in NHPs. However, in SD rats, doses of ≥ 1.5 mg/kg resulted in early mortality of several females after > 3 doses, which was attributed to hepatocellular degeneration/necrosis. In rats, the secreted antibody encoded by mRNA-XXX is not pharmacologically active and the same LNP formulation was well tolerated at the same doses and frequency. It was hypothesized that the hepatotoxicity was related to immunogenicity to the expressed protein which was foreign to rats. All SD rats dosed with mRNA-XXX tested positive for anti-protein antibodies on Day 29. A follow-up mechanistic study was conducted with immunodeficient SD Rag2/IL2rg (SRG) female rats at the same dose level (5 mg/kg) and frequency (once weekly for four weeks) as the SD study. Unlike immunocompetent SD rats, which experienced early mortality, all SRG rats survived to scheduled necropsy without any findings. No SRG rats tested positive for anti-protein antibodies, indicating that the hepatotoxicity in SD rats was secondary to immunogenicity, and irrelevant to humans. These data demonstrate the value of immunodeficient animal models in the early stages of development for identifying species-specific immunogenicity-related toxicities and choosing the appropriate animal species for toxicology studies.

P127

Mono-(2-ethylhexyl) Phthalate (MEHP) Exposure Suppresses Slit2 Expression in Peritubular Myoid Cells Primary Isolates from the Rat Testis

Narayan Acharya ¹ , and John Richburg ¹

¹ The University of Texas at Austin, Austin, Texas, United States

In mammals, the Slit2 neuropeptide is reported to serve as a cellular migration guidance cue, particularly for neuronal and immune cells. Previously, we observed that acute exposure of peri-pubertal Fischer 344 rats to MEHP (700 mg/kg, p.o.) increased testis peritubular macrophages in the area specific to differentiating and undifferentiated spermatogonia. The cellular mechanisms for recruiting peritubular macrophages after toxicant-induced injury are unknown. We aimed to identify whether the Slit2 neuropeptide ligand is produced by extra-neuronal cells in the testis and to understand its relationship to MEHP-induced testicular injury. Primary peritubular myoid cells (PTMCs) isolated from rat testis were prepared and incubated for 7 days to confluency. At this time, the PTMCs were exposed to 200 µM MEHP, which is known to mimic the in vivo dosing concentration. After 48 hours, cells were collected for RNA analysis. After qRT-PCR, RNA sequencing was executed, and the data were utilized to interrogate the regulation mechanism of Slit2 ligand expression under the MEHP exposure. Enrichment analysis indicates a significant enrichment of genes involved in the regulation of the expression of Slits and Robos in PTMCs upon exposure to MEHP for 48h. The substantial downregulation of Slit2 in the treatment group indicates that MEHP exposure potentially affects its expression. These findings suggest that PTMCs of testis express Slit2 and that the MEHP exposure suppresses its expression. Further experiments will focus on understanding the functional consequence of alterations in Slit2 signaling to the pathogenic sequence of events responsible for MEHP-induced testicular injury.

P128

Withdrawn

P129

A Comparison of Jacketed External Telemetry Procedures on NHP Pretreatment Heart Rate: Can Different Acclimation Procedures Provide Similar Results?

Sara Christiansen ¹ , Fernanda Alvarez ¹ , Katarina Haist ² , Courtney Jenkins ³ , Ghislaine Boudreau ⁴ , Douglas Regalia ¹ , and Stephen Tichenor ¹

¹ Charles River Laboratories, Reno, Nevada, United States

² Charles River Laboratories, Mattawan, Michigan, United States

³ Charles River Laboratories, Ashland, Ohio, United States

⁴ Charles River Laboratories, Senneville, Quebec, Canada

Habituation and acclimation procedures of laboratory animals in safety assessment studies, including cardiovascular endpoints, are known to be important to mitigate the effects of stress on study parameters. Abstaining from either could result in cardiovascular values outside of the expected physiological range. This is especially the case for External Telemetry (ET) recordings in Non-Human Primates (NHPs). While it is widely accepted to acclimate NHPs to study procedures, there is no set way described on how best to acclimate animals for the Jacketed External Telemetry (JET) apparatus. The purpose of this data review was to determine if certain acclimation procedures are better suited at mitigating the effects of stress on JET recordings. Heart rate (HR) data were collected from approximately 150 animals at each of the four different Charles River Laboratories and were compared after differing acclimation practices: 1) two cycles of four, eight and 24-hours/cycle; 2) one cycle of four-, eight- and 24 hours; 3) 72-hour continuous acclimation; 4) 24-hour acclimation. All four acclimation procedures produced mean heart rates within the physiological range for NHPs during day and night. The two cycles of four-, eight-, and 24-hour acclimation, one cycle of four-, eight-, and 24-hour, and single 24-hour acclimation paradigms had a slightly decreased (∼10 to 20 beats/min) heart rates compared to the continuous 72-hour paradigm. In conclusion, these data supports that no single acclimation paradigm is better than another as all heart rates are similar in comparison. The acclimation itself is more important than the method of acclimation.

P130

Immunomodulatory Approaches in Preclinical Gene Therapy Studies

Nirmala Chinnappareddy ¹ , Kelsey Brooks ¹ , Julie Forget ¹ , Narine Lalayeva ¹ , and Norbert Makori ¹

¹ Altasciences, Preclinical Seattle, Everett, Washington, United States

In-vivo gene therapy (GT) utilizing Adeno-Associated Virus (AAV) vectors or Lipid Nanoparticles (LNPs) has made considerable progress over the past several years; nonetheless, activation of both the innate and adaptive immune systems remains a critical obstacle often necessitating premedication with immunomodulating or suppressive agents to mitigate adverse responses. Given the unique challenges of executing successful preclinical GT studies, we reviewed data from 30 studies conducted in the last 3 years to identify the frequency of premeditation use, common single drug and combination pretreatment regimes, and incidence of clinical observations indicative of immune responses in nonhuman primates. Of the studies reviewed, ∼50% utilized a pretreatment regimen prior to dosing with AAV/LNPs, sometimes in combination with other supportive medications. Clinical signs indicative of an innate immune response were not observed with premedication regimes of (A) dexamethasone at 2 mg/kg via intravenous (IV) bolus at least 1 hour prior to test article administration in 6 studies, (B) prednisolone at 1 or 3mg/kg over the duration of the dosing period in 3 studies, (C) Diphenhydramine at 2 mg/kg administered intramuscularly 15 to 30 minutes prior to dosing in 1 study or (D) in studies (1 each) utilizing multiple premedication drugs including 3 mg/kg prednisolone and 5 mg/kg diphenhydramine, 8 mg/kg Tocilizumab with and without 0.05 mg/kg Tacrolimus, or 750 mg/m2 Rituximab with 2 mg/m2/day Sirolimus and 4 mg/kg Diphenhydramine. In conclusion, this data review can serve as a guide to inform future GT study designs and identify suitable premedication regimes for adequate clinical translation.

P131

Nonclinical Vehicle Formulations: HP-β-CD (Hydroxypropyl-Beta-Cyclodextrin) and the Risk of Hemolysis

Claire Trincot ¹ , Mark Andracki ¹ , Alexandra Austenova ² , JDS Holt ² , Stewart Jacobson ¹ , Holden Janssens ² , Supida Monaikul ¹ , Brian Morrison ¹ , and Rebecca Prince ¹

¹ Frontage Laboratories, Chicago, Illinois, United States

² Atai Therapeutics Inc., New York, New York, United States

Beta-cyclodextrin is a commonly used excipient in nonclinical vehicle formulations. In a recent study, vehicle formulated with 35% HP-β-CD resulted in hemolysis after a 5-10 minute intravenous (IV) infusion in Sprague Dawley rats, as evidenced by red urine, increased serum bilirubin levels, and increased clinical pathology hemolysis scores. During subcutaneous (SC) administration using 11.6% HP-β-CD, hemolysis did not occur, though histopathological changes at the injection site indicated local toxicity (edema, mixed inflammation, hemorrhage, and myofiber degeneration) 24 hours post dose. During IV bolus administration at both 35% and 11.6% HP-β-CD, hemolysis was only present in test article-treated animals with 35% HP-β-CD (not controls), indicating that certain test articles may exacerbate hemolysis caused by HP-β-CD. In vitro rat blood analysis showed visible dose-dependent evidence of hemolysis caused by HP-β-CD, a 35% HP-β-CD solution resulted in visible hemolysis, while a diluted HP-β-CD solution (11.6% HP-β-CD) did not result in hemolysis. In conclusion, HP-β-CD levels should be carefully considered in nonclinical vehicle formulations in Sprague Dawley rats, as 35% HP-β-CD can cause hemolysis after intravenous administration, potentially exacerbated by the presence of test article.

P132

Occupational Exposure Banding (OEB) for Mutagenic Compounds in Absence of Robust Dataset

Sanket Gadhia ¹ , Martin Kohan ¹ , and Janet Gould ¹

¹ SafeBridge Consultants, Inc., New York City, New York, United States

Potent compound manufacturing exposes healthy individuals to several compounds (e.g., active pharmaceutical ingredients, intermediates), including some that are genotoxic. When adequate preclinical and clinical data are available, robust occupational exposure limits (OELs) are developed to protect the workers from adverse effects of these compounds as well as their overexposure. But in the absence of a robust dataset, it is recommended that health-based exposure range (e.g., OEB) or default OEL are used to protect the workers. For potentially mutagenic compounds lacking carcinogenicity data, an exposure-based approach of threshold of toxicological concern (TTC) is often employed to limit the exposure to 1.5 µg/day with a risk tolerance of 1:100,000 as per ICH M7, which is then adjusted for breathing volume and exposure duration. For worker safety, the NIOSH Cancer Policy recommends calculating the risk estimates from 1:100 to 1:1,000,000 over a 45-year working period to determine a risk management limit. Assuming that a worker is exposed to a mutagen every day of their working life is overtly conservative. A more pragmatic approach by extending the principles of NIOSH Cancer Policy to a more realistic exposure scenario, wherein we hypothesize that cumulative exposure to a mutagen for any worker would not exceed half of their work life (i.e., 20 years). This combined with intact DNA repair mechanisms and a limited exposure duration (5 days/week, 8-hours/day) allows for a higher risk tolerance. Based on this paradigm, a default OEB and OEL for mutagenic compounds of 1-10 µg/m3 and 1 µg/m3, respectively, is health protective.

Regulatory Toxicology

P201 - International Travel Grant Recipient

Retrospective Evaluation of the Use of Nonhuman Primates for Fertility Assessment of Marketed Medicinal Products

Puck Roos ¹ , Diann Blanset ² , Peter van Meer ³ , Shermaine Mitchell-Ryan ⁴ , Peter Theunissen ⁵ , Katie Turner ⁶ , Ronald Wange ⁷ , and Christopher Bowman ⁸

¹ Utrecht University (NL), Medicines Evaluation Board (NL), Utrecht, Utrecht, Netherlands

² Boehringer Ingelheim, Ridgefield, Connecticut, United States

³ Medicines Evaluation Board, Utrecht, Utrecht, Netherlands

⁴ Health and Environmental Sciences Institute, Washington, District of Columbia, United States

⁵ Medicines Evaluation Board, Utrecht, Netherlands

⁶ Johnson & Johnson Innovative Medicine, Spring House, Pennsylvania, United States

⁷ US Food and Drug Administration, Center for Drug Evaluation and Research, Silver Spring, Maryland, United States

⁸ Pfizer Worldwide Research, Groton, Connecticut, United States

Reproductive toxicity assessment is conducted to detect potential adverse effects of medicinal products on fertility. When non-human primates (NHP) are the only pharmacologically relevant test species, effects on fertility parameters can be evaluated in NHP. We have retrospectively evaluated fertility assessment in NHP. Our database consists of compounds submitted for marketing approval to the European Medicines Agency or approved by the US Food and Drug Administration between 2011-2022. Compounds with ≥1 GLP-compliant repeat-dose toxicity or fertility study in NHP were included. Information on compound characteristics, study design, effects on fertility parameters and labeling was obtained from publicly available documents. In total, 263 compounds with 668 NHP studies were included. NHP age was reported for 49.9% of studies. NHP sexual maturity was reported for 18.0% of studies. Adverse effects on fertility parameters in NHP were reflected in the label of 22 compounds (8.4%). For none of these 22 compounds were sexually mature NHP vital to inform infertility risk assessment. Effects were predictable based on mechanism of action, or adverse effects were also observed in rodent studies. In conclusion, the value of fertility assessment in NHP for approved medicinal products appears to be low. For small molecules, rodent studies were sufficient for risk assessment. For biologicals, a weight-of-evidence approach could be used. This would reduce the need for sexually mature NHP in drug development. Additionally, when there are scientifically valid reasons to use NHP for fertility assessment, sexual maturity status should be established prior to study initiation and reported in public documents.

P202

Adrenal and Splenic Findings in a Toxicity Study of the Antibody GEN3009 May Be Related to ADA-Enhanced Hypotension

Joshua Gamse ¹ , Jens Mortensen ² , and Theodora Salcedo ¹

¹ Genmab, Plainsboro, New Jersey, United States

² Genmab, Valby, Denmark

GEN3009 is a bispecific antibody that targets CD37, intended for treating B cell malignancies. Hexamers of GEN3009 induce complement-dependent cytotoxicity (CDC) of B cells. In a GLP toxicity study, infusion of GEN3009 in monkeys for 5 weekly doses at 3, 10, or 30 mg/kg transiently decreased B cells and consumed complement consistent with its mode of action. Exposure in monkeys was lower after the second weekly infusion consistent with anti-drug-antibody (ADA)-mediated clearance and presence of ADA in all treated monkeys. ADA-reduced exposure was reflected by loss of pharmacology (B cell CDC) after the 3rd and subsequent doses. However, complement consumption gradually increased after the 3rd and subsequent doses, likely due to formation of GEN3009/ADA immune complexes. Starting with the 1st dose, clinical observations at all dose levels included transient facial swelling, emesis, pupil dilation, and hypoactivity. After the 4th dose, these findings became more severe, and additionally severe transient hypotension was seen shortly after dosing in all 30 mg/kg monkeys. At necropsy, non-adverse splenic congestion with dose-dependent severity was seen at all doses, indicating hypotension reached down to 3 mg/kg. In addition, adverse dose-dependent medullary congestion and cortical hemorrhage occurred in the adrenal gland at ≥10 mg/kg. Together, the evidence suggests that high complement activation by GEN3009/ADA immune complexes resulted in hypotensive stress, leading to splenic congestion and adrenal congestion/hemorrhage. This toxicity is specific to the preclinical model as ADA formation in monkeys is not translatable to humans.

P203

A Data-Driven Approach to Determine Benchmark Response (BMR) for Continuous Endpoints

Kan Shao ¹ , Andrew Weissmann ² , Mario Di Salvo ² , and Owen Gordon ²

¹ Indiana University, Bloomington, Indiana, United States

² DREAM Tech, LLC, Bloomington, Indiana, United States

Benchmark dose (BMD) modeling has become the predominant method for deriving a point of departure (POD) and further a reference dose (RfD) in risk assessment. One important aspect of BMD modeling is choosing an appropriate benchmark response (BMR) to define adversity. Unlike dichotomous data, in which the BMR can be expressed in terms of risk, for continuous data, the typically used BMR definitions based on central tendency (e.g., relative change or standard deviation shift) are straightforward to use but much less interpretable as adversity or risk. In this study, we aim to develop BMR value guidance for various continuous endpoints that can help users define the adversity of BMR as risk based on relative change in central tendency. We utilized a large dose-response database with more than 75,000 historical toxicological datasets separated based on species, strain, and sex of the animal studied to determine endpoint-specific adversity and understand how each endpoint responds to exposure at different dose levels. Utilizing the historical data, bootstrap and Monte Carlo simulation methods were applied to (1) properly define the adversity based on control group data and (2) plausibly estimate the relative change in central tendency corresponding to various risk levels for different endpoints. The methodology was first applied to nine organ weights to develop risk-based BMR values for these endpoints to make BMD modeling using continuous data more convenient and consistent. The results suggest that the BMR value is quite endpoint-dependent and should be carefully chosen.

P204

Developmental Neurotoxicity of Opioid Medication Assisted Treatments (MAT) in Long-Evans Rats

David Ndambuki ¹ , and Andrew Shen ¹

¹ US FDA, NCTR, Jefferson, Arkansas, United States

The incidence of neonatal opioid withdrawal syndrome (NOWS), which includes increased irritability, high-pitched crying, decreased sleep, and respiratory distress, has increased because of the ongoing opioid epidemic. Clinical management of opioid addiction during pregnancy relies on off-label prescription of FDA approved medication-assisted treatments (MATs) buprenorphine and methadone, both of which can contribute to NOWS. Clinical studies that compare buprenorphine and methadone neonatal outcomes are confounded by polysubstance use in patients hindering clear conclusions over which drug is more neurotoxic. Hence, preclinical models are necessary to investigate and compare short- and long-term neurotoxicity of perinatal MATs. We hypothesize that buprenorphine and methadone cause relatively different short-term and long-term neurochemical and neurobehavioral toxicity. Female Long-Evans rats were treated with either vehicle, morphine, buprenorphine, or methadone prior and during pregnancy until parturition. On postnatal day one, offspring were cross fostered to naïve unexposed dams and then received opioid treatment corresponding to maternal treatment. Tissue collection at developmental timepoints for histopathological, neurochemical, and neurobehavioral analyses were performed. Preliminary results suggest that: (1). MAT affects neonatal growth and development that results in low birth weight and delayed developmental milestones, and (2). histopathology of brains collected from postnatal seven to 21 observed region-specific abnormalities such as decreased myelination. Our preliminary findings suggest that perinatal exposure to opioid MAT causes neurotoxicity that is likely drug, developmental stage, and brain region specific. These data are guiding efforts to recapitulate these findings with larger sample sizes and direct the behavioral assays to reveal lesion specific changes in behavioral outcomes.

P205

Environmental Risk Assessment Approaches of Human Pharmaceuticals in the US and EU: A Case Study with Neuroactive Steroid

Benjamin Burruss ¹ , Allan Ader ² , and Janet Gould ¹

¹ SafeBridge, New York City, New York, United States

² SafeBridge, Mountain View, California, United States

Regulations in the United States (US) and European Union (EU) require environmental risk assessments of active pharmaceutical ingredients (APIs) intended for human use. While regulations from these jurisdictions share similar objectives, notable differences in their application exist as highlighted in a case study with a neuroactive steroid API. Under the US framework, the estimated exposure in the environment was < 1 ppb, and no extraordinary circumstances were identified based on the API’s toxicity profile and mechanism of action. Therefore, the requirements for a categorical exclusion were satisfied. Under the EU framework, the screening-level exposure estimate triggered aquatic toxicity testing and a higher-tiered risk assessment where a risk quotient (RQ) of < 1 is acceptable. The neuroactive steroid was not readily biodegradable nor was it inhibitory to microbial function in activated sludge up to 1000 mg/L, indicating negligible risk (RQ = 4.5 x 10-7) to sewage treatment plant function. In aquatic organisms, fish were the most sensitive (NOEC of 2.8 µg/L) in a fish early life stage study (OECD 210), which also indicated low environmental risk (RQ = 0.16). The API was shown to rapidly partition to and degrade in sediment. Therefore, a spike sediment-water toxicity study with the midge (Chironomus riparius) was conducted in which no effects were observed up to 910 mg/kg sediment [dry weight] resulting in low environmental risk (RQ = < 0.00183). Using the US and EU frameworks, the API was considered of low risk to the aquatic environment including sediment-water systems based on pharmaceutical use.

P206

Comparison of Mutagenic Potency of Three Positive Controls in the Enhanced Ames Test

Leon Stankowski ¹ , and Jacqueline Banks ¹

¹ Charles River Laboratories, Skokie, Illinois, United States

The EMA, Health Canada, and the US FDA have proposed a modified “enhanced Ames test” (EAT) to improve detection of mutagenic nitrosamines. Mutagenicity of two suggested nitrosamine-specific positive controls, N-nitrosodimethylamine (NDMA) and 1-cyclopentyl-4-nitrosopiperazine (CPNP), as well as our standard 2-aminoanthracene (2AA) positive control, were compared in two or three trials using EAT conditions (strains TA98, TA100, TA1535, TA1537 and WP2 uvrA pKM101; liquid preincubation treatment; 30% v/v induced rat and hamster S9s). NDMA was reproducibly positive (inducing dose-dependent >2- or 3-fold increases in revertant frequencies) in WP2 uvrA pKM101 with rat and hamster S9s, but in TA100 and TA1535 only with hamster S9. In contrast, CPNP was reproducibly positive in TA100 and TA1535 with rat and hamster S9s, but in WP2 uvrA pKM101 only with hamster S9. Negative results were observed for NDMA and CPNP, at dose levels up to 5000 µg/plate, in TA98 and TA1537 with both S9s. 2AA was positive in all five strains with both S9s, and at much lower dose levels than the nitrosamines. 2AA induced base pair substitution and frameshift mutations (i.e., in all strains), while NDMA and CPNP induced only base pair substitutions. The most potent responses observed (revertants/µg/plate), based on the initial linear portion of the dose-response curves, were: NDMA in WP2 uvrA pKM101 with rat (0.21-0.49) or hamster (1.4-3.1) S9s; CPNP in TA1535 with rat (0.48-0.53) or hamster (3.9-5.5) S9s; 2AA in TA100 with rat (320-460) or hamster (1800-2100) S9s. The EAT is well suited to detect NDMA, CPNP and 2AA.

P207

Influence of Vaccine-Induced Anti-PEG Antibodies on In Vitro Measurement of Pegylated Drug Concentrations

Elizaveta Svyatova ¹ , Zhuoming Liu ¹ , Robyn Becker ¹ , Jungeun Sung ¹ , Adetunji Fayomi ¹ , Jeffry Florian ¹ , James Weaver ¹ , Rodney Rouse ¹ , and Kristina Howard ¹

¹ US FDA, Silver Spring, Maryland, United States

The adoption of mRNA-based technology for vaccine creation has introduced new vaccine components, notably polyethylene glycol (PEG), to global populations. We investigated the presence of anti-PEG antibodies produced in response to mRNA-based vaccines in human serum and evaluated their impact on our ability to measure drug concentration in human serum as would be completed in the assessment of new and biosimilar pegylated drugs. We analyzed serum samples from individuals who received COVID-19 vaccines, as well as unvaccinated controls, for the presence of anti-PEG antibodies and their impact on the quantification of a pegylated drug product (pegfilgrastim). We detected increased anti-PEG antibody titers, persisting for extended periods, in individuals who received the mRNA-1273 vaccine compared to those who received the BNT162b2 and Ad26.COV2.S vaccines. Among recipients of the mRNA-1273 vaccine, anti-PEG antibody levels exhibited an approximate 15-fold increase post-vaccination compared to pre-vaccination levels in approximately 33% of individual donor samples. In some individuals, elevated anti-PEG antibody levels remained observable for six months. Serum from donors with elevated anti-PEG antibodies displayed reduced detection of the pegylated drug, pegfilgrastim, measuring up to a 50% reduction in multiple donor sera, using two distinct in vitro bioanalytical methods. This study underscores that anti-PEG antibodies generated by mRNA-based vaccines have the potential to impact bioanalytical assays during the development of pegylated pharmaceutical products.

P208

FDA Montelukast Working Group Studies: Identification of Off-Target Neurological Receptors and Potential Molecular Mechanisms of Drug-Related Neuropsychiatric Adverse Effects

Jessica Oliphant ¹ , Bounleut Phanavanh ² , Sandeep Kondala ² , Svetoslav Slavov ² , Elizabeth Jones ² , Lisa Pence ² , Andrew Goodwin ³ , David Klein ³ , Timothy Robison ³ , and Li-Rong Yu ⁴

¹ FDA, NCTR, Benton, Arkansas, United States

² FDA, NCTR, Jefferson, Arkansas, United States

³ FDA, CDER, Silver Spring, Maryland, United States

⁴ FDA, NCTR, Silver Spring, Maryland, United States

Montelukast (Singulair, CAS #158966-92-8) has been widely used to manage asthma symptoms among millions of adults and pediatrics. Although use has been associated with neuropsychiatric adverse events, a direct link between montelukast exposure and precipitation of neuropsychiatric-related events has not been clearly established, and potential mechanisms of drug-related effects on the central nervous system (CNS) have remained largely unknown. Following the 2019 FDA Advisory Committee Meeting related to this topic, the FDA Montelukast Working Group (MWG) was established to investigate potential for drug-related CNS effects, with the first of several studies being identification of potential CNS targets of montelukast that could contribute to neuropsychiatric safety concerns. Secondary pharmacology studies were conducted using radioligand binding assays and functional assessments, which identified several candidate off-target G-protein coupled receptors (GPCRs) and neurotransmitter transporters expressed in the brain. Emerging data from confirmatory pharmacology studies in primary human CNS cells treated with montelukast and endogenous receptor ligands provide further insight into potential molecular mechanisms of drug-related CNS effects. A pharmacokinetic model mimicking clinically relevant exposure levels in rats was also developed to evaluate in montelukast-related CNS effects, drug exposure levels, and CNS accumulation potential in vivo. This abstract is the first public release of findings from the FDA MWG studies.

P209

A Systematic Evaluation of Off-Target Binding in Antibody-Based Molecules during Clinical and Preclinical Development

Diana Norden ¹ , Benjamin Doranz ¹ , Carmen Navia ¹ , and Jonathan Sullivan ¹

¹ Integral Molecular, Philadelphia, Pennsylvania, United States

Specificity profiling is a requirement for monoclonal antibodies (MAbs) and antibody-based biotherapeutics such as CAR-T prior to initiating human trials. Traditional approaches to assess MAb specificity, primarily tissue cross-reactivity (TCR) studies, have been unreliable and have resulted in undetected off-target binding. Cell-based protein arrays represent an alternative and improved assessment tool for MAb specificity that have been recommended in recent FDA guidance. We developed a cell-based protein array, the Membrane Proteome Array (MPA), to assess binding across the full human membrane proteome. The MPA encompasses ∼6,000 membrane proteins, each individually expressed in their native structural configuration within live or unfixed cells. The MPA enables quantitative and high sensitivity detection using flow cytometry. A systematic review of 250 customer antibodies that were screened using the MPA indicates a surprisingly high off-target rate, with 32% of lead candidates displaying off-target binding. Moreover, about 20% of therapeutic MAbs in clinical development and currently on the market displayed off-target binding. To directly compare cell-based protein arrays to TCR studies, we compared our own MPA data to publicly available TCR data and in several cases identified discrepancies in off-target binding results. Overall, the presented case studies and off-target rates at the different phases of drug approval suggest that off-target binding is a major cause of adverse events and drug attrition that could be ameliorated with better specificity screening.

P210

Considerations and Challenges for Acute Inhalation Toxicity Testing and Classification of Zinc Sulphide under REACH

Daria Gödecke ¹ , Daria Gödecke ¹ , Noömi Lombaert ² , Arne Burzlaff ³ , Christine Spirlet ² , and Paula Janssen ¹

¹ Fraunhofer ITEM, Hannover, Niedersachsen, Germany

² International Zinc Association, Brussels, Belgium

³ EBRC Consulting GmbH, Hannover, Niedersachsen, Germany

Zinc sulphide is a widely used inorganic powder, and its production has reached quantities greater than 1000 t/year. Previous assessments using a read-across approach with other inorganic zinc substances have proven unreliable for predicting toxicity. Therefore, in accordance with OECD guideline 436, an acute inhalation test was implemented to provide more accurate data. This study is crucial for ensuring the safety of workers exposed to zinc sulphide dust and complying with regulatory requirements for occupational health. Due to particle specific properties the maximum attainable concentration of zinc sulphide for an inhalation study was not certain. Two dry dispersion systems were used to aerosolize the zinc sulphide powder, and the generated aerosol was supplied to a nose-only inhalation exposure system. The chamber concentration was measured using an aerosol photometer and a glass fiber filter. The mass median aerodynamic diameter (MMAD) and geometric standard deviation (GSD) of the aerosol were also measured using a cascade impactor. The results showed a maximum attainable chamber concentration of 0.82 mg/L at an MMAD of 1.5 µm consistent over a 4-hour exposure period. The test evaluated respiratory tract effects in rats, as the preferred species following OECD 436. In the study all 6 rats showed no specific symptoms, a good general health status and survived post-exposure observation period up to 14 days. From the results observed the status Unclassified was derived according to GHS. Based on the experimental results, an LC50 was not determined but is considered higher than 0.82 mg/L (the maximum achievable concentration).

P211

Considerations For Establishing Health-Based Exposure Limits (HBELs) For Small Molecule Protein Kinase Inhibitors

Tyler Nicholas ¹ , Alexandre Beaulieu ¹ , Eric Morinello ¹ , Chandrika Moudgal ¹ , and Jessica Graham ¹

¹ Genentech Inc, South San Francisco, California, United States

Protein kinases are involved in various cell cycle functions such as cell growth, differentiation, and development. Consequently, dysregulation of these kinases is frequent in many diseases and small molecule kinase inhibitors (SMKIs) are a growing class of therapeutics. Research and development activities for SMKIs require speculative health-based exposure limits (HBELs) (including occupational exposure limits [OELs]) and an understanding of the health hazards presented by this class of molecules; information which is supplied by occupational toxicologists. Since molecules in early research often lack data, it is imperative to understand the common hazards presented by SMKIs to properly assess the potential risk of new molecules. This work presents a consolidated overview of the health hazard data (nonclinical and clinical) for FDA approved (n=80) SMKIs, their common health hazards from an occupational standpoint, and guidance for establishing OELs for this chemical class. The majority of SMKIs are compounds with oncology indications (88%; 70/80) and administered orally (96%; 77/80). In line with their indirect anticancer activity, most approved SMKIs are not mutagenic (98%; 78/80). However, 13% (10/79) were positive in an in vivo micronucleus or chromosome aberration assay and 99% (79/80) caused developmental toxicity in animal studies. Overall, this information assists in understanding the potential human health hazards associated with SMKIs that can be leveraged when establishing speculative HBELs for SMKIs lacking compound-specific data.

P212

Characterization and Toxicological Assessment of Heated Tobacco Product Aerosol through Nontargeted Analyses

Chanel Griggs ¹ , Virginia Bass ¹ , Clarissa Tatum ¹ , Owen Bussey ¹ , Nicola Gibbins ² , and Mark West ²

¹ RAI Services Company, Winston Salem, North Carolina, United States

² BAT, Southampton, United Kingdom

The FDA’s 2021 final rule on Premarket Tobacco Product Applications requires a full statement of the constituents, including [harmful and potentially harmful constituents] and other constituents. Therefore, the chemical composition of glo Heated Tobacco Products (HTP) was characterized through a non-targeted analysis of the aerosol emissions. A toxicological assessment of the identified compounds was conducted to understand the potential toxicological implications of glo HTP aerosols compared to combustible cigarette smoke. Our study evaluated the aerosol of multiple glo HTP variants using direct inject gas chromatography mass spectrometry (GC-MS), headspace GC-MS, and liquid chromatography tandem mass spectrometry (LC-MS-MS) and identified 708 compounds across all samples. A subset of 362 compounds, identified with higher confidence scores, were prioritized based on the level of toxicological concern. Priority 1 compounds are formally classified as carcinogenic, mutagenic, reprotoxic, or respiratory sensitizers. Priority 2 compounds had positive Derek alerts for carcinogenicity, mutagenicity, and reproductive toxicity. Priority 3 included compounds where no specific toxicological concerns could be identified. 313 compounds (86%) were prioritized as Priority 3, 29 compounds (8%) as Priority 2, and 20 compounds (6%) as Priority 1. Nineteen Priority 1 compounds are known constituents of cigarette smoke and/or tobacco. Only a small subset of compounds identified have the potential to impact the toxicological profile of glo HTP aerosols. Overall, glo HTP aerosols contain fewer and significantly lower levels of constituents than combustible cigarette smoke. This study supports the conclusion that glo HTPs may present substantially lower individual human health risks than combustible cigarettes.

P213

Retrospective Analysis of Nonclinical Regulatory Strategy for Approved Oncology Antibody-Drug Conjugates

Laura Aschenbrenner ¹ , Alyssa Boler ² , and Laurie Baldwin ¹

¹ Labcorp, Madison, Wisconsin, United States

² Labcorp, Bartlett, Tennessee, United States

Antibody-drug conjugates (ADCs) have been in development for the treatment of advanced stage cancer for over 25 years. Currently, there are 13 approved ADC’s that have been approved by FDA and EU. While there are no specific regulatory guidance documents that provide detailed guidance on the nonclinical development of ADCs, several regulatory guidance documents provide a recommended framework for the nonclinical development of ADCs including: ICH S6(R1) Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals, and ICH S9 Nonclinical Evaluation for Anticancer Pharmaceuticals and the associated Questions and Answers Guidance Document. In May of 2023, a new guidance, Generally Accepted Scientific Knowledge (GASK) in Applications for Drug and Biological Products: Nonclinical Information was published by the FDA. We reviewed all 13 of the publicly available pharmacology and toxicology reviews for approved ADCs. Of these 13 ADC’s, 7 were products that utilized microtubulin inhibitor payloads and all of these nonclinical safety programs included either stand-alone payload only studies or additional dose groups that were administered the payload alone in rats, dogs, and/or NHP. Given the robust assessment of mictotubulin inhibitors and similarity in toxicological findings of the microtubulin inhibitors in approved ADC drugs and the literature, we propose that developers of novel ADC’s, could leverage the GASK guidance to develop a rationale for health authorities to replace the need to conduct characterization of previously characterized microtubulin inhibitors. If accepted by regulators, this could accelerate the nonclinical development of ADC’s, reduce animal use, and bring life-saving medicines to patients faster.

Safety Evaluation Nonpharmaceuticals

P301

Comprehensive Evaluation of CAR T-Cell Efficacy and Safety in a Mouse Model of Acute Lymphoblastic Leukemia (ALL) Using Integrated Biodistribution and Kinetics Analysis

Janet Olugbodi ¹ , Arundhati Ghosh ¹ , Kali Cook ¹ , Pang-Kuo Lo ¹ , Tracey Papenfuss ² , Yongping Chen ¹ , and Todd Parsley ¹

¹ Noble Life Sciences Inc, Sykesville, Maryland, United States

² Stage Bio, Mount Jackson, Virginia, United States

Assessing the biodistribution and in vivo kinetics of cell therapy products is crucial for evaluating their efficacy and safety. This study presents a standardized methodology for evaluating CAR T-cell therapy in a xenogeneic acute lymphoblastic leukemia (ALL) model using multiple techniques. NSG mice implanted with human NALM6-GFP/Luc tumor cells received anti-CD19 CAR T-cells intravenously after day 7 post- tumor engraftment. The biodistribution and kinetics of tumor and CAR T-cells were analyzed using flow cytometry, in vivo bioluminescent imaging (IVIS), digital PCR (dPCR), and in situ microscopic pathology. Tumor spread and burden were monitored weekly using IVIS. Flow cytometry quantified CAR T-cells and tumor cells splenocytes. dPCR detected and quantified human tumor DNA and CAR T-cell DNA in various tissues. Pathological evaluations of FFPE tissue blocks provided in situ detection and distribution data. Result showed that CAR T-cell therapy significantly reduced tumor growth in NSG mice compared to untreated controls. dPCR indicated that CAR T-cells were present predominantly in the liver and lungs, with lower levels in the spleen and kidneys. Tumor cells were most abundant in the liver, spleen, lungs, and ovaries. CAR T-cell treatment reduced tumor DNA copies by 50% in the lungs, 82% in the liver, and 25% in the kidneys. Pathologic evaluations confirmed reduced tumor presence in liver sinusoids and extensive tumor burden in the lungs, consistent with IVIS findings. This study utilized multiple methods to evaluate biodistribution and decreased tumor burden following anti-CD19 CAR T-cell immunotherapy in an NSG mouse model of ALL.

P302

Genotoxicity and Cardiotoxicity Studies of Novel Peroxy Acid-Based Alternative Sodium Hypochlorite Candidates

Seung Eui Min ¹ , Seunghee Oh ² , Byung Sun Min ³ , and Kang-Hyun Han ¹

¹ University of Science and Technology, Yuseong-gu, Daejeon, Republic of Korea

² Korea Institute of Toxicology, Yuseong-gu, Daejeon, Republic of Korea

³ Daegu Catholic University, Gyeongbuk, Republic of Korea

Sodium hypochlorite is known as a disinfectant, oxidizer, and bleach. It is highly reactive and volatile, making it an effective disinfectant. It is used extensively across both living and industrial environments, and its market share is increasing. However, when exposed to high concentrations of sodium hypochlorite or if handled improperly, it can have harmful effects on the human body and the environment. Therefore, caution is required, and the development of alternative materials is necessary. This study presents the results of a toxicity screening of sodium dodecanoyloxybenzenesulfonate (LOBS) and 3-(dodecylamino) propane-1,2-diol (LAMP), candidates developed as alternatives to sodium hypochlorite. As a result of the study, in the Ames test, both LOBS and LAMP did not cause reverse mutations regardless of the with or without of the S9 enzyme and were evaluated as negative. In the chromosome aberration test, CHL cells treated for 6 hours with the S9 enzyme and for 6 and 22 hours without metabolic activation did not cause chromosomal abnormalities and were evaluated as negative. As a result of the hERG assay, the inhibition rates of the hERG potassium channel current for LOBS were 6.2%, 11.7%, and 14.3% at concentrations of 1, 10, and 50 μM, respectively. For LAMP, the inhibition rates were 11.1%, 34.9%, and 62.6% at concentrations of 1, 3, and 10 μM, respectively. LAMP was confirmed to have potential cardiotoxicity. Funding Source: This research was funded by grants from the Korea Ministry of Environment (1485018893) and Korea Institute of Toxicology, Republic of Korea (1711195885).

P303

Withdrawn

P304

The Effects of Carbon Nanodots on Ultrasonic Vocalizations and Other Behaviors of C57BL/6J and LDLr -/- Mice

Delicia Esther Cardenas Vasquez ¹ , Radmila Petric ¹ , Zhenquan Jia ¹ , Isabella Pack ¹ , Keith Erikson ¹ , Steve Fordahl ¹ , Kristina El-Khouri ¹ , Tucker Philbrook ¹ , and Chenhao Tang ¹

¹ University of North Carolina at Greensboro, Greensboro, North Carolina, United States

Carbon Nanodots (CNDs) are emerging nanomaterials valued in biomedicine for their cost-effectiveness and low toxicity, particularly in bioimaging and drug delivery. Although short-term studies have demonstrated their low toxicity, the long-term effects of CNDs on living organisms remain unclear. This study hypothesized that prolonged CND administration affects behavior and physiology in lab mice. We tested this administering 2.5 mg/kg CNDs on C57BL/6J and LDLr -/- (Low density lipoprotein receptor knockout) mice during eight weeks and assessing behavioral and physiological changes. Our primary focus was evaluating the effects of CNDs on Ultrasonic Vocalizations (USVs) as indicators of communicative behavior. We also used the Open Field Test and the Elevated Plus Maze Test to evaluate behavioral changes, and a neuromuscular test to assess physical strength. Results indicated that CNDs treated C57BL/6J mice produced fewer USVs than controls, with lower end and maximum frequencies. In LDLr -/- mice, although number of USVs did not differ, vocalizations showed altered spectral characteristics, including lower end frequencies, maximum frequencies, start frequencies, frequency at maximum amplitude, and minimum frequencies. Behaviorally, CNDs treated C57BL/6J mice exhibited increased floor velocity in the Open Field Test. No significant differences were observed between treatment groups in the Elevated Plus Maze Test or neuromuscular tests. These findings suggest that CNDs may induce stress responses, evident through altered vocalization patterns and increased activity levels, particularly in C57BL/6J mice. This study highlights the need for further research into the long-term impacts of CNDs on behavior and physiology in vivo.

Safety Evaluation Pharmaceuticals

P401 - International Travel Grant Recipient

Administration of Sorafenib, an Anticancer Agent, Caused Oxidative Stress and Reproductive Dysfunction in Male Wistar Rats

Raphael Olatoye ¹ , Mary Aisedion ¹ , and Oluwatosin Adaramoye ¹

¹ University of Ibadan, Ibadan, Oyo, Nigeria

Sorafenib (SRB), a multikinase inhibitor is a chemotherapeutic agent against liver, kidney, and thyroid cancers. Its application has been reported to induce toxicological responses in these organs. Still, the knowledge of its toxicity and the mechanism of action on the testis is yet to be fully determined, thus the need for this study. Fourteen adult male Wistar rats were assigned equally into two groups (n=7). Group 1 received vehicle (corn oil) and group 2 received SRB (10mg/kg) thrice weekly for six consecutive weeks. Administration of SRB caused significant increase in the levels of testicular malondialdehyde (an index of oxidative stress) by 65% with a concomitant decrease in testicular activities of catalase and superoxide dismutase by 44% and 25%, respectively when compared to the control. Additionally, markers of inflammation, myeloperoxidase, and nitric oxide increased significantly in the testes by 27% and 21%, respectively. Furthermore, administration of SRB decreased sperm motility, count, and live-dead ratio by 37%, 18%, and 21%, respectively when compared to the control. Evaluation of serum Potassium, Sodium, chloride, and Bicarbonate ions showed that the potassium ion level decreased by 73% in SRB-administered rats. Also, SRB administration decreased the levels of WBC and neutrophils in rats by 45% and 22%, respectively. Histology of testes of the rats treated with SRB showed degeneration of seminiferous tubules with distortion of cyto-architecture of the basement membrane of testes. Generally, prolonged administration of sorafenib to rats induced male reproductive dysfunction leading to poor semen quality via mechanism that involves oxidative stress and inflammation.

P402 - North American Travel Grant Recipient

Effects of BMS 204352, DEC and GoSlo-SR-5-69 on Ovo-SLO-1A’s Response to Emodepside

Charity Njeshi ¹ , Shivani Choudhary ¹ , Sudhanva Kashyap ² , Richard Martin ¹ , Mark McHugh ¹ , and Alan Robertson ¹

¹ Iowa State University, Iowa, United States

² Creighton University, Omaha, Nebraska, United States

Onchocerciasis, caused by Onchocerca volvulus, remains a major health concern especially in Sub-Saharan Africa, despite elimination efforts. Given the absence of a vaccine and the high cost of drug development, drug repositioning has gained attention. Emodepside, a veterinary anthelmintic, has shown promise due to its activity on O. volvulus' SLO-1K channels and has recently completed Phase II clinical trials for onchocerciasis. In this study, we explored the effects of Diethylcarbamazine (DEC), (3S)-(+)-(5-Chloro-2-methoxyphenyl)-1,3-dihydro-3-fluoro-6-(trifluoromethyl)-2H-indole-2-one (BMS 204352), and Sodium 1-Amino-9,10-dioxo-4-((5,6,7,8-tetrahydronaphthalen-2-yl)amino)-9,10-dihydroanthracene-2-sulfonate (GoSlo-SR-5-69), alone and in combination with emodepside, on Ovo-SLO-1A, a specific O. volvulus isoform. We hypothesized that DEC, BMS 204352, and GoSlo-SR-5-69 would enhance emodepside's effect on Ovo-SLO-1A. We expressed Ovo-slo-1a in Xenopus laevis oocytes and using the two-electrode voltage clamp electrophysiology, we tested the compounds individually and in combination with emodepside. Results showed concentration-dependent outward currents with emodepside (EC50 = 0.50 ± 0.08 µM). While GoSlo-SR-5-69 alone didn't activate the channel, it enhanced emodepside's effect when combined, with a significant 130% increase in maximum response. Specifically, maximum responses were 2402.00 ± 85.68 nA with emodepside alone and 5511.00 ± 723.10 nA when combined with GoSlo-SR-5-69, though the EC50 remained similar at 0.42 ± 0.04 µM. BMS 204352 and DEC did not activate or enhance emodepside's effect on the channel. The study suggests that GoSlo-SR-5-69 can enhance emodepside's effectiveness against O. volvulus, offering a potential treatment for river blindness. This finding aids in developing improved anthelmintic strategies and addressing drug resistance in onchocerciasis treatment. Acknowledgment: NIH: R01AI047194 & R01AI155413 and E.A. Benbrook Foundation

P403

Comparison of Time-Dependent Variation of Respiratory Function Parameters in Wistar Rats

Sudhakar Jadhav ¹ , Jeshica Bulsara ² , Kunjan Shah ² , and Manish Patel ²

¹ Master, Vapi, Gujarat, India

² Jai Research Foundation, Vapi, Gujarat, India

Whole-body plethysmography (WBP) is the gold standard non-invasive lung function measurement method in animals. Evaluation of respiratory functions is a prime requirement for regulatory submission of new drugs to assess the secondary pharmacodynamic effect. Standardization of control group data compared to positive and treatment groups at each evaluation time point is essential for statistical assessment. The respiratory data of rats from control groups of different experiments were collected to characterize time-dependent differences. Respiratory parameters such as frequency (Fre), tidal volume (TV), minute volume (MV), inspiratory time (TI), and expiratory time (TE) were selected for evaluation. Data recordings were performed at 0, then continuously for 2 hr, and then for 30 minutes at 24 hr in conscious rats using the WBP. These data were evaluated statistically at 0, 0.5, 1, 1.5, 2, and 24 hr to assess effect. During this evaluation, an increase in the TI, TE, and TV with a simultaneous decrease in the MV and Fre was noted as the time of recording or housing in the animal chamber of WBP increased (i.e., continuous 2 hr). Again, data recorded at a 24-hour interval was comparable to the 0-hour interval. Based on the results, the control rats revealed normal respiratory parameters with time-dependent variation during continuous recording. It is recommended that such time dependent variations be considered as normal while interpreting the outcomes of respiratory safety pharmacology studies of any test product. Consideration should also be focused on using defined time intervals in each experiment to construct strong historical data.

P404

Preclinical Development of LFA-9, a Dual Inhibitor of mPGES-1 and 5-LOX, for Prevention of Colorectal Cancer: GLP-Compliant, IND-Enabling 28-Day Toxicology and Pharmacokinetics Studies (with Recovery) in Rats and Dogs

Harriet Gaskell ¹ , Genoveva Murillo ¹ , Thomas Horn ¹ , Sudhakar Parthasarathy ¹ , Sheila Huang ¹ , Matthew Reed ² , Adrienne Schucker ³ , Brian Cholewa ⁴ , C Rao ⁵ , and David McCormick ¹

¹ IIT Research Institute, Chicago, Illinois, United States

² Charles River - Pathology Associates, Skokie, Illinois, United States

³ Charles River - Pathology Associates, Frederick, Maryland, United States

⁴ National Cancer Institute, Rockville, Maryland, United States

⁵ University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, United States

Inhibition of arachidonic acid metabolism to biologically active lipids such as prostaglandins (PGs) and leukotrienes (LTs) is an effective means to prevent cancer in the colon, rectum, and other sites in preclinical models. Membrane prostaglandin E synthase-1 (mPGES-1) is a key pathway in PG synthesis; 5-LOX is a major pathway in LT and hydroxyeicosatetraenoic acid (HETE) synthesis. LFA-9 is a licofelone analog that inhibits both mPGES-1 and 5-LOX and prevents colorectal carcinogenesis in rodent models. As part of the preclinical development of LFA-9, GLP-compliant, IND-enabling oral 28-day toxicity and pharmacokinetics studies were performed in rats and dogs. LFA-9 was administered to rats by gavage at doses of 0 (vehicle control), 25, 75, and 225 mg/kg/day. The high dose of LFA-9 induced small but statistically significant reductions in body weight gain in both sexes; these were partially reversed during the recovery period. Confirming and extending range-finding data, LFA-9 induced dose-related but reversible elevations in ALP and ALT in both sexes. Rats in the high dose group demonstrated high incidences of hepatic centrilobular hypertrophy and periportal vacuolation; all microscopic changes in rats were completely reversed during the recovery period. In the toxicity and PK study in dogs, LFA-9 was administered by capsule at doses of 0 (capsule control), 10, 30, and 90 mg/kg/day. No clinical evidence of toxicity and no microscopic evidence of liver toxicity was seen in any animal. These data confirm the results of range-finding studies that identified the liver as a target of LFA-9 toxicity. (NCI HHSN261201500024I).

P405

Species-Specific Liver Microtissues: A Set of Microphysiological Systems to Assess Translational Hepatotoxicity in Drug Development

Maria Vittoria Colombo ¹ , Friederike Wenz ¹ , Lola Fäs ¹ , Agnieszka Pawlowska ¹ , Armin Wolf ¹ , and Bruno Filippi ¹

¹ InSphero AG, Schlieren, Zurich, Switzerland

The toxicity of substances in animals and humans can be different. For example, fialuridine is more hepatotoxic in humans than in monkey, dog, and rat, due to the expression of a drug transporter on the mitochondrial membrane in humans. Therefore, understanding species-specific differences in toxicities is key to the safety assessment process of drug candidates. To this aim, microtissues composed of Macaca fascicularis (Cynomolgus), Canis familiaris (Beagle), or Rattus norvegicus (Sprague Dawley) primary parenchymal and non-parenchymal liver cells were developed and characterized for viability, morphology, albumin production and temporal stability. We then evaluated cytotoxicity and transcriptomic response to two FDA-approved drugs, fialuridine and chlorpromazine. Human, monkey, dog and rat liver microtissues are viable and functional for up to seven days of culture. Fialuridine is markedly more cytotoxic in human liver microtissues than in their animal counterparts, hence recapitulating the in vivo observation. Conversely, chlorpromazine, which is not reported to be more hepatotoxic in a specific species, is cytotoxic at similar drug concentrations across human, monkey, dog, and rat liver microtissues. Transcriptomic analysis of fialuridine response in liver microtissues revealed a different signature specifically in human liver microtissues, with significant changes in genes involved in DNA damage response, in agreement with different work highlighting the mutagenesis potential of fialuridine in human hepatocytes. Altogether, the results of this work suggest species-specific liver microtissues form a set of micro-physiological systems to study the translation of the hepatotoxicity, or the innocuity, of a compound in one or more animal species in humans.

P406

Repeated Intraperitoneal Injections of a Liposomal Suspension in Newborn Minipigs to Evaluate Its Safety Before Pediatric Use

Hervé Giorgi ¹ , Meriam Kabbaj ² , and Diana Evstafeva ²

¹ Genfit SA, Loos, Nord-Pas-de-Calais, France

² Versantis AG, Zurich, Switzerland

VS-01 is an intraperitoneal (i.p.) liposomal suspension for the treatment of hyperammonemia caused by inborn errors of metabolism affecting the urea cycle. Accumulation of ammonia, produced by the breakdown of proteins, can manifest shortly after birth and correlates with a high mortality rate and a poor neurological prognosis. Citric acid-based transmembrane pH-gradient liposomes in VS-01 rapidly capture ammonia that passively diffuses from the blood into the peritoneal cavity thereby increasing its body clearance. The safety of daily three-hour i.p. sessions of VS-01 for ten days was confirmed in adolescent Göttingen minipigs. Herein, the feasibility and preliminary tolerability of daily i.p. administrations of VS-01 in newborn minipigs for seven days were assessed. Six Göttingen piglets (four males and two females) were treated from post-natal day (PND) 2 to PND9. VS-01 was administered at 30 mL/kg under full isoflurane anesthesia in the lower right or left abdominal quadrant. The localization of the formulation and potential erroneous dosing in the bladder was examined using ultrasound imaging. Overall, the dosing procedure was proven successful, and VS-01 was well tolerated. Piglets demonstrated expected growth for their age and no severe clinical signs or local reaction to the treatment. Vocalization at handling was observed in one female, while watery feces were observed in one male and one female animals. Daily i.p. dosing with VS-01 in piglets from PND2 over seven days is therefore considered feasible up to 30 mL/kg. These results will have to be confirmed in a GLP pivotal juvenile toxicity study before clinical use.

P407

Benefits of Early In Vitro Screening for Seizure Liability in Problem Solving and Decision Making

Kimberly Rockley ¹ , Michael Morton ² , and Ruth Roberts ²

¹ ApconiX, Alderley Edge, Cheshire, England, United Kingdom

² ApconiX, Macclesfield, United Kingdom

Seizure liability remains a significant cause of attrition throughout drug development in pre-clinical and clinical studies. We have developed an approach utilizing hiPSC-neuronal cell microelectrode array (MEA) and ion channel screening for early seizure prediction. In our MEA assay, seizurogenic compounds were identified correctly with high predictivity, and correlations were observed between the in vitro and clinical exposures of many therapies known to cause seizure. In the early phase of nonclinical testing MEA studies can be used to derisk and prioritize a chemical series. E.g., after testing several compounds, one was identified with low seizure risk compared to the others in the series – this compound had distinct structural features. In another study of compounds undergoing nonclinical testing, exposures that caused no CNS signs or convulsions in rats, aligned with the results of the MEA study. Conversely, where convulsions were reported in rats, seizurogenic responses were present in the MEA study at comparable concentrations. These studies can also be used to determine the human relevance of seizures observed in nonclinical studies. E.g., nonclinical testing of a compound caused convulsions only in dogs. Testing metabolites in the MEA assay revealed only the dog-specific metabolite caused a seizurogenic phenotype. In addition, screening this metabolite against the seizure-related ion channel targets revealed a hit, providing mechanistic insight and also the opportunity for compound redesign. Collectively, these studies demonstrate the utility of this approach for early seizure prediction to provide mechanistic information, early de-risking, and support optimal drug design using human in vitro models.

P408

Assessment of Genotoxicity Prediction Using New Approach Methodologies (NAMs): A Comparative Study with In Vitro Micronucleus Assays on Small Molecule Pharmaceuticals

Tomas Lagunas ¹ , Gabby Cole ¹ , Yodi Melnikov ¹ , Catrin Hasselgren ¹ , Aaron Fullerton ¹ , and Yu Zhong ¹

¹ Genentech, South San Francisco, California, United States

Genotoxicity evaluation to identify genetic changes that may lead to carcinogenic potential is critical in non-clinical safety assessment for small molecule drug development. According to the International Conference on Harmonisation S2(R1) regulatory guidance, genotoxicity is assessed using a battery of in vitro and in vivo assays. The micronucleus (MN) assay detects micronuclei formation, with Fluorescence In Situ Hybridization staining elucidating clastogenicity versus aneugenicity mechanisms. This study assessed the genotoxic potential of 22 anonymized small molecules in development using Litron MicroFlow® and MultiFlow® assays in human lymphoblastoid TK6 cells. Currently, the high content micronucleus (HC-MN) assay in Chinese Hamster Ovary cells serves as our screening assay, while the human peripheral blood lymphocytes micronucleus (HPBL-MN) assay is used for confirmatory testing. To evaluate the predictive value of the Litron assays as screening tools, we retrospectively analyzed the concordance between the Litron assays and the manual MN assay in HPBLs. Our findings indicate that Litron assays, with sensitivity ∼0.8 and specificity ∼0.7, outperform the standard HC-MN screening assay, with sensitivity ∼0.7 and specificity ∼0.2, in predicting HPBL-MN results for these small molecules. Additionally, the MultiFlow® assay provided mechanistic insights for MN-positive compounds, offering a comprehensive understanding of genetic damage mechanisms. We conclude that integrating New Approach Methodologies (NAMs) enhances mechanistic insights and predictive capabilities, refining genotoxicity screening processes and improving accuracy and reliability in toxicological evaluations. This approach shows promise in effectively categorizing compounds and delivering valuable insights to early-stage project teams for molecule optimization.

P409

Toxicology Assessment of High Concentration XeriJect® Trastuzumab Biosimilar Administered Subcutaneous in Cynomolgus Nonhuman Primates

Rick Fitch ¹ , and Gabriela Toomer ²

¹ Xeris Pharmaceuticals, O'Fallon, Missouri, United States

² Xeris Pharmaceuticals, Chicago, Illinois, United States

Background: The proprietary XeriJect® technology (Xeris Pharmaceuticals, Inc., Chicago, IL, USA) uses a novel viscoelastic suspension (VES) for delivery of high-concentration, low-volume subcutaneous (SC) injections of therapeutic antibodies. Trastuzumab (TmAb) is a monoclonal antibody that selectively binds and inhibits human epidermal growth factor receptor 2 protein (HER2) used for the treatment of certain types of breast, stomach, and esophageal cancer, and is commercially available for both intravenous (IV) infusion and SC injection (Herceptin® IV and Herceptin Hylecta® SC, Genentech Inc., South San Francisco, CA, USA). XeriJect TmAb is a stable, high concentration (417 mg/mL) formulation of a trastuzumab biosimilar for SC administration, thus enabling lower injection volumes. The aim of the study was to evaluate potential toxicity of XeriJect TmAb SC compared to US Herceptin IV infusion administered at 50 mg/kg/day weekly for five doses to cynomolgus non-human primates (NHP). Results: There were no XeriJect TmAb-related mortality or clinical signs, or effects on body weights, ophthalmology, electrocardiographic examinations, clinical pathology parameters, organ weights, or macroscopic and microscopic examinations. Slight edema was noted following dosing, but the incidence was similar between the XeriJect Vehicle and XeriJect TmAb. No erythema was noted following dosing of either the XeriJect Vehicle or XeriJect Trastuzumab. No anti-drug antibodies for trastuzumab were observed for either XeriJect TmAb or Herceptin IV. Conclusion: Weekly administration of XeriJect TmAb at 50 mg/kg/day SC to cynomolgus NHP for five consecutive doses was well tolerated when compared to US-Herceptin administration.

P410

Integrative Analysis of Human and Cynomolgus Monkey Peripheral Blood Mononuclear Cells after T Cell Stimulation via scRNA-seq

Lindsey Mack ¹ , Zimei Huang ¹ , Kathleen Hoyt ¹ , Lindsey Dzielak ¹ , April Masters ¹ , Ernest Raymond ¹ , and Kari Neier ¹

¹ Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, United States

T cell engagers are biologics that bind to a tumor antigen and bind to and activate T cells inducing a tumor-targeted adaptive immune response. The binding specificity of these molecules necessitates the use of non-human primates (NHPs) in preclinical toxicology studies. The cynomolgus macaque (cyno) is the most used NHP in drug safety assessment due to its small size, non-seasonal reproductive cycle, and high homology with humans. However, the differences between cyno and human adaptive immune responses have not been fully characterized, limiting translatability of safety-related findings to the clinic. Therefore, we conducted a single cell RNA-sequencing (scRNA-seq) study assessing peripheral blood mononuclear cells before and 24, 48, and 72 hours after T cell receptor mediated stimulation to characterize the adaptive immune response in both species. Cross-species data integration was performed to control for species specific gene expression signatures. Data integration was successful at each time point, but 24 hours after stimulation proved the most divergent between species. This is consistent with the knowledge that macaques have a higher T cell activation threshold than humans, resulting in slower activation when stimulated through the T cell receptor. Further analysis will reveal cell-type- and species-specific gene expression signatures and overall functional characteristics different between species. Understanding this response in vitro will provide information about the cascade of events that occur due to T cell activation, similarly to the response to T cell engagers. An in-depth understanding of this response will improve interpretation and translation of pre-clinical study results to the clinic.

P411

Quantitative Systems Toxicology Modeling of Otenaproxesul Liver Enzyme Elevations for Predicting Liver Safety of Acute Otenaproxesul Dosing

Brett Howell ¹ , Nader Hamzavi ² , Pallavi Bhargava ² , Vinal Lakhani ² , and Jeffrey Woodhead ²

¹ Simulations Plus Inc., Salisbury, North Carolina, United States

² Simulations Plus Inc., Durham, North Carolina, United States

Otenaproxesul is a drug candidate for pain consisting of naproxen bound to a 4-hydroxythiobenzamide moiety that releases the antioxidant, hydrogen sulfide (H2S), which makes otenaproxesul less gut-toxic than naproxen. Liver enzyme elevations were observed (mostly after cessation of dosing) in some clinical trials. Quantitative systems toxicology (QST) modeling using DILIsym was undertaken to explain the observed liver signals and predict the safety of acute protocols. A PBPK model for several formulations of otenaproxesul was constructed in GastroPlus. DILIsym was modified to represent the scavenging of reactive oxygen species (ROS) by the H2S released by otenaproxesul. In vitro experiments describing the potential mechanism of drug-induced liver signals for both otenaproxesul and its main metabolite, M25 (naproxen), were conducted and used. Simulation results suggested that the post-treatment liver enzyme elevations were due to the replacement of endogenous antioxidant mechanisms with the drug-supplied H2S. In individuals with pre-existing adaptation to elevated ROS, the influx of H2S would lead to a decrease in ROS and eventually the cessation of the adaptive mechanisms (i.e. de-adaptation) that enabled the liver to manage the elevated ROS. Upon cessation of dosing, endogenous antioxidant mechanisms were slow to recover (i.e. re-adapt) while ROS returned to pre-treatment levels, leading to liver effects. This hypothesis was able to explain all the post-treatment enzyme elevations observed in the clinic. The proposed acute dosing protocols were predicted to avoid full de-adaptation (i.e. liver safe). Subsequent clinical experience validated the prediction of liver safety for one of the proposed acute dosing protocols.

P412

Establishing a Decision Tree Model for Developmental and Reproductive Toxicology Strategy for Gene Therapies

Christopher Drummond ¹ , Heather Jewell ² , and Justine Cunningham ²

¹ Regenxbio, West Olive, Michigan, United States

² RegenxBio, Rockville, Maryland, United States

Risk for germline transmission is a safety concern for the application of gene therapies (GT) in patients of reproductive potential. Irrespective of the route of administration most viral vector systems used in GT distribute to the systemic circulation, thereby carrying an inherent risk for off-target germ cell transduction. Examination of the gonads is required during safety assessment of GT but conduct of developmental and reproductive toxicity (DART) studies are not common. No clear strategy or guidance addresses how to assess the reproductive risk to patients if vector and or transgene DNA is present within the gonads. We developed a decision tree model to frame the reproductive safety of GT taking into consideration 4 key factors underlying this risk: 1) patient population, 2) viral vector, 3) presence of vector DNA, and 4) type of transgene. We applied this decision tree to assess reproductive risk for three different AAV-delivered GT programs that are delivered intra-ocular, intra-thecal, or systemically for use in elderly, reproductively mature, juvenile, or neonatal populations. Application of this model allowed us to quickly stratify reproductive risk. In gonads, where vector DNA signal was identified, we used tissue in situ hybridization to examine germ cell transduction. For products where the transgene represented a reproductive risk, we developed probes to examine transgene expression. This decision tree model may be useful to safety scientists to assess reproductive risk of an intended GT and identify if in situ hybridization is sufficient to de-risk or if DART studies are warranted to assure patient safety.

P413

Development of an Electrochemiluminescence-Based Immunoassay to Evaluate T Cell-Dependent IgG and IgM Responses to Keyhole Limpet Hemocyanin (KLH) Administration in the Juvenile New Zealand White Rabbit

Eric Bruder ¹ , and Desiree Flouro ¹

¹ Labcorp Drug Development, Madison, Wisconsin, United States

An important component of the safety assessment of new therapeutics is the examination of off-target immune modulation. A well-documented approach relies on the assessment of the T cell-dependent antibody response (TDAR). The present study sought to validate TDAR assessment in New Zealand white rabbits. After drawing pre-dose samples, weanling kits (five kits/sex/dose) at postnatal day (PND) 49 were administered KLH at two different doses (300 and 1000 µg/kit). Serial blood sampling was performed over a period of three weeks, followed by a second KLH challenge on PND80 and subsequent sampling. Anti-KLH IgG and IgM were measured using an electrochemiluminescence-based immunoassay that was developed and validated in-house. Briefly, pooled pre-dose serum (naïve control) and a pooled subset of post-dose serum (positive control) were isolated, with the intent of capturing high titers of both IgG and IgM in the positive control. Two sets of goat anti-rabbit IgG/IgM antibodies and other assay parameters (e.g. dilution series) were assessed. Using assay response values from naïve controls, cut-point titer for IgG and IgM could then be calculated, and the temporal TDAR response was subsequently evaluated. There was a profound increase in anti-KLH IgG following the first dose of KLH, and values plateaued around PND60. There was a minimal IgG response to the second dose. Anti-KLH IgM responses mirrored that of IgG, except the magnitude of increased IgM was smaller and responses displayed more biological variability. The present data indicate that the rabbit is an attractive option when assessing the TDAR and drug-induced immunogenicity.

P414

Advancing Integrated Omics for Decision-Making and Prediction of Renal Toxicity, Hepatic Toxicity, and Carcinogenicity—An International, Cross-Industry, Academia, and Government Initiative

Deidre Dalmas ¹ , Scott Auerach ² , Chrissy Crute ³ , and Connie Mitchell ³

¹ Independent (on behalf of the HESI eSTAR Committee), Collegeville, Pennsylvania, United States

² National Institute Environmental Health Sciences, Durham, North Carolina, United States

³ Health and Environmental Sciences Institute, Washington, District of Columbia, United States

Transcriptomics has been a valuable tool in toxicology for over two decades, enhancing traditional animal studies and deepening our understanding of underlying mechanisms of toxicity. However, integrating transcriptomics and other emerging omics data—such as metabolomics and high content image-based analysis (e.g., Cell Painting)—into decision-making processes still poses significant challenges. Innovative approaches in systems toxicology for risk assessment are being pioneered by a cross-sector group spanning academia, industry, and government (HESI Emerging Systems Toxicology for the Assessment of Risk (eSTAR) committee). This collaboration has made significant progress and fueled projects aimed at promoting identification/adoption of new translational/predictive risk assessment tools, as well as identifying genomic biomarkers in biofluids and tissues that are crucial for predictive toxicology. The first initiative focuses on deriving microRNA (miR) biomarkers for prediction of renal toxicity by mining existing animal studies in parallel with new advanced in vitro research to develop novel genomic signatures (miR-34a/miR-34c/miR29c-3p/miR486m/miR-192/miR-21) that predict kidney damage. The second initiative focuses on development of mechanism-based transcriptomic-signatures (e.g.,AhR/ER/PXR) from short-term in vivo studies that predict rat liver tumors and inform on relevance to human carcinogenicity risk. The final project focuses on Cell Painting and integration with transcriptomics/proteomics using CIVMs through evaluation of >1,500 data-rich compounds associated with liver toxicity to generate more sensitive/predictive methods for hepatotoxicity risk-assessment. These efforts culminated in creation of a comprehensive-toolkit of biomarkers/analytical pipelines that not only elucidate novel mechanisms but enhance the use of genomic tools in risk-assessment and decision-making processes for pharmaceutical compound development that are more efficient, and less animal-reliant.

P415

Withdrawn

P416

Nonclinical Assessment of KER-012, a Novel Modified ActRIIB Ligand Trap, for the Treatment of Pulmonary Arterial Hypertension

Sebastien Sannajust ¹ , ffolliott Fisher ¹ , Jennifer Lachey ¹ , Jasbir Seehra ¹ , and Francis Wolenski ¹

¹ Keros Therapeutics, Lexington, Massachusetts, United States

Pulmonary arterial hypertension (PAH) is a rare medical condition characterized by vascular resistance and high blood pressure in the lungs that can lead to heart failure. KER-012 is a novel modified activin receptor type IIB (ActRIIB) fusion protein trap designed to target select TGF-β ligands, including activin A, activin B, GDF8 and GDF11, to rebalance the defective activin receptor type II signaling observed in PAH. To support initiation of a Phase 2 clinical trial, the nonclinical safety package included GLP-compliant general toxicity studies in rats and cynomolgus monkeys of up to 6-months in duration, embryo-fetal developmental studies in rats and rabbits, and an in-vitro cytokine release assay. An overview of the study designs, key endpoints, results, and first-in-human dose selection rationale are described. Overall, KER-012 was well-tolerated with no cardiovascular, respiratory, or neurological effects. Toxicological findings included reversible non-adverse mammary gland hyperplasia and non-remarkable clinical chemistry changes. Sotatercept, a TGF-β ligand trap approved for the treatment of PAH in the United States, binds activins, GDF8 and 11, in addition to tightly binding multiple BMP ligands. Key published nonclinical toxicological findings of sotatercept include effects on the kidney, reproductive tract, and hematology parameters. The pharmacology and toxicology profile of KER-012 was designed to have a selective, BMP-sparing ligand binding profile with the potential to provide benefit in multiple tissues in patients with PAH, including in the vasculature, lung and heart. The differences in nonclinical safety profiles observed between KER-012 and sotatercept may potentially be due to differential inhibition of multiple BMPs.

P417

ICH S1B Weight-of-Evidence Carcinogenicity Assessment for GLP-1RA Drugs

Alex Gauthier ¹ , Manoochehr Khazaee ² , and Isaac Mohar ³

¹ Gradient, Boston, Massachusetts, United States

² Gradient, Charlottesville, Virginia, United States

³ Gradient, Seattle, Washington, United States

Glucagon-like peptide 1 receptor agonists (GLP-1RA) are peptide drugs used to treat diabetes and obesity. Most GLP-1RA drugs have a boxed warning for risk of thyroid C-cell tumors. We conducted a critical analysis of nonclinical toxicology for US FDA-approved GLP-1RA drugs with the objectives of understanding their genotoxicity and carcinogenicity potential and determining whether an ICH S1B-based weight-of-evidence carcinogenicity assessment is appropriate for new drugs in this class. The following GLP-1RA drugs were evaluated: exenatide, dulaglutide, semaglutide, tirzepatide, liraglutide, lixsenantide, and albiglutide. Standard genotoxicity tests were negative across GLP-1RA drugs (dulaglutide was not evaluated). Carcinogenicity studies showed increased incidence in treatment-related thyroid C-cell tumors in all rat studies and in select mouse studies with no increased tumors in rasH2 transgenic mice. Thyroid C-cell tumor incidence correlated with hyperplasia and was reported to be associated with GLP-1 receptor-mediated proliferative response in C-cells. Thyroid C-cell hyperplasia was evident in subchronic and chronic repeat-dose rodent studies. In contrast, C-cell hyperplasia was not reported in chronic studies in nonhuman primates (NHPs) (≤ 82 weeks) or dogs (≤ 12 months). Furthermore, histopathological studies have reported a lack of GLP-1 receptor expression on C-cells of normal thyroid tissue in NHPs and humans. Overall, available data show that GLP-1RA drugs can experimentally induce thyroid C-cell hyperplasia and tumor formation in rodents by a nongenotoxic mechanism with uncertain human relevance. Therefore, related to a regulatory requirement for a boxed carcinogenicity warning, a weight-of-evidence approach to carcinogenicity assessment is appropriate for new in-class GLP-1RA drugs.

P418

Nonclinical Determination of Pharmacodynamic, Pharmacokinetic, and Toxicology Characteristics of an Anti-TNFα Monoclonal Antibody ZB002

Pallavi Limaye ¹, Mark Matijevic², Minggeng Gao², and Xiao Feng²

¹ Zenas BioPharma, Overland Park, Kansas, United States

² Zenas BioPharma, Waltham, Massachusetts, United States

ZB002 is a potential best-in-class recombinant human monoclonal antibody directed against human tumor necrosis factor alpha (TNFα) with a modified Fc region for higher FcRn binding to extend the half-life than that of adalimumab. The nonclinical pharmacodynamic, pharmacokinetic, and toxicology characterization of ZB002 is completed. The pharmacodynamic studies confirmed that ZB002 binds to and neutralizes human TNFα and exhibited higher FcRn binding than adalimumab. No antibody-dependent cell-mediated cytotoxicity (ADCC) and an attenuated complement-dependent cytotoxicity (CDC) by ZB002 compared to adalimumab were observed. Surface plasmon resonance and species cross-reactivity assessments showed the highest binding affinity in human and cynomolgus monkey supporting monkey as the relevant toxicology species. After single SC dose in mice the half-life of ZB002 was two-fold greater than that of adalimumab. Following single and weekly SC dosing for 13-weeks in monkeys low serum clearance, low volume of distribution, dose-proportional increase in systemic exposure, no sex-dependence in systemic exposure, and >62% bioavailability were observed. There was no marked accumulation following the weekly SC dosing for 13-weeks in monkey. Antidrug antibodies were detected without any impact on ZB002 serum concentrations. ZB002 was well tolerated in monkeys following a single SC dose with the maximum tolerated dose (MTD) of ≥ 300 mg/kg. Once weekly SC dosing for 13-weeks with a 4-week recovery period was well tolerated with the no-observed-adverse-effect-level (NOAEL) of 200 mg/kg (highest tested dose). In conclusion, ZB002 has a favorable safety profile with similar pharmacological activities to adalimumab but with longer half-life to support ongoing clinical development.

P419

Embryo-Fetal Development and Pre- and Postnatal Development Toxicity Study in Mice Following a Single Intravenous Administration of DTX301

SunJung Kim ¹ , Derek Kennedy ² , Tim Chiu ² , Dimosthenis Giamouridis ³ , Elise Lewis ³ , and Kelly Flagella ²

¹ Ultragenyx Pharmaceutical Inc., Fairfax, Virginia, United States

² Ultragenyx Pharmaceutical Inc., Novato, California, United States

³ Charles River Laboratories, Horsham, Pennsylvania, United States

DTX301 is a non-replicating, self-complementary, recombinant adeno-associated virus serotype 8 (AAV8) vector that contains a codon-optimized, wild-type human ornithine transcarbamylase (hOTC) coding sequence for the treatment of OTC deficiency. In this study, the potential impact on maternal toxicity and embryo-fetal development (EFD), as well as pre- and post-natal development (PPND) during gestation and lactation in C57BL/6 mice, were evaluated. DTX301 or vehicle control were intravenously administered to females on Gestation Day (GD) 6 at doses up to 7E13 GC/kg. Half of the pregnant females were assigned to the EFD Cohort 1, and all remaining pregnant females were assigned to the PPND Cohort 2. Standard readouts for EFD and PPND were performed. Additionally, DTX301 vector DNA was examined in pregnant females and the F1 generation. There were no effects on gestation, parturition, lactation, or maternal behavior in the F0 generation dams. Also, no effects on fetal development, mortality, growth, or sexual maturation in their F1 generation offspring at any dose level were observed. Despite the presence of vector DNA in the blood and placenta of F0 females, none of the tissues evaluated (e.g., liver and gonad) of the F1 generation male and female mice from either dose group contained DTX301 vector DNA. Thus, the maternal no-observable-adverse-effect level (NOAEL) for maternal toxicity and reproductive toxicity, viability, and growth in the F1 offspring is 7E13 GC/kg, the highest dose tested. Therefore, the risks of DTX301 for pregnant females and the F1 generation are considered low.

P420

The ABCs of Deriving Health-Based Exposure Limits for Antibody-Drug Conjugates: Adjustment Factors, Best Practices, and Challenges in Quantitative Hazard Assessment

Nicholas Griggs ¹ , and Janet Gould ²

¹ SafeBridge Consultants, Inc., Mountain View, California, United States

² SafeBridge, New York City, New York, United States

Antibody-drug conjugates (ADCs) are targeted therapeutics composed of monoclonal antibodies coupled to cytotoxic small molecules, offering significant promise in oncology. Their complex chemical, pharmacokinetic, and pharmacodynamic properties pose unique challenges in the development of health-based exposure limits (HBELs). Gemtuzumab ozogamicin (Mylotarg), the first ADC approved by the FDA in 2000, targets human CD33 antigen expressed on tumor cells where it delivers N-acetyl gamma calicheamicin dimethyl hydrazide, a DNA-damaging cytotoxic agent, and was chosen as a representative example to guide discussion about variations of the HBELs, typically 0.01 to 1 µg/m3 for an occupational exposure limit (OEL) and 0.1 to 10 µg/day for a permitted daily exposure (PDE) value. This range occurs based on the selection of the critical effect dose and use of adjustment factors influenced by clinical dosing regimens (0.25 to 9 mg/m2 every 1 to 4 weeks), estimated bioavailability for different routes of exposure (20-50% inhalation vs. 100% parenteral), potential for direct lung effects, influence of the antibody on non-cancer target organs, and pervasiveness of free drug. For ADCs in clinical development, additional adjustment factors ranging from 3-10 to account for limited human exposure, database completeness, and severe toxicities (e.g., reproductive, developmental, mutagenicity, carcinogenicity) can result in even greater variation in the derived HBELs. Lastly, challenges in classifying ADCs using the Globally Harmonized System (GHS) of Classification and Labelling of Chemicals, generally corresponding to the severe toxicities mentioned above, are discussed. This presentation provides a comprehensive framework for accurate quantitative safety evaluation of ADCs in development.

P421

Cellular Characterization of the Comprehensive In Vitro Proarrhythmia Compound Library Using 2D vs. 3D Human iPSC Cardiomyocytes Models

Nicholas Radio ¹ , Claudia Jaffe ² , Brent Samson ¹ , and Elizabeth Vargas ²

¹ Lumencor, Pittsburgh, Pennsylvania, United States

² Lumencor, Beaverton, Oregon, United States

The ability to evaluate the safety potential of pharmaceuticals using cardiomyocytes from induced pluripotent stem cells (iPSCs) has advanced cardiac therapeutic potential. To better understand iPSC-cardiomyocytes’ value as a 2D and 3D screening model, we set out to compare the Comprehensive in Vitro Proarrhytmia (CiPA) compound effects using the automated Volta Scanner and Berkely Red Sensor of Transmembrane potential (BeRST) voltage-sensitive dye. Human iPSCs were dispensed into 96-well microtiter plates in both a monolayer (2D) and cardioid (3D) format according to manufacturer’s instructions. The CiPA library contains compounds generally known to be safe through compounds known to be highly toxic to human cardiomyocytes and is therefore useful for characterizing sensitivity and selectivity for safety characterization purposes. The results were collected in a double-blind fashion, with cardiac toxicity potential determined for each unique compound. Both 2D and 3D cardiac models demonstrated high selectivity for those compounds known to be highly toxic to cardiomyocytes. Interestingly, the 3D cardiac model demonstrated improved sensitivity for intermediate toxicity class. Both the 2D and 3D cardiac models demonstrated high specificity for compounds generally considered safe. While both 2D and 3D models demonstrated utility for highly toxic compounds, the 3D model demonstrated increased sensitivity for intermediate classed CiPA compounds. Accordingly, this study is consistent with previous in vitro toxicology results demonstrating the added value of 3D cellular models for safety assessment purposes.

P422

Historical Control Data for Phototoxicity Safety Assessments Performed in Compliance with ICH S10

Mathew Carlson ¹ , Christopher Cangelosi ¹ , Michelle Elliott ¹ , Kelli Smith ¹ , Joanne Bonner-Bowen ¹ , and Douglas Learn ²

¹ Charles River Laboratories, Horsham, Pennsylvania, United States

² Consultant, Doylestown, Pennsylvania, United States

ICH Guidance for Industry S10 Photosafety Evaluation of Pharmaceuticals recommends a stepwise approach in the evaluation of photosafety for therapeutic candidates. The ability to absorb light between the wavelengths of 290 and 700 nm, as well as a determination of the molar extinction coefficient (MEC), is performed in an ultraviolet-visible (UV-VIS) absorbance assay. Compounds that absorb light between 290 and 700 nm and with an MEC > 1000 L mol-1 cm-1 have photoreactive potential. A subsequent in vitro assessment in the 3T3 neutral red uptake assay using mouse fibroblast cells (OECD Test No. 432) determines the inherent cytotoxic potential of the compound in the absence of irradiation, as well as enhanced cytotoxicity following irradiation (phototoxic potential). A positive phototoxic result in the 3T3 assay requires either mitigation of risk for patients in the clinic or an in vivo phototoxicity study in animals that assesses responses following administration and irradiation. Compounds evaluated in the in vivo phototoxicity assessment that result in dermal responses, assessed using a modified Draize scale, and/or ocular responses, assessed through ophthalmologic examinations and microscopic examinations, are considered phototoxicants. We present the number of studies of each type that have been performed at the Testing Facility, the number of positive in vitro assays that have triggered in vivo assessments, and the number of in vivo assessments that either confirmed or dis-proved the preceding in vitro assessment, in turn demonstrating the sensitivity of the 3T3 assay to predict in vivo phototoxicity and the relevance of in vivo phototoxicity assessments.

P423

Development of Fit-for-Purpose In Silico Models for the Assessment of Extractables and Leachables

Kevin Cross ¹ , Arianna Bassan ² , Manuela Pavan ² , and Candice Johnson ¹

¹ Instem, Columbus, Ohio, United States

² Innovatune, Padova, Veneto, Italy

The fitness for purpose given a specific use context determines the acceptability of in silico models. To support the assessment of extractables and leachables (E&L), several endpoints are evaluated including mutagenicity, and skin sensitization. Leachable reactivity with a protein/DNA based active pharmaceutical ingredient (API) may also be relevant. In silico models are applied to the assessment of E&L genotoxicity in line with the ICH M7 procedures. This work first describes the maturation of the mutagenicity knowledgebase by refinement of expert alert rules and previous work on aromatic amines (a representative class of E&L) is used as an example. The development of a protocol to implement genotoxicity in silico evaluations is also described. We then present the incorporation of knowledge derived from proprietary data into models which predict potency categories for skin sensitization and demonstrate high sensitivity and accuracy of the high potent class (82%) towards 925 non-training structures which were experimentally assessed. Finally, we discuss the use of in silico approaches to investigate the potential interaction of leachables with API biomolecules within a rapid and conservative framework. Using models that lay their foundation in electrophilicity (e.g., direct acting mutagenicity alerts, direct peptide reactivity and reaction domain profiling), a consensus approach was shown to be fit for purpose (sensitivity of 90% and negative predictivity of 88%), when benchmarked against 22 experimentally determined leachable-insulin glargine reactivity values. The models described herein can be used to support E&L assessments, such as described in ISO-10993 regarding medical devices and to address emerging regulatory questions.

P424

A Balancing Act—Optimizing Efficacy vs. Safety with Better Tissue-Level PK/PD Data

Wen-Kai Hsiao ¹ , Thomas Birngruber ¹ , and Sylvia Furst ²

¹ Joanneum Research, Graz, Steiermark, Austria

² Certara, Durham, North Carolina, United States

Targeted drug delivery, such as topical soft drugs for dermatology indications, can significantly reduce systemic exposure hence toxicological liabilities. However, achieving the desired clinical efficacy while minimizing systemic toxicity can be challenging. In addition, using animal models to evaluate drug absorption can be costly and time-consuming, with uncertain translatability to human. We present a case study in which time-resolved, cutaneous pharmacokinetic (PK) data in human skin, obtained by the novel Open Flow Microperfusion (OFM) technology, was used to benchmark the PK performance of a topical Janus kinase inhibitor (JAKi) soft drug candidate against the reference drug product (Opzelura®, ruxolitinib cream 1.5%). An ex-vivo OFM study involving fresh (< 3 h) human skin explants was carried out. For 24 hours OFM probes implanted in the dermis were used to continuously sample interstitial fluid (ISF) after topical drug applications (drug candidate and Opzelura®). The ISF and skin biopsy samples (the latter collected at the end of the OFM sampling from the drug application sites) were analyzed for the concentrations of the parent/metabolite of the soft drug candidate, as well as ruxolitinib for Opzelura®. The data and the associated interpretation were decisive in determining the development priority for the drug candidate. Complemented by the Simcyp™ PBPK Simulator’s mechanistic dermal model which provides an in-silico method to assess drug absorption, data from the ex-vivo skin OFM model can be a powerful tool to determine the right dosing strategy for the IND-enabling studies without the need for expensive animal models.

P425

Specificity Testing of Trispecific Antibodies for Treating Multiple Myeloma Using the Membrane Proteome Array

Michael Phelan ¹ , Benjamin Doranz ¹ , Carmen Navia ¹ , Amelia Snyder ¹ , Rachel Fong ¹ , and Jonathan Sullivan ¹

¹ Integral Molecular, Philadelphia, Pennsylvania, United States

Assessment of off-target antibody reactivity is a regulatory requirement prior to clinical development; however, conventional screening methods are frequently ineffective in screening newer therapeutic modalities including multispecifics and CAR-T cell therapies. We developed the Membrane Proteome Array—containing 6,000 human proteins—as a comprehensive approach to identify off-target protein interactions. The MPA is a cell-based protein array that assesses binding interactions in native, unfixed cells, allowing for quantitative and high sensitivity detection using flow cytometry. The MPA has been used to successfully screen numerous emerging biotherapeutic modalities, including scFv, VH/VHH, and bispecific molecules, as well as other protein modalities. Recently, we used the MPA to evaluate the specificity of several multispecific antibody candidates against the G protein-coupled receptor GPRC5D, a promising target for multiple myeloma. Here we tested a panel of GPRC5D bispecifics using a CD3 targeting arm, as well as trispecific antibodies targeting GPRC5D, BCMA, and CD3. These T-cell engaging therapeutics require exquisite specificity, as even minimal cross-reactivity could result in non-target cell killing accompanied by serious or even life-threatening consequences. The MPA was successfully able to assess specificity of these molecules, showing for the first time compatibility with screening trispecifics. As a result, lead candidates devoid of cross-reactivity were identified for further development. We will also present case studies describing successful use of the MPA in regulatory IND filings for antibody-based therapies and discuss its ongoing development as an FDA-qualified Drug Development Tool.

P426

Spontaneous Convulsions and Tremors in Control Animals from Regulatory Toxicology Studies: A Multi-Site Retrospective Analysis Comparing Nonhuman Primates, Dogs, Minipigs, Rabbits, Rats, and Mice

Simon Authier ¹ , Raquel Lorenzo ¹ , Christian Li ¹ , Dominique Paquette ² , Mylene Pouliot ¹ , Kim Bujold ¹ , Karen Wong ¹ , and Raluca Kubaszky ³

¹ Charles River Laboratories, Laval, Quebec, Canada

² Centre Veterinaire DMV, Lachine, Quebec, Canada

³ BioCryst Pharmaceuticals, Inc., Blainville, Quebec, Canada

Convulsions are life threatening events that can be test article related or spontaneous background. The incidence of convulsions differs between species and the interpretation of toxicology and safety pharmacology studies relies on the availability of robust historical control data. This is paramount for low incidence observations such as convulsions. Tremors are commonly observed but can be considered as a premonitory sign to seizure. We aimed to characterize and compare the incidence of spontaneous convulsions and tremors in control non-human primates, dogs, minipigs, rabbits, rats and mice. A retrospective analysis was conducted with data from GLP facilities in North America and Europe including non-human primates (n=8805), dogs (n=24553, minipigs (n=2359), rabbits (n=21476), rats (n=312261) and mice (n=131272). For rats and mice, the incidence of convulsions was lowest at less than 6 weeks of age, was stable from 6 to 26 weeks of age and then increased progressively for older animals reaching 0.66% in rats and 0.60% in mice above 38 weeks. When comparing species, the incidence of spontaneous convulsion was lowest in minipigs (0%) followed by mice (0.03%), rats (0.06%), rabbits (0.07%), dogs (0.11%) and non-human primates (0.17%). While tremors can be monitored as a premonitory sign of seizures, the incidence of tremors in normal healthy animals is generally high and this observation is most often considered physiological. Consequently, other premonitory clinical signs such as myoclonus present higher specificity than tremors to predict test article related seizurogenic effects.

P427

Blood Transfusions in Cynomolgus and Rhesus Monkeys: A Retrospective Analysis Over a 20-Year Period

Pramila Singh ¹ , Karen Wong ² , Cristina Ballesteros ² , Raluca Kubaszky ³ , Christian Li ² , Dina Mayameei ² , Catherine Beaulieu ² , and Simon Authier ²

¹ Charles River Laboratories, Evreux, Haute-Normandie, France

² Charles River Laboratories, Laval, Quebec, Canada

³ BioCryst Pharmaceuticals, Inc., Blainville, Quebec, Canada

An increasing number of cell therapies require myelo-conditioning to increase cell engraftment. Antibiotics and blood transfusions are the main supportive care to mitigate pancytopenia following myelosuppression. Platelet (PLT) characteristics have a 10-day maturation period, a ∼4.5 day half-life, a lifespan of 8-11 days, and ∼60-75% PLT are present in circulation (with remaining PLT mostly in the spleen). Transfused PLT have shortened half-life (2.6 days). A retrospective analysis of blood transfusion data in non-human primates over a 20-year period at a single research facility was conducted. An important benefit from transfusion is a proportional increase in PLT, red blood cells (RBC) and hematocrit (HCT). The number of transfusions (6-13 per animal), the time between required transfusions (0-2 days between transfusions), and the elevation of the PLT, RBC and HCT vary, depending on the recipient’s condition and severity of the myelosuppression. With over 300 serological cross matched assessments performed, between donors and recipients, all evaluations resulted in cross matched compatibility. Our data suggest that donor to recipient cross match is not required for significantly immunosuppressed cynomolgus and rhesus monkeys. A blood transfusion volume of 7-14 mL/kg infused over 30-60 minutes was well tolerated in 99% of the animals. All blood transfusions that were not tolerated could be halted during the initial transfusion phase at a lower infusion rate, hence preventing complications and severe adverse reactions.

P428

Evaluation of (Q)SAR In Silico Mutagenicity Prediction Models Using a Proprietary Chemical Dataset

Teresa Wegesser ¹ , Ben Brooks ¹ , Aldo Coppi ² , and Emily de Koning ¹

¹ Amgen, Thousand Oaks, California, United States

² Amgen, Cambridge, Massachusetts, United States

It is widely recognized that the ICH M7 (R2) guidance is a framework for the assessment and control of mutagenic impurities in pharmaceuticals. This ICH guidance suggests using computational tools, such as in silico models, that can be used to evaluate potential mutagenicity effects without the need for in vitro or in vivo studies. The guidance recommends utilizing two (Q)SAR (Quantitative Structure Activity Relationship) models, one expert (rule-based) and one statistical-based, to collectively predict the outcome of a bacterial mutagenicity assay. There are various commercially available rule- and statistical-based in silico models that are widely employed in the pharmaceutical field; such models should be reviewed to make sure they comply with the OECD (Q)SAR Assessment Framework (OECD 386) prior to implementation. As these models are often trained and validated using publicly available datasets (> 10k compounds), further validation with proprietary chemical datasets should be employed. In this evaluation, a proprietary dataset of diverse synthetic chemicals, not included in the training of the models, was evaluated against three commercially available rule-based, and statistical-based models to assess their mutagenicity prediction performance. The results of this evaluation demonstrated that the models had > 70% sensitivity, specificity ranging from 35-65%, and accuracy ranging from 47-73%. Similar evaluations, using both public and proprietary chemical datasets, using commercially available in silico tools have reported sensitivity 30-72%, specificity 60-93%, and accuracy 60-88%. Regular evaluations of in silico models using proprietary datasets provides a valuable benchmark of performance metrics across tools and against bacterial mutagenicity data.

P429

Framework Approach for Priority Testing and Risk Assessment of N-Nitrosamine Degradants in Active Pharmaceutical Ingredients

Andrey Massarsky ¹ , Marisa Kreider ² , John Reichard ³ , and Michael Kovochich ⁴

¹ Stantec ChemRisk, Aliso Viejo, California, United States

² Stantec ChemRisk, Pittsburgh, Pennsylvania, United States

³ NASA Johnson Space Center; University of Cincinnati, Cincinnati, Ohio, United States

⁴ Stantec ChemRisk, Chicago, Illinois, United States

N-Nitrosamines are a class of organic compounds and source of potential mutagenic impurities in active pharmaceutical ingredients (APIs). The formation of one such impurity, N-nitrosodimethylamine (NDMA), has gained recent attention and can form from the manufacturing process as well as degradation processes. Predicting the potential formation of nitrosamines from degradation processes can aid in effective allocation of resources for further testing and risk assessment of contaminants. Therefore, a framework was developed, which utilized in silico tools (Zeneth, DEREK, and SARAH by Lhasa) to rank potential degradation pathways for additional testing and risk assessment. The framework includes assessment of degradants across a range of conditions (e.g., temperature, pH, presence/absence of water). Priority ranking for testing and risk assessment was established based on formation of degradants under different conditions, presence of NDMA or other N-nitrosamines, as well as mutagenic properties. Several APIs reported to contain NDMA as an impurity (e.g., valsartan) or as a degradant (e.g., afatinib) were evaluated. The framework displayed differences in degradant pathways and priority ranking for different APIs, such that degradants with predicted N-nitrosamine formation or other mutagenic properties ranked higher for further evaluation and risk assessment. It was concluded that the framework can be a helpful tool for priority testing and risk assessment of potential mutagenic degradants including N-nitrosamines.

P430

Mouse versus Rabbit: Time Course of Corneal Ocular Toxicity of an Antibody Drug Conjugate (ADC) with a Maytansinoid Payload

Shana Dalton ¹ , Olga Ab ² , Grace Lytle ² , Paul Miller ³ , Edward Stevens ⁴ , Ali Warbington ² , Cynthia Willard-Mack ⁵ , and Patrick Zweidler-McKay ²

¹ Labcorp Early Development Laboratories Inc., Madison, Wisconsin, United States

² Immunogen (now part of Abbvie), Waltham, Massachusetts, United States

³ Ocular Services On Demand (OSOD, LLC), Madison, Wisconsin, United States

⁴ Labcorp Early Development Laboratories Inc., Chantilly, Virginia, United States

⁵ Labcorp Early Development Laboratories Inc., East Millstone, New Jersey, United States

Microcystic-like epithelial changes (MECs) in the cornea (keratopathy) are a known ocular toxicity risk of antibody-drug conjugates with microtubulin-acting payloads. MECs in rabbits were previously evaluated using a single 12 mg/kg intravenous injection (IV) of M9346A-SPDB-DM4, an ADC consisting of a non-crossreactive humanized monoclonal antibody (M9346A) conjugated through a cleavable disulfide linker to a maytansinoid payload (DM4). Here we evaluate if mice are a useful preclinical model for predicting keratopathy and compare the onset, severity, and characterization of MECs in rabbits to mice. Female CD-1 mice were administered 12, 50, or 80 mg/kg M9346A-SPDB-DM4 once IV. For mice, corneal MECs were identified with a confocal microscope as early as Day 2 and through Day 4 at 12 mg/kg or Day 8 at ≥50 mg/kg, correlating with the same onset and findings in rabbits administered 12 mg/kg. Histopathologic findings noted in the cornea in rabbits and mice included: increased epithelial mitosis, single cell epithelial necrosis, epithelial atrophy, and/or erosion at the limbus region at the edge of the cornea. The kinetics of these histopathologic changes reflected the confocal microscopy MEC scores over the 8-day period. On Day 8, the microscopic corneal findings were decreased in severity in mice administered ≥50 mg/kg, indicating a trend for reversal not observed in rabbits at 12 mg/kg in a similar post-dose interval. Collectively, the results demonstrated that corneal MECs were detected by confocal microscopy and presented with similar histopathological changes in both species; however, reversal of the findings were earlier in mice than in rabbits.

P431

In Vitro Platelet Function Assessment of Oligonucleotides and Antibodies by Flow Cytometry during Drug Development

Carolyne Dumont ¹ , Kelly Hatch ¹ , Rachel Bourgeois ¹ , Anthony Pincon ¹ , Florence Poitout ¹ , and Marie-Soleil Piché ¹

¹ Charles River Laboratories, Senneville, Quebec, Canada

During new drug development, effects on platelets function can be a cause for concern for unwanted inhibition or enhancement of activation, resulting in risk of increased bleeding or thrombotic events. In the present study, a human whole blood flow cytometry based in-vitro platelets activation assay was used to determine the appropriate assay conditions (time course and concentrations) for the detection of potential off-target platelets activation using different biotherapeutics. Activation was measured using antibodies for surface expression of p-Selectin (CD62p) and active Glycoprotein IIb/IIIa complex (PAC-1) on platelets (CD61). Adenosine diphosphate, ADP (small molecule) was used as agonist, and different compounds, ODN 2395 (ASO), anti-CD226 (monoclonal antibody) and CD40 ligand (human recombinant protein) were tested as a proof of concept for the assay design. Furthermore, to evaluate the inhibition format of the assay, blood was treated with different concentrations of antagonistic molecules prior to stimulation with one of the relevant aforementioned molecules. The results confirmed that dose-dependent platelet activation could be detected as soon as 2 minutes (up to 60 minutes) post stimulation, with ADP and the different compounds. Furthermore, inhibition of the agonist induced activation could be achieved with the antagonists tested. This data confirmed that the in-vitro human whole in-vitro platelets activation flow cytometry assay was sensitive and suitable to screen for platelet activation or inhibition effects of drug candidates during the development phase.

P432

Leveraging Nonclinical Safety Evaluation Findings to Expedite Next-Generation Glucagon-Like Peptide-1 Receptor Agonists Development for Metabolic Disorders and Beyond

Yafei Chen ¹ , Tara Arndt ¹ , and Wendell Davis ¹

¹ Altasciences, Seattle, Washington, United States

Glucagon-like peptide-1 receptor agonists (GLP-1RA) represent an attractive class of medications for the management of type-2-diabetes and obesity, with 15 USFDA-approved GLP-1RA drugs from short-acting (exenatide/lixisenatide) to long-acting analogs (liraglutide/dulaglutide/semaglutide/tirzepatide). Several key differences exist between these products in terms of molecular structure, administration route, dosing frequency, and pharmacokinetics, and this likely led to discordant nonclinical toxicity profiles in early literature. We conducted a systemic and retrospective review of the USFDA CDER GLP-1RA toxicology documents to compile a database that includes each marketed GLP-1RA drug's adverse findings. By utilizing this regulatory information and knowledge, we aimed to identify potential class or specific adversity of novel GLP-1RA candidate drugs under early development at Altasciences. Although frequency/rates of toxicological effects differ between specific agents, noteworthy findings include 1) chronic pancreatic injury/pancreatic neoplasia in islet beta cells and/or exocrine duct cells in various animal models; 2) thyroid C-cell hyperplasia and neoplasia in 2-year rodent carcinogenicity studies. In contrast, the most common adverse effects in humans are gastrointestinal-related symptoms and injection site reactions. There is no direct evidence of thyroid carcinoma in humans due to the lower density of GLP-1 receptors on C-cells compared to rodent models. With considerable heterogeneity/complexity across the class of GLP-1RA, each candidate agent should be evaluated independently, as opposed to assuming a class effect. As an increasing number of newer-generation GLP-1 RA-based therapies are being developed for chronic metabolic, cardiovascular, and neurodegenerative disorders, this review provides insights and considerations that can help accelerate the development of next-generation therapeutic GLP-1RAs.

P433

Influence of Electrocardiogram and Hemodynamic Recording Methods in Nonrodent Toxicology Studies on Statistical and Pharmacological Sensitivity

Jean-Pierre Valentin ¹ , Annie Delaunois ² , Michael Foley ³ , Kim Henderson ⁴ , Emma Pawluk ⁵ , Sridharan Rajahmani ⁶ , Christopher Regan ⁷ , Rebecca Ross ⁸ , Stephen Tichenor ⁹ , and Todd Wisialowski ¹⁰

¹ UCB Biopharma, Braine l'Alleud, Brabant Wallon, Belgium

² UCB Biopharma, Braine l'Alleud, Brabant Wallon, Belgium

³ Abbvie, North Chicago, Illinois, United States

⁴ Amgen, Thousand Oaks, California, United States

⁵ UCB Biopharma SRL, Braine-l'Alleud, Brussels Hoofdstedelijk Gewest, Belgium

⁶ Johnson & Johnson, San Francisco, California, United States

⁷ Merck, West Point, Pennsylvania, United States

⁸ Labcorp, Madison, Wisconsin, United States

⁹ Charles River Laboratories, Reno, Nevada, United States

¹⁰ Pfizer, Groton, Connecticut, United States

Cardiovascular (CV) parameters [e.g., blood pressure (BP), electrocardiogram (ECG), heart rate (HR)] are recorded in non-rodent toxicology studies to support drug development. However, the quality of the measurements varies depending on the use of restraint or telemetry (implanted or jacketed)-based methods. The present analysis assessed the influence of recording method on the statistical and pharmacological sensitivities to detect BP-HR-ECG effects. Baseline BP-HR-ECG values were sourced from 5 sponsors and 2 CROs from studies performed in dogs or non-human primates (NHP) with treatment durations up to 52 weeks. A survey was conducted to ascertain current statistical analysis practices for CV parameters in those studies. Lastly, literature-based and proprietary unpublished examples were used to establish baseline values and determine the sensitivity of different methodologies to detect drug-induced CV effects. The de novo analysis and literature-based search showed that baseline BP/HR values were highly variable with consistently higher means under restraint versus telemetry methods. The root mean square errors for BP/HR were larger under restrained conditions, in both species. Under restrained conditions, the use of fixed formulae for HR-corrected QT resulted in inconsistent QTc values across sponsor/CROs. The survey showed that statistical analysis of ECG/BP data was infrequently performed under restrained conditions contrasting with telemetry-based methods. Literature and proprietary examples showed that drug mediated BP elevation or QTc prolongation were detected in the clinic and in NHP or dog via telemetry, but not under restrained conditions. The data illustrate that animal restraint reduces the pharmacological and statistical sensitivities to detect CV effects.

P434

To Seize or Not to Seize

Ashley McCord ¹ , and Marc Bailie ¹

¹ Certara, Ann Arbor, Michigan, United States

Convulsions are a major human safety concern, especially for brain-penetrant therapies. Therefore, convulsions in animal studies present a significant challenge in development. Sometimes convulsions indicate seizure activity; however, convulsions also occur in animals in the absence of seizure activity. If convulsions are observed in a study, it is critical to determine if they are indicative of seizures and if they are drug-related which may require additional studies. In this poster, two cases involving convulsions are described. In the first case, a single dog in a 28-day study exhibited convulsions which resulted in the program being put on clinical hold. Additional data, including external electroencephalography monitoring in a subsequent study, contributed to removal of the clinical hold. In the second case, several dogs in a dose range-finding study were noted with non-sustained convulsions. After further review, it was determined that these dogs were shivering/trembling. Data from a cardiovascular telemetry study showing decreased body temperature at the time when shivering/trembling occurred, aided in removing the clinical hold. The impact of these observations on development are described as are the steps that were taken to determine 1) if the convulsions were indicative of seizures and 2) whether the convulsions were drug-related. These cases demonstrate that if convulsions are observed in nonclinical studies, and the risk to humans is not adequately characterized, this can result in a clinical hold. Additionally, these cases demonstrate how to manage unexpected convulsions in animal studies and the type of data that might contribute to removing a clinical hold.

P435

Intracerebroventricular (ICV) Route in Mice for Administration of Gene Therapy Products

Julie Forget ¹ , George De Los Santos ¹ , Norbert Makori ¹ , Serah Tanno ¹ , and Adeline Varnet ¹

¹ Altasciences Preclinical Seattle, Everett, Washington, United States

Gene therapy (GT) has emerged as a promising approach for treating a wide array of genetic disorders and diseases. One of the innovative delivery methods in GT is the intracerebroventricular (ICV) route, which involves administering therapeutic agents directly into the cerebrospinal fluid (CSF) within the brain's ventricles. This method provides a unique opportunity to target the central nervous system (CNS) efficiently, especially for conditions that affect the brain and spinal cord. This direct injection approach can overcome the challenges posed by the blood-brain barrier (BBB), which often limits the effectiveness of systemic delivery methods. Since intrathecal injection in mice is not practical, ICV is the more widely used method of choice. While under anesthesia, the mouse is secured in a stereotaxic frame, the bregma (reference point on the skull) is located, and stereotaxic equipment is used to inject the therapeutic agent (up to 10 uL) at a rate of approximately 1 uL/second. To confirm suitable delivery and potential effects on behavior post administration, functional observational battery (FOB) was performed at 3- and 24- hours post dose to assess animal's behavior, sensorimotor functions, and physiological responses. FOB assessments indicated that the selected pain management post injection was the main contributor for the abnormal behavior observed, when compared to a positive control (pain management only). By overcoming the challenges posed by the BBB, this approach holds great potential for treating a variety of neurodegenerative diseases and genetic disorders, offering hope for improved outcomes and quality of life for patients.

P436

Nonclinical Toxicology Profile of ALG-055009, a Novel and Potent Thyroid Hormone Receptor β Agonist, for the Treatment of Metabolic Dysfunction-Associated Steatohepatitis (MASH)

Dinah Misner ¹ , Lawrence Blatt ¹ , Sushmita Chanda ¹ , Doug Clark ¹ , Kha Le ¹ , Tse-I Lin ¹ , Xuan Luong ¹ , David Smith ¹ , Julian Symons ¹ , and Kusum Gupta ¹

¹ Aligos Therapeutics, South San Francisco, California, United States

MASH is characterized by hepatic inflammation/damage as a reaction to build-up of fat in the liver. Recently, RezdiffraTM (resmetirom) was approved to treat MASH, demonstrating the activity of thyroid hormone receptor β (THR-β) agonists in MASH. Additional THR-β agonists have demonstrated reduced liver fat, restored liver function and reversed inflammation/fibrosis in clinical trials. Here we present the preclinical development of ALG-055009, a second-generation THRβ agonist with improved potency and favorable selectivity. In vivo pharmacokinetic studies were conducted in mice, rats, rabbits, dogs, and non-human primates. Repeat-dose toxicology studies were conducted in rats and dogs using once daily (QD) oral gavage administration, up to 6 months in rats and 9 months in dogs. Clinical pathology endpoints including thyroid hormones were assessed at 6-, 13-, 26-, and 39-weeks (dogs only), as well as following 4-weeks of recovery. Pharmacodynamic endpoints included total/LDL cholesterol, liver enzymes, and thyroid hormones. ALG-055009 showed favorable pharmacokinetic properties across species following single or repeat doses. ALG-055009 was well tolerated in both rats and dogs in repeat dose toxicology studies up to the highest doses tested. Dose-dependent, reversible changes in lipid parameters were observed in repeat-dose toxicology studies, whereas changes in total circulating thyroid hormones levels were observed at supratherapeutic exposures. ALG-055009 was generally well-tolerated in developmental and reproductive toxicity studies in rats and rabbits. Taken together, ALG-055009 is a potent and selective THR-β agonist with favorable in vitro safety and ADME properties and repeat-dose toxicity profile in rats and dogs.

P437

Determining a Class-Specific Parenteral Acceptable Daily Limit for Sugars

Ernest Fung ¹ , Courtney Callis ² , Lisa Cooper ² , Jordan Kozal ³ , Lisa Yang ⁴ , Andrew Maier ⁵ , and Robert Jolly ²

¹ Integral Consulting, Irvine, California, United States

² Eli Lilly & Company, Indianapolis, Indiana, United States

³ Integral Consulting, San Francisco, California, United States

⁴ Stantec, San Francisco, California, United States

⁵ Integral Consulting, Cincinnati, Ohio, United States

Sugars are often identified impurities in biological therapeutics. While sugars are endogenous and dietary, they aren’t therapeutics and lack clinical/nonclinical data. While dietary intake values are available, they don’t inform on parenteral safety. Together, parenteral acceptable daily intakes (ADIs) for sugars are not established. The goal of this study is to establish a parenteral ADI for sugars where compound-specific information isn’t available. Parenteral ADIs were determined for three monosaccharides (glucose, fructose, galactose) and three disaccharides (sucrose, trehalose, lactose). Generally, overt toxicity was not identified, and clinical/nutritional information was used as points of departure (POD). However, adversity was noted for galactose and was used as the POD. Parenteral ADIs for monosaccharides varied (280-1000 mg/day), whereas parenteral ADIs for disaccharides varied less (375-540 mg/day). Importantly, ADIs of disaccharides are in alignment with their monosaccharide components on a mass basis. For example, the ADI for lactose (galactose and glucose) is 375 mg/day; this roughly coincides with the ADIs of glucose (1000 mg/day) and galactose (300 mg/day). The ADIs determined are well below and added only a small fraction to dietary intake. Together, a preliminary parenteral ADI of 250 mg/day is proposed for sugars where compound-specific information isn’t available. The proposed ADI protective of all sugars is evaluated in this study. If possible, compound-specific parenteral ADIs should be determined. Importantly, only evaluated monosaccharides and disaccharides were evaluated; thus, the proposed ADI may not be applicable to polysaccharides or sugar derivatives. Additional sugars must be evaluated to validate and further refine the proposed ADI.

P438

Inhaled Delivery of a Novel Broad Spectrum Immune Modulator to Minipigs Produces a Favorable Toxicity Profile

Hayley Myles ¹ , David Coleman ² , Petr Sova ³ , Emily Resseguie ⁴ , and Roland Franke ³

¹ Labcorp Early Development Laboratories Ltd, Huntingdon, England, United Kingdom

² Labcorp, Huntingdon, England, United Kingdom

³ OCI Biotech Limited, Wilmington, Delaware, United States

⁴ Labcorp, Somerset, New Jersey, United States

At the start of the COVID-19 pandemic there was a call for promising therapies to be fast tracked. OCI Biotech Limited were developing a small molecule therapy for acute lung inflammation which had potential use as a therapy against the symptoms of COVID-19. The molecule, OCI-B2, was intended for inhalation administration for delivery directly to the site of action - the lung. This poster describes the nonclinical package of studies with particular focus on use of the minipig as the non-rodent species. In vitro hepatocyte metabolic profiling was used for species selection. Safety pharmacology was evaluated in separate studies in rat (CNS, respiratory) and minipig (cardiovascular) and there were no OCI-B2-related findings. GLP-compliant 28-day repeat-dose inhalation toxicity studies with toxicokinetics show the maximum feasible dose of 12.7 mg/kg/day was the No Observed Adverse Effect Level (NOAEL) for minipig and a dose of 54 mg/kg/day was the NOAEL for rat. The mass median aerodynamic diameter of the OCI-B2 aerosol was 2.0-3.4 microns in the minipig system and 3.7-3.9 microns in the rat system. These dose levels provided adequate safety margins based on systemic and local doses to support a Phase 1 clinical trial.

P439

A Novel Approach Method Combining GFP Reporter Cell Lines with Error Corrected Next Generation Sequencing (ecNGS) for In Vitro Genotoxicity Assessment of N-Nitrosamines

Daniel Roberts ¹ , Giel Hendriks ² , Amy Gernaat ³ , Nynke Moelijker ³ , Marit Hoogenboom-Geijer ³ , Inger Brandsma ³ , Sam Sharifi ⁴ , Christopher Bradley ⁴ , and Alexander Maslov ⁴

¹ Toxys, Canandaigua, New York, United States

² Toxys, Oegstgeest, Zuid-Holland, Netherlands

³ Toxys, BV, Oegstgeest, Netherlands

⁴ Mutagentech, Bronx, New York, United States

N-Nitrosamines (NAs) are considered probable human carcinogens with ubiquitous exposure from food, water, cosmetics, pharmaceuticals and tobacco products. ToxTracker is a GFP reporter assay that accurately predicts genotoxicity and discriminate between DNA-reactive and secondary genotoxic mechanisms, including aneugenicity and oxidative stress. NAs require phase 1 metabolization to form reactive intermediates, which is limited when using standard exogenous metabolic activation. Hence, we optimized a hamster S9-based preincubation protocol to assess the genotoxic potency of 8 different NAs. All tested NAs were classified as genotoxic in ToxTracker with a clastogenic MoA. The smaller NAs did induce DNA strand breaks but no detectable inhibition of DNA replication, likely related to the size of the induced DNA lesions. Next, we investigated if NA-induced DNA damage leads to induction of gene mutations. The genotoxic MoA assessment in ToxTracker was combined with Single-Molecule Mutation sequencing (SMM-seq). The small NA dimethylnitrosamine induced a 5-fold increase in single nucleotide substitutions, primarily GC>AT mutations arising from O6-methylguanine lesions. Also exposure to the larger NA nitrosofluoxitine resulted in a strong increase in gene mutations. Six other genotoxic substances with a clastogenic MoA in ToxTracker, ENU, MMS, B[a]P, AFB1, CisPt and KBrO3, induced 3-15 fold increases in gene mutations by SMM-seq. The obtained mutational spectra aligned with expected reactivity and mutagenic MoA of these compounds. Overall, this pilot study provides evidence that combining ToxTracker reporter cell lines with ecNGS can accurately predict primary genotoxic potential and provide deep insight into genotoxic MOA.

P440

Characterization of a CNS Liability Observed in Dogs after Administration of a LPAR1 Antagonist

Ellen McGlinchey ¹ , Kevin Holsapple ¹ , Monica Metea ² , Anna Zagorska ¹ , Yiding Hu ¹ , and Jim Hartke ¹

¹ Gilead Sciences, Inc., Foster City, California, United States

² Preclinical Electrophysiology Consulting, Mattapoisett, Massachusetts, United States

GS-2278 is a selective lysophosphatidic acid receptor 1 (LPAR1) antagonist that was being developed for the treatment of nonalcoholic steatohepatitis (NASH) and idiopathic pulmonary fibrosis (IPF). GS-2278 was preclinically evaluated for efficacy and safety. Central nervous system (CNS) toxicity, characterized by convulsions and other CNS-related clinical signs, was observed in dogs at high exposures after a single oral dose of GS-2278 in the 7-day repeat dose toxicology study. Further studies to characterize the CNS toxicity, with the goal to enable further development, included dedicated electroencephalography (EEG) studies and 4-week FIH-enabling toxicology studies with safety pharmacology endpoints. In the EEG studies in dogs, seizures were confirmed at dose levels lower than previously observed by clinical signs. The CNS-related toxicity was observed at exposures margins below the recommended 10-fold safety margin over the predicted clinical efficacious dose. Additionally, the variability in the exposures did not provide a clear threshold for the CNS effect. The FIH-enabling studies in dogs confirmed the CNS liability as the dose-limiting toxicity for GS-2278. Overall, the weight of evidence and an undetermined mechanism for the CNS toxicity resulted in termination of GS-2278 development.

Toxicology Methods

P501

Enhancing Drug Safety with Organoids

Sylvia Boj ¹ , Farzin Pourfarzad ¹ , Merel Derksen ¹ , Alessandra Merenda ¹ , Francisco Morales Rodriguez ¹ , and Robert Vries ¹

¹ HUB Organoids, Utrecht, Netherlands

Selecting the optimal in vivo model for toxicology preclinical studies and making informed decisions regarding dose escalation in clinical trials has been a formidable challenge. This challenge arises from inter-species variability, limited human relevance, cost constraints, and lengthy timelines. The FDA Modernization Act 2.0 has introduced a transformative avenue by endorsing alternative in vitro models. In this context, organoids, self-organized and self-renew in vitro models derived from primary tissues have emerged as invaluable tools. With remarkable accuracy in recapitulating tissue physiology, organoids are amenable to high-throughput screening and offer a promising platform for assessing drug toxicity. Adult stem cell-derived organoids were established from intestinal epithelium from human and preclinical animal models, such as dogs, rats, and minipigs. Cell viability assays were developed to assess compound toxicity. Compound cytotoxicity in rat and dog organoids significantly correlated with in vivo preclinical toxicity. When extending the assessment of compound toxicity, we confirmed species-specific sensitivity differences between human, rat, minipig and dog organoids. Furthermore, the determination of IC50/Cmax demonstrated that organoids effectively identify inter-species and regional gastrointestinal tract toxicity profiles. Our results underscore the value of human and preclinical animal organoids as in vitro platforms for comprehensively investigating gastrointestinal drug toxicity endpoints. The unique ability to utilize organoids from diverse species, organs, and segments holds immense promise for enabling cross-species and regional comparisons, shedding new light on the selection of the most appropriate animal model for preclinical in vivo studies and aiding in defining the starting point for dose escalation in clinical trials.

P502

An 8-Week Repeat-Dose Intravenous Infusion Study in CD-1 Mice Using Peripherally Placed (Tail Vein) Pediatric Catheters

JoAnna Bultman ¹ , Teresa Gleason ¹ , Brad Buczynski ¹ , Jason Flavin ² , and Jeff Foy ²

¹ Charles River, Ashland, Ohio, United States

² PepGen, Boston, Massachusetts, United States

Intravenous (IV) infusion in mice can be a technically difficult procedure, often requiring surgical implantation of jugular catheters. In our experience, this model shows approximately 20% catheter dislodgement or lost patency at around 8 weeks, limiting the utility of this model to studies of short duration and leaving an unmet need within nonclinical safety assessment. To address this limitation, we conducted a feasibility study to determine if using a 26-gauge 1/2” over the needle pediatric IV catheter was a viable alternative to the standard surgical mouse model of IV infusion. We successfully dosed 5/6 mice once weekly for 7 weeks and considered a once monthly regimen to be achievable. We followed-up with an 8-week toxicity study using the pediatric catheter, which included 48 Main Study and 168 Toxicokinetic Study male and female mice (1 control and 3 treated groups). Mice were appropriately restrained and dosed once every 4 weeks (3 doses) at an infusion rate of 50 mL/kg/hour for 30 minutes. Study endpoints included assessment of mortality, clinical observations, body weights, clinical pathology, toxicokinetics, organ weights, and macroscopic and microscopic examinations. No procedure-related mortality occurred in the Main Study animals; however, there were 3 unscheduled deaths among the Toxicokinetic Study animals that were considered potentially procedure-related. There were no procedure-related effects on any of the other endpoints evaluated. Therefore, we successfully demonstrate that using the pediatric catheter is a viable and durable alternative to surgical mouse models of IV infusion in studies lasting up to 8 weeks in duration.

P503

Withdrawn

P504

Advancing Noninvasive Electroencephalogram (EEG) Seizure Detection Methods Using Göttingen Minipigs

Emily Fitch ¹ , Hadasha Cortes-Luciano ¹ , Kharizmara Kuamua ¹ , Jeffrey Langston ¹ , Mark Moffett ¹ , Todd Myers ¹ , Kevin Nusser ¹ , Subramaniam Uthayathas ¹ , Camille Walker ¹ , and Katherine Wesley ¹

¹ USAMRICD, Edgewood, Maryland, United States

Seizures induced by chemicals (such as the nerve agent soman) remain a significant threat to our civilian and military personnel, and improved models for seizure analysis and treatment are desired. Although rodents and primates have historically been used for such studies, miniature swine (such as the Göttingen minipig) are increasing in popularity because they are more readily available and affordable than non-human primates, and also closely resemble human anatomy and physiology (including a gyrencephalic brain and larger body mass). Implanted telemetric recording of EEG signals through skull screw electrodes is a reliable technology for seizure recording, but requires expensive transmitters be placed through invasive surgical methods. The use of non-invasive EEG recording via scalp electrodes (as is common in human EEG recording), would preclude the need for surgery and prevent the need for lengthy surgical recovery (often two weeks or more), enhancing throughput and reducing infection risks inherent to surgical procedures. Here, we elaborate and compare two different non-invasive EEG systems and discuss the pros and cons of each. Detailed baseline, seizure onset, status epilepticus, and post-seizure periods were analyzed to assess signal strength, quality, movement artifact, and general usability. Göttingen minipigs tolerated the scalp electrodes and both methods provided satisfactory EEG recording to assess seizure onset, severity, and cessation. Non-invasive EEG proved comparable to a standard implanted telemetric method. It is our hope that non-invasive approaches such as those detailed here could be adopted more widely because they are less expensive, less invasive, and more humane.

P505

African Green Monkeys (AGM) Quarantine Colony Census: Antiparasitic Treatment and Husbandry

M.S. Lawrence ¹ , Sven Korte ¹ , Katalina Cruz Cruz ² , Murti Amanda ¹ , Calvin Carpenter ¹ , L Dzikiti ¹ , and Quincy Edwards ¹

¹ Virscio, New Haven, Connecticut, United States

² Virscio, Bassettere, Saint Kitts and Nevis

Nonhuman primates, like humans, are natural hosts to a range of enteric parasites in the setting of omnivorous foraging in the absence of veterinary management. We evaluated responsiveness of this natural state to therapeutic intervention in St. Kitts origin African green monkeys (AGM; Chlorocebus sabaeus) by conducting fecal egg counts (FECs) using the double centrifugation technique (5 min, 500 g) with Sheather’s sugar solution (spg 1.28), pre and post anthelmintic and husbandry intervention over 2 years (n=760 AGMs). At baseline 100% of adult AGMs had evidence of nematode infections. When treated with ivermectin (300 µg/kg subcutaneously), albendazole (10 mg/kg orally daily for 3 days), and Praziquantel (5 mg/kg IM at intake and before 4 weeks) and maintained in individual or pair housed cages with cleaning and feeding regimens to prevent reinfections, 75.4% of the AGMs were negative at a follow-up FEC at 1-2 months post-treatment. Capillaria sp (8%) was most prevalent, followed by strongylid species (6.9%) specifically (hookworm), and Trichuris trichiura (3.8%), but also Strongyloides fuelleborni (2.7%). No trematode or cestode infections were identified. Post treatment positive animals were retreated with the appropriate anthelminthic regimen with a follow-up analysis 7-14 days post retreatment to confirm clearance, which was achieved in 96.9% of animals in follow-up. The ivermectin, albendazole, and praziquantel treatment regimen, coupled with good husbandry practices to prevent reinfections, effectively controls nematode infections in AGMs.

P506

Comparison of Intratracheal and Inhaled AAV Pulmonary Deposition in Two Species

Emily Resseguie ¹ , Phillip Gouldthorpe ¹ , Justine Damiano ¹ , Simon Moore ² , Mayuri Prasad ³ , Ambre Brink ³ , and Brian McIntosh ³

¹ Labcorp, Somerset, New Jersey, United States

² Labcorp, Huntingdon, England, United Kingdom

³ Labcorp, Madison, Wisconsin, United States

Cell and gene therapies offer an attractive approach to treating monogenic pulmonary diseases. Standard nonclinical pharmaceutical inhalation exposure systems require greater volumes of test article than other parenteral routes of administration because multiple animals are simultaneously dosed on a single exposure system with innate inefficiencies. Because clinical treatment with these modalities will likely only require single or infrequent dose administration, intratracheal (IT) administration offers an alternative route for pulmonary delivery; however, IT dosing requires anesthesia. Therefore, customized administration techniques for IT administration to cynomolgus monkey and nose-only inhalation to mice were set up for delivery of an adeno-associated virus (AAV). For the monkey, achieved aerosol concentrations from nebulized AAV for the IT delivery method were compared to theoretical values from oronasal (facemask) delivery with the same nebulizer. Assuming 100% lung deposition from IT and 25% from facemask delivery, IT dosing provides a 1.3-log greater lung delivery and requires 0.9-log less AAV for an equivalent lung dose as oronasal inhalation for 3 animals. For the nose-only inhalation in mouse, calculated achieved deposited doses were compared to theoretical IT values (assuming 10% and 100% lung deposition, respectively). Inhalation dosing requires only 0.7-log more AAV than IT dosing for an equivalent lung dose in the same number of animals. Because inhalation dosing requires more AAV for both species compared to IT dosing, logistics (such as group size) and welfare considerations should also be considered for selecting the best route of viral gene therapy delivery to lungs.

P507

Sinclair Nanopigs as an Alternative Model for Implanted Cardiovascular Telemetry Data Assessment

Rommel Interino Matheson ¹ , Alexander-Yap Sia ¹ , Frédérick Thouin ¹ , Sherwin Silo ¹ , Joseph Younan ¹ , Bassem Attalla ¹ , Djamal Bennamane ¹ , and Rocky DiFruscia ¹

¹ITR Laboratories Canada Inc., Baie D'Urfe, Quebec, Canada

Historically, non-human primates (NHP), dogs and pigs have been mainly used to conduct large-scale cardiovascular telemetry studies. As an alternative model to the more traditional dogs, NHPs and minipigs, the Sinclair Nanopig offers several advantages in this context. Its compact size and physiological similarities to humans make it a valuable model for preclinical cardiovascular telemetry assessments. In order to demonstrate that cardiovascular telemetry data could be acquired, Sinclair Nanopigs were surgically implanted with DSI PhysioTel Digital L11 transmitters and were administered negative and positive control items (L-name, Medetomidine and Sotalol). The animals were then monitored for changes in arterial blood pressure, pulse rates, electrocardiograms (heart rates, RR, PR, QT, QTcV intervals, QRS complex durations), body temperature and locomotor activity. The readings were then assessed quantitatively and qualitatively for changes indicative of cardiac abnormalities. Telemetry data was continuously recorded from each animal during each treatment session for all parameters detailed above from 2 hours prior to dosing until 24 hours post dosing. The data obtained showed expected effects of the positive control items, in monitored parameters. L-Name resulted in hypertension along with a decrease in body temperature. Sotalol and Medetomidine produced bradycardia, whereas Sotalol resulted also in a prolongation of the QTcV interval. The changes above confirm the effectiveness and the suitability of Sinclair Nanopigs as a viable non rodent species for preclinical cardiovascular telemetry assessments.

P508

Validation of the Flow-Cytometric Peripheral Blood Micronucleus Assay in Mouse

Mingzhu Kang ¹ , Ning Xu ¹ , Lingling Cao ¹ , Xiang Shan ¹ , Liwen Gao ¹ , and Millie Chen ¹

¹ WuXi AppTec, Suzhou, Jiangsu, China (People’s Republic)

Traditionally, the mammalian erythrocyte micronucleus assay employs bone marrow collection and manual scoring for micronucleus frequency detection, which is labor-costing and time-consuming. The flow cytometry (FCM) method, which owns the advantages of time-efficient and data robustness and consistency, was also adopted by ICH-S2(R1) guideline and OECD Testing Guideline 474. This abstract presents the in-house validation results of a peripheral blood micronucleus assay in mouse, using FCM based method and analyzed under Good Laboratory Practice (GLP) conditions. A routine three daily administration was determined, including negative control (0 mg/kg/day), hydroquinone (20, 40 and 80 mg/kg/day), positive control article cyclophosphamide monohydrate was administrated once at 75 mg/kg, samples were collected at 20-24 hours post last dosing and processed following Litron’s commercial analysis kit manual. Percentage of reticulocyte among total erythrocytes (%RET) and micronucleus frequency (%MN-RET) were evaluated per the evaluation criteria. Several validation parameters were evaluated: 1) The consistency of positive and negative control over 2 days was within or comparable with the kit reference data range; 2) carryover effect of negative control over positive samples was less than 5%; 3) intra-day and inter-day precision verification, the coefficient of variation (%CV) of both MN-RET% and %RET were both less than 20%; 4) stability in fixative solution (1 month) or Long-term storage solution (LTSS, 1 month) was evaluated, the mean difference of both MN-RET% and %RET were all within 20%. All validation specifications met the acceptance criteria. These solid data had proved WuXi’s capability in tracking and developing methodology for better regulatory compliance.

P509

Assessment of Leading Regulatory T Cell Immunophenotyping Methods using Flow Cytometry

Michael Shea ¹ , Amanda Lucchini ² , and James Munday ³

¹ Labcorp Early Development Laboratories Inc, Madison, Wisconsin, United States

² Labcorp, Madison, Wisconsin, United States

³ Labcorp Early Development Laboratories Ltd, Harrogate, England, United Kingdom

Regulatory T cells (Tregs) are widely recognized as a critical component of the immune system and an increasing number of studies are identifying new roles these cells play in immune tolerance and the success of clinical immunotherapy treatment. Due to the importance of accurately measuring Tregs in the modern drug development landscape, greater scrutiny is being placed on the methods used to identify them. Immunophenotyping via flow cytometry is a widely recognized technique for analyzing Tregs in whole blood from both preclinical and clinical samples, however the preferred set of markers for identifying Tregs is debated. Here, our laboratory examined three widely used Treg phenotypes (CD25+FoxP3+, CD25+CD127-, and CD25+FoxP3+CD127-) in parallel using whole blood from up to 32 Cynomolgus Macaques, a common non-human primate model in toxicological safety assessment studies. Our results show that Treg phenotypes using either FoxP3 or CD127 represent similar proportions of CD4+ T cells (3.15% and 2.85% respectively). Furthermore, when labelled with both markers, greater than 95% of Tregs share both phenotypes leading us to conclude that all methodologies tested identify equivalent Treg cell populations. Comparing the methods, FoxP3 is a nuclear transcription factor requiring cell permeabilization to be properly detected in flow cytometry experiments; whereas CD127 is a surface marker and does not require permeabilization. In conclusion, we recommend laboratories utilize the CD25+CD127- Treg phenotype due to its efficient assay design and the similarity of performance of each method analyzed.

P510

Retrospective Immunophenotyping Analysis of Peripheral Blood Samples from Preclinical Species

Kristina Schmunk ¹ , Bhavana Bhatt ¹ , Lei Zhu ¹ , Francisco Sanchez-Conde ¹ , Emma Anziano ¹ , Nianyu (Jason) Li ¹ , and Raymond Gonzalez ¹

¹ Merck & Co., Rahway, New Jersey, United States

Effects of drugs on immune cells is commonly studied in clinical and nonclinical studies by immunophenotyping analysis. Measured by flow cytometry, immunophenotyping analysis allows for identification and quantification of distinct immune cell subsets at a single-cell level. Although immunophenotyping allows rapid profiling of immune cells, the requirement for fresh samples limits retrospective analysis of previously stored samples. The purpose of this study was to enable retrospective immunophenotyping analysis of cryopreserved samples. A peripheral blood-based cryopreservation method was developed that allowed for long-term storage of both nonhuman primate and rat samples suitable for retrospective flow cytometry analysis. Peripheral blood samples were cryopreserved in either 10% DMSO in FBS or CryoStor® CS10 and stored in liquid nitrogen. Samples were thawed and immunophenotyping analysis was performed up to 10 months post collection. Total and subsets of T (CD4, CD8, naïve, memory, and regulatory T cells), B (naïve and memory), and NK cells were analyzed by flow cytometry. Our result showed absolute count of lymphocyte subsets from samples stored up to 10 months had comparable changes (<20%) in the absolute count and surface protein expression level to those stored overnight at room temperature. Overall, we identified cryopreservation procedures that enabled a retrospective immunophenotyping analysis of nonhuman primate (rhesus and cynomolgus macaques) and rat peripheral blood samples. This will benefit nonclinical studies when retrospective immunophenotyping is needed such as facilitating sample shipment across sites and allow for the ability to analyze samples from different timepoints concurrently, potentially minimizing interday variability.

P511

Evaluation of NHP Exposure Systems—Pharmacokinetic, Respiratory, and Behavioral Quantification between Head Dome and Face Mask Exposures in NHPs

Philip Kuehl ¹ , Brent Barre ¹ , Edward Barrett ¹ , Jabari Farrar ¹ , Brandon Hillhouse ¹ , Jacob Jantzi ¹ , Conor Ruzycki ¹ , and Julie Strickland ¹

¹ Lovelace Biomedical, Albuquerque, New Mexico, United States

Inhalation exposures in NHPs are often a required portion of non-clinical drug development and can include pharmacokinetic, pharmacology, and toxicology studies. For NHP exposures conducted in alert animals, animals are preconditioned to mechanical restraints, and aerosols are administered via engineered delivery systems using either face masks or head domes. These aerosol delivery systems are designed to mitigate back pressure and facilitate tidal breathing, with flows balanced to support minute volume needs of all animals on-system. To compare and contrast key parameters between face mask and head domes (exposure, animal behaviour and stress) a pilot study was conducted in N = 4 NHPs using budesonide as a model compound, with a cross-over design specifically comparing head dome, face mask, and IV exposure routes. Key endpoints included clinical observations, pharmacokinetics, and standard respiratory parameters (measured via external RIP system). Blood samples were collected at 8 timepoints out to 6 hours post exposure to compare PK (Cmax, Tmax and AUC). RIP data showed that breathing rate, tidal volume and respiratory minute volume were unchanged between exposure methods for all animals. Hudson MicroMist jet nebulizers were used to aerosolize a budesonide suspension for inhalation into a four animal NHP exposure system. The total aerosol concentration was ∼ 0.15 mg/L and budesonide aerosol concentration was ∼ 2.4 µg/L, with particle size in the 1.3 µm MMAD range (1.5 GSD). Estimated pulmonary deposited dose (25% DF) was 7 µg/kg.

P512

A Sensitive and Specific Human Primary Stem Cell-Based Assay for Predicting Diarrheagenic Potential of Therapeutic Agents

Colleen Pike ¹ , James Levi ¹ , Swetha Peddibhotla ¹ , Lauren Boone ¹ , Reganne Lorichon ¹ , Mariana Castillo ¹ , William Thelin ¹ , Maureen Bunger ¹ , and Liz Boazak ¹

¹ Altis Biosystems, Durham, North Carolina, United States

Gastrointestinal (GI) toxicities are the most frequently reported drug-induced adverse events in human clinical trials. The consequences of these toxicities can range from dose-limitations that reduce drug efficacy to chronic GI disease, hospitalizations, and even death. Animal models poorly predict these toxicities, leaving few options for early detection of GI side effects prior to human studies. RepliGut® Planar systems were developed to provide human relevant gut modeling in a format that can detect clinical outcomes earlier in the discovery and development process. A unique feature of this model is the ability to test drug-induced effects on proliferating primary human cells. Drugs targeting mitotic cells (colchicine and idarubicin) exhibited selective toxicity towards proliferative cells, whereas the 26S proteosome inhibitor, bortezomib, was toxic to all cell types. We tested the ability of a transverse colon stem cell platform to predict diarrheagenic potential of a reference set of 30 drugs with known clinical diarrhea incidence. Using three human donors, drugs were dosed over 6-orders of magnitude, overlapping the Cmax. Dose-response curves were generated for each donor utilizing endpoints that assess cell viability (DAPI quantification), proliferative capacity (EdU incorporation) and epithelial barrier formation (TEER). This model predicted diarrheagenic potential with high accuracy for each endpoint (91% DAPI, 90% EdU, 87% TEER). Variation in donor response was minimal. This model system provides informative screening for acute GI toxicity that integrates sensitive and robust endpoints, correctly identifies drugs associated with clinical diarrhea incidence and discerns toxicity towards specific cell populations of the intestinal epithelium.

P513

High Content Assessment of GI Toxicity in a Self-Renewing Human Intestinal Epithelium Model

Ben Scruggs ¹ , Maureen Bunger ¹ , Ashley Spreen ¹ , Lauren Boone ¹ , Yuli Wang ¹ , Shimeng Liu ² , Maureen Bunger ¹ , Chris Sims ¹ , and Liz Boazak ¹

¹ Altis Biosystems, Durham, North Carolina, United States

² AstraZeneca, Boston, Massachusetts, United States

Gastrointestinal (GI) side effects are the most frequently reported adverse events in clinical trials. A mechanistic understanding of GI toxicity, along with preclinical data supporting precise dosing, could improve clinical trial outcomes. We aimed to develop an in vitro model that mimics gut turnover processes to enable more physiologically relevant drug testing. RepliGut® 2D Crypt is a human stem-cell derived intestinal epithelial model that mimics cell turnover and spatial organization of the in vivo epithelium. Cells are plated on engineered impermeable cell culture inserts that are laser drilled with 91 microholes. A growth factor gradient establishes a self-organized tissue spatially segregated into a stem cell niche alongside absorptive and secretory cell types. Culture can be maintained ≥ 15 days, supporting sub-chronic, multiple, or pulsed dosing regimens. To evaluate model performance, crypts were treated for five days with IL-22, a reparative cytokine known to increase cell proliferation. Quantification of enterocytes (ALP), proliferative cells (EdU), and goblet cells (MUC2), revealed IL-22 treatment significantly reduced ALP+ enterocytes and increased EdU+ cells. These changes were partially reversed after a five-day washout. Assessment of two preclinical drugs in development revealed that a hyper-proliferative drug increased EdU+ cells and decreased ALP+ and MUC2+ cells whereas a cytostatic drug induced the opposite effect relative to vehicle treatment. The 2D Crypt model represents a significant advancement in the development of physiologically relevant gut microphysiological systems that permits repeat and long-term dosing to enable assessment of both adverse and therapeutic responses to gut-targeted interventions.

P514

Osteogenic Differentiation of Human Pluripotent Stem Cells as a Model for Assessing Developmental Bone Toxicity

Paula Van Rossum ¹ , Giel Hendriks ¹ , Amer Jamalpoor ¹ , Iris Müller ² , José Manuel Horcas Nieto ¹ , and Connor Philippo ¹

¹ Toxys, Oegstgeest, Zuid-Holland, Netherlands

² Unilever, Shambroom, United Kingdom

Evaluating the adverse effects of chemicals on human embryonic bone formation is a crucial component of regulatory in vivo studies addressing prenatal developmental toxicity. A highly promising alternative to reduce the need for in vivo testing is the use of stem cell-based in vitro methods. Here, we established a novel human induced pluripotent stem cell (hiPSCs)-based assay to predict developmental bone toxicity in vitro. Kinetic analysis of biomarkers involved in bone development (RUNX2, COL1A1, BGLAP) confirmed the presence of a differentiation pattern characteristic for osteogenesis. Moreover, immunostaining demonstrated the development of a dense, collagen-rich extracellular matrix that mineralized. The effect of teratogenic compounds was assessed by their ability to decrease the gene expression of different bone biomarkers, as well as their impact on mineralization. Cells were differentiated in the presence of test chemicals from day 0 to 21 and analyzed for osteogenic differentiation. Treatments with the developmental toxicants (thalidomide, warfarin, and lead-acetate) led to a decrease in the expression of different biomarkers including mesoderm (BRACHYURY), osteoblast progenitor marker RUNX2 and mature osteoblast-specific markers (COL1A1 and BGLAP) and affected the mineralization at subtoxic concentrations. In contrast, the non-developmental toxicants, saccharin, and penicillin had no significant impact on gene expression nor on mineralization. Here we demonstrated that in vitro differentiation of hiPSCs towards osteoblasts can be used to assess the teratogenic effect of chemicals on the bone development. Nevertheless, further studies are warranted to validate and establish the predictive capacity of the assay in assessing developmental bone toxicity.

P515

Early Prediction of Drug-Induced Cardiotoxicity (DICT) Using Human Reporter Cell Lines

Giel Hendriks ¹ , Bas Ter Braak ¹ , Liesanne Loonstra-Wolters ¹ , Eva De Tombe ¹ , Bensu Tangil ¹ , and Amer Jamalpoor ¹

¹ Toxys, Oegstgeest, Zuid-Holland, Netherlands

Cardiotoxicity poses a major challenge in drug development, often leading to the withdrawal of promising therapeutic agents. To address this issue, it is essential to develop robust and predictive in vitro assays that can identify potential cardiotoxicants early in the development process. In this study we applied ToxProfiler on a set of 60 drugs with known FDA DICT liabilities to gain insight into their toxicity signatures and link to in vivo cardiotoxicity. The ToxProfiler assay consists of a panel of seven GFP reporter cell lines that are specifically activated upon induction of oxidative stress, ER stress, cell cycle stress, ion stress, protein stress, mitochondrial stress, autophagy and inflammation. Using live cell confocal imaging of these seven reporters, toxicological profile of the compounds was obtained. Point of Departure concentrations for the different toxicological responses were calculated and used in combination with exposure (human plasma levels) information to potency rank and assess the performance of the assay in predicting DICT. We revealed that in vivo cardiotoxicants with distinct pharmacological mode-of-actions have a similar toxicity signature with different potency in ToxProfiler. Cardiotoxicants like calcium channel blockers (e.g., diltiazem, verapamil) or tyrosine kinase inhibitors (e.g., imatinib, ibrutinib and sorafenib) primarily induced autophagy activation in combination with a weak ER and oxidative stress at their clinically relevant concentrations. Overall, ToxProfiler identified most compounds with DICT liabilities, demonstrating that the assay can be implemented in the early phase of drug development to enhance the predictive accuracy of cardiotoxic potential, and minimizing adverse cardiac events in vivo.

P516

Human and Preclinical Animal Microphysiological Systems for Assessing Drug-Induced Liver Injury during Drug Development

Alysha Bray ¹ , Ovidiu Novac ¹ , Emily Richardson ¹ , and Tomasz Kostrzewski ¹

¹ CN Bio Innovations, Cambridge, England, United Kingdom

Drug induced liver injury (DILI) remains a common cause for acute liver failure and the attrition of compounds in drug development. New approach methodologies (NAMs) offer promise to increase the efficiency of drug development by detecting DILI earlier. The use of microphysiological systems (MPS) address the shortcomings in current drug development, by improving translation of preclinical results and thus accelerating safety evaluation. The adoption of these more advanced in vitro approaches is therefore inevitable to de-risk workflows and offset the trend towards more costly and limited animal use. The PhysioMimix® Organ-on-a-Chip (OOC) system was used to develop DILI assays using human or preclinical animal liver MPS. Perfused scaffolds were utilized to generate 3D microtissues from primary hepatocyte and Kupffer cells. Plates containing either 12 or 48 individual chips were deployed during drug development to accurately predict the human DILI risk of candidates. Superior sensitivity and specificity versus alternative models were demonstrated (sensitivity 100%, accuracy 85%) using reference compounds (IQ MPS Consortium) using the Liver-12 plate. ALT levels matched clinically based severity gradings, defined by the DILI Network (DILIN). Dog and rat MPS were tested against a panel of compounds to qualify their use in predicting species specific effects, mimicking known in vivo and clinical data. The preclinical animal DILI assays can inform the design of in vivo studies to predict interspecies differences. Together, these MPS systems bridge the gap by providing data to refine pre-clinical in vivo experimental design, reduce drug attrition rates and safeguard animal use.

P517

Assessment of the Cardiovascular Impact of E-Vapour and Heated Tobacco Products Using New Approach Methodologies

Matthew Stevenson ¹ , Fiona Chapman ² , Roman Wieczorek ³ , Edgar Trelles Sticken ³ , Sarah Jean Pour ³ , Luuk de Haan ⁴ , Linda Gijzen ⁴ , Matthias Gossman ⁵ , Bettina Lickiss ⁵ , and Liam Simms ²

¹ Imperial Brands PLC, Bristol, England, United Kingdom

² Imperial Brands PLC, Bristol, United Kingdom

³ Reemtsma Cigarettenfabriken GmbH, An Imperial Brands PLC Company, Hamburg, Germany

⁴ Mimetas BV, Oegstgeest, Netherlands

⁵ InnoVitro GmbH, Julich, Germany

Cigarette smoking is a cause of serious diseases in smokers, including heart disease. As e-vapour (EVP) and heated tobacco (HTP) products gain popularity among adult smokers, there is a need for rapid and sensitive methods to evaluate their biological impact. Here, we present data comparing the cardiovascular effects of cigarettes, EVPs and HTPs using a variety of in vitro New Approach Methodologies (NAMs). Our initial assessments utilised high throughput NAMs using: human endothelial cells in the scratch wound assay, and cardiomyocytes to assess contractility (FLEXcyteTM) and cytotoxicity (real-time cell analysis). Follow-on mechanistic studies utilised high-content screening technology on Human Coronary Artery Endothelial Cells (HCAECs), with a focus on oxidative stress. Since shear stress is an important component of disease modelling, we also explored organ-on-a-chip technologies to replicate key stages of atherosclerosis development. Using HCAECs cultured on an OrganoPlate® 2-lane chip (Mimetas BV), endothelial damage (cytotoxicity, oxidative stress), adhesion molecule expression (ICAM-1), cytokine expression and immune cell adhesion were modelled. We exposed the various NAMs to both cigarette smoke and aerosol (EVP & HTP) bubbled extracts, to replicate systemic exposures. The test samples used the 1R6F reference cigarette, Pulze&iD HTP and blu EVP (both obtained from EU market). Across the range of NAMs used, we consistently observed that cigarette smoke extracts elicited strong biological responses at low concentrations, whilst HTP and EVPs elicited weaker to no responses. These results add to the weight-of-evidence that the tested EVP and HTPs have the potential to make a positive contribution to tobacco harm reduction.

P518

Gnawing Device Enrichment Consumption in Rats: A Possible Indicator for Predicting Emetogenic Potential?

Shinichi Oshida ¹ , and Taku Masuyama ¹

¹ Japan Tobacco, Yokohama, Kanagawa, Japan

In the clinical and nonclinical development of pharmaceuticals, nausea/emesis can be dose-limiting toxicity. Since the rodent lacks a vomiting reflex, it is impossible to evaluate emetogenic potential directly as vomiting in rodents. Pica behavior, consumption of nonnutritive substances which are not normal diets, is known to be associated with various emetogenic stimuli in rats. For the evaluation of pica behavior in rats, kaolin, a clay mineral, is mainly used. In our laboratories, gnawing devices made of plastic are provided as environmental enrichment to rats. The purpose of this study was to investigate whether consumption of gnawing devices could be an indicator of pica behavior in rats. Three known emetogenic agents, cisplatin (dissolved in saline, 10 mg/kg, intraperitoneally), apomorphine (dissolved in saline, 10 mg/kg, intraperitoneally) and copper sulfate (dissolved in distilled water, 40 mg/kg, orally), were administered once daily for three/four days to male Sprague-Dawley rats (6 weeks of age, five animals per group) and the amount of gnawing devices consumed was measured daily. The consumption of the gnawing devices statistically significantly increased in cisplatin- or apomorphine-treated rats and tended to increase in copper sulfate-treated rats when compared with the corresponding vehicle control groups, indicating that gnawing device consumption was a possible indicator of pica behavior. The results obtained suggest the possibility that the measurement of gnawing device consumption can contribute to the prediction of emetogenic potential in the early stage of pharmaceutical development.

P519

Magnetic Resonance Imaging: Repeated Intrathecal Administration of Gadolinium in Adult and Juvenile African Green Monkeys (AGM)

Sven Korte ¹ , Michael Weed ¹ , Viviana Chaseau ¹ , M.S. Lawrence ¹ , Shervin Liddie ¹ , Lucy Nemchek ¹ , and Alexis Nisbett ¹

¹ Virscio, New Haven, Connecticut, United States

The African green monkey (AGM; Chlorocebus sabeus) shares close central nervous system (CNS) physiology and anatomy to humans and other Old-World primates and has been increasingly utilized as a preclinical test system for CNS route of administration, pharmacokinetic, pharmacodynamic and safety assessments. Central administration of compounds which do not cross the blood brain barrier such as antisense oligonucleotides (ASOs) is commonly delivered via intrathecal (IT) route into the cerebrospinal fluid (CSF). The current study demonstrates effective repeat percutaneous lumbar IT dosing to adult and juvenile AGMs. Four (4) animals, two (2) adults (3.5-6.7 kg, 8.5 yr.) and two (2) juveniles (1.7-1.8 kg, 21-22 mo.), were anesthetized, secured to a back-stretching device, sterilely prepped, and dosed with a 5-10 mM gadolinium (GAD) solution (Gadavest©, Bayer) into the intrathecal space between the L4 and L5 vertebrae. A CSF sample was drawn, followed by 0.5 mL of GAD solution and 2 mL/kg saline flush. MRI scans before and after infusion documented GAD distribution throughout the spinal column from tail to cisterna magna. Repeat dosing was conducted after 7 days. These studies demonstrate the effectiveness of lumbar bolus IT administration in delivering test articles to the spine and brain, and relevance and utility of the AGM in evaluating distribution, efficacy and safety of candidates delivered by this clinically important route of administration.

P520

Establishing In Vitro MultiFlow Assay for Identifying the Mode of Action of Potential Genotoxicants at ITR Laboratories Canada Inc

Rambabu Naravaneni ¹ , Klara Louise Bauermeister ¹ , Steven Bryce ² , Christian Giordano ¹ , and Joseph Younan ¹

¹ ITR Laboratories Canada Inc., Baie D’Urfe, Quebec, Canada

² Litron Laboratory, Rochester, New York, United States

The aim of this study is to establish endpoints indicating DNA damage (p53, γH2AX), frequency of mitotic cells (Phospho-Histone H3), cytotoxicity (nuclei counts), and polyploidy in TK6 cells treated with potential genotoxicants (clastogens or aneugens) and non-genotoxicants. Exponentially growing TK6 cells were treated in 96-well plates with various clastogens (Etoposide, Hydroxyurea, Methotrexate, 5-Fluorouracil, MMS, menadione, Stavudine), aneugens (Mebendazole, Nocodazole, Tozaserib, ZM-447439, AMG-900, Carbendazim, Crizotinib, Griseofulvin,) or non-genotoxic agents (Clofibrate, Gleevec, Hexachloroethane, Phenanthrene, CCCP, Cyclohexamide, Tunicamycin, Brefedine A) at various dose levels for 24 continuous hours with sampling at 4- and 24- hours of treatment. The highest concentration chosen for measuring multiplexed biomarkers aimed to achieve cytotoxicity of approximately 80% after 24 hours of continuous treatment. Concurrent vehicle/negative (DMSO) and positive controls (MMS & Carbendazim) were included in each treatment plate. The treated cell culture samples were collected and analyzed using flow cytometry at 4- and 24- hours post-treatment. Multiplexed machine learning models and the global evaluation factors approach accurately classified each compound as clastogenic, aneugenic, or non-genotoxic. Based on the results obtained in our study, we conclude that this method is useful for studying endpoints associated with DNA damage response pathways. It can effectively distinguish between clastogenic and aneugenic modes of action while identifying potentially misleading in vitro positives arising from cytotoxicity. Additionally, it is highly valuable for follow-up to positive in vitro micronucleus test results or as a primary screening tool, requiring minimal quantities of the test substance (e.g. 5 mg or 200 µL of a100 mM solution).

P521

Withdrawn

P522

A 29 Day Two Dose and 3 Day One Dose Research and Development Study of Intrathecal Injection or Epidural Infusion in Rats

Joseph Jilek ¹ , Kyle Wegner ¹ , James Nelson ¹ , Michael Taschwer ¹ , Sarah Newman ¹ , Joshua Ryan ¹ , Connor Faivre ¹ , Sara Bernhagen ¹ , Todd Quackenboss ¹ , and James Shupryt ¹

¹ Labcorp Early Development Laboratories Inc., Madison, Wisconsin, United States

The purpose of this study was to evaluate the tolerability of repeat intrathecal injections or continuous epidural infusion in rats. Male and female Crl:WI(Han) rats underwent either sham surgery; two intrathecal (IT) injections on Days 1 and 29 via direct puncture or catheterization; or a single continuous epidural infusion (72 hours) via implanted catheter. Dose volumes were 30 µL for IT procedures and 0.15 µL/hour or 5 µL/hour for epidural infusion. Control articles used included Omnipaque™ 300 (to confirm dose placement via fluoroscopy), artificial cerebral spinal fluid; or 0.9% sodium chloride. IT procedure-related, transient clinical observations included decreased activity, increased sensitivity to touch, self-mutilation of the front digits, or pica behavior. Epidural infusion-related clinical observations included red penis discharge/red urine, vocalization, or pica behavior. Procedure-related decreased forelimb or hindlimb grip strength was documented for all groups. For repeat IT dosing, no increase in the frequency or severity of clinical observations was noted, suggesting that repeat dosing is tolerated. Inflammation-related microscopic findings were noted in the skin/subcutis of the surgical site or at the catheter implant site and findings consistent with a degenerative process were noted at the entrance into the spinal canal. Within neural tissues, noteworthy findings included slight to moderate axonal degeneration and gliosis. These findings suggest that administration into the IT or epidural space of rats is viable via either single or repeat dose procedures; however, given the microscopic findings observed at the dose site, the potential for degenerative effects in the neural tissues should be considered.

P523

Use of Buccal Swabs for Gene Expression Analysis as a Minimally Invasive Endpoint in Nonhuman Primate Toxicology Studies

James Boyer ¹ , K. Borrelli ¹ , A. Ejaz ¹ , D. Maurer ² , K. McManus ¹ , N. Montanari ¹ , M. Rocheleau ¹ , and L. Thomas ¹

¹ Celldex Therapeutics, Inc., Fall River, Massachusetts, United States

² Celldex Therapeutics, Inc., New Haven, Connecticut, United States

The principles of the 3R’s are a guiding philosophy in the design of toxicology studies. In addition, there is also a desire to have experimental endpoints that can be assessed longitudinally with minimal impact on other endpoints. We investigated the use of buccal swabs as a minimally invasive technique to collect gene expression data in cynomolgus macaques. In the first study, buccal swabs were collected from 5 cynomolgus macaques at 3 timepoints of a pilot study. Animals were anesthetized and a sterile cotton swab was rubbed over the buccal mucosa for about 5 seconds near the front of the oral cavity. The swab was placed into RNAlater solution (ThermoFisher) for RNA expression analysis using Nanostring technology (Canopy Biosciences). This method resulted in the capture of monkey RNA, but the quantity proved variable between timepoints and animals. Regardless, expression of selected genes was able to be detected and quantitated. To refine the technique, buccal swabs were collected from 4 cynomolgus macaques at 3 timepoints of a second study. Here, a sterile cotton swab was rubbed over the buccal mucosa 8-10 times, rotating the swab during the collection, for about 30 seconds, inserting further into the oral cavity, near the cheek pouch. This method yielded more consistent amounts of nucleic acid between samples. Expression of selected genes were able to be evaluated across the study. In conclusion, buccal swabbing can prove to be a valuable minimally invasive method to collect longitudinal data in primate studies for assessing gene expression.

P524

A Review of Exploratory Dose Range Finding Study Outcomes

Aimee Altemus ¹ , Pamela Foppiano ¹ , Matthew Palmer ¹ , Jennifer Byerley ¹ , and Michael Thibodeau ¹

¹ Boehringer Ingelheim, Ridgefield, Connecticut, United States

The main objectives of a dose range finding (DRF) study are to determine the maximum tolerated dose (MTD) and support the dose selection for the Phase 1 enabling GLP studies. In 2018, our facility introduced a streamlined DRF study design termed an exploratory DRF (eDRF) design applicable to rodent and nonrodent small molecule programs. The purpose of this initiative was to retool the rodent and nonrodent designs to minimize animal use according to the 3Rs principles and improve study efficiency by limiting endpoints to a focused list and shortening the dosing phase duration without jeopardizing study objectives. We reviewed 17 eDRF studies conducted since the roll-out to assess if study designs 1) resulted in reduced animal usage and an improvement in study efficiency, and 2) adequately informed selection of dose levels for those compounds advancing to Phase 1 enabling GLP toxicology studies. We found that the study designs applied in practice reduced animal use, number of endpoints, and dosing phase duration, thus improving study efficiency compared to previous DRF designs. The eDRF study design also allowed for test item-specific endpoints as warranted. In summary, rodent and nonrodent eDRF studies adequately informed dose selection for the Phase 1-enabling toxicology studies. However, there were challenges in evaluating tolerability in a limited time frame prior to escalation and after resumption of dosing following dose holidays in the nonrodent design. Further application of these designs will help refine the approach.

P525

Evaluation of Endogenous and Ex Vivo Stimulated Matrices for Validation of Cytokine Expression Assessment

Allison Repic ¹ , Amanda Lucchini ¹ , and Andrew Mcdermott ¹

¹ Labcorp, Madison, Wisconsin, United States

Elevated concentrations of cytokines indicate activation of pathways associated with inflammation, and the measurement of cytokines has been widely used to understand and to monitor the effects of therapeutic treatments. Validated methods for the evaluation of biomarkers such as cytokines require a biological matrix relevant to the physiology and distribution of the biomarker. The source of the biomarker can potentially influence the validation process and subsequently the robustness of the method. Endogenous non-human primate (NHP) cytokine samples, and ex vivo stimulated NHP samples [whole blood and peripheral blood mononuclear cells (PBMCs)] were generated and compared to recombinant cytokines. These samples were assessed across kits and assay format for cytokine expression to evaluate the utility of these matrices in validating methods. The NHP endogenous matrix, stimulated matrix, and spiked matrix using recombinant material all generated methods that reliably detected cytokines in study samples and demonstrated similar patterns of cytokine expression. Samples were assessed on the Bioplex 200 and MSD® s600 Sector imager instruments using the Millipore® NHP multiplex and MSD® PIP VPlex cytokine kits. Different ex vivo stimulation conditions resulted in cytokine levels both above and below endogenous expression in NHPs administered 0.002 mg/kg LPS; all matrices demonstrated similar percentage of samples passing precision for proinflammatory cytokines. While endogenous samples are preferred it can be difficult to obtain an identical matrix to that of the study samples. The comparison data demonstrate that the cytokine expression in ex vivo generated matrices is acceptable and scientifically appropriate for method validation.