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-thus, the actual challenge is to identify early factors of progression. We present a method of detection of CTNNB1 mutations using a targeted digital droplet PCR (ddPCR) on cell-free DNA (cfDNA) extracted from blood samples of 31 DT patients. Furthermore, we analyzed the correlation between DT evolution and plasmatic concentration of total and mutated cfDNA at the time of diagnosis. Circulating copies of CTNNB1 mutants (ctDNA) were detected in the plasma of 6 patients (33%) but their concentration was not correlated with evolution of the tumor. Concentration of total cfDNA was higher in the plasma of patients with progressive desmoids (p = 0,0009). Using a threshold <900 copies/mL of plasma to detect indolent desmoid and a threshold >1375, it was possible to predict desmoid evolution for 65% of patients by measuring the quantity of circulating DNA in their plasma as early as the time of diagnosis. Albeit showing that the detection of CTNNB1 mutants is possible in the plasma of patients harboring a desmoid tumor, the results of this preliminary study raise the hypothesis that most of the circulating DNA detected in their plasma is derived from non-neoplastic cells, most likely normal neighboring tissues being actively invaded. Our results open the perspective of using cfDNA as a biomarker to predict prognosis at the time of diagnosis and assess tumor dynamics to optimize the treatment strategy.
Keywords: CTNNB1; cfDNA; ddPCR; desmoid; prognosis.