Abstract
Introduction:
Radiation-induced oral mucositis (RIOM) may cause severe pain and impaired quality of life during head and neck radiotherapy (RT). Preventive photobiomodulation (PBM) has shown promising clinical benefits.
Materials and Methods:
Eligible patients had oropharyngeal or oral cavity cancer and were opioid-naïve at the start of curative-intent RT. Patients in the PBM group were treated between 2020 and 2023. PBM was applied to the entire oral mucosa at 2.25 J/cm2 (λ = 660 nm). Extraoral PBM was delivered using an LED cluster device at 6 J/cm2 (λ = 810 nm) to target the pharyngolaryngeal mucosa. PBM was administered three times per week by trained radiation therapists. A historical control cohort treated between 2012 and 2019 was matched 1:1 based on tumor site, concurrent systemic therapy, and prior surgery. The primary objective was to evaluate differences in daily morphine consumption at the end of RT according to the preventive use of thrice-weekly oral PBM in patients receiving curative-intent RT for head and neck cancer.
Results:
A total of 280 patients were included: 45% had oral cavity cancers, 43% were treated postoperatively, and 42% received RT without systemic therapy. Morphine use was significantly lower in the PBM group at the end of RT (22.9 vs. 37.4 mg/day; p = 0.002) and 1 month post-RT (10.3 vs. 27.2 mg/day; p = 0.0001). No morphine was required in 60% of PBM patients compared with 47% of controls. PBM appeared particularly beneficial in oropharyngeal cancer, in patients receiving definitive RT, and with concurrent treatment. In contrast, rates of grade 3–4 RIOM at the end of RT were identical (63% in both groups), and grade 3–4 mucositis-free survival did not differ (p = 0.99).
Conclusion:
In this large matched cohort, preventive PBM reduced morphine consumption during and after RT, supporting its use to alleviate RIOM-related pain in head and neck cancer patients.
Introduction
Radiotherapy (RT) is a cornerstone in the management of head and neck cancers, used either as adjuvant treatment in locally advanced stages or as definitive therapy, with or without concurrent chemotherapy. However, it is frequently associated with acute toxicities, among which radiation-induced oral mucositis (RIOM) is particularly debilitating.1,2
RIOM is a delayed and progressive inflammatory condition that typically develops after cumulative radiation exposure and may persist for several weeks following treatment completion. Severe RIOM can cause intense pain, weight loss, and nutritional impairment, sometimes requiring enteral feeding, especially in patients receiving concurrent chemotherapy. Its pathogenesis involves radiation-induced epithelial basal cell damage, triggering excessive production of reactive oxygen species and activation of Nuclear Factor-kappa B-mediated inflammatory pathways. This leads to the upregulation of proinflammatory cytokines (tumor necrosis factor-α, interleukin-1β, interleukin-6), amplification of tissue injury, disruption of epithelial turnover, and impaired mucosal regeneration due to altered stem cell, fibroblast, and endothelial function. 1
The severity of RIOM depends on multiple factors, including radiation dose and irradiated volume, concomitant chemotherapy, baseline oral and nutritional status, comorbidities, and the effectiveness of preventive strategies. Current Multinational Association of Supportive Care in Cancer and International Society of Oral Oncology (MASCC/ISOO) guidelines therefore recommend a multi-modal preventive strategy combining optimal oral care, nutritional support, and evidence-based supportive interventions. 2 Among these, photobiomodulation (PBM) has emerged as a well-established and increasingly used approach. PBM acts through mitochondrial photoreception, enhancing ATP production, modulating reactive oxygen species, and regulating inflammatory pathways, thereby promoting mucosal healing and pain relief.1–3 In line with this rationale, MASCC/ISOO guidelines now specifically endorse intraoral PBM for the prevention of RIOM in patients undergoing head and neck RT. 2
Although PBM has demonstrated clinical benefits, treatment protocols vary widely across studies, highlighting the need for real-world evaluation. 4 A recent review also reported a decreasing effect of PBM on RIOM severity in more recent studies (2016–2023) compared with earlier publications (2000–2015), while the analgesic benefit appeared unchanged. 5 Moreover, a recent trial in a curative setting failed to demonstrate mucositis-related benefit. 6 These discrepancies underscore the importance of further assessing PBM in contemporary clinical practice.
Our center introduced PBM three times per week, consistent with earlier MASCC recommendations and chosen to balance feasibility, patient burden, and resource allocation. 7 Most published studies, however, use a five-times-per-week schedule, as recommended by updated MASCC guidelines. 2 Evaluating the real-world effectiveness of this intermediate schedule in a large, homogeneously treated cohort therefore appeared clinically relevant.
Given the retrospective design of the study and the variability of mucositis grading in routine practice, daily morphine consumption at the end of RT was selected as an objective and unbiased primary endpoint in initially opioid-naïve patients. The aim of this study was to determine whether preventive PBM administered three times per week was associated with reduced morphine use at the end of RT compared with a matched historical cohort not receiving PBM.
Material and Methods
Study design
This was a single-center, retrospective, descriptive, paired-cohort study conducted at Bordeaux University Hospital.
Population
Treatment procedures
Therapeutic decisions were made in a multi-disciplinary tumor board.
All patients received normofractionated intensity-modulated radiation therapy using VMAT on an Elekta Synergy machine. Treatment consisted of 20–35 fractions to a total dose of 55–70 Gy depending on tumor site and clinical context. Histological confirmation was obtained by biopsy or surgical resection.
Preventive measures for RIOM
All patients underwent a dental assessment before RT, with necessary dental care completed to reduce the risk of odontogenic infection. Sodium bicarbonate mouth rinses were prescribed 6–12 times per day to prevent fungal superinfection, facilitate clearance of thick saliva, and avoid self-medication with alcohol-based mouthwashes. Analgesic management was carefully reviewed at least weekly and whenever required, based on patient needs and assessments by nurses or radiation therapists. Gastrostomy was generally recommended for patients receiving concurrent chemotherapy. No benzydamine mouthwash, topical morphine mouthwash, sucralfate, “magic mouthwash,” or honey was used in routine practice. Suspected Candida albicans infections were treated with fluconazole, adjusted according to antifungal susceptibility results. Oral corticosteroids were rarely prescribed for symptomatic indications. Antibiotics were not administered in the absence of clinically documented infection.
PBM procedure
PBM was prescribed by a radiation oncologist and administered by specifically trained radiation therapists three times per week (Monday, Wednesday, Friday). At least 13 intraoral treatment spots were applied, carefully avoiding the visible tumor. The dose per spot was 2.25 J/cm2 (power: 75 mW, 30 sec, λ = 660 nm), initiated before the completion of the first week of irradiation. Extraoral PBM was delivered using an external infrared LED cluster positioned bilaterally on the neck to cover the remaining pharyngolaryngeal mucosa with 6 J/cm2 (power: 5 × 200 mW, 30 sec, λ = 810 nm). Additional intraoral treatment spots were applied as needed, depending on tumor location and RIOM progression. All participants wore dark safety goggles, and each session lasted approximately 15 min.
The intraoral PBM dose of 2.25 J/cm² was consistent with the 2 J dose recommended by Zecha et al. 8 The extraoral dose of 6 J/cm² corresponded to twice the 3 J/cm² dose recommended in these guidelines to account for interindividual variability in cervical soft-tissue thickness and to maintain a uniform 30-sec application time across PBM components, reducing the risk of dosing errors. PBM parameters and prescription details are summarized in Figure 1. 9

Photobiomodulation (PBM) protocol and technical characteristics.
Endpoints
The primary endpoint was daily morphine consumption at the end of RT, expressed as oral morphine-equivalent dose according to Caraceni et al. 10 Secondary endpoints included morphine use during RT, morphine-equivalent daily consumption 1 month after RT, the difference in morphine dose between end-of-RT and 1 month, RIOM severity using the RTOG scale, weight loss, and the use of enteral nutrition. Data were collected weekly.
Statistical analysis
Continuous variables are presented as median and interquartile range and were compared using Student’s t-test. Categorical variables are reported as frequencies and percentages and were analyzed using the chi-square test. As morphine dose distribution was non-normal, comparisons between groups were performed using the non-parametric Mann–Whitney test. Morphine-free survival was estimated using the Kaplan–Meier method, and groups were compared using the log-rank test. A p value <0.05 was considered statistically significant.
Ethics
All data were collected retrospectively in compliance with institutional ethical policies. Based on the documents presented, the Research Ethics Committee of the Health and Research Ethics Center of Bordeaux approved the study (reference CER-BDX 2025-126).
Results
Patient and tumor characteristics
Between 2020 and 2023, 286 patients received PBM during RT for head and neck cancer. Of these, 140 met the inclusion criteria and were matched with 140 patients from the historical cohort (Supplementary Fig. S1). Baseline characteristics were comparable between groups (Table 1): mean age was 63.0 years, 69% were male, 26% were never smokers, and 51% had no history of alcohol consumption. Tumor distribution was comparable between groups, except for nodal invasion which was more frequent in the PBM group (85% vs. 71%, p = 0.004). Overall, 45% had oral cavity cancer, 19% p16-negative oropharyngeal cancer, and 36% p16-positive oropharyngeal cancer. Surgery was performed in 43% of patients, while 57% received definitive RT.
Patient, Tumor, and Treatment Characteristics
An asterisk (*) indicates the categories used for matching.
t-Test was used for continuous variable and chi-square for qualitative variables.
AJCC, American Joint Committee on Cancer; N, lymph node affected; T, tumor extension.
Conventional fractionation (1.6–2.12 Gy per fraction) was used in 96% of patients. Among them, definitive RT to 70 Gy was delivered in 156 patients, whereas postoperative RT consisted of 60 Gy (N = 63) or 66 Gy (N = 47) according to pathological risk factors such as incomplete resection or extracapsular spread. 11 Twelve patients were treated with 55 Gy in 2.75 Gy fractions according to the ELAN-RT protocol (3 PBM; 9 control). 12 Mean overall treatment time was 46.5 days in the control group and 45.7 days in the PBM group (p = 0.07). Concurrent systemic therapy was administered in 68% of patients, with a nonsignificant trend toward lower cetuximab use in the PBM group (10% vs. 20%, p = 0.054).
Morphine consumption at the end of RT
Daily morphine consumption at the end of RT was significantly lower in the PBM group (p = 0.002) (Fig. 2A). Mean doses were 22.9 mg/day (range, 0–360) in the PBM group and 37.4 mg/day (range, 0–200) in the control group. No morphine was required in 60% of PBM patients versus 47% of controls. Among patients requiring opioids, mean consumption was lower with PBM (57 mg/day) than in controls (71 mg/day). Subgroup analyses of stratified factors (p16-negative and p16-positive oropharyngeal tumors combined) showed consistent trends; however, statistical significance was only observed for oropharyngeal cancers, definitive RT, and concurrent chemotherapy (Fig. 2B; Supplementary Fig. S2).

Effect of PBM on individual daily morphine consumption at the end of radiotherapy.
Morphine-free survival
Morphine-free survival was significantly longer for patients in the PBM group (p = 0.025) (Fig. 3). At 21, 35, and 42 days after RT initiation, morphine-free survival was 86% (PBM) versus 79% (control), 76% versus 70%, and 61% versus 51%, respectively.

Morphine-free survival according to intervention by PBM and stratification factors.
According to stratification factors, concurrent systemic therapy was associated with shorter morphine-free survival (p = 0.005). There was a trend toward shorter morphine-free survival in patients without surgery (p = 0.095). Tumor site had no significant impact (p = 0.33).
Morphine consumption at 4 weeks after RT
Morphine daily consumption 4 weeks after RT was available for 137 controls (98%) and 139 PBM patients (99%). A higher proportion of PBM patients remained opioid-free (80% vs. 61%, p = 0.0001). Among those requiring opioids, the median morphine consumption was lower in the PBM group (37 mg) than in controls (60 mg).
Radiomucositis
Grade 3–4 RIOM at the end of RT occurred in 52 control patients (37.1%) and 51 PBM patients (36.4%). Grade 3 mucositis-free survival curves were nearly identical (p = 0.99) (Supplementary Fig. S3). No significant differences were observed across stratification factors, including chemotherapy, surgery, or tumor site.
Discussion
This study highlights the benefit of PBM in reducing morphine requirements among patients undergoing RT for head and neck cancer. Fewer patients required opioids, morphine initiation was delayed, and the overall doses were lower both at the end of RT and four weeks after treatment. Although randomized studies typically use RIOM severity as the primary endpoint and pain as a secondary outcome,6,13–16 opioid consumption represents a more objective and clinically relevant measure in retrospective settings. Unlike mucositis grading, which is subject to inter-observer variability, morphine use provides a standardized indicator of clinically significant pain management. To further minimize bias, we restricted inclusion to oropharyngeal and oral cavity cancers, unlike many prior studies with broader eligibility criteria.6,13,14,16 The stratification factors employed also proved pertinent, as morphine consumption differed or tended to differ across stratification subgroups based on the timing of morphine initiation.
The two groups were generally comparable, although three differences deserve consideration. First, nodal involvement was more frequent in the PBM group (85% vs. 71%, p = 0.004). Second, median treatment duration was slightly shorter in the PBM group due to a shift from 35 fractions of 2 Gy to 33 fractions of 2.12 Gy between 2018 and 2020 to avoid exceeding a total treatment duration of 49 days. Third, delineation methods evolved over time, gradually transitioning from a compartment-based approach before 2018 and the publication by Grégoire et al. to a geometric approach.17,18 Increased nodal involvement and modest treatment acceleration could theoretically increase toxicity in the PBM group, whereas reductions in prophylactic target volumes might have mitigated it.
Pain reduction and decreased morphine consumption are consistent with most published studies evaluating PBM for the prevention of RIOM in head and neck cancer. 19 In the triple-blind randomized trial by Gautam et al., use of PBM significantly reduced pain (p < 0.0001) and opioid use (9% vs. 26%, p < 0.0001) in 221 patients treated with concurrent chemoradiotherapy using cisplatin. 13 Similar findings were observed by Antunes et al. in 94 head and neck cancer patients treated with radiochemotherapy. 14 Kauark-Fontes et al. also documented a drop in opioid use from 27.3% in the placebo group to 4.0% with PBM (p = 0.02) in a randomized trial involving 55 patients with oropharyngeal and oral cavity cancer treated with RT alone or in combination with chemotherapy. 15 In contrast, Gouvêa de Lima et al. found no significant difference in severe pain or opioid use, although overall analgesic drug use in their study was markedly lower (8% for opioids in both groups) than in our real-world cohort. 16 In our study, we observed a reduction in the number of patients requiring morphine, from 53% in the control group to 40% in the PBM group. Other retrospective analyses have also reported high opioid reliance in head and neck cancer radiochemotherapy, with, for example, 80% of the 261 patients requiring opioids in the cohort by Ganju et al., 20 and 83% of the 976 patients in the SEER analysis by McDermott et al. 21 In our population, where a large proportion of patients require morphine, this outcome is clinically relevant.
In contrast to opioid use, PBM did not reduce the incidence of grade 3–4 RIOM in our study. The frequency of PBM administration may have contributed to this finding. While most published studies administer PBM five times per week, this is not consistent across all studies.22,23 At the time we initiated PBM at our center, existing guidelines recommended performing PBM two to three times per week, up to once daily. 8 For feasibility reasons, a three-times-per-week schedule was adopted to ensure long-term sustainability of the intervention. 24 Most recent guidelines now favor daily administration. 2 Notably, a recent review by Shen et al. reported a decreasing effect of PBM on RIOM in more recent studies, although analgesic benefit remained consistent; no clear explanation has yet been identified. 5 Our findings support the hypothesis that PBM exerts a frequency-dependent effect and should be administered more than three times per week to achieve a reduction in RIOM.
This study has several limitations. First, its retrospective design limits causal inference despite careful matching. Moreover, given current guidelines recommending daily PBM, conducting a prospective trial comparing three versus five weekly sessions would be ethically questionable, as it could expose patients to a potentially less effective schedule. Our findings nonetheless support the rationale that daily PBM may be required to achieve optimal mucositis prevention. Second, the primary endpoint of morphine consumption was intentionally selected for its objectivity yet remains an indirect surrogate for pain. Mucositis grading, although widely used, is inherently less reliable outside prospective controlled trials. Nevertheless, the near-superimposition of grade 3 mucositis-free survival curves suggests consistent assessment across groups. Despite these limitations, this study provides robust real-world evidence supporting PBM implementation, particularly at higher treatment frequencies.
In conclusion, this large matched-cohort analysis shows that preventive PBM is associated with reduced morphine use at the end of RT and 1 month after treatment in opioid-naïve patients with head and neck cancer. This benefit was most pronounced in patients with oropharyngeal cancer treated with definitive chemoradiotherapy, compared with those with oral cavity cancer treated in the postoperative setting without chemotherapy. PBM administered three times per week appears to confer measurable analgesic benefit, although daily PBM should be preferred as it may provide greater efficacy for patients.
Authors’ Contributions
Conceptualization: N.G., A.-C.D., A.L., J.-P.S., M.Y., C.R., A.D., K.S., V.V., and C.D.; data curation: N.G. and C.D.; formal analysis: N.G. and C.D.; funding acquisition and investigation: A.-C.D., A.L., J.-P.S., M.Y., C.R., A.D., K.S., and V.V.; methodology: N.G., A.-C.D., A.L., J.-P.S., M.Y., C.R., A.D., K.S., V.V., and C.D.; project administration: C.D.; resources, software, and supervision: V.V. and C.D.; validation: N.G., V.V., and C.D.; visualization: N.G., A.-C.D., A.L., J.-P.S., M.Y., C.R., A.D., K.S., V.V., and C.D.; writing—original draft: N.G., A.-C.D., A.L., J.-P.S., M.Y., C.R., A.D., K.S., V.V., and C.D.; writing—review and editing: N.G., V.V., and C.D.
Footnotes
Acknowledgments
Radiation therapist team: Sabrina Berlureau, Cécile Cavalieri, Anaïs Chapeleau, Maeva Gomez, Anne-Laure Juguin, and Sophie Laroche for their implication and their enthusiasm for treating patients during all these years.
Author Disclosure Statement
The authors declare that they have no conflict of interest related to this work.
Funding Information
The authors received no financial support for the research, authorship, and publication of this article.
Supplemental Material
Supplemental Material
Supplemental Material
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
