Gene-expression Prognostic Signatures Guide Adjuvant Chemotherapy of Early-stage Breast Cancer
Gene expression prognostic signatures continue to evolve to provide greater accuracy for both short- and long-term risk of recurrence.
Patients with early-stage breast cancer (stage I, IIA, IIB, or IIIA) have varying residual risk of recurrence, despite initial treatment. Adjuvant chemotherapy is commonly used to reduce this risk, though “nearly 60% of all patients with early-stage breast cancer receive adjuvant chemotherapy, but only 2% to 15% of these patients derive an important benefit and others only experience its toxic effects,” wrote Domen Ribnikar, MD, of the Institute of Oncology in Slovenia, and Fatima Cardoso, MD, of the Champalimaud Cancer Center in Portugal, in an article published in the 2016 American Society of Clinical Oncology (ASCO) Educational Book based on their presentation at the ASCO Annual Meeting.1
Historically, the decision to use adjuvant chemotherapy in early-stage breast cancer was based on clinicopathologic factors or algorithms derived from these factors, such as Adjuvant! Online.2-4 These methods are limited, however, owing to the heterogeneousness of breast cancer.
Molecular profiling led to intrinsic molecular subtyping that identified 4 molecular subtypes associated with different levels of risk and responsiveness to chemotherapy: luminal, HER2-enriched, basal-like, and normal breast–like.5,6 About 30% of breast cancers do not fit within 1 of the 4 subtypes, which together constituted the first-generation of gene-expression prognostic signature.7
First-generation Prognostic Signatures
These signatures were developed from microarray or quantitative reverse transcription polymerase chain reaction (qRT-PCR) data that generally identify tumors with a high expression of proliferation-related genes that confer poor prognosis, and therefore indicate that adjuvant chemotherapy may be of benefit. The type of chemotherapy regimen of most benefit is not, however, predicted by these scores. Importantly, Drs Ribnikar and Cardoso pointed out that, “the accuracy of the outcome prediction of all first-generation prognostic signatures is time-dependent, with more accurate prediction during the first 5 years after diagnosis and less accurate predictions for late relapses.”
The first-generation prognostic signatures include the 70-gene profile (MammaPrint), Oncotype DX (recurrence score), the 76-gene signature (Veridex), and the Genomic Grade Index (GGI). Of these, MammaPrint is approved by the U.S. Food and Drug Administration (FDA) for determining the prognosis of patients with stage I or II, node-negative or node 1- to 3-positive, invasive breast cancer.8,9
In early 2016, data from the randomized, phase 3 MINDACT trial demonstrated that patients with a high risk of recurrence benefited from chemotherapy, whereas patients at intermediate risk experienced a similar distant metastasis–free survival with or without chemotherapy.10
Oncotype DX assigns patients to 1 of 3 risk groups; patients at low risk of recurrence did well with endocrine therapy alone,11 and patients at high risk of recurrence are recommended to receive chemotherapy. The best course of treatment for patients with intermediate risk is under investigation in the randomized Tailor X trial.12
RELATED: ABP 980 Biosimilar Not Inferior to Trastuzumab for Treatment of Breast Cancer
The Veridex signature is able to accurately determine risk of recurrence within 5 years for patients with either estrogen receptor positive or negative disease. But the assay requires the use of fresh or frozen samples, and its ability to predict prognosis of triple-negative breast cancer is unknown.
The CGI assigns patients with estrogen receptor–positive grade 2 tumors into grade 1–like or grade 3–like, with grade 3–like tumors most likely to benefit from chemotherapy.