Abstract Background: Immunotherapies such as immune checkpoint inhibitors (ICIs) and cancer vaccines have transformed the treatment of several cancers, yet pancreatic ductal adenocarcinoma (PDAC) remains largely refractory. PDAC has a dismal 5-year survival rate of ∼13%, and therapeutic resistance has been attributed to the profoundly immunosuppressive tumor microenvironment (TME). Multiple preclinical strategies have been proposed to overcome stromal and immune barriers in PDAC, but translation to patients has not yet successfully sensitized PDAC to immunotherapy, underscoring the need for new approaches. PDAC is a disease of aging, with a median age at diagnosis of 71 and 90% of cases diagnosed after age 55. Normal biological aging alters both stromal and immune compartments: immunosenescence reduces the pool of naïve T cells and diminishes antigen-presenting cell (APC) function, while fibroblast remodeling alters paracrine signaling and stromal interactions in normal tissues. Our prior work established that aging also alters the PDAC tumor microenvironment. We found that as pancreatic fibroblasts age, they secrete ligands that directly enhance PDAC cell growth and invasion, leading to faster tumor growth and increased metastatic spread. This matches with clinical data showing that older patients with PDAC have more aggressive disease at diagnosis and inferior outcomes compared to younger patients even when controlling for all other relevant clinical factors. Our group has also recently defined age-associated differences in the circulating and tumor-infiltrating T-cell pool across patients with cancer. Older individuals harbor fewer naïve T cells that have higher baseline expression of immune checkpoints, suggesting that these aged T cells appear less capable of mounting new antigen-specific responses. We therefore hypothesize that immune aging directly limits the efficacy of cancer vaccines in older patients and that interventions restoring antigen presentation and T-cell priming may be required in patients with PDAC. Despite these observations, most preclinical PDAC models fail to incorporate the biological effects of aging, limiting translational relevance. To model this biology directly, we developed aged-host PDAC systems and observed that aging drives expression of multiple immunosuppressive cytokines in the TME. Among these, growth differentiation factor-15 (GDF-15), a stress- and aging-induced member of the TGF-β superfamily, has emerged as a candidate mediator of age-associated immune dysfunction, which may limit the efficacy of cancer vaccine therapies. Our work has established that aged pancreatic fibroblasts express high levels of GDF-15, which promotes progression of PDAC tumors. GDF-15 is known to suppress dendritic cell maturation, limit antigen presentation, and restrict CD8⁺ T-cell infiltration in other contexts. These observations raise the possibility that GDF-15 contributes to T-cell exclusion and diminished vaccine responsiveness in older hosts, which represents the overwhelming majority of patients with PDAC. Methods and Results: To interrogate how aging shapes the PDAC tumor immune microenvironment, we implanted orthotopic KPC and KPCY tumors into young (8-12 weeks) and aged (12 months) C57BL/6 mice, including cohorts with systemic knockout of GDF-15. Tumors were profiled by flow cytometry, immunohistochemistry, and imaging mass cytometry to quantify immune compartments. Across multiple PDAC models in aged hosts, tumors contained significantly fewer dendritic cells (DCs) than those in young mice, suggesting reduced antigen-presenting capacity. We also identified a significant decrease in both CD4+ and CD8+ T cell infiltration in tumors from aged compared to young mice. These deficiencies were associated with impaired vaccine immunogenicity: aged mice given a pooled mutant KRAS (mKRAS) vaccine, incorporating peptides from the six most common alleles, which are present in ∼90% of PDAC patients, mounted much weaker interferon-γ responses on ELISpot than young controls, consistent with impaired generation of new antigen-specific immunity during aging. Cytokine profiling revealed that aged PDAC tumors expressed higher levels of multiple immunosuppressive mediators, with growth differentiation factor-15 (GDF-15) emerging as a dominant stromal signal. Aged pancreatic fibroblasts expressed abundant GDF-15, and circulating GDF-15 was elevated in aged PDAC-bearing mice compared to young. Importantly, analysis of the Human Protein Atlas demonstrated that serum GDF-15 is highly elevated in patients with PDAC compared to both healthy volunteers and patients with other types of cancer, underscoring its disease relevance and the importance of age-informed PDAC models. To test GDF-15’s functional impact on the tumor immune microenvironment, we administered recombinant GDF-15 to young mice bearing PDAC tumors, which resulted in accelerated tumor growth and reduced intratumoral CD8⁺ T-cell infiltration. Conversely, GDF-15 knockout in aged mice reduced tumor growth and significantly increased CD8⁺ infiltration, providing evidence that expression of GDF-15 in the microenvironment shapes the immune landscape in PDAC. Therapeutically, treatment with a neutralizing antibody against GDF-15, currently in clinical development, significantly reduced tumor size in aged, but not young, mice, emphasizing the age-dependence of this phenotype. However, GDF-15 blockade alone did not produce complete regressions or durable control. We therefore combined GDF-15 blockade with immune checkpoint inhibition and the mKRAS vaccine to determine if we could overcome age-induced immunosuppression and sensitize tumors to immunotherapy strategies. The combination produced significant reductions in tumor growth and was associated with marked increases in neoantigen-specific T-cell responses, as measured by ELISpot assays compared to these therapies given in isolation. Conclusions and Impact: Our work demonstrates that aging reshapes the PDAC immune microenvironment through elevated GDF-15, promoting tumor growth and impairing vaccine-elicited T-cell responses and immunotherapy responses. Neutralization of GDF-15 significantly reduced tumor size in aged but not young models of PDAC, highlighting the age-dependence of this phenotype and the necessity of age-informed models to recapitulate human disease. While GDF-15 blockade alone did not yield durable tumor regressions, combinations with checkpoint inhibitors and mutant KRAS vaccination produced substantial tumor growth inhibition and robust neoantigen-specific T-cell responses, supporting a novel therapeutic strategy to overcome resistance. This body of work establishes immune aging as a barrier to immunotherapy and identifies GDF-15 as a key mediator that links stromal remodeling to impaired antitumor immunity. The findings provide immediate translational potential. Neutralizing antibodies targeting GDF-15 are in clinical development, and our data directly support their study in PDAC, particularly in older patients who shoulder the greatest disease burden. Treating aging as a biological determinant of therapy resistance may help identify additional pathways that can be therapeutically targeted. More broadly, our work provides a framework for integrating aging biology into the design of cancer immunotherapies across solid tumors, ensuring that future interventions account for the molecular features of cancer in older patients. Ultimately, this approach lays the groundwork for age-adapted immunotherapy strategies that could extend the benefits of vaccination and checkpoint blockade to patients who currently derive diminished benefit from these transformative therapies. Citation Format: Emma Kartalia, Sara Young, Edwin He, John DeBetta, Harshada Bharadwaj, Benjamin Barrett, Neeha Zaidi, Daniel J. Zabransky. Overcoming aging-driven barriers to effective immunotherapy in pancreatic cancer abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts) ; 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86 (8Suppl): Abstract nr NG09.
Building similarity graph...
Analyzing shared references across papers
Loading...
Emma Kartalia
Sara Young
Edwin He
Cancer Research
Johns Hopkins University
Building similarity graph...
Analyzing shared references across papers
Loading...
Kartalia et al. (Fri,) studied this question.
www.synapsesocial.com/papers/69e4745f010ef96374d90193 — DOI: https://doi.org/10.1158/1538-7445.am2026-ng09