2720 - Effect of Tumor Volume and Dose on Local and Regional Control in HPV-Related Oropharyngeal Squamous Cell Carcinoma
Presenter(s)
X. S. Chen1, B. M. Anderson2, F. Dong3, R. Morse4, V. Xu4, M. C. Repka2, S. Sheth1, S. Patel5, J. Blumberg6, C. Lumley7, W. Yarbrough7, T. Hackman8, J. Weiss9, B. S. Chera10, C. Shen2, and P. Mavroidis2; 1University of North Carolina at Chapel Hill, Chapel Hill, NC, 2Department of Radiation Oncology, University of North Carolina, Chapel Hill, NC, 3University of North Carolina Department of Radiation Oncology, Chapel Hill, NC, United States, 4Department of Radiation Oncology, University of North Carolina School of Medicine, Chapel Hill, NC, 5University of North Carolina Hospitals, Chapel Hill, NC, 6UNC, Chapel Hill, NC, 7UNC School of Medicine, Chapel Hill, NC, 8Department of Otolaryngology, University of North Carolina School of Medicine, Chapel Hill, NC, 9University of North Carolina Lineberger Comprehensive Cancer Center, Chapel Hill, NC, 10Department of Radiation Oncology, University of North Carolina at Chapel Hill, charleston, SC
Purpose/Objective(s): Optimal radiation dose for HPV-related (HPV+) oropharyngeal squamous cell carcinoma (OPSCC) is not well defined. Although initial trials of reduced dose showed high rates of tumor control, a recent randomized controlled trial failed to confirm equivalence of 60 vs 70 Gy in outcomes, particularly locoregional control. We hypothesize that tumor volume and dose affect local and regional control in HPV+ OPSCC.
Materials/Methods: We retrospectively identified a cohort of patients with HPV+ OPSCC treated with definitive (chemo)radiotherapy (RT) at a single center between 2010 and 2021. All patients received daily radiation prescribed to either 60 or 70 Gy at 2 Gy per fraction. Local (LR) and regional recurrences (RR) were confirmed with tissue biopsy or resection. Radiation dose parameters and gross tumor volumes for primary (GTVp) and nodal (GTVn) disease were extracted from original RT plans. For patients who experienced LR or RR, doses to the recurrent tumor were identified by registering diagnostic CT or PET/CT at the time of recurrence to the treatment dose. Cumulative incidences of LR or RR were estimated non-parametrically using death as a competing risk. Multivariable Fine-Gray competing risk regressions (CRR) were performed for LR and RR using GTV volume and dose as independent variables. Hazard ratios (HR) and 95% confidence intervals (CI) are reported.
Results: This cohort included 421 patients, of whom 44% never smoked and 20% had minimal smoking history (<10 pack-year). 18% had T4 and 29% had N2-3 disease (AJCC 8th ed). 91% received concurrent chemotherapy (74% cisplatin). Prescription dose was 60 Gy for 51%. Median minimum dose (Dmin) to GTVp and GTVn were 61.1 Gy (range 41.8-72.7) and 61.0 Gy (28.6-72.2), respectively. In total, 18 LR and 33 RR occurred (3 with both). Most recurrences (10 LR and 25 RR) occurred in field within the prescription dose. Relative to the high dose area, 3 LR were marginal and 3 outside, while 4 RR were marginal and 4 outside. Cumulative incidences of LR and RR were 3.6% and 7.8% at 5 years. GTVp was larger for those with LR (median 34.5 cm3, interquartile range (IQR) 12.4-69.6) than those without (median 16.8 cm3, IQR 9.2-29.5). On multivariable CRR, both GTVp volume (HR 7.1 for > vs = 32 cm3, 95% CI 2.2-22.8, p=0.001) and Dmin (HR 0.90 for every Gy, 95% CI 0.81-0.99, p=0.04) were associated with LR. An additional 6.0 Gy to Dmin was required to mitigate the increase in LR for every doubling of GTVp. Similarly for the neck, those with RR had larger GTVn (median 33.2 cm3, IQR 17.5-54.0) than those without (median 19.4 cm3, IQR 9.9-31.9). Both GTVn volume (HR 3.4 for > vs = 32 cm3, 95% CI 1.7-6.8, p=0.0004) and Dmin (HR 0.95 for every Gy, 95% CI 0.93-0.98, p=0.002) were associated with RR on multivariable CRR. For the neck, an additional 8.9 Gy was needed to counter a doubling of GTVn.
Conclusion: Tumor volume and minimum dose significantly affect local and regional control for HPV+ OPSCC. Dose higher than 60 Gy may be required for optimal locoregional control in patients with larger tumors.