Immunohistochemical correlates of TP53 somatic mutations in cancer

Abstract

Despite controversy on the correlation between p53 accumulation and TP53 mutational status, immunohistochemical (IHC) detection of overexpressed protein has long been used as a surrogate method for mutation analysis. The aim of our study was to characterise the IHC expression features of TP53 somatic mutations and define their occurrence in human cancers. A large-scale database analysis was conducted in the IARC TP53 Database (R17); 7878 mutations with IHC features were retrieved representing 60 distinct tumour sites. The majority of the alterations were immunopositive (p <0.001). Sex was known for 4897 mutations showing a female dominance (57.2%) and females carrying negative mutations were significantly younger. TP53 mutations were divided into three IHC groups according to mutation frequency and IHC positivity. Each group had female dominance. Among the IHC groups, significant correlations were observed with age at diagnosis in breast, bladder, liver, haematopoietic system and head & neck cancers. An increased likelihood of false negative IHC associated with rare nonsense mutations was observed in certain tumour sites. Our study demonstrates that p53 immunopositivity largely correlates with TP53 mutational status but expression is absent in certain mutation types.Besides, describing the complex IHC expression of TP53 somatic mutations, our results reveal some caveats for the diagnostic practice.

Keywords: IARC TP53 Database; TP53 mutations; cancer; p53; immunohistochemistry.

Figure 1. Classification of TP53 somatic mutations based on frequency and IHC positivity in IARC TP53 Database

Somatic TP53 alterations were categorized into three groups: ‘hot spot’ mutations (marked with blue triangle) are frequent in Group A (mutation frequency ≥ 0.013 and IHC positivity ≥ 0.85); common nonsense mutations (blue circles) are in Group B (mutation frequency ≤ 0.011 and IHC positivity ≤ 0.44); less frequent and mostly positive missense mutations (yellow square) fall into Group C (mutation frequency ≤ 0.072 and IHC positivity ≥ 0.74). Each data point represents distinct TP53 mutations and only individual mutations with 15 or more IHC results were considered.

Figure 2. Significant differences of mean age of TP53 mutation carriers between IHC groups were found in five tumour sites

Bar graphs represent data in mean ± standard deviation (SD). Level of significance: * (p < 0.05), ** (p < 0.01), and *** (p < 0.001).

Figure 3. Frequency differences of nonsense TP53 (Group B) mutations across tumour sites

The ‘heat map’ shows the different frequency of the seven nonsense TP53 mutations (ranging from −3 to 3) across tumour sites (compared to their overall frequency in IARC database). A heat-map of nonsense TP53 mutations were constructed using Gene-E version 3.0.204 (http://www.broadinstitute.org/cancer/software/GENE-E/index.html). Level of significance: * (p < 0.05), ** (p < 0.01), and *** (p < 0.001).