Molecular profiling and the impact of treatment on outcomes in adenoid cystic carcinoma (ACC) type-I and II

Clin Cancer Res. 2024 Feb 28. doi: 10.1158/1078-0432.CCR-23-3182. Online ahead of print.

Abstract

Background: Adenoid cystic carcinoma (ACC) is an uncommon salivary gland cancer with no approved therapies available to treat advanced, incurable disease. Recent molecular profiling efforts have identified two important subtypes: the more aggressive ACC-I is characterized by Notch pathway alterations and MYC amplification while ACC-II demonstrates a more indolent phenotype and TP63 overexpression.

Methods: This retrospective observational cohort study involved de-identified samples from 438 ACC patients with tumor samples sent for commercially-available molecular profiling (Caris Life Sciences). Next-generation whole-exome and whole-transcriptomic sequencing was performed on primary and metastatic samples. Immunostaining for PD-L1 and RNA deconvolution (quanTIseq) was used to explore the tumor immune microenvironment (TME). Real-world clinical and survival outcome metrics were extracted from insurance claims data.

Results: MYC expression was 1.61-fold higher (39.8 vs. 24.7, p<0.0001) among NOTCH1-mutant ACC-I tumors, while MYB/L1 fusion rates were similar among ACC-I/II. The median B cell fraction in the TME was higher among ACC-II (7.1 vs. 5.8%; p<0.01), though infiltrating T cells subsets were low among either ACC subgroup (both <1%). When pooling systemic treatment categories, ACC-I patients had worse outcomes with available therapies (HR 3.06, 95%CI 1.65-5.68; p<0.01), with no significant difference in overall survival between ACC-I/II based on chemotherapy or VEGFR tyrosine kinase inhibitor exposure in smaller subsets.

Conclusions: We confirmed the previously reported associations with MYC and TP63 in the prognostically relevant subgroups of ACC-I and II, respectively, and report immunologic differences among these subtypes. Survival outcomes are comparatively worse in ACC-I regardless of treatment type.