Can Immunohistochemistry Accurately Identify Patients with TP53 Mutation?

Introduction:

Mutated TP53 correlates with worse outcomes and treatment resistance in acute myeloid leukemia (AML). In addition, studies have shown that hypomethylating agents (HMA) are more effective than chemotherapy in inducing remission in patients with TP53 mutations. Although next generation sequencing (NGS) is an accurate method to assess TP53 mutations, results may take up to two weeks which could delay initiating treatment, impact treatment choice, and potentially affect outcome. In this retrospective analysis, we evaluated immunohistochemical (IHC) staining in correlation with TP53 mutations by NGS in patients with AML.

Methods:

A retrospective analysis was conducted at SUNY Upstate University Hospital. Records were reviewed from November 2020 to March 2024. Subjects were at least 18 years old and diagnosed with AML. NGS was performed at diagnosis on bone marrow aspirate or peripheral blood obtained at the same time as the bone marrow biopsy. NGS was performed at one of three laboratories: NeoGenomics, FoundationMedicine, or the molecular laboratory at SUNY Upstate. Each laboratory analyzed a panel of genes via NGS, including TP53. Patients with any type of TP53 mutation/deletion were selected as the positive control group while patients from the same timeframe with a wild-type P53 (wtp53) were selected as the negative control group.

P53 immunohistochemical staining (D07 mouse monoclonal antibody from Cell Marque, TM) was performed on all patient samples. Two hematopathologists at SUNY Upstate, blinded to the patient information, but not to the study design, interpreted and classified IHC staining based on stain strength into 3 categories: +1 (<10%), +2 (10-90%), +3 (>90%). A strong diffuse pattern (>90%) was considered positive, while weaker staining (<10%) was negative. For samples categorized as +2, attempts were made to classify them as positive or negative. The Negative Predictive Value (NPV) of IHC for predicting TP53 mutation was calculated considering repeated measures for stain strength. NPV was calculated using formula where TN are true negatives and FN are false negatives and aggregating the data from repeated measures across subjects.

Results

Thirty patients were included, 14 with a TP53 mutation (positive control group) and 16 with a wtp53 (negative control group). The median age of the patients was 69 years (range 26-88 years), with a male-to-female ratio of 2:1. In the positive control group, TP53 mutations identified by NGS included 13 missense mutations and 1 deletion with frameshift.

P53 immunohistochemical staining was performed on all 30 samples, either the bone marrow core biopsy (7 positive control cases, 11 negative control cases) or clot aspirate (7 positive control cases, 5 negative control cases). Out of the thirty samples, 9 were interpreted as +3, 9 as +1, and 12 as +2. The positive predictive value (PPV) for +3 IHC staining in correlation with positive TP53 on NGS was 100%. The NPV for +1 IHC staining in correlation with negative TP53 on NGS was 89%. Patient samples with +2 IHC staining had a PPV of 43% and NPV of 80%. There was discordance between pathologists' interpretations in 4 of 30 cases, all in samples with +2 staining.

Conclusion:

In our small study, p53 IHC staining with a strength of 3+ correlated with a 100% likelihood that the patient truly had a TP53 mutation. Conversely, p53 IHC results with a stain strength of 1+ had an 89% probability that the individual did not have a TP53 mutation. Results with p53 IHC intermediate staining of 2+ had a higher NPV (80%) compared with PPV (43%) suggesting these patients were more likely not to have a TP53 mutation. These results support using positive (3+) and negative (1+) p53 IHC staining at the time of AML diagnosis to guide treatment choice and initiate treatment in a timely manner. Given the lower PPV and NPV with intermediate (2+) staining, NGS would offer more conclusive results on which to base these decisions. This study provides preliminary results which should be validated in a larger prospective trial.

No relevant conflicts of interest to declare.