In It for the Long Haul: Long-Term Benefit With Adjuvant Endocrine Therapy for Premenopausal Women With Early-Stage Steroid Receptor–Positive Breast Cancer

Optimal management for premenopausal women with early breast cancer has been under study for nearly 75 years, but our understanding continues to be refined through the cumulative findings from iterative clinical trials. These findings have become increasingly sophisticated, thanks to recognition of the importance of the estrogen receptor–signaling pathway, the development of multiple agents targeting that pathway, and the realization that, in these women, chemotherapy may function via endocrine and nonendocrine mechanisms. The long natural history of endocrine-responsive breast cancer, which necessitates long-term follow-up, has also become evident.¹

THE TAKEAWAY

• In the article that accompanies this editorial, Johansson et al² report better distant recurrence free interval at 20 years in a randomized trial for premenopausal women with early‐stage estrogen receptor–positive breast cancer after only 2 years of tamoxifen, goserelin, or the combination when compared with placebo. The results confirm the long-term value of endocrine therapy for these young women and suggest that further work using multigene assays may help to tailor selection of endocrine therapy in this setting.

In the article that accompanies this editorial, Johansson et al² add another piece to this puzzle through a secondary analysis of the Stockholm trial (STO-5, a subset of the ZIPP trial) of 20-year benefit from adjuvant goserelin and tamoxifen in premenopausal patients with breast cancer. The analysis focused on the 584 patients with estrogen receptor (ER)–positive tumors (identified from 924 randomly assigned patients) who were randomly assigned to 2 years of goserelin, 2 years of tamoxifen, 2 years of combined tamoxifen and goserelin, or no adjuvant endocrine therapy. Lymph node–positive patients (N = 301) also received cyclophosphamide, methotrexate, and fluorouracil chemotherapy concurrent with endocrine therapy. Banked tumor specimens were centrally assayed in 2020 for ER, progesterone receptor, human epidermal growth factor receptor 2, and Ki67 expression by immunohistochemistry and transcriptional profiling using the 70-gene Agilent microarray assay. The analysis showed that goserelin, tamoxifen, and the combination significantly improved distant recurrence-free interval compared with control (multivariable hazard ratio [HR], 0.49; 95% CI, 0.32 to 0.75; HR, 0.57; 95% CI, 0.38 to 0.87; and HR, 0.63; 95% CI, 0.42 to 0.94, respectively). A provocative finding was that genomic low-risk patients as defined by the Agilent platform had long-lasting benefit from tamoxifen, whereas genomic high-risk patients benefited from goserelin, especially in the early years. Finally, the possibility of increased risk from the addition of tamoxifen to goserelin (HR, 3.36; 95% CI, 1.39 to 8.07) was seen in genomic high-risk patients.

The strengths of this report are many. They include very long follow-up thanks to an exceptional registry; the decision and ability to perform assays on banked tissues for key biomarkers in central laboratories using contemporary techniques including a commonly used genomic test; standardized use of cyclophosphamide, methotrexate, and fluorouracil chemotherapy in node-positive patients; and the inclusion of a control arm without endocrine therapy. Limitations include lack of information about survival or menopausal status after study enrollment and the relatively small size of the study. Key trial attributes to consider are the definition of ER- or progesterone receptor–positivity as ≥ 10% which probably enriched the likelihood of endocrine responsiveness of participants,³ the 2-year duration of adjuvant endocrine therapy, concurrent rather than sequential use of chemotherapy and endocrine therapy, and the use of a nonstandard tamoxifen dose of 40 mg once daily. The study design is also unusual as the study population could be viewed as a study within a study—receptor–positive patients who were identified retrospectively from the STO-5 population which is itself a subset of the ZIPP trial.

A key finding is the heartening and enduring distant recurrence-free interval benefit of only 2 years of adjuvant endocrine therapy of any type which one hopes translated into a survival benefit without excess toxicity. Optimal duration of adjuvant tamoxifen from 1 to 10 years has been studied intensively through randomized trials over many years, and the results in aggregate suggest that longer duration is better (reviewed in Davies et al⁴). Preferred duration of adjuvant ovarian function suppression (OFS) has not been systematically studied; rather, trials have generally selected its duration arbitrarily, often based on the preferred duration of tamoxifen in the trial.⁵ Trials that also use chemotherapy have been inconsistent regarding when endocrine therapy, especially OFS, is introduced. A recent update of the ASTRRA study, which randomly assigned premenopausal women with early-stage ER-positive breast cancer who had received adjuvant chemotherapy to tamoxifen for 5 years or tamoxifen for 5 years plus goserelin for 2 years showed a significant DFS advantage (but no difference in overall survival) at a median follow-up of about 8 years with addition of goserelin to tamoxifen, further supporting the value of a shorter course of OFS.⁶

An unexpected but exploratory finding of this late analysis of STO-5 is the suggestion of differential response to the type of endocrine therapy depending on genomic risk assessment using a multigene assay. Genomic low-risk patients benefited most from tamoxifen while genomic high-risk patients benefited from goserelin. Furthermore, increased risk of recurrence was seen in high-risk genomic risk patients from tamoxifen plus goserelin compared with monotherapy. These findings should be viewed as hypothesis generating given the small numbers of patients, retrospective nature of the analysis, and the use of a nonstandard dose of tamoxifen according to today's practice. An adverse interaction between tamoxifen and OFS has not been observed in other adjuvant trials such as INT 0101 (E5188) which randomly assigned 1,503 premenopausal women with node-positive, receptor–positive breast cancer to cyclophosphamide, doxorubicin, and fluorouracil (CAF) chemotherapy or CAF followed by 5 years of goserelin (CAFZ) or CAF followed by 5 years of goserelin and tamoxifen at 20 mg once daily (CAFZT).⁷ With 9.6 years of median follow-up, no significant benefit was seen with the use of CAFZ compared with CAF, and recipients of CAFZT had the best outcomes for time to recurrence and disease-free survival. No adverse interaction between tamoxifen and goserelin was identified although this larger patient population was not selected for genomic risk. Nonetheless, findings of differential efficacy as demonstrated by Johansson et al,² if replicated in other data sets, would be extremely useful to try to tailor adjuvant endocrine therapy for premenopausal women in a more precise way than through reliance on traditional clinical markers such as stage and conventional biomarkers alone. A challenge for any validation is likely to be the availability of tissues for analysis from some of the older trials of OFS and selection of the best multigene test.

An area of uncertainty is how to view the STO-5 results in the context of contemporary adjuvant endocrine therapy which often partners OFS with an aromatase inhibitor rather than tamoxifen. A recent meta-analysis used individual patient data from the four randomized trials comparing an aromatase inhibitor versus tamoxifen for 3 or 5 years in premenopausal women with ER-positive breast cancer receiving OFS or ovarian ablation.⁸ With a median follow-up of 8 years, the rate of breast cancer recurrence was lower for women randomly assigned to aromatase inhibitor than those assigned to tamoxifen (first event relative ratio, 0.79; 95% CI, 0.69 to 0.90; P = .0005). The main benefit was in years 0-4 when the therapy differed, and there was no increase or loss of benefit in subsequent years. No significant differences between treatments were observed for breast cancer mortality, death without recurrence, or all-cause mortality. Whether such an advantage will emerge with longer follow-up remains to be seen. In addition, given the large number of trials underway to define the benefit of use of extended adjuvant aromatase inhibitor for postmenopausal women with early receptor–positive breast cancer, it is not known if extended estrogen deprivation strategies would bring even more value to certain premenopausal women.⁹ How to factor in use of adjuvant bisphosphonates and cyclin-dependent kinases 4/6 inhibitors brings yet another level of complexity.

Finally, this late analysis of STO-5 must cause us to reflect on the arc of clinical progress in adjuvant therapy of breast cancer. This trial started in 1990 at a time when adjuvant therapies were relatively short in duration. Over the years longer, more intensive strategies became the norm. In the recent years, our relative success coupled with concern about toxicity has caused us to question our philosophy that more is better, leading to the current efforts to develop rational de-escalation trials to identify minimal biologically effective strategies.¹⁰ The success of these efforts will rest in part on the availability of long-term outcomes from previous trials such as STO-5. Let us recognize then that long-term follow-up, but not necessarily long-term treatment, is essential for our efforts to optimize care for breast cancer survivors.

AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

In It for the Long Haul: Long-Term Benefit With Adjuvant Endocrine Therapy for Premenopausal Women With Early-Stage Steroid Receptor–Positive Breast Cancer

The following represents disclosure information provided by the author of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

No potential conflicts of interest were reported.