Abstract C19: Use of liquid biopsies in clinical oncology: the UCSD Moores Cancer Center experience in 168 patients with diverse malignancies

Abstract

Abstract There is a growing interest in using methods such as liquid biopsies to assess genomics of circulating tumor DNA (ctDNA) in patients with advanced cancer. Liquid biopsies are non-invasive, may provide access to shed DNA from multiple metastases, and the dynamic results of genomic evolution can be followed with repeat blood sampling. We report our initial experience using ctDNA testing in 168 consecutive patients with diverse cancers. Testing was performed using a panel of 54 cancer-related genes (digital sequencing), examining mutations (in all 54 genes) as well as amplifications in ERBB2, EGFR, and MET (Guardant Health, Inc. (Guardant360 test)). Fifty-eight percent of patients (98/168) had ≥ 1 alteration(s) identified in their plasma. In a multivariable analysis, gastrointestinal cancer (N = 13) was an independent predictor of a higher number of alterations (median = 2 alterations/patient; P = 0.001), while primary brain tumors (N = 56) correlated with fewer alterations (median = 0 alterations/patient; P = 0.019) detected in the plasma. Even so, 16 of 56 patients (28.6%) with primary brain tumors had discernible ctDNA alterations. Of the 98 patients with alterations, 71.4% had ≥ 1 anomaly potentially actionable by an FDA-approved drug. When comparing results of patients (N = 63) who had both a tissue and ctDNA test performed (and had aberrations in the tissue biopsy that were detectable by the ctDNA panel), 22 individuals (35%) had ≥ 1 alteration in common between the tissue and ctDNA test (median time interval between the tissue biopsy and the blood draw = 2.7 months for these 22 patients). In contrast, 41 patients had no alterations in common between tissue and ctDNA, and in these 41 individuals, the median time between tissue biopsy and blood draw was 14.4 months (P = 0.006). Overall concordance rates were 70.3% for TP53 and EGFR, 88.1% for PIK3CA, and 93.1% for ERBB2 alterations. In addition, we observed a strong correlation between cases with ≥ 1 alteration reported with ctDNA ≥ 5% and shorter overall survival (median = 4.0 months versus not reached; P<0.001 in multivariable analysis; median follow up time = 6.1 months). Finally, 5 of 12 patients (42%) matched to a therapy according to their ctDNA test results achieved stable disease ≥ 6 months (N = 2) or partial response (N = 3). Taken together, ctDNA testing via liquid (blood) biopsies demonstrated frequent potentially actionable aberrations, including in primary brain tumors. The longer the time elapsed between tissue and ctDNA procurement for testing, the less chance that genomic alterations would be consistent between the two specimens. Higher percentage of ctDNA was an independent predictor for shorter survival. Finally, preliminary results suggest that patients matched to cognate targeted treatments based on ctDNA can achieve clinical benefit. Overall, these observations indicate that ctDNA genomic analysis may be a valuable tool for evaluating patients with cancer. Citation Format: Maria C. Schwaederle, David E. Piccioni, Santosh Kesari, Hatim Husain, Sandip P. Patel, Razelle Kurzrock. Use of liquid biopsies in clinical oncology: the UCSD Moores Cancer Center experience in 168 patients with diverse malignancies. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C19.