A Setback for TNBC Monitoring With Circulating Tumor DNA in ZEST Trial

SAN ANTONIO — A first-in-kind trial of molecular residual disease (MRD)-guided therapy in breast cancer came to an early end because of low randomization, but insights can still be gained on the potential use for circulating tumor DNA (ctDNA) to predict disease recurrence.

The objective of the ZEST trial was to evaluate the potential of the PARP inhibitor niraparib (Zejula) to delay or prevent breast cancer recurrence in patients with MRD, defined as the presence of ctDNA after the completion of a recommended treatment course. However, phase III was terminated because it was not feasible to complete enrollment due to the low rate of ctDNA detected, reported Nicholas Turner, MD, PhD, of the Royal Marsden Hospital and Institute of Cancer Research in London.

“This largely reflected broad entry criteria that allowed a relatively large percentage of low-risk patients to enroll that resulted in a low-rate of ctDNA detection,” he explained in a presentation at the San Antonio Breast Cancer Symposium (SABCS).

“There was a tension in enrollment in this study,” he pointed out. “The study was designed in a way to allow low-risk patients to enroll, and of course these patients have a high fear of recurrence, but their ctDNA detection rates were very low, and this is a real problem for a study that wants to randomize patients between treatment arms.”

A lesson to be learned here is that “in a clinical trial that ultimately is interested in how we treat people on ctDNA detection, we need to design our studies to enroll a relatively high-risk patient population so that we have a relatively high conversion of patients into the study, and to make sure as much as possible that the patients who enroll in ctDNA surveillance are then going to be eligible for the randomization,” Turner emphasized.

ZEST included patients with stage I-III triple negative breast cancer (TNBC), or BRCA-mutated HER2-BC. Testing for ctDNA-based MRD began any time after the end of definitive treatment, and those with HR-positive disease were allowed concurrent endocrine therapy. Upon ctDNA detection, radiographic staging was performed, and those without evidence of metastatic disease were randomized to either niraparib or placebo.

As of May 2024, 2,746 patients were prescreened, 1,901 of whom had a ctDNA test result. Of these 1,901 patients, the median age was 52, 89% had TNBC, and the remaining 11% had BRCA-mutated HR-positive disease. A total of 147 (8%) were ctDNA-positive (n=135 with TNBC; n=12 with BRCA-mutated HR-positive BC).

Of those patients who had ctDNA detected, 107 either had radiologic recurrence when DNA was detected (n=73), or did not meet inclusion/exclusion criteria, leaving just 40 for randomization (n=18 to niraparib; n=22 to placebo).

While the trial was not powered to evaluate the effect of niraparib versus placebo due to early termination, recurrence-free interval was greater with niraparib, Turner and colleagues reported.

As for the timing of ctDNA detection after the end of definitive treatment, ctDNA detection rates were highest within the first 3 months of definitive treatment (6.7%), and continued to fall in the periods after definitive treatment. In TNBC, 60% of the patients who were ctDNA positive were detected within the first 6 months after the end of treatment, “consistent with the early recurrence typical of triple negative breast cancer,” Turner observed.

Turner theorized that the high rate of early recurrence at the time of a positive ctDNA test “provides support for strategies to start ctDNA testing earlier in the disease trajectory of triple-negative breast cancer.”

But SABCS invited discussant Ian Krop, MD, PhD, of the Yale School of Medicine in New Haven, Connecticut, noted that many of the patients in the study had ctDNA detected within 3 months, and that there was a high prevalence of metastatic disease at the time of that positive test.

“So it is a little worrisome that it may not be possible to get fewer patients with metastatic disease, even if we move earlier in the course of triple-negative disease,” he said.

Still, he suggested “we already have the tools” to use ctDNA to identify high-risk patients in order to evaluate a novel therapy, as attempted in ZEST. He added that in select relatively high-risk patients, testing as early as possible from the completion of standard therapy, and improving the sensitivity of ctDNA testing, “should improve the feasibility of this.”

“But I just want to be clear that we do need to formally test this approach before we start using it in the clinic,” Krop said. “It is possible that outcomes, such as survival, cannot be meaningfully changed in patients who are already significantly ctDNA positive. I don’t think that’s the case based on the data we have, but it should be formally tested.”

Krop also pointed out that like any intervention, there are potential harms associated with ctDNA testing. “We can imagine increased toxicity from starting a therapy earlier than needed or the therapy is ineffective in this situation, or patients with positive ctDNA could potentially discontinue a therapy that’s actually effective but hasn’t had a chance to clear their ctDNA at that particular moment,” he cautioned.

He also suggested this kind of testing will be associated with significant anxiety, particularly if a patient receives a positive test.

Please enable JavaScript to view the comments powered by Disqus.