Antiemetic prophylaxis during chemoradiation: Sub-study of the GAND-emesis trial identifying dosimetric predictors for vomiting

BackgroundNausea and vomiting are common and distressing side effects of chemoradiation. This study aims to identify patients at increased risk of vomiting by analysing dosimetric factors of the bowel while considering the antiemetic treatment applied. The study is based on the randomised double blind placebo controlled GAND-emesis trial (PMID: 26952945) demonstrating the superiority of adding fosaprepitant (FOS) to palonosetron and dexamethasone during radiation and weekly cisplatin in cervical cancer patients.MethodsPlanning CT scans and dosimetry data were collected and stored pseudonymised prospectively during the trial at the Danish sites. Blinded to treatment (FOS or placebo) and outcome, the contouring of the bowel was standardised to ensure consistency across the cohort. A predictive model for vomiting was developed using 14 variables as potential predictors. Model selection involved cross-validation of all subsets of variables using bootstrap resampling. The primary outcome was time to first vomit as in the initial study.ResultsOf 184 patients, 55 (29.9%) experienced vomiting; 40% and 60% in the FOS and placebo group, respectively. The best-performing model included three variables: treatment (FOS), BMI, and a cross-term variable between FOS and bowel volume receiving ≥15 Gy (V15). Hazard ratios and 95% CI for the three variables were: 0.28 (0.08-0.72), 0.81 per 5 units (0.56-1.07), and 1.56 per 500 cm3 (1.05-2.40). A dosimetric factor (V15) was significant only in patients receiving FOS, suggesting that optimal prophylaxis for cisplatin-induced vomiting may uncover the emetic impact of the pelvic radiation.ConclusionsThis is the first study demonstrating clinical meaningful and implementable dosimetric predictors for vomiting during chemoradiation. The study confirms the overall superiority of the triple antiemetic regimen and indicates that within these patients, we can further identify those being at increased emetic risk due to the radiation. Applying a soft constraint for V15 during radiation planning, the risk could potentially be reduced, and a tailored antiemetic regimen could be offered to the patients at increased risk.Clinical trial identificationEudraCT 2009-014691-21; NCT 01074697.Legal entity responsible for the studyC. H. Ruhlmann.FundingHas not received any funding.DisclosureAll authors have declared no conflicts of interest.