Scholarlink

Background <p>M2 macrophages significantly contribute to the advancement of prostate cancer (PCa). This research aims to pinpoint M2 macrophage-associated genes (M2RGs) by leveraging single-cell analyses, with a focus on evaluating their prognostic and therapeutic implications in PCa.</p> Methods <p>We utilized transcriptomic and scRNA-seq datasets sourced from GEO and TCGA, analyzing both PCa and nearby non-cancerous tissues. M2 macrophage infiltration levels were quantified through “Cibersort” and “xCell” algorithms, followed by assessing their relationship with PCa outcomes. We identified M2RGs using differential expression analysis from scRNA-seq data. A risk score model (M2GS) was subsequently developed using COX and LASSO regression to predict biochemical recurrence-free survival (BRFS) and drug sensitivity. ROC curve analysis and subgroup assessments were conducted to evaluate model performance. Additionally, a nomogram integrating M2GS and clinical parameters was created to refine prediction accuracy.</p> Results <p>Higher infiltration levels of M2 macrophages were linked to poorer outcomes in patients with prostate cancer (PCa). Using COX regression and LASSO analyses, we identified seven M2 macrophage-related genes (M2RGs) with prognostic significance: MTUS1, NFE2L2, CD9, NOP56, KIF22, RBM3, and RALGDS, which were incorporated into an M2-related gene signature (M2GS). ROC analysis affirmed the model’s predictive capabilities, yielding AUC values of 0.702, 0.752, and 0.831 for predicting 1-, 3-, and 5-year survival, respectively. Subgroup analysis and violin plot comparisons highlighted distinct drug sensitivity patterns between high- and low-risk groups defined by M2GS. Both M2GS and T stage were independently validated as prognostic indicators. The nomogram demonstrated consistent calibration and strong predictive performance.</p> Conclusion <p>Our prognostic risk scoring model effectively predicts BRFS and drug responsiveness in prostate cancer, providing clinicians with valuable guidance for tailoring individualized treatment strategies and follow-up protocols for patients.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

M2 macrophage related genes predict prognosis and drug response in prostate cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月27日

查看全文
  • Zhikai Wu,
  • Jianxin Li,
  • Jintao Hu,
  • Cong Lai,
  • Zhuohang Li,
  • Hao Yu,
  • Zhihan Yuan,
  • Mingzhou Dai,
  • Juanyi Shi,
  • Cheng Liu,
  • Kewei Xu

摘要

Background

M2 macrophages significantly contribute to the advancement of prostate cancer (PCa). This research aims to pinpoint M2 macrophage-associated genes (M2RGs) by leveraging single-cell analyses, with a focus on evaluating their prognostic and therapeutic implications in PCa.

Methods

We utilized transcriptomic and scRNA-seq datasets sourced from GEO and TCGA, analyzing both PCa and nearby non-cancerous tissues. M2 macrophage infiltration levels were quantified through “Cibersort” and “xCell” algorithms, followed by assessing their relationship with PCa outcomes. We identified M2RGs using differential expression analysis from scRNA-seq data. A risk score model (M2GS) was subsequently developed using COX and LASSO regression to predict biochemical recurrence-free survival (BRFS) and drug sensitivity. ROC curve analysis and subgroup assessments were conducted to evaluate model performance. Additionally, a nomogram integrating M2GS and clinical parameters was created to refine prediction accuracy.

Results

Higher infiltration levels of M2 macrophages were linked to poorer outcomes in patients with prostate cancer (PCa). Using COX regression and LASSO analyses, we identified seven M2 macrophage-related genes (M2RGs) with prognostic significance: MTUS1, NFE2L2, CD9, NOP56, KIF22, RBM3, and RALGDS, which were incorporated into an M2-related gene signature (M2GS). ROC analysis affirmed the model’s predictive capabilities, yielding AUC values of 0.702, 0.752, and 0.831 for predicting 1-, 3-, and 5-year survival, respectively. Subgroup analysis and violin plot comparisons highlighted distinct drug sensitivity patterns between high- and low-risk groups defined by M2GS. Both M2GS and T stage were independently validated as prognostic indicators. The nomogram demonstrated consistent calibration and strong predictive performance.

Conclusion

Our prognostic risk scoring model effectively predicts BRFS and drug responsiveness in prostate cancer, providing clinicians with valuable guidance for tailoring individualized treatment strategies and follow-up protocols for patients.