Session Item

A retrospective study based on 655 patients treated in a single center in Switzerland suggested that head and neck squamous cell cancer (HNSCC) patients receiving radiotherapy (RT) during “dark” season (fall/winter) may have better outcomes than those treated during ”light” season (spring/summer) (Elicin et al, Radiother Oncol 2021). The possible rationale is that seasons have an impact on cell cycle progression. We tested this hypothesis on large databases of randomized controlled trials (RCT).

The databases of the meta-analysis of chemotherapy in head and neck carcinoma (MACH-NC) and of the meta-analysis of radiotherapy in head and neck carcinoma (MARCH) were queried for individual patient data from RCT of non-metastatic HNSCC patients with available RT duration. For MACH-NC, we selected only patients from the arms of concomitant chemo-RT. Treatment season was defined by RT mid-treatment date being during dark season (from autumn equinox to spring equinox) or during light season (from spring equinox to autumn equinox). As RT dates were often not available, RT mid-treatment date was estimated for each patient using the date of randomisation and RT duration. Progression free survival (PFS) was the primary endpoint. In each meta-analysis, the hazard ratio (HR) of dark season versus light season and its 95% confidence interval (95%CI) were estimated by a Cox model stratified by trial, and, as in the Elicin et al publication, adjusted on sex, tumor site, stage. In the MARCH subset, we also adjusted for the type of RT fractionation (standard versus modified). Sensitivity analyses were performed, restricted to patients who received at least 60 Gy in less than 10 weeks (“complete” RT) and, in the MARCH subset, to trials performed in Northern Europe where differences in daylight between winter/summer are maximal. Statistical and sensitivity analyses were planned a priori.

We included 11,906 patients from 35 trials of MARCH and 3,289 from 31 trials of MACH-NC. In both patient subsets, patient sex, tumor site and tumor stage were strong prognostic factors for PFS. There was no evidence of a benefit on PFS of the dark season for RT delivery in the MARCH (HR: 1.02 [95%CI 0.98; 1.07], p=0.33) or in the MACH-NC (HR: 0.99 [95%CI 0.91; 1.07], p=0.79) subset. Similar results were observed in sensitivity analyses: in patients with “complete” RT, HR were 1.01 [95%CI 0.96; 1.05] in MARCH (10,725 patients) and 0.97 [95%CI 0.87; 1.09] in MACH-NC (1,733 patients); in 2,634 patients from Northern trials with “complete” RT in MARCH, HR was 0.93 [95%CI 0.85; 1.02].

Using two large databases of non-metastatic HNSCC RCT, we did not find any effect of season of RT delivery on PFS. We could not confirm the findings from the single center retrospective study, which is possibly a false positive result.