A random survival forest-based pathomics signature classifies immunotherapy prognosis and profiles TIME and genomics in ES-SCLC patients

Yuxin Jiang ¹, Yueying Chen ², Qinpei Cheng ²

¹School of Medicine, Southeast University, Nanjing, 210000, China.
²Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China.
³Department of Respiratory and Critical Care Medicine, Jinling Hospital, Affiliated Hospital of Nanjing Medical School, Nanjing, 210002, China.
⁴Jinling Clinical Medical College, Nanjing University of Chinese Medicine, Nanjing, 210002, China.
⁵Department of Pathology, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China.
⁶Department of Respiratory and Critical Care Medicine, School of Medicine, Jinling Hospital, Southeast University, 305 Zhongshan East Road, Nanjing, 210002, China.
⁷Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China. wq.026@163.com.
⁸School of Medicine, Southeast University, Nanjing, 210000, China. bairoushui@163.com.
⁹Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China. bairoushui@163.com.
¹⁰Department of Respiratory and Critical Care Medicine, Jinling Hospital, Affiliated Hospital of Nanjing Medical School, Nanjing, 210002, China. bairoushui@163.com.
¹¹Department of Respiratory and Critical Care Medicine, School of Medicine, Jinling Hospital, Southeast University, 305 Zhongshan East Road, Nanjing, 210002, China. bairoushui@163.com.
¹²School of Medicine, Southeast University, Nanjing, 210000, China. yong.song@nju.edu.cn.
¹³Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China. yong.song@nju.edu.cn.
¹⁴Department of Respiratory and Critical Care Medicine, Jinling Hospital, Affiliated Hospital of Nanjing Medical School, Nanjing, 210002, China. yong.song@nju.edu.cn.
¹⁵Department of Respiratory and Critical Care Medicine, School of Medicine, Jinling Hospital, Southeast University, 305 Zhongshan East Road, Nanjing, 210002, China. yong.song@nju.edu.cn.
¹⁶School of Medicine, Southeast University, Nanjing, 210000, China. zhanping207@163.com.
¹⁷Department of Respiratory and Critical Care Medicine, Affiliated Hospital of Medical School, Jinling Hospital, Nanjing University, Nanjing, 210002, China. zhanping207@163.com.
¹⁸Department of Respiratory and Critical Care Medicine, Jinling Hospital, Affiliated Hospital of Nanjing Medical School, Nanjing, 210002, China. zhanping207@163.com.
¹⁹Department of Respiratory and Critical Care Medicine, School of Medicine, Jinling Hospital, Southeast University, 305 Zhongshan East Road, Nanjing, 210002, China. zhanping207@163.com.

⁰School of Medicine, Southeast University, Nanjing, 210000, China.

Cancer Immunology, Immunotherapy : Cii +

|

October 2, 2024

View abstract on PubMed

Summary

This study developed a machine learning model using H&E images to predict survival in extensive-stage small cell lung cancer (SCLC) patients treated with chemoimmunotherapy, identifying potential biomarkers for better treatment strategies.

Area Of Science

Oncology
Computational Pathology
Immunotherapy

Background

Small cell lung cancer (SCLC) is aggressive with poor prognosis.
Limited biomarkers exist for predicting chemoimmunotherapy response in extensive-stage SCLC (ES-SCLC).
Pathomics analysis of H&E images offers a novel approach for prognostic assessment.

Purpose Of The Study

To develop a machine learning model for predicting overall survival in ES-SCLC patients receiving first-line chemoimmunotherapy.
To explore the association of pathomics features with genomic alterations and the tumor immune microenvironment (TIME).
To identify reliable noninvasive biomarkers for guiding precision medicine in ES-SCLC.

Main Methods

Retrospective analysis of 118 ES-SCLC patients treated with first-line chemoimmunotherapy.
Extraction of pathomics features from H&E-stained whole-slide images.
Development of a random survival forest (RSF) model incorporating clinical and pathomics data.
Genomic analysis (NGS, PD-L1, multiplex IHC) and single-cell RNA sequencing were performed on a subset of patients.

Main Results

The RSF model, using three pathomics features and clinical factors (liver/bone metastases, smoking, LDH), accurately predicted survival in ES-SCLC patients.
A higher RSF-Score correlated with increased CD8+ T cell infiltration and higher probabilities of MCL-1 amplification and EP300 mutation.
Single-cell analysis linked MCL-1 to TNFA-NFKB signaling and apoptosis.

Conclusions

A novel, noninvasive pathomics-based machine learning model can predict chemoimmunotherapy survival in ES-SCLC.
The model's findings suggest associations between pathomics, immune infiltration, and specific gene alterations (MCL-1, EP300).
This approach holds promise for facilitating precision medicine strategies in ES-SCLC management.

Keywords:

Biomarkers Immunotherapy Machine learning Small cell lung cancer Tumor Microenvironment