Bone Marrow Patterns of Expression of p53, Caspase-3, Ki-67, P21^waf and WT-1 Proteins May Provide a Tool To Differentiate Inherited from Acquired Cytopenias: A Study of 199 Patients.

Cell cycle regulation plays a major role in controlling hematopoiesis. Loss of appropriate mechanisms regulating cell cycle may lead to increased bone marrow apoptosis with subsequent cytopenias or to the accumulation of DNA-damaged cells, with an associated increased risk of leukemogenesis. p53 protein, a regulator of cell cycle, induces maturation arrest and apoptosis through its downstream mediator p21^waf or other pathways. Cleaved caspase-3 is a common pathway for most apoptotic mechanisms. Wilms’ tumor 1 (WT-1) protein, which is normally not detected in the bone marrow, has been correlated with disease progression in myelodysplastic syndrome. Ki-67 is a marker of cell proliferation. To study bone marrow (BM) cell cycle patterns in inherited and acquired cytopenias, we used immunohistochemical methods against p53 (DO-7, Dako, Carpinteria, CA), WT-1 (6F-H2, Dako), cleaved caspase-3 (Cell Signaling, Beverly, MA), Ki-67 (Dako), and p21^waf (Dako) in BM biopsies from patients with various inherited bone marrow failure syndromes (IBMFS, 68), low blast refractory cytopenias (RC, 29), acquired aplastic anemia (AA, 10) and acquired cytopenias (AC, 92 of infectious, endocrine, hepatic, and renal origins). The IBMFS included: 16 Shwachman-Diamond syndrome, 18 Diamond Blackfan anemia, 15 Fanconi’s anemia, 4 dyskeratosis congenita, 1 Kostman’s syndrome, and 14 unclassified inherited neutropenias, as well as 21 control subjects. p53 protein was significantly overexpressed (≥1% positive marrow cells) in 68% of IBMFS, 52% of RC, and none of the AA, AC, or control subjects. Intense p53 protein overexpression (≥5%) was detected in 83% of the p53-positive IBMFS but in only 26% of the p53-positive RC patients (P<0.0001). None of the subjects expressed WT-1 protein while the overexpression of p21^waf was rare. Detectable caspase-3 (any positive cells) was seen in 40% of IBMFS, 55% of RC, 20% of AA, 33% of AC, and 52% of the controls with no statistical significance among the groups. Ki-67 staining was detected in 64% of IBMFS (p<0.00001 compared to controls), 28% of RC, 0% of AA, 7% of AC, and 9% of the controls. Most IBMFS marrows with p53 protein overexpression had concomitant Ki-67 expression (80%) compared to 40% of RC suggesting deregulation of cell cycle control pathways. p53 protein overexpression and caspase-3 expression were not correlated indicating a defective p53 apoptotic pathway in IBMFS and RC. The following immunohistochemical patterns are observed: IBMFS p53+ intense, Ki67+; RC p53+ weak, Ki67±; AA p53-, Ki-67-; AC p53-, Ki-67+. WT-1 protein plays no role in the development of RC or the RC stage of IBMFS. The p53 pathway may play a role in leukemogenesis in RC and IBMFS. The absence of a positive correlation between p53 protein overexpression and increased caspase-mediated apoptosis in RC and IBMFS suggests that a dysfunctional p53 apoptotic pathway may play a role in the pathogenesis of these disorders. Disease-specific patterns of bone marrow cell cycle characteristics may provide an additional tool for discrimination of bone marrow failure disorders.