Gut Microbiome Is Associated with Radiotherapy Response in Lung Cancer Patients with Brain Metastases

Fei Liang ¹ Yichu Sun ¹ Jing Yang ² Ziqiang Shen ¹ Feng Guang Wang ¹ Rui Jiang Zhu ² Chong Zhou ^3* Youyou Xia ^1*

• ¹ Lianyungang Clinical Medical College, Nanjing Medical University, Lianyungang, China
• ² The Affiliated Lianyungang Hospital of Xuzhou Medical University/The First People's Hospital of Lianyungang, Lianyungang, Jiangsu, China
• ³ Xuzhou Central Hospital, Xuzhou, Jiangsu Province, China

Purpose: To investigate the gut microbiome of lung cancer patients with brain metastases undergoing radiotherapy, identify key microorganisms associated with radiotherapy response, and evaluate their potential as biomarkers.This study enrolled 55 newly diagnosed lung cancer patients with brain metastases. Fecal samples were collected before radiotherapy and analyzed by 16S rRNA sequencing to assess the gut microbiome's composition and function. Patients were categorized into response (n=28) and non-response (n=27) groups based on treatment efficacy, and α-diversity, β-diversity, and functional pathways were compared between them. Linear discriminant analysis was used to identify microbial features associated with treatment efficacy. Logistic regression analyses were performed to evaluate the predictive capacity of clinical and microbial factors for treatment outcomes.No significant difference in α-diversity was observed between the groups (P > 0.05), but βdiversity differed significantly (P = 0.036). Twelve characteristic microorganisms were identified in the response group, including g_ Oscillibacter and g_ Blautia, and nine in the non-response group, such as f_ Desulfovibrionaceae and g_ Megamonas. Metabolic pathways associated with treatment response included ketone body metabolism and pathways related to amyotrophic lateral sclerosis.Multivariate analysis identified g_Flavonifractor (odds ratio [OR] = 6.680, P = 0.004), g_Negativibacillus (OR = 3.862, P = 0.014), C-reactive protein (OR = 1.054, P = 0.017), and systemic inflammation response index (OR = 1.367, P = 0.043) as independent predictors of radiotherapy response. The nomogram and microbiome models achieved area under the curve (AUC) values of 0.935 and 0.866, respectively, demonstrating excellent predictive performance. Decision curve analysis further confirmed these models provided significant net benefits across risk thresholds.The composition and functional characteristics of the gut microbiome in lung cancer patients with brain metastases prior to radiotherapy are associated with therapeutic response and possess potential as predictive biomarkers. Further studies are warranted to validate these findings.