Page 124 - Read Online
P. 124

Page 4 of 13                                                 Fang et al. Cancer Drug Resist. 2025;8:42
































               Figure 1. Somatic mutations of ROS1 in patients with HNC. Lollipop plots show somatic mutation resulting in amino acid changes in (A)
               MSKCC HNC cohort and (B) TCGA HNC cohort. HNC: Head and neck cancer; TCGA: The Cancer Genome Atlas.

               rank-sum test. Overall survival (OS) differences were evaluated using Kaplan-Meier curves (log-rank test)
               and Cox proportional hazards models. Multivariate analyses were adjusted for age, gender, metastatic status,
               TMB, and treatment regimen. PD-L1 expression across ROS1 subgroups was compared using Fisher’s exact
               test, and the association between ROS1 status and TMB was assessed using the Mann-Whitney U-test.
               Statistical significance was defined as two-sided P <​ 0.05.


               RESULTS
               Clinicopathologic characteristics of ICI-treated HNC patients
               A total of 139 patients with advanced HNC were enrolled (median age 62 years; range, 17-81 years; 78.4%
               male). Patients were classified according to primary tumor site: oropharynx squamous cell carcinoma (37,
               26.6%), oral cavity squamous cell carcinoma (25, 18.0%), larynx squamous cell carcinoma (9, 6.5%),
               nasopharyngeal carcinoma (9, 6.5%), sinonasal squamous cell carcinoma (6, 4.3%), hypopharynx squamous
               cell carcinoma (5, 3.6%), and unspecified head and neck squamous carcinoma (48, 34.5%).


               Regarding ICI treatment, 131 patients (94.2%) received PD-1/PD-L1 inhibitor monotherapy, while 8 patients
               (5.8%) received combination therapy with CTLA-4 inhibitors. Tumor samples included 42 primary tumors
               (30.2%) and 97 metastatic lesions (69.8%), all analyzed by DNA-based next-generation sequencing (NGS).
               Detailed clinicopathologic characteristics are listed in Supplementary Table 1.


               ROS1 mutations in MSKCC and TCGA HNC cohorts
               ROS1-Mut were detected in 5.0% (7/139) of the MSKCC cohort and 7.2% (36/502) of the TCGA-HNC
               cohort. Unlike ROS1 kinase domain mutations that drive tyrosine kinase inhibitor (TKI) resistance in
               fusion-positive cancers , somatic ROS1-Mut in both the MSKCC [Figure 1A] and TCGA [Figure 1B]
                                    [26]
               cohorts were distributed broadly across the protein (e.g., extracellular, transmembrane, cytoplasmic, and
               kinase domains). Most variants were classified as variants of unknown significance (VUS) according to
               current CAP/AMP guidelines, and their clinical relevance requires functional validation.


               ROS1 mutations are associated with poor survival in the ICI-treated HNC cohort
               To assess the impact of ROS1-Mut on treatment outcomes, we examined their association with response to
               ICI therapy. Patients with ROS1-WT mutations exhibited a median OS of 11.0 months, whereas patients with
               ROS1-Mut had a significantly shorter median OS of 5.0 months [Figure 2A]. The hazard ratio (HR) for OS

                                                           117
   119   120   121   122   123   124   125   126   127   128   129