100P - Predicting tumor ER and HER2 status using a cell-free RNA liquid biopsy assay

Background

Liquid biopsy for circulating tumor DNA is a valuable tool for cancer treatment selection and monitoring but other technologies must be explored to reveal actionable biomarker expression in the blood. We present a novel method to infer biomarker expression from cell-free non-coding small RNAs (smRNAs), including orphan non-coding RNAs (oncRNAs), which are actively secreted and stable in blood. Our platform uses smRNAs as surrogates for tumor transcriptomes and can infer biomarkers in breast cancer patients from 1 ml of plasma.

Methods

We developed a cell-free smRNA assay with an automated, CLIA-certified lab workflow. We screened the smRNA transcriptome to identify smRNAs associated with ESR1 and ERBB2 mRNA expression across breast cancer subtypes in tumor tissue (n=540). We developed a model in tumor and validated our expression scores with published RNA-seq in tumor tissue (TCGA), breast cancer cell lines with conditioned media (RNA-seq from DepMap) (n=10), and plasma from Stage III and IV breast cancer patients (n=75) with immunohistochemistry (IHC) derived hormone receptor (ER and HER2) status.

Results

From 3856 smRNA features, we developed separate smRNA scores for ESR1 and ERBB2 expression. In tumor tissue, we observed strong correlation between smRNA score and mRNA-seq for ESR1 (r=0.81, 95% CI: 0.78-0.84) and ERBB2 (r=0.79, 0.72-0.84), as well as with IHC status (ER, AUC = 0.96, 0.94-0.98; HER2, AUC = 0.87, 0.8-0.94). We confirmed that our scores were calibrated for conditioned media from common breast cancer cell lines with known ER and HER2 status, such that extracellular smRNA scores correlated with mRNA levels (r = 0.83 ERBB2, r = 0.79 ESR1). smRNA scores for baseline patient plasma predicted ER status in HER2- patients (AUC=0.79, 0.62-0.96) and HER2 status in HR- patients (AUC=0.85, 0.45-0.89).

Conclusions

We present the first use of cell-free small RNAs in an automated liquid biopsy assay to accurately infer breast cancer biomarkers from transcriptional profiles. This approach provides a sensitive option to assess potential drug target activity from a single blood draw.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Exai Bio.

Funding

Exai Bio.

Disclosure

L.S. Schwartzberg, E. Boyle, T. Cavazos, M. Karimzadeh, L. Fish, F. Hormozdiari: Financial Interests, Institutional, Full or part-time Employment: Exai Bio. J. Yen: Financial Interests, Institutional, Full or part-time Employment: Exai Bio; Financial Interests, Personal, Stocks/Shares, Former Employee: Guardant Health. H. Heydari: Financial Interests, Institutional, Financially compensated role: Exai Bio. R. Trivedi: Financial Interests, Institutional, Full or part-time Employment: Exai Bio; Financial Interests, Personal, Stocks/Shares, Former Employee: Illumina. A. Lazar: Financial Interests, Institutional, Advisory Role: Exai Bio. B. Alipanahi: Financial Interests, Institutional, Full or part-time Employment: Exai Bio; Financial Interests, Personal, Stocks/Shares: AstraZeneca, Ionis Pharmaceuticals, Moderna Therapeutics, Guardant Health, 23andMe.