Circulating cell free tumor DNA detection as novel biomarkers to monitor desmoid tumors evolution.

Abstract

11063 Background: Since desmoid tumors (DT) exhibit an unpredictable clinical course, with stabilization and/or spontaneous regression, an initial “wait-and-see” policy is the new standard of care to select best indications of active treatments in case of significant evolution. Therefore, translational research is crucial to identify predictive factors of progression. Most DT are characterized by CTNNB1 mutation (CM) in exon 3 (T41A, S45F, S45P). Circulating cell-free tumoral DNA (ctDNA), named "liquid biopsy", has emerged as a new promising non-invasive tool to detect biomarker in several cancers. Methods: We present a method of detection of DT-specific CM using a targeted strategy digital droplet PCR (ddPCR) on cell-free DNA (cfDNA) extracted from blood samples of 31 DT patients (pts). T41A, S45P, S45F and their respective CTNNB1 wild-type probe were designed for ddPCR. Furthermore, we analyzed the correlation of cfDNA levels (CTNNB1 wt/ml plasma) and evolution of the tumor. Results: Initial DT CM statuswas known for 28 pts and unknown for 3 pts. 24 pts presented a CM (17 pts T41A, 6 pts S45F and 1 pt S45P), 2 pts a mutation of APC, 2 pts were wild-type, and 3 pts were undetermined. Among pts with a CM, CTNNB1 mutants were detected in the cfDNA of 6 patients (19.4%). CM detection was not correlated with the quantity of cfDNA analyzed (p = 0,7263–Mann-Whitney (MW)). Absolute quantification of cfDNA (CTNNB1 wt) normalized by mL of plasma displayed higher levels for patients with progressive DT (p = 0,0009 - MW), this difference of cfDNA quantity was also present between progressive, stable and self-regressive DT (p = 0,0012 - MW). A threshold of 875 CTNNB1 copies/mL predicted DT progression with a sensibility of 100% (CI95%: 59-100) and a specificity of 76.5% (CI95%: 50.1-93.2). Absolute cfDNA quantity was also higher in patients harboring multiple desmoids (p = 0.0292 - MW). Conclusions: The absolute quantification of normalized cfDNA is correlated with evolution of the disease, independently of the initial tumor type of CM. This study opens the perspective of using cfDNA as a genomic biomarker to assess the tumor dynamics at initial diagnosis, and to monitor treatment strategy in case of tumor evolution.