Loss of Heterozygosity in Bone Marrows of Patients with Iron-Deficiency Anemia.

The management of patients with low-risk myelodysplastic syndromes (LR-MDS) transitioned from supportive care to active therapeutic interventions focused on the improvement of hematopoiesis. Thus, it is now critical to early identify patients with this disorder for optimal management. The diagnosis of myelodysplastic syndrome is based on clinical data, morphologic features of the bone marrow and conventional cytogenetic analysis. However, the majority of patients with LR-MDS do not demonstrate abnormal marrow karyotypes. In addition, changes in bone marrow morphology similar to those seen in LR-MDS, i.e. mild dysplasia with no increase in blasts, occur in a variety of systemic illnesses. Thus, there is a need for more objective diagnostic methods. The detection of loss of heterozygosity (LOH) can be supportive of the diagnosis of MDS as it is widely accepted that genetic lesion(s), including loss of tumor suppressor genes, lead to clonal expansion of hematopoietic populations in myelodysplasia. Indeed, previous studies and our pilot series showed high incidence of allelic imbalance in MDS. However, there is no data on the baseline LOH in bone marrows of age-matched controls without primary bone marrow disorder. To further explore the utility of LOH analysis in refining the diagnosis of LR-MDS, we investigated the incidence of allelic imbalance in bone marrows of patients with iron-deficiency anemia. The LOH analysis was performed using DNA extracted from formalin-fixed, paraffin-embedded bone marrow clot sections. Unrelated non-neoplastic tissues from the same patients served as controls. The oligonucleotide primers were selected based on previously reported high frequency of involvement in MDS and AML (D1S450, D11S1363, IRF1, D11S1338 and WT1). Nineteen patients were included in the study [median age 71 years, range 38–81 years; 8 males; median hemoglobin and MCV of 9.4 g/dL and 86 fL (normal range 80–94 fL)]. All patients showed depleted iron stores on bone marrow aspirate smears. Review of bone marrow morphology and subsequent follow-up showed no evidence of primary bone marrow disorder. Karyotypes, available in 7 patients, were normal, with the exception of one case showing loss of chromosome Y. Ninety two percent of the samples were informative. The overall frequency of LOH for all loci was 16% (12–21%, Tab. 1). LOH was seen in 10 cases (1 locus involved in 6 cases, 2 loci affected in 4 patients). LOH was not seen at any loci in the control samples from 19 non-bone marrow tissues from the same patients. In conclusion, we demonstrated a significant rate of LOH in non-neoplastic bone marrow tissue. For selected loci, the frequency of LOH approximates the rate seen in MDS samples. However, in the contrary to MDS group, no more than two markers were involved in any one patient with iron-deficiency anemia. The use of LOH analysis in the diagnosis of MDS may require selection of the primers based upon the background frequency of LOH in normal populations. The detection of LOH at numerous loci (more than two), may still serve as a valuable ancillary diagnostic tool. A prospective study of larger series will be necessary to confirm these findings and to address their clinical and biological significance.