Predictive molecular tumor response through circulating DNA (cDNA) measurements and correlation with established prognostic markers in a series of solid tumors treated with multitargeted epigenetic therapy (MTET).

Abstract

e13015 Background: Epigenetic dysregulation of certain genes contribute to the progression of cancers. Plasma ctDNA concentrations can reflect these altered genes and have been shown to correlate well with tumor burden. Methods: We planned to assess the correlation value of c DNA on known factors for prediction of overall survival (OS), recognized as surrogates for survival. Ten( 10) genes of interest, namely NOTCH1, FGF2, MYC, RB1, ATM, MET, BRCA1, BRCA2, ATM and, VHL, were selected as being highly influenced by epigenetic control. We included 54 patients with reported cDNA on the genes of interest for this study evaluating the predictive impact of early changes in cDNA in response to multitargeted epigenetic treatment and the patient survival. There were 12 cases with NOTCH1 mutations, 20 cases with MYC mutations, 5 cases with RB1, 15 cases with BRCA1 or BRCA2 mutations, 26 cases with FGF2 mutations, 18 cases with MET mutations, 2 with VHL mutations and 2 cases with ATM mutations, for a total of 100 data points. Tests for statistical significance were completed using the Fisher Exact probability test calculator found at vassarstats.net/tab2x2.html. Significance level was set at α = .05. Results: There was statistically significant (p-value: .046) correlation on the response of the allele frequency of the observed NOTCH1 alteration with similar response in the angiogenesis biomarkers (Plasma VEGF) and presence of circulating tumor cells. There was a significant correlation (p-value: .014) between a decrease in AF of the observed MYC alteration and biomarker/imaging response on follow-up. In the group of 26 identified with FGF2 mutations, 22 saw cDNA alteration fraction(AF) reduction post therapy. 16/26 of them had either c Myc or NOTCH1 alterations. Also 8 /18 cases with c Met alterations shared NOTCH1 or Myc alteration. This represents a statistically significant (p-value: .0001) correlation on the response of the allele frequency of the observed FGFR2 alteration with the response of other biomarkers. Conclusions: There were a number of significant correlations observed between early changes in cDNA allele fractions of specific genes and the later observations of established prognostic markers. We conclude that changes in cDNA mutated allele fractions observed in liquid biopsy may provide an early useful independent prognostic marker for tumors carrying epigenetically influenced targets and their alterations.