ctDNA in the de-escalation of breast cancer therapy. A puzzle not solved.

Treatment of ERBB2-positive (formerly known as HER-2) breast cancer has changed dramatically to less harsh regimens that home in on disabling ERBB2 rather than attacking cancer cells more broadly. Now researchers are refining treatment further by exploring “de-escalation” strategies that will pare back the number of drugs, the amount of time they are used, or both. Along with studies of de-escalation are studies of biomarkers such as circulating tumor DNA (ctDNA) could guide treatment decisions. As the name suggests, ctDNA are pieces of DNA shed by cancer cells into the bloodstream.

Results of a secondary analysis of the DAPHNe trial reported online yesterday in JAMA Oncology showed positive results with just 12 weeks of neoadjuvant (before surgery) treatment with the combination of paclitaxel, sold under Taxol and other names; trastuzumab, sold under Herceptin and other names; and Perjeta (pertuzumab). Twelve weeks is a de-escalation from 18 to 24 weeks of “dual ERBB2 blockade” treatment that the National Comprehensive Cancer Network recommends. Both trastuzumab and Perjeta are believed to work by blocking the ERBB2 pathway that spurs cancer growth.

Lead author Paolo Tarantino, M.D., Ph.D., a clinical research fellow at the Dana-Farber Cancer Institute in Boston, and his colleagues also reported findings that suggest that measuring ctDNA might play a role as a prognostic biomarker for such patients, although they were careful to qualify that observation.

“Ultrasensitive ctDNA analyses demonstrated high baseline detection rates and near-universal clearance after abbreviated neoadjuvant therapy, supporting further investigation of ctDNA-guided de-escalation strategies in this setting,” they wrote in the conclusion of the abstract of the study, which was published as a brief report in the journal.

In an accompanying editorial, Naoto T. Ueno, M.D., Ph.D., director of the University of Hawaii Cancer Center, noted that several patients in the DAPHNe trial had ctDNA detected after surgery, but no recurrences happened. One explanation is that adjuvant therapy eradicated the cancer. Another is that such low levels of ctDNA “do not carry the same clinical risk as higher-level or persistent positivity. " Ueno sounded a note of caution. “Without prospective evidence that acting on these signals improves outcomes, early molecular detection may risk molecular overdiagnosis.” He said that thresholds, timing and other issues must be set before ctDNA results are used to change therapy.

Even so, the DAPHNe trial results are another piece of evidence in support of de-escalating breast cancer treatment and ctDNA levels having role in those efforts.

DAPHNe was a single-arm prospective phase 2 nonrandomized trial that included 98 patients (97 of whom were women) with stage 2 or stage 3 ERBB2-positive breast cancer. Patients were enrolled at academic medical centers and affiliated community practices between November 2018 and January 2020. More than half (55 of 97) of the patients in the primary trial analysis had a pathologic complete response after the abbreviated 12 weeks of treatment with paclitaxel, trastuzumab and Perjeta. The results reported yesterday showed that after approximately five years of follow-up, the five-year event-free survival in the group was 99% and the overall survival was 99%. The results from other ways of analyzing the data were similar.

Among the study participants, 57 were included in the ctDNA analysis. At baseline, 51 of the 57 had detectable ctDNAand 16 had levels that would only be detected with ultrasensitive tests. After 12 weeks of paclitaxel, trastuzumab and Perjeta and before surgery, only two had had detectable minimal residual disease. After surgery, ctDNA was detected in 5 of 56 at the initial postsurgical timepoint, they all received adjuvant therapy. Tarantino and his colleagues noted that all of the ctDNA detected after surgery was identified at low concentrations, which highlights, they wrote, “the importance of ultrasensitive detection methods for accurate monitoring.”

Ueno highlighted some of the unknowns about ctDNA testing in the context of de-escalation therapy. While baseline testing establishes detectability, testing during treatment after one or two cycles may be informative. Complete clearance may be less important to clinical outcomes and treatment choices than speed and pitch of the decline, he wrote. Ueno argued that ctDNA must be interpreted along with other factors, such as pathologic complete response, nodal status (whether the cancer has spread to lymph nodes), and genomic classification.

Ueno also advocated for a nuanced consideration of de-escalation. “The future of ERBB2-positive breast cancer treatment is not simply escalation or de-escalation,” he wrote. “It is adaptive treatment: the right intensity, at the right time, for the right patient.”