Introduction
Acute myeloid leukaemia (AML) is a heterogenous disorder that arises from clonal expansion of malignant hematopoietic precursor cells. The tumor suppressor p53 gene plays an important role in regulation of the cell cycle and apoptosis. Somatic mutations of the p53 gene have been reported in 5-10% of all patients with AML, although the rate is higher in therapy-related disease and elderly patients. Alteration or loss of p53 is one of the most powerful independent indicators of poor outcome. BCL2 positivity in AML and its impact on prognosis is less well described.
Methods
This is a single-centre, retrospective analysis of AML patients treated over a ten year period (2006-2016). Patients were identified through our pathology database. P53 expression (as a surrogate marker for tp53 mutation) and BCL2 expression were analysed by immunohistochemical (IHC) analysis of marrow biopsies (greater than 30% was considered positive). We analyzed AML presentation, management and overall survival (OS), and correlated these with presence of p53 mutation and BCL2 expression.
Results
We identified 48 patients; the majority were elderly (median age 68.5 years). Thirty-four patients (71%) haddenovopresentation of AML, 6 had underlying Myelodysplastic Syndrome (13%), 4(8%) progressed to AML from underlying Myelofibrosis, and 8% had therapy-associated AML. Patient and disease characteristics are outlined inTable 1.
Median blast count by IHC of bone marrow biopsies was 70%. The results of cytogenetic analysis were available for 38 patients (79%). Fourteen patients (37%) had normal cytogenetics, but a wide range of genetic abnormalities were identified(Table 1). NPM1 (for 22 patients) and FLT3 mutation analysis (for 33 patients) were available, and reported as positive in 4(18%) and 5(15%) respectively.
p53 overexpression was identified in 6 patients (12.5%). 50% of these had secondary AML, and the remainder weredenovopresentations. Four of these patients (66%) had complex cytogenetics, one had deletion 5q and cytogenetic analysis was not available for the remaining patient. Median age of patients who were p53 positive was 60 years, compared to 70 years in those without p53 overexpression.
BCL2 expression analysis showed a median value of 60% (range 5-100%). BCL2 positivity was seen in 39 patients (81.2%); half of this cohort (15 patients) had normal cytogenetics and only five patients (16%) had abnormalities that indicated poor risk cytogenetics (complex, deletion 7q, etc.).
Thirty-nine patients (83%) received active treatment and nine received supportive care only due either to frailty or patient choice. Twenty-two patients (60%) had induction chemotherapy, 7(19%) received Azacitadine, and a further 7 patients had low-dose cytarabine. Nineteen patients (51%) had relapsed/refractory disease. Five patients (13.5%) with BCL2 positive disease received Venetoclax as a single agent for relapsed or refractory disease.
Median OS was 11.5 months (range 0-72); 39 patients (81%) had died at the time of analysis. 26 died from disease progression (79%), and 7(19%) of sepsis. One patient was lost to follow up. OS in those with wild-type p53 was 11.5 months, which compared favorably to 7.5 months in those with p53 mutation. BCL2 overexpression also conferred a poorer prognosis; median OS was 10 months compared to 17 months in BCL2 negative disease.
Conclusion
This study reflects the potential use of p53 and BCL2 in real-life practice in assessing prognosis in AML patients outside of a clinical trial setting. We confirmed the previously reported poor prognosis of p53 alterations in AML (OS 7.5 months), and the study also indicates a poorer prognosis in patients with BCL2 positivity. The widespread availability of IHC techniques for trephine biopsies and the potential for rapid turn-around times in routine clinical laboratories suggests that further studies of the prognostic impact of such gene alterations are warranted.
Disclosures
No relevant conflicts of interest to declare.
PRIMA-1 induces apoptosis in acute myeloid leukaemia cells with p53 gene deletion