297 - Comparison of Image-Guided Intensity Modulated Radiotherapy (IG-IMRT) vs. Conventional Radiotherapy (4FRT) for Locoregionally Advanced Cervical Cancer on the NRG GY006 Trial

L. K. Mell¹, W. Deng², L. Gao¹, C. M. Kunos³, J. P. Chino⁴, A. M. Quick⁵, J. M. Feddock⁶, E. D. Donnelly⁷, K. V. Albuquerque⁸, J. L. Leif⁹, D. L. Richardson¹⁰, S. R. Amarnath¹¹, N. Jegadeesh¹², L. J. Santos-Reyes¹³, H. MacKay¹⁴, C. Aghajanian¹⁵, B. J. Monk¹⁶, A. Miller¹⁷, J. S. Mayadev¹⁸, and C. Leath¹⁹; ¹University of California San Diego, La Jolla, CA, ²NRG Oncology, Buffalo, NY, ³University of Miami, Miami, FL, ⁴Department of Radiation Oncology, Duke University Medical Center, Durham, NC, ⁵Department of Radiation Oncology, The Ohio State University Wexner Medical Center, Columbus, OH, ⁶University of Kentucky, Lexington, KY, ⁷Department of Radiation Oncology, Northwestern University Feinberg School of Medicine, Chicago, IL, ⁸Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, ⁹MD Anderson, Houston, TX, ¹⁰University of Oklahoma, Oklahoma City, OK, ¹¹Department of Radiation Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, ¹²Winship Cancer Institute of Emory University, Atlanta, GA, ¹³Puerto Rick MU NCORP, San Juan, PR, ¹⁴Divison of Medical Oncology and Hematology, Sunnybrook Health Sciences Centre, Odette Cancer Centre, University of Toronto, Toronto, ON, Canada, ¹⁵Memorial Sloan Kettering Cancer Center, New York, NY, ¹⁶University of Arizona College of Medicine Creighton University School of Medicine Phoenix, Phoenix, AZ, ¹⁷Roswell Park Cancer Institute, Buffalo, NY, ¹⁸Department of Radiation Medicine and Applied Sciences, University of California San Diego, La Jolla, CA, ¹⁹ONeal Comprehensive Cancer Center, University of Alabama at Birmingham, Birmingham, AL

Purpose/Objective(s): To test the hypothesis that IG-IMRT reduces toxicity and improves cisplatin delivery, compensatory hematopoiesis, and outcomes compared to 4FRT for definitive treatment of locoregionally advanced cervical cancer. Materials/

Methods: We included patients with stage IB2-IVA cervical or vaginal carcinoma enrolled on NRG-GY006 randomized to RT + cisplatin +/- concurrent triapine. 4FRT was 45 Gy in 25 fractions +/- parametrial boost. IG-IMRT was 45 Gy-47.6 Gy in 25-28 fractions +/- integrated boost 54.0-59.4 Gy to gross nodal disease. Targets were tumor, cervix, uterus, parametria, upper vagina, and pelvic and presacral nodes. Para-aortic fields were not used. Inguinal fields were used for vaginal primaries. IG-IMRT used daily online imaging (kV or CBCT). Brachytherapy was primarily HDR 27.5-30.0 Gy, 4-5 fractions. IG-IMRT plans were approved following credentialing and pre-treatment central review. Key objectives of IG-IMRT were bowel and active bone marrow (ABM) sparing, using baseline ¹⁸F-FDG PET/CT (custom) or atlas-based avoidance structures. Custom ABM was defined as regions with SUV above the mean SUV within whole bone. Atlas methodology was per Yusufaly et al. Field size was defined as craniocaudal length of the 40 Gy isodose. Hematopoietic compensatory response was defined per Noticewala et al. Statistical tests were 2-sided stratified by treatment arm. Significance level was 5% with no adjustment.
Results: Of 450 patients enrolled, 445 were eligible for analysis (4FRT: 116; custom IG-IMRT: 127; atlas IG-IMRT: 202) and 417 were assessable for toxicity (4FRT: 105; custom IG-IMRT: 118; atlas IG-IMRT: 194). Comparisons of dosimetry, toxicity, and cisplatin delivery are shown in Table. There were no significant differences in PFS or OS with IG-IMRT (HR 1.16, 95% CI 0.67-1.91 and HR 1.13, 95% CI 0.76-1.77, resp., adjusted for treatment arm and stage). Field lengths were significantly larger in patients receiving IG-IMRT (21.1 vs 18.8 cm, p<0.01). Triapine did not impact GI or hematologic toxicity. Comparison of pre vs post-therapy ABM distributions showed increased compensatory hematopoiesis (mean 0.50, SD 0.26), unrelated to treatment arm or RT technique.
Conclusion: Despite reducing dose to organs at risk, IG-IMRT did not reduce GI or overall hematologic toxicity compared to 4FRT, possibly due to increased cisplatin delivery and different field sizes. IG-IMRT had no impact on long-term outcomes.

Abstract 297 – Table 1

*Breslow-Day test suggested unequal odds ratio across stratum

Endpoint	4FRT	IG-IMRT			p (4F vs. IG-IMRT)
		Custom	Atlas	All
Bowel V45 (cc)	344	171	152	159	<0.001
BM V20 (%)	83	59	56	57	<0.001
ABM V20 (%)	85	65	64	65	<0.001
GI Toxicity (%)
Grade = 2	48	54	48	50	0.63
Grade = 3*	9	13	11	12	0.35
Hematologic Toxicity (%)
Grade = 2	52	53	66	61	0.13
Grade = 3*	30	32	36	35	0.34
Cisplatin Cycles
Mean	4.5	4.8	4.8	4.8	0.11
% = 6	24	41	30	34	0.04