3686 - Association of the Number of Weekly Cisplatin Cycles with Outcomes among Patients with Head and Neck Cancer Treated with Definitive Chemoradiation

S. J. Ma¹, S. Zhu², P. Bhateja³, E. Gogineni⁴, S. Baliga⁴, D. J. Konieczkowski⁴, D. L. Mitchell⁴, J. C. Grecula⁴, M. Old⁵, J. W. Rocco⁵, M. Bonomi⁶, and D. M. Blakaj⁴; ¹The Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, The Ohio State University Comprehensive Cancer Center, Columbus, OH, ²University of Florida, Gainesville, FL, ³Ohio State University, Columbus, OH, ⁴Department of Radiation Oncology, James Cancer Hospital, The Ohio State University, Columbus, OH, ⁵The Ohio State University Department of Otolaryngology - Head & Neck Surgery, Columbus, OH, ⁶Department of Medical Oncology, The Ohio State University Wexner Medical Center, Columbus, OH

Purpose/Objective(s): RTOG 0129 suggested 200 mg/m2 as a minimum cumulative dose to be effective when using concurrent cisplatin given every 3 weeks. However, it remains unclear whether this threshold applies to weekly cisplatin. We performed an observational cohort study of patients with head and neck cancer treated with definitive chemoradiation receiving weekly cisplatin to determine if the number of cycles were associated with oncologic outcomes. Materials/

Methods: A single-institution database was queried for patients with head and neck cancer diagnosed between December 2011 and March 2020 who underwent chemoradiation and had weekly cisplatin of 40 mg/m2 with at least 200 mg/m2 total. Patients with <5 cycles of weekly cisplatin were excluded. Cox multivariable analysis (MVA), Fine-Gray MVA, and Kaplan-Meier were performed to evaluate the treatment outcomes among >6, 6, and 5 cycles. Reasons for missing cisplatin cycles were also recorded. For subgroup analysis, Cox and Fine-Gray MVA were repeated among those with p16-negative and p16-positive subgroups. Bonferroni correction was performed, with p value of 0.025 considered as statistically significant.
Results: A total of 142 patients were identified (n=34 with 5 cycles [n=8 for p16-negative; n=26 for p16-positive], n=58 with 6 cycles [n=10 for p16-negative; n=48 for p16-positive], and n=50 with >6 cycles [n=14 for p16-negative, n=36 for p16-positive]). Median follow up was 46.8 months (interquartile range 40.8-55.6). Among 92 patients with missing cisplatin cycles, 79 patients (85.9%) had 92 reasons for lower adherence to chemotherapy available for review. Three most common reasons were low blood counts (n=42/92, 45.7%), failure to thrive (n=26/92, 28.3%), and fever or infection (n=13/92, 14.1%). Outcomes were comparable between >6 cycles and 6 cycles (overall survival [OS]: adjusted hazards ratio [aHR] 0.62, p=0.39; progression-free survival [PFS]: aHR 0.92, p=0.86; locoregional failure [LRF]: aHR 0.29, p=0.20; distant failure [DF]: aHR 0.83, p=0.83). However, OS was worse for patients who received 5 cycles than those who received >6 cycles (aHR 3.37, p=0.02), while PFS (aHR 2.72, p=0.05), LRF (aHR 0.38, p=0.42), and DF (aHR 2.03, p=0.38) outcomes were comparable. On subgroup analysis, among those with p16-negative tumors (n=32), 5 cycles were associated with worse OS (aHR 145, p=0.02) than >6 cycles, but comparable PFS (aHR 56, p=0.05). However, among those with p16-positive tumors (n=110), OS and PFS outcomes were comparable between those treated with 5 cycles versus >6 cycles (OS: aHR 0.95, p=0.95; PFS: aHR 1.36, p=0.68).
Conclusion: Patients who received 5 weekly cisplatin cycles had lower overall survival than those with >6 cycles, while oncologic outcomes were comparable. This finding was primarily driven by patients with p16-negative tumors, while outcomes for patients with p16-positive tumors were not associated with the number of cisplatin cycles. Cytopenia represented the most common reason for missing cisplatin cycles.