Therapeutic Advances in Gastroenterology (May 2023)

The efficacy of immune checkpoint inhibitors in biliary tract cancer with KRAS mutation

  • Sun Young Jeong,
  • Jung Yong Hong,
  • Joon Oh Park,
  • Young Suk Park,
  • Ho Yeong Lim,
  • Jae Yeon Jang,
  • Youngkyung Jeon,
  • Seung Tae Kim

DOI
https://doi.org/10.1177/17562848231170484
Journal volume & issue
Vol. 16

Abstract

Read online

Background: With a 15% incidence, KRAS is one of the most common mutations in biliary tract cancer (BTC) and is a poor prognostic factor. Immune checkpoint inhibitors (ICIs) as salvage therapy have modest activity in BTC. Objectives: There are limited data on the efficacy of ICIs according to KRAS mutation in BTC. We evaluated the efficacy of ICIs in BTC patients with or without KRAS mutations. Design: Retrospective observational study. Methods: We conducted molecular profiling in BTC patients who received ICIs as salvage therapy. The expression of programmed death ligand 1 (PD-L1) on tumor cells was assessed using immunohistochemistry. The TruSight TM Oncology 500 assay from Illumina was used as a cancer panel. We analyzed overall survival (OS) and progression-free survival (PFS) of ICI in BTC patients according to KRAS mutation and PD-L1 expression. Results: A total of 62 patients were included in this analysis. The median age was 68.0 years; 47 patients (75.8%) received pembrolizumab and 15 (24.2%) received nivolumab as salvage therapy. All patients received gemcitabine plus cisplatin as the frontline therapy, and 53.2% received fluoropyrimidine plus oxaliplatin (FOLFOX) before ICI. The median number of lines of prior chemotherapy was 2.5. The KRAS mutation was found in 13 patients (19.1%), and 28 patients (45.2%) showed 1% or more of tumor cells out of visible tumor cells positive for PD-L1. There was no statistical correlation between KRAS mutation and PD-L1 expression. The median OS and PFS with ICI were 5.6 [interquartile range (IQR): 3.3–8.0] and 3.8 (IQR: 3.0–4.5) months, respectively. There were no statistically significant differences in PFS with ICIs according to KRAS mutation (mutant type versus wild type) and PD-L1 expression (positive versus negative). In subgroup analysis, patients with both KRAS mutation and PD-L1 positivity had longer PFS compared with patients with KRAS mutation and PD-L1 negativity (10.1 versus 2.6 months, p = 0.047). This finding was not shown in patients with wild-type KRAS. Conclusion: Our analysis suggested that PD-L1 expression might be a useful biomarker for ICIs in BTC patients with KRAS mutation but not in those with wild-type KRAS.