ASCO 2017: Comparison of somatic mutation profiles from cell free DNA versus tissue in metastatic urothelial carcinoma

Chicago, IL ( In this single-institution study, the authors aim to establish the feasibility of using next-generation sequencing (NGS) to analyze cell-free DNA (cfDNA), which represents DNA fragments released into circulation. While NGS has become an integral component of tissue analysis in the search for biomarkers, predictors and potential genetic causes of cancer, NGS on cfDNA is less well established. By analyzing cfDNA, the authors feel they can overcome the spatial and temporal limitations of tissue analysis.

To that effect, they proceeded to complete NGS on cfDNA of patients with metastatic urothelial carcinoma (mUC), and compared it to the genomic profile of tissue obtained in the clinical setting. As described in a prior abstract from the same group1, the MSKCC group utilized an established pipeline tool called MSISensor to evaluate NGS data from prospectively collected tissue, as part of the MSK IMPACT study, in which they target 341-468 genes. Both cfDNA and previously collected tissue were analyzed for somatic mutations, and the outputs were then compared.

In this proof of concept study, 26 pts were included. NGS analysis of cfDNA detected ≥1 somatic mutations (range 1-21) in 69% (18/26). For 15 pts, NGS data was available from archival tissue (11 primary tumors, 3 metastases, and matched primary/metastatic tissue in 1 case). The interval between cfDNA and tissue collection ranged from 35 days to > 4 yrs. Eleven patients (73%) received intervening treatment, including 47% (7/15) with chemotherapy, 67% (10/15) with immunotherapy, and 40% (6/15) with both. In 40% (6/15), cfDNA harbored alterations not found in archival tumor tissue.

Comparison of cfDNA and archival tissue
In 73% (11/15), some mutations within archival tissue were not detected in cfDNA, including hotspot HER2 S310F and FGFR3 S249C mutations.
Tumor and cfDNA mutation profiles were identical in 20% (3/15), with the tumor/cfDNA interval in this group ranging from 35 days to < 1.5 yrs.
Somatic alterations including hotspot ERCC2 P463A and PIK3CA E545K mutations were detected in cfDNA from 3 pts where archival tumor tissue NGS failed. Thus, cfDNA identified new mutations in 50% (9/18) of pts for whom cfDNA identified somatic mutations and tissue NGS was previously attempted.

1. The time between collection of tissue and cfDNA ranged to > 4 years. It is possible that some discordance may be due to intervening treatments and mutations within the tumor itself. Identical mutation profiles were found in patients with shorter interval times.

Despite its limitation, this proof-of-concept study is important to establish a new potential source of tumor genomics. This may lead to new biomarkers and therapies for mUC. Further studies, preferably prospective with interval cfDNA collections, may help address some of the issues with the current study.

Presented By: Michael L. Cheng

Co-Authors: Catharine Kline Cipolla, Samuel Funt, Maria E. Arcila, Hikmat Al-Ahmadie, Jonathan E. Rosenberg, Dean F. Bajorin, Michael F. Berger, Dana Tsui, David B. Solit, Gopa Iyer

Institution(s): Memorial Sloan Kettering Cancer Center, New York, NY; Memorial Sloan-Kettering Cancer Center, New York, NY

Written By: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Princess Margaret Cancer Centre
Twitter: @tchandra_uromd

at the 2017 ASCO Annual Meeting - June 2 - 6, 2017 – Chicago, Illinois, USA

Abstract 4511. Iyer G. “Mismatch repair (MMR) detection in urothelial carcinoma (UC) and correlation with immune checkpoint blockade (ICB) response.”