TP53 Mutational Analysis Enhances the Prognostic Accuracy of IHC4 and PAM50 Assays

Breast cancer is one of the leading malignancies in women, with an increasing incidence over the past 2 decades. With the advent of screening, a large number of patients are being diagnosed at an early stage and have a favorable prognosis. However, these patients have a certain recurrence rate, depending on the clinicopathological features. Adjuvant chemotherapy can reduce the recurrence risk, but it has moderate adverse effects. The established clinicopathological features are insufficient to guide patients for adjuvant chemotherapy, and therefore substantial over- or under-treatment can occur. Several multi-gene tests, such as MammaPrint1,2, Oncotype DX3,4, and PAM505,6, have been shown to provide additional prognostic information over classical clinicopathological factors. To improve the applicability, Cuzick et al. constructed the IHC4 score by combining 4 widely examined immunohistochemical markers, namely, estrogen receptor (ER), progesterone receptor (PgR), human epidermal growth factor receptor 2 (HER2; including fluorescent in situ hybridization in the IHC 2+ group), and Ki-677. Previous studies have shown that the IHC4 score provides similar prognostic information compared with the Oncotype DX recurrence score (RS)7 and PAM50 risk of recurrence score (ROR)8, and one study suggested that IHC4 has the potential to be the most cost-effective prognosis tool9. Spearman correlations of risk groups defined by the RS, ROR, or IHC4 score are only modest to moderate (RS and IHC4 scores: r = 0.72; ROR and IHC4 scores: r = 0.48; RS and ROS scores: r = 0.39)7,8. Although a considerable difference exists among these 3 assays, combinations of 2 assays out of the 3 assays did not significantly improve the prognostic value7,8. A similar prognostic power, modest-to-moderate concordance, and the absence of a significant additive effect of these assays suggest the need for incorporating other factors to improve the prognostic accuracy. The PIK3CA and TP53 somatic mutations are the 2 most frequently mutated genes in breast cancer, and their frequencies are much higher than other somatic mutations10,11. In contrast to the conflicting findings about prognostic value of PIK3CA mutation12,13,14,15, TP53 mutation has been consistently shown to predict poor outcomes in 2 meta-analyses (hazard ratio [HR] = 2.0 and 2.27)16,17. Mutant P53 in breast cancer may act at various cancer stages, such as early tumorigenesis, tumor growth, and metastasis. TP53 mutation in breast cancer is associated with high-grade tumor behavior, and is molecularly distinct from wild type tumor18,19. MammaPrint, Oncotype DX, PAM50, and IHC4 mainly focus on measuring tumor proliferation on the basis of protein or mRNA expressions. The broad-spectrum effects of TP53 mutation may be more likely to show additive effect on these assays. We determined whether incorporating TP53 mutation can enhance the prognostic accuracy of the IHC4 and PAM50 assays.