Early Use of ctDNA Testing Can Identify Likelihood of Relapse in TNBC

Physicians treating patients with early-stage triple-negative breast cancer (TNBC) should employ circulating tumor DNA (ctDNA) testing early and often, according to Nicholas Turner, MD, PhD. He delivered results from the prospective, multicenter phase 2 cTRAK TN trial (NCT03145961), which sought to determine whether ctDNA is predictive for relapse in this patient population, during the 2021 San Antonio Breast Cancer Symposium.

“Our findings have implications for future trial design, most importantly, to start ctDNA testing early and likely to employ more sensitive ctDNA assays that track multiple variants,” said Turner, an academic consultant medical oncologist specializing in the treatment of breast cancer and a team leader in the Breast Cancer Now Toby Robins Research Centre at The Institute of Cancer Research in London, United Kingdom. “Potentially, more frequent testing is advisable in the 0- to 6-month time points, given the rapidity of disease evolution.”

He added that it may be appropriate to start ctDNA surveillance during administration of adjuvant capecitabine, and that patients with high-risk disease may not derive benefit from ctDNA surveillance because high-risk status may provide enough insight to guide treatment decisions.

In previous retrospective findings, ctDNA detection was highly predictive for relapse in patients with TNBC who have completed primary treatment. Turner and colleagues conducted what he called “the first prospective study to assess whether ctDNA assays can guide therapy for patients with early stage TNBC.”

A total 208 patients with trackable mutations who completed surgery and adjuvant chemotherapy underwent ctDNA testing at baseline, and every 3 months for 1 year. Assessment could continue for up to 18 months if COVID-19 interfered with sample collection. Investigators designed personalized polymerase chain reaction ctDNA assays tracking up to 2 mutations per patient.

“Patients who were ctDNA positive were randomized, double-blind, between observation and continued standard follow-up or intervention,” Turner said. “In the treatment arm, patients had staging, and those without metastatic disease were offered the immune checkpoint antibody pembrolizumab [Keytruda] every 3 weeks for a year.”

Investigators hypothesized 30% of patients with be positive for ctDNA within 12 months. Of those, they expected that 20% would have metastatic disease. The treatment and observations groups were powered to describe rates of ctDNA clearance, not comparison.

Turner noted that investigators made 3 changes to the protocol during recruiting: ctDNA testing was suspended during the first COVID-19 wave and active ctDNA surveillance was extended for 15 or 18 months to replace missed samples; after 154 patients were recruited, investigators mandated study entry within 6 weeks of surgery following neoadjuvant chemotherapy or before the third cycle of adjuvant chemotherapy after primary surgery; and the observation group was closed after all patients were recruited and 39 were determined to be ctDNA-positive and all subsequent ctDNA-positive patients were assigned to pembrolizumab.

The coprimary end points were rates of ctDNA detection 12 or 24 months after starting surveillance. The rates of sustained ctDNA clearance in the pembrolizumab arm were defined as the absence of detectable ctDNA 6 months after starting pembrolizumab or nonrecurrence. Secondary end points included time to ctDNA detection, detection of overt metastatic disease at the time of first ctDNA detection in the pembrolizumab arm and lead-time between ctDNA detection and recurrence in both arms.

Of 208 patients initially registered from January 2018 to December 2019, 77 were high risk and 131 were moderate risk. High-risk patients were those who had neoadjuvant chemotherapy with residual disease in the auxiliary lymph node, those who had a node-positive tumor greater than 50 mm, or those with 4 or more nodes.

Eligible patients who had received neoadjuvant chemotherapy had to have residual disease. Those who underwent primary surgery had to have a tumor of at least 20 mm or an involved lymph node.

A total of 161 patients entered ctDNA surveillance. Fifty-three were tracked for 2 mutations while the rest were tracked for 1 mutation.

Turner said that, as expected, high-risk patients had larger tumors and were more likely to have node-positive disease.

At a median follow-up of 20.4 months, all patients had completed active surveillance. At 12 months, 27.3% of patients were ctDNA-positive (95% CI, 20.6%-24.9%). Patients in the high-risk group were much more likely to be positive for ctDNA at 12 months than those at moderate risk, at 55.7% (95% CI, 42.8-69.5) vs 11.8% (95% CI, 6.9-19.8). Seven patients relapsed without prior ctDNA detection.

Forty-five ctDNA-positive patients went into the therapeutic portion of the trial; 14 were assigned to observation and 32 were assigned to pembrolizumab, including 1 patient who was initially assigned to observation. Investigators observed a rate of overt metastatic disease on staging at time of ctDNA detection of 71.9% (23/32; 95% CI, 53.3%-86.3%), substantially higher than they expected.

Nine patients were ctDNA-positive and staging negative. Four declined treatment, largely due to concerns over COVID-19. The remaining 5 started treatment with pembrolizumab. One patient remains on treatment. Three discontinued due to recurrence and 1 discontinued who later had recurrence discontinued because of adverse events (AEs).

None of these patients exhibited ctDNA clearance after 6 months on pembrolizumab. However, Turner highlighted 1 patient who had sustained ctDNA suppression and remains on treatment following treatment interruption because of an immune-related AE.

For the 14 patients assigned to observation, lead time to relapse was a median 4.1 months (95% CI, 3.2–not defined). That is shorter than what has been reported in previous retrospective results, which Turner attributed to the high proportion of high-risk patients in the cohort (64.2%).

Four patients in the observation group remain recurrence-free. All became ctDNA negative after an initial ctDNA-positive result, although 2 have since become ctDNA positive again. The 6-month recurrence-free survival rate with ctDNA clearance was 21.4% (95% CI, 4.7-50.8).

Turner noted that early ctDNA detection was associated with high rates of metastatic disease. Investigators reanalyzed ctDNA to account for the variability of time on study in calendar windows for time from surgery for the patients on neoadjuvant chemotherapy or the time from the last cycle of adjuvant chemotherapy for primary surgery patients. They found that an estimated 1.4% of patients in the moderate-risk group and 28.5% in the high-risk group had undiagnosed metastatic disease at baseline ctDNA assessment.

Further, 18.2% of patients tested ctDNA positive after 3 months on capecitabine.

“At enrollment, patients had a relatively high rate of undiagnosed metastatic disease when imaged, especially high-risk patients,” Turner said. “Relatively few patients commenced pembrolizumab, precluding an assessment of potential activity.”

In the treatment arm, 1 patient experienced 2 serious AEs during therapy, which included diarrhea, enteritis, dyspnea, and dizziness.

Reference

Turner N, Swift C, Jenkins B, et al. 1Primary results of the cTRAK TN trial: A clinical trial utilising ctDNA mutation tracking to detect minimal residual disease and trigger intervention in patients with moderate and high risk early stage triple negative breast cancer. Presented at: 2021 San Antonio Breast Cancer Symposium; December 7-10, 2021; virtual. Abstract GS3-06.