Shi Wei, Zhang Li, Li Zhiming, Zhao Xu, Lui Wailok, Meng Jin, Chen Xingxing, Mei Xin, Ma Jinli, Yang Zhaozhi, Xia Jingjing, Wang Jiucun, Zhang Zhen, Shao Zhimin, Yu Xiaoli, Guo Xiaomao
Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China; Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China; Shanghai Clinical Research Center for Radiation Oncology, Shanghai, China; Shanghai Key Laboratory of Radiation Oncology, Shanghai, China.
State Key Laboratory of Genetics and Development of Complex Phenotypes, Fudan University, Shanghai, China.
Int J Radiat Oncol Biol Phys. 2025 May 6. doi: 10.1016/j.ijrobp.2025.03.077.
The clinical significance of multi-kingdom skin microbiota in acute radiation dermatitis (ARD) is not well understood. We hypothesized that skin microbiota is associated with ARD in patients with breast cancer (BC) undergoing radiation therapy (RT) after reconstructive surgery.
A total of 412 skin microbiota samples from 103 patients, taken before and after RT, from both the treated and contralateral healthy sides, were analyzed using bacterial 16S ribosomal RNA (rRNA) V3-V4 region and fungal rRNA internal transcribed spacer (ITS) sequencing. ARD was graded using the Toxicity Criteria of the Radiation Therapy Oncology Group (RTOG). Patients were divided into 2 groups: no or mild ARD subgroup (N_MD, RTOG grade 0-1) and significant ARD subgroup (SD, RTOG grade ≥ 2).
Significant differences in skin microbiota were observed between the N_MD and SD subgroups, with Staphylococcus, Cutibacterium, and Malassezia genera enriched in SD and Ralstonia and Methyloversatilis enriched in N_MD. Network analysis revealed that interkingdom and intrakingdom ecological interactions were more notable and stable in N_MD than SD over the course of RT. Importantly, 2 dermotypes with robust patterns of microbial networks were identified, with the "D-dermotype" (highly diversified and dominated by Devosiaceae) composing entirely of N_MD. Dermatitis-prediction classifiers were constructed. Classifiers I and III, which included bacterial variables with or without fungal variables, performed significantly better than classifier II, which relied solely on fungal variables. Bacteria-based classifier I yielded the best area under the curve in the test set of 94.64% (95% confidence interval, 83.58%-100%).
This prospective longitudinal study indicated an association between multi-kingdom skin microbiota and the development of significant ARD in patients with BC undergoing RT after reconstructive surgery, implying the possible application of skin microbiota in the prediction of ARD and microbial therapy in the management of ARD.
多界皮肤微生物群在急性放射性皮炎（ARD）中的临床意义尚未完全明确。我们推测，在接受重建手术后放疗（RT）的乳腺癌（BC）患者中，皮肤微生物群与ARD有关。
对103例患者放疗前后治疗侧和对侧健康侧共412份皮肤微生物群样本，采用细菌16S核糖体RNA（rRNA）V3-V4区域和真菌rRNA内转录间隔区（ITS）测序进行分析。使用放射治疗肿瘤学组（RTOG）的毒性标准对ARD进行分级。患者分为两组：无或轻度ARD亚组（N_MD，RTOG 0-1级）和重度ARD亚组（SD，RTOG≥2级）。
N_MD和SD亚组之间观察到皮肤微生物群存在显著差异，SD组中葡萄球菌属、棒状杆菌属和马拉色菌属富集，而N_MD组中罗尔斯通氏菌属和甲基嗜食菌属富集。网络分析显示，在放疗过程中，N_MD组的界间和界内生态相互作用比SD组更显著、更稳定。重要的是，鉴定出了两种具有强大微生物网络模式的皮肤型，“D-皮肤型”（高度多样化，以德沃斯氏菌科为主）完全由N_MD组成。构建了皮炎预测分类器。分类器I和III，包括有或没有真菌变量的细菌变量，其表现明显优于仅依赖真菌变量的分类器II。基于细菌的分类器I在测试集中的曲线下面积最佳，为94.64%（95%置信区间，83.58%-100%）。
这项前瞻性纵向研究表明，多界皮肤微生物群与重建手术后接受放疗的BC患者中重度ARD的发生有关，这意味着皮肤微生物群在ARD预测中的可能应用以及在ARD管理中的微生物治疗。
