ImportanceCetuximab-based and carboplatin-based chemoradiotherapy (CRT) are often used for patients with locally advanced head and neck cancer who are ineligible for cisplatin. There are no prospective head-to-head data comparing cetuximab-based and carboplatin-based regimens for radiosensitization.ObjectiveTo compare survival with cetuximab-based and carboplatin-based CRT in locally advanced head and neck squamous cell carcinoma (HNSCC).Design, Setting, and ParticipantsThis cohort study included US veterans who received a diagnosis of HNSCC between January 2006 and December 2020 and were treated with systemic therapy and radiation. Data cutoff was March 1, 2022 and data analysis was conducted from April-May 2022.ExposuresCisplatin, cetuximab, or carboplatin-based systemic therapy as captured in VA medication data and cancer registry.Main Outcomes and MeasuresOverall survival by systemic therapy was estimated using Kaplan-Meier methods. We used propensity score and inverse probability weighting to achieve covariate balance between cetuximab-treated and carboplatin-treated patients and used Cox regression to estimate cause-specific hazard ratios of death associated with carboplatin vs cetuximab. We also performed subgroup analyses of patients with oropharynx vs nonoropharynx primary sites.ResultsA total of 8290 patients (median [IQR] age, 63 [58-68] years; 8201 men [98.9%]; 1225 [15.8%] Black or African American and 6424 [82.6%] White individuals) with nonmetastatic HNSCC were treated with CRT with cisplatin (5566 [67%]), carboplatin (1231 [15%]), or cetuximab (1493 [18%]). Compared with cisplatin-treated patients, patients treated with carboplatin and cetuximab were older with worse performance status scores and higher comorbidity burden. Median (IQR) overall survival was 74.4 (22.3-162.2) months in patients treated with cisplatin radiotherapy (RT), 43.4 (15.3-123.8) months in patients treated with carboplatin RT, and 31.1 (12.4-87.8) months in patients treated with cetuximab RT. After propensity score and inverse probability weighting, carboplatin was associated with improved overall survival compared with cetuximab (cause-specific hazard ratio, 0.85; 95% CI, 0.78-0.93; P = .001). This difference was prominent in the oropharynx subgroup.Conclusions and RelevanceIn this cohort study of a US veteran population with HNSCC undergoing treatment with CRT, almost a third of patients were ineligible to receive treatment with cisplatin and received cetuximab-based or carboplatin-based radiosensitization. After propensity score matching, carboplatin-based systemic therapy was associated with 15% improvement in overall survival compared with cetuximab, suggesting that carboplatin may be the preferred radiosensitizer, particularly in oropharynx cancers.

10549 Background: Germline genetic testing is a guideline-based practice for men with pancreas, breast and metastatic prostate cancers. Emerging evidence in female breast and ovarian cancers suggests that Black patients are less likely to receive germline genetic testing, but it is unclear how disparities extend to male cancers and vary by health system. We examined differences by race in germline genetic testing for men with pancreas, breast and metastatic prostate cancers in equal-access and non-equal access systems. Methods: We conducted a retrospective cohort study of men with newly diagnosed pancreas, breast and metastatic prostate cancers between January 1, 2019, and September 30, 2021. We studied two national cohorts: 1) Veterans receiving care in the Veterans Health Administration (VHA), an equal-access health system, and 2) commercially insured beneficiaries. Data consisted of claims and electronic health record data from the VHA Corporate Data Warehouse and claims data from Optum’s De-Identified Clinformatics Data Mart Database (2007-2021). Current Procedural Terminology codes ± laboratory orders and results, and genetic services notes were used to ascertain germline testing prior to the end of the study period. Cox proportional hazards models were used to test the association of non-Hispanic White vs. Black race with the primary outcome of germline genetic testing completion, adjusted for baseline covariates (age, cancer subtype, census region, diagnosis year, Elixhauser comorbidity index and Agent Orange exposure [VHA only]). Results: Our cohort consisted of 7,894 men (5,142 commercially-insured; 2,752 Veterans), including 1,589 Black men (779 [15.1%] commercially-insured; 810 [24.9%] Veterans). Among commercially-insured beneficiaries, one-year germline genetic testing rates were higher in White compared with Black men (18.8% vs. 13.1%; log-rank p < 0.001). Among Veterans, one-year germline genetic testing rates were similar for White and Black Veterans (13.1% vs. 16.9%; log-rank p = 0.335). After adjusting for baseline covariates, Black race was associated with a lower hazard of testing among commercially-insured beneficiaries (adjusted hazard ratio [aHR] 0.72; 95% confidence interval [CI] 0.58 – 0.90; p = 0.004), but not among Veterans (aHR 0.99; 95% CI: 0.79 – 1.25; p = 0.960). Conclusions: This is the first study to examine disparities in germline genetic testing across equal-access and non-equal access healthcare systems, and the largest national study examining germline genetic testing completion in men. While overall rates of testing were similar between Veterans and commercially-insured beneficiaries, racial disparities in testing were observed in non-equal-access settings but not in equal-access settings. Access to care, cost, and other sequelae of structural racism may impact access to guideline-based germline testing.