Promising clinical trial results have been seen with the exploration of new and existing treatment approaches for patients with triple-negative breast cancer (TNBC) on several fronts, including chemotherapy, immunotherapy, novel antibodies, and agents targeting the PI3K/mTOR/AKT pathway, according to David W. Miles, MB BS, BSc, MD. During a presentation at the <em>1st Annual </em>Paris Breast Cancer Conference hosted by the Physicians Education Resource®, LLC, Miles, a consultant medical oncologist at West Hertfordshire Hospitals in England, provided his perspective on potential new treatment options for patients with TNBC.
David W. Miles, MB BS, BSc, MD
Promising clinical trial results have been seen with the exploration of new and existing treatment approaches for patients with triple-negative breast cancer (TNBC) on several fronts, including chemotherapy, immunotherapy, novel antibodies, and agents targeting the PI3K/mTOR/AKT pathway. Despite this progress, problems still impede future advancements, including aspects of clinical trial designs and analysis for these patients, and the lack of biomarkers to predict for responders, according to David W. Miles, MB BS, BSc, MD.
In a presentation during the1st AnnualParis Breast Cancer Conference hosted by the Physicians Education Resource®, LLC, Miles, a consultant medical oncologist at West Hertfordshire Hospitals in England, provided his perspective on potential new treatment options for patients with TNBC.
In everyday practice, treatment for patients with TNBC typically includes chemotherapy. “[For] those of us working out in the peripheral hospitals and seeing lots of patients, we are still stuck with chemotherapy,” said Miles. “We say to our patients, ‘I’m sorry the endocrine treatment isn’t going to work, I’m sorry HER2 isn’t a target, so we’re really going to hit as hard as we can with chemotherapy,’ even for those in relatively low-risk situations in the adjuvant setting.” In the phase II tnAcity study, 3 combinations of these drugs were tested in the frontline setting for patients with metastatic TNBC. The combination of nab-paclitaxel and carboplatin showed the highest progression-free survival (PFS) rate at 7.4 months, compared with 5.4 months with nab-paclitaxel and gemcitabine and gemcitabine plus carboplatin at 6.0 months.1Overall survival (OS) was also increased with the nab-paclitaxel/ carboplatin combination.
Miles noted that the trial was not progressing to phase III due to the evolving treatment landscape, but he suggested considering a platinum-containing regimen for these patients.
Miles also argued that there was a role for bevacizumab (Avastin) in TNBC, especially for patients who have received a taxane, based on prior subgroup analyses. In the phase III MERiDiAN trial, patients with HER2-negative metastatic breast cancer, including those with TNBC, received paclitaxel with or without added bevacizumab. In the overall population, the median PFS was 11.0 months with the combination versus 8.8 months with paclitaxel and placebo (hazard ratio [HR], 0.68; 95% CI, 0.51-0.91; P = .0007).2
In searching for a biomarker for response for this regimen, Miles et al tested patients for their plasma VEGF-A levels in the study. Yet the results indicated that baseline plasma VEGF-A levels are not predictive of a clinically relevant difference in treatment effect with the addition of bevacizumab. “Angiogenesis is a very complex behavior, [so] it may be naïve to think that 1 biomarker is going to do it for us,” Miles said.
IMMUNOTHERAPY AND COMBINATIONS
“I would have to describe immunotherapy as the dominant strategy at the moment,” commented Miles, as several PD-1/ PD-L1 checkpoint inhibitors are currently being tested for patients with TNBC.
In a study of atezolizumab (Tecentriq) in the frontline or in later lines of therapy, patients with TNBC showed an objective response rate (ORR) of 26% (95% CI, 9%-51%) with the PD-L1 inhibitor in the frontline; in the second line and beyond, the ORR was 7%.3He also noted that with traditional measurements in clinical trials, such as PFS, immunotherapy agents may not demonstrate significant benefit. In this trial of atezolizumab, for example, the median PFS was 1.4 months by RECIST v1.1 criteria and 1.9 months by immune-related response criteria. The benefit seen, however, was much greater when looking at duration of response, which was 21.1 months by both sets of criteria.
Additionally, the study also showed a higher ORR among patients with positive PD-L1 expression compared those with negative expression, yet Miles noted that he found PD-L1 expression to be a transient biomarker of utility. Other biomarkers tumor-infiltrating lymphocytes, CD8-positive T cells, and PD-L1 expression on immune cells—may be associated with better outcomes, but they are not yet predictive. Instead, he stressed, they may simply identify patients with a better prognosis.
Combination trials of chemotherapy with immunotherapy are inevitable, according to Miles, and have shown some encouraging findings, yet these are still only in small numbers of patients. A phase Ib/II study of pembrolizumab (Keytruda), a PD-1 inhibitor, combined with eribulin (Halaven) demonstrated promising responses in 39 evaluable patients with metastatic TNBC in the frontline setting and beyond. The ORR was 41.2% in the frontline and 27.3% in the second line and beyond, demonstrating a clear benefit earlier in the course of disease.4Miles noted that the use of eribulin here was important because many of these patients would have received a platinum-containing regimen in the first line.
“When you’ve got chemotherapy in the backbone, PD-L1 expression doesn’t differentiate those patients who respond versus those who don’t,” he noted, as the ORR in PD-L1positive patients was 29.4% compared with 33.3% in PD-L1–negative patients.
The results of several ongoing trials of immunotherapy agents may help shape the course of future trials for patients with TNBC, according to Miles. The ongoing trials include 2 studies of pembrolizumab: 1 phase III trial compares pembrolizumab with single-agent chemotherapy in previously treated patients (KEYNOTE-119; NCT02555657), and 1 compares pembrolizumab with nab-paclitaxel or paclitaxel or gemcitabine/carboplatin in the frontline (KEYNOTE-355; NCT02819518).
There are also 2 ongoing studies of atezolizumab: 1 is studying nab-paclitaxel with or without added atezolizumab in the frontline (IMpassion 130; NCT02425891), and 1 is examining paclitaxel plus or minus atezolizumab (IMpassion 131; NCT03125902) for patients with no prior therapy for metastatic or locally advanced disease.
The subset of TNBC patients who express the androgen receptor (AR) generally demonstrate a longer disease-free interval in early-stage disease, Miles said. In a single-arm, 2-stage, phase II trial of hormonal therapy with enzalutamide (Xtandi), the median PFS was 12.6 weeks (95% CI, 8.1-15.7).5For patients with AR expression ≥10%, the median PFS was 14.7 weeks, and in patients with low AR expression <10%, the median PFS was 8.1 weeks. “You can see some possible correlation between the degree of androgen receptor expression and the resultant progression-free survival,” he said.
In the patients who expressed the PREDICT AR genomic signature of an AR-driven biology, about 50% of patients in this study, the median PFS was 3.7 months, compared with 1.8 months in patients who did not express this signature. “These are what people would call hypothesis-generating for the purpose of the next study,” said Miles. The next step with this agent was supposed to be the phase III ENDEAR study of enzalutamide with or without paclitaxel in advanced TNBC, but this study was discontinued.
Glembatumumab vedotin, a glycoprotein NMB (gpNMB)-targeting antibodydrug conjugate, was investigated in the small phase II EMERGE study compared with investigator’s choice of chemotherapy, and it showed a small ORR of 6% versus 7% with chemotherapy.6However, in patients expressing gpNMB in ≥25% of their tumor cells, the ORR was 30% with glembatumumab vedotin and 9% with chemotherapy. In patients with the gpNMB marker, not just gpNMB expression, the ORR was 40% with glembatumumab vedotin versus 0% with chemotherapy in a posthoc analysis. “This to me seems very promising,” Miles said. “It seems like a validatable and an identifiable target that could be done in a relatively straightforward way, in a less complex fashion than immunology or angiogenesis.”
Findings from the EMERGE study led to the opening of the METRIC trial, which will randomize patients 2:1 to glembatumumab vedotin or capecitabine and use PFS as the primary endpoint (NCT01997333). Miles said that METRIC’s results are highly anticipated.
Sacituzumab govitecan (IMMU-132), a humanized immunoglobulin G antibody against Trop-2, received a breakthrough therapy designation for the treatment of previously treated patients with metastatic TNBC in February 2016. The phase II study showed that the ORR was 30%, the median PFS was 6.0 months (95% CI, 5.0-7.3), and the median OS was 16.6 months (95% CI, 11.1-20.6) with sacituzumab govitecan.7Miles said this looked like a promising approach.
Of particular interest to Miles was the use of MEK inhibition with cobimetinib (Cotellic), as MEK may confer resistance to taxanes. In the COLET trial, preliminary efficacy results from the safety run-in stage with paclitaxel with or without added cobimetinib showed 6 confirmed partial responses and 2 unconfirmed responses.8He noted that Adam Brufsky, MD, PhD, will present further findings from this trial during the 2017 San Antonio Breast Cancer Symposium. This combination is now being investigated with added atezolizumab and nab-paclitaxel in the first-line setting (NCT02322814).
In the LOTUS trial, patients with metastatic TNBC were treated with the AKT inhibitor ipatasertib. The study looked at PFS overall and in patients with low PTEN expression, but the study also examined OS in patients with PI3K/ AKT pathway-activated tumors as a secondary endpoint.9Miles noted that, consistent with literature, a majority of patients with PTENlow tumors by immunohistochemistry testing do not have a genetic alteration.
In the overall population (n = 124), the stratified HR for PFS was 0.60 (90% CI, 0.40-0.91), favoring the combination with ipatasertib; in the PTEN-low subgroup (n = 48), the HR was 0.59 (90% CI, 0.30-1.16); and for patients with PIK3CA/AKT1/ PTEN-altered tumors, the unstratified HR was 0.44 (90% CI, 0.20-0.87). While Miles commented that a valiant attempt has been made to explore these targets of the PI3K/mTOR/AKT pathway, “[we may] need to look at a more molecular level at that pathway and at what drives it,” he said.