Prognostic factors associated with survival in patients with pancreatic cancer treated on early phase immune-checkpoint inhibitor clinical trials.

Abstract

577 Background: Immune-checkpoint inhibitors (ICI) have altered the treatment landscape of various solid tumors, but clinical outcomes in patients with pancreatic ductal adenocarcinoma (PDAC) continue to remain dismal. Data are needed to determine prognostic factors of survival to ICI-therapies in pancreatic cancer to better inform future recruitment in immunotherapy-based early clinical trials. Methods: This is a single center, retrospective analysis of patients with PDAC treated with ICI-based therapy as part of phase I trials at MD Anderson Cancer Center. Patients enrolled in a phase I study and treated between 2014 to 2021 were included in the analysis. Descriptive analysis included patient characteristics, progression free survival (PFS) and overall survival (OS). Cox proportional hazards were used to assess associations between patient or tumor characteristics and survival endpoints. Results: One hundred and twenty-two patients were included in this study and received a median of two prior lines of therapy (IQR 1-3). The most commonly utilized ICI were PD-L1 inhibitors (56%) and PD-1 inhibitors (34%). Most patients (96%) received ICI in combination with another investigational agent and 9% of patients received ICI in combination with chemotherapy. The median duration of treatment was 1.38 months (IQR 0.69-2.56). The median PFS was 1.81 months (95% CI 1.78-2.04) and the median OS was 4.83 months (95% CI 4.08-5.92). Univariate analysis showed improved PFS in patients with an albumin of ≥3.5 (1.94 vs 1.32 months, HR 0.29, 95% CI 0.14-0.60, p = 0.001) and Hgb ≥10.5 (1.87 vs 1.81 months, HR 0.63, 95% CI 0.4-0.98, p = 0.039); a longer OS was demonstrated in patients who had lung only metastases (7.92 vs. 4.18 months, HR 0.4, 95% CI 0.21-0.76, p = 0.006), albumin ≥3.5 (5.52 vs. 1.68 months, HR 0.4, 95% CI 0.2-0.8, p = 0.010), and a Hgb ≥10.5 (5.29 vs 4.11 months, HR 0.59, 95% CI 0.38-0.93, p = 0.022). Patients with an ANC/ALC ratio of ≥5 had a shorter OS compared to patients with an ANC/ALC ratio of < 5 (3.98 vs 6.54 months, HR 1.78, 95% CI 1.17-2.69, p = 0.007). Multivariate analysis showed that an albumin ≥3.5 predicted improved PFS (HR 0.41, 95% CI 0.19-0.90, p = 0.026) and an ANC/ALC ratio of ≥5 predicted shorter OS (HR 2.59, 95% CI 1.59-4.19, p < 0.001). Conclusions: Results of this large, single center retrospective study demonstrate that most patients with PDAC treated with ICI based therapies as a part of early phase clinical trials experience disease progression shortly after initiation of therapy and have a limited overall survival. Pretreatment albumin ≥3.5 predicted improved PFS and ANC/ALC ratio ≥5 predicted shorter OS.