CHICAGO – Homologous recombination (HR) deficiency is a biomarker for benefit from neoadjuvant chemotherapy in women with triple-negative breast cancer (TNBC), suggest data from the GeparSixto trial presented at the annual meeting of the American Society of Clinical Oncology.
Roughly two-thirds of women were found to have HR-deficient tumors, impairing their ability to repair DNA and thereby rendering them susceptible to DNA-damaging agents. These women had more than double the odds of a pathologic complete response (pCR).
“HR deficiency in triple-negative breast cancer appears to be an independent predictor of high pCR rates to the chemotherapies that were given in this study,” summarized Dr. Gunter von Minckwitz on behalf of the German Breast Group and Arbeitsgemeinschaft Gynakologische Onkologie-B (AGO-B) study group.
Dr. Gunter von Minckwitz
However, the findings were somewhat inconsistent. Additionally, although the highest pCR rate, at about 65%, was seen in women who had HR-deficient tumors and received carboplatin, women whose tumors were nondeficient also had some benefit from addition of this agent.
“This is why these data have to be confirmed by other studies, for example, the same measurements are currently ongoing in the CALGB 40603 study,” commented Dr. von Minckwitz, a senior physician at the University Women’s Hospital in Frankfurt. “And finally, they have to be set into the context with survival data that we expect at the end of 2015 or beginning of 2016.”
Invited discussant Dr. Pamela N. Munster, a professor at the University of California, San Francisco, noted, “Homologous recombination deficiency mediated through the host or tumor predicts for high responses to chemotherapy and platinum salts in early-stage triple-negative breast cancer.” However, “the role of carboplatin and its optimal setting remains complex.”
She asked about the prevailing clinical practice regarding this agent’s use. “Based on your presentation and your work, what’s the landscape in Europe on the incorporation of carboplatin in the neoadjuvant therapy in triple-negative breast cancer or a subselect group?”
“We made a survey in our group half a year ago, and all members said that they are using carboplatin in triple-negative disease when they get neoadjuvant treatment,” Dr. von Minckwitz replied. “I’m not sure if this was 100% of patients, but I believe it was a more general quote, and this is in concordance with our AGO guideline, which allows the use of carboplatin in general in triple-negative [breast cancer]. It still says in patients with germline mutations, there is a somewhat stronger recommendation to use it, but it’s of course not a must.”
Session attendee Dr. Rebecca Dent of the National Cancer Centre Singapore and the University of Toronto asked whether oncologists should begin clinically applying HR deficiency for patient selection.
“I don’t think that currently these data or the sample size is sufficient to support clinical use tomorrow. … We have to wait for a confirmative study,” Dr. von Minckwitz replied.
Dr. Munster, the discussant, agreed, saying, “I think the HRD percentage is actually quite high in [this] set, so the test may not be as robust as we like to see. So I think part of it is refinement of biomarkers has to be the focus of what we do in the next 10 years.”
Previous results of the randomized phase 2/3 GeparSixto trial have shown that adding carboplatin to a neoadjuvant chemotherapy backbone (paclitaxel, liposomal doxorubicin, and bevacizumab) improves the pCR rate in patients with triple-negative breast cancer, but at the cost of added toxicity (Lancet Oncol. 2014;15:747-56). “Therefore it is of importance to better define the group of patients that might have a higher benefit from the addition of carboplatin,” Dr. von Minckwitz explained.
“We know from previous studies that tumors with a decreased DNA repair capacity, for example, due to a mutation of the BRCA1 or BRCA2 gene are expected to have a higher sensitivity to DNA-damaging agents like platinum compounds. A more extensive way to measure DNA repair capacity is now possible using the HRD assay,” he noted.
The investigators assessed tumor HR deficiency among the 315 trial participants with triple-negative breast cancer. Overall, 70.5% had tumors that were HR deficient (meaning they had a high HRD score or a tumor BRCA mutation).
In the entire cohort, women with HR-deficient tumors were more likely to have a pCR, defined as absence of invasive residual disease in the breast or nodes (ypT0/is ypN0), than peers with HR-nondeficient tumors (55.9% vs. 29.8%). In multivariate analysis, HR deficiency independently predicted pCR (odds ratio, 2.51; P = .009).
Adding carboplatin improved the pCR rate significantly in women who had HR-deficient tumors (from 45.2% to 64.9%, P = .025) but also somewhat in women who had HR-nondeficient tumors (from 20.0% and 40.7%, P = .146). And there was no significant interaction between HR deficiency and carboplatin benefit.