Gene Expression Signatures and Immunohistochemical Subtypes Add Prognostic Value to Each Other in Breast Cancer Cohorts

Abstract

Purpose Gene signatures and Ki67 stratify the same breast tumor into opposing good/poor prognosis groups in approximately 20% of patients. Given this discrepancy, we hypothesized that the combination of a clinically relevant signature and IHC markers may provide more prognostic information than either classifier alone. Experimental Design We assessed Ki67 alone or combined with ER, PR and HER2 (forming IHC subtypes), and the research versions of the Genomic Grade Index, 70-gene, cell-cycle score, recurrence score (RS), and PAM50 signatures on matching TMA/whole tumor sections and microarray data in two Swedish breast cancer cohorts of 379 and 209 patients, with median follow-up of 12.4 and 12.5 years, respectively. First, we fit Cox proportional hazards models and used the change in likelihood ratio (Δ LR) to determine the additional prognostic information provided by signatures beyond that of (i) Ki67 and (ii) IHC subtypes. Second and uniquely, we then assessed whether signatures could compete well with pathology-based IHC classifiers by calculating the additional prognostic information of Ki67/IHC subtypes beyond signatures. Results In cohort 1, only RS and PAM50 provided additional prognostic information beyond Ki67 and IHC subtypes (Δ LR-χ2 Ki67: RS = 12.8, PAM50 = 20.7, IHC subtypes: RS = 12.9, PAM50 = 11.7). Conversely, IHC subtypes added prognostic information beyond all signatures except PAM50. Similar results were observed in cohort 2. Conclusions RS and PAM50 provided more prognostic information than the IHC subtypes in all breast cancer patients; however, the IHC subtypes did not add any prognostic information to PAM50.

Role of dynamic contrast‐enhanced MRI in evaluating the association between contralateral parenchymal enhancement and survival outcome in ER‐positive, HER2‐negative, node‐negative invasive breast cancer

“…Although tumor grade is also a marker of proliferation, gene expression signatures have been reported to add prognostic information beyond that provided by tumor grade . In addition, a recent study reported that gene expression signatures failed to provide significant prognostic information beyond Ki‐67 in ER + /LN – patients, whereas all signatures demonstrated prognostic strength beyond Ki‐67 or immunohistochemical subtypes in ER + /LN + patients . This may imply that contralateral BPE may provide less additional information after consideration of tumor proliferation activity, and further risk stratification beyond Ki‐67 would be even more difficult in ER‐positive, LN‐negative patients, who already have a good prognosis.…”

A pan-cancer analysis of the frequency of DNA alterations across cell cycle activity levels

“…We found that our cell cycle score gene expression signature, previously established in a breast cancer setting [11,12], provided independent prognostic information on a pan-cancer level. This signature was originally conceived as simple biological measure of cell cycle activity in response to the dependence of more established commercial gene expression signatures on multiple cell cycle/cell proliferation genes for their prognostic capacity [19].…”

“…Clusters with less than 20 tumours were excluded from further analysis. Cell Cycle Score (CCS) signature was applied as previously published [11,12]. Briefly, we extracted gene expression data from 433 of 463 signature CCS genes from all pan-cancer tumours and summed their values on an individual tumour basis to derive a single score of cell cycle activity for each sample.…”

“…The Cancer Genome Atlas (TCGA). Specifically, we examine gene expression levels, gene mutational frequency and chromosome arm-level alterations across pan-cancer tumours grouped into tertiles of cell cycle activity on the basis of our cell cycle score (CCS) gene signature [11,12]. Finally, we also determine the clinical relevance of this signature across and within cancer types using survival analyses including Kaplan-Meier graphs and multivariable Cox proportional hazards modeling adjusting for patient and tumour characteristics.…”

PAM50 Provides Prognostic Information When Applied to the Lymph Node Metastases of Advanced Breast Cancer Patients

“…Signatures were chosen on the basis of their relevancy in an on-going Swedish clinical trial (20) and owing to their use in a routine clinical setting. The cell-cycle score (CCS) was derived by adding the expression of cell-cycle genes identified from three different databases (KEGG, HGNC, Cyclebase) and splitting the resulting continuous variable into tertiles of low, intermediate, and high cell-cycle activity, further details here (21). …”