The potential of circulating tumor DNA (ctDNA) to reshape the design of clinical trials testing adjuvant therapy in patients with early stage cancers

BACKGROUND: The conventional approach to testing the benefit of adjuvant therapies in patients (pts) with relatively favorable prognoses is to follow a large number of pts for long periods of time, hoping that mature outcome data will document an improved outcome compared to control pts. We reasoned that the design of such trials could be improved if pts with minimal residual disease could be identified a priori through the presence of ctDNA, and the effects of adjuvant therapy then assessed through serial ctDNA assays. METHODS: We carried out a prospective trial in 231 pts with Stage II colon cancer. Serial plasma samples were collected every 3 months starting 4-10 weeks after surgery. Somatic mutations in pts' tumors were identified via sequencing of 15 genes commonly mutated in colon cancer. We then designed personalized assays to quantify ctDNA in plasma samples. Adjuvant chemotherapy was administered at clinician discretion, blinded to ctDNA analysis. RESULTS: Somatic mutations were identified in 230 (99.6%) of tumors. Matching ctDNA was detected in the immediate post-operative period in 14 of 178 (8%) pts not treated with chemotherapy, 11 of whom had recurred (79%) at a median follow-up of 27 months. In contrast, recurrence occurred in only 16 (10%) of the 164 pts with negative ctDNA not treated with chemotherapy (HR 15.66, log-rank P