Predictive Assays Help Personalize Care in Early-Stage HR+ Breast Cancer

Stephanie L. Graff, MD, discusses the importance of using genomic risk to tailor treatment to patients with early-stage hormone receptor-positive, HER2-negative disease, the benefit of extended endocrine therapy, and ongoing research with CDK4/6 inhibitors and immunotherapy.

Stephanie L. Graff, MD

The 21-gene recurrence score assay demonstrated predictive value in identifying whether patients with early-stage hormone receptor (HR)—positive, HER2-negative breast cancer will require chemotherapy in addition to endocrine therapy, according to the results of the phase III TAILORx trial, said Stephanie L. Graff, MD.

“[Results of] the TAILORx study showed us that genomic risk prediction is really a much stronger tool to utilize than clinical risk,” said Graff, who is the director of the Breast Program at Sarah Cannon Cancer Institute, HCA Midwest Health. “We know that the genomic score outperforms tumor size, grade, and many of the features we have classically looked at—especially in postmenopausal women.”

In the trial, women were assigned to 1 of 4 treatment arms: Those with a recurrence score of ≥10 were assigned to receive endocrine therapy alone, women with a score of ≥26 were assigned to receive chemotherapy plus endocrine therapy, and women with a midrange score of 11 to 25 were randomized to receive either endocrine therapy alone or chemotherapy plus endocrine therapy. The trial's objective was to determine whether chemotherapy provides benefit to women with a midrange recurrence score of 11 to 25.

Results of TAILORx (NCT00310180) showed that adjuvant endocrine therapy alone is noninferior to adjuvant chemotherapy plus endocrine therapy in patients with HR-positive, HER2-negative, node-negative, early-stage breast cancer who have an intermediate risk of distant recurrence based on the assay (HR, 1.08; 95% CI, 0.94-1.24; P = .26).1

A prespecified secondary analysis of the trial looked at whether clinical risk provides additional prognostic or predictive information to the recurrence score results. Data presented at the 2019 ASCO Annual Meeting showed that clinical risk alone was not predictive of chemotherapy benefit in the overall population; this was true for most women over age 50.2

“In 2020, ovarian suppression is a very reasonable strategy for women under the age of 50 with those intermediate Oncotype recurrence scores, said Graff. “It gives us an opportunity to personalize care based on patient preference and overall clinical risk in addition to their genomic assay.”

Once patients complete 5 years of endocrine therapy, providers must work with them to determine whether extending therapy to 10 years is the appropriate next step in the treatment journey.

In an interview during the 2020 OncLive State of the Science Summit on Breast Cancer, Graff, who is also associate director of the Breast Cancer Research Program at Sarah Cannon Research Institute, discussed the importance of using genomic risk to tailor treatment to patients with early-stage HR-positive, HER2-negative disease; the benefit of extended endocrine therapy; and ongoing research with CDK4/6 inhibitors and immunotherapy.

OncLive: Could you provide an overview of the TAILORx trial in early-stage HR-positive, HER2-negative breast cancer and the data we have seen thus far?

Graff: Since the 2018 ASCO Annual Meeting readout on the TAILORx study, which looked at the 21-gene recurrence score in women with early-stage estrogen receptor (ER)—positive breast cancer, genomic profiling has become the clear standard of care to determine who should receive chemotherapy and in whom we can safely omit chemotherapy. [Results from] TAILORx showed us that women with recurrence scores of <10 have an exceptional outcome, with <5% risk of distant recurrence at 8 years. This became [a new addition] in our staging criteria, which was exciting.

In 2018, we got the update for the midrange [recurrence score] group, which was really the intention-to-treat population in the study; these were patients who had recurrence scores of 11 to 25. In these women, we saw no difference between chemotherapy plus endocrine therapy and endocrine therapy alone. Because TAILORx was a trial that included over 10,000 women, we did a subset analysis, in which we saw that patients under the age of 50 still appeared to get some benefit [from chemotherapy] if they had an intermediate recurrence score of 16 to 25.

Joseph A. Sparano, MD, of Montefiore Medical Center, gave an update [at the 2019 ASCO Annual Meeting] that incorporated clinical factors. A patient’s clinical risk profile is really simple. If their tumor is <1 cm and grade 3, if it's <2 cm and grade 2, or if it's <3 cm and grade 1, they are low risk. If you take women with recurrence scores of 16 to 20 and superimpose their clinical risk, you see populations that get a smaller and smaller benefit with chemotherapy. The benefit of adding chemotherapy in many of those populations is about 6%, which closely parallels what we saw in the SOFT and the TEXT trials with the benefit of ovarian suppression.

How are you balancing clinical and genomic risk in practice? Should the assay results be interpreted definitively?

The role of the clinical predictors is important in personalizing care. In women with intermediate Oncotype recurrence scores between 14 and 20, maybe 16 and 20 in particular, the benefit of chemotherapy starts to become incrementally larger based on their overall clinical risk profile.

What do you recommend in terms of extended endocrine therapy?

Extended adjuvant therapy is an evolving question. At the end of the day, all patients benefit [from extended therapy]. When I'm starting a patient on endocrine therapy, I always tell them that the clear goal is to get to 5 years and then reassess how they're doing and decide whether

to push to 10 years. The benefit of 10 years of tamoxifen is clear, but more women are converting at some point from tamoxifen to an aromatase inhibitor (AI), or are just starting on an AI. For patients who are on AIs, the [rationale] to extend therapy to 10 years isn't quite as strong yet. That’s probably because of where we are in the data reporting.

As we extend an AI, we don't see the sharp incline in benefit that we do when we go from tamoxifen to an AI. When we have 15- or 20-year follow-up on 10 years of an AI, that benefit will continue to exist. For patients who are tolerating the regimen well, particularly patients with significant risk factors, I believe that extended therapy is the way to go.

The Breast Cancer Index (BCI) showed predictive value for extended adjuvant therapy. What would be its utility be if it becomes commercially available?

We have seen some really nice reviews comparing all the various genomic assays that have been performed, including the PAM50 (Prosigna) score, the 21-gene recurrence score, BCI, and EndoPredict. All the assays have been compared head-to-head, and they all have clear prognostic implications. In 2020, the standard is to do the profile and use it to select patients who potentially won't benefit from chemotherapy. Now, the 21-gene recurrence score is the only one that can really show that [patients] are predictive for response to chemotherapy. All of the other assays commercially available are still prognostic. Of course, MammaPrint is closer [to becoming predictive]. We're seeing emerging evidence that many of these assays have similar value to our patients in determining what's in their best interest.

What are your thoughts on intermittent dosing for patients who have trouble with extended therapy?

We still don’t have really clear evidence about the value of intermittent dosing. We know that patients who are nonadherent to therapy have a worse distant recurrence-free survival, with a reduction of about 5%. For patients who discontinue [treatment] early, we also see a reduction in their survival indices. Intermittent dosing is not the standard of care. I try to work with a patient to develop a strategy to make the medication and the adverse events associated with it tolerable. I do de-escalate therapy to tamoxifen if a patient is intolerant of an AI. The standard is still to take treatment continuously and to complete at least 5 years of therapy.

How are bone-preserving agents being used in practice?

Bone-preserving agents are really important. Abundant evidence shows that zoledronic acid improves breast cancer—specific survival. Because the AIs are also potentially harmful to bone health, a medicine that can strengthen bones and improve breast cancer outcomes is a really valuable tool. It's important that we as oncologists are champions for our patients’ bone health as well.

What work is being done to bring CDK4/6 inhibitors into the early-stage setting?

CDK4/6 inhibitors are definitely on the horizon in the early-stage setting. We have already seen the monarchE study looking at abemaciclib (Verzenio). The PALLAS study looking at palbociclib (Ibrance) finished accrual. The NATALEE trial with ribociclib (Kisqali) is still accruing patients, so it’s a great option for patients who are interested [in enrolling]. Given the remarkable benefits we've seen with CDK4/6 inhibitors in the metastatic setting, we're expecting [to see benefit] in the high-risk, ER-positive, early-stage breast cancer space as well. We'll just have to wait on those results. I’d encourage patients and physicians to look for those trials for appropriate patients.

What are your expectations for the future treatment of these patients?

ER-positive breast cancer has not typically been a forum where we think a lot about immunotherapy or being immunologically active. We're seeing more and more of a signal that the earlier we treat patients who have other subtypes, such as triple-negative or HER2-positive disease, the more robust the response will be. [I’m hoping to see] trials looking at immunotherapy in carefully selected patients with HR-positive disease. [Likely, we will also] start to consider how other genetic and genomic risk profiles impact our treatment decisions, and [we will] move other lines of therapy up into the earlier-phase setting.


  1. Sparano JA, Gray RJ, Makower DF, et al. Adjuvant chemotherapy guided by a 21-gene expression assay in breast cancer. N Eng J Med. 2018;379(2):111-121. doi: 10.1056/NEJMoa1804710.
  2. Sparano JA, Gray RJ, Makower DF, et al. Impact of clinical risk category on prognosis and prediction of chemotherapy benefit in early breast cancer (EBC) by age and the 21-gene recurrence score (RS) in TAILORx. J Clin Oncol. 2019;37(suppl 15; abstr 503). doi: 10.1200/JCO.2019.37.15_suppl.503.