Predictive tests, to refine the estrogen receptor assay, for the adjuvant treatment of breast cancer with tamoxifen and oral selective estrogen receptor degraders (SERD) are required. A splice variant of the corepressor NCOR2, BQ2313636.1 predicts tamoxifen resistance to adjuvant tamoxifen and AZ9496, the first oral SERD, completes phase I studies. Clin Cancer Res; 24(15); 3480–2. ©2018 AACR.
In this issue of Clinical Cancer Research, Gong and colleagues (1) and Hamilton and colleagues (2) addressed the following question: How can we improve upon the estrogen receptor (ER) as a predictive test for adjuvant tamoxifen therapy and how can we advance therapeutics for breast cancer beyond tamoxifen, aromatase inhibitors (AI), and the injectable selective estrogen receptor degrader (SERD) fulvestrant? Initial answers to these questions hold the potential to unravel the Gordian knot of cancer complexity.
The ER (3) is the key to the success or failure of adjuvant tamoxifen therapy. The translational research strategy (4, 5) of long-term adjuvant tamoxifen treatment increases survival for patients with ER-positive breast cancer. However, acquired resistance to tamoxifen develops during continuous therapy, and this resistance is unique. Laboratory studies demonstrate that ER-positive breast tumors eventually grow in tamoxifen-treated immunodeficient mice. This is a demonstration of resistance to treatment. Nevertheless, retransplantation of tumors to new generations of immunodeficient mice demonstates that the breast cancer cells are actually dependent on tamoxifen for growth (6). Surprisingly, these same tumors will also grow with estrogen treatment (6). This observation provided a scientific explanation for the clinical phenomenon of a withdrawal response following tamoxifen failure (7). Mechanisms for this dualist action of estrogen and tamoxifen on breast cancer growth have subsequently been deciphered (8). However, as with all cancers, the mechanisms of resistance are multifaceted.
The SERD fulvestrant binds to ER, and the disrupted complex is targeted for ubiquitinylation and proteasomal destruction. The steroidal SERDs were first tested in the tamoxifen-stimulated immunodeficient mouse breast cancer model (9). The overall laboratory conclusion for second-line therapies following acquired tamoxifen resistance was to use a SERD (fulvestrant) or an AI to provide no estrogen signaling for the tumor to grow. Clinical trials subsequently demonstrated the veracity of the translational science (10). These treatment strategies became the standard of care.
Coregulatory molecules bind to the liganded ER complex either to enhance cell replication (coactivators) or prevent cell replication (corepressors; ref. 11). The tamoxifen/ER complex recruits dimerized NCOR2 to block growth. The finding (1) that a novel splice variant of NCOR2, BQ323636.1 (BQ), is present in some breast cancers is an interesting observation. The variant BQ dimerizes with NCOR2, thereby creating a flawed platform to recruit the necessary additional coregulatory proteins. This resistance mechanism is novel and has potential for clinical applications. Presumably, if the tamoxifen/ER complex is not emasculated by recruitment of dimerized NCOR2, then the tamoxifen/ER complex becomes stimulatory (Fig. 1).
Gong and colleagues (1) assembled 358 breast cancer cases that could be scored for BQ and were also ER positive. Despite this limitation, low and high nuclear BQ scores were used to predict overall survival or disease-specific survival. High nuclear BQ undermined the antitumor actions of adjuvant tamoxifen, and these data are highly significant over 20 years.
The new orally active SERD AZD9496 (Fig. 1; ref. 12) is the first to complete a phase I clinical trial (2). The novel acrylic acid side chain in AZD9496 is a common feature of a number of new orally active SERDs (13). However, the medicinal chemistry has its origins in the selective ER modulator (SERM) GW5638 (Fig. 1; ref. 14). Dose escalation of AZD9496 (2) up to 600 mg twice daily, administered to 45 heavily pretreated patients (none of whom took tamoxifen) with ER-positive/HER2-negative disease, showed one with stable disease at 1 year and one partial response. Adverse events (AE) were seven patients with ≥grade 3 AEs, three patients with dose-limiting toxicity, and one third of patients experiencing diarrhea and fatigue.
One advantage of AZD9496 is the finding that the oral SERD is effective against wild-type ER– and mutant ER (D538G and Y537S)–containing tumors (12). The mutant ERs (D538G and Y537S) are responsible for closing helix 12 of the unliganded ER during AI adjuvant therapy to D351, which produces an autostimulatory unliganded ER complex that drives recurrence (15). This combined quality of oral activity and SERD action (12) against a known mechanism of AI resistance (15) are the necessary hallmarks for the success of a new oral SERD in the treatment of recurrent breast cancer during AI adjuvant therapy.
Nevertheless, the one thing we learned about the strategic use of tamoxifen was that when treatment was given to patients with metastatic breast cancer (MBC), everyone died within 2 to 3 years. In contrast, administration of tamoxifen, as a long-term (10 year) adjuvant therapy after surgery (4), significantly increases survival. Unfortunately, the limiting factor with long-term adjuvant endocrine therapy is the real or perceived side effects. This factor decreases persistence and compliance with therapy.
A new possible approach to adjuvant therapy would be to leverage current combinations of mTOR inhibitors or CDK4/6 inhibitors with AIs used successfully for the treatment of AI-resistant recurrences. Responsiveness in the treatment of MBC is a good predictor of efficacy as an adjuvant therapy. However, current enormous cost of both mTOR inhibitors and CDK4/6 inhibitors, a lack of a predictive test for efficacy, and in the case of CDK4/6 inhibitors, up to 50% grade 3 and 4 neutropenia (discussed in ref. 13) will together have a major effect on noncompliance during long-term adjuvant therapy. Another path could be taken.
With our aging population, a 50% increase in ER-positive breast cancer will occur over the next 30 years. The incidence of ER-positive breast cancer is age related. Any oral antihormone therapy, destined to replace AIs as a long-term adjuvant therapy, should provide documented health benefits for the patients and be side effect free to maintain compliance. Future new long-term medicines to treat breast cancer must now address health care issues for elderly women.
Medicinal chemists have already devised a SERM that can reduce the incidence of osteoporotic fractures; the incidence of breast cancer; and coronary heart disease, strokes, and endometrial cancer. Perhaps it is time to return to first principles and develop the ideal SERM, but with an acrylic acid side chain that programs the ER complex for destruction. The contortion produced by an acrylic acid SERM/ER complex (16) will destroy the overproduction of tumor ER caused by current AI adjuvant therapy or by ER gene amplification. As a result, no mutated ER would occur (∼25% in AI recurrences), so there would be a 25% decrease in recurrences with a new SERM compared with AIs during adjuvant therapy. This would be a wise public health strategy proven that health benefits would enhance compliance. Savings for health care systems globally would be immense in prevention of diseases associated with menopause, as well as maintaining more patients disease free during long-term adjuvant breast cancer treatment.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
This work was supported by The University of Texas MD Anderson Cancer Center support grant to Peter Pisters (CA 016672), the benefactors of the Dallas/Ft. Worth Living Legend Chair of Cancer Research, and the George and Barbara Bush Endowment for Innovative Cancer Research.