Fig. 5: Regulon activity signatures identify key biological processes in PPBC that predict poor prognosis in Young Women’s Breast Cancer.
From: Postpartum breast cancer has a distinct molecular profile that predicts poor outcomes
RNA expression profiles between postpartum breast cancer (PPBC, red, n = 9) and nulliparous breast cancer (NPBC, blue, n = 7) were evaluated for a transcriptional network activity through regulon analysis and b compared with regulon results comparing FFPE derived ER-negative to ER-positive breast cancer specimens. Most differentially active ER- regulons are highlighted as green boxes (a) and green circles (b). Most differentially active regulons between PPBC and NPBC are highlighted by bolded circles in red (upregulated) or blue (downregulated). c Single sample ESR1 regulon activity scores for PPBC (n = 9) and NPBC (n = 7) were evaluated. ICGC identified TP53 mutations are noted by orange-filled circles. Data are presented as a minimum to maximum with a median value marked by a line within the depicted interquartile range. d Gene expression values for the expressed (49) genes of the PAM50® subtype determination assay were evaluated to determine intrinsic subtype for each sample assessed for determination of sample clustering in PPBC and NPBC samples. e Pseudo-Oncotype Dx® recurrence scores were derived from RNA expression values for each sample and compared between cohorts (NPBC n = 7, PPBC n = 9). Data are presented as mean values ±SEM. f A postpartum breast cancer regulon-based gene expression signature was composed incorporating immune exhaustion (Fig. 4d), proliferation (E2F1), P53, and ESR1 regulon activity values and evaluated for prognostic significance from a multi-study accumulated cohort of Young Women’s Breast Cancer (n = 311) composed of female breast cancer patients whose primary breast cancer diagnosed occurred at the age of 45 or under. g Subset analysis in ER-positive cases (n = 214) from this YWBC cohort. Cohorts were split into PPBC signature high (hi, red, n = 107) or low (lo, black, n = 107) cohorts based upon the median value of the group plotted. P values were determined by Students’ unpaired two-tailed t test with Welch’s correction (c, e) or by two-tailed log-rank (Mantel–Cox) evaluation for survival plots (f, g). Log-rank evaluated Hazard Ratios (HR) are depicted.
