ALBERT

Description: Iron deficiency (ID) is generally easily diagnosed by a serum ferritin concentration less than 12 μg/L. However, in patients with concomitant pathologies, sensitivity of ferritin for diagnosing ID is low, since inflammation or liver disease can lead to normal or increased ferritin values, even when iron deficiency. In this study, we propose a new marker for diagnosing ID in patients with chronic disease: the Transferrin/Albumin ratio (Tf/A ratio). Indeed, the synthesis and the elimination of albumin and transferrin are regulated by similar processes, excepted in ID. In the latter, the synthesis of transferrin is stimulated, whereas that of albumin is not. Therefore, the Tranferrin/Albumin ratio increases in case of ID, even when transferrin values are within the normal range. To determine the accuracy of this novel parameter, we studied 75 patients with chronic disease who were submitted to bone marrow aspirates and iron staining. Iron stores depletion was defined by less than 10% sideroblasts, without extra-cellular iron nor siderocytes. Blood samples were routinely undertaken at the time of the medullar sampling for determination of hematological and biochemical parameters. Iron status including ferritin, Tranferrin Saturation (TfSat), soluble Transferrin Receptor (sTfR), sTfR/log ferritin and Tf/A ratio, was determined. The diagnostic accuracy of the Tf/A ratio was compared to previously described parameters of iron status that we cited above. Receiver Operating Characteristics (ROC) curves were built to determine the best cut-off values for the prediction of iron deficiency. According to the Perls’ reaction, 25 of the 75 patients (33%) had depleted iron stores and 50 had normal or increased iron stores. Sixteen iron-depleted patients (67%) had anemia. Mann and Whitney U test showed that parameters significantly associated with ID were: Tf, Tf/A ratio, ferritin, TfSat, sTfR, sTfR/log ferritin, mean corpuscular volume, mean corpuscular hemoglobin, red blood cell and reticulocyte counts. In a multivariate analysis, the only significant, independent predictor of iron depletion was the Tf/A ratio (r = 0.637, p 〈 0.005). The sensitivity/specificity of Tf/A ratio at a cut-off point of 6.4% as given by ROC curve were 80%/88%. In conclusion, the Tf/A ratio is useful in the detection of iron depletion in patients with chronic disease and could dispense with bone marrow aspirate and Perls’ reaction in more than 80% of cases.

Description: Abstract 3534 Sex chromosome loss is observed in the healthy population and increases with advancing age. It is also observed in hematological malignancies with variable clinical value. Its significance as a disease marker in CLL is not clearly defined. We undertook this study to evaluate whether Y or X loss is an age- and/or a CLL-associated abnormality, since it has not yet been reported. Out of 198 patients with stage B/C CLL analyzed by karyotype (K), 6 (3%) exhibited a loss of sex chromosome (Sutton et al, blood 2011). We also identified 14/745 (2%) additional patients in our CLL-database (2002–11). All patients had a Matutes score ≥ 4. K was performed on peripheral blood using TPA until 2005, and then CPG + IL2. FISH analyses were carried out to detect trisomy 12 and 11q22 (ATM), 13q14, 17p13 (TP53), X/Y deletions. IGVH gene mutational status was analyzed by PCR with a sequence homology cutoff of 98%. B- and T-cells were sorted by flow cytometry from peripheral blood of 5 patients (purity ≥ 94%) using CD19+ and CD3+ respectively. Polynuclear and mononuclear cells were enriched by Ficoll (purity ≥ 99%) for 4 patients. There were 13 males and 7 females (n =20). At the time of the K, there were 11 Binet stage A and 9 stage B, and median age was 64 (36–78). Regarding IGVH status, 9/12 (75%) cases were mutated. We observed 13 -Y and 7 -X, chromosome loss being the sole abnormality in 10 (50%) cases. The percentage of cells with -X or -Y ranged between 3 to 20/20 mitoses. K was complex (≥ 3 abnormalities) in 4 cases. When associated with another chromosomal abnormality, -X or -Y was the primary change in 2 cases, in the same clone in 7 cases, and as a sub-clone change in one case. FISH identified 16/20 (80%) del13q, 4/20 (20%) tri12, 3/20 (15%) delATM, 2/20 (10%) delTP53. Among del13q, there were 8 monoallelic, 7 concomitant mono- and biallelic, and 1 biallelic deletion. Sex chromosome loss was confirmed by interphase FISH in 14 patients, with percentage of loss between 5 to 84% (median: 68%). Co-hybridization with two probes including the X/Y probe was performed for 13 patients. The two analyzed abnormalities were present in independent clones in 4 cases, in the same cells in 2 cases; -X or -Y appeared as the primary change in 4 cases, and as a sub-clone change in 3 cases. The analyses of polynuclear and mononuclear populations in 4 patients showed a maximum of 2% of polynuclear cells with X or Y loss, whereas the mononuclear cells exhibited a significant higher loss frequency (range 6–87%, p =.03). The analyses of B- and T-cells in 5 samples showed a significant increase in losses frequency in B-cells (range 88–96%) compared to T-cells (2%, 2%, 3%, 4%, 6%) (p =.008). The incidence of -X in peripheral blood lymphocytes is about 3% of the cells in healthy women aged 16–50, rising to 5% in women aged over 85 years. The incidence of Y loss in men is distinctly lower, with a frequency less than 1% of the cells before 85 years (Wodja et al, J Appl Genet, 2003). The median of cells with sex chromosome loss in our series is 68%, with only 2 patients with low percentage: a 66-year old female (5%) and a 68-year old male (6%). However these percentages are higher than those observed in healthy people. Moreover, we never observed the loss of a sex chromosome in more than 2% in polynuclear cells of any of the patients, indicating that sex chromosome loss in CLL is associated with the tumor cells. It is not clear whether the low proportion of loss of sex chromosome in CD3+ cells that exceed in one case only slightly our laboratory threshold of 5% is due to contamination. Sex chromosome loss may occur during K evolution, as well as the acquisition of additional abnormalities may occur in cells with primary sex chromosome loss. Even if loss appears through an age-related effect in elderly cells, our data support an oncogenic property of this abnormality. One patient with –Y as the sole aberration in CLL, has developed a myelodysplastic syndrome upon treatment, with complex karyotype without -Y, suggesting that Y loss had been associated to CLL. In addition, del13q, especially when biallelic, appears to be closely associated to the loss of sex chromosome, as compared to our database (del13q: 16/20 (80%) vs 141/279 (50%), p =.01; biallelic del13q: 8/16 (50%) vs 24/141(17%), p =1×10−5). In conclusion, sex chromosome loss has to be considered as a clonal abnormality in CLL, is significantly associated with biallelic 13q deletion, and may participate to oncogenic transformation. Disclosures: No relevant conflicts of interest to declare.

Description: Transcription deregulation resulting from genomic rearrangement is frequently observed in human lymphoid malignancies. For example, chromosomal aberrations, such as translocations or inversions, involving the immunoglobulin (Ig), T cell receptor (TCR) and the BCL11B loci, are expected to juxtapose the strong regulatory elements of these genes to the affected genes on the other chromosomes, resulting in such activation. Recently, ectopic activation of two orphan homeobox genes located ~2Mb apart on chromosome 5q35, HOX11L2/TLX3 and CSX1/NKX2-5 have been described, as a result of their fusion to BCL11B on chromosome 14q32 in acute lymphoblastic leukemia of T-lineage. Fusion of TLX3 to TCRδ has also been reported. We describe a novel translocation t(5;14)(q35;q11) identified in a case of B-cell chronic lymphoproliferative disorder. FISH and molecular studies located the chromosome 14 breakpoint within the TCRδgene, at the Dδ3 segment, and the chromosome 5 breakpoint 2.5kb downstream (telomeric) of the NKX2-5 gene. As a result of this translocation, both NKX2-5 and TLX3 were transcribed, whereas only NKX2-5 was detected at the protein level. To investigate the expression of these two genes in B-cell chronic proliferations, we screened 20 B-cell chronic lymphocytic leukemias (B-CLL) and 21 atypical B-CLL samples by RT-PCR and RQ-PCR. No sample demonstrated NKX2-5 expression and only one patient showed TLX3 expression. This patient had an evolutive B-CLL, with complex chromosomal abnormalities, without involvement of the band 5q35. In conclusion, this study identifies a new translocation involving a NK-like homeobox gene in a mature B lymphoproliferative disorder, underscoring precedent identification of NKX2-5 activation in two human lymphoid leukemia cell lines. It also demonstrates that deregulation of homeobox encoding genes is not a specific feature of acute leukemic proliferations, but is also observed in chronic malignancies.