Platinum/taxane/pembrolizumab vs platinum/5FU/pembrolizumab in patients with recurrent/metastatic (r/m) head and neck squamous cell carcinoma (HNSCC).

6039 Background: Pembrolizumab +/- chemotherapy is standard of care for patients (pts) with r/m HNSCC. Despite approval of platinum/5FU/pembrolizumab based upon the KN-048 study, a taxane is often substituted for 5FU due to ease of administration and tolerability. No studies have directly compared these approaches. We describe nationwide patterns of taxane vs 5FU chemoimmunotherapy in r/m HNSCC and compare survival and time on treatment between the two. Methods: This study used the nationwide Flatiron Health electronic health record (EHR)-derived de-identified database, and included pts treated with frontline pembrolizumab plus platinum-based chemotherapy (either taxane or 5FU) for r/m HNSCC. Demographic, cancer, and clinical outcomes were summarized. Categorical variables were compared using Pearson’s chi-square test; continuous variables were compared using T-test. Overall survival (OS) was estimated using Kaplan Meier methodology, and compared using log-rank test. Multivariable Cox regression for overall survival adjusting for age, sex, race, year of diagnosis, ECOG performance status (PS), smoking history, primary tumor site, PD-L1, HPV status, cisplatin use, socioeconomic status, insurance, and practice type was performed, with multiple imputation by chained equations for missing variables. Results: Of 444 pts, 321 (72%) received 5FU and 123 (28%) received taxane (108 paclitaxel, 15 docetaxel). Use of cisplatin rather than carboplatin was higher in the 5FU group than the taxane group (16.8% vs 4.9%, p=0.001). Taxane use was higher in academic vs community practices (50.7% vs 23.9%, p<0.001); within community sites, Northeast states had highest taxane use (27.5%) and Western states had lowest taxane use (16.7%). Pts with Medicare or Medicaid were more likely to receive taxane (35.5%) vs commercial (23.6%) or other (26.7%) insurance. OS did not differ between taxane and 5FU groups (mOS 12.7 vs 13.5 months, p=0.743). On multivariable Cox regression, HR for death associated with taxane vs 5FU was 1.01 (95%CI 0.73-1.41). HPV positive status was associated with improved survival (HR 0.58, 95%CI 0.40-0.83). PS>1 (HR 1.38, 95%CI 0.96-1.98) and cisplatin use (HR 1.37, 95%CI 0.95-1.98) showed non-significant trends towards worse survival. Taxane-treated pts received more treatment cycles (mean 11.3 vs 8.8 cycles, p=0.015) and were on treatment slightly longer (mean 6.9 vs 6.0 months, p=0.289) than 5FU-treated pts. Conclusions: In US pts with r/m HNSCC undergoing chemoimmunotherapy, rates of taxane vs 5FU use vary by academic setting, geographic region, and insurance status. There was no difference in survival between 5FU and taxane, and taxane-treated pts received more cycles of therapy. Platinum-taxane appears to be a non-inferior alternative to platinum-5FU with pembrolizumab for r/m HNSCC, and should be allowed in clinical trials.