Abstract
HFO-1234ze (E) is proposed as a near zero global warming propellant for use in metered dose inhaled (MDI) products. This paper describes the non-clinical safety assessment in mice, rats, and dogs and supplements previously reported data (genetic toxicology, short-term toxicology, and reproductive toxicology). In all species, HFO-1234ze (E) was only detectable in blood for a short period after dosing with no evidence of accumulation. HFO-1234ze (E) was without any toxicological effects at very high doses in subchronic (13-week mouse) and chronic (39-week dog) studies. Chronic (26-week) administration to rats at very high doses was associated with an exacerbation of rodent progressive cardiomyopathy, a well-documented background finding in rodents. In a 2-generation study, extremely high doses were associated with the early euthanasia of some lactating female rats. This finding was considered to be significantly influenced by a state of negative energy balance, reflecting the specific vulnerability of rats during lactation. These findings are considered to not pose a risk to humans with typical MDI use given they occurred at doses which far exceed those expected in patients. Overall, the nonclinical safety data for HFO-1234ze (E) support its further development as an MDI propellant.
Keywords
Introduction
A critical component of a metered dose inhaler (MDI) formulation is the propellant (liquified compressed gas) which facilitates the generation of an aerosol cloud containing the micron-sized particles of pharmaceutical ingredient(s), ultimately inhaled by the patient. 1 Hydrofluoroalkane (HFA) propellants (subsequently referred to as HFAs) were first introduced into MDI applications in 1994 with Proventil-HFA with the aim of replacing chlorofluorocarbon (CFC) propellants due to concerns about their ozone-depleting properties. 2 Currently used pharmaceutical propellant gases include HFA-134a and HFA-227ea. Since their introduction, HFA MDIs have been safely used by millions of people with respiratory diseases such as asthma and chronic obstructive pulmonary disease (COPD). 3
Recent environmental initiatives, such as the European Union F-gas regulations, have identified that certain HFAs have a high global warming potential (GWP) leading to restrictions in their use in applications such as refrigeration and fire suppression. 4 Although these initiatives have not specifically targeted the use of HFA in MDIs, and noting that, while MDI use accounts for only a very small proportion of global HFA emissions, it is appropriate to replace the currently used propellant gases with alternatives that have a reduced potential for environmental impact but without compromising patient outcomes.
Propellant Material Properties for HFO-1234ze and HFA-134a.
Prior to human administration, the safety of any new pharmaceutical ingredient must be fully characterized in accordance with regulatory guidance. HFO-1234ze has therefore been evaluated in a series of toxicology studies in rodents and nonrodents as well as reproductive toxicology, safety pharmacology (cardiovascular, central nervous system, and respiratory functional endpoints), and genetic toxicology assessments. Two-year carcinogenicity studies in mice and rats are ongoing (data not yet available). Despite the significant challenges associated with analyzing a gaseous test material such as HFO-1234ze in biological matrices, toxicokinetic (TK) data was generated in whole blood from mice, rats, and dogs. A 3-month drug product “combination” toxicology study was conducted in dogs which demonstrated that the toxicology and toxicokinetic performance of the active ingredients for a HFO-1234ze-based MDI drug product was comparable to the reference MDI drug product (HFA-134a based) (data not shown).
Nonclinical Studies Conducted With HFO-1234ze.
All inhalation studies conducted by snout only (rat and mouse) or face mask (dog) exposure unless otherwise stated.
h (hours), m (minutes), GD (gestational day), MDI (metered dose inhaler), and API (active pharmaceutical ingredient).
aIncluding HFO-1234ze TK.
bIncluding electrocardiography.
cStudies reported by Rusch et al, 2013.7
dIncluding electrocardiography (telemetry) and respiratory inductive plethysmography.
eUnscheduled DNA synthesis and micronucleus assessments performed as part of the 4-week study in rats.
fIncluding Irwin (central nervous system function) and head-out plethysmography (respiratory function) assessments.
With the exception of a 2-generation study, all studies which were conducted to chemical guidelines were reported previously. 7 In this manuscript, we report data from repeat dose toxicology (with safety pharmacology endpoints) and support toxicokinetic data in mice (13 weeks), rats (26 weeks), and dogs (39 weeks). Data from the manufacturer sponsored 2-generation study in rats is also reported.
In order to summarize the overall safety of HFO-1234ze, key findings from the previously reported studies 7 are described and integrated with other available data; otherwise, these study designs and data are not reproduced.
It is pertinent to note some of the key differences (historically and more recently) in study designs required under chemicals vs pharmaceuticals testing guidelines, particularly for the inhalation route of administration: • High doses for OECD chemical guideline studies are generally set to ensure a maximum tolerated dose is achieved. This helps the chemical industry in risk assessment for long-term occupational exposure or for short-term accidental exposures. For pharmaceuticals, dose selection is designed to characterize toxicology dose responses and achieve sufficient safety margins to the anticipated exposure in the clinical setting. • Guidelines for chemical testing are based on exposure durations of 6 h/day for either 5 days/week for standard toxicology studies or 7 days/week for reproductive studies to ensure all aspects of the reproductive cycle are covered. • For pharmaceuticals, ICH guidelines are relevant (including ICH M3R2, ICH S5 (R3), and ICH S7A), for the inhalation route of administration, and taking into account animal welfare considerations, it is preferable to limit the exposure duration for rodents to ideally 60 min or less (although up to 6 h is feasible, at least for studies up to 13 weeks) and less than 60 min for dogs (although up to 2 h is feasible for dogs with adequate training).
Where necessary, the design and conduct of some studies with HFO-1234ze conducted under chemical guidance were reviewed to ensure their conduct was sufficient to meet pharmaceutical guidelines. In particular, the embryo-fetal development studies in rats and rabbits and a 2-generation reproductive performance study in rats to support the reproductive toxicology package were reviewed against the ICH S5 (R3) guidance for the detection of reproductive and developmental toxicity for human pharmaceuticals. 11 The review confirmed that all reproductive toxicology studies were robust and generally met or exceeded the endpoints required by ICH guidance.
It is noted that whole body exposure was employed for some early studies, 7 including the 2-generation study. With this method, an animal is exposed via inhalation as well as other routes, such as oral or dermal. 12
Methods
Test material
Trans-1,3,3,3-tetrafluoropropene (HFO-1234ze; CAS no.: 29118-24-9) is a colorless, non-flammable liquefied gas with a boiling point of −19°C. For the studies sponsored by the manufacturer, HFO-1234ze was supplied in cylinders by Honeywell (Buffalo, New York, USA) at a minimum purity of 99.8%. For subsequent studies, purified HFO-1234ze of >99.99% was used. All batches were analyzed by gas chromatography (GC)/mass spectroscopy (MS).
Atmosphere Generation, Exposure Systems, and Atmosphere Analysis
Different exposure levels were achieved by varying the HFO-1234ze concentration in the exposure systems, whilst keeping the duration of exposure constant within (based on the study design/species). Control animals were exposed to air only at the same flow rate as the high dose animals. All animals were exposed once daily for the predefined exposure period (see study summaries).
HFO-1234ze atmosphere concentrations were measured by gas chromatography using a ZB-5, 30 m × .32 mm ID, .25 mm film thickness analytical column and flame-ionization detection.
For toxicology and toxicokinetic studies in rodents the inhalation exposure system comprised a pressurized gas cylinder containing the test item, a snout-only inhalation exposure chamber, restraining tubes, and a pre-chamber. The snout-only inhalation chamber (flow through chamber) was a modular apparatus of aluminum alloy construction comprising a base unit, an animal exposure section, and a top section incorporating a central inlet. A separate chamber was used for each group. Each exposure system was housed in a separate extract cabinet to avoid possible cross-contamination between groups. During exposure, the mice/rats were held in restraining tubes with their snouts protruding into the exposure chamber. Exposure ports not in use were closed with blanking plugs. Prior to exposure, animals on study were acclimated to the snout-only method of restraint over a 5-day period immediately preceding the first test substance exposure.
For all dog studies, the inhalation exposure system for treated groups comprised a pressurized gas cylinder containing the test item, a conditioning chamber fitted with 2 outlets, tubing to sub-chambers and face masks, and further tubing to an extract system. The conditioning chambers used on the study connected to two sub-chambers, each with multiple take-off points. Each take-off point supplied one face mask position. Up to eight mask positions were used simultaneously to expose the animals. The face masks were constructed of plastic with a flexible rubber seal. The mask was secured over the dog’s nose and mouth by a muzzle. The animals on study were acclimated to the method of light restraint (tether), for at least 14 consecutive days before treatment started, immediately preceding the first test substance exposure, including at least 12 days minimum acclimatization with air exposure.
For the 2-generation reproductive study, rats were exposed in whole body exposure chambers: .75 m3 chamber of stainless-steel and glass construction, with animals held in individual stainless steel mesh compartments.
Determination of Estimated Achieved Dose
For all nonclinical studies, dose equivalents, in terms of mg/kg day, were calculated based on the exposure concentration in parts per million (ppm), a conversion factor to w/v of 1 ppm = 4.66 µg/L (molecular weight of HFO-1234ze of 114 g/mol), the duration of exposure, and average bodyweight of each animal species according to the equation below
13
; a deposition factor was not applied due to the gaseous nature of HFO-1234ze.
where
C = concentration in air
RMV = Respiratory minute volume = .608 × body weight (Kg)0.852
D = Duration of exposure.
Toxicokinetic (TK) Assessment
TK data was initially generated in a 7-day rat study (at the same doses evaluated in the 26-week toxicology study); this study was preceded by a 7-day feasibility study (sample collection and bioanalysis, data not reported). Subsequently, TK data was generated in mice (13-week study) and dogs (39-week study). Blood sampling and handling was similar across all 3 species; animals were restrained (under isoflurane anesthesia for mice only) during the sampling procedure and the blood samples (100 µL), without anticoagulant, were obtained without overnight deprivation of food via the jugular vein. Single aliquots of the 100 µL, accurately measured, were immediately transferred from the blood sampling syringe to a sealed 20 mL headspace glass vial containing 10 mL of water:dimethylsulfoxide (90:10 v:v). Blood was placed at room temperature following collection and stored for up to 48 h. The presence of the test item was determined by headspace GC-MS. The linearity of the method was confirmed over the target concentration range of .5 to 500 μg/mL. The lower limit of quantification was < 0.474 µg/mL.
Exposure of animals to HFO-1234ze was expressed as the maximal blood HFO-1234ze concentration (Cmax) for all species. As described in the results section, an area under the blood concentration-time curve from zero to the last quantifiable time point, (AUC), was determined in some dogs only, and based on very limited data. Given the sparsity of TK data, no AUC data are presented.
Mouse (13-Week Study)
The design of the 13-week mouse study is described in the repeat dose section. Blood sampling timepoints were immediately after exposure (IAD) (all groups including controls) and 30, 60, and 120 min after exposure for all HFO-1234ze exposed animals.
Rat (7-Day TK Study)
TK data was not generated on any of the repeat dose toxicology studies in rats and was therefore generated in a subsequent, stand-alone, 7-day study at the same target concentrations as those employed in the 26-week rat study. Three groups of RccHan™:WIST rats were therefore exposed to HFO-1234ze for 7 days, 4 h/day, at target concentrations of 1500 ppm (3/sex) or 5000 or 15,000 ppm (6/sex). The achieved concentrations of HFO-1234ze were 1550, 5280, and 15,900 ppm with an estimated delivered dose of 1380, 4690, and 14,100 mg/kg/day for Groups 1 to 3 respectively. Sampling timepoints were IAD, 15 and 30 min (1500, 5000, and 15,000 ppm), 60 min (5000 and 15,000 ppm), and 90 min (15,000 ppm) after exposure on Days 1 and 7.
Dog (39-Week Study)
The design of the 39-week dog study is described in the repeat dose section. Blood sampling timepoints were IAD, 15 min after exposure during Weeks 2, 7, 14, 28, and 38 (control) or IAD, 15, 30, 60, and 90 min after completion of exposure during Weeks 2, 7, 14, 28, and 38 or 39 (treated).
Repeat Dose Studies
All studies were conducted in the United Kingdom to Good Laboratory Practice (GLP) standards in accordance with the respective institutional UK Home Office project licenses and current guidelines for the nonclinical safety evaluation of pharmaceuticals. All animal studies were conducted in in AAALAC accredited facilities and accordance with UK Home Office legislation (Animals [Scientific Procedures] Act 1986) and AstraZeneca’s Bioethics Standard.
Animals and Animal Management
Mice and Rats
For the 14-day and 13-week studies in mice (six to seven weeks old at the start of dosing) and the 26-week study in rats (seven to eight weeks old at start of dosing), animals were housed in groups of up to three of one sex/cage and provided access to standard rodent diet (Teklad 2014) and water (public supply via polycarbonate bottles fitted with sipper tubes). Mice had free access to food and water throughout the studies except during the exposure periods. For rats, food was freely available except during exposure periods and urine collection (4 h and overnight respectively) while water was freely available except during exposure periods and overnight for blood sampling for hematology or blood chemistry/during urine collection. For mice and rats, the rooms were maintained on a 12-hour light/dark cycle and temperature and relative humidity were 20 to 24°C and 40 to 70%, respectively, with only minor deviations for short periods observed. The cages were made of a polycarbonate or polypropylene body with a stainless steel mesh lid, with environmental enrichment provided (wooden balls and a plastic shelter, plus bedding for nesting, with additional aspen chew blocks for rats). Food hoppers and water bottles were changed at appropriate intervals. At necropsy, main study animals were terminated one day after administration of the last dose, using an overdose of intraperitoneal pentobarbitone sodium followed by exsanguination.
Dogs
For the 14-day and 13- and 39-week studies, dogs (seven to 12 months old at start of dosing) were housed as two or three of the same group/sex in pens with softwood-based granules as litter. Dogs were allowed social interaction except for the period just before each exposure period up until shortly after food residues were removed. Group housing of animals was carried out by removal of partitions between adjacent kennels. Additional environmental enrichment included daily exercise, provision of a variety of toys in pens and the exercise area, a comfortable resting area, and treats following daily dosing. Each dog was offered approximately 400 g of Teklad 2025C daily at approximately the same time of day and allowed access to it for at least one hour. During the training for dosing and dosing periods, food was offered on return to the pen following completion of training/dosing. Uneaten food was removed and weighed. Water (public supply) was available ad libitum except during dosing and the period of urine collection (approximately 16 h overnight). The room was maintained on a 12 h light/dark cycle and target values for temperature were 15 to 24°C. Relative humidity was monitored continuously but not controlled. Necropsy was conducted one day after the final dose; animals were given overdose by intravenous injection of sodium pentobarbitone solution (200 mg/mL) by followed by exsanguination.
2-Generation Reproduction Study in Rats
Adult selected female rats were housed in groups of up to 4/cage during the premating and post weaning periods. During the mating period unrelated males and females from the same treatment group were pair housed. After mating (from GD0), the females were singly housed throughout gestation until their litters were born at which time they were housed singly, with their respective litter. Between weaning and termination unselected offspring where housed as a litter. Adult selected male rats housed in groups of up to 4/cage except during the mating period. F0 rats were 40 to 46 days of age at the start of the treatment period. Rats were allowed free access to a standard rodent diet (SDS VRF1 Certified diet). This diet contained no added antibiotic or other chemotherapeutic or prophylactic agent. The F0 animals (parent generation) were allowed to acclimatize to the conditions described below for 18 days before treatment commenced. The room was maintained on a 12 h light/dark cycle and target values for temperature and relative humidity were 19 to 25°C and 29 to 70%, respectively, with only minor deviations for short periods of time throughout the study. The cages (polycarbonate with polycarbonate floors, except during mating) were distributed on the racking to equalize, as far as possible, environmental influences amongst the groups. Wood-based bedding was used in cages with solid floors and was sterilized by autoclaving and changed at appropriate intervals each week. The gridded cages were suspended over trays covered with absorbent paper which was changed daily during mating. IMS Nestlets were provided from Day 15 after mating and throughout parturition and lactation. Cages, cage-trays, food hoppers and water bottles were changed at appropriate intervals. Each cage of animals was provided with an Aspen chew block and a plastic shelter for environmental enrichment. Chew blocks were provided throughout the study (except during mating and littering), and were replaced when necessary. Plastic shelters were provided throughout the study (except during mating and littering) and were replaced at the same time as the cages.
Study Designs
Repeat Dose Toxicology Studies
Exposure Concentrations and Doses for Studies in Mice.
14-day and 13-week studies in mice employed an exposure period of 2 h/day.
Exposure Concentrations and Doses for Studies in Rats.
aExposure duration of 4 h/day.
bExposure duration of 6 h/day.
cMean of F0 and F1 achieved concentrations.
dAchieved concentrations for F0 and F1 generations, respectively.
Exposure Concentrations and Doses for Studies in Dogs.
aLower estimated achieved doses in the 14-day study represent the 1 h exposure period, relative to that employed in the subsequent 13- and 39-week studies (2 h).
Mouse (14-Day and 13-Week Studies)
CD-1 (Crl:CD1(ICR)) mice in the 14-day study (3/sex/group) and 13-week study (10/sex/group) were exposed to air alone (control) or HFO-1234ze by nose only inhalation at the concentrations/doses shown in Table 3. In each study, the exposure period was for 2 h/day. Prior to exposure, animals were acclimated to the method of restraint, over a 5-day period immediately preceding the first dose. Main study animals were terminated for scheduled necropsy one day after the final dose by an overdose of intraperitoneal pentobarbitone sodium followed by exsanguination. In both studies, animals were assessed for clinical condition, body weight, food consumption, hematology (14-day study only), blood chemistry, organ weight, macropathology, and histopathology. TK was performed on the 13-week study (Weeks 4, 9, and 13) as detailed previously.
Rat (26-Week Study)
Rats (RccHan™:WIST) (15/sex/group) were exposed to air alone (control) or HFO-1234ze by nose only inhalation at the concentrations/doses shown in Table 4. The exposure period was 4 /day. Prior to exposure, animals were acclimated to the method of restraint, over a 5-day period immediately preceding the first dose. Animals were terminated for scheduled necropsy one day after the final dose was given by an overdose of intraperitoneal pentobarbitone sodium followed by exsanguination. Animals were assessed for clinical condition, body weight, food consumption, water consumption, ophthalmoscopy, hematology (peripheral blood), blood chemistry, urinalysis, organ weight, head-out plethysmography (respiratory function), Irwin screen (Central nervous system (CNS) function), macropathology, and histopathology.
Dog (14-Day and 13- and 39-Week Studies)
Beagle dogs in the 14-day study (2/sex/group, RCC (HsdRcc:DOBE)), 13-week study (3/sex/group, RCC (HsdRcc:DOBE)) or 39-week study (4/sex/group, Marshall) were exposed to air alone (control) or HFO-1234ze at the concentrations/dose shown in Table 5. Prior to exposure, animals were acclimated to the method of restraint, for up to 14 days before the first dose. In the 14-day study, telemetry transmitters were surgically implanted into each dog prior to the pretreatment period for the measurement of electrocardiography and blood pressure. Across the studies, animals were assessed for clinical condition, body weight, food consumption, ophthalmic examination, electrocardiography, blood pressure (14-day study), respiratory measurements (14-day study only, using respiratory inductive plethysmography), hematology, blood chemistry, urinalysis, organ weight, macropathology, and histopathology.
Rat (2-Generation Reproduction Toxicity Study)
In this study, the effect of HFO-1234ze on the integrity and performance of the male and female reproductive systems, and the growth and development of the offspring was assessed through two successive generations of Crl:CD (SD) rats. The study design is shown in Figure 1. Inhaled 2-Generation reproductive performance study design.
For the F0 generation, rats (28/sex/group) received HFO-1234ze by whole body inhalation administration at target exposure levels of 0 (air control), 2000, 5000 or 20,000 ppm for 10 weeks before pairing, throughout pairing, gestation, lactation, and until termination.
The selected F1 generation comprised of 24 male and 24 female progeny from each group, and these juvenile animals, which had been potentially exposed in-utero or via the milk also directly received air alone (control) or HFO-1234ze, as per the F0 generation, from weaning for a minimum of 10 weeks before pairing, throughout pairing, gestation and lactation and until termination. Dosing of the high dose F1 generation females given a target concentration of 20,000 ppm stopped after 16 weeks of treatment (approximately 2 weeks into the lactation phase for the majority of the females) as some unscheduled deaths occurred (described in the Results section).
Adults reaching the end of the scheduled treatment period were killed by intraperitoneal injection of sodium pentobarbitone followed by exsanguination. Adult animals not completing the planned treatment period and unselected offspring killed after Day 14 of age were killed by carbon dioxide asphyxiation. Offspring killed before Day 14 of age (including Day 4 culls; litters were standardized to 10 pups (5 males and 5 females) where possible) received an intraperitoneal injection of sodium pentobarbitone.
During the study, data was recorded on clinical condition, bodyweight, food consumption, estrous cycles, mating performance and fertility, gestation length and parturition observations and reproductive performance. Sperm analysis (including count, motility, and morphology), organ weight, and macroscopic and microscopic pathology investigations were undertaken on each adult generation. The clinical condition of offspring, litter size and survival, sex ratio, physical development, sexual maturation (selected F1 generation only), and bodyweight gain were assessed and organ weight and macroscopic and microscopic pathology investigations were undertaken on surplus unselected offspring from each set of litters at approximately 4 weeks of age.
Results
TK Assessment of HFO-1234ze
Blood HFO-1234ze Exposure in Toxicology Studies.
aMouse exposure period = 2 h/day.
bachieved HFO-1234ze concentrations within +/− 10% of those achieved in the 26-week rat toxicity study.
cRat exposure period = 4 h/day.
dDog exposure period = 2 h/day.
Mouse (13-Week Study)
On each occasion, HFO-1234ze was only quantifiable in blood samples collected immediately after dosing (which corresponded with Tmax). Since only IAD values were obtained, no AUC or T1/2 could be calculated. Cmax increased with the increase in HFO-1234ze concentration. Cmax values were generally higher during Weeks 9 and 13 than during Week 4.
Rat (7-Day Study)
On each sampling occasion, HFO-1234ze was only quantifiable in blood samples collected immediately after dosing (which corresponded with Tmax). Since limited values were obtained, no AUC or T1/2 could be calculated. Tmax was observed immediately after the end of exposure (IAD) on Days 1 and 7. Concentrations declined rapidly thereafter and were not quantifiable in any group after 30 minutes post exposure. Cmax, increased with the increase in HFO-1234ze concentration.
Dog (39-Week Study)
On each occasion, blood concentrations were generally quantifiable up to a maximum of 30 min after the end of exposure at 1680 or 4850 ppm and up to 90 min (the last sampling time) after the end of exposure at 14,700 ppm. Tmax was observed immediately after dosing. HFO-1234ze concentrations in blood generally increased proportionally with dose. Mean concentrations of HFO-1234ze were similar during Weeks 2, 7, 14, 28, and 38/39.
Toxicology
Given the variation in exposure concentration (ppm) and duration, and to facilitate comparison to expected human dose via MDI use, all results are described in terms of the estimated achieved doses (mg/kg/day).
Mouse (14-Day and 13-Week Studies)
There were no toxicologically significant HFO-1234ze-related effects, including any microscopic findings, up to the highest estimated doses in the 14-day (28,000 mg/kg/day) and 13-week (29,000 mg/kg/day) studies in mice. In the 13-week study, only minor clinical observations were noted (also noted in controls) alongside occasional small perturbations in hematology parameters and some evidence of a stress response.
Dog (14-Day and 13- and 39-Week Studies)
There were no toxicology findings, including any microscopic findings in dogs at the highest estimated doses in the 14-day (1740 mg/kg/day), 13-week (3620 mg/kg/day), or 39-week (3647 mg/kg/day) studies. This also included an absence of effect on functional respiratory or cardiovascular safety pharmacology parameters.
Rat (26-Week Study)
Exposure to HFO-1234ze at estimated doses of 1220, 4280, and 12,000 mg/kg/day was well tolerated and there were no effects on the safety pharmacology endpoints related to respiratory, or neurobehavioral or physiological function (plethysmography and Irwin assessments). Animals at the highest dose showed evidence of an increased incidence/severity of rodent progressive cardiomyopathy, with an increase in severity seen in males (minimal-slight) vs females (minimal). This was consistent with effects seen in previous studies in rats of shorter duration (2, 4, and 13 weeks). 7 Sporadic increases in plasma aspartate transaminase levels were noted but with no clear association to the myocardial findings on an individual animal basis. Other differences in hematology and urinalysis were minor and considered not to be of toxicological significance. For all other parameters, there were no toxicologically significant changes considered to be related to HFO-1234ze. Based on the increased incidence of minimal rodent progressive cardiomyopathy in both sexes given 12,000 mg/kg/day (3/15 males showed slight severity), a conservative no observed adverse effect level (NOAEL) was considered to be the intermediate dose of 4280 mg/kg/day.
Rat (2-Generation Reproduction Study)
There were no effects on the reproductive endpoints examined, including fertility, or general reproductive performance as assessed by mating performance and the ability to induce or maintain pregnancy. There were no treatment-related macroscopic or microscopic changes in the male or female reproductive organs of either generation and no effect of treatment on sperm motility, counts, or morphology in F0 or F1 generation males.
Effects in this study were limited to the highest dose tested (23,000 mg/kg/day). There were no effects of treatment at 2400 or 5800 mg/kg/day.
At 23,000 mg/kg/day, bodyweight gain of the F1 generation from Day 7 (males) to Day 4 (females) up to weaning was slightly lower than the controls. Slightly lower bodyweights persisted throughout the F1 generation. At this dose, there were seven F0 generation female decedents during late lactation (between Days 11 and 21), and two F1 generation female decedents during late lactation (Days 14 and 15). Following the F1 decedents, treatment was withdrawn for the remaining females in this group; it is noted that one F1 female was killed 12 days after cessation of treatment due to its poor condition and was considered not to be associated with HFO-1234ze. Three of the affected F0 generation decedents presented with hindlimb paralysis. Histopathologically, HFO-1234ze-related findings were observed in the heart, brain and spinal cord of F0 females given 23,000 mg/kg/day only. No HFO-1234ze-related histopathology findings were noted in F1 animals. In the brain of F0 females, minimal to marked neuronal necrosis with or without slight to moderate gliosis and/or neuronal loss was observed in the cerebellum of four animals. In one decedent animal, the marked neural necrosis and moderate gliosis was associated with moderate swelling of the astrocytes, vacuolation of the neuropil, and slight hemorrhage. This animal also had findings in the lumbar spinal cord (moderate gliosis, vacuolation of the neuropil, and neuronal loss); these findings were the most severe observed and were believed to have contributed to the clinical observations that necessitated euthanasia.
In the heart, an increased incidence and severity of rodent progressive cardiomyopathy was observed in males and females exposed to HFO-1234ze at 23,000 mg/kg/day. Compared with controls, this finding was more commonly multifocal in treated animals and due to the severity was considered adverse.
In conclusion, based on the unscheduled deaths in the F0 and F1 generations during late lactation, brain lesions, and rodent progressive cardiomyopathy observed in animals exposed to 23,000 mg/kg/day, the no observed effect level (NOEL) for parental (systemic) toxicity was 5800 mg/kg/day. In the absence of any reproductive effects, the NOEL for reproductive performance of the male and female reproductive system, and growth and development of the offspring, was 23,000 mg/kg/day, the highest dose tested.
Discussion
HFO-1234ze is being developed as a propellant with near zero global warming potential, for use in inhaled MDI drug products; initial clinical studies with a HFO-1234ze-based MDI formulation are complete. 14 HFO-1234ze has been evaluated in a series of inhalation toxicology studies in rodents and nonrodents (including TK assessment), as well as reproductive toxicology, safety pharmacology and genetic toxicology assessments. Early studies were conducted to satisfy chemicals regulations and explored very high HFO-1234ze concentrations while longer term studies were performed in accordance with appropriate pharmaceutical guidelines. With the exception of the 2-generation reproductive study (described in this paper), early studies were previously reported by Rusch and demonstrated that HFO-1234ze was not genotoxic and was not associated with any effects on embryo-fetal development. 7 Key findings in these studies are summarized below, where necessary to enable contextualization to subsequent data.
Toxicokinetics and Metabolism
In accordance with standard inhalation toxicology study protocols, blood samples for TK were collected following completion of the dosing period only. Consequently, these samples only capture the very rapid elimination phase of HFO-1234ze and do not account for the prolonged exposure periods (2 to 4 hours) in these studies. Systemic exposure to HFO-1234ze is therefore likely to be significantly underestimated. Given the unavoidable delay between the end of dosing and collection of the first blood sample, it is likely that the reported Cmax values are also underestimated. Overall, Cmax (and Tmax) was observed immediately after dosing in all animals; HFO-1234ze was rarely detectable in blood samples collected beyond 30 min after dosing demonstrating that HFO-1234ze does not accumulate, despite the very high doses tested. The animal exposure data with HFO-1234ze are generally in-line with that observed for the existing propellant, HFA-134a. 15 These data are also in-line with data generated by Schuster et al 16 ; this group evaluated the fate of inhaled HFO-1234ze in urine samples collected from rats and mice and demonstrated that less than 1% of the inhaled HFO-1234ze dose was metabolized. These data are in general accordance with that reported for HFA-134a and HFA-227ea 17 and HFA-152a. 18
Human exposure data was considered not to add value to the overall risk assessment for HFO-1234ze and has not been generated. This position considered the rapid elimination of HFO-1234ze observed in toxicology studies in which HFO-1234ze was administered at doses (based on long exposure periods at very high concentrations) far in excess of those possible in humans with typical MDI use. For example, the exposure periods in animals (two to four hours) are in stark contrast to the clinical situation whereby a patient will likely complete a single dose (1 or 2 inhalations from an MDI) in around 1 minute.
Toxicology
HFO-1234ze was very well tolerated in nonclinical studies in mice, rats, dogs, and rabbits. In short-term studies in rats, very high concentrations/doses of HFO-1234ze were associated with microscopic findings in the nasal passages, liver, and heart. 7 Of these, only the heart findings were reproduced in longer term rat studies. These findings were not observed in mice or dogs. In addition, an extremely high dose (23,000 mg/kg/day) was also not tolerated by some lactating female rats in the 2-generation reproductive toxicology study.
The myocardial findings in rats are entirely consistent with the acute and chronic stages of RPCM. RPCM is a well-documented spontaneous finding in the heart of rats, with a predisposition for male animals, and a higher incidence in the Sprague Dawley rat compared to the Han Wistar. 19 RPCM is characterized by a spectrum of histopathology changes that progress with increasing animal age and may show exacerbation with drug or chemical administration.19-21 Findings in young rats are usually composed of one or more scattered foci of mononuclear cells with occasional degenerate or necrotic cardiomyocytes which may progress to interstitial cell proliferation and fibrosis in aged animals.22,23 The aetiology of RPCM is not fully understood although the distribution of lesions in the acute stages has led some authors to speculate that it may involve vascular dysfunction; 19 studies in hypertensive rats may support this hypothesis. 24 Myocardial findings similar to those observed with HFO-1234ze have been reported in Sprague Dawley rats exposed to other HFCs such as HFC-245fa and HCFO-1233zd. 25 However, there were no HFO-1234ze related findings in the heart of dogs in the 39-week study or in the 13-week study in CD1 mice (for which this is also a common background finding) at any dose tested; it is important to note that the highest dose in the 13-week mouse study was 2.4-fold higher than in the 26-week rat study (29,000 mg/kg/day vs 12,000 mg/kg/day). While large animals were not routinely used in the toxicological assessment of HFCs, no cardiac findings were present in minipigs exposed to HFO-1234yf. 25 The lack of cardiac findings in mice, dogs, and minipigs strongly suggests that the rat may be uniquely sensitive to the development of structural cardiac lesions such as RPCM following exposure to very high HFC doses; the effect dose in the 6 month rat study (12,000 mg/kg/day) is approximately more than 3000-fold higher than that expected in humans with typical use of an MDI (2 × 2 inhalations per day); the lower dose tested was the NOAEL (4280 mg/kg/day) and associated with a margin of approximately 1000 times (see section on Exposure and Margins of Safety).
The provision of milk through lactation places significant metabolic demands on the dam and is associated with transient, yet notable, alterations in metabolic, hormonal, and tissue morphology. 29 Daily metabolizable energy demands in female rats during peak lactation are around 3-fold higher than those required for maintenance of non-lactating females. 30 While lactating rats compensate for this by consuming two to four times more food than non-lactating females, these animals are still in a state of negative energy balance at peak lactation. 30 Increased energy demand may drive higher respiration rates or tidal volumes in rats, which could lead to higher inhaled exposures to test materials during lactation. For example, a difference in pulmonary response to inhaled ozone has been reported for lactating rats when compared with post-lactating animals. 31 This was considered to be due to a higher total inhaled ozone dose in lactating rats, given their considerably larger tidal volume and suggests that during the lactational period, inhaled doses of HFO-1234ze may be higher than during other periods of the study.
There are a number of additional study design-aspects that may have also exacerbated the metabolic stresses in the lactating females in the 2-generation reproduction study with HFO-1234ze and other HFCs. Firstly, the effect dose, exposure concentration, and duration in the 2-generation study of 23,000 mg/kg/day (target 20,000 ppm for 6 h/day) were notably higher than the NOAEL in the 26-week rat study of 12,000 mg/kg/day (target 5000 ppm for 4 h/day). Secondly, the normal behaviors of pups and dams are significantly disrupted, based on the need to remove the dams from their pups for the duration of the inhalation exposure (in the case of HFO-1234ze study for 6 h/day, plus additional time for loading/unloading the animals from the chambers and performing routine observations). On return to the home cage, rather than feeding and watering themselves, the dams tend to feed the pups first (unpublished observation, Labcorp) leading to large periods of time when dams are unable to replenish their energy stores. Indeed, food availability (or rather lack of it) is listed as a potential additional stressor limiting the ability to cope with the extra metabolic demands in lactating animals, 32 which is not generally relevant for humans.
An absence of deaths or CNS findings in non-pregnant female and male rats exposed to HFO-1234ze at high concentrations in acute and chronic exposure scenarios adds to the weight of evidence that the CNS lesions observed in lactating rats reflect the unique energy demands of lactation. CNS lesions were not observed in mice (13 weeks) at higher doses (29,000 mg/kg/day) than the effect level in rats (23,000 mg/kg/day) or in dogs (to 39 weeks). There was no evidence for an effect on the CNS as assessed by the modified Irwin screen in the chronic rat study.
Exposures and Margins of Safety
HFO-1234ze: Summary of Safety Margins From Key Toxicology Studies.
2-Gen = 2-generation reproductive toxicology study.
aHFO-1234ze emitted dose = 58.6 mg/actuation; human doses calculated based on a body weight of 60 kg.
Relative to the doses of HFO-1234ze expected in patients following MDI use, in terms of body weight (mg/kg), very large margins of safety have been demonstrated. The NOELs and NOAELs for key studies relative to the expected dose of HFO-1234ze through MDI use are shown in Table 7.
Conclusion
As previously reported, HFO-1234ze is not genotoxic and is not associated with any effects on male or female reproduction. There were no toxicology findings in any study in mice or dogs, even at the very high doses tested. Following completion of dosing, systemic exposure in all animals was extremely short lived, with no evidence of accumulation, consistent with data for existing MDI propellants and supporting very low exposure in humans. While there were some toxicological effects observed in rats, these effects were noted at doses which far exceed those expected, or indeed achievable, in humans with use of MDI products. Effects in the nasal cavities and liver in early studies were not reproduced at high doses in long-term studies. The main effects included an exacerbation of rodent progressive cardiomyopathy and the early termination of some lactating female rats (some with CNS pathology) in a 2 generation rat study. The findings in the heart reflected an exacerbation of a widely reported background finding in rodents while the poor condition of the lactating females at excessively high doses was considered to be significantly influenced by a state of energy deprivation. Overall, these findings, and the doses at which they were observed, are considered not to reflect any risk to patients and support further development of HFO-1234ze as an MDI propellant. Indeed, the nonclinical package has been reviewed by a range of regional health authorities and considered sufficient to support the conduct of large clinical studies intended to provide long-term human safety data (ongoing).
Footnotes
Acknowledgements
The authors would like to acknowledge colleagues at Honeywell International Inc. and the department of Clinical Pharmacology and Safety Sciences at AstraZeneca for their support in the review of data and discussions throughout the lifecycle of this project, notably, Ian Robinson (retired) who provided important initial guidance. The authors would also like to acknowledge the input of the LabCorp Corporation who oversaw the conduct of the toxicology studies described in this manuscript and who provided assistance regarding clarifying questions relating to data generation and study conduct.
Author Contributions
All authors - Contributed to acquisition, analysis and interpretation, drafted manuscript.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: ApconiX (Joanne Kilgour) was paid by AstraZeneca for inhalation toxicology consulting services. PG, MJ, KT, and AAH are employees of AstraZeneca and may hold stock.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Toxicology studies reported were all funded by AstraZeneca with the exception of the 2-generation study which was funded by Honeywell International Inc.
