Multigene assays have become central to managing HR+/HER2- early-stage breast cancer by refining genomic risk and supporting treatment decisions. Different assays answer different clinical questions, from predicting chemotherapy benefit to estimating late recurrence risk, and results must be interpreted alongside clinical factors such as nodal status, tumor size, age, and menopausal status.

It has been more than 25 years since we saw the first applications of gene expression profiling in breast tumors. These tests extract messenger RNA from cancer and look at which genes in that tumor are expressed and at what levels. This information is used to help come up with an overall assessment of a tumor’s genomic risk.

Fairly early in the development of these assays, we started to see their commercialization. We currently have several gene expression assays available for use in the clinical setting. If we look at all the different assays that are available, we see a fairly consistent and highly reproducible assessment and identification of patients who are at high, low, or intermediate risk. All the assays that are clinically available are quite good at prognosis assessment, but they are used in various ways to answer different clinical questions.

For example, some tests have really robust evidence to help guide chemotherapy decision making, such as the 21-gene assay (ie, the Oncotype DX Breast Recurrence Score [Exact Sciences Corporation]). This assay was prospectively evaluated in both lymph node–negative and –positive early-stage breast cancer in the TAILORx and RxPONDER trials, respectively. These trials showed that the overall assessment of genomic risk can help identify which patients are likely or unlikely to derive benefit from adjuvant chemotherapy. Oncotype DX has really been the assay that is best known for predicting chemotherapy benefit, not only because of its prognostic value but also because of its predictive value in terms of the estimation of benefit of adjuvant chemotherapy in HR+/HER2- early-stage disease.

There are other assays, such as the MammaPrint (Agendia) 70-gene signature. The level of evidence from the MINDACT trial for MammaPrint was not as high as that from the TAILORx and RxPONDER trials for Oncotype DX, but, certainly, there are data showing that patients with low MammaPrint scores who do not receive chemotherapy can have excellent long-term outcomes.

There has also been a lot of interest in looking at late recurrence risk, which involves looking further along in the treatment trajectory to determine whether we should extend a 5-year course of adjuvant endocrine therapy. Certain tests seem to perform better in the recurrence risk setting. For example, the Prosigna Breast Cancer Prognostic Gene Signature (Veracyte, Inc) and the Breast Cancer Index (BCI; Biotheranostics, Inc) assays are used in this setting.

Overall, we have learned that we cannot just look at the results of a multigene assay in isolation and make all decisions based on one test. It is one piece of information, and we need to also look at the age of the patient, their menopausal status, their overall health status, the size of the tumor, and the extent of the nodal involvement—integrating all these pieces to make the best decision for a patient. Over the years, a number of investigators and companies have developed tools that use molecular information from multigene assays and integrate the patient’s clinical information to develop a more accurate risk prediction.