During a recent OncLive Peer Exchange®, a panel of breast cancer experts discussed emerging agents in triple-negative breast cancer, which include an anti–PD-L1 antibody, 2 PARP inhibitors, and a novel antibody–drug conjugate.
Joyce A. O’Shaughnessy, MD
Although triple-negative breast cancer (TNBC) remains challenging to treat, promising clinical trial findings are paving the way for new therapies. During a recent OncLive Peer Exchange®, a panel of breast cancer experts discussed emerging agents, which include an anti—PD-L1 antibody, 2 PARP inhibitors, and a novel antibody–drug conjugate.
The need for new strategies to treat TNBCs is particularly pressing. These tumors are associated with higher rates of relapse, greater metastatic potential, and shorter overall survival (OS) compared with most other breast cancer subtypes.1 Treatment options have been limited, as agents targeting specific molecular markers have often failed to achieve clinically meaningful improvements in outcomes. Subsequently, chemotherapy has remained the standard of care. Now, recent and pending FDA approvals may change that paradigm.IMpassion130 Findings
The combination of atezolizumab (Tecentriq) plus nab-paclitaxel (Abraxane) is undergoing FDA review as a frontline therapy for TNBC. In November 2018, the FDA granted the combination a priority review designation for this indication in patients with unresectable locally advanced or metastatic PD-L1—positive tumors, based on data from the phase III IMpassion 130 trial.2 The FDA is expected to make its decision by March 12, 2019.2 If approved, atezolizumab would become the first immune checkpoint inhibitor approved for the treatment of breast cancer. IMpassion 130 (NCT02425891) randomly assigned women with untreated metastatic TNBC (N = 902) to receive the anti—PD-L1 antibody atezolizumab plus standard chemotherapy with nab-paclitaxel (n = 451) or nab-paclitaxel plus placebo (n = 451).3 Patients were treated until disease progression or unacceptable toxicity. “The population was notable for a few different reasons, one of which is that 40% were treatment-naïve, so [it was] a higher population of de novo disease than one might normally expect in their own practice. And 40% of the patients were PD-L1 positive by the VENTANA assay, which directly impacts the interpretation of the results,” Heather L. McArthur, MD, MPH, said.
The study had 2 primary endpoints: progression-free survival (PFS) in both the intention-to-treat (ITT) population and the PD-L1— positive subgroup and OS in the ITT population, with OS formally assessed in the PD-L1–positive population only upon the ITT findings showing significance. At a median follow-up of 12.9 months, atezolizumab-treated patients in both subgroups had a statistically significant PFS benefit. In the ITT population, the median PFS was 7.2 months in the atezolizumab group versus 5.5 months in the placebo group (HR for progression or death, 0.80; 95% CI, 0.69-0.92; P = .0025). In the PD-L1—positive population, the median PFS was 7.5 months in the atezolizumab group and 5.0 months in the placebo group (HR, 0.62; 95% CI, 0.49-0.78; P <.001) (Table).3 “[This] 40% reduction [in risk of progression or death] was impressive,” McArthur said. In contrast, the OS benefit in the ITT population did not reach statistical significance, with a median OS of 21.3 months in the atezolizumab group and 17.6 months in the placebo group (HR for death, 0.84; 95% CI, 0.69-1.02; P = .08). “Because the survival benefit seen in the intention- to-treat population was not significant, the trial investigators were not able to do formal testing; however, the improvement in overall survival in the PD-L1—positive subset was 9.5 months [25.0 months for atezolizumab-treated vs 15.5 months for placebo-treated], which is unprecedented. So this really represents an incredible innovation in an otherwise aggressive and devastating disease,” McArthur said, adding that the trial is “practice-changing.”
The panelists proceeded to express considerable excitement over the trial findings because atezolizumab plus nab-paclitaxel is the first immunotherapy and targeted treatment to show improved survival in patients with metastatic TNBC. “I didn’t think the results were going to be that good. I really thought that maybe we’d see some kind of PFS benefit, maybe longer in the PD-L1—positive population but no real OS benefits like we’re seeing now,” Adam M. Brufsky, MD, PhD, said. He added that the results were being driven by patients with PD-L1–positive tumors, as the greatest benefit was observed in this subset.
“I think there is something to be learned here about how we interpret studies in the TNBC space—that maybe the rules we’ve applied to cytotoxics don’t apply here. Because if the primary endpoint had been PFS alone, that may not have been as impactful. The fact that a coprimary endpoint of OS was chosen really drove home that impact,” McArthur said.
She emphasized that the atezolizumab combination may not have been explored further had OS not been an endpoint, and that “would have represented a lost opportunity.” Although it remains unclear why the OS benefit was so much greater than the PFS benefit, she noted that this phenomenon has been observed in other tumor types.
Identifying the Right Candidates for Treatment
“As soon as the FDA expands the label to add atezolizumab and nab-paclitaxel, we’ll be all ready to go,” said Joyce A. O’Shaughnessy, MD, who served as the panel moderator. She acknowledged that there are challenges with PD-L1 as a biomarker, making it difficult to identify appropriate candidates for this treatment. One such challenge is that PD-L1 positivity and negativity can result from a variety of biological mechanisms, each of which can affect the efficacy of anti—PD-L1 therapy.4
PD-L1 positivity may result from genetic events that lead to constitutive PD-L1 expression on tumor cells or inducible PD-L1 expression on tumor or noncancer cells in response to a T-cell infiltrate, whereas PD-L1 negativity may result from a tumor having no T-cell infiltrate or being unable to express PD-L1 secondary to a genetic event.4
Another challenge is that even with standardized reagents, tissue processing, and assay performance, there is no consensus on what level of PD-L1 expression serves as an appropriate cutoff between a positive and negative result. Subsequently, cutoffs in studies have varied widely, ranging from 1% to 50%.4
In IMpassion 130, the VENTANA PD-L1 SP142 assay was used to determine PD-L1 status, with expression ≥1% on tumor or immune cells considered positive.5 It is not known whether the FDA will require use of a companion diagnostic, such as VENTANA, as part of the label for atezolizumab plus nab-paclitaxel in TNBC; however, an immune biomarker subanalysis of the IMpassion 130 trial results, presented by Emens and colleagues at the SABCS meeting, revealed an unusual finding that may affect how PD-L1 expression is assessed in the future in TNBC and other malignancies.5
McArthur said this subanalysis found PD-L1 expression in only 9% of tumor cells compared with 40% of immune cells. In most cases, when tumor cells expressed PD-L1, there was also expression of PD-L1 by immune cells, and patients with BRCA mutations and/or CD8-positive cells were found to have PD-L1 positivity of immune cells.
“It really seems to be driven by the immune cell PD-L1 status rather than the tumor cell PD-L1 status, and I think that we will probably be establishing guidelines based on immune cells alone, rather than incorporating tumor cells, ” McArthur said. Historically, PD-L1 expression of tumor cells has been used to determine positivity, but in patients with TNBC, immune cell PD-L1 was found to be the most robust predictive biomarker of response to atezolizumab plus nab-paclitaxel.
“[This finding] will require some changing of the pathology methodology, but we’ll wait for the FDA guidance on that…It’s going to be about immune cells,” O’Shaughnessy said. The panelists noted that data from several other studies examining anti—PD-L1 strategies are forthcoming and will shed more light on PD-L1 as a predictive biomarker in TNBC.OlympiAD and EMBRACA Trials
In January 2018, the FDA approved olaparib (Lynparza) for patients with germline BRCA-mutated (gBRCAm), HER2-negative metastatic breast cancer who have been treated with chemotherapy in the neoadjuvant, adjuvant, or metastatic setting, making the small molecule the first PARP inhibitor in this tumor type.6 Patients with gBRCAm hormone receptor—positive disease can be considered for olaparib therapy if they receive prior endocrine therapy or are not candidates for endcrine therapy. In October 2018, the agency approved talazoparib (Talzenna) for patients with gBRCAm, HER2-negative locally advanced or metastatic breast cancer.7
The pivotal trials for the 2 drugs were similarly designed randomized phase III studies that assessed PARP inhibitors for patients with BRCA mutations: OlympiAD (NCT02000622), which compared olaparib with physician’s choice of chemotherapy (capecitabine, eribulin, or vinorelbine) for patients with gBRCAm HER2-negative (HER2-) metastatic breast cancer that was hormone receptor—positive or triple-negative, and EMBRACA (NCT01945775), which compared talazoparib with physician’s choice of chemotherapy (capecitabine, eribulin, gemcitabine, or vinorelbine) in patients with gBRCAm advanced or metastatic breast cancer.8,9
In OlympiAD, 205 patients received olaparib and 97 received chemotherapy. The median PFS was 7.0 months in the olaparib group and 4.2 months in the chemotherapy group (HR for disease progression or death, 0.58; 95% CI, 0.43- 0.80; P <.001).8 The final OS and tolerability results showed no statistically significant improvement in median OS with olaparib compared with chemotherapy (19.3 vs 17.1 months with chemotherapy at 64% data maturity, respectively; HR, 0.90; 95% CI, 0.66-1.23; P = .513).10
However, women with TNBC had a better PFS delta, Brufsky said. He added that women who received prior platinum treatment did not have as good a response. In a subgroup analysis, a median difference in OS of 7.9 months was observed with olaparib in patients who had not received chemotherapy in the metastatic setting (median OS of 22.6 vs 14.7 months with chemotherapy; HR, 0.51; 95% CI, 0.29-0.90; nominal P = .02).11
In EMBRACA, 287 patients received talazoparib and 144 received chemotherapy. The median PFS was 8.6 months and 5.6 months, respectively (HR for disease progression or death, 0.54; 95% CI, 0.41-0.71; P <.001).9 The numbers were slightly higher than observed with olaparib because the study included first-line patients, Brufsky said.
The study included patients with TNBC (n = 190) and hormone receptor—positive disease (n = 241). The PFS benefit was less pronouced among those with TNBC (HR, 0.60; 95% CI, 0.41-0.87) compared with patients with hormone receptor–positive breast cancer (HR, 0.47; 95% CI,0.32-0.71).9
Meanwhile, investigators are looking at ways to boost the efficacy of PARP inhibitors. “One combination that will be interesting for our BRCA-positive patients will be combining a PARP [inhibitor] with a checkpoint inhibitor. That would make a lot of sense because we want to try to use the PARP inhibitors as soon as we can in the natural history of the BRCA1/2 patient population, and if they’re PD-L1—positive, we want to be using checkpoint inhibitors, too,” O’Shaughnessy, said.The panelists proceeded to discuss when they use PARP inhibitors in patients with gBRCAm and PD-L1—positive tumors, a category that Brufsky said includes 8% to 10% of women with TNBC. He favors using atezolizumab and nab-paclitaxel first in these patients. “I think that there’s a clear survival benefit to the first-line therapy with atezolizumab and chemo…[and] there does seem to be benefit in later lines with a PARP inhibitor, where there was not benefit in later lines with immunotherapy,” he said.
O’Shaughnessy said she likes to use PARP inhibitors in the first line for patients with TNBC because these agents have been available, whereas checkpoint inhibitors have not. She again emphasized the potential to combine checkpoint inhibitors with PARP inhibitors. She noted that there are various trials (eg, MEDIOLA, DORA, TOPACIO) assessing such combinations across a variety of tumor types and there have not been many “surprising toxicities.”
“We also need to remember the eligibility criteria for all these different trials,” Tiffany A. Traina, MD, said. “So, for the PARP inhibitors, while patients might have had prior exposure to a platinum, they could not have progressed on a platinum. We also have to remember the disease-free interval on a study like IMpassion 130. I’m sure there will be more data coming, but I’m hoping that some of those characteristics might help in deciding which way to go for those few patients [who] have both biomarkers.”Finally, the panelists discussed sacituzumab govitecan (IMMU-132), a novel antibody—drug conjugate consisting of SN-38, the active metabolite of irinotecan, conjugated to a humanized monoclonal antibody targeting trophoblast antigen-2 (Trop-2). Trop-2 is overexpressed in many epithelial cancers and has been associated with aggressive tumor growth and decreased survival. “[It’s] overexpressed in more than 95% of triple-negative breast cancers, so it’s an antigen that’s universally expressed in [this disease],” Aditya Bardia, MD, MPH, said. He explained that sacituzumab govitecan enables higher doses of chemotherapy to be delivered to cancer cells while helping spare normal cells.
Bardia studied sacituzumab govitecan as part of a phase I/ II basket trial (NCT01631552) that included patients with metastatic TNBC, the findings of which led to the FDA to grant the agent breakthrough designation status in 2016.12 However, after discussions with the FDA, the trial was modified to only evaluate this agent in the third-line setting and beyond, and trial enrollment resumed focusing on this now more defined patient population.
The refocused study included 110 patients with metastatic TNBC, which included 53 from the previous cohort of patients who received ≥2 prior lines of therapy for metastatic disease.12 “The overall response rate was approximately 30%... The progression-free survival was in the range of 6 months. With standard chemotherapy, it’s usually in the range of 2 to 3 months,” Bardia said, noting that a global confirmatory randomized phase III trial (ASCENT; NCT02574455) is currently underway.
Bardia said that sacituzumab govitecan was granted a priority review designation for its biologics license application (BLA) and that a decision was expected in January 2019. On January 17, 2019, the drug’s manufacturer, Immunomedics, announced that it had received a complete response letter from the FDA regarding its BLA and the drug’s approvability.13 The letter cited problems regarding chemistry, manufacturing, and control matters. Immunomedic’s President and Chief Executive Officer Michael Pehl assured that no new clinical or preclinical data were needed and that the company is requesting a meeting with the FDA to resolve the identified issues, “with a goal of bringing this important medicine to patients as soon as possible.”13
During the Peer Exchange, O’Shaughnessy expressed a similar sentiment, hoping to have this treatment option available soon. “I’m personally hoping that we’re going to be able to use it sooner rather than later in triple-negative breast cancer because of the durability of the responses,” she said, adding that she would eventually like to see this agent moved up to earlier lines.