新辅助免疫疗法为非小细胞肺癌(NSCLC)患者带来了新的希望。然而，由于缺乏临床可行的标记物，目前仍然难以在治疗前选择对新辅助免疫疗法有良好反应的NSCLC患者，并预测患者的临床结果。在治疗前，我们从78名接受新辅助免疫疗法的NSCLC患者的三个队列(发现队列、训练队列和验证队列)中分离出血浆细胞外囊泡(EVs)。我们在发现队列中使用RNA测序技术(RNA-seq)鉴定差异表达的长EV核糖核酸(exLRs)。然后，我们在另外两个队列中使用定量逆转录聚合酶链式反应(qRT-PCR)建立和验证预测标记。我们从27个差异表达排名前的exLRs中鉴定出8个候选exLRs，并在训练队列中使用qRT-PCR测试其表达。我们最终确认H3C2 (P = 0.029)，MALAT1 (P = 0.043) 和 RPS3 (P = 0.0086)在对疗效有反应的患者中的显著表达，用于建立预测标记。结合PD-L1的表达情况，我们的标记在训练队列(AUC=0.892)和验证队列(AUC=0.747)中表现良好。此外，我们的标记被证明是预测接受新辅助免疫疗法的患者有利预后的因素，这证明了我们标记在临床实践中的可行性(P = 0.048)。我们的结果表明，基于exLR的标记可以准确预测NSCLC患者对新辅助免疫疗法的反应和预后。Copyright © 2023. Elsevier Ltd.出版。

Neoadjuvant immunotherapy has brought new hope for patients with non-small cell lung cancer (NSCLC). However, limited by the lack of clinically feasible markers, it is still difficult to select NSCLC patients who respond well and to predict patients' clinical outcomes before the treatment. Before the treatment, we isolated plasma extracellular vesicles (EVs) from three cohorts (discovery, training and validation) of 78 NSCLC patients treated with neoadjuvant immunotherapy. To identify differentially-expressed EV long RNAs (exLRs), we employed RNA-seq in the discovery cohort. And we subsequently used qRT-PCR to establish and validate the predictive signature in the other two cohorts. We have identified 8 candidate exLRs from 27 top-ranked exLRs differentially expressed between responders and non-responders, and tested their expression with qRT-PCR in the training cohort. We finally identified H3C2 (P = 0.029), MALAT1 (P = 0.043) and RPS3 (P = 0.0086) significantly expressed in responders for establishing the predictive signature. Integrated with PD-L1 expression, our signature performed well in predicting immunotherapeutic responses in the training (AUC=0.892) and validation cohorts (AUC=0.747). Furthermore, our signature was proven to be a predictor for favorable prognosis of patients treated with neoadjuvant immunotherapy, which demonstrates the feasibility of our signature in clinical practices (P = 0.048). Our results demonstrate that the exLR-based signature could accurately predict responses to neoadjuvant immunotherapy and prognosis in NSCLC patients.Copyright © 2023. Published by Elsevier Ltd.