Sherene Loi, MD, PhD, on Early-Stage Breast Cancer: Weighing the Prognostic Value of ctDNA Detection

The monarchE study was a trial which was looking specifically a high-risk node-positive estrogen receptor positive early-stage breast cancer patients. We know that these patients have an extremely high risk of recurrence, particularly those that have four or more nodes or larger tumors with one to three positive nodes or high grade. The monarchE study showed the benefit of two years of adjuvant abemaciclib when added to endocrine therapy. In this particular study, we looked at 910 patients from the monarchE study. These patients were cross-treatment arms, and this group was enriched for invasive disease events. This is because we wanted to look at the sensitivity of the assay. The assay's tumor-informed, which means whole exome sequencing was performed on the primary tumor, and then each assay was designed specifically for the individual patient using that information. Patients had blood samples taken at baseline, so prior to treatment, or at three, six, or at the end of 24 months of treatment. What we found in these 910 patients was that the detection of ctDNA positivity at baseline occurred in about 8% of patients. This is a very low frequency of patients had positive ctDNA by this assay compared to the 27% of patients who had events that was enriched in this study. This was despite a high-risk population. What we did find, however, that serial sampling was very informative. Some patients that were undetectable at baseline became detectable. This was a further 10% of patients, so around 17% of patients of the 27% of patients who had an event were picked up by this assay. Of most interest was the patients that were negative or undetectable at baseline that became positive. We know that they did poorly actually. Most of these patients recurred, 100% of patients recurred. There were patients that were positive at baseline that remained positive. Those patients also did poorly, 100% rates. Of most interest to me were the patients who were positive at baseline, but then became undetectable using serial sampling. These patients had a better outcome than those patients who remained positive by the ctDNA. This suggests that the treatment was having an effect on the disease, perhaps curing some patients. This suggests that clearance of ctDNA may be a useful surrogate endpoint for evaluation of future therapies in the future or in future clinical trials. Our data also suggests that patients who are positive for ctDNA at baseline and remain positive have very aggressive disease and are resistant to treatment and in the future will need some sort of rethink or different types of therapy quite early on.