This article is more than 5 years old. Information may no longer be current.
Genomic test may predict recurrence risk in early-stage breast cancer
Use of the Genomic Grade assay, as well as central review of Ki67 data, both significantly improved clinicopathological models for determining risk for distant recurrence in patients with breast cancer, according to a substudy of phase 3 trial results.
However, compared with histological grade, the Genomic Grade assay (Mapquant Dx, Ipsogen SA) identified a higher proportion of women treated with endocrine therapy who had a very low risk for 10-year distant recurrence.
The Genomic Grade assay measures the expression of 97 genes that characterize high-grade vs. low-grade tumors.
Christos Sotiriou, MD, PhD, head of Breast Cancer Research Foundation’s Translational Research Laboratory in Belgium, and colleagues sought to demonstrate the clinical validity of the assay by comparing it with immunohistochemical staining of Ki67 antigen proliferation for detection of breast cancer recurrence risk.
Sotiriou and colleagues conducted a substudy of a large phase 3 adjuvant trial of women with ER-positive and node-positive or node-negative nonmetastatic breast cancer. Patients in this study — who had available formalin-fixed, paraffin-embedded samples of their primary tumors — had been randomly assigned to 5-year monotherapy with tamoxifen or letrozole (Femara, Novartis).
Distant recurrence-free interval served as the primary endpoint.
The mean follow-up was 8.1 years
The researchers obtained Genomic Grade assay data in 61% of evaluable patients (n = 883 of 1,440; 49% treated with tamoxifen). Ten percent of women (n = 84) had distant recurrences at follow-up.
Sotiriou and colleagues observed that increasing assay-measured grade or Ki67 expression both appeared significantly associated with shorter distant recurrence-free intervals, while also adding independent prognostic information to the clinicopathological prognostic factors.
After excluding patients with missing information, the researchers observed more patients who were histological grade 3 (27.3% vs. 16.4%; P < .001), high Ki67 (47.9% vs. 28.4%; P < .001), PR-positive tumors (90.7% vs. 59.5%; P < .001) and node-negative tumors (59.1% vs. 56%) in the GG assay.
The 8-year distant recurrence-free interval was 90% (95% CI, 88-92) for the substudy and 86% (95% CI, 85-88) for the remaining patient population.
Among women treated with endocrine therapy alone (n = 467), 38% were Genomic Grade 1, with a 10-year distant recurrence-free interval of 99% (95% CI, 97-100). In comparison, 18% were histological grade 1, with a 10-year distant recurrence-free interval of 100% (95% CI, 100-100).
Genomic Grade equivocal patients had a 10-year distant recurrence-free interval of 94% (95% CI, 90-98). Among patients who were Genomic Grade 3, the 10-year distant recurrence-free interval was 87% (95% CI, 80-94).
The researchers acknowledged a high failure rate as a study limitation. They attributed it to poor RNA quantity or quality of the collected samples.
“The test showed that it can refine the prediction of risk for distant recurrence and is able to identify a larger proportion of patients who have excellent distant recurrence-free interval at 10 years with just 5 years of endocrine therapy compared with centrally reviewed immunohistochemical testing of Ki67 and histological grade,” Sotiriou and colleagues wrote. “The feasibility of performing the assay on formalin-fixed, paraffin-embedded samples, and the ongoing testing in a randomized phase 3 trial, would make the case for the Genomic Grade assay emerging as a useful genomic tool for clinical use.”
Further investigation of gene expression tests are needed before they become part of standard risk modeling, Jee Ye Kim, MD, Seung-il Park, MD, PhD, and Soonmyung Paik, MD, all of Yonsei University College of Medicine in Seoul, South Korea, wrote in an accompanying editorial.
“While all current gene expression-based prognostic tests use within-sample reference gene normalization to be able to compare between cases, no reference gene is perfect, and this could be a source of the lack of agreement among the tests,” they wrote. “In that regard, the current generation of gene expression-based prognostic tests … can be regarded as interim with Ki67 as a low-cost alternative until we are ready for a universal low-cost [RNA sequencing]-based test.” – by Cameron Kelsall
Disclosure: Sotiriou and one other study researcher report inventing the Genomic Grade Index. Other researchers report consultant fees from Deerfield Advancing Healthcare, honoraria from Nanostring and advisory board positions with PAMSO ROR (Prosigna). Paik reports co-invention of and the issuance of patents for a 21-gene recurrence score test (OncotypeDx, Genomic Health); however, he reports that he holds no financial interest in his invention. Kim and Park report no relevant financial disclosures.