基于机器学习的标准组织学图像上的肿瘤浸润淋巴细胞评估与小细胞肺癌患者的免疫治疗结果的关系

SCI 1 December 2022

Association of Machine Learning–Based Assessment of Tumor-Infiltrating Lymphocytes on Standard Histologic Images With Outcomes of Immunotherapy in Patients With NSCLC 

(JAMA Oncol IF: 31.78)

Mehrdad Rakaee, PhD; Elio Adib, MD; Biagio Ricciuti, MD; Lynette M. Sholl, MD; Weiwei Shi, MD, PhD; Joao V. Alessi, MD; Alessio Cortellini, MD, PhD; Claudia A. M. Fulgenzi, MD; Patrizia Viola, MD, PhD; David J. Pinato, MD, PhD; Sayed Hashemi, MD; Idris Bahce, MD, PhD; Ilias Houda, MSc; Ezgi B. Ulas, MSc; Teodora Radonic, MD, PhD; Juha P. Väyrynen, MD, PhD; Elin Richardsen, MD, PhD; Simin Jamaly, PhD; Sigve Andersen, MD, PhD; Tom Donnem, MD, PhD; Mark M. Awad, MD, PhD; David J. Kwiatkowski, MD, PhD 

Corresponding Author: David J. Kwiatkowski, MD, PhD, Cancer Genetics Laboratory, Brigham and Women’s Hospital, 20 Shattuck St, Thorn Research Building, Boston, MA 02115 (dk@rics.bwh.harvard.edu). 

IMPORTANCE 重要性  

Currently, predictive biomarkers for response to immune checkpoint inhibitor (ICI) therapy in lung cancer are limited. Identifying such biomarkers would be useful to refine patient selection and guide precision therapy.

目前，对肺癌免疫检查点抑制剂（ICI）治疗反应的预测性生物标志物是有限的。识别这种生物标志物将有助于完善患者选择和指导精准治疗。

OBJECTIVE 目的

 To develop a machine-learning (ML)-based tumor-infiltrating lymphocytes (TILs) scoring approach, and to evaluate TIL association with clinical outcomes in patients with advanced non–small cell lung cancer (NSCLC). 

开发一种基于机器学习（ML）的肿瘤浸润淋巴细胞（TILs）评分方法，并评估TIL与晚期非小细胞肺癌（NSCLC）患者临床结果的关系。

DESIGN, SETTING, AND PARTICIPANTS 设计、设置和参与者

This multicenter retrospective discovery-validation cohort study included 685 ICI-treated patients with NSCLC with median follow-up of 38.1 and 43.3 months for the discovery (n = 446) and validation (n = 239) cohorts, respectively. Patients were treated between February 2014 and September 2021. We developed an ML automated method to count tumor, stroma, and TIL cells in whole-slide hematoxylin-eosin– stained images of NSCLC tumors. Tumor mutational burden (TMB) and programmed death ligand-1 (PD-L1) expression were assessed separately, and clinical response to ICI therapy was determined by medical record review. Data analysis was performed from June 2021 to April 2022. 

这项多中心回顾性发现-验证队列研究包括685名ICI治疗的NSCLC患者，发现队列（n = 446）和验证队列（n = 239）的中位随访时间分别为38.1和43.3月。患者在2014年2月至2021年9月期间接受治疗。我们开发了一种ML自动方法来计算NSCLC肿瘤的全切片苏木精-伊红染色图像中的肿瘤、基质和TIL细胞。肿瘤突变负担（TMB）和程序性死亡配体-1（PD-L1）的表达被分别评估，对ICI治疗的临床反应通过病历审查来确定。数据分析在2021年6月至2022年4月进行。

EXPOSURES 暴露

 All patients received anti–PD-(L)1 monotherapy. 

所有患者均接受抗PD-(L)1单药治疗。

MAIN OUTCOMES AND MEASURES 主要结果和测量方法

 Objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) were determined by blinded medical record review. The area under curve (AUC) of TIL levels, TMB, and PD-L1 in predicting ICI response were calculated using ORR. 

客观反应率（ORR）、无进展生存期（PFS）和总生存期（OS）由盲法医疗记录审查。使用ORR计算TIL级、TMB和PD-L1预测ICI反应的曲线下面积（AUC）。

RESULTS 结果

 Overall, there were 248 (56%) women in the discovery cohort and 97 (41%) in the validation cohort. In a multivariable analysis, high TIL level (250 cells/mm2) was independently associated with ICI response in both the discovery (PFS: HR, 0.71; P = .006; OS: HR, 0.74; P = .03) and validation (PFS: HR = 0.80; P = .01; OS: HR = 0.75; P = .001) cohorts. Survival benefit was seen in both first- and subsequent-line ICI treatments in patients with NSCLC. In the discovery cohort, the combined models of TILs/PD-L1 or TMB/PD-L1 had additional specificity in differentiating ICI responders compared with PD-L1 alone. In the PD-L1 negative (<1%) subgroup, TIL levels had superior classification accuracy for ICI response (AUC = 0.77) compared with TMB (AUC = 0.65). 

总体而言，发现队列中有248名女性（56%），验证队列中有97名（41%）。在多变量分析中，在发现组（PFS：HR，0.71；P = .006；OS：HR，0.74；P = .03）和验证组（PFS：HR = 0.80；P = .01；OS：HR = 0.75；P = .001），高TIL级（250细胞/mm2）与ICI反应独立相关。在NSCLC患者的一线和后续的ICI治疗中都能看到生存获益。在发现队列中，与单独的PD-L1相比，TILs/PD-L1或TMB/PD-L1的联合模型在区分ICI反应者方面具有额外的特异性。在PD-L1阴性（<1%）的亚组中，与TMB（AUC=0.65）相比，TIL级对ICI反应的分类准确性更高（AUC=0.77）。

CONCLUSIONS AND RELEVANCE 结论和意义

In these cohorts, TIL levels were robustly and independently associated with response to ICI treatment. Patient TIL assessment is relatively easily incorporated into the workflow of pathology laboratories at minimal additional cost, and may enhance precision therapy. 

在这些队列中，TIL级与ICI治疗的反应有很强的独立关联。患者的TIL评估相对来说很容易纳入病理实验室的工作流程，只需最低限度的额外费用，并可能加强精准治疗。

打赏