2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
New strategies to increase the uptake of these proven breast cancer risk-reduction interventions in general clinical practice are needed.
Rowan T. Chlebowski, MD, PhD
Updated findings from breast cancer chemoprevention trials inform our understanding regarding benefits and risks of available interventions. In full-scale, randomized chemoprevention trials, the selective estrogen receptor modulators tamoxifen and raloxifene and the aromatase inhibitors exemestane and anastrozole all reduce breast cancer incidence in postmenopausal women. However, long-term follow-up of the International Breast Cancer Intervention Study I and the National Surgical Adjuvant Breast and Bowel Project Study of Tamoxifen and Raloxifene chemoprevention trials question the primacy of tamoxifen use in the prevention setting. Long-term follow-up of the Women’s Health Initiative hormone therapy clinical trials identifies complex and changing influences on breast cancer risk during intervention and postintervention periods, but confirms findings that estrogen plus progestin increases breast cancer incidence, whereas estrogen alone decreases risk. New strategies to increase the uptake of these proven breast cancer risk-reduction interventions in general clinical practice are needed.
Both tamoxifen and raloxifene have FDA approval for breast cancer risk reduction. These recommendations are endorsed by an American Society of Clinical Oncology (ASCO) guideline,1 by the National Comprehensive Cancer Center Network (NCCN) guideline,2 and, importantly, by the US Preventive Services Task Force (USPSTF).3 Despite these recommendations, uptake of selective estrogen receptor modulators (SERMs) for breast cancer chemoprevention in clinical practice has been extremely limited.4 In addition, updated, long-term follow-up of two of these chemoprevention trials raises questions regarding the role of tamoxifen in this setting. Updated analyses from the Women’s Health Initiative (WHI) randomized hormone therapy trial (Table 2) have informed understanding of the relationships between exogenous estrogen and exogenous progestin and breast cancer risk, with relevance for breast cancer prevention.The SERM tamoxifen has been compared with placebo in 4 breast cancer prevention trials (National Surgical Adjuvant Breast and Bowel Project [NSABP] P-1,5 International Breast Cancer Intervention Study I [IBIS-I],6 the Royal Marsden Hospital Tamoxifen Prevention Trial,7 and the Italian Randomized Tamoxifen Prevention Trial),8 where, when combined in an updated meta-analysis, a statistically significant 38% reduction in invasive breast cancer incidence was seen.9 The SERM raloxifene has been compared with placebo in 3 trials (Raloxifene Use for The Heart [RUTH],10 Multiple Outcomes of Raloxifene Evaluation [MORE],11 and Continuing Outcomes Relevant to Evista [CORE]).12 In these three trials, a statistically significant reduction in breast cancer incidence was seen. Finally, the NSABP Study of Tamoxifen and Raloxifene (STAR) P-2 trial13 directly compared these agents in a breast cancer prevention setting (Table 1).47 Newer SERMs have been evaluated in full-scale trials,14,15 but, for various reasons, are unlikely to receive approval in the United States.16
In IBIS-I,6 7154 women were randomized to either tamoxifen or placebo for 5 years in a primary breast cancer prevention trial. Study findings were recently updated with 16 years’ follow-up.17 A reduced breast cancer incidence persisted throughout follow-up (214 vs 289 cases; hazard ratio [HR], 0.73; 95% CI, 0.61-0.84; P <.0001). However, there were more deaths in the tamoxifen group (187 [5.1%]) vs 166 deaths [4.6%], respectively; odds ratio [OR], 1.10; 95% CI, 0.88-1.37)17 (Table 1). In an accompanying commentary, Chlebowski questioned the greater mortality in the tamoxifen group and suggested that although the findings could reflect the play of chance in a small sample, less favorable alternative explanations include tamoxifen only decreasing breast cancers with favorable prognosis or tamoxifen increasing breast cancers with unfavorable prognosis.18
The NSABP STAR trial for breast cancer prevention also recently was updated.19 Intervention was originally for 5 years with median 9.7 years follow-up. Tamoxifen reduced invasive breast cancers compared with raloxifene (375 vs 453 cases; relative risk [RR], 1.19; P = .01).19 However, there were 5 more deaths from breast cancer in the tamoxifen group and more deaths from all causes as well (413 vs 364 deaths, respectively), a finding of borderline statistical significance (RR, 0.87; 95% CI, 0.75-1.00). Taken together, the findings from IBIS-I and the STAR trials challenge the primary role of tamoxifen for chemoprevention in postmenopausal women. These findings are difficult to understand since in an adjuvant setting, after 15 years follow-up, tamoxifen not only reduced breast cancer recurrence risk but improved overall survival, as well.20Two full-scale trials have evaluated aromatase inhibitors (AIs) for breast cancer prevention based on their efficacy in reducing contralateral breast cancer in adjuvant breast cancer trials.21 In the Mammary Prevention 3 (MAP.3) trial,22 postmenopausal women were eligible based on age alone (≥60 years) or increased breast cancer risk. A total of 4560 women were randomized to exemestane 25 mg or placebo. Exemestane reduced breast cancer risk by 65% (HR, 0.35; 95% CI, 0.18-0.70; P = .002).22 Similarly, in the International Breast Cancer Intervention Study II (IBIS-II),23 a randomized, placebo-controlled breast cancer prevention trial with 3864 postmenopausal women, the AI anastrozole reduced breast cancer risk by 53% (HR, 0.47; 95% CI, 0.32-0.68; P <.00001). In addition, there were significantly fewer other cancers in the anastrozole group, including skin and colorectal cancers (RR, 0.53; 95% CI, 0.28-0.99).
In adjuvant breast cancer trial reports, compared with tamoxifen, AIs increased fractures and substantially increased musculoskeletal complaints. In the placebo-controlled breast cancer prevention trials, a different toxicity profile emerges.22,23 In the MAP.3 trial,24 no increase in fractures was seen, the frequency of musculoskeletal side effects was lower than in adjuvant trials, and global quality of life did not differ between randomization groups. Similar findings were reported in the IBIS-II study,23 with no increase in fractures with anastrozole use and modest differences in musculoskeletal complaints between anastrozole and placebo users (7% vs 5%, respectively).
With respect to fracture-risk differences seen in the adjuvant versus prevention setting, the adjuvant trials were conducted when there was limited understanding of bone health. Subsequently, bone mineral density (BMD) monitoring and bone-targeted therapies have come into standard clinical practice. In both the MAP.3 and the IBIS-II trials, serial BMD monitoring was not a protocol requirement and bisphosphonate use was not protocol-defined. Nonetheless, in these trials, the 15% to 25% use of bisphosphonates reflected current clinical practice largely directly by primary care physicians.22,23 Thus, current AI use would not be expected to increase fracture risk in women receiving current medical management.
The difference in musculoskeletal symptoms for AI use in the adjuvant setting compared with the primary prevention setting could reflect differences in characteristics of participants. Woman in prevention trials were older; had not received therapy likely to exacerbate joint symptoms, such as chemotherapy and radiation therapy; and many were taking medications for prevention, and thus might expect few additional problems when adding another medication.25When the 2 WHI hormone therapy clinical trials, separately evaluating estrogen plus progestin in women with a uterus and estrogen alone in women with prior hysterectomy, were initiated, an increase in breast cancer was anticipated in both—but the cancers were anticipated to have favorable characteristics. The surprising results of these 2 studies have fundamentally changed understanding of the relationship of exogenous estrogen and progestin to breast cancer.26,27
The WHI estrogen-plus-progestin trial was stopped when more harm than benefit emerged for combined hormone therapy use.28 Estrogen plus progestin significantly increased breast cancer incidence,29 interfered with mammographic cancer detection,29,30 and significantly increased breast cancer mortality.31 In contrast, the WHI trial evaluating estrogen alone in postmenopausal women with prior hysterectomy found that estrogen alone significantly decreased breast cancer incidence32,33 and significantly decreased deaths from breast cancer.33
When the findings for estrogen-plus-progestin use were reported in 2003, a rapid decrease in hormone therapy use occurred in the United States and around the world,34,35 which was associated with the first decrease in breast cancer incidence in the United States in over 20 years.36,37 While subsequent reports generally supported the original findings,35 there were questions about whether the rapid drop in breast cancer was biologically feasible or was related to a reduction in mammography use.
In the WHI clinical trial, all participants were instructed to stop their study medication when the intervention ended with 98% compliance. In addition, mammography frequency was similar immediately before and after the intervention ended in both randomization groups. The end of the intervention was followed by a rapid and statistically significant reduction in breast cancer incidence.38 Taken together, these findings supported the prior proposed hypothesis. The immediate reduction in breast cancer incidence was felt to reflect the impact on preclinical but already established breast cancers from a sudden change in hormone environment, similar to an oophorectomy or AI effect in women with established breast cancer.38
After 5.6 years of estrogen-plus-progestin use, with an additional 8 years of postintervention follow-up, a persistent elevation in breast cancer risk of about 24% developed.39 A year-to-year analysis comparing the intervention period to the post- intervention period identified a complex pattern. There was a significant increase in year-to-year risk during intervention, a sudden decrease in risk post-intervention for about 2 years, followed by sustained increase in breast cancer incidence afterwards (Figure 1).27
In an editorial, Joshi and colleagues40 posed a biologically plausible explanation for these findings. As previously suggested,38 the initial post-intervention decrease was felt to be related to a reduction in estrogen exposure and resulting inhibition of growth of hormone receptor (HR)-positive preclinical breast cancers. However, the authors postulated that progestin, demonstrated to stimulate breast mammary stem cells in preclinical studies, then results in an excess of stem cells responsible for the long-term, sustained increase in breast cancer risk. As a result, the long-term risk of breast cancer for estrogen plus progesterone for about 5 years of use is substantially greater than previously thought. The findings regarding breast cancer for estrogen-alone use in women with prior hysterectomy were opposite those in the combined-hormone-therapy trial. Estrogen-alone use significantly decreased breast cancer incidence and significantly decreased deaths from breast cancer.33 When examined for year-to-year influence on breast cancer incidence during the intervention and post-intervention period, a lower breast cancer incidence was seen throughout the intervention, which persisted for about 4 years post intervention (Figure 2).27
The effect appeared to be more pronounced in black women, particularly those with more than the median percentage of African ancestry (>80%) in whom a 68% reduction in breast cancer incidence was seen.41 Thus, the favorable breast cancer effect of estrogen-alone use in black women has identified a potentialintervention strategy for addressing the disparity in breast cancer mortality risk seen in black compared with white women in the United States.
In the chemoprevention trials, estrogen reduction with AIs resulted in a reduction in breast cancer incidence.23,24 However, in the WHI trial, estrogen addition with conjugated equine estrogen also resulted in a reduction in breast cancer incidence.27,33 A likely explanation for this apparent paradox has been put forward by Jordan and colleagues.50 Estrogen typically simulates mammary epithelium and inhibits apoptosis to prepare the breast for milk production. However, after a period of estrogen deprivation, gene-expression profile change results in estrogen functioning as an apoptosis stimulant.42 These time-dependent, exposure level—dependent, complex interactions have practical clinical implications. Ellis and colleagues42 provided proof-of-principle in a study where postmenopausal women with HR-positive, advanced breast cancer refractory to AI use had some demonstrated activity to relatively low doses of estradiol (2 mg/ day). Taken together, these findings suggest that breast cancers that are HR-positive can survive only in an environment with a relatively narrow estrogen range.
Breast cancer prevention is commonly felt to be quite distinct from breast cancer therapy. However, a recent report by Santen and colleagues43 calls that concept into question. A tumor-growth kinetic model, based on preclinical and clinical findings of tumor-doubling time, size detection threshold, and tumor prevalence based on autopsy series was used to model the percentage of breast cancers in the WHI trial evaluating estrogen plus progestin, where age and eligibility criteria are similar to those in most breast cancer chemoprevention trials. Findings from the model analysis indicated that 94% of cancers detected were already established but preclinical, whereas only 6% were de novo tumors. Thus, it is likely that breast cancer chemoprevention in prevention trials can be, at least in part, therapy of preclinical breast cancer.The findings from the WHI hormone therapy randomized trials on breast cancer incidence and outcome differ in many ways from many observational studies of the same issue. For example, in the Million Women’s Study, although no increase in breast cancer is seen with estrogen-alone use in women starting use in their fifth decade, no decrease in incidence is seen.48 The discussion of bases of such differences is beyond the scope of the present report, but the issue has been addressed elsewhere.26,27,47Randomized clinical trial findings have identified a linear relationship, in that agents effective in the adjuvant setting that also reduce contralateral breast cancer risk have shown reduced breast cancer incidence in primary prevention studies. We have seen this model for tamoxifen and now for AIs.1,9,22,23 One could propose that an agent with a well-established toxicity profile that is effective as an adjuvant therapy and reduces incidence of contralateral breast cancer could be proposed as a breast cancer prevention agent.
Based on preclinical and emerging observational study evidence,44-46 metformin is currently undergoing trial as an addition to standard adjuvant breast cancer therapy in a randomized trial with over 3500 randomized patients. Because metformin has a well-established safety profile, if this adjuvant trial is positive with an impact on contralateral breast cancer, one could reasonably propose metformin for breast cancer prevention use.Although tamoxifen, raloxifene, and the AIs exemestane and anastrozole have all demonstrated an ability to reduce breast cancer incidence, emerging evidence suggests that raloxifene or AIs may be better choices in postmenopausal women. Women avoiding combined hormone therapy with estrogen plus progestin will have lower breast cancer risk. Some of these conclusions differ from the most recent ASCO breast cancer prevention guidelines, as the current report incorporates more recent emerging clinical trial information.17,19,23 New strategies to broaden the uptake of the available breast cancer risk reduction interventions need to be developed.
Affiliation: Rowan T Chlebowski, MD, PhD; Radhika Prabhaker, MD; and Tomoko Tagawa, MD; are from the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA.
Disclosure: Dr. Chlebowski is a consultant for AstraZeneca, Novartis, Amgen, Genentech Genomic Health, and Novo Nordisk, and serves on the speaker’s bureaus for Novartis and Genentech.
Funding/Support: The WHI program is reported by the National Heart, Lung and Blood Institute, National Institutes of Health, Department of Health and Human Services through contracts N01WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-32119, 32122, 42107-26, 42129-32, and 44221.
Author Contributions: Dr. Chlebowski wrote the initial draft of the report and takes full responsibility for the report. Drs. Prabhaker and Tagawa provided critical revision of the manuscript for important intellectual content.
Acknowledgment: We thank the Women’s Health Initiative investigators, staff, and the trial participants for their outstanding dedication and commitment.
Women’s Health Initiative Investigators
Program Office: (National Heart, Lung, and Blood Institute, Bethesda, MD); Jacques Roscoe, Shari Ludlum, Dale Burden, Joan McGowan, Leslie Ford, and Nancy Geller;
Clinical Coordinating Center: (Fred Hutchinson Cancer Research Center, Seattle, WA); Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg
Investigators and Academic Centers: (Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn E. Manson; (Med Star Health Research Institute/Howard University, Washington, DC) Barbara V. Howard; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A. Thompson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Iowa, Iowa City/Davenport, IA) Robert Wallace; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA) Rowan T. Chlebowski; (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker.
Women’s Health Initiative Memory Study: (Wake Forest University School of Medicine, Winston Salem, NC) Sally Shumaker.
Additional information: A full list of all the investigators who have contributed to Women’s Health Initiative science appears at https://www.whi.org/researchers/Documents%20%20Write% 20a%20Paper/WHI%20Investigator%20Long%20List.pdf.
The WHI program results in this review were funded by the National Heart, Lung, and Blood Institute at the National Institutes of Health, US Department of Health and Human Services.