PURPOSE:Recurrent and metastatic human papillomavirus (HPV)-associated head and neck squamous cell carcinoma (R/M HPV+ HNSCC) remains largely incurable, with genetic drivers incompletely defined. We profiled the genetic landscape of R/M HPV+ HNSCC and functionally characterized genetic alterations strongly enriched in this cancer type. EXPERIMENTAL DESIGN:We identified genetic alterations uniquely enriched in 159 R/M HPV+ tumors. High-priority alterations were functionally modeled, examining proliferation, clonogenicity, migration/invasion, apoptosis, therapy response, in vivo growth, metastasis, and immune contexture. RESULTS:Compared with HPV+ primary tumors, R/M HPV+ tumors were enriched for TP53 mutations (prespecified FDR threshold met; OR, 6.23; P = 0.02) and were associated with poorer survival. Within R/M disease, cylindromatosis lysine 63 deubiquitinase (CYLD) alterations were specific to HPV+ tumors (21% vs. 0% in HPV-). TP53 mutations were predominantly clonal and associated with whole-genome duplication. Expression of TP53 gain-of-function (GOF) mutants (R175H, G245C, R273C) in HPV+ HNSCC cells increased clonogenic survival, migration/invasion, lung metastatic burden in vivo, and cisplatin IC50, without altering radiation sensitivity. CYLD knockdown accelerated cellular growth yet increased radiosensitivity. Transcriptomic analyses linked CYLD loss to NF-κB/TNFα pathway activation, a T cell-inflamed microenvironment, and checkpoint upregulation. CONCLUSIONS:R/M HPV+ HNSCC is genomically and functionally shaped by two axes with therapeutic implications: TP53 GOF mutations promote metastatic phenotypes and cisplatin resistance, whereas CYLD loss defines an HPV-specific subset with enhanced radiation sensitivity and immune activation. These data support using TP53 and CYLD as predictive biomarkers to guide investigation into precision strategies for systemic therapy choices, p53-targeted/Wee1 strategies, and radiotherapy-immunotherapy combinations in high-risk or R/M HPV+ HNSCC.

Abstract Chromosomal instability (CIN) signatures are DNA copy number-based genomic biomarkers with emerging evidence for predicting treatment sensitivity across multiple drugs and cancer types. Modern computational approaches enable the quantification of CIN signatures from clinically validated targeted DNA sequencing panels such as MSK-IMPACT. In this study, we derived copy number profiles from 63,630 tumor-normal MSK-IMPACT pairs using FACETS and computed pan-cancer CIN signatures across 73 cancer types. Analyses were restricted to pre-treatment samples, and CIN signature exposures were used to build three therapy-specific biomarkers: (i) a new biomarker of sensitivity for PARP inhibitors (PARPi) trained on progression-free survival (PFS) data from BRCA wild-type (BRCAwt) high-grade serous ovarian cancer (HGSOC) (ii) a new biomarker of sensitivity to platinum-based chemotherapies trained on disease-free survival (DFS) data from locally advanced rectal cancer (LARC) and (iii) an existing biomarker for resistance to anthracycline-based chemotherapies, now derived from targeted sequencing data for the first time. Each biomarker was validated in at least one independent MSK-IMPACT cohort of a different cancer type, using PFS calculated from treatment initiation and assessed at 12 months. The PARPi biomarker performed similarly to the Myriad MyChoice Genomic Instability Score (GIS) in identifying BRCAwt HGSOC patients with longer PFS after treatment (CIN signatures: n=84, HR=0.38, p=0.005; GIS: n=84, HR=0.41, p=0.004) and fully identified responders in an independent BRCAwt prostate adenocarcinoma (PRAD) cohort (Sensitivity 100%, Specificity 60%, AUC=0.73). The platinum biomarker predicted longer PFS in primary head and neck squamous cell carcinoma (HNSCC) (n=82, HR=0.43, p=0.007). The anthracycline biomarker predicted shorter PFS in hormone-receptor positive HER2 negative (HR+/HER2-) breast cancer - both primary (n=191, HR=1.72, p=0.015) and metastatic (n=216, HR=1.72, p=0.003) - and in primary soft-tissue sarcoma (n=251, HR = 1.81, p=0.006). These results highlight the feasibility and clinical potential of CIN signatures at a pan-cancer level, illustrating how the frequently overlooked complexity of genome-wide copy number alterations contained in routine targeted panels can be transformed into interpretable biomarkers to guide therapy selection. This proof-of-concept establishes that CIN signatures can be extracted from regulatory-approved assays such as MSK-IMPACT, although further validation is needed to facilitate clinical adoption. Citation Format: David Gómez-Sánchez, Adam Price, Max Schmidt, Sharafudeen Abubakar, Farheen Shah, Christina Lee, Chin-Tung Chen, Barbara Hernando, Daniel Muldoon, Areej Alsaafin, Evan Seffar, George Li, Subhiksha Nandakumar, Wassim Abida, Stephen Graves, Mackenzie Sullivan, Rachel N. Grisham, Britta Weigelt, Luc GT Morris, Nadeem Riaz, Pedram Razavi, Allison L. Richards, Mark Donoghue, Walid Khaled Chatila, Chaitanya Bandlamudi, Nikolaus Schultz, Michael F. Berger, Sohrab Shah, Geoff Macintyre, Julio Garcia-Aguilar, Francisco Sanchez-Vega. CIN signatures as biomarkers of drug sensitivity: Real-world evidence from DNA targeted sequencing data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1024.

The co-occurrence of germline and somatic oncogenic alterations is frequently observed in breast cancer, yet their combined influence on tumour evolution and therapy resistance remains poorly defined. Through an integrated clinicogenomic analysis of more than 5,800 patients, we show that germline (g) pathogenic variants dictate the evolutionary trajectory of acquired resistance. We specifically find that gBRCA2-associated tumours are uniquely predisposed to develop acquired RB1 loss-of-function alterations, resulting in poor outcomes on standard-of-care frontline CDK4/6 inhibitor (CDK4/6i) combinations. This vulnerability is driven by a dual mechanism: baseline RB1 hemizygosity (heterozygous loss resulting in a single functional RB1 allele), which lowers the evolutionary barrier to biallelic inactivation, and ongoing homologous recombination deficiency, which promotes acquisition of RB1 loss-of-function alterations under the selective pressure of CDK4/6i. Preclinical models from gBRCA2 carriers showed near-uniform resistance to CDK4/6i, with consistent post-treatment Rb loss. Across multiple independent models and in our clinical data, PARP inhibition consistently outperformed CDK4/6i. Our findings suggest that prioritizing PARP inhibition in gBRCA2 carriers may intercept RB1-loss trajectories and delay resistance. More broadly, we establish a predictive framework for forecasting drug-resistant trajectories based on pre-treatment allelic configuration and mutational signatures.

The oncogenic impact of somatic driver alterations is shaped by tissue context. Classifying alterations by cancer type and evaluating their context-specific properties requires large cohorts of genomically profiled and clinically annotated tumors. Here, we define cancer type-specific patterns of driver alterations, including 164 newly identified hotspots, in 54,331 tumors from 48,179 patients spanning 448 histological cancer subtypes. One-third of all drivers arose in non-canonical contexts and exhibited distinct features, including increased subclonality, later emergence, and divergent biological properties. Within cancer types, gene fusions and other distinct patterns of co-occurring drivers are indicative of earlier age of disease onset. We also identify ancestry-specific differences in human leukocyte antigen (HLA)-restricted driver neoantigens affecting T cell receptor therapy eligibility, and demonstrate cancer-type-specific patterns of intrinsic resistance via somatic HLA loss. Our findings highlight that functional roles of driver alterations depend on the cancer types and clinical contexts in which they arise.

The phase 2 POLAR trial evaluated maintenance pembrolizumab plus olaparib in 63 participants with metastatic pancreatic cancer with disease control on platinum-based chemotherapy. Participants were prospectively stratified into three cohorts by type of HRD: Cohort A (homologous recombination deficient [HRD] by BRCA1/2 or PALB2 mutations, N=33), Cohort B (non-core HRD mutations, N=15), and Cohort C (platinum-sensitive without HRD mutations, N=15). Cohort A used a two-stage design with co-primary endpoints of objective response rate (ORR) ≥43% and 6-month progression-free survival ≥77% per Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. For Cohort A, ORR was 35% (95% CI: 15-59) and 6-month-PFS rate was 64% (95% CI: 49-82), not meeting the primary endpoint. Among surviving participants (N=17), the median follow-up was 26.0 months (range: 1.4-52.5), and the 2-year overall survival rate was 56% (95% CI: 41-76). Median PFS for Cohort A was 8.3 months (95% CI: 5.3-not reached), 4.8 months (95% CI: 4-12) for Cohort B, and 3.3 months (95% CI: 1.9-4.8) for Cohort C. Pre-planned translational profiling demonstrated that molecular response by circulating cell-free DNA (cfDNA), high tumor-infiltrating lymphocyte (TIL) density, and increased abundance of frameshift indels and neoantigens were associated with durable benefit, particularly in HRD tumors. These findings support a precision immunotherapy approach for biomarker-defined subsets of pancreatic cancer. ClinicalTrials.gov identifier: NCT04666740.

Supplementary Figure 17. (A) Objective response rate, (B) progression-free survival, and (C) overall survival to PD-(L)1 inhibition among patients with advanced/metastatic NSCLC, according to ATM expression by immunohistochemistry.

List of 301 genes included in all versions of MSK-IMPACT and DFCI OncoPanel NGS platforms used for all analyses.

Kaplan-Meier curves of overall survival of patients with long-term response (LTR), short-term response (STR), and progressive disease (PD) from individual sites.

Supplementary Figure 9. (A) Disease free- and overall survival among patients with (A) stage I, and (B) stage II NSCLC according to ATM mutation status.

Time on treatment among patients with long-term response (LTR).

Kaplan-Meier curve of progression-free survival (PFS) among patients with long-term response (LTR) in the combined cohort.

Univariable and multivariable analyses of association of clinical characteristics with patients achieving complete response (CR) compared to patients not achieving complete response (non-CR).

Supplementary Figure 19. (A) Objective response rate, (B) progression-free survival, and (C) overall survival to PD-(L)1 immune checkpoint blockade monotherapy among patients with ATMMUT NSCLC, according to KEAP1 mutation. (D) Objective response rate, (E) progression-free survival, and (F) overall survival to PD-(L)1 immune checkpoint blockade in combination with platinum doublet chemotherapy among patients with ATMMUT NSCLC, according to KEAP1 mutation status

Supplementary Figure 29. Box plot showing immune cell subsets significantly enriched among ATMMUT versus ATMWT NSCLC as assessed by deconvoluting bulky RNAseq data from the TCGA-NSCLC dataset. Non-significant results are not shown.

Supplementary Figure 18. (A) Objective response rate, (B) progression-free survival, and (C) overall survival to PD-(L)1 immune checkpoint blockade monotherapy among patients with ATMMUT NSCLC, according to STK11 mutation. (D) Objective response rate, (E) progression-free survival, and (F) overall survival to PD-(L)1 immune checkpoint blockade in combination with platinum doublet chemotherapy among patients with ATMMUT NSCLC, according to STK11 mutation.

Supplementary Figure 6. (A) Tumor mutational burden and (B) PD-L1 tumor proportion score distributions according to ATM protein expression by IHC (lost versus intact).

Supplementary Figure 11. Overall survival since the date of initial diagnosis among patients with stage IV NSCLC according to ATM mutation status.

Supplementary Figure 1. (A) ATM mutation classification schema (B) Lollipop plot of all 714 mutations in the ATM gene identified in the combined DFCI/MSK cohort. (C) Lollipop plot of all 152 benign mutations in the ATM gene identified in the combined DFCI/MSK cohort.