Photobiomodulation (PBM) for radiodermatitis in patients with head and neck cancer (HNC) undergoing radiotherapy: A large real-world cohort and benchmarking against historical data.
Abstract
365 Background: Radiation dermatitis is nearly universal among patients with HNC receiving radiotherapy, with 73-80% developing grade ≥2 and 14-25% developing grade ≥3 toxicity in contemporary trials. Recovery typically occurs 2-4 weeks after treatment ends, with complete resolution by 6-8 weeks. PBM has demonstrated preventive benefits, yet evidence on its therapeutic role once dermatitis is established remains limited. We evaluated the clinical course, severity distribution, and recovery time among patients referred for PBM to treat acute radiodermatitis. Methods: We conducted a retrospective cohort study including consecutive patients with HNC referred for PBM for clinically diagnosed radiation dermatitis (N=181). Radiodermatitis grade (NCI-CTCAE v5.0), supportive-care needs, treatment interruptions, and recovery times were extracted from clinical records. Primary outcomes were severity distribution (Grades 1-3) and time to first improvement and complete resolution. Multivariable Cox proportional hazards modeling assessed whether clinical-related characteristics predicted time to improvement. Observed proportions were compared with published historical incidence rates using one-sample binomial tests. Results: Median age was 62 (IQR 53-71), 57% were male, and 62% white. Most had Stage III-IV disease (56%), ECOG 0-1 (89%), and received IMRT (84%) with a mean prescribed dose of 70 Gy. Radiodermatitis severity at PBM referral was Grade 1 in 74%, Grade 2 in 20%, and Grade 3 in 6%. The incidence of grade ≥2 dermatitis was 26% (95% CI 19.8-32.3%), significantly lower than historical rates of 73-80% (p<0.0001). Grade ≥3 occurred in 6% (95% CI 2.8-9.7%), lower than the expected 14-25% (p≤0.002). Median time to first improvement was 10 days (Mean 13 ± 11.7) and median time to resolution was 18 days (mean 20 ± 13.1), substantially faster than the typical 2-4-week and 6–8-week recovery windows, respectively. Most patients required no systemic therapy (96% no antibiotics; 92% no analgesics) and maintained uninterrupted RT (89% no treatment break). In multivariable Cox regression, no variable (age, sex, total dose, fraction dose, number of fractions, or RT technique) significantly predicted time to improvement, suggesting PBM benefit across subgroups (global p = 0.351). Conclusions: In this large real-world cohort, PBM for established radiodermatitis was associated with lower-than-expected rates of grade ≥2/≥3 dermatitis and accelerated clinical recovery compared with historical expectations. Improvement trajectories did not vary by demographic or treatment-related factors, supporting PBM as a broadly effective supportive care strategy in HNC radiotherapy. These findings support PBM as a promising therapeutic strategy for radiation dermatitis and justify prospective controlled trials.