绝经前女性辅助内分泌疗法：过犹不及

J Clin Oncol. 2016 May 16. [Epub ahead of print]

Still Refining Adjuvant Endocrine Therapy in Premenopausal Women: Not Too Much, Not Too Little.

Antonio C. Wolff, Hope S. Rugo.

The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD; University of California San Francisco Comprehensive Cancer Center, San Francisco, CA.

Oophorectomy was the first treatment described for young women with advanced breast cancer.[1] In 1995, the selective estrogen receptor (ER) modulator tamoxifen was recognized as effective adjuvant therapy for premenopausal ER-positive disease.[2] One hundred twenty years after Beatson's seminal observation, ovarian function suppression (OFS)—achieved reversibly with luteinizing hormone-releasing hormone agonists or permanently with bilateral oophorectomy—has become an established therapeutic option for the treatment of hormone receptor (HR) –positive early-stage breast cancer in premenopausal women, now further supported by two recent studies, the Suppression of Ovarian Function Trial (SOFT; ClinicalTrials.gov identifier NCT00066690) and the Trial of Exemestane and Tamoxifen (TEXT; ClinicalTrials.gov identifier NCT00066703).[3,4]

The complex history of breast cancer endocrine therapy was made more complex by several challenges. One was the accurate testing and interpretation of ER and progesterone-receptor status, which partly explain the now-defunct belief that endocrine therapy offered benefit for patients with HR-negative disease.[5,6] Another was the prevailing view that adjuvant polychemotherapy should be recommended to young women with localized breast cancer, regardless of risk.[7] This led to poor accrual into trials testing endocrine therapy without chemotherapy in premenopausal women.[8,9] Indeed, the Premenopausal Endocrine Responsive Chemotherapy Trial (PERCHE; ClinicalTrials.gov identifier NCT00066807), a study designed to test the benefit of chemotherapy added to OFS/antiestrogen therapy, closed prematurely without results.

SOFT compared 5 years of tamoxifen, OFS plus tamoxifen, and OFS plus the aromatase inhibitor (AI) exemestane in women who did not receive chemotherapy or who continued to have premenopausal estradiol levels after chemotherapy. In TEXT, premenopausal women were randomly assigned to OFS with either tamoxifen or the same AI. In broad terms, these trials demonstrated that adjuvant OFS/AI significantly reduced recurrence compared with OFS/tamoxifen.[3] Interestingly, adding OFS to tamoxifen only improved disease outcomes in women who received adjuvant chemotherapy and remained premenopausal, with further improvement if OFS was partnered with an AI.[4]

The improvements observed in SOFT and TEXT come with trade-offs, particularly in terms of toxicity and potential future adverse events.[10,11] Treatment decisions are further complicated by the fact that the decision to give chemotherapy in SOFT and TEXT was driven by the treating physician and the patient, on the basis of provider assessment of recurrence risk. Consequently, the greater absolute improvements in outcome observed in the chemotherapy cohorts were potentially driven by higher individual patient risk, as determined by routine histopathologic and clinical parameters.

In the article that accompanies this editorial, Regan et al [12] evaluated the magnitude of absolute benefit of the therapeutic options in SOFT and TEXT (tamoxifen, OFS/tamoxifen, and OFS/exemestane) against a composite measure of recurrence risk, using an approach similar to their prior report on the selection of endocrine therapy in postmenopausal women.[13] The authors evaluated absolute treatment effects as the impact on breast cancer-free interval using a continuous composite measure of recurrence risk on the basis of a Cox model including age, node status, tumor size, centrally confirmed HR-positive/human epidermal growth factor receptor 2–negative status, and centrally tested Ki-67. In the SOFT trial, they found an absolute 10% to 15% improvement in 5-year breast cancer-free interval with OFS/exemestane compared with OFS/tamoxifen or tamoxifen in those with intermediate-to-high composite risk disease who received chemotherapy. Similar improvements occurred in relatively lower-risk TEXT patients given chemotherapy who were treated with OFS/exemestane versus OFS/tamoxifen, whereas the nonchemotherapy cohorts with the lowest composite risks from both studies did well with all endocrine options.[12]

These data reinforce the importance of our ability to individualize therapy decisions, guided by biology and its impact on efficacy, toxicities, and patient preferences. Should OFS always be preferentially partnered with an AI and not tamoxifen? Should all patients who remain premenopausal after adjuvant chemotherapy receive OFS? The answer to these two questions is a definitive no; treatment choices should be tailored to risk and preference. A third and more complex question is whether patients treated with OFS plus endocrine therapy could be spared chemotherapy. Although the PERCHE trial closed prematurely, recent data indicate that clinical and molecular risk parameters can be used to identify patients with HR-positive disease who can be spared chemotherapy.[14,15]

As we attempt to optimize and maximize the role of endocrine therapy in HR-positive premenopausal breast cancer, it is illustrative to revisit the tortuous path to acceptance of OFS as effective adjuvant therapy. Studies of OFS conducted in the first half of the twentieth century led to the first randomized trial of OFS in 1948, but by the 1960s, these efforts were eclipsed by interest in the studies of adjuvant chemotherapy.[16] Much later, it became apparent that modern adjuvant chemotherapy regimens were unlikely to cause permanent amenorrhea in most women younger than age 40 years (and almost all those younger than age 35 years).[17] Additional data supported the hypothesis that cessation of ovarian function (temporary or permanent) might partly or fully explain the observed chemotherapy benefits in premenopausal women.[18] In other words, for women with lower-risk HR-positive breast cancer, adjuvant chemotherapy might simply be a toxic and possibly ineffective form of ovarian suppression.[19]

These observations, plus the now-defunct notion that adjuvant tamoxifen was ineffective in premenopausal women, led to a reawakening of interest in OFS. By the 1990s, evidence from a new generation of trials comparing OFS with no therapy in women younger than age 50 years showed that OFS improved survival, at least in the absence of chemotherapy. However, questions remained about the contribution of OFS to standard adjuvant therapies.[20] Unfortunately, the study designs testing various OFS methods alone or after chemotherapy had a fundamental historical design flaw in that none of them included tamoxifen in both arms.[21,22]

Still, these trials offered crucial new insight about OFS. For instance, the Zoladex Early Breast Cancer Research Association (ZEBRA) study showed similar disease-free survival with monthly goserelin for 2 years or six cycles of cyclophosphamide, methotrexate, and fluorouracil (CMF) in women with node-positive breast cancer, with a significant interaction between treatment and HR status.[23] Amenorrhea occurred in almost all goserelin patients by 6 months, and in more than half of those treated with CMF. By year 3, fewer goserelin patients remained amenorrheic compared with those treated with CMF (22.6% v 76.9%, respectively), [23] suggesting that continued OFS may not be required for efficacy. In fact, allowing women to resume menses could also minimize the negative cardiovascular disease and bone health effects associated with early menopause, including worse all-cause mortality and cardiovascular disease events—as observed with bilateral oophorectomy among women younger than age 50 years [24] without breast cancer—along with reducing other negative effects from premature menopause.[19]

Other issues must also be considered if, indeed, most of the benefit of adjuvant chemotherapy in premenopausal women with HR-positive disease is indirectly mediated through OFS and recovery of menses has a favorable impact on long-term health. Although women who recover some ovarian function after starting adjuvant chemotherapy (defined as an episode of menstrual bleeding during months 12-24) were shown to generally become menopausal 3 years earlier than expected, [25] most women treated with a luteinizing hormone-releasing hormone agonist for OFS recover menstrual function after completing therapy, although this is clearly age dependent.[26] Women younger than age 35 years treated with chemotherapy without endocrine therapy faired more poorly than did older women, [18] whereas other studies suggested that the benefit from OFS given after chemotherapy was limited to women younger than age 40 years, [26,27] reaffirming the notion that chemotherapy may function primarily as a form of OFS for at least some cancers.

Questions about how to optimally integrate OFS and chemotherapy in premenopausal women receiving tamoxifen, along with growing interest on testing AIs plus OFS, triggered the large international collaboration led by the International Breast Cancer Study Group to help isolate the contribution from each of these therapeutic elements. Although the PERCHE trial failed to accrue, the SOFT and TEXT trials provide definitive evidence about the benefit of OFS as well as AIs in premenopausal women with HR-positive early-stage breast cancer. This article by Regan et al [12] further defines the patients most likely to benefit from this therapy, sparing others from toxicity.

The results from SOFT/TEXT and the modeling of risk presented by Regan et al [12] impart important information that can be applied to clinical practice. First, clinicians have a good internal molecular gauge allowing stratification between major differences in risk, reflected in the use of chemotherapy in SOFT and TEXT, and defining those most likely to benefit from more aggressive hormone therapy. Second, using a fairly simple measure of recurrence risk, physicians can recommend appropriate hormone therapy, reserving more intensive treatment for women most likely to benefit. Indeed, tamoxifen alone is excellent therapy for women with lower-risk disease, but is less effective than combined OFS/AI in those at high risk for breast cancer recurrence in the first 5 years after diagnosis.

What these data do not tell us is the role of different hormonal therapies in the reduction of longer-term risk of recurrence—that occurring after 5 years. We need the results of ongoing studies to better understand the benefit of continuing AIs after 5 years, although the impact will clearly be related to clinicopathologic risk. MA17R (ClinicalTrials.gov identifier NCT00754845) evaluated continued AI for a second 5-year course versus placebo, with data to be presented in the near future. Retrospective analyses show that genomic tests may help identify patients with residual late-recurrence risk who could benefit from longer-duration hormone therapy.[28]

In today's world of genomic tests to assess the benefit of chemotherapy, it is clear that clinicopathologic risk factors retain significance. Studies such as Microarray in Node-Negative Disease May Avoid Chemotherapy (MINDACT; ClinicalTrials.gov identifier NCT00433589) are evaluating how to integrate molecular and clinical profiling, in this case, whether patients with zero to three positive lymph nodes, a low-risk molecular prognosis on the basis of a microarray-based genomic assay, and a high-risk clinical prognosis have a good outcome regardless of chemotherapy use. In the end, it is all about balancing risks and benefits. The article by Regan et al [12] provides tools to get us closer to this balance when selecting endocrine therapy for premenopausal women.

DOI: 10.1200/JCO.2016.67.0463