Nishiwada Satoshi, Nakamura Kota, Ozu Naoki, Terai Taichi, Kohara Yuichiro, Nagai Minako, Sakata Takeshi, Doi Shunsuke, Matsuo Yasuko, Yasuda Satoshi, Tanaka Toshihiro, Sho Masayuki
Department of Surgery, Nara Medical University, 840 Shijo-cho, Kashihara, Nara, 634-8522, Japan.
Department of Surgery, Minami-Nara General Medical Center, Nara, Japan.
J Gastroenterol. 2025 May 10. doi: 10.1007/s00535-025-02260-w.
Complete oncological local control is essential for a potential cure in patients with pancreatic ductal adenocarcinoma (PDAC), but predicting local recurrence following curative surgery remains clinically challenging. In this study, we performed comprehensive biomarker discovery to identify an Axon guidance-related miRNA panel (AGMP) for risk-stratification of perivascular plexus recurrence (PPR) following curative surgery in patients with PDAC.
To identify axon guidance-related microRNAs, we performed the pathway-miRNA interaction analysis using the miRPathDB2.0. Subsequently, the predictive performance of the miRNAs was trained and validated in three independent clinical surgically resected sample cohorts and one pretreatment blood sample cohort with different disease statuses [upfront surgery cohort: n = 162 (training: n = 103, internal validation: n = 59), neoadjuvant chemoradiotherapy (NACRT) cohort: n = 217, arterial invasion cohort: n = 62, pretreatment blood sample cohort: n = 53].
The pathway-miRNA interaction analysis identified 13 miRNAs related to axon guidance pathway. Subsequently, we trained a 13-miRNA risk-prediction model, AGMP, which robustly distinguished PPR after surgery in the training cohort (AUC = 0.95). The AGMP was successfully validated in three independent cohorts (AUC: validation = 0.94, NACRT = 0.94, Arterial invasion = 0.90). Furthermore, we additionally validated the performance of AGMP in a pretreatment blood cohort, which again confirmed the robustness of risk-stratification for PPR (AUC = 0.86).
We developed a novel biomarker, AGMP that demonstrated remarkable predictive performance for PPR following curative surgery in patients with PDAC; highlighting the clinical importance of the nerve-cancer cross-talk and the hopefulness as a guidepost for designing future clinical and basic research to establish individualized treatment strategies.
完整的肿瘤学局部控制对于胰腺导管腺癌（PDAC）患者的潜在治愈至关重要，但预测根治性手术后的局部复发在临床上仍然具有挑战性。在本研究中，我们进行了全面的生物标志物发现，以确定一个轴突导向相关的miRNA面板（AGMP），用于对PDAC患者根治性手术后血管周围丛复发（PPR）进行风险分层。
为了鉴定轴突导向相关的微小RNA，我们使用miRPathDB2.0进行了通路-miRNA相互作用分析。随后，在三个独立的临床手术切除样本队列和一个具有不同疾病状态的预处理血液样本队列中对这些miRNA的预测性能进行了训练和验证[ upfront手术队列：n = 162（训练：n = 103，内部验证：n = 59），新辅助放化疗（NACRT）队列：n = 217，动脉侵犯队列：n = 62，预处理血液样本队列：n = 53]。
通路-miRNA相互作用分析确定了13个与轴突导向通路相关的miRNA。随后，我们训练了一个13-miRNA风险预测模型AGMP，该模型在训练队列中能够有力地区分术后的PPR（AUC = 0.95）。AGMP在三个独立队列中成功得到验证（AUC：验证 = 0.94，NACRT = 0.94，动脉侵犯 = 0.90）。此外，我们还在预处理血液队列中验证了AGMP的性能，再次证实了其对PPR风险分层的稳健性（AUC = 0.86）。
我们开发了一种新型生物标志物AGMP，它在PDAC患者根治性手术后对PPR表现出显著的预测性能；突出了神经-癌症相互作用的临床重要性，以及作为设计未来临床和基础研究以建立个体化治疗策略的指导的希望。
