Proteomic Subtyping of Small Cell Lung Cancer: Analysis of Prognosis and Treatment Strategies

Research Background

Small Cell Lung Cancer (SCLC) is a highly malignant and heterogeneous cancer characterized by rapid growth, early metastasis, and drug resistance, which limits treatment options and challenges prognostic prediction models. Current genomic analyses mainly focus on TP53 and RB1 inactivation, which are abnormal in over 98% of SCLC patients. Additionally, gene mutations in the PI3K/AKT/mTOR pathway are frequently observed. However, effective subtyping markers and therapeutic targets for SCLC remain limited. This has resulted in no significant improvement in overall patient survival despite numerous clinical trials of chemotherapy regimens and biological agents over the past decades. The current 5-year survival rate is approximately 20%-25% for limited-stage SCLC (LS-SCLC) and 1%-2% for extensive-stage SCLC (ES-SCLC), making SCLC one of the deadliest cancers.

Early cell line morphology studies classified SCLC into classic and variant subtypes, which were further validated in clinical samples. As research progressed, SCLC classification incorporated the expression of neuroendocrine transcription factors ASCL1 and/or NEUROD1 to define different subtypes. POU2F3 expression was used to define the non-neuroendocrine tuft cell variant subtype. Additionally, YAP1 was proposed as a potential subtype marker but has not been confirmed. Recent transcriptome data has reclassified SCLC into four subtypes, notably discovering the new SCLC-I subtype characterized by inflammatory gene expression.

These genomic and transcriptomic studies have greatly enhanced our understanding of SCLC but still show poor correlation with clinical prognosis. Proteomic subtyping has demonstrated excellent clinical potential in predicting prognosis, chemotherapy sensitivity, and therapeutic targets in many cancers, including gastric, liver, ovarian, colorectal, and non-small cell lung cancer (NSCLC). Therefore, the aim of this study is to improve prognostic prediction and precision treatment for SCLC patients through a proteomic subtyping model.

Paper Source

This paper was jointly authored by Zitian Huo, Yaqi Duan, Dongdong Zhan, Xizhen Xu, and Nairen Zheng, among others. The research comes from multiple institutions, including the Institute of Pathology and Department of Pathology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology; Beijing Zhongshun Diagnostic Co., Ltd.; and the Tumor Biobank of Shanxi Cancer Hospital. The study was published in “Genomics, Proteomics & Bioinformatics” on April 18, 2024.

Research Process

Research Subjects and Sample Processing

The study used 75 surgically resected SCLC FFPE (Formalin-Fixed Paraffin-Embedded) samples for proteomic analysis to develop a proteomic subtyping model. These samples were primarily analyzed using label-free quantitative mass spectrometry, detecting 7,028 high-confidence protein products, of which 2,957 proteins were identified in over 50% of the samples. Finally, 445 proteins with a coefficient of variation (CV) greater than 1.9 were selected for non-negative matrix factorization (NMF) consensus clustering analysis.

Proteomic Subtyping

NMF consensus clustering analysis identified three subtypes: S-I (28 cases), S-II (20 cases), and S-III (27 cases). These proteomic subtypes were highly correlated with overall survival (OS). Multivariate Cox regression analysis further confirmed that proteomic subtyping was an independent prognostic factor.

Chemotherapy Response

The study found significant differences in chemotherapy response among different SCLC proteomic subtypes. The S-I subtype was most sensitive to chemotherapy, while the S-III subtype had the worst prognosis and was insensitive to chemotherapy.

To validate the effectiveness of this subtyping model, the study conducted validation in an independent cohort (52 FFPE samples from the First Affiliated Hospital of Henan University). The validation results showed that the accuracy of the proteomic subtyping model could reach 90.8%.

Biological Processes and Immunotherapy Prediction

Comparing protein expression among subtypes, subtype-specific significantly altered proteins were identified. Functional enrichment analysis showed that the S-II subtype was significantly enriched in core matrix components, interferon signaling, and immune-related processes, particularly with higher expression of antigen presentation-related MHC-I molecules in S-II.

To validate the hypothesis that the S-II subtype might benefit from immunotherapy, the study collected samples from 52 extensive-stage SCLC patients who received immune checkpoint inhibitor (ICI) treatment. Analysis found that the S-II subtype showed the best progression-free survival (ICI-PFS) when receiving first-line immunotherapy.

New Therapeutic Targets

As the S-III subtype had the worst prognosis and was insensitive to chemotherapy, the study further explored potential new drug targets. Proteins highly expressed in this subtype include EGFR, AURKB, BCL-2, and EZH2, all of which have been studied as targets in various cancer treatments.

Research Results

The study demonstrated that proteomic subtyping could be highly correlated with SCLC patient prognosis and chemotherapy sensitivity. Notably, the S-I subtype showed significant survival advantages in chemotherapy, while the S-II subtype was more likely to benefit from immunotherapy. The S-III subtype exhibited the worst prognosis, indicating an urgent need for new treatment approaches. By targeting highly expressed markers in the proteome, treatment strategies for SCLC patients can be more precise.

Research Value

This study not only broadens the application of proteomics in cancer prognosis but also provides new insights for clinical treatment of SCLC. Through proteomic subtyping, patient prognosis can be better predicted, guiding the selection of chemotherapy and immunotherapy, thereby improving patient survival rates and quality of life.

Research Highlights

1. Development and Validation of Proteomic Subtyping Model: For the first time, a proteomic subtyping model was developed using label-free quantitative mass spectrometry, which was validated in an independent cohort with an accuracy of 90.8%.

2. Different Subtype Responses to Chemotherapy and Immunotherapy: Clarified the different responses of proteomic subtypes to chemotherapy and immunotherapy, particularly finding that the S-II subtype has significant survival advantages in immunotherapy.

3. New Drug Targets: Identified subtype-specific highly expressed therapeutic targets in the S-III subtype, such as EGFR, AURKB, BCL-2, and EZH2, providing new directions for precision treatment of SCLC.

Through this research, proteomic subtyping has the potential to become a new tool for clinical treatment of SCLC patients, providing theoretical basis and practical guidance for achieving personalized treatment and improving patient prognosis.