P53 Protein Expression By Immunohistochemistry and Its Impact on Survival in Adults with Acute Lymphoblastic Leukemia

Introduction

The prognostic impact of TP53 mutations has been described for multiple hematological malignancies, however its role on adult acute lymphoblastic leukemia (ALL) remains unclear. These have been linked to worse outcomes in non-adolescent and young adult (AYA) patients with ALL, but this association has not been found in the AYA group. On the other hand, overexpression of p53 protein on immunohistochemistry (IHC) has been associated to heterozygous TP53 mutations, as well as decreased survival in multiple myeloma and acute myeloid leukemia. In ALL patients, this has only been evaluated in pretransplant evaluation, but its prognostic role at diagnosis is not known.

Methodology

We conducted a retrospective, cross-sectional observational study on patients newly diagnosed with ALL at a single Mexican third-level institution from 2011 to 2022, whose bone marrow biopsies were available. We performed indirect immunostaining on bone marrow biopsies using an anti-p53 antibody. Positive nuclei were counted in a field with at least 200 blasts and classified according to an intensity scale (0: no staining, 1+: weak staining, 2+: intermediate staining, and 3+: intense staining). Samples were considered positive if ≥5% of the analyzed cells had 3+ staining. We analyzed percentage of IHC reactivity and its association with disease outcomes. A 7% p53 expression was considered the ideal cutoff value, based on reports in medical literature. We obtained survival curves, according to this cutoff value, as well as for groups of AYA and patients aged >40.

Results

A total of 116 patients were included, with a median age of 34.5 years (17-79), 45.7% of whom were female. Median leukocytes at diagnosis were 10.4 x 10³/µl (0.4-488.0), with 39.7% of patients having a count >30 x10³/µl. A normal karyotype was found in 56.2% of patients, Philadelphia chromosome in 14.9%, and del17p in 4.1%. For p53 protein expression, 31.9% of patients had expression above 5%, and 29.3% above 7%. Among those who had a positive p53 expression (≥5%) the median percentage of expression was 15.8% (5-72). A higher median leukocyte count was found in patients with p53 protein expression ≥7% (46.3 x10³/µl[0.7-436.4] vs 5.06 x10³/µl [0.4-488], p=0.008). Patients with p53 expression ³7% more frequently had a leukocyte count >30 x10³/µl (55.9% vs. 32.9%, OR 2.58 [95% CI 1.14-5.85], p=0.036) and Philadelphia chromosome (26.8% vs.10.8%, OR 3.24 [95% CI 1.13-9.31], p=0.041, while the presence of del17p was not different. In patients with p53 expression ≥7%, overall survival (OS) was not different than those expressing <7% (HR 1.139 [95% CI 0.691-1.877], p=0.610).

We made a subgroup analysis between AYA and patients aged ≥40. Among AYA, p53 expression ³7% did not have an impact on OS (median 37.36 [0.31-74.99] vs. 32.16 [17.59-46.73] months, p=0.623), 2-year OS of 59.9% vs. 51.9%. Among the patients ≥40 years, OS was inferior in patients with p53 expression ³7% (median 5.50 [4.95-5.96] vs. 12.88 months [9.42-16.34], p=0.045), 2-year OS: 12.5% vs. 35.5%. In a multivariate analysis including Philadelphia chromosome and leukocyte count > 30 x10³/µl in the non-AYA population, p53 expression ≥7% was independently associated with mortality (HR 2.53 [95% CI 1.038-6.133], p=0.041).

Conclusions

According to our study, p53 protein expression on IHC has an independent prognostic impact for overall survival in ALL patients exclusively in the non AYA population, with expression ³7% on bone marrow blasts having a negative impact. This is associated with high-risk features, such as leukocyte count >30 x10³/µl at diagnosis and Philadelphia chromosome. Further studies are needed to confirm whether IHC p53 protein expression can serve as a surrogate for TP53 mutation detection via DNA analysis.