Proteomic Stratification of Prognosis and Treatment Options for Small Cell Lung Cancer

Abstract

Small cell lung cancer (SCLC) is a highly malignant and heterogeneous cancer with limited therapeutic options and prognosis prediction models. Here, we analyzed formalin-fixed, paraffin-embedded (FFPE) samples of surgical resections by proteomic profiling, and stratified SCLC into three proteomic subtypes (S-I, S-II, and S-III) with distinct clinical outcomes and chemotherapy responses. The proteomic subtyping was an independent prognostic factor and performed better than current tumor-node-metastasis or Veterans Administration Lung Study Group staging methods. The subtyping results could be further validated using FFPE biopsy samples from an independent cohort, extending the analysis to both surgical and biopsy samples. The signatures of the S-II subtype in particular suggested potential benefits from immunotherapy. Differentially overexpressed proteins in S-III, the worst prognostic subtype, allowed us to nominate potential therapeutic targets, indicating that patient selection may bring new hope for previously failed clinical trials. Finally, analysis of an independent cohort of SCLC patients who had received immunotherapy validated the prediction that the S-II patients had better progression-free survival and overall survival after first-line immunotherapy. Collectively, our study provides the rationale for future clinical investigations to validate the current findings for more accurate prognosis prediction and precise treatments.

Keywords: Chemotherapy response; Immunotherapy; Prognosis; Proteomics; Small cell lung cancer.

Figure 1

Study design and key clinical characteristics of the SCLC patients from the discovery cohort A. A flowchart of our study scheme. B. Clinical characteristics of the patients from the discovery cohort. C. The Kaplan–Meier plots showing the limited predictive values on patient OS of the VALG, TNM, and ASCL1/NeuroD1 staging and classification methods. SCLC, small cell lung cancer; TNM, tumor–node–metastasis; VALG, Veterans Administration Lung Study Group; OS, overall survival; PFS, progression-free survival; ICI, immune checkpoint inhibitor; LNM, lymph node metastasis; LS, limited stage; ES, extensive stage.

Figure 2

Proteomics stratifies SCLC into subtypes that correlate with clinical outcomes A. NMF clustering yielded three subgroups in SCLC in the discovery cohort. B. The three subtypes based on proteomic profiling were associated with different clinical outcomes. C. The three proteomic subtypes exhibited different responses to chemotherapy. D. The predictive classifier model showing the prognosis trend similar to the discovery cohort. In the validation cohort, biopsy samples were obtained from the First Affiliated Hospital of Henan University, China. P values were determined by log-rank test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s., not significant). NMF, non-negative matrix factorization.

Figure 3

Differential signature proteins and enriched biological processes in proteomic SCLC subtypes A. Heatmap of signature proteins in each subtype. B. Metascape functional enrichment analysis identified different biological processes or pathways activated in the three subtypes. C. and D. The MHC-I (C) rather than MHC-II (D) molecules were highly expressed in S-II compared with S-I and S-III. P values were determined by rank sum test between two groups and Kruskal-Wallis rank sum test for all. MHC-I, major histocompatibility complex class I; MHC-II, major histocompatibility complex class II; FOT, fraction of total.

Figure 4

The S-II subtype patients are more likely to benefit from immunotherapies A. The clinical characteristics of the patients from the ICI cohort. B. and C. In the ICI cohort, the S-II patients had the best PFS time when receiving first-line immunotherapy (B) but showed no statistically obvious difference compared with S-I/S-III patients when receiving second-line immunotherapy (C). D. In all S-II patients, patients after immunotherapy had better OS than those without immunotherapy. S-II patients in the ICI cohort and S-II patients who did not receive immunotherapy in the discovery and validation cohorts were used for analysis. P values were determined by log-rank test (*, P < 0.05; ****, P < 0.0001; n.s., not significant).

Figure 5

The atlas of potential drug targets for individual patients in S-III FC, fold change; UPS, ubiquitin–proteasome system.