Clinicopathological factors related to survival in advanced pancreatic cancer in elderly patients at the American-British Cowdray Medical Center.

Abstract

701 Background: Although the majority of pancreatic cancer (PC) patients worldwide are elderly, this population remains underrepresented in clinical trials. Understanding the clinicopathological factors related to oncological outcomes in older adults is essential, particularly the impact of the geriatric profile, comprehensive geriatric assessment, dose adjustments, and chemotherapy (CT) de-escalation. Methods: Retrospective cohort study at the American British Cowdray Medical Center, Mexico City, including patients ≥65 years with histologically confirmed metastatic or unresectable pancreatic cancer diagnosed between 2014–2025. Clinical, pathological, and geriatric factors at diagnosis were analyzed. Survival was assessed with Kaplan-Meier curves and log-rank tests. Results: Were identified 196 patients with PC, of whom 127 (65%) were aged ≥65 years; 58% were women. The median age was 73 years (IQR 68–79), 6 (4.7%) had locally advanced/unresectable disease, and 121 (95.3%) metastatic disease. The median body mass index (BMI) was 22 (IQR 22–25). Only 5.5% underwent an initial geriatric assessment. Polypharmacy was present in 63.5%, 40.9% were classified as frail, 80% had at least one comorbidity, 1.6% had fall syndrome, 10.2% underwent next-generation sequencing, and 22% had BRCA status determined. Most patients (94.5%) received first-line CT: 46% with FOLFIRINOX, 18.9% with gemcitabine plus nab-paclitaxel, 25.2% with gemcitabine monotherapy, and 16% with other regimens. Initial dose adjustment was performed in 30.7% of patients, and 25% underwent chemotherapy de-escalation at treatment initiation. The median follow-up was 6 months. Median PFS was 4 months (IQR 1–8), and mean OS was 13 months (SD 8–17). The ORR to first-line CT was 20%, and 57% (n=73) experienced clinical benefit. Univariate analysis showed that the type of first-line CT was not associated with differences in ORR (p=0.66) but was significantly associated with toxicity (p=0.001). Patients with unresectable disease had significantly better OS compared to those with metastatic disease (p=0.01), while low BMI was associated with poorer PFS (p=0.04). First-line chemotherapy (p=1.00 for OS, p=0.58 for PFS), chemotherapy de-escalation (p=0.58 for OS, p=0.28 for PFS), and dose adjustment (p=0.99 for OS, p=1.00 for PFS) did not impact in OS or PFS; however, these factors did significantly affect ORR (p=0.03). Conclusions: Most PC patients are elderly, and the majority received first-line CT with clinical benefit. The choice of first-line, dose adjustments, or de-escalation did not significantly affect OS or PFS; however, it did influence the ORR and toxicity profile. Geriatric assessment remains underutilized. This study offers important insights into the characteristics of elderly pancreatic cancer patients within a Mexican population.