Introduction: The Challenge of the Cold Tumor
Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal malignancies worldwide, characterized by a dismal five-year survival rate and a notorious resistance to conventional and modern therapeutic interventions. While immune checkpoint inhibitors (ICIs), such as those targeting the PD-1/PD-L1 axis, have revolutionized the treatment of various solid tumors, they have largely failed to produce meaningful clinical responses in PDAC. This resistance is primarily attributed to the unique, highly immunosuppressive tumor microenvironment (TME) of pancreatic cancer, which is often described as a “cold” tumor due to its dense desmoplastic stroma and lack of infiltrating effector T cells.
To overcome this barrier, researchers have turned to combinatorial strategies designed to prime the TME for immunotherapy. One promising target is Heat Shock Protein 90 (HSP90), a molecular chaperone that stabilizes various oncogenic proteins. Preclinical models have suggested that HSP90 inhibition can limit the activation of cancer-associated fibroblasts (CAFs) and enhance T cell infiltration, potentially sensitizing PDAC to PD-1 blockade. A recent Phase Ib/II trial investigated this hypothesis using the HSP90 inhibitor XL888 in combination with pembrolizumab.
Study Design and Methodology
The study was a single-center, open-label, nonrandomized, dose-escalation trial focused on an expansion cohort of 16 patients with advanced, metastatic PDAC. The primary objective was to evaluate the safety and clinical efficacy of the combination therapy, while secondary objectives included translational immune profiling to identify measurable changes in the TME and peripheral blood.
Dosing and Administration
Patients were enrolled in a 21-day treatment cycle. During the first cycle, patients received either pembrolizumab (200 mg intravenously) as a monotherapy or in combination with XL888 (90 mg orally, twice per week). Following the initial cycle, all patients crossed over to the combination therapy. This design allowed researchers to compare the early immunological effects of the combination against the checkpoint inhibitor alone.
Correlative Studies
To understand the biological impact of the treatment, the research team collected peripheral blood mononuclear cells (PBMCs) and performed image-guided liver biopsies at baseline (C1D1) and on-treatment (C1D15). These samples underwent extensive profiling, including cytokine analysis and flow cytometry, to track changes in immune cell populations and signaling pathways.
Clinical Results: Safety and Efficacy
The combination of XL888 and pembrolizumab was found to be generally well-tolerated. The safety profile was consistent with the known side effects of the individual agents, and no unexpected grade 3 or 4 adverse events occurred that would necessitate trial cessation. This confirms that HSP90 inhibition can be safely combined with anti-PD-1 therapy in this patient population.
Survival and Response Data
Despite the manageable safety profile, the clinical efficacy was disappointing. Of the 15 evaluable patients, none achieved an objective response (0% ORR). Two patients (13.3%) achieved stable disease, while the vast majority (86.7%) experienced rapid disease progression. The median progression-free survival (PFS) was a mere 2.0 months, and the median overall survival (OS) was 4.4 months. These figures align closely with the natural history of refractory metastatic PDAC, suggesting that the addition of XL888 provided no significant clinical benefit over standard care or pembrolizumab alone.
Translational Insights: A Disconnect Between Blood and Tumor
The most intriguing findings of the trial came from the translational immune profiling, which revealed a stark contrast between systemic immune activation and the local tumor environment.
Systemic Immune Modulation
Analysis of peripheral blood showed that the combination therapy did indeed induce biological changes. Patients receiving XL888 and pembrolizumab exhibited increased levels of circulating Th1-associated cytokines and chemokines, which are typically associated with an active immune response. Furthermore, PBMC analysis revealed an elevation in terminal effector CD8+ T cells in the combination arm compared to the monotherapy arm. However, there was also a concomitant increase in CD4+ regulatory T cells (Tregs), which may have served to dampen any potential anti-tumor immune response.
The TME Barrier
Crucially, the liver biopsies (representing the TME of metastatic sites) showed no significant changes in immune cell infiltration or stromal composition regardless of the treatment group. This suggests that while the drugs were active in the systemic circulation, they failed to penetrate or effectively modulate the dense, immunosuppressive environment of the pancreatic cancer metastases. The “cold” nature of the PDAC TME remained unchanged, explaining the lack of clinical response despite the systemic increase in effector T cells.
Expert Commentary and Clinical Implications
The failure of the XL888 and pembrolizumab combination highlights a recurring theme in pancreatic cancer research: the disconnect between preclinical success and clinical reality. In mouse models, HSP90 inhibitors successfully remodeled the stroma and allowed T cells to flood the tumor. In human patients, however, the desmoplastic reaction is far more complex and robust.
The increase in regulatory T cells (Tregs) observed in the peripheral blood may also be a critical factor. If the therapy increases both effector cells and suppressive cells simultaneously, the net effect on the tumor may be neutral or even detrimental. Future trials may need to consider “triple-threat” strategies that combine stromal remodeling, checkpoint blockade, and specific depletion of regulatory immune subsets.
Furthermore, the lack of change in liver biopsies underscores the necessity of developing drugs with better tissue penetration or those that can more effectively target the specific signaling pathways utilized by human CAFs. The study also reinforces the importance of using paired biopsies in early-phase trials to confirm whether a drug is reaching its target and achieving the intended biological effect within the tumor itself.
Conclusion
In summary, the Phase Ib/II trial of XL888 and pembrolizumab in metastatic PDAC demonstrated that while the combination is safe and capable of inducing systemic immune modulation, it does not translate into clinical efficacy or meaningful changes in the tumor microenvironment. For clinicians, these results serve as a reminder that the barriers to immunotherapy in pancreatic cancer are multi-faceted and cannot be overcome by HSP90 inhibition alone at the doses tested. Future research must continue to explore more potent ways to “heat up” the PDAC TME and ensure that systemic immune activation can be successfully directed into the tumor tissue.
References
1. Horvat NK, Diab M, Phillips MJ, et al. A phase Ib/II trial of XL888 (HSP90 inhibitor) and pembrolizumab in metastatic pancreatic cancer with translational immune profiling. Cancer Lett. 2025;639:218233. doi:10.1016/j.canlet.2025.218233.
2. Royal RE, Levy C, Turner K, et al. Phase 2 trial of single agent ipilimumab (anti-CTLA-4) for locally advanced or metastatic pancreatic adenocarcinoma. J Immunother. 2010;33(8):828-833.
3. Feig C, Jones JO, Kraman M, et al. Targeting CXCL12 from FAP-expressing carcinoma-associated fibroblasts synergizes with anti-PD-L1 immunotherapy in pancreatic cancer. Proc Natl Acad Sci U S A. 2013;110(50):20212-20217.
