BACKGROUND: In estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer, first-generation genomic signatures serve predominately as prognostic biomarkers and secondarily as predictors of response to chemotherapy. We compared both the prognostic and predictive value of histologic grades and genomic markers. METHODS: We retrieved publicly available cDNA microarray data from 1373 primary ER(+)/HER2(-) breast cancers and developed a genomic signature simulated from Recurrence Online (http://www.recurrenceonline.com/) to calculate the recurrence score and risk using predefined sets of genes in the cDNA microarray. We then compared the prognostic and predictive information provided by histologic grade and genomic signature. RESULTS: Based on genomic signatures, 55%, 28%, and 17% of breast cancers were classified as low, intermediate, and high risk, respectively, whereas the histologic grades were I, II, and III in 22%, 59%, and 19% of breast cancers, respectively. Univariate analysis in the untreated cohort revealed that both histologic grade (overall P = .007) and genomic signature (P < .001) could predict prognosis. Results were similar using the genomic signature, with pathologic complete response rates of 4.6%, 5.7%, and 16.5% for low-, intermediate-, and high-risk cancers, respectively. Neither biomarker was statistically significant in multivariate analysis for predictive response to neoadjuvant chemotherapy (NAC). CONCLUSION: Genomic signature was better at identifying low-risk cases compared to histologic grade alone, but both markers had similar predictive values for NAC response. Better predictive biomarkers for NAC response are still needed.
BACKGROUND: In estrogen receptor (ER)-positive, humanepidermal growth factor receptor 2 (HER2)-negative breast cancer, first-generation genomic signatures serve predominately as prognostic biomarkers and secondarily as predictors of response to chemotherapy. We compared both the prognostic and predictive value of histologic grades and genomic markers. METHODS: We retrieved publicly available cDNA microarray data from 1373 primary ER(+)/HER2(-) breast cancers and developed a genomic signature simulated from Recurrence Online (http://www.recurrenceonline.com/) to calculate the recurrence score and risk using predefined sets of genes in the cDNA microarray. We then compared the prognostic and predictive information provided by histologic grade and genomic signature. RESULTS: Based on genomic signatures, 55%, 28%, and 17% of breast cancers were classified as low, intermediate, and high risk, respectively, whereas the histologic grades were I, II, and III in 22%, 59%, and 19% of breast cancers, respectively. Univariate analysis in the untreated cohort revealed that both histologic grade (overall P = .007) and genomic signature (P < .001) could predict prognosis. Results were similar using the genomic signature, with pathologic complete response rates of 4.6%, 5.7%, and 16.5% for low-, intermediate-, and high-risk cancers, respectively. Neither biomarker was statistically significant in multivariate analysis for predictive response to neoadjuvant chemotherapy (NAC). CONCLUSION: Genomic signature was better at identifying low-risk cases compared to histologic grade alone, but both markers had similar predictive values for NAC response. Better predictive biomarkers for NAC response are still needed.