The biological, clinical and prognostic implications of p53 transcriptional pathways in breast cancers

We hypothesized that the functional status of p53 transcriptional pathways, rather than p53 protein expression alone, could accurately discriminate between low- and high-risk breast carcinoma (BC) and inform about individuals' tumour biological behaviour. To test this, we studied a well-characterized series of 990 BCs with long-term follow-up, immunohistochemically profiled for p53, its main regulators and downstream genes. Results were validated in an independent series of patients (n = 245) uniformly treated with adjuvant anthracycline-based chemotherapy. Eleven p53 transcriptional phenotypes were identified with just two main clinical outcomes. (a) Low risk/good prognosis group (active/partially inactive p53 pathways), defined as p53(+/-)/MDM4(+)/MDM2(+/-)/Bcl2(+/-)/p21(+/-), p53(-)/MDM4(-)/MDM2(+)/Bcl2(+)/p21(+/-) and p53(+/-)/MDM4(-)/MMD2(-)/Bcl2(+)/p21(+/-). These tumours had favourable clinicopathological characteristics, including ER(+) and long survival after systemic adjuvant-therapy (AT). (b) High risk/poor prognosis group (completely inactive p53 pathways), defined as p53(+/-)/MDM4(-) MDM2(-)/Bcl2(-)/p21(-), p53(-)/MDM4(-) MDM2(+)/Bcl2(-)/p21(-) and p53(+/-)/MDM4(-)/MDM2(-)/Bcl2(-)/p21(+). These tumours were characterized by aggressive clinicopathological characteristics and showed shortened survival when treated with AT. Completely inactive p53 pathways but intact p21 axis p53(+/-)/MDM4(-)/MDM2(-)/Bcl2(-)/p21(+) had the worst prognosis, particularly patients who received AT. Multivariate Cox regression models, including validated prognostic factors for both test and validation series, revealed that the functional status of p53 transcriptional pathways was an independent prognosticator for BC-specific survival (HR 2.64 and 4.5, p < 0.001, respectively) and disease-free survival (HR 1.93 and 2.5, p < 0.001, respectively). In conclusion, p53 functional status determined by assessment of p53 regulatory and downstream targets provides independent prognostic value and may help determine more adequate therapeutic regimens for specific subgroups of breast cancer patients.