Negative prognostic implications of splenomegaly in nivolumab-treated advanced or recurrent pancreatic adenocarcinoma.

Immune checkpoint inhibitors have limited efficacy in the treatment of pancreatic ductal adenocarcinoma (PDAC). We investigated prognostic markers for nivolumab-based therapy in advanced or recurrent PDAC. Consecutive patients receiving nivolumab-based therapy at our institution between 2015 and 2020 were evaluated. Overall survival (OS) was analyzed through univariate and multivariate analyses. Spleen volume was estimated from the width, thickness, and length of the spleen. A total of 45 patients were identified. Biweekly nivolumab was administered as monotherapy (n = 5) or in combination with chemotherapy or targeted therapy (n = 40). Among 31 evaluable patients, the response and disease control rates were 7% and 36%, respectively. The baseline median spleen volume was 267 (110-674) mL. Patients with spleens ≥267 mL had significantly shorter median OS (1.9 months, 95% confidence interval [CI], 1.0-2.7) than did those with smaller spleens (8.2 months, 95% CI, 5.6-10.8; P = .003). In the multivariate analysis, spleen volume of <267 mL, ≤2 lines of prior chemotherapy, ECOG performance status of 0-2, add-on nivolumab with stable disease after prior therapy, concomitant or sequential cell therapy, high lymphocyte count, and total bilirubin <1 mg/dL were independent favorable prognostic factors for OS. In the control groups of patients receiving gemcitabine-based chemotherapy (n = 142) or FOLFIRINOX regimen (n = 24), spleen volume exhibited no prognostic significance. In heavily pretreated PDAC, a large spleen may predict poor OS following nivolumab-based immunotherapy. Studies with larger cohorts should confirm the prognostic value of spleen volume.