Prognostic Significance Of TP53 mutations and Single Nucleotide Polymorphisms In Acute Myeloid Leukemia

The response to treatment and overall survival (OS) of patients with acute myeloid leukemia (AML) is variable, with a median OS ranging from several months to more than 10 years. Age at diagnosis, performance status (PS), and karyotype expression have long been established in prognostication. Loss of TP53, a tumor suppressor gene located on the short arm of chromosome 17, is one of the most frequent genetic abnormalities in human cancer and is one of the more promising prognostic markers for AML. Studies have shown that TP53 mutations are present in 5-25% of all AML patients, in 70% of those with complex karyotypes, and are associated with old age, chemotherapy resistance, and worse OS. Single nucleotide polymorphisms (SNPs), changes in DNA seen in an appreciable amount of the population, have been examined in AML and studies have suggested a possible correlation with worse outcomes. Using genetic sequencing, we set out to look at our own experience with AML, and hypothesized TP53 mutations and SNPs would mimic the literature, occurring in a minority of patients, and conferring a worse OS.

We performed a pilot study of randomly selected, newly diagnosed AML patients at Mount Sinai Medical Center, diagnosed from 2005-2008 (n =10). Immunohistochemical (IHC) analysis of bone marrows and peripheral blood smears was assessed via DO-1 antibody on paraffin embedded tissue. Conventional cytogenetic analyses were performed on short-term cultured bone marrow and peripheral blood cells with the use of the GTG-banding technique. TP53 PCR sequencing was performed using DNA from bone marrow smears using the Sanger sequencing platform and resolved by capillary electrophoresis. Analysis was performed using Mutation Surveyor software with confirmation of the variants using the COSMIC and dbSNP databases. Descriptive frequencies and median survivals were calculated for demographic information, prognostic factors, and treatment variables. A univariate analysis was performed.

The majority of patients in our pilot study were older than age 60 (80%), male (60%), Hispanic (60%), and had a poor PS (ECOG 2-3: 60%). Most patients had de-novo AML (50%) with an intermediate (50%) non-complex (70%) karyotype and a TP53 P72R SNP (50%). Fewer than half of these patients harbored TP53 mutations (40%). There was no significant difference in OS based on sex, AML history, risk-stratified karyotype, or TP53 mutation. There was a trend toward improved survival among patients younger than age 60 (11, 4 mo, p = 0.09), of Hispanic ethnicity (8, 1 mo, p = 0.11), and those not harboring P72R (8, 2, p = 0.10). There was a significant improvement in survival among patients with a better PS (28, 4 mo, p = 0.01) and those who did not have a complex karyotype (8, 1 mo, p = 0.03). Among patients with a TP53-mutation, there were a larger number of individuals who were younger than age 60 (25.0, 16.7%), who were male (75.0, 50.0%), had a good performance status (ECOG 0-1: 50.0, 16.7%), had de-novo AML (50.0, 66.7%), and who had an adverse karyotype (50.0, 33%). Patients with a P72R SNP were more often male (80, 40%) and had a worse PS (ECOG 2-3: 80, 40%) with AML secondary to MDS (60, 20%) and a complex karyotype (40, 0%). The most commonly observed TP53 mutation was a missense N310K (40%) and the most commonly observed SNP was P72R (100.0%). Patients with more than one TP53 mutation had a worse clinical course than those with only a single mutation.

Our study demonstrated that poor PS and the presence of a complex karyotype were associated with a decreased OS. TP53 mutations were relatively uncommon, occurring more frequently in male patients with an adverse karyotype. Although there was no significant difference in survival between TP53 mutated and un-mutated patients, there was a trend toward worse OS among patients with a specific SNP. These results suggest that different TP53 mutations and SNPs should not be treated the same, and that some may confer a worse prognosis than others. Larger studies are needed to validate these findings.

No relevant conflicts of interest to declare.