The Promise of Molecular Breast Tumor Subtyping
Will molecular-genomic panels like MammaPrint® overcome the limitations of histo- and clinicopathological breast cancer subtyping, to better personalize patients' treatment strategies and prognoses?
Newly published data are promising, and prospective clinical trials are under way. However, critics caution that the field needs more standardization.
Breast cancer is a complexly heterogeneous malignancy encompassing at least 18 distinct tumor types.1 Correctly predicting tumor responses to different treatment regimens is dependent upon accurate tumor subtyping.
Breast tumors are widely subtyped using immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) for prognosis and treatment planning based on tumor histological type, grade, stage, and biomarker status (eg, estrogen receptor [ER], progesterone receptor [PR], and HER2-neu). These tests provide valuable information for clinical decision making.1
Despite the unquestioned utility of histological typing, however, IHC tests only capture a fraction of breast tumor subtype diversity, as evidenced by the variability of outcomes among patients whose tumors fall within particular subtypes under the existing classification system.1
Some patients with HER2-positive tumors “show either de novo resistance or will develop recurrent disease,” note the authors of a 2012 review in the journal Diagnostic Histopathology.1 “Additionally, approximately 15% of patients with breast carcinomas categorized as low-grade with a low risk of aggressive behavior will develop recurrent disease.”
“There is a need to provide better and more detailed prognostic information in the clinical setting” and to incorporate “conventional histopathological parameters with molecular data derived from the patient's own tumor,” the coauthors argued.1
Furthermore, the interpretation of histochemical lab-test results heavily depends primarily on the training, experience, and skill of the lab personnel performing them.2 Errors in the performance and interpretation of ER/PR IHC testing can—and do—occur in the lab, with potentially devastating consequences for patients whose tumor subtypes are misidentified and who may consequently receive inappropriate treatment.2
Lab errors in the late 1990s and early 2000s in Newfoundland and Labrador, Canada, appear to have contributed to the misdiagnosis and incorrect treatment of 383 women with breast cancer; in turn, this contributed to more than 100 patient deaths, according to a government inquiry.2
The clinical implications of IHC testing errors are “huge,” said Allen Gown, MD, of PhenoPath Laboratories in Seattle, WA, who testified before the Canadian commission in charge of the lab errors inquiry. False-negative ER test results such as those that occurred in the Canadian labs mean that patients “might be denied drugs such as tamoxifen, which could extend survival time and potentially cure their disease.”2
Even when IHC testing is well performed and properly interpreted, it is noteworthy that tumors with similar histologies can exhibit different clinical responses.
Molecular Tumor Subtyping Moving Forward
Genetic and genome-wide molecular classifications of breast carcinomas using high-resolution, microarray-based gene expression profiling are quickly maturing and might prove to be a viable clinical complement or alternative to IHC subtyping in the future (See Box to right).1
MammaPrint: 70-gene prognostic signature
Oncotype DX: 21-gene prognostic signature for ER-positive patients treated with tamoxifen5
MapQuant Dx: 97-gene prognostic signature for classifying grade 2 tumors into low- or high-risk for relapse-free survival5
Breast Cancer Index (BCI): Incorporates MapQuant Dx test with a 2-gene test to classify ER-positive, node-negative patients with breast cancer into high vs low-risk groups for 10-year post-tamoxifen recurrence rates5
EndoPredict (EP): 11-gene expression array identifies patients with ER-positive, HER2-negative early-stage breast cancer with good prognosis following adjuvant endocrine therapy alone.5
A retrospective study published in July 2013 in the journal Breast Cancer Research and Treatment found that molecular subtyping of biopsied early-stage breast tumor tissue with the customized BluePrint™ and MammaPrint® assays was able to identify a proportion of patients (21%) with a Luminal A molecular subtype.
This group of patients did not experience improved pathologic complete response (pCR) after preoperative chemotherapy but nevertheless exhibited excellent 5-year distant metastases–free survival (DMFS) rates.3
MammaPrint and BluePrint are components of Agendia®'s Symphony™ panel of molecular tests for subtyping tumor tissue; Agendia supported the study. MammaPrint was approved by the U.S. Food and Drug Administration (FDA) in 2007 as a test of metastasis risk among breast cancer patients with lymph node–negative tumors less than 5 cm in diameter.3
Compared with IHC and FISH, molecular subtyping with BluePrint and MammaPrint offered “a more precise classification of patients with early-stage breast cancer and a better correlation with long-term clinical treatment,” reported lead author Stefan Glück, MD, PhD, of the University of Miami's Sylvester Comprehensive Cancer Center, and colleagues.3