Predicting clinical benefit response after neoadjuvant chemotherapy in locally advanced gallbladder cancer: retrospective analysis.

BACKGROUND

Neoadjuvant chemotherapy is increasingly used in patients with locally advanced gallbladder cancer (LAGBC). This study investigated factors affecting clinical benefit response (CBR) to neoadjuvant chemotherapy for LAGBC.

METHODS

All consecutive patients with LAGBC following neoadjuvant chemotherapy, from January 2013 to December 2022, were analyzed for clinical and radiological responses as well as survival outcomes. CBR rates, resectability, and their impact on survival were evaluated. In addition, factors predicting CBR were identified and a predictive nomogram model was developed.

RESULTS

Of 401 patients with LAGBC undergoing neoadjuvant chemotherapy, 303 (75.5%) exhibited a CBR. The median overall survival (OS) in patients with a CBR was 25 months, compared with 8.5 months for those without a CBR. Factors predicting a worse CBR rate included age ≥ 55.5 years (hazard ratio (HR) 2.17; 95% confidence interval (c.i.) 1.29 to 3.65), Eastern Cooperative Oncology Group (ECOG) performance status ≥ 1 (HR 2.5; 95% c.i. 1.117 to 5.59), platelet count ≥ 468 × 109/l (HR 2.86; 95% c.i. 1.12 to 6.74), tumour (T) size ≥ 2.1 cm (HR 3.4; 95% c.i. 1.70 to 6.80), T stage ≥ T3 (HR 3.26; 95% c.i. 1.22 to 8.74), and a systemic immune-inflammation index (SII) ≥ 1265.90 (HR 2.34; 95% c.i. 1.27 to 4.30). Of the patients with a CBR, 86% underwent curative surgical resection, with median OS improved to 29.54 months, compared with 11.86 months for those without resection (P < 0.01).

CONCLUSION

A CBR was achieved in 75.5% of patients, with curative surgical resection in 86%. A CBR was associated with improved OS. Anatomical (T size, T stage) and immune-inflammation markers (platelet count, SII) were found to predict a CBR, and could help identify responders to neoadjuvant chemotherapy. This could have implications for treatment strategies, but requires validation in further prospective studies.