4P - Optimizing utilization of antibody-drug conjugates in NSCLC by identification of subsets using RNA sequencing

Background

Antibody-drug conjugates (ADCs), composed of monoclonal antibodies linked to cytotoxic drugs, are currently approved and standard-of-care for many malignancies including primary lung adenocarcinoma (LUAD). With several ADCs targeting different tumor antigens currently in clinical trials, it is important to characterize ADC targets to optimize drug development for LUAD.

Methods

RNA-sequencing data for 537 primary LUAD and 59 normal lung tissue samples were obtained from The Cancer Genome Atlas. Profiles of ADC-targetable gene expression including ERBB2 (HER2), ERBB3 (HER3), TACSTD2 (Trop2), MET (c-Met), CEACAM5, CD276 (B7-H3) and NECTIN4 within each tumor were assessed by comparison of transcripts per million. Recurrent patterns of ADC-targetable gene expression profiles were identified by hierarchical clustering. Differential gene expression was performed using the DESeq2 software package.

Results

Differential gene expression analysis demonstrated higher expression of CEACAM5 (P=3.72E-53), TACSTD2 (P=9.62E-4) and MET (P=8.06E-10) in tumors compared to normal lung tissue. Interestingly, TACSTD2 (Trop-2) expression was inversely correlated with CEACAM5. On hierarchical clustering, we identified four distinct clusters based on ADC-targetable gene expression profiles: (1) CEACAM5 high/TACSTD2 low, (2) CEACAM5 low/TACSTD2 high, (3) MET high, (4) CEACAM5 mid/TACSTD2 mid. A subset of tumors had high expression of either CD276, ERBB2, or ERBB3; these tumors also had low expression of TACSTD2, CEACAM5, and MET.

Conclusions

In primary LUAD, CEACAM5, TACSTD2 and MET are significantly overexpressed in distinct segregable patterns. Particularly, CEACAM5 and TACSTD2 expression showed an inverse correlation and appeared to be nearly mutually exclusive biomarkers. This emphasizes the need for biomarker-based approaches for optimal selection of ADCs based on expression of target antigens.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

A. Desai: Financial Interests, Personal, Advisory Board: Amgen, Sanofi. All other authors have declared no conflicts of interest.