Lynn G. Dressler

CONTEXT: Gene expression analysis has identified several breast cancer subtypes, including basal-like, human epidermal growth factor receptor-2 positive/estrogen receptor negative (HER2+/ER-), luminal A, and luminal B. OBJECTIVES: To determine population-based distributions and clinical associations for breast cancer subtypes. DESIGN, SETTING, AND PARTICIPANTS: Immunohistochemical surrogates for each subtype were applied to 496 incident cases of invasive breast cancer from the Carolina Breast Cancer Study (ascertained between May 1993 and December 1996), a population-based, case-control study that oversampled premenopausal and African American women. Subtype definitions were as follows: luminal A (ER+ and/or progesterone receptor positive [PR+], HER2-), luminal B (ER+ and/or PR+, HER2+), basal-like (ER-, PR-, HER2-, cytokeratin 5/6 positive, and/or HER1+), HER2+/ER- (ER-, PR-, and HER2+), and unclassified (negative for all 5 markers). MAIN OUTCOME MEASURES: We examined the prevalence of breast cancer subtypes within racial and menopausal subsets and determined their associations with tumor size, axillary nodal status, mitotic index, nuclear pleomorphism, combined grade, p53 mutation status, and breast cancer-specific survival. RESULTS: The basal-like breast cancer subtype was more prevalent among premenopausal African American women (39%) compared with postmenopausal African American women (14%) and non-African American women (16%) of any age (P<.001), whereas the luminal A subtype was less prevalent (36% vs 59% and 54%, respectively). The HER2+/ER- subtype did not vary with race or menopausal status (6%-9%). Compared with luminal A, basal-like tumors had more TP53 mutations (44% vs 15%, P<.001), higher mitotic index (odds ratio [OR], 11.0; 95% confidence interval [CI], 5.6-21.7), more marked nuclear pleomorphism (OR, 9.7; 95% CI, 5.3-18.0), and higher combined grade (OR, 8.3; 95% CI, 4.4-15.6). Breast cancer-specific survival differed by subtype (P<.001), with shortest survival among HER2+/ER- and basal-like subtypes. CONCLUSIONS: Basal-like breast tumors occurred at a higher prevalence among premenopausal African American patients compared with postmenopausal African American and non-African American patients in this population-based study. A higher prevalence of basal-like breast tumors and a lower prevalence of luminal A tumors could contribute to the poor prognosis of young African American women with breast cancer.

PURPOSE: Expression profiling studies classified breast carcinomas into estrogen receptor (ER)+/luminal, normal breast-like, HER2 overexpressing, and basal-like groups, with the latter two associated with poor outcomes. Currently, there exist clinical assays that identify ER+/luminal and HER2-overexpressing tumors, and we sought to develop a clinical assay for breast basal-like tumors. EXPERIMENTAL DESIGN: To identify an immunohistochemical profile for breast basal-like tumors, we collected a series of known basal-like tumors and tested them for protein patterns that are characteristic of this subtype. Next, we examined the significance of these protein patterns using tissue microarrays and evaluated the prognostic significance of these findings. RESULTS: Using a panel of 21 basal-like tumors, which was determined using gene expression profiles, we saw that this subtype was typically immunohistochemically negative for estrogen receptor and HER2 but positive for basal cytokeratins, HER1, and/or c-KIT. Using breast carcinoma tissue microarrays representing 930 patients with 17.4-year mean follow-up, basal cytokeratin expression was associated with low disease-specific survival. HER1 expression was observed in 54% of cases positive for basal cytokeratins (versus 11% of negative cases) and was associated with poor survival independent of nodal status and size. c-KIT expression was more common in basal-like tumors than in other breast cancers but did not influence prognosis. CONCLUSIONS: A panel of four antibodies (ER, HER1, HER2, and cytokeratin 5/6) can accurately identify basal-like tumors using standard available clinical tools and shows high specificity. These studies show that many basal-like tumors express HER1, which suggests candidate drugs for evaluation in these patients.

The International Cancer Genome Consortium (ICGC) was launched to coordinate large-scale cancer genome studies in tumours from 50 different cancer types and/or subtypes that are of clinical and societal importance across the globe. Systematic studies of more than 25,000 cancer genomes at the genomic, epigenomic and transcriptomic levels will reveal the repertoire of oncogenic mutations, uncover traces of the mutagenic influences, define clinically relevant subtypes for prognosis and therapeutic management, and enable the development of new cancer therapies.

BACKGROUND: Validation of a novel gene expression signature in independent data sets is a critical step in the development of a clinically useful test for cancer patient risk-stratification. However, validation is often unconvincing because the size of the test set is typically small. To overcome this problem we used publicly available breast cancer gene expression data sets and a novel approach to data fusion, in order to validate a new breast tumor intrinsic list. RESULTS: A 105-tumor training set containing 26 sample pairs was used to derive a new breast tumor intrinsic gene list. This intrinsic list contained 1300 genes and a proliferation signature that was not present in previous breast intrinsic gene sets. We tested this list as a survival predictor on a data set of 311 tumors compiled from three independent microarray studies that were fused into a single data set using Distance Weighted Discrimination. When the new intrinsic gene set was used to hierarchically cluster this combined test set, tumors were grouped into LumA, LumB, Basal-like, HER2+/ER-, and Normal Breast-like tumor subtypes that we demonstrated in previous datasets. These subtypes were associated with significant differences in Relapse-Free and Overall Survival. Multivariate Cox analysis of the combined test set showed that the intrinsic subtype classifications added significant prognostic information that was independent of standard clinical predictors. From the combined test set, we developed an objective and unchanging classifier based upon five intrinsic subtype mean expression profiles (i.e. centroids), which is designed for single sample predictions (SSP). The SSP approach was applied to two additional independent data sets and consistently predicted survival in both systemically treated and untreated patient groups. CONCLUSION: This study validates the "breast tumor intrinsic" subtype classification as an objective means of tumor classification that should be translated into a clinical assay for further retrospective and prospective validation. In addition, our method of combining existing data sets can be used to robustly validate the potential clinical value of any new gene expression profile.