Avoiding overtreatment in breast cancer: Lessons from three trials

Alan P. Lyss, MD

Three recent studies suggest certain patients with early breast cancer can achieve good outcomes with less intense therapy.

One study, PRIME-2, suggested that omitting whole-breast adjuvant radiotherapy does not affect long-term survival in women ages 65 and older who have had surgery for early-stage, hormone receptor–positive (HR+) breast cancer.

Another study, RxPONDER, showed that adding chemotherapy to endocrine therapy did not improve outcomes for postmenopausal women with low-risk, node-positive, HR+, HER2-negative (HER2-) breast cancer.

A third study, ADAPT HR+/HER-, indicated that patients with high-risk, HR+/HER2- breast cancer may be able to forgo neoadjuvant chemotherapy, depending on their 21-gene recurrence score.

Results from all three studies were presented at the San Antonio Breast Cancer Symposium 2020.

PRIME-2 results

Ian Kunkler, MRCPUK, FRCR, of Western General Hospital, University of Edinburgh in Scotland, presented PRIME-2 results as abstract GS2-03.

PRIME-2 is a phase 3 study that enrolled 1,326 women with invasive breast cancer. Eligible patients had to be 65 years or older and have planned breast-conserving surgery and postoperative endocrine therapy. The patients’ tumors were required to be HR+, 3 cm or less, and node-negative.

Participants were randomized to receive no radiotherapy or adjuvant whole-breast irradiation at a dosing schedule of 40-50 Gy in 15-25 fractions. Adjuvant endocrine therapy was required.

Dr. Ian Kunkler
The primary endpoint was ipsilateral breast cancer recurrence at 10 years. There was a significantly lower rate of ipsilateral recurrence with radiotherapy than without it (0.9% vs. 9.8%; P = .00008). Similarly, the 10-year rate of regional recurrence was significantly lower with radiotherapy than without it (0.5% vs. 2.3%, P = .014).

However, there was no significant difference in the radiotherapy and no-radiotherapy arms with respect to the rates of distant metastases (3.6% vs. 1.9%, P = .07), contralateral breast tumors (2.2% vs. 1.2%, P = .20), or subsequent non–breast cancers (8.7% vs. 10.2%, P = .41).

Most importantly, the overall survival estimate at 10 years was 81% with radiotherapy and 80.4% without it (P = .68). Likewise, the metastasis-free survival rate was not significantly different between the arms (96.4% and 98.1%, respectively; P = .28).

Implications for practice

Based on results from the CALGB 9343 study, National Comprehensive Cancer Network guidelines permit the omission of radiotherapy in women who are at least 70 years of age and who are receiving endocrine therapy, but only for women with T1, clinical N0 tumors.

The CALGB 9343 study has not had a major impact on the use of radiotherapy, despite long-term follow-up data showing no survival benefit.

Although the median follow-up of PRIME-2 was only 7 years, and there were few high-grade tumors, the results corroborate the CALGB 9343 results.

PRIME-2 suggests the risks of omitting radiotherapy in women who are at least 65 years old and have node-negative, endocrine-responsive tumors up to 3 cm are modest and not life-threatening. Deaths of the participants were due to competitive causes of mortality in an older population.

Among women who wish to lower their risk of an in-breast recurrence and elect to receive radiotherapy, more abbreviated treatment courses may be beneficial. The FAST-Forward trial showed that a regimen of hypofractionated radiotherapy at 26 Gy in 5 fractions over 1 week was safe and noninferior to a regimen of 40 Gy in 15 fractions over 3 weeks for local tumor control. Partial breast irradiation is also an option for selected patients.

Top image credit: BrianAJackson/Getty Images
Image credit: FatCamera/Getty Images

RxPONDER results

Kevin Kalinsky, MD, of Winship Cancer Institute of Emory University in Atlanta, presented results from RxPONDER as abstract GS3-00.
He presented data for 5,015 HR+, HER2- breast cancer patients with 1-3 involved nodes and a 21-gene recurrence score of 25 or lower. The patients were randomized to endocrine therapy alone or adjuvant chemotherapy followed by endocrine therapy. Randomization was stratified by recurrence score (0-13 vs. 14-25), menopausal status, and type of axillary surgery.

In both arms, 67% of patients were postmenopausal, 57% had a recurrence score of 14-25, and 63% had full axillary lymph node dissection.

The median follow-up was 5.1 years. In the overall population, the 5-year invasive disease-free survival (IDFS) rate was 92.4% with chemotherapy and 91.0% with endocrine therapy alone (adjusted hazard ratio [aHR], 0.81; P = .026).

Dr. Kevin Kalinsky

The investigators found that chemotherapy use and recurrence score were independently prognostic for IDFS, but recurrence score did not predict the relative benefit of chemotherapy for IDFS. On the other hand, the chemotherapy benefit was different depending on menopausal status.

For postmenopausal patients, the 5-year IDFS rate was not significantly different with endocrine therapy alone and with chemotherapy — 91.9% and 91.6%, respectively (aHR, 0.97, P = .82). For premenopausal patients, the difference was significant — 94.2% and 89.0%, respectively (aHR, 0.54, P =. 0004).

When postmenopausal patients were divided by recurrence scores (0-13 vs. 14-25), there was no significant difference in IDFS by treatment arm.

In contrast, there was a significant difference for premenopausal patients. In premenopausal patients with a recurrence score of 0-13, the 5-year IDFS rate was 96.5% with chemotherapy and 92.6% with endocrine therapy alone (aHR, 0.46, P = .04). In premenopausal patients with a recurrence score of 14-25, the 5-year IDFS rate was 92.8% and 86.6%, respectively (aHR, 0.57, P = .005).

Implications for practice

The RxPONDER data establish that postmenopausal women with 1-3 involved nodes and a 21-gene recurrence score of 25 or less do not benefit from postoperative adjuvant chemotherapy and can be treated with endocrine therapy alone.

For physicians who have been giving those patients chemotherapy, these data are immediately practice-changing. Since most postmenopausal women in this trial had 1 or 2 involved nodes, one could legitimately question whether the initial results will apply to women with 3 involved nodes. Those data will be forthcoming in later analyses.

Based on the results of RxPONDER, however, for premenopausal women with 1-3 involved nodes and a 21-gene recurrence score of 25 or lower, adjuvant chemotherapy should still be recommended.

Additional analyses may determine whether the benefit of chemotherapy in premenopausal patients is related to chemotherapy-induced menopause. If so, ovarian function suppression may produce benefits of similar magnitude, with less potential toxicity, inconvenience, and expense.

Since the overall survival information is immature and more than half of recurrences occur in this type of breast cancer beyond the 5-year mark, follow-up in this trial will extend for 15 years.

ADAPT HR+/HER2- results

In HR+/HER2- early breast cancer, node-negative patients with a low 21-gene recurrence score do not benefit from adjuvant chemotherapy.

Among women with a small number of positive nodes, the RxPONDER trial indicated that postmenopausal women with a low-to-intermediate recurrence score may be treated with postoperative endocrine therapy alone and avoid adjuvant chemotherapy, but chemotherapy should still be offered to premenopausal patients.

The question researchers sought to answer with the ADAPT HR+/HER2- trial is whether endocrine therapy response-guided criteria could also play a role in guiding subsequent, potentially curative, therapy.

Nadia Harbeck, MD, PhD, of the University of Munich, presented results from the trial as abstract GS4-04.

ADAPT HR+/HER2- enrolled 5,625 patients with HR+, HER2- early breast cancer who were candidates for adjuvant chemotherapy based on conventional criteria.

Investigators attempted to use a patient’s 21-gene recurrence score at baseline and Ki-67 response to a 3-week course of preoperative endocrine therapy to personalize adjuvant therapy.

Those patients with a pretreatment recurrence score of 12-25 had a needle biopsy of the residual tumor. If the post-endocrine therapy core biopsy showed that the Ki-67 on the tumor cells had dropped to less than 10%, the patient continued endocrine therapy alone.

If the Ki-67 remained 10% or higher or if there was a large bulk of remaining clinically involved lymph nodes, the patient was assigned to neoadjuvant chemotherapy.

Dr. Nadia Harbeck

The primary objective was to determine if the 5-year IDFS rate for patients with an intermediate recurrence score and a response to induction endocrine therapy was the same as it was for patients who presented with a low recurrence score of 0-11.

Dr. Harbeck reported results in 2,290 patients, including 868 patients with a recurrence score of 0-11 and 1,422 patients with recurrence score of 12-25 who had a response to preoperative endocrine therapy.

The 5-year IDFS rate was 93.9% in patients with recurrence scores of 0-11 and 92.6% in patients with recurrence scores of 12-25 and a Ki-67 response (P = .05). Therefore, the prespecified criteria to accept non-inferiority were met.

The 5-year distant disease-free survival was 96.3% in patients with recurrence scores of 0-11 and 95.6% in patients with recurrence scores of 12-25 and a Ki-67 response (P = .247). The 5-year overall survival rates were 98.0% and 97.3%, respectively (P = .160).

Patients with recurrence scores of 0-11 had excellent IDFS, distant disease-free survival, and overall survival, regardless of nodal involvement.

However, in patients with recurrence scores of 12-25 and Ki-67 response, those with 3 involved nodes did not do well with endocrine therapy alone. In that small group (7% of study participants), the 5-year distant disease-free survival was just 75.9%, much lower than for patients with 0-2 involved nodes (92.4%-96.6%).

Image credit: FatCamera/Getty Images

Implications for practice

This trial gives us confidence that subsequent chemotherapy can be safely avoided in patients with HR+/HER2- breast tumors with recurrence scores of 12-25 who are willing to have a few weeks of preoperative endocrine therapy and undergo a repeat biopsy, have a low number of clinically involved nodes, and have a tumor-associated Ki-67 below 10%. The 5-year IDFS rate in this study was great with adjuvant endocrine therapy alone.

The outcome for patients with endocrine-responsive tumors was comparable to the outcome for tumors with a better genomic profile, regardless of age, if the patient did not have 3 involved axillary nodes. The implication is that genomic profiling could spare the majority of HR+/HER2- patients with 0-3 involved nodes from receiving chemotherapy.

Study discussant Lajos Pusztai, MD, DPhil, of Yale Cancer Center in New Haven, Conn., noted that it remains unclear just how much predictive value early response assessment could add to gene expression profiling alone in clinical practice, since:

• About 75% of patients with a recurrence score of 12-25 had a Ki-67 response by week 3-4 with endocrine therapy.

• The trial did not show how well the group without a Ki-67 response did in comparison with those who responded.

• Ki-67 is not a strong predictor of pathological complete response rates.

• Measurement of Ki-67 in community practice is unreliable, with a high degree of inter-observer variation.

Dr. Lajos Pusztai
Despite these concerns about the metric chosen in this trial, the issue of response-guided criteria for assessing subsets of patients who are likely to do well with lower intensity therapy and for guiding selection of subsequent treatment is one of high scientific interest. Response-guided criteria for tailoring the intensity of adjuvant treatment to patients will be a topic of ongoing study.

ADAPT HR+/HER2- was sponsored by the West German Study Group and Roche Pharma AG. Dr. Harbeck disclosed relationships with Agendia, Amgen, AstraZeneca, and several other companies. Dr. Pusztai disclosed relationships with AstraZeneca, Merck, Novartis, and several other companies.

RxPONDER was sponsored by the National Cancer Institute, the Susan G. Komen for the Cure Research Program, the Hope Foundation for Cancer Research, the Breast Cancer Research Foundation, and Exact Sciences. Dr. Kalinsky disclosed relationships with Grail, Pfizer, Lilly, and several other companies.

PRIME-2 was funded by the Chief Scientist Office (Scottish Government) and the Breast Cancer Institute at the Western General Hospital in Edinburgh, Scotland. Dr. Kunkler had no relevant disclosures.

Dr. Lyss was a community-based medical oncologist and clinical researcher for more than 35 years before his recent retirement. His clinical and research interests were focused on breast and lung cancers, as well as expanding clinical trial access to medically underserved populations. He is based in St. Louis. He has no conflicts of interest.