预测晚期非小细胞肺癌免疫治疗反应的深度学习模型

只有一小部分晚期非小细胞肺癌患者对免疫检查点抑制剂 (ICI) 治疗有反应。为了获得最佳的个性化非小细胞肺癌治疗，确定最有可能受益于免疫治疗的患者是非常必要的。

To develop a supervised deep learning-based ICI response prediction method; evaluate its performance alongside other known predictive biomarkers; and assess its association with clinical outcomes in patients with advanced NSCLC.

建立一种基于监督深度学习的 ICI 反应预测方法，评估其与其他已知预测性生物标志物的效果，并评估其与晚期非小细胞肺癌患者临床结局的关系。

This multicenter cohort study developed and independently validated a deep learning-based response stratification model for predicting ICI treatment outcome in patients with advanced NSCLC from whole slide hematoxylin and eosin-stained images. Images for model development and validation were obtained from 1 participating center in the US and 3 in the European Union (EU) from August 2014 to December 2022. Data analyses were performed from September 2022 to May 2024.

这项多中心队列研究开发并独立验证了一个基于深度学习的反应分层模型，用于预测来自全载玻片苏木精和伊红染色图像的晚期非小细胞肺癌患者的 ICI 治疗结果。从 2014 年 8 月至 2022 年 12 月，从美国的 1 个参与中心和欧洲联盟 (EU) 的 3 个参与中心获得了用于模型开发和验证的图像。数据分析于 2022 年 9 月至 2024 年 5 月进行。

Monotherapy with ICIs.

ICI 单药治疗。

Model performance measured by clinical end points and objective response rate (ORR) differentiation power vs other predictive biomarkers, ie, programmed death-ligand 1 (PD-L1), tumor mutational burden (TMB), and tumor-infiltrating lymphocytes (TILs).

空白通过临床终点和客观缓解率 (ORR) 分化能力相对于其他预测性生物标志物 (即程序性死亡配体 1 (PD-L1) ，肿瘤突变负荷 (TMB) 和肿瘤浸润淋巴细胞 (TIL)) 测量的模型性能。

A total of 295 581 image tiles from 958 patients (mean [SD] age, 66.0 [10.6] years; 456 [48%] females and 502 [52%] males) treated with ICI for NSCLC were included in the analysis. The US-based development cohort consisted of 614 patients with median (IQR) follow-up time of 54.5 (38.2-68.1) months, and the EU-based validation cohort, 344 patients with 43.3 (27.4-53.9) months of follow-up. The ORR to ICI was 26% in the developmental cohort and 28% in the validation cohort. The deep learning model's area under the receiver operating characteristic curve (AUC) for ORR was 0.75 (95% CI, 0.64-0.85) in the internal test set and 0.66 (95% CI, 0.60-0.72) in the validation cohort. In a multivariable analysis, the deep learning model's score was an independent predictor of ICI response in the validation cohort for both progression-free (hazard ratio, 0.56; 95% CI, 0.42-0.76; P < .001) and overall survival (hazard ratio, 0.53; 95% CI, 0.39-0.73; P < .001). The tuned deep learning model achieved a higher AUC than TMB, TILs, and PD-L1 in the internal set; in the validation cohort, it was superior to TILs and comparable with PD-L1 (AUC, 0.67; 95% CI, 0.60-0.74), with a 10-percentage point improvement in specificity. In the validation cohort, combining the deep learning model with PD-L1 scores achieved an AUC of 0.70 (95% CI, 0.63-0.76), outperforming either marker alone, with a response rate of 51% compared to 41% for PD-L1 (≥50%) alone.

分析中共纳入了958例接受 ICI 治疗 NSCLC 的患者（平均 [SD] 年龄，66.0[10.6] 岁；456例 [48%] 女性和502例 [52%] 男性）的295 581个图像片段。基于美国的开发队列包括614例患者，中位 (IQR) 随访时间为54.5(38.2-68.1) 个月，基于欧盟的验证队列包括344例患者，随访时间为43.3(27.4-53.9) 个月。开发队列中 ICI 的 ORR 为26%，验证队列中为28%。在内部测试集中，深度学习模型的 ORR 受试者工作特征曲线下面积 (AUC) 为0.75(95%CI，0.64-0.85)，在验证队列中为0.66(95%CI，0.60-0.72)。在多变量分析中，深度学习模型的评分是验证队列中无进展 ICI 反应的独立预测因素（风险比，0.56；95%CI，0.42-0.76；P < .001）和总生存期（风险比，0.53；95%CI，0.39-0.73；P < .001）.调谐深度学习模型获得的 AUC 高于内部集的TMB、TIL和PD-L1；在验证队列中，优于TIL，与 PD-L1 相当 (AUC，0.67；95%CI，0.60-0.74)，特异性提高10个百分点。在验证队列中，结合深度学习模型与 PD-L1 评分获得的 AUC 为0.70(95%CI，0.63-0.76)，优于任一标志物单药治疗，应答率为51%，而 PD-L1 单药治疗的应答率为41%(≥50%)。

A total of 295 581 image tiles from 958 patients (mean [SD] age, 66.0 [10.6] years; 456 [48%] females and 502 [52%] males) treated with ICI for NSCLC were included in the analysis. The US-based development cohort consisted of 614 patients with median (IQR) follow-up time of 54.5 (38.2-68.1) months, and the EU-based validation cohort, 344 patients with 43.3 (27.4-53.9) months of follow-up. The ORR to ICI was 26% in the developmental cohort and 28% in the validation cohort. The deep learning model's area under the receiver operating characteristic curve (AUC) for ORR was 0.75 (95% CI, 0.64-0.85) in the internal test set and 0.66 (95% CI, 0.60-0.72) in the validation cohort. In a multivariable analysis, the deep learning model's score was an independent predictor of ICI response in the validation cohort for both progression-free (hazard ratio, 0.56; 95% CI, 0.42-0.76; P < .001) and overall survival (hazard ratio, 0.53; 95% CI, 0.39-0.73; P < .001). The tuned deep learning model achieved a higher AUC than TMB, TILs, and PD-L1 in the internal set; in the validation cohort, it was superior to TILs and comparable with PD-L1 (AUC, 0.67; 95% CI, 0.60-0.74), with a 10-percentage point improvement in specificity. In the validation cohort, combining the deep learning model with PD-L1 scores achieved an AUC of 0.70 (95% CI, 0.63-0.76), outperforming either marker alone, with a response rate of 51% compared to 41% for PD-L1 (≥50%) alone.

分析中共纳入了958例接受 ICI 治疗 NSCLC 的患者（平均 [SD] 年龄，66.0[10.6] 岁；456例 [48%] 女性和502例 [52%] 男性）的295 581个图像片段。基于美国的开发队列包括614例患者，中位 (IQR) 随访时间为54.5(38.2-68.1) 个月，基于欧盟的验证队列包括344例患者，随访时间为43.3(27.4-53.9) 个月。开发队列中 ICI 的 ORR 为26%，验证队列中为28%。在内部测试集中，深度学习模型的 ORR 受试者工作特征曲线下面积 (AUC) 为0.75(95%CI，0.64-0.85)，在验证队列中为0.66(95%CI，0.60-0.72)。在多变量分析中，深度学习模型的评分是验证队列中无进展 ICI 反应的独立预测因素（风险比，0.56；95%CI，0.42-0.76；P < .001）和总生存期（风险比，0.53；95%CI，0.39-0.73；P < .001）.调谐深度学习模型获得的 AUC 高于内部集的TMB、TIL和PD-L1；在验证队列中，优于TIL，与 PD-L1 相当 (AUC，0.67；95%CI，0.60-0.74)，特异性提高10个百分点。在验证队列中，结合深度学习模型与 PD-L1 评分获得的 AUC 为0.70(95%CI，0.63-0.76)，优于任一标志物单药治疗，应答率为51%，而 PD-L1 单药治疗的应答率为41%(≥50%)。