Does Time of Daily Treatment Matter in Radiation Therapy for Head and Neck Cancer?
D. Jiang,M. L. Mierzwa,B. Huth,C. Xie,W. L. Barrett,K. P. Redmond,N. Forster
DOI: https://doi.org/10.1016/j.ijrobp.2015.07.1434
2015-01-01
Abstract:Cell biology research has provided evidence that cell cycle checkpoints and DNA damage response pathways are coupled to circadian clocks. Accordingly, radiation therapy (RT) may result in variable tumor cell insult or normal tissue repair depending on the time of day delivered. Here, we reviewed the outcomes of patients treated for HNSCC with RT paying attention to the time of day patients receive RT. We reviewed patients definitively treated for HNSCC between 2006 and 2012 who met the following criteria: daily radiation treatment times/complete smoking history/staging/ECOG PS/RTOG toxicity scoring available, at least 2 yrs follow-up for alive patients. A total of 240 patients were included in the analysis. Endpoints included LRC, OS, CSS, and highest RTOG acute toxicity. Cox model was used to test the effect of the time of day that a patient was treated adjusted for age, sex, race, T stage, N stage, use of chemotherapy, smoking status, tumor site, RT modality, days elapsed during treatment and ECOG performance status. We calculated the median treatment time of day for each patient, and then we created a binary variable using noon as a cut point. We additionally calculated the standard variation in daily treatment time for each patient and considered correlation between median time of day or variation in daily treatment time and greatest RTOG toxicity score using spearman correlation. Data was analyzed using a statistical software package. Patient characteristics were as follows: median age=64 years; ECOG PS: 0-50%, 1-41%, 2-9%. Smoking status: 38% current smokers, 15% never smokers, 47% former smokers, median number of pack years =34. Site of disease: 48% OP, 12% OC, 29% larynx, 2% NP, 6%HP, 3% nasal cavity/paranasal sinus. RT modality: 57% IMRT, 43% 3DC. All patients completed 100% of prescribed dose which was 63-70Gy. At median follow-up of 24 months, overall LRC was 63%, CSS was 60%, and OS was 49%. Time of daily treatment significantly correlated with CSS (p=0. 0.0262, HR 1.837, 95% CI 1.075, 3.141) with AM patients having improved CSS. Additionally, there was a trend towards improved LRC (p= 0.0641, HR 1.569, 95% CI 0.974, .528) and OS (P=.1144, HR 1.531, 95% CI 0.902, 2.597) in the AM patients. Highest RTOG toxicity score was not significantly different between AM and PM patients. No correlation was seen between variation in daily treatment time and LRC, CSS, OS. Using Spearmen correlation coefficients, we saw a trend of increased RTOG toxicity in patients with greater variability in their daily treatment times (P=.0575). CSS was significantly improved in patients treated in the AM time period after correction for other pretreatment and treatment variables. While time of day that a patient is treated may be driven by patient specific socio-economic factors not easily captured in the medical record, we cannot rule out a biologic basis for these provocative findings.