Subtype-Specific Immune Signatures Found in BTC Profiling Data

Specific biliary tract cancer clusters have distinct clinical and biologic features which may provide opportunities for therapeutic development.

After analyzing deidentified clinical, genomic, and transcriptomic data of patients with biliary tract cancers (BTC) from the Tempus database, investigators found that BTC subtypes exhibit diverse DNA alterations, RNA inflammatory signatures, and immune biomarkers.1

The study evaluated 454 samples of BTC with matched RNA and clinical data. The most common alterations reported were TP53 (42.5%), CDKN2A (23.4%), ARID1A (19.6%), BAP1 (15.5%), KRAS (15%), CDKN2B (14.2%), PBRM1 (11.7%), IDH1 (11.7%), TERT (8.4%), KMT2C (10.4%) and LRP1B (8.4%), and FGFR2 fusions (8.7%). Further, investigators identified potentially actionable molecular alterations in 30.5% of the BTC samples. These alterations were also found within separate site origins: in 39.1% of the intrahepatic cholangiocarcinomas, in 29.6% of extrahepatic cholangiocarcinomas, and in 15% of gallbladder cancers.

The study included patients who had confirmed histologic diagnosis of cholangiocarcinoma. Most patients had matched DNA sequencing (n = 367). The assay primarily identified mutations such as germline and/or somatic single-nucleotide polymorphisms, insertions/deletions, fusions, and copy number variations. An agglomerative clustering method was used to stratify patients on the presence of driver mutations. Genomic sequencing methods were used to derive tumor mutational burden (TMB), and all reported immune-related biomarkers were discovered from RNA expression data.

A limitation of the study is that the data used are heterogenous. Differences across demographic and clinical aspects of the patients can obscure or bias the results. Also, because a portion of the study focused on RNA, it is important to note that RNA and abundancies of protein do not precipitate 1-to-1 correspondences. The ultimate objective of the study was to assess differences in molecular- and genome-level features across the BTC subtypes, but investigators did not consider clinical outcomes or end points.

TMB and PD-L1 gene expression were found to be highly variable within theintrahepatic, extrahepatic, and gallbladder BTC subtypes. Median TMB values for gallbladder subtypes were 1.92 m/MB for intrahepatic, 2.63 m/MB for extrahepatic, and 2.5 m/MB for gallbladder. The median PD-L1 expression for each subtype were 0.875 for intrahepatic, 1.01 for extrahepatic, and 0.997 for gallbladder.

The intrahepatic subtype had a median cytolytic activity (CYT) score of 1.75, the extrahepatic subtype had a median of 1.86, and the gallbladder subtype had a 1.89. The median neoadjuvant signature (NRS) scores were 1.98 for intrahepatic, 2.02 for extrahepatic, and 2.09 for gallbladder. There were significant differences found between the gallbladder and intrahepatic subtypes for both CYT and NRS scores (P < .01). The immune-predictive score for each subtype was 9.0 for intrahepatic, 10.0 for extrahepatic, and 9.0 for gallbladder.

Previous attempts to sequence BTC showed that targetable genomic alterations occur with moderately high frequency, including alterations in genes such as FGFR2, IDH1, BRAF, and ERBB2. Other common BTC alterations have not been targetable therapeutically, but they supplied important prognostic information.

Immunotherapies targeted for the PD-1 axis and other immune checkpoints have altered the treatment of many cancer but have not demonstrated significant clinical activity in BTC. Further comprehension of immunotherapy treatment in molecularly defined subsets of BTCs could strengthen patient selection for immunotherapy and possibly aid the development of combination treatment strategies for these defined cancer subsets.

There was a high frequency of potentially actionable molecular alterations in BTC, and molecular profiling should be considered for all patients who may benefit from the discovery of a potentially actionable mutation. Specific BTC clusters have distinct clinical and biologic features which may provide opportunities for therapeutic development.


Mody K, Jain P, El-Refai SM, et al. Clinical, genomic, and transcriptomic data profiling of biliary tract cancer reveals subtype-specific immune signatures. JCO Precis Oncol. 2022;6:e2100510. doi:10.1200/PO.21.00510