Effects of combining durvalumab with chemotherapy in individuals suffering from advanced biliary tract cancer: an investigation into the prognostic nutritional index, systemic immune-inflammation index and patient survival outcomes.

Objective

To evaluate the dynamic effects of durvalumab plus chemotherapy on the prognostic nutritional index (PNI) and systemic immune-inflammation index (SII) in patients with advanced biliary tract cancer (BTC), and to develop a nomogram for individualized prognostic assessment.

Methods

Clinical data from 417 patients with advanced BTC treated between October 2019 and October 2022 were retrospectively analyzed. Patients were assigned to a training cohort (n=271) and a validation cohort (n=146) based on treatment timeline. The study group received durvalumab combined with gemcitabine/cisplatin, while the control group received gemcitabine/cisplatin monotherapy. Alterations in PNI and SII were compared between the two groups. Short-term antitumor response was assessed according to the Response Evaluation Criteria in Solid Tumors version 1.1. Survival analysis was performed using the Kaplan-Meier method and Cox proportional hazards regression. A prognostic nomogram was constructed and validated via bootstrap resampling.

Results

Post-treatment PNI was significantly higher, whereas post-treatment SII was markedly lower in the study group (both P<0.001). The study group achieved a higher objective response rate (P=0.003), with more prominent clinical benefits observed in patients with low post-treatment SII. Median overall survival (OS) was substantially prolonged in the study group (21.00 vs 13.00 months; hazard ratio (HR)=0.555, 95% confidence interval (CI): 0.408-0.755, P<0.001). Independent prognostic factors for OS included treatment regimen, tumor stage, ECOG performance status (PS), pretreatment PNI, pretreatment SII, and carbohydrate antigen 19-9. The established nomogram exhibited favorable discriminative ability and satisfactory internal calibration in the training cohort, with a bias-corrected Harrell's concordance index (C-index) of 0.758 and a calibration slope of 0.917. In the validation cohort, acceptable discrimination was retained (C-index=0.711), whereas calibration efficacy declined, with a calibration slope of 0.677, suggesting overestimation of survival probability among high-risk patients.

Conclusion

Compared with chemotherapy alone, the combined therapy ameliorated patients' nutritional status, mitigated systemic inflammatory response, and prolonged survival. The nomogram yielded acceptable discriminative power and favorable internal calibration but achieved suboptimal external calibration. Further model recalibration and prospective multicenter validation are warranted prior to its widespread clinical application.