Targeting Apoptosis: A New Paradigm for the Treatment of Estrogen Receptor–Positive Breast Cancer

Approximately 70% of breast cancers express the estrogen receptor, and antiestrogen therapy is the mainstay of treatment for estrogen receptor–positive (ER⁺) patients. In contrast to traditional cytotoxic chemotherapies, which function largely by triggering cell damage and eliciting apoptosis, antiestrogen therapies work by causing cell-cycle arrest in G₁, leading to inhibition of neoplastic proliferation and macroscopic tumor stasis (1). When used as monotherapy, antiestrogen treatments such as selective estrogen receptor modulators (SERM), selective estrogen receptor degraders (SERD), or aromatase inhibitors (AI) can result in prolonged clinical responses, even in patients with widely metastatic disease.

Recent clinical advances in the treatment of metastatic ER⁺ breast cancer have built directly on this principle by adding inhibitors of CDK4/6 or mTOR to antiestrogen therapies. These newer combination treatment regimens work by reinforcing the pressure to remain in cell-cycle arrest, and even encouraging G₁ exit into a state of quiescence/senescence. This approach has resulted in a marked improvement in progression-free survival in the first- or second-line metastatic setting, and underlies the current standard of care of combining either palbociclib, ribociclib, or abemaciclib (the three FDA-approved CDK 4/6 inhibitors) with an AI or the ER downregulator fulvestrant (2–4).

Despite these significant advances, resistance to antiestrogen combination therapies is clinically inevitable, largely via mechanisms that allow for bypass of the G₁–S checkpoint and reentry into the cell cycle (5). The emergence of endocrine therapy resistance is arguably the central challenge in the treatment of ER⁺ breast cancer, and few therapeutic options exist to treat these patients. Other attempts to combat endocrine resistance have focused on targeting various hallmarks of cancer biology such as angiogenesis, immune evasion, and growth factor signaling. However, none of these strategies have led to FDA-approved therapies in ER⁺ breast cancer or moved the needle significantly with respect to clinical outcomes.

In this issue of Cancer Discovery, Lok and colleagues offer a new mechanistic strategy to address the problem of endocrine therapy resistance, building on strong preclinical work to demonstrate that the apoptotic cascade is a targetable pathway in breast cancer. This novel approach comes in the form of a phase Ib dose escalation and expansion study of the BCL2 inhibitor venetoclax, given along with tamoxifen in patients with ER⁺, BCL2-expressing metastatic breast cancer (6).

By way of background, BCL2 is a prosurvival protein and a member of the apoptosis-regulating apparatus called the BCL2 family, which plays a key role in regulating mitochondrial outer membrane permeability (MOMP), considered the point of no return in intrinsic apoptosis (7). BCL2 is expressed in approximately 75% of breast cancers and is associated with low tumor grade and ER positivity (8). Inhibition of BCL2 has been shown to be cytotoxic in breast cancer cell lines and xenograft models when used in combination with antiestrogen therapy, perhaps by exploiting mitochondrial priming caused by estrogen deprivation, and tipping the scales for cells to proceed toward MOMP and cell death (9). Venetoclax is FDA-approved for the treatment of chronic lymphocytic leukemia (CLL), small lymphocytic lymphoma (SLL), and acute myeloid leukemia (AML), but has never been studied in solid tumors.

The primary endpoint of the study was to determine the maximum tolerated dose (MTD) and dose-limiting toxicities (DLT) of venetoclax given with tamoxifen in patients with hormone receptor–positive metastatic breast cancer. In all, the drug appeared relatively well tolerated. No DLTs were observed at the maximum planned dose of 800 mg daily, and the observed toxicity profile was similar to that reported in the original phase I study of venetoclax (400 mg daily) in CLL (10). As might be expected, cytopenias predominated, with lymphopenia occurring in 29 of 33 cases (88%), neutropenia occurring in 24 of 33 cases (73%), and anemia occurring in 13 of 33 cases (39%). Gastrointestinal side effects also occurred, including nausea, vomiting, and diarrhea. MTD was not technically reached, and pharmacokinetic studies were not performed.

Secondary endpoints in the study included overall response rate (ORR) and progression-free survival (PFS). Caution must be taken when interpreting clinical outcomes from a small, nonrandomized trial; however, these early results are promising. The group observed an ORR in 15 of 33 patients (45%), with an additional 8 patients demonstrating stable disease lasting more than 24 weeks. The overall clinical benefit rate reached 75% when analysis was limited to patients receiving the maximum dose, and PFS for patients in this group was 51 weeks. Because response rates appeared better at higher doses, the question remains of whether efficacy might be improved further at doses >800 mg per day.

Correlative studies were also performed, including post-treatment biopsies in a small subset of patients and measurement of circulating tumor DNA (ctDNA). Although post-treatment biopsies were obtained too late (after 28 days) to shed light on the mechanism of treatment effect, biopsies obtained at progression interestingly suggested that loss of BCL2 expression could not account for treatment resistance. ctDNA genotyping revealed mutations in PIK3CA and ESR1 that did not seem to modulate response to therapy in this small sample; however, significant reductions in ctDNA were observed in responders, further validating the role of ctDNA in monitoring treatment effect.

These results are compelling for several reasons. First, we draw special attention to the clinical benefit that was demonstrated in 8 of 12 patients (67%) who had received >3 lines of prior therapy for metastatic disease. This finding is encouraging, as is the finding that venetoclax seemed to be able to resurrect the efficacy of tamoxifen in several previously resistant patients. It would be especially promising if this drug combination is shown to be active in patients with prior exposure to CDK4/6 inhibitors, reflecting its most likely clinical niche. To that end, a phase II trial of fulvestrant with or without venetoclax is currently ongoing in patients who have progressed on or after a CDK4/6 inhibitor (clinicaltrials.gov identifier: NCT03584009).

Looking forward, it will be interesting to explore whether expression levels of BCL2 (or other members of the BCL2 protein family) might help to predict treatment response. Developing a clinically meaningful definition of BCL2 expression will also be important as we step into an era in which biomarker-stratified treatment defines the standard of care. For instance, one could foresee a future in which patients harboring PIK3CA mutations could be treated with a SERD plus alpha-specific PI3K inhibitor (if approved based on results from the phase III SOLAR-1 trial; ref. 11), whereas PIK3CA wild-type, BCL2-overexpressing patients could be treated with a SERD plus venetoclax.

Although more robust prospective data are forthcoming, the paradigm of using BCL2 inhibitors to sensitize susceptible cells to drug-induced stress and apoptosis is intriguing and may have broad potential application in breast cancer and beyond. Within the breast cancer space specifically, future studies could potentially investigate the safety and efficacy of BCL2 inhibitors in combination with more potent antiestrogen drugs such as SERDs, or alongside PI3K inhibitors, mTOR inhibitors, or CDK4/6 inhibitors, which may synergize to elicit apoptosis. No doubt many of these trials will be under way soon.

In conclusion, this laudable translational effort by Lok and colleagues generates exciting questions about the modulation of apoptotic signaling in the treatment of cancer, and sparks optimism about the therapeutic potential of this approach. We eagerly anticipate the results of later-phase studies to elucidate the true clinical role for BCL2 as a target in breast cancer.

S. Chandarlapaty reports receiving a commercial research grant from Daichi Sankyo and is a consultant/advisory board member for Eli Lilly, Sermonix, Context Therapeutics, Sun Pharma, Revolutions Medicine, Novartis, AstraZeneca, and Chugai. K. Jhaveri reports receiving commercial research support from Novartis, Genentech, Pfizer, Novita Pharmaceuticals, ADC Therapeutics, Debio Pharmaceuticals, and Lilly Pharmaceuticals and is a consultant/advisory board member for Pfizer, Bristol-Myers Squibb, ADC Therapeutics, Jounce Therapeutics, Taiho Oncology, AstraZeneca, Novartis, and Spectrum Pharmaceuticals. No potential conflicts of interest were disclosed by the other author.

This work was supported by NCI Cancer Center Support Grant P30CA08748.