Background: Radiation dermatitis (RD) secondary to radiation therapy (RT) to treat cancer reduces quality of life (QoL) and can lead to treatment interruption. The exact etiology of RD is unknown, and bacteria play a role in other inflammatory dermatoses. As our group recently showed nasal colonization with Staphylococcus aureus (SA) prior to RT was an independent predictor of grade ≥2 RD, we conducted a randomized controlled trial evaluating efficacy of bacterial decolonization (BD) to prevent RD and improve QoL. Methods: This is a randomized phase II trial comparing BD to standard of care (SC) for adult patients with breast cancer or head and neck cancer to receive fractionated (≥ 15 fractions) RT. Patients were randomized 1:1 to the BD intervention of intranasal mupirocin ointment twice daily and chlorhexidine body wash once daily for 5 consecutive days before RT start and repeated for 5 days every other week during RT or the SC arm of emollient use as needed. The primary endpoint was development of grade ≥2 RD using Criteria for Adverse Events v4.03, and planned sample size was 80 patients. Evaluation of a preliminary cohort showed wide variability of disease in grade 2 RD, so grade 2 RD was further differentiated for more refined statistical analysis: “moderate to brisk erythema” defined as grade 2 and “patchy moist desquamation” defined as grade 2 with moist desquamation (2-MD). The secondary endpoint was patient-reported QoL assessed via the SKINDEX-16 (SD-16) questionnaire before and after RT. Bacterial culture swabs of the nares and skin at RT beginning, middle, and end were obtained for both groups. Results: 80 patients were randomized 1:1 to each arm (40 BD, 40 SC) June 2019-August 2021. 78 breast and 2 head and neck cancer patients were enrolled instead of the projected 40 of each cancer type. 76 patients were included for analysis (38 BD, 38 SC). Clinical and demographic characteristics were well balanced between arms. We demonstrated prevention of RD grades 2-MD or higher in the BD arm compared to the SC arm (0/38, 0% vs 9/38, 23.68%; P=0.002). Additionally, BD resulted in a significantly lower median RD grade compared to SC (1.19±0.7 vs 1.58±0.75, P=0.019). Finally, a linear regression model showed a significant association between BD and decreased RD grade (estimate=-0.431, 95% CI: -0.7516, -0.1054; p=0.010), even when adjusting for other RD risk factors. Most patients reported no difficulty with BD and only one patient discontinued due to itch. There was no difference in QoL outcomes between arms. Conclusions: Our results support the use of a BD regimen to prevent moist desquamation in patients receiving RT for breast or head and neck cancer. Our study included mainly breast cancer patients; thus BD efficacy needs to be tested in other solid tumors receiving RT. This is the first study demonstrating efficacy of BD to reduce RD. Given the safety and availability of this regimen, we suggest adding BD to RD prophylaxis protocols. Clinical trial information: NCT03883828.