Long Term Toxicity And Oncologic Outcomes Of De-Intensified Chemoradiation In Early Stage Oropharyngeal Carcinoma

Human papilloma virus (HPV) – associated oropharyngeal squamous cell carcinomas (OPSCC) have a favorable prognosis leading to efforts at treatment de – intensification (DI) to reduce toxicity while maintaining oncologic outcomes. Therefore we compared outcomes of patients treated with DI chemoradiation (CRT) to those treated with standard of care (SOC) CRT with the hypothesis DI-CRT would lead to similar oncologic outcomes while reducing treatment related toxicity. We included T0-2 N0-2b M0 (AJCC 7) OPSCC treated with CRT. Those treated with RT alone were excluded. DI-CRT patients were treated prospectively on a phase II trial where the primary PTV was treated with 70 Gy, but reduced to 63 Gy in areas of overlap with the pharyngeal constrictors, larynx, or parotids. SOC-CRT was treated with a minimum of 70 Gy to the PTV. Both groups were treated with 63 Gy to high risk subclinical disease. The DI-CRT group also had dose reduction from 58.1 Gy to 50.75 Gy for low risk subclinical disease. Kaplan Meier curves and log rank test compared overall (OS) and progression free survival (PFS). Toxicity was compared at early (< 18 months) and late (> 18 months) time points with the Kruskal Wallis test using prospectively collected patient reported outcome measures of the MD Anderson Dysphagia Inventory (MDADI), Sydney Swallow Questionnaire (SSQ) and Xerostomia Questionnaire (XQ). A total of 126 patients (97 SOC-CRT and 29 DI-CRT) were included. The two groups were well balanced with regards to T and N stage, HPV status, and smoking history. Median follow up was 3.5 years. At last follow up 5 deaths occurred – 4 in the SOC-CRT and 1 in the DI-CRT. Median OS was not met in either group and was not significantly different between the groups (p = 0.94). 2 year OS was 95.2% with SOC-CRT and 96.6% with DI-CRT. 11 treatment failures occurred with SOC-CRT and 5 with DI-CRT. The predominant pattern of failure was regionally in the neck or distantly. Only 3 local failures occurred, 2 with SOC-CRT and 1 with DI-CRT. Median PFS was not yet reached in either group, but did not significantly differ (p = 0.78). 2 year PFS with SOC-CRT was 87.8% compared to 89.5% with DI-CRT. Early median SSQ scores with DI-CRT were lower (144.8 vs 217.7) compared to SOC-CRT (p = 0.069), however MDADI scores (82.1 vs 81.6) were not significantly different (p = 0.29). Late swallow toxicity also showed a benefit for DI-CRT; median MDADI scores were higher (93.7 vs 90.5; p = 0.004), however median SSQ scores were not significantly different (136.1 vs 105.6; p = 0.88). Median late XQ scores were improved in patients treated with DI-CRT (18.8 vs 25; p = 0.017). DI-CRT did not result in inferior oncologic outcomes and was associated with modest benefits in early and late toxicity. Given the conservative nature of this DI protocol, future investigations at further DI are warranted to continue to reduce treatment morbidity.