SS 40 - H&N 2: Predicting the Future with Immunologic, Genomic, and Imaging Metrics
332 - Chromosomal Instability as a Prognostic Biomarker for Local Failure in Laryngeal Carcinoma Patients Treated with Definitive Radiation in NRG/RTOG 9512
University of Wisconsin Madison Madison, WI, United States
P. Cosper1,2, J. Harris3,4, K. Jones1, A. Bryan1, M. Paracha1, L. Henderson5, R. Hu6, P. M. Harari1,2, B. Emami7, S. M. Sagar8, S. Firat9, J. Caudell10, D. J. Ma11, C. U. Jones12, M. K. Garg13, S. J. Feigenberg14, T. J. Galloway15, H. E. Kim16, and Q. T. Le17; 1Department of Human Oncology, University of Wisconsin School of Medicine and Public Health, Madison, WI, 2University of Wisconsin Carbone Cancer Center, Madison, WI, 3American College of Radiology, Philadelphia, PA, 4NRG Oncology Statistics and Data Management Center, Philadelphia, PA, 5Wisconsin State Laboratory of Hygiene, University of Wisconsin, Madison, WI, 6Department of Pathology and Laboratory Medicine, University of Wisconsin, Madison, WI, 7Loyola University Medical Center, Department of Otolaryngology and Head and Neck Surgery, Maywood, IL, 8Professor, McMaster University, Hamilton, ON, Canada, 9Medical College of Wisconsin, Milwaukee, WI, 10Moffitt Cancer Center, Tampa, FL, 11Mayo Clinic Comprehensive Cancer Center, Rochester, MN, 12Sutter Medical Center Sacramento, Roseville, CA, 13Montefiore Einstein Comprehensive Cancer Center, Bronx, NY, 14Department of Radiation Oncology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, 15Department of Radiation Oncology, Fox Chase Cancer Center, Philadelphia, PA, 16Karmanos Cancer Institute, Detroit, MI, 17Stanford University, Stanford, CA
Purpose/Objective(s): Chromosomal instability (CIN) is an ongoing rate of chromosome missegregation events over the course of multiple cell divisions and is common in cancer. There appears to be a range of CIN that allows for optimal tumor cell fitness, and when increased beyond a maximally tolerated threshold, can lead to cell death due to loss of both copies of one or more essential chromosomes. Radiation induces CIN and we have shown that head and neck cancer cells engineered to have higher CIN at baseline are more sensitive to radiation than their isogenic, wild-type counterparts. We hypothesized that laryngeal cancers from patients enrolled on RTOG 9512 with higher levels of CIN at baseline would be more sensitive to radiation therapy (RT), as reflected by decreased local failure (LF) and locoregional failure (LRF). Materials/
Methods: RTOG 9512 randomizedpatients with T2 squamous cell carcinoma of the glottic larynx to hyperfractionated (twice daily) versus standard fractionation (once daily) RT. CIN was quantified by 6-chromosome fluorescence in situ hybridization (FISH). Samples were stained with centromeric probes for chromosomes 3, 4, 7, 9, 10 and 17 and CIN was calculated as the average percentage of cells that deviate from the modal number for each centromeric probe. LF/LRF rates were estimated by the cumulative incidence method. Cox proportional hazards models (stratified by treatment) were used to associate (two-sided 0.10) CIN with LF, LRF, and overall survival (OS). Hazard ratios (HRs) for CIN are presented per 1-standard deviation (SD) increment. Results: Of 239 eligible patients from RTOG 9512, 129 had samples available and 107 (44.8%) were of sufficient quality to quantify CIN. There were no notable differences in treatment assignment, patient characteristics, or outcomes between patients included in or excluded from analysis. Mean CIN was 0.48 (SD 0.09) and there was no difference between treatment arms. 5-year LF/LRF rates in this cohort were 23.5% [95% confidence interval (CI) 15.9, 32.0) and 26.3% (95% CI 18.4, 35.0). CIN was significantly associated with LF (p=0.04), HR 0.67 (90% CI 0.48, 0.92), and LRF (p=0.04), HR 0.68 (90% CI 0.50, 0.92). After adjustment for covariates, CIN remained significantly associated with LF (p=0.01), HR 0.58 (90% CI 0.40, 0.82) and LRF (p=0.009), HR 0.59 (90% CI 0.42, 0.82). Before or after adjusting for covariates CIN was not significantly associated with OS (p=0.61, p=0.36) with HR 1.07 (90% CI 0.85, 1.36) and 0.87 (90% CI 0.68, 1.12). Conclusion: Higher CIN in laryngeal tumors from RTOG 9512 was associated with decreased LF/LRF, potentially indicating enhanced radiation sensitivity. This suggests that CIN may serve as a prognostic biomarker for LF/LRF in laryngeal cancers treated with definitive radiation, though further validation is required.