Precision immunotherapy in metastatic pancreatic cancer: OPTIMIZE-1 biomarker-driven patient stratification for future studies

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal human malignancies, with a 5-year overall survival rate of 3.2 in the metastatic setting (1). Despite the adoption of mFOLFIRINOX based on the landmark 2011 FOLFIRINOX trial as the first-line standard of care in chemotherapy-naïve patients with metastatic disease, median overall survival (mOS) has plateaued at approximately 9–11 months over the past decade (2). Subsequent trials with modified regimens and combination approaches, such as NALIRIFOX and Gemcitabine/nab-Paclitaxel, have achieved comparable outcomes, with mOS ranging from 10.4 to 11.7 months, depending on the specific regimen and patient population studied (3,4). The desmoplastic, immunosuppressive tumor microenvironment (TME) of PDAC represents a fundamental barrier to therapeutic efficacy, necessitating novel approaches that reprogram tumor-associated myeloid compartments and promote antitumor immunity (5). The recent publication of the OPTIMIZE-1 biomarker analysis in Cell Reports Medicine (6) represents a critical step forward, providing compelling evidence that CD40 agonism with optimized dosing can overcome prior limitations of this therapeutic class while generating actionable biomarkers for patient selection in the planned Phase 3 confirmatory trial.

The clinical case for CD40 agonism in PDAC

CD40 is a tumor necrosis factor (TNF) superfamily receptor expressed on antigen-presenting cells (APCs), including B cells, dendritic cells (DCs), and tumor-associated macrophages. Upon ligation by a CD40 agonist, it triggers both canonical and noncanonical NF-κB signaling pathways, leading to activation of DCs, maturation of antigen presentation machinery (MHC-I/II, CD80, CD86), and critically, reprogramming of the immunosuppressive TME through M2-to-M1 macrophage polarization (7,8). Preclinical studies have demonstrated that CD40 agonism promotes stromal degradation and decreases collagen deposition, potentially enhancing chemotherapy penetration into previously inaccessible tumor regions (7).

However, earlier clinical attempts to develop CD40 agonists in PDAC have largely failed to demonstrate clinical efficacy. The PRINCE trial, which evaluated sotigalimab (another CD40 agonist) plus gemcitabine/nab-paclitaxel as first-line therapy, reported negative results (9). Several factors likely contributed to this negative outcome: suboptimal dosing, late sequencing of the CD40 agonist (administered after chemotherapy initiation rather than as a tumor-priming strategy), and limitations in identifying patients most likely to benefit from this immunomodulatory approach. OPTIMIZE-1 addressed each of these shortcomings through a thoughtfully designed trial that incorporated a tumor-priming mitazalimab dose prior to chemotherapy initiation, dose optimization to 900 µg/kg (following earlier exploration of 450 µg/kg, which proved inferior), and comprehensive biomarker analyses examining both tumor-intrinsic and immune-activated signatures (6).

OPTIMIZE-1: efficacy and safety in context

The Phase 1b/2 OPTIMIZE-1 study enrolled 70 chemotherapy-naïve patients with metastatic PDAC, of whom 57 were evaluable at the optimal dose of 900 µg/kg. The trial met its primary endpoint, with a confirmed objective response rate (ORR) of 42.1% [95% confidence interval (CI): 29.4–52.1%], exceeding the historical 31.6% threshold for mFOLFIRINOX alone (2). These comparisons to historical benchmarks should be interpreted cautiously as cross-trial comparisons are inherently limited by differences in patient selection, eligibility criteria, supportive care, and the potential enriching effect of biomarker-selected populations, and do not substitute for a randomized control arm. With that caveat the trial demonstrated durable responses and a survival benefit: mOS reached 14.9 months, with a 24-month OS rate of 29.4%, comparing favorably to the 8% historical 24-month OS for chemotherapy alone. Median progression-free survival was 7.8 months, and median duration of response was 12.6 months, suggesting that responders to this combination maintain benefit for extended periods.

The safety profile was manageable and consistent with mFOLFIRINOX expectations. The most common grade ≥3 adverse events included neutropenia (25.7%), hypokalemia (15.7%), anemia (11.4%), and thrombocytopenia (11.4%), with no treatment-related deaths. Critically, only 2 of 57 patients (3.5%) discontinued treatment due to adverse events, indicating that the combination did not introduce unmanageable toxicity relative to chemotherapy alone. These data support the feasibility of advancing mitazalimab to Phase 3 with the 900 µg/kg dose, which the FDA subsequently endorsed based on dose characterization data presented at ESMO 2025.

KRAS genotype as a predictive biomarker: hypothesis-generating stratification potential

One of the interesting exploratory findings from the OPTIMIZE-1 biomarker analysis that warrants prospective validation is the potential for differential clinical outcomes based on KRAS mutational status. Among the 57 efficacy-evaluable patients, mutational analysis revealed that KRAS G12V and KRAS G12R mutations were associated with increased response rate, and when data were combined with KRAS G12V, improvement in PFS, but there was no statistical difference, although a positive trend, in OS when the aforementioned variants were compared to KRAS G12D, contradictory to prior reports. The molecular biology underlying these exploratory findings is increasingly understood: KRAS G12D mutations preferentially activate the MAPK/ERK pathway and correlate with an immunosuppressive TME characterized by elevated regulatory T cells (Tregs) and a low Th1/Th2 ratio. Conversely, KRAS G12V and G12R variants are associated with reduced PD-L1 expression, lower checkpoint receptor expression on immune cells, and a relatively more inflammatory immune contexture (10). These biological differences have profound implications for immunotherapy responsiveness: patients with G12D mutations may require more aggressive immune priming or combinatorial approaches, whereas G12V/R-mutant tumors appear more permissive to CD40-mediated immunomodulation.

The OPTIMIZE-1 biomarker data align with recent large-scale genomic analyses demonstrating that KRAS G12R-mutant tumors have superior overall survival compared to G12D across multiple treatment modalities, and that KRAS G12V/R mutations are associated with improved outcomes following adjuvant chemotherapy (11-13). These convergent findings should be cautiously interpreted in the absence of randomized control arm and inability to determine whether these KRAS-associated differences represent true predictive biomarkers of benefit from mitazalimab or simply reflect the prognostic biology of each KRAS variant independent of the treatment. Nonetheless, these findings suggest that KRAS genotype warrant evaluation as a primary stratification variable in the planned Phase 3 trial, with potential pre-planned analyses examining efficacy separately in G12V/R-mutant versus G12D-mutant disease.

Molecular subtypes: prognostic value of the classical phenotype

Consistent with established transcriptomic classifications of PDAC, OPTIMIZE-1 evaluated the impact of molecular subtypes, specifically the classical and basal-like (or squamous) phenotypes, on treatment efficacy. The study demonstrated that patients with the classical subtype achieved significantly better clinical outcomes, with an ORR of 42.1% and mOS of 14.9 months, compared to those with basal-like tumors. These findings suggest that while mitazalimab plus mFOLFIRINOX provides robust efficacy in the classical population, the basal-like subtype remains a distinct therapeutic challenge.

Tumor-intrinsic biomarkers: baseline fibrosis signature

Beyond KRAS genotype, the OPTIMIZE-1 biomarker analysis identified a novel baseline tumor fibrosis-related gene signature associated with improved overall survival. This exploratory finding is particularly intriguing given CD40’s mechanistic role in stromal remodeling and desmoplastic fibrosis depletion. Patients whose baseline tumors expressed elevated fibrosis-related transcripts paradoxically demonstrated better responses to mitazalimab plus mFOLFIRINOX, potentially reflecting the ability of CD40 agonism to degrade and repurpose protective stromal elements into a therapeutically vulnerable state. This observation challenges the conventional assumption that extensive desmoplasia universally confers a poor prognosis (14). Rather, the data suggest a nuanced biology wherein baseline fibrosis-high tumors represent an opportunity for CD40-mediated stromal degradation, provided the immune microenvironment can be sufficiently activated. Moreover, whether this association is truly predictive of benefit from mitazalimab—as opposed to a prognostic marker of favorable tumor biology—cannot be resolved without a randomized comparator arm. Future Phase 3 biomarker analyses should incorporate quantitative transcriptomic or radiological assessments of baseline fibrosis burden to validate this signature in an independent cohort.

Immunodynamic biomarkers: evidence for CD40-mediated immune activation

The unique trial design of OPTIMIZE-1, incorporating a tumor-priming mitazalimab dose administered on days 1 and 10 of the first cycle prior to chemotherapy initiation, permitted interrogation of CD40-induced immunological changes in an unconfounded setting (i.e., before chemotherapy-mediated immune perturbations). This design revealed several interesting associations between immune cell dynamics and clinical outcomes, that warrant further investigation and validation.

Notably, increasing percentages of proliferating T cell and NK cell populations, including Ki67+ T cells, Ki67+ CD8 T effector memory cells, Ki67+ CD4 T central memory cells, and Ki67+ natural killer T (NKT) cells, were associated with longer OS. Patients demonstrating robust expansion of CD4+ effector T cells following the first mitazalimab administration achieved an mOS of 14 months compared to 8.8 months in low expanders. This finding validates CD40’s theoretical mechanism of action—ligation activates DCs to provide enhanced costimulation (via CD80/CD86 engagement of CD28) and IL-12 production, driving Th1 differentiation and CD4+ effector expansion.

Activated myeloid populations, including increased frequency of DCs and pro-inflammatory macrophages (M1 phenotype), correlated with improved survival. These changes reflected CD40-mediated reprogramming of tumor-associated macrophages from an M2 immunosuppressive phenotype toward M1 pro-inflammatory states, consistent with the proposed mechanism of action. Notably, these immunodynamic changes were evident within days of mitazalimab administration, suggesting that CD40 exerts its effects rapidly and that early circulating immune biomarkers might serve as real-time pharmacodynamic readouts of drug engagement and efficacy.

B-cell compositional changes, including increased frequency of activated (CD54+) B cells and overall changes in B-cell compartment composition, were also associated with improved survival. While the precise role of CD40-activated B cells in antitumor immunity remains incompletely understood, these associations suggest that comprehensive immune profiling—extending beyond traditional T-cell and myeloid assessments—may be necessary to fully capture CD40’s immunomodulatory effects.

Clinical implications for Phase 3 trial design and patient stratification

The OPTIMIZE-1 biomarker findings provide a roadmap for designing the confirmatory Phase 3 trial with several key recommendations:

Primary stratification by KRAS genotype

Pre-planned analysis stratifying efficacy by KRAS G12V/R versus G12D mutation status is essential. This approach would identify whether G12D-mutant patients require augmented immunomodulation (e.g., addition of a checkpoint inhibitor like pembrolizumab or nivolumab) to achieve meaningful benefit; and generate actionable information for real-world clinical practice, enabling oncologists to counsel patients on genotype-dependent treatment efficacy.

Secondary stratification by baseline tumor fibrosis

This can be achieved radiologically (e.g., T2-weighted imaging, diffusion-weighted imaging, or computed tomography-derived texture analysis) or by incorporating circulating biomarker-based assessments of baseline fibrosis burden. This would validate whether the OPTIMIZE-1 finding of improved outcomes in fibrosis-high patients is reproducible and might identify a particularly enriched population for CD40 agonism.

Real-time immunodynamic biomarker monitoring

Incorporation of early post-first-dose mitazalimab circulating biomarkers (CD4 effector T-cell expansion, activated myeloid populations) could enable early identification of patients unlikely to benefit and facilitate adaptive trial designs or companion diagnostics in clinical practice. Sequential circulating immune profiling at defined time points (e.g., baseline, day 2–4 post-first mitazalimab dose, post-first chemotherapy, post-second cycle) would generate mechanistic insights and identify potential resistance mechanisms.

ctDNA as a surrogate efficacy marker

The OPTIMIZE-1 finding that circulating tumor DNA (ctDNA) clearance correlates with improved survival warrants prospective validation in Phase 3. ctDNA assessment is a non-invasive, real-time measure of treatment response and may inform decisions on treatment continuation, intensification, or switching earlier than radiological imaging.

Translating biomarkers into clinical practice

While OPTIMIZE-1 provides compelling evidence for CD40 agonism in PDAC, several translational gaps require bridging before routine clinical implementation:

Practical KRAS genotyping

Routine reflex KRAS genotyping (not merely KRAS mutation status) of metastatic PDAC samples should be incorporated. This may require updating institutional pathology protocols and ensuring access to next-generation sequencing (NGS) platforms capable of distinguishing G12 codon substitutions. Implementing such protocol is currently facing many challenges including: turnaround times for comprehensive NGS can range from days to weeks at academic centers and may be substantially longer in community settings; adequate tissue availability from biopsies or surgical specimens may be insufficient for comprehensive molecular profiling; insurance coverage and reimbursement for extended KRAS subtype genotyping remain inconsistent and institution-dependent; and access to NGS infrastructure is geographically uneven, with community oncologists frequently lacking in-house capability and relying on commercial laboratories with variable quality standards. These barriers must be addressed systematically if biomarker-driven trial enrollment and clinical practice integration are to be feasible at scale.

Companion biomarker development

Prospective clinical trials should explore whether simple, scalable biomarkers (e.g., cfDNA-based KRAS genotyping, peripheral immune profiling, or imaging-derived fibrosis quantification) might be developed as rapid, inexpensive companion diagnostics to guide patient selection and monitoring. However, the path from promising biomarker signal to validated companion diagnostic is long and requires careful attention to assay standardization and harmonization across participating sites—a particular challenge for immune profiling assays such as flow cytometry or mass cytometry, which are highly sensitive to pre-analytical variables. Radiological fibrosis assessment methods [e.g., diffusion-weighted magnetic resonance imaging (MRI), computed tomography (CT)-texture analysis] similarly require protocol standardization before they can function as reproducible clinical tools.

Combination strategies for G12D-mutant disease

The modest activity of mitazalimab monotherapy in G12D-mutant PDAC raises the question of whether rational combinations might improve outcomes. Addition of a checkpoint inhibitor (anti-PD-1/PD-L1 antibody) is biologically plausible, given that CD40 drives effector T-cell expansion while potentially upregulating PD-L1 on DCs and tumor cells. Alternatively, TGF-β pathway inhibition (galunisertib or similar agents) might synergize with CD40 agonism to overcome the immunosuppressive stromal environment characteristic of G12D-driven tumors. Phase 1b trials exploring these combinations in selected G12D patients are warranted.

Study limitations

Several important limitations of the OPTIMIZE-1 biomarker data must be acknowledged before the findings discussed in this commentary can be placed in clinical context. First, the study enrolled only 70 patients, of whom 57 were evaluable at the optimal dose, representing a small sample size with limited statistical power for subgroup analyses. Second, the multiple correlative biomarker analyses conducted across KRAS subtypes, molecular phenotypes, immune cell dynamics, fibrosis signatures, and ctDNA raise concerns about false-positive findings due to multiple testing; none of these associations have been corrected for multiplicity and none should be considered validated. Third, without a concurrent control arm, it is not possible to distinguish biomarkers that are truly predictive of differential benefit from mitazalimab (i.e., predictive biomarkers) from those that simply identify patients with more favorable tumor biology who would do better on any treatment (i.e., prognostic biomarkers). Fourth, biomarker instability over the course of treatment, assay variability between laboratories, and limited generalizability of results from a specialized academic cohort to broader real-world populations further constrain the immediate clinical utility of these findings. These limitations underscore the imperative to treat all biomarker associations from OPTIMIZE-1 as hypothesis-generating signals rather than practice-defining evidence.

Concluding remarks

The OPTIMIZE-1 biomarker analysis represents an important, albeit early and exploratory, step in immunotherapy development for pancreatic cancer. This study demonstrates that rigorous trial design, comprehensive biomarker collection, and detailed correlative analysis can generate hypothesis-driving signals regarding patient populations and mechanisms of action to inform next-generation therapeutic strategies. The observed differential efficacy across KRAS genotypes, combined with evidence for rapid CD40-mediated immune activation and baseline tumor-intrinsic predictive signatures, provides a potential foundation for biomarker integration into the Phase 3 trial—with the essential caveat that all of these findings require prospective, pre-specified validation in a randomized setting. Critically, these results may inform future patient stratification strategies, supporting a precision medicine approach in which KRAS genotype and baseline tumor characteristics guide treatment selection. For patients with KRAS G12V/R-mutant metastatic PDAC, the OPTIMIZE-1 data are hypothesis-generating and suggest that mitazalimab plus mFOLFIRINOX warrants evaluation as a potential first-line option. In contrast, for those with G12D-mutant disease, the OPTIMIZE-1 data support the investigation of rationally designed combination regimens that utilize CD40 agonism as an immunopriming strategy within multimodal immunotherapy platforms.

If a randomized Phase 3 trial is planned, it should prospectively incorporate KRAS genotyping and comprehensive biomarker monitoring to validate these findings and establish the immunological predictors that translate between OPTIMIZE-1 and real-world clinical populations. Success in these efforts would represent not merely an incremental therapeutic advance but rather a paradigm shift toward molecularly informed, immune-driven treatment of a disease that has historically resisted therapeutic innovation. However, it will be interesting to see how the development of multi-select RAS (ON) and selective RAS inhibitors will change the paradigm in the treatment of patients with pancreatic adenocarcinoma. With the revolutionary discovery of these effective molecules and new data demonstrating immune activation of the TME, it can rationalize investigating the combination of CD40 agonists, such as mitazalimab, and KRAS inhibitors specifically in the immunosuppressive KRAS G12D variant. The development of CCR8 inhibitors with positive preclinical results and early clinical efficacy in pancreatic adenocarcinoma can also be combined with other immunotherapeutic molecules, such as CD40 agonists, specifically in KRAS G12D, to augment immune activation. We are very excited about the burgeoning development of efficacious novel molecules to fight this lethal disease with poor prognosis and look forward to the future for the betterment of our patients.

Acknowledgments

H.M.B. acknowledges support awarded under the K-12 NCI grant program (K12CA090628).

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Pancreatic Cancer. The article has undergone external peer review.

Peer Review File: Available at https://apc.amegroups.com/article/view/10.21037/apc-26-0014/prf

Funding: This work was supported by the K-12 NCI grant program (K12CA090628).

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://apc.amegroups.com/article/view/10.21037/apc-26-0014/coif). H.M.B. serves as an unpaid section editor of Annals of Pancreatic Cancer from January 2026 to December 2027. H.M.B. received research funding from Spirita Oncology (Inst), Novocure (Inst), AstraZeneca (Inst), JSI (Inst), Incyte (Inst), Qurient (Inst), HiFiBiO Therapeutics (Inst), Revolution Health Care (Inst), Elevation Oncology (Inst), Dragonfly Therapeutics (Inst), Zelbio (Inst), BMS (Inst), Mirati Therapeutics (Inst), and Strategia (Inst). H.M.B. received consulting fee from Endocyte, Celgene, Idera, Myovant Sciences, Novocure, Ipsen, Caris MPI, Incyte, and Guardant Health. The other author has no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Kratzer TB, Wagle NS, et al. Cancer statistics, 2026. CA Cancer J Clin 2026;76:e70043. [Crossref] [PubMed]
2. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med 2011;364:1817-25. [Crossref] [PubMed]
3. Wainberg ZA, Melisi D, Macarulla T, et al. NALIRIFOX versus nab-paclitaxel and gemcitabine in treatment-naive patients with metastatic pancreatic ductal adenocarcinoma (NAPOLI 3): a randomised, open-label, phase 3 trial. Lancet 2023;402:1272-81. [Crossref] [PubMed]
4. De Dosso S, Siebenhüner AR, Winder T, et al. Treatment landscape of metastatic pancreatic cancer. Cancer Treat Rev 2021;96:102180. [Crossref] [PubMed]
5. Li Y, Wang J, Wang H, et al. The Interplay Between Inflammation and Stromal Components in Pancreatic Cancer. Front Immunol 2022;13:850093. [Crossref] [PubMed]
6. Van Laethem JL, Geboes K, Borbath I, et al. CD40 agonist mitazalimab with mFOLFIRINOX in untreated metastatic pancreatic cancer: Biomarkers associated with outcomes from OPTIMIZE-1. Cell Rep Med 2025;6:102407. [Crossref] [PubMed]
7. Beatty GL, Chiorean EG, Fishman MP, et al. CD40 agonists alter tumor stroma and show efficacy against pancreatic carcinoma in mice and humans. Science 2011;331:1612-6. [Crossref] [PubMed]
8. O'Hara MH, O'Reilly EM, Varadhachary G, et al. CD40 agonistic monoclonal antibody APX005M (sotigalimab) and chemotherapy, with or without nivolumab, for the treatment of metastatic pancreatic adenocarcinoma: an open-label, multicentre, phase 1b study. Lancet Oncol 2021;22:118-31. [Crossref] [PubMed]
9. Padrón LJ, Maurer DM, O'Hara MH, et al. Sotigalimab and/or nivolumab with chemotherapy in first-line metastatic pancreatic cancer: clinical and immunologic analyses from the randomized phase 2 PRINCE trial. Nat Med 2022;28:1167-77. [Crossref] [PubMed]
10. Ardalan B, Ciner A, Baca Y, et al. Distinct Molecular and Clinical Features of Specific Variants of KRAS Codon 12 in Pancreatic Adenocarcinoma. Clin Cancer Res 2025;31:1082-90. [Crossref] [PubMed]
11. Ardalan B, Ciner A, Baca Y, Darabi S, et al. Prognostic indicators of KRAS G12X mutations in pancreatic cancer. J Clin Oncol 2023;41:735.
12. Varghese AM, Perry MA, Chou JF, et al. Clinicogenomic landscape of pancreatic adenocarcinoma identifies KRAS mutant dosage as prognostic of overall survival. Nat Med 2025;31:466-77. [Crossref] [PubMed]
13. Norton C, Shaw MS, Rubnitz Z, et al. KRAS Mutation Status and Treatment Outcomes in Patients With Metastatic Pancreatic Adenocarcinoma. JAMA Netw Open 2025;8:e2453588. [Crossref] [PubMed]
14. Wang LM, Silva MA, D'Costa Z, et al. The prognostic role of desmoplastic stroma in pancreatic ductal adenocarcinoma. Oncotarget 2016;7:4183-94. [Crossref] [PubMed]