Factors associated with grade ≥2 acute radiation dermatitis after adjuvant radiotherapy in breast cancer patients with immediate breast reconstruction: a single-center study using penalized regression.

BACKGROUND: Immediate breast reconstruction is increasingly performed in women with breast cancer, yet many such patients also require post-mastectomy radiotherapy (PMRT), during which acute radiation dermatitis (RD) is common and may compromise reconstruction outcomes and quality of life. Risk factors for RD have been characterized mainly in non-reconstructed or breast-conserving populations, and data specific to immediately-reconstructed patients are scarce. METHODS: We retrospectively analyzed 61 women with immediate reconstruction (implant/expander or autologous tissue) who completed adjuvant radiotherapy. All patients received intensity-modulated radiotherapy (IMRT) delivered in 25 fractions with a routine 0.5-cm chest-wall bolus; the prescribed chest-wall/breast target dose was ≥50 Gy (EQD2 50 Gy) in 57 of 61 patients (93%) and slightly lower (<50 Gy) in the remaining 4. Acute RD was graded with the Radiation Therapy Oncology Group (RTOG) scale and dichotomized at grade ≥2. Univariable comparisons used the Mann -Whitney U and Fisher exact tests. Because of the small sample, imbalanced outcome and quasi-separation, Firth penalized-likelihood logistic regression with profile penalized-likelihood 95% confidence intervals (CIs) was used as the primary method. A parsimonious model (age plus supraclavicular irradiation) was pre-specified from clinical priors and univariable P<0.10, and evaluated by likelihood-ratio test, area under the receiver-operating-characteristic curve (AUC) and 1000-fold bootstrap optimism correction. RESULTS: Grade ≥2 RD occurred in 72.1% of patients (44/61). In the multivariable model, older age (odds ratio [OR] 2.30 per 10 years, 95% CI 1.11 -5.87; P=0.023) and supraclavicular irradiation (OR 6.40, 95% CI 1.37 -37.37; P=0.018) were associated with grade ≥2 RD; the event rate was 76.9% (40/52) with versus 44.4% (4/9) without supraclavicular irradiation. The model likelihood-ratio χ2 was 16.30 (P=0.0003); the apparent AUC was 0.719 and the optimism-corrected AUC 0.705. Findings were consistent in an exploratory sensitivity analysis additionally adjusting for tumour laterality. CONCLUSION: In conclusion, grade ≥2 acute RD was common after adjuvant radiotherapy in immediately-reconstructed patients, and older age and supraclavicular (regional nodal) irradiation were associated with the outcome. These hypothesis-generating findings support intensified skin-protective strategies for high-risk patients and warrant validation in larger, multicenter cohorts.