Prostatype® P-score在多族裔美国退伍军人队列中预测前列腺癌特异性死亡率的验证
本研究在Durham VA新诊断前列腺癌患者的多族裔队列中验证Prostatype® P-score对前列腺癌特异性死亡率(PCSM)的预测性能。P-score由三个干细胞基因(IGFBP3、F3、VGLL3)表达量与PSA、分期、分级组合计算。结果显示P-score每增加1单位,PCSM风险显著升高(HR=1.48,95%CI 1.20-1.84,p<0.001),c-index=0.87,在中危患者中仍显著预测PCSM(HR=1.43,95%CI 1.09-1.86,p=0.009)。
要点
- P-score每增加1单位,PCSM风险显著升高(HR=1.48,95%CI 1.20-1.84,p<0.001)。
- P-score准确估计PCSM(c-index=0.87),加入临床变量仅轻微提高准确性。
- 在中危患者中P-score仍显著预测PCSM(HR=1.43,95%CI 1.09-1.86,p=0.009)。
核心结论
- 在美国退伍军人队列中,P-score显著预测前列腺癌特异性死亡率。
- 加入临床变量仅轻微提高准确性,在中危患者中准确性仍高。
- P-score可作为新诊断患者风险分层的新工具,减少不匹配治疗。
研究设计与样本
- 回顾性队列研究,纳入Durham VA新诊断前列腺癌患者,经病理复核确保肿瘤组织充足。
- 对FFPE活检组织切片提取RNA,行RT-qPCR检测IGFBP3、F3、VGLL3及GAPDH(对照)。
- 结果与临床数据结合生成P-score,分析P-score与PCSM的关联。
主要发现
- P-score每增加1单位,PCSM风险显著升高(HR=1.48,95%CI 1.20-1.84,p<0.001)。
- P-score准确估计PCSM(c-index=0.87)。
- 加入临床变量仅轻微提高准确性。
- DCA显示P-score在PCSM风险5%至约50%的患者中提供净临床获益。
- 在中危患者中P-score显著预测PCSM(HR=1.43,95%CI 1.09-1.86,p=0.009)。
临床怎么用
- P-score可用于新诊断前列腺癌患者的风险分层,尤其在中危患者中辅助识别PCSM高风险者。
- P-score可通过标准化RT-qPCR assay在院内快速生成。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: The Prostatype® Test evaluates expression levels of three stem cell genes (IGFBP3, F3, and VGLL3), which are combined with PSA, stage, and grade to calculate P-score. Previous research found P-score accurately predicts prostate cancer (PC) specific mortality (PCSM) in patients with newly diagnosed clinically localized PC. We evaluated the performance of P-score to predict PCSM in a large, multiethnic cohort from the Veterans' Administration (VA). METHODS: After pathologic review to ensure sufficient tumor tissue, formalin-fixed paraffin-embedded (FFPE) biopsy cores from patients with newly diagnosed PC at the Durham VA were sent to an academic medical center. There, cores were sectioned, RNA extracted, and reverse transcription quantitative polymerase chain reaction (RT-qPCR) tests conducted for IGFBP3, F3, VGLL3, and GAPDH (control). Results were combined with clinical data to generate P-scores. The association between P-score and PCSM was evaluated using c-index, Cox and Fine-Gray models, and decision curve analysis (DCA). RESULTS: Higher P-scores were significantly associated with a higher risk of PCSM (HR = 1.48 per 1 unit increase in P-score, 95% CI: 1.20-1.84, p <0.001) and accurately estimated PCSM (c-index = 0.87). Adding clinical variables to P-score only incrementally improved accuracy. The DCA indicated P-score provided net clinical benefit for patients with PCSM risk between 5% and ~50%. As P-score strongly correlated with risk group, we tested the value of P-score in intermediate-risk patients specifically, where it significantly predicted PCSM (HR 1.43, 95% CI: 1.09-1.86, p = 0.009). CONCLUSION: In this American cohort of veterans, P-score significantly predicted PCSM. Adding clinical variables minimally improved accuracy. Accuracy remained high in intermediate-risk patients, wherein there is arguably the greatest need for better risk stratification. Given P-scores can be generated rapidly in-house using a standardized RT-qPCR assay, P-score represents a robust new tool to risk-stratify newly diagnosed patients for PC death, thereby minimizing mismatched treatments.
引用
- Mack A, Tran TD, Berglund E, et al. Validation of the Prostatype® P-score for predicting prostate cancer specific mortality in a multiethnic U.S. veterans cohort. Prostate Cancer Prostatic Dis. 2026. PMID: 41501560. doi:10.1038/s41391-025-01070-8
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。