Next-Generation Sequencing of Tumor Tissue Versus Blood in Patients with Advanced Urothelial Cancer

By Pedro Barata, MD, MSc, Brian Rini, MD, Jorge Garcia, MD, Pauline Funchain, MD, and Petros Grivas, MD, PhD

The ability to sequence DNA in cancer cells has helped identify unique molecular alterations, for which there are an increasing number of potential therapies that merit evaluation. Multiple next-generation sequencing (NGS) platforms are now available, but little is known about the concordance of genomic alterations between different tests, especially when different samples, e.g., tumor tissue or circulating tumor DNA (ctDNA), are used.

In our paper titled “Next-generation sequencing in cell-free circulating tumor DNA and tumor tissue in patients with advanced bladder cancer: a feasibility pilot assessment,” published recently in Annals of Oncology, we aimed to assess the concordance of genomic alterations found in two frequently used commercially available NGS assays either in blood (“liquid biopsy”, Guardant360™) or tumor tissue (FoundationOne™), in consecutive patients seen at Cleveland Clinic with advanced urothelial cancer. In our cohort, both tests were able to detect a relatively high number of genomic alterations; however, only a relatively small proportion suggested potential therapies that could be attempted. Furthermore, we found a significant discordance (> 80 percent) between the two tests, even when both tumor tissue and circulating tumor DNA were collected around the same time (which was the case in a small number of patients).

Complementary but not interchangeable

This study is one of the first to calculate concordance between NGS tests in advanced urothelial cancer while the same concept has been reported in other cancers with higher concordance rates (e.g., lung and breast cancers). There are several reasons that may explain the low concordance reported, including intra- and intertumoral heterogeneity, different assays and bioinformatics used, intervening therapies and timing of specimen collection. This low concordance suggests that while these NGS platforms may provide complementary and not interchangeable information, the integration of the data into clinical practice remains challenging, especially in the broader oncologic community. Because of this, multidisciplinary genomics tumor boards that discuss in detail NGS reports become much more frequent. Until recently, archival tumor tissue NGS was the most commonly used genomic test for clinical trial eligibility, as it has been commercially available for longer and generally has included a broader panel of genes tested. Nonetheless, “liquid biopsies” are technically and logistically easier, faster, less invasive, and may be used in a dynamic serial manner to assess resistance mechanisms, treatment response and clonal evolution.

While the number of genomic platforms is rising and we have more genomic-driven biomarker-based clinical trials, further analysis can help data interpretation and optimize clinical applications. The many ongoing prospective clinical trials that incorporate NGS screening (and possibly monitoring) in their design will likely help us better understand the results and ultimately may contribute to more available therapies for our patients.

Dr. Barata is a medical oncologist and experimental therapeutics fellow at Cleveland Clinic Cancer Center. Dr. Rini directs the Cancer Center’s Genitourinary Cancer Program. Dr. Garcia is staff in the Department of Hematology and Medical Oncology. Dr. Funchain is associate staff in the Department of Hematology and Medical Oncology.