Identification of del(20q) in a subset of patients diagnosed with idiopathic thrombocytopenic purpura

Abstract

The American Society of Haematology practice guidelines for idiopathic thrombocytopenic purpura (ITP) suggests a bone marrow examination in patients older than 60 years of age, in whom splenectomy is considered, or in whom an appropriate response to therapy is never achieved (George et al, 1996). Thrombocytopenia is the most frequent cytopenia in myelodysplastic syndrome (MDS) patients presenting with del(20q) on karyotypic analysis (Menke et al, 1992; Sashida et al, 2003). In these cases, the dysplasia is often subtle and distinction between MDS and ITP may be difficult (Kurtin et al, 1996; Gupta et al, 2007). Given the association between del(20q) and a thrombocytopenic presentation of MDS and the subtle nature of dysplasia in such cases, we hypothesized that some cases of ITP with normal karyotype on routine cytogenetic analysis may harbor a del(20q) that is detectable solely by interphase fluorescence in situ hybridization (FISH) and may represent a form of MDS that is missed on routine evaluation. Interphase FISH analysis was performed on unstimulated 24-h cultures of thawed frozen bone marrow cell pellets from 23 adult patients diagnosed with ITP. The reasons for bone marrow examination included age >60 years (12 patients), history of lymphoma (one patient), and failure to respond to ITP therapy (five patients); the reason for bone marrow examination was unknown in five patients. FISH was performed using the Vysis LSI D20S108 probe (Abbott Molecular, Des Plaines, IL, USA) to the commonly deleted region on 20q12. One hundred nuclei were analysed in each case. Cut-off values for the 20q12 probe were determined on 20 control thawed frozen cell pellets from negative staging bone marrow samples obtained from lymphoma patients. A positive del(20q) result was determined to be >6/100 nuclei with loss of a 20q12 signal (mean of all control samples +3 standard deviations). Bone marrow karyotypic analysis was normal in all cases. Bone marrow flow cytometry had been performed in all cases and did not show any clonal B-cell population, abnormal T-cell population, or increased CD34+ blast population. Haematoxylin and Eosin-stained bone marrow trephine biopsy sections and Wright-Giemsa-stained aspirate smears were analysed for cellularity, megakaryocyte number and clustering, and morphological dysplasia in the erythroid, myeloid, and megakaryocytic lineages. Peripheral blood counts (at baseline and follow-up), treatment, and clinical outcome were compared between the groups with and without del(20q). Mean, median, and Standard Deviation (SD) were calculated for continuous numeric data. The student’s t-test was used to compare numerical laboratory data. The Fischer’s exact test was used to compare morphological features. The P values reported are two sided and all values <0·05 were considered significant. We found that 12/23 (52%) cases were positive for a del(20q) (range 7–23/100 cells, median 10/100 cells). The median age for the del(20q) cases was 57 years (range: 27–83 years), which was not significantly different from the median age of 67 years (range: 24–78 years) in the non-del(20q) group. The male to female ratio was 8:4 in the del(20q) group and 8:3 in the non-del(20q) group. The median bone marrow cellularity was 50% in both groups. There were no significant differences between the groups in the quantity or clustering of megakaryocytes nor was there sufficient dysplasia to warrant a diagnosis of MDS in any case in either group. Small lymphoid aggregates were more commonly present in the del(20q) group (five cases) compared to the non-del(20q) group (one case), but this finding did not reach statistical significance (P = 0·15). There were no major differences in the types of ITP treatments given to the two groups of patients (treatment data summarized in Table I); patients treated with vincristine, cyclosporine, or azathioprine had received these therapies after the bone marrow examination. Peripheral blood count data at baseline and latest followup (median follow-up 26 months, range 3–173 months) for both groups are shown in Table II. At latest followup after ITP therapy, only 1/9 del(20q) patients had achieved a normal platelet count compared to 7/11 non-del(20q) patients (P = 0·02). The 3 del(20q) group patients who underwent splenectomy had no significant response (baseline platelet counts 11–40 × 109/l and post-splenectomy platelet counts 32–41 × 109/l), while both patients from the non-del(20q) group had complete response to splenectomy (baseline platelet counts 5 and 9 × 109/l and post-splenectomy platelet counts 290 and 320 × 109/l, respectively). The apparently high incidence of the del(20q) abnormality in our cohort of patients may reflect some degree of selection bias, as many of the patients in our series were older and some had already shown poor response to standard ITP therapies. The presence of del(20q) raises the possibility that these patients may actually have a form of indolent MDS that is not morphologically evident on bone marrow examination (Bizzoni et al, 2006; Cooper & Bussel, 2006). In the current study however, dysplasia was insufficient in the del(20q) cases to allow a cytomorphological diagnosis of MDS. Del(20q) is a relatively common finding in MDS as well as in other myeloid neoplasms, but, unlike abnormalities such as −7 and del(5q), it is not considered a diagnostic cytogenetic marker of myeloid neoplasia in the absence of morphological dysplasia (Orazi A et al, 2008). Rare cases of isolated del(20q) have been associated with autoimmune disease in the absence of morphological dysplasia (Kurtin et al, 1996) and del(20q) is also the most common cytogenetic abnormality in cases of unexplained cytopenias that lack concomitant dysplasia on bone marrow examination, occurring in 18–21% of such cases with a cytogenetic abnormality (Steensma et al, 2003; Han & Theil, 2007). As the cases in our series were karyotypically normal, the identification of del(20q) by FISH presumably represents a small clone within the bone marrow. In spite of receiving similar treatments to the non-del(20q) group, the del(20q) group patients showed significantly more persistent thrombocytopenia and the 3 del(20q) group patients who underwent splenectomy did not respond. Thus, FISH for del(20q) is a potentially useful test in adult patients under evaluation for ITP, as it identifies a subset of patients unlikely to respond to standard ITP therapies, including splenectomy.