Abstract
Pancreatic ductal adenocarcinoma (PDAC) remains a leading cause of cancer-related death with a rising incidence in younger individuals and no standard early detection exam. Kirsten rat sarcoma viral oncogene homolog (KRAS) is the most frequently mutated gene of the RAS family, and KRAS mutations are found in approximately 85%–90% of PDAC. Long considered undruggable due to its molecular structure, the advent of sotorasib and adagrasib has ushered in multiple novel therapeutics targeting the RAS pathway, including mutation-selective, pan-KRAS, and pan-RAS inhibitors. Combination strategies using chemotherapy and directed against EGFR, SOS1, SHP2, and immune pathways, among others, aim to overcome resistance to RAS inhibitor monotherapy and enhance the depth and duration of clinical benefit. The minority of PDAC tumors that are KRAS-wildtype are enriched for rare, actionable alterations, including Neuregulin-1, HER2, and BRAF amenable to targeted treatments. For patients with metastatic disease carrying a germline BRCA1/2 or PALB2 pathogenic variant, PARP inhibitors remain an option as maintenance treatment after achieving at least stable disease with platinum-based chemotherapy. Importantly, ongoing and future clinical trials are shifting the use of targeted therapies from the refractory setting to earlier lines and the perioperative setting with promising results. Here, we detail the different areas of investigation targeting KRAS and other precision-based therapies in PDAC, as well as the potential emerging roles of local interventions (radiation, surgery) for select patients with oligometastatic disease. Composite predictive biomarkers using genomic, proteomic, and radiographic factors are needed to refine and individualize treatment selection and ultimately improve patient outcomes.
Introduction
Pancreatic ductal adenocarcinoma (PDAC) remains a formidable cancer with an almost invariably poor prognosis. In the United States, the incidence of PDAC has been increasing by 0.5%–1.0% each year, and PDAC is estimated to become the second leading cause of cancer-related mortality by 2030. 1 The growing incidence has been greatest in younger patients (under 55 years of age), independent of multiple risk factors, including demographics (age, black race, male sex), tobacco use, and comorbidities such as obesity, periodontitis, precancerous mucinous lesions, chronic pancreatitis, diabetes mellitus, Helicobacter pylori infection, and germline predisposition. 2 Most patients present with locally advanced or metastatic disease, and those with resectable tumors commonly recur despite the use of perioperative chemotherapy. With currently approved therapies, the 5-year survival remains 10%–15%, and there remains a large unmet need for precision-based therapies to improve outcomes.
Kirsten rat sarcoma viral oncogene homolog (KRAS) is the most frequently mutated gene of the RAS family, and KRAS mutations are found in approximately 85%–90% of PDAC. A mutation in KRAS promotes the development of the PDAC precursor lesion, pancreatic intraepithelial neoplasia (PanIN). 3 Subsequent acquisition of mutations in the tumor suppressor genes, CDKN2A, TP53, and SMAD4 allow for progression from PanIN to PDAC.3 –5 Patients with KRAS-wildtype (WT) PDAC have a better prognosis, with a longer overall survival (OS) compared to those with KRAS-mutant disease. 4 Other RAS homologs within the RAS family include NRAS and HRAS, and mutations of these genes are exceedingly rare (<1%) in PDAC.
While chemotherapy remains the mainstay of systemic treatment, precision-based therapeutics are playing an increasing role in the management of patients with PDAC. When given to suppress activation of redundant parallel signaling pathways, small molecules and other inhibitors can prevent or delay resistance to chemotherapy. 6 Herein, we review the current state of RAS inhibitors and other targeted therapies and modalities in PDAC (Table 1). Chemical structures of certain drugs described are included in Appendix 1.
Select enrolling trials of targeted therapies in PDAC.
KRAS, Kirsten rat sarcoma viral oncogene homolog; NRG1, neuregulin-1; PDAC, pancreatic ductal adenocarcinoma; VSV, vesicular stomatitis virus.
KRAS: A diverse and druggable mutation
The KRAS protein exists in two configurations. In the active or (ON) state, KRAS is bound to guanosine triphosphate (KRAS-GTP) and leads to activation of multiple downstream signaling pathways, including RAF-MEK-ERK and PI3K-AKT-AKT-mTOR, which regulate cell proliferation and differentiation. To prevent uncontrolled activation of these pathways, KRAS hydrolyzes its GTP to guanosine diphosphate (GDP) by way of the GTPase-activating protein (GAP), converting KRAS to its inactive or (OFF) state (KRAS-GDP). The guanine nucleotide exchange factor exchanges GDP for GTP, changing KRAS back to its active state. When KRAS is mutated, GAP activity is impaired, leading to constitutive activation of the KRAS pathway, tumor initiation, and progression. 7
About 90% of KRAS mutations in PDAC occur at codon 12, with the remainder of KRAS alterations comprising G13D or Q61X mutations and KRAS amplification. The WT KRAS sequence is GGT encoding glycine. The most common mutation subtype in PDAC is KRAS G12D, which is present in 40% of tumors. The G12V subtype refers to a GTT sequence producing valine and is present in 33% of PDAC. KRAS G12R refers to a CGT replacement sequence producing arginine, which is present in 15% of PDAC tumors and is more unique to pancreatic cancer as compared to other cancers.3,8 KRAS G12C mutations replace glycine with cysteine and are the rarest in codon 12, accounting for 1.7% of KRAS mutations in PDAC.
KRAS mutation subtypes have been shown to be prognostic among patients with PDAC. In a real-world cohort, patients whose tumors harbored KRAS G12R mutations had longer median OS compared to those with KRAS G12D-mutant PDAC. No significant difference was noted in outcomes between patients with KRAS G12R, G12V, or G12C-mutant PDAC. 9 Other studies are examining the impact of specific KRAS mutations on the benefit from chemotherapy and other systemic treatments.
KRAS-specific therapies in oncology have long been difficult to develop due to the intrinsic characteristics of the KRAS protein.4,9,10 Specifically, the native KRAS protein has a smooth surface, without a groove or pocket easily amenable to drug binding. 4 Furthermore, KRAS has a very high affinity for GTP, which makes competitive inhibition challenging. 4 Importantly, there are multiple upstream and downstream collateral pathways that can circumvent KRAS inhibition and mediate treatment resistance. 4 Indeed, initial attempts to block KRAS signaling in PDAC, including blocking EGFR and MEK, were not successful. 4
KRAS-directed therapies can be classified based on the specific mutations targeted and their mechanism of action. While the two currently approved small molecule inhibitors, sotorasib and adagrasib, are mutation-selective, multiple other agents in development are pan-KRAS or pan-RAS inhibitors. Furthermore, while most inhibitors block KRAS in the (OFF) or GDP-bound inactive state, others target the (ON) or GTP-bound state. Mechanistically, RAS inhibitors bind the protein to block its activity, while RAS protein degraders mark it for ubiquitination and proteasome-mediated degradation.
Mutation-selective KRAS-directed therapies
KRAS G12C
The KRAS G12C mutation involves a glycine-to-cysteine substitution at codon 12. 11 This leads to a conformational change in the GAP binding site, which prevents GAP-mediated GTP hydrolysis and results in constitutive KRAS pathway activation. 4
Sotorasib, an oral KRAS G12C inhibitor, selectively and irreversibly inhibits this protein, thereby blocking downstream signaling and proliferation. 11 In the phase I/II CodeBreak 100 clinical trial, 38 patients with metastatic pancreatic cancer previously treated with a median of two prior lines of therapy received sotorasib, with a recommended phase II dose of 960 mg. The confirmed overall response rate (ORR) was 21%, with a median progression-free survival (PFS) of 4.0 months and a median OS of 6.9 months. 12 Sotorasib had an acceptable safety profile, with any grade treatment-related adverse events (trAEs) occurring in 42% of patients, and the most frequent grade 3 trAEs being diarrhea and fatigue (5% each).
Similarly, adagrasib is a covalent KRAS G12C inhibitor that irreversibly binds KRAS G12C in its inactive state. In the KRYSTAL-1 clinical trial, adagrasib was evaluated at a dose of 600 mg twice daily in patients with KRAS G12C mutant advanced solid tumors, including 21 patients with PDAC. 13 The ORR in this cohort was 33.3%, with a median PFS of 5.4 months and a median OS of 8.0 months. Across all patients, the most common any-grade trAEs were nausea (49.2%), diarrhea (47.6%), fatigue (41.3%), and vomiting (39.7%). The most common grade 3 trAEs were fatigue (6.3%) and QT prolongation (6.3%). 14
Based on the clinical benefit seen in these studies, sotorasib and adagrasib were integrated into the National Comprehensive Cancer Network guidelines for patients with previously treated metastatic PDAC harboring the KRAS G12C mutation.
Glecirasib (JAB-21822) is another covalent KRAS G12C inhibitor being developed by Jacobio Pharma (Beijing, China) which recently received FDA orphan designation for PDAC. In a pooled analysis of two phase I/II trials, which included 32 patients with PDAC, the ORR was 46.9% and the disease control rate (DCR) 93.8%, with a median PFS of 5.5 months and a median OS of 10.8 months. 15 Among all patients, the most common any grade trAEs were anemia, bilirubin increase, aspartate aminotransferase (AST) increase, and asthenia; the most common grade 3 trAE was anemia. A related single-arm, multicenter phase II study is evaluating the safety and efficacy of glecirasib monotherapy in patients with advanced PDAC (NCT06008288). Another phase I/II study is combining glecirasib with sitneprotrafib (JAB-3312), a potent oral SHP2 inhibitor, in patients with KRAS G12C-mutant solid tumors (NCT05288205). Preliminary data support that this combination is well-tolerated and has promising efficacy with regard to both overall response and DCRs.
Olomorasib (LY3537982) is a second-generation, oral GDP-bound KRAS G12C inhibitor that has demonstrated marked potency in preclinical models and subsequent efficacy in clinical trials. In KRAS G12C mutant cell lines, olomorasib inhibited growth with greater efficiency than sotorasib and adagrasib, which is possibly due to high target occupancy at low exposures. 16 Early phase clinical trials have since demonstrated the safety, tolerability, and efficacy of olomorasib. In a phase I/II study of olomorasib monotherapy, which included 24 patients with PDAC, the ORR was 35% and the median PFS was 7.1 months. Patients with gastrointestinal tumors, including colorectal cancer (CRC), PDAC, and biliary tract cancers, were treated with olomorasib monotherapy or in combination with cetuximab (CRC). Frequent trAEs included diarrhea, dermatitis acneiform, dry skin, hypomagnesemia, and fatigue. 17 Preliminary results from this study demonstrate a tolerable safety profile and efficacy in gastrointestinal malignancies. 17
In contrast to the first and second-generation covalent inhibitors, elironrasib (RMC-6291) is an oral KRAS G12C (ON) inhibitor that utilizes a tri-complex configuration composed of the drug, intracellular chaperone protein cyclophilin A (CypA), and KRAS G12C in the GTP-bound active state (NCT061285). 18 This platform, developed by Revolution Medicines, remodels the surface of CypA to create an interface with high affinity and selectivity for the active state of the KRAS protein, leading to target protein inactivation upon drug binding. 19 In preclinical studies, elironrasib achieved durable regression and response, especially in comparison to adagrasib, a KRAS G12C (OFF) inhibitor. 18 Preliminary data from a phase I clinical trial demonstrated promising anti-tumor activity in patients with non-small cell lung cancer previously treated with a KRAS G12C (OFF) inhibitor, as well as in patients with CRC who had no prior exposure to the KRAS G12C (OFF) inhibitor. 18 Adverse effects included nausea (34%), diarrhea (30%), QTc prolongation (21%), and vomiting and fatigue (15% each). 18 Grade 3 adverse events subsequently resolved after dose reduction or dose hold, indicating a tolerable safety profile. 18
KRAS G12D
The KRAS G12D mutation is the most common KRAS mutation seen in PDAC, present in about 40% of patients, and is associated with worse prognosis across disease stages. 8 In one retrospective cohort study, the presence of KRAS G12D was associated with shorter survival in patients with resectable PDAC. 8 The three leading KRAS G12D-specific small molecule drugs in development are: setidegrasib (ASP3082), zoldonrasib (RMC-9805), and HRS-4642.
Setidegrasib (ASP3082) represents a first-in-class KRAS G12D selective protein degrader known as a proteolysis targeting chimera or PROTAC. 20 PROTACs consist of a ligand for the protein of interest (POI), a ligand for an E3 ligase, and a linker, which function by inducing the formation of a ternary complex with POI/PROTAC/E3 ligase, which ultimately inhibits all functions of the target protein. 21
In the first-in-human phase I study, setidegrasib was administered to 98 patients, including 67 patients with previously treated KRAS G12D mutant PDAC at doses of 10–300 mg given as a weekly infusion. 20 The maximum tolerated dose (MTD) was not reached, and the predicted lowest dose for clinical efficacy was >100 mg. The most recent trAEs were fatigue (15.3%), infusion-related reactions (14.3%), pruritus (9.2%), nausea (7.1%), urticaria (7.1%), AST increase (7.1%), alanine aminotransferase (ALT) increase (6.1%), and vomiting (5.1%). At a dose of 300 mg, three of seven patients with PDAC had a partial response, and the DCR was 75%. The study is ongoing, examining setidegrasib in combination with FOLFIRINOX and gemcitabine/nab-paclitaxel in patients with PDAC (NCT05382559), and a phase III trial is planned.
Zoldonrasib (RMC-9805) is another RAS(ON) inhibitor and has demonstrated tumor regression in PDAC with the KRAS G12D mutation in preclinical and early phase trials. This KRAS G12D-specific orally bioavailable small molecule functions by forming a stable, high-affinity tri-complex between the drug, active KRAS G12D, and CypA, resulting in suppression of downstream KRAS signaling and durable tumor regression in preclinical cancer models of PDAC. 22 In an ongoing phase I study (NCT06040541), 104 patients with PDAC were enrolled, and among response-evaluable patients, the ORR was 30% and DCR 80%. 23 Notably, most patients demonstrated marked reductions in KRAS G12D variant allele frequency in circulating tumor DNA (ctDNA), confirming on-target effects. In addition to the monotherapy cohort, the study is evaluating the combination of zoldonrasib and daraxonrasib (RMC-6236), a multi-selective (G12X, G13X, Q61X) RAS(ON) inhibitor.
HRS-4642 is another KRAS G12D-directed therapy that is distinguished by its non-covalent binding and preclinical synergy with proteasome inhibitors. 24 In preclinical studies, HRS-4642 demonstrated regression of PDAC xenograft models, and genome-wide CRISPR-Cas9 screening revealed the proteasome as a sensitization target. Indeed, adding the proteasome inhibitor, carfilzomib, improved the antitumor efficacy of HRS-4642. This combination helped make the tumor microenvironment more immune-permissive, reflected by increased CD4/CD8 T-cells and M1 macrophages. 25 In a phase I study of HRS-4642, which included 24 patients with PDAC, the confirmed ORR was 20.8% and DCR 79.2%, with complete clearance of KRAS G12D ctDNA associated with response24,26; (NCT05533463). HRS-4642 is administered intravenously in a liposomal formulation and given weekly. The most frequent any-grade trAEs included hypercholesterolemia, anemia, AST/ALT increase, and neutropenia.
Another area of KRAS G12D-directed drug development involves immunotherapies. The phase Ib KISIMA-02 trial is exploring a heterologous prime boost vaccine both independently and in conjunction with the PD-1 inhibitor, ezabenlimab (NCT05846516). KISIMA-02 (ATP150/ATP152/ATP162) is a platform comprising cell penetrating peptide for antigen delivery; a domain of multiple tumor-associated antigens, and a toll-like receptor (TLR) agonist for self-adjuvanticity. 27 In combination with an oncolytic vesicular stomatitis virus with or without ezabenlimab, KISIMA-02 is being administered in patients with locally advanced or metastatic PDAC, as well as resectable PDAC after surgery (NCT05846516), and has been shown to provoke a potent anti-tumor immune response.
Another phase I/II clinical trial is focusing on peripheral blood lymphocytes transduced with a murine T-cell receptor that specifically recognizes the KRAS G12D mutation in HLA-A*11-01 positive patients (NCT03745326).
Emerging therapeutic options also include exosomes that are specifically designed to target actionable mutations. For example, one study is evaluating mesenchymal stem cell-derived exosomes loaded with small interfering RNA (siRNA) against KRAS G12D in metastatic pancreatic cancer, known as iExosomes (NCT03608631). Preclinical studies in murine models have demonstrated that these exosomes were able to effectively deliver siRNA targeting KRAS G12D, ultimately leading to tumor control. 28 In the phase I study, there were no dose-limiting toxicities (DLTs) or MTD identified in 12 patients with metastatic PDAC. Stable disease was observed in 6 of 12 patients. Immune profiling suggests synergy with checkpoint inhibitors, including anti-CTLA4 antibodies. 29
KRAS G12V
The KRAS G12V mutation is present in about 32% of PDAC tumors. 8 In one pathologic analysis, KRAS G12V-mutant PDAC was associated with more frequent papillary/large duct rather than conventional histology, and papillary/large duct histology was associated with a significantly longer OS than tumors with conventional histology. 30 In contrast, in a correlative ctDNA analysis conducted in patients with metastatic PDAC undergoing first-line chemoimmunotherapy, ctKRAS G12D was a negative prognostic marker for OS; however, no significant association between ctKRAS G12V and outcomes was detected. 31
One new technology targeting the KRAS G12V mutation utilizes peripheral blood leukocytes transduced with anti-KRAS G12V murine T-cell receptors to mediate tumor regression (NCT05933668). Clinical trials are also underway to determine the safety of administering peripheral blood leukocytes transduced with anti-KRAS G12V murine T-cell receptors simultaneously with lymphodepletion and high-dose interleukin-2 (NCT03190941). Another clinical trial is exploring T-cell receptor transduced T-cell therapy for advanced pancreatic cancer patients who are HLA-A*11:01 positive, with the postulation that mutant KRAS (mKRAS) antigen-specific T-cell receptor transduced autologous T-cells will specifically target HLA-matched mKRAS cancer cells (NCT04146298).
EFTX-G12V is a first-in-class siRNA inhibitor of KRAS G12V that is uniquely delivered through the EGFR receptor to improve tumor-to-normal payload delivery. 32 EFTX-G12V was found to significantly silence KRAS G12V mRNA and suppress PDAC growth in cell lines and spheroids. 32 In preclinical in vivo models, EFTX-G12V was found to outperform pan-KRAS siRNA due to superior suppression of EGFR signaling, MAPK/ERRK activation, and tumor angiogenesis. 32 First-in-human clinical trials of this novel compound are anticipated.
KRAS G12R KRAS G12R involves a glycine-to-arginine substitution, occurs in about 15% of PDAC, and primarily affects downstream signaling of the KRAS pathway. KRAS G12R is different from KRAS G12C and G12V because it fails to engage with downstream PI3Ka. Upregulation of KRAS independent of PI3Ka compensates for PI3Ka deficiency and macropinocytosis and is seen specifically in PDAC. KRAS G12R mutant tumors show limited sensitivity to MAPK/ERK inhibitor monotherapy but have demonstrated sensitivity to combined autophagy and MAPK/ERK inhibition. 33
While the KRAS G12R mutation is found in a notable proportion of patients with PDAC, there are a few clinical trials underway specifically targeting this variant. These therapies include small molecules such as selumetinib (AZD6244), which inhibits the MEK1/2 protein within the MAPK/ERK signaling pathway, thereby inhibiting cell line proliferation and increasing apoptosis, and is already approved for use in children with neurofibromatosis 1 with inoperable neurofibromas. 34
Pan-K/RAS directed therapies
In addition to mutation-selective KRAS-directed agents, pan-KRAS inhibitors, a class of therapeutics with broader inhibition, offer a promising strategy to overcome emerging resistance to more targeted KRAS drugs. 35 Current pan-K/RAS inhibitors are being investigated both as monotherapy or in conjunction with other therapies and are being explored in both the advanced and peri-operative settings. 35 To date, the most promising and developed novel therapeutics in this class are the oral RAS(ON) inhibitor, daraxonrasib (RMC-6236), and the cancer vaccine, ELI-002.
Daraxonrasib (RMC-6236) is a broad-spectrum RAS(ON) covalent tri-complex inhibitor with activity against KRAS G12X, G13X, Q61X, and RAS-WT variants as well as resistance mechanisms, including secondary RAS mutations. This compound demonstrated profound tumor regression in mouse models of different tumor types with diverse KRAS mutations, and this efficacy has been borne out in clinical trials. In a phase I/Ib study, 127 patients with RAS-mutant metastatic PDAC were treated with escalating doses (160–300 mg QD) of daraxonrasib in the second-line setting. 36 The ORR was 36% and 27% in the KRAS G12X-mutant and broader RAS-mutant population at the 300-mg dose, respectively, with DCRs of 91% and 95%. Among patients with KRAS G12X-mutant PDAC, the median PFS was 8.5 months, and the median OS was 14.5 months. The most common any-grade trAEs were rash (91%), diarrhea (48%), nausea (43%), vomiting (31%), stomatitis (31%), fatigue (20%), paronychia (13%), mucosal inflammation (13%), decreased appetite (11%), and peripheral edema (10%). 36 Over 90% of evaluable patients had at least 50% reduction in RAS mutant ctDNA. 36 As a result of these findings, daraxonrasib received FDA orphan drug designation for pancreatic cancer. Building on this clinical efficacy, the global, randomized, phase III RASolute 302 trial is evaluating daraxonrasib in the second-line setting compared to chemotherapy in patients with metastatic PDAC (NCT06625320). Other studies are examining daraxonrasib in the adjuvant setting for resected PDAC, first-line metastatic PDAC, and in combination with different chemotherapy regimens in multiple gastrointestinal cancers (NCT06445062).
ELI-002 is a lymph-node targeted cancer vaccine comprising two components: mKRAS peptides targeting G12X or G13X (2 or 7 peptide formulations) and a CpG DNA adjuvant, which serves as a potent TLR9 agonist. Both components are attached to an albumin-binding lipid for lymph node targeting. 37 In the phase I AMPLIFY-201 trial, 20 patients with pancreatic cancer who completed locoregional therapy received the ELI-002 2P (2 peptide) vaccination, including mKRAS amphiphile-peptide antigens (G12D, G12R) and amphiphile-adjuvant, CpG-7909. Clinical benefit was significantly associated with mKRAS-specific T-cell responses above or below a threshold 9.17-fold increase over baseline, with median radiographic relapse-free survival not reached versus 3.02 months (hazard ratio (HR) = 0.12, p = 0.0002) and median OS not reached versus 15.98 months (HR = 0.23, p = 0.0099). 37 The majority of patients (71%) induced both CD4+ and CD8+ T-cell responses. Notably, antigen spreading was also observed, with increased T-cells reactive to personalized, tumor antigens absent from the initial ELI-002 2P vaccine.
Another novel immunotherapy employs an autologous chimeric antigen receptor-engineered dendritic cell (CAR-DC)-based cancer vaccine and is being evaluated in combination with anti-PD1 and anti-CTLA4 immune checkpoint inhibitors (ICIs; NCT05631899). This CAR-DC vaccine targets EphA2, CD8 T-cells, and dendritic cells and is loaded with KRAS-mutant peptides. EphA2 belongs to one of the largest families of tyrosine kinase receptors and mediates oncogenic cell–cell interactions, cell migration, and angiogenesis. 38 EphA2 expression has been shown to be a negative prognosticator in PDAC, and preclinical models have shown PDAC tumor regression and improved chemotherapy delivery with EphA2 inhibition. 38
Other small molecule pan-RAS inhibitors include YL-17231 (NCT06096974), which is currently in phase I dose-escalation trials in patients with solid tumors harboring mutations in KRAS, HRAS, or NRAS, as well as JYP0015, an oral small molecule pan-RAS inhibitor that is currently in phase I and II trials in PDAC and other advanced solid tumors. BGB-53038 is currently in phase I trials and has selectivity for KRAS mutations and amplifications over HRAS or NRAS alterations, with the aim of optimizing efficacy and limiting toxicity (NCT06585488).
Other investigational agents are directed at broader MAPK inhibition downstream of RAS. For example, IMM-1-104 inhibits MEK phosphorylation downstream of the RAS protein and has demonstrated efficacy and tolerable safety in phase I clinical trials (NCT05585320). 39 BDTX-4933, a BRAF inhibitor that has the potential to broadly inhibit RAF activation downstream of activated MAPK signaling via aberrant KRAS signaling, is also under investigation (NCT05786924). 40 This molecule has thus far demonstrated effective inhibition of ERK phosphorylation in mouse models across various BRAF targets and is currently in phase I clinical trials. 40
Mechanisms of resistance to KRAS inhibition and the role of combination strategies
While KRAS-directed therapies have shown promise, emerging resistance is inevitable. For instance, resistance to sotorasib and adagrasib has been observed in patients with KRAS G12C-mutated PDAC, postulated due to acquired PIK3CA and KRAS mutations. 41 In a study examining in vitro models of acquired KRAS mutations, concordant copy number gain or amplification of CDK6 or ABCB1A/B (drug efflux proteins) and KRAS were found to be associated with resistance to KRAS inhibition. 41 Other causes of resistance to KRAS inhibition have been identified and are related to activation of collateral pathways. One such mechanism is the conversion of KRAS G12C into a drug-insensitive state, mediated through constitutive EGFR activation. 42 Furthermore, KRAS G12D inhibition has been shown to reprogram the microenvironment of early and advanced pancreatic cancer, promoting FAS-mediated killing by CD8+ T cells. 43 Overcoming therapeutic resistance can be accomplished by strategic sequencing of regimens to optimize synergy, synchronously targeting multiple collateral pathways or targeting acquired resistance pathways. As such, combination regimens that include EGFR inhibition and immunotherapy in conjunction with KRAS inhibitors are being evaluated in clinical trials as a means of achieving more durable clinical benefit. 42
Targeting RAS-WT disease
RAS is a family of genes with three isoforms (HRAS, KRAS, and NRAS) in which activating mutations are responsible for oncogenesis in multiple solid tumors. 43 Approximately 10%–15% of PDACs bear the WT variant of the KRAS gene, 44 and these tumors tend to be enriched for other actionable alterations. In a single-institution analysis of 785 PDAC tumors studied via a targeted multigene sequencing panel, 73 patients (9.2%) had KRAS-WT pancreatic cancer. Within this cohort, 43.8% had activation of an alternative signaling pathway, such as the MAPK pathway, including BRAF mutations, as well as activating alterations in other oncogenes such as GNAS, MYC, PIK3CA, and CTNNB1. 44 Others have shown the actionability of HER2 overexpression in PDAC. 45 Also significant in PDAC is the association with downregulation of tumor suppressor genes, such as TP53, CDKN2A, and SMAD4; although one single-center study found a lower incidence of loss-of-function mutations in TP53 in KRAS WT tumors than in KRAS mutant tumors. 44 Multi-selective targeting higher up in the RAS oncogene pathway has been postulated as a potential therapeutic strategy.
EGFR-directed therapies
PDAC RAS-WT disease has been the focus of numerous clinical trials. These include agents already approved for use in other types of solid cancers, such as nimotuzumab. A phase III clinical trial evaluated OS in patients with locally advanced or metastatic PDAC treated with nivolumab in combination with gemcitabine (NCT02395016). Results for this study indicate that patients treated with nimotuzumab-gemcitabine combination therapy had statistically significant longer OS than those treated with gemcitabine alone (10.9 vs 8.5 months, HR = 0.5, confidence interval (CI): 0.06–0.94, p = 0.025), as well as longer PFS (4.2 vs 3.6 months, HR = 0.56, 95% CI: 0.12–0.99, p = 0.013), 46 indicating the promise of combination treatments in KRAS WT PDAC. This study had two primary limitations: the inclusion of both patients with locally advanced and metastatic disease and the use of gemcitabine monotherapy as the chemotherapy backbone, rather than more commonly used regimens of either gemcitabine/nab-paclitaxel or FOLFIRINOX (or NALIRIFOX).
Neuregulin-1-directed therapies
Neuregulin-1 (NRG1) belongs to a family of physiological ligands that contain an EGFR-like domain, which mediates binding to cellular receptors involved in the development of the nervous and cardiovascular systems. 47 NRG1 was first described in its role in breast cancer oncogenesis and has since been identified in several other solid tumors, including pancreatic cancer. 47 NRG1 rearrangements play a role in oncogenesis via pathologic activation of the PI3K-AKT and MAPK signaling pathways, leading to continued cell proliferation. 48 NRG1 fusions result in activation of the ErbB-mediated pathway, leading to the development of targeted therapeutics aimed at ErbB inhibition in affected solid tumors.47,48 Most NRG1 fusion partners contain a transmembrane domain near its primary receptor, the human epidermal growth factor receptor (HER)3/ERBB3. 49 In PDAC, implicated NRG1 fusion genes include CD74-NRG1, ROCK1-NRG1, ATP181-NRG1, APP-NRG1-APP, SARAF-NRG1-CHD6, CDH1-NRG1, VTCN1-NRG1. The overall incidence of NRG1 gene fusions is rare and estimated to be around 0.5%–3% of PDAC.47,48
Zenocutuzumab (MCLA-128) is a bispecific antibody that targets HER2 and HER3 as well as NRG1 binding and has demonstrated marked efficacy. In the registrational phase II study, 99 patients with NRG1 fusion-positive tumors were enrolled and received zenocutuzumab as an intravenous infusion every 2 weeks. There was an objective response in 15 of 36 patients with pancreatic cancer (ORR 42%), with an overall median duration of response of 11.1 months and median PFS 6.8 months. 50 The treatment was well-tolerated, with less than 5% of participants experiencing grade 3 or higher adverse events. 50 The most common trAEs included diarrhea (18%), fatigue (12%), nausea (11%), and infusion-related reactions (14%). Based on these findings, on December 4, 2024, the FDA granted zenocutuzumab accelerated approval for patients with previously treated advanced, unresectable, or metastatic PDAC. 51
Seribantumab is a human monoclonal antibody against HER3, which blocks the ligand-dependent activation of HER3 and HER3-HER2 dimerization, subsequently inhibiting PI3K/Akt signaling (PMID: 36455193). There are reported cases of clinical benefit with this agent in patients with PDAC (NCT04790695), including one patient with metastatic disease that had progressed on FOLFIRINOX and gemcitabine/nab-paclitaxel who achieved a serologic and radiographic response. 49
Beyond RAS-mutation status: Clinically meaningful pathways in PDAC
Immunotherapy: Checkpoint inhibitors, mRNA vaccines
While successful in other cancers, ICIs have had limited efficacy and use in PDAC. In the small subset of patients with microsatellite instability high (MSI-H) PDAC, ICIs may be clinically beneficial. 52 It has been estimated that about 1%–2% of PDAC tumors are MSI-H, making this a rare entity compared to MSI-H colorectal, endometrial, small bowel, and gastric cancers. 53
A multi-institutional study of 16 centers of patients with MSI-H PDAC treated with ICI demonstrated that ICI are well-tolerated and effective in patients with MSI-H PDAC. In this study, of the 31 MSI-H PDAC patients identified, 25 received single-agent anti-PD-1 antibodies, 3 received combination nivolumab and ipilimumab, and 3 received immunotherapy in combination with chemotherapy, of which nearly half (15, or 48.4%) had an objective response to therapy and 6 (19.4%) and stable disease, with objective response defined as complete response (present in 3 patients) and partial response (present in 12 patients). 54 Checkpoint inhibitors continue to be explored in PDAC in clinical trials in combination with other agents (Table 1).
Second-generation CTLA-4 antibodies have shown promise in cancer types that have traditionally responded poorly to immunotherapy. One such drug is botensilimab, an Fc-enhanced anti-CTLA-4 antibody, which has been demonstrated to enhance the FcyR-dependent response on antigen-presenting cells, thus enhancing T-cell recruitment and anti-tumor activity. 55 In tumor models of refractory PDAC comprising epithelial cancer cells and cancer-associated fibroblasts, anti-CTLA-4 therapy performed better than traditional chemotherapy. 55 This provided the impetus for a randomized phase II clinical trial of gemcitabine/nab-paclitaxel with or without botensilimab in second-line PDAC, which has completed enrollment (NCT05630183).
While initial vaccine studies failed to ultimately demonstrate clinical benefit, there has been a reemergence of novel and promising vaccine therapies. Recent studies have demonstrated that certain long-term survivors of PDAC harbor neoantigens that can stimulate T-cells, providing an individualized target for therapy. 56 The mRNA neoantigen vaccine, autogene cevumeran, is one adjuvant therapy that delivers neoantigens to tumor cells, helping induce neoantigen-specific T-cell responses. 56 In a study evaluating the safety and efficacy of autogene cevumeran in patients with surgically resectable PDAC, cevumeran, in combination with atezolizumab and subsequent modified FOLFIRNOX chemotherapy, generated a neoantigen-specific response in 50% of the patients enrolled, with persistence of the vaccine-expanded T-cells for up to 2 years despite post-vaccination FOLFIRINOX treatment. 56 An ongoing phase II study is randomizing patients with resected PDAC to modified FOLFIRINOX with or without atezolizumab and autogene cevumeran (NCT05968326).
Claudin 18.2
Claudin 18.2 is transmembrane tight junction protein which regulates epithelial cell polarity, adhesion, and permeability and is implicated in carcinogenesis of multiple gastrointestinal tumors, including PDAC. 57 In PDAC, claudin 18.2 is frequently overexpressed and expression has been associated with histologic features and OS. 57 Importantly, understanding the level of tumor Claudin 18.2 expression necessary for effective clinical intervention varies with each novel therapeutic and is an area of active investigation.
Zolbetuximab is an IgG1 monoclonal antibody directed against Claudin 18.2, which mediates tumor cell death via antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity, and is approved for advanced gastric cancer. 58 The phase II GLEAM study randomized 393 patients with metastatic claudin 18.2 positive PDAC (defined as ⩾75% of tumor cells demonstrating moderate-to-strong membranous CLDN18 staining using IHC) to gemcitabine/nab-paclitaxel with or without zolbetuximab as first-line therapy. Unfortunately, the study did not meet the primary endpoint of OS. 59
Spevatamig (PT886) is a bispecific antibody directed against Claudin 18.2 and CD47. Spevatamig works through both ADCC by natural killer (NK) cells and antibody-dependent cellular phagocytosis by macrophages. Targeting CD47 blocks the “don’t eat me” signal, enhancing macrophage anti-tumor activity, and may have a better safety profile than CD47 monotherapy. Spevatamig (PT886) is currently being evaluated in patients with PDAC as both monotherapy and in combination with gemcitabine/nab-paclitaxel in the front-line setting (NCT05482893). 60
Satricabtagene autoleucel (“satri-cel,” CT041) is an autologous CAR T-cell product against Claudin 18.2. In results from the phase I study, Satri-cel demonstrated safety and feasibility as well as clinical activity among patients with heavily pre-treated metastatic Claudin 18.2-positive gastric and pancreatic cancers. 61 Satri-cel was examined as monotherapy and in combination with anti-PD1 therapy. No dose-limiting toxicities, treatment-related deaths, or immune effector cell-associated neurotoxicity syndrome were reported, and all instances of cytokine release syndrome were grade 1–2. The ORR and DCR in all 98 patients were 38.8% and 91.8%, respectively, and the median PFS and OS were 4.4 months (95% CI: 3.7–6.6) and 8.8 months (95% CI: 7.1–10.2), respectively. 61
DNA repair pathways
Approximately 5%–9% of patients have a pathogenic germline or somatic mutation in a DNA repair gene, including BRCA or PALB2. 62 In this subset of patients with germline BRCA-mutated metastatic pancreatic cancer, the POLO trial demonstrated that active maintenance of olaparib after at least 16 weeks of first-line platinum-based chemotherapy had no statistically significant OS benefit compared to placebo. However, there were statistically significant benefits for other key endpoints, such as time to treatment discontinuation and time to first and second subsequent therapies, suggesting that olaparib in BRCA-mutated PDACs is clinically beneficial in these patients. 63
Building on this strategy are two key clinical trials adding PD1 to PARP inhibition in the maintenance setting: POLAR and SWOG 2001. POLAR enrolled three cohorts of patients with metastatic PDAC who achieved stable disease on induction platinum-based chemotherapy: (A) homologous recombination (HRD) mutations (e.g., BRCA 1/2, PALB2); (B) non-core HRD mutations (e.g., CHEK2 and ATM); (C) no HRD but an exceptional response to platinum-based chemotherapy. Patients were treated with olaparib and pembrolizumab. In data presented at ESMO 2024, 63 total patients have been enrolled, with the best outcomes seen in cohort A of patients with HRD-positive tumors (n = 33). Among these patients, the ORR was 35%, DCR 90%, median PFS 8.2 months, and median OS not reached. 64 The ongoing SWOG 2001 phase II trial is randomizing patients with metastatic PDAC and germline BRCA 1/2 mutations to olaparib with or without pembrolizumab 65 (NCT04548752).
Shifting to the adjuvant setting, the ongoing APOLLO (EA2192) double-blind phase II trial is randomizing patients with resected PDAC and a pathogenic BRCA 1/2 or PALB2 mutation in a 2:1 fashion to 1 year of olaparib or placebo after completion of ⩾3 months of curative intent adjuvant chemotherapy. 66 The primary endpoint is relapse-free survival; OS is a secondary endpoint (NCT04858334).
Beyond systemic therapy: Role of locoregional modalities for advanced disease
Emerging radiation techniques as targeted therapy
The EXTEND trial, a phase II multicenter randomized basket trial for patients with various solid tumors, including PDAC, evaluated whether the addition of metastasis-directed radiotherapy to systemic therapy was superior compared to systemic therapy alone. 67 The primary endpoint was PFS, with an exploratory endpoint of systemic immune response measures. Metastasis-directed therapy in this context included local radiotherapy, and patients with five or fewer sites of metastasis were eligible. The study demonstrated that patients who received locoregional radiotherapy targeted at metastatic lesions in addition to systemic therapy had a significantly longer PFS compared to those who received systemic therapy alone (10.3 vs 2.5 months, HR 0.43, p = 0.030). Furthermore, radiotherapy resulted in immune activation as evidenced by activated CD25+ T-cells and NK-type CD56+ T-cells, as well as increased T-cell receptor repertoire diversity and chemotaxis signaling, which were significantly associated with improved PFS and OS. EXTEND serves as a landmark study as the first randomized trial to examine the role of metastasis-directed therapy in oligometastatic PDAC and suggests a role for radiation in carefully selected patients. 67
Surgical resection in patients with oligometastatic PDAC
Oligometastatic pancreatic cancer is a unique biologic entity. Definitions across studies vary with respect to what qualifies as oligometastatic disease including synchronous versus metachronous, number/size of lesions, number of organs involved, site of disease (liver, peritoneum, lung, etc.), Ca 19-9 levels, response to chemotherapy, and performance status. 68 Patients with oligometastatic disease, particularly lung metastases, have a tendency toward improved prognosis, and this is thought to be due to more favorable tumor biology marked by fewer driver mutations and a less aggressive tumor phenotype. 68 Multiple small single-arm retrospective and randomized studies have demonstrated a survival benefit to metastatectomy in patients with lung, liver, or peritoneal metastases. However, the role of metastatectomy remains to be defined, and decisions should be made in the context of multidisciplinary and goals of care discussions and ideally in a clinical trial.
To this end, multiple randomized studies are ongoing, utilizing different chemotherapy regimens, evaluating the role of resection of oligometastatic disease. 68 Among them, METAPANC is the first randomized phase III clinical trial examining the role of curative intent surgical resection of oligometastatic pancreatic cancer, enrolling in multiple centers in Europe and America. 69 In this study, patients with oligometastatic liver-limited PDAC, defined as no more than three liver metastases, will undergo eight cycles of induction mFOLFIRINOX and then be randomized to resection of the primary tumor along with resection or ablation of all liver metastases versus standard of care systemic therapy alone. All patients who do not demonstrate disease progression will undergo 12 cycles of mFOLFIRINOX. The primary endpoint is OS. Secondary endpoints include safety and health-related quality of life. Importantly, tissue, blood, and fecal samples will be examined to identify potential biomarkers.
Conclusion
PDAC remains a leading cause of cancer-related mortality. Traditionally, systemic treatment options have been largely limited to cytotoxic chemotherapy. We are now beginning to scratch the surface in the era of precision therapeutics for pancreatic cancer. Although KRAS G12C inhibitors are furthest along in development, targeting KRAS G12D, G12V, and G12R represents the greatest unmet need in PDAC. To this end, clinical trials of mutation-selective or pan-K/RAS directed agents are actively changing the landscape of PDAC therapy. In KRAS-WT PDAC, actionable alterations along multiple pathways, including DNA repair, RAF, HER2, and NRG1, lend themselves to existing and novel treatments. Furthermore, emerging targets, including Claudin 18.2, promise to expand the reach of what is actionable. Immunotherapy is currently reserved for the rare MSI-high PDAC tumors, but second-generation checkpoint inhibitors, various vaccines, and cellular-based therapies could soon be integrated into the standard of care. In patients with oligometastatic PDAC, locoregional treatments with radiation and/or surgery are being increasingly considered in select patients and may synergize with immunotherapy and other systemic agents. Ultimately, composite predictive biomarkers, combination strategies, and shifting targeted approaches to earlier stages of disease are needed to optimize outcomes in patients with pancreatic cancer.
