Leukemia Research 43 (2016) 18–23
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
T-cell large granular lymphocyte proliferation in myelodysplastic syndromes: Clinicopathological features and prognostic significance Xiaohui Zhang a , Lubomir Sokol b , John M. Bennett c , Lynn C. Moscinski a , Alan List b , Ling Zhang a,∗ a
Department of Hematopathology and Laboratory Medicine, United States Department of Malignant Hematology, H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL, United States c Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, New York, United States b
a r t i c l e
i n f o
Article history: Received 1 October 2015 Received in revised form 5 February 2016 Accepted 14 February 2016 Available online 17 February 2016 Keywords: Myelodysplastic syndromes T-cell large granular lymphocyte proliferation Prognosis
a b s t r a c t Inflammatory and immune dysregulation are crucial in the initiation and development of myelodysplastic syndromes (MDS). It is noted that clonal T-cell large granular lymphocyte (T-LGL) proliferation associated with MDS is not uncommon. However, clinicopathological features, and prognostic and predictive value of presence of T-LGL proliferation in MDS patients is not very clear. This study compared 35 MDS patients with T-LGL proliferation with 36 MDS patients without T-LGL proliferation and summarized clinicopathologic features, including peripheral blood LGL cell counts, immunophenotype, T cell receptor gene rearrangement, bone marrow hematopoietic status, and adjuvant immunosuppressive therapy. The peripheral blood CD3+/CD57+ cell counts were significantly different (p < 0.01) between the two groups. Notably, on examination of the bone marrow, MDS patients with T-LGL proliferation showed more frequent hypocellularity and/or lineage hypoplasia, particularly erythroid hypoplasia. On survival analysis, no overall difference was noted between MDS patients with T-LGL proliferation and those without T-LGL proliferation, and between the patients who received therapy for LGL and those who did not receive adjuvant therapy for LGL in the same risk group. In conclusion, T-LGL proliferation present in MDS patients can be associated with bone marrow hypocellularity and lineage hypoplasia. Although immunosuppressive therapy to eliminate T-LGL cells is potentially beneficial to the MDS patients with associated T-LGL proliferation, there is no overall survival benefit to the patients who received such treatment. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematologic disorders of hematopoietic stem cell characterized by peripheral blood cytopenias due to ineffective hematopoiesis in bone marrow, often accompanied by a variable increase in myeloblasts (1–19%). Intrinsic genetic and epigenetic alterations and extrinsic abnormalities of bone marrow microenvironment are both critical in MDS evolution that may involve dysregulation of hematopoietic stem cell proliferation, differentiation and apoptosis. There is increasing evidence suggesting that
Abbreviations: MDS, myelodysplastic syndromes; T-LGL, T-cell large granular lymphocyte; IPSS, International Prognostic Scoring System. ∗ Corresponding author at: Department of Hematopathology and Laboratory Medicine, H. Lee Moffitt Cancer Center, MCC-LAB, 12902 USF Magnolia Dr., Tampa, FL 33612, United States. E-mail address: ling.zhang@moffitt.org (L. Zhang). http://dx.doi.org/10.1016/j.leukres.2016.02.006 0145-2126/© 2016 Elsevier Ltd. All rights reserved.
altered body immunity affects bone marrow microenvironment and contributes to the pathogenesis of MDS [1]. It has been noted that some MDS patients have expansion of cytotoxic (CD8+) T cells. The expanded and dysregulated CD8 + cells can be detrimental to hematopoiesis[2,3]. In accordance with these observations, immunosuppressive treatment with cyclosporine or antithymocyte globulin (ATG) significantly improves anemia and neutropenia in a subgroup of MDS patients [4,5]. T-cell large granular lymphocytes (T-LGL) are post-thymic antigen-primed, cytotoxic CD8+ lymphocytes, morphologically characterized by moderate amounts of cytoplasm with azurophilic, cytotoxic granulation. T-LGL comprise 10–15% of total peripheral blood mononuclear cells in normal adults [6]. The proliferation of TLGL can be observed in a spectrum of clinical settings, and can range from mild lymphocytosis to asymptomatic clonal T-LGL proliferation or even clinically overt leukemic process [7]. Per literature, clonal LGL cell expansion was detected in as many as 50% of MDS bone marrows [8,9]. Up to date, there are limited data on the clinicopathological features of T-LGL cell proliferation in MDS patients,
X. Zhang et al. / Leukemia Research 43 (2016) 18–23
and mechanism of the interference between T-LGL cells and the bone marrow hematopoietic and stromal niches. The main objective of present study is to examine the T-LGL proliferation in MDS patients and to explore clinical and pathological features of MDS patients with T-LGL proliferation, comparing them with control group with MDS and absence of T-LGL proliferation. 2. Patients and methods 2.1. Patients and controls We searched Moffitt Cancer Center hematopathology database and identified 71 patients with a diagnosis of MDS between 1/2006 and 12/2011. Flow cytometry on peripheral blood mononuclear cells with LGL monoclonal antibody panel was available from these patients. Clinical and pathologic data from these patients were retrieved and analyzed. Peripheral blood samples were obtained at routine patient visits. The patients who had a history of autoimmune disorders and presented with sustained cytopenia without sufficient evidence to diagnose MDS were excluded. Sampling and handling were in accordance with the guidelines of the Institutional Review Board at University of South Florida College of Medicine. In parallel, peripheral blood samples from 32 control healthy subjects were analyzed following the same protocol. Flow cytometry data obtained from 17 cases with T-LGL leukemia without simultaneous MDS were also analyzed. 2.2. T cell receptor gene rearrangement analysis Total cellular DNA was extracted and PCR amplification performed in five multiplex PCR tubes with ASR Biomed-2 primers (Invivoscribe Technologies, San Diego, California, USA) targeting TCR V, D, J, V␥, and J␥ regions [10]. The products were separated and detected by capillary gel electrophoresis on the ABI Prism 3130xl genetic analyser (Applied Biosystems, Warrington, UK). 2.3. Flow cytometry analysis Peripheral blood cell counts were determined by an XE-2100 automated hemocytometer (Sysmex, Hyogo, Japan). Aliquots of 50 uL whole blood in sodium heparin were incubated in the dark for 15 min at room temperature with combinations of four monoclonal antibodies conjugated to fluorescein isothiocyanate (FITC), phycoerythrin (PE), peridinin chlorophyll protein-Cy 5.5 (PerCP) or allophycocyanin (APC). Red blood cells were lysed with BD FACS lysing solution (BD Biosciences, San Jose, California) and nucleated cells were resuspended in phosphate buffered saline containing 2% paraformaldehyde. At least 200,000 events were acquired on a FACSCalibur flow cytometer with CellQuest software (Becton Dickinson, San Jose, CA). One panel using the four-color antibody combinations was available at MCC: A ‘T cell lymphoma panel’ was available since 2002 and consisted of an isotype control tube plus 8 tubes (CD14/CD7/CD5/CD45, CD3/CD4/CD5/CD8, CD25/HLADR/CD5/CD8, CD56/CD16/CD3/CD5, CD57/CD3/CD5/CD8, CD38/CD62L/CD5/CD19, CD44/CD11b/CD5/CD11c, TCR␣/TCR␥␦/CD3/CD5). All antibodies were obtained from BD Biosciences. Lymphoid populations were identified and categorized. Isotype controls were performed in all cases. 2.4. Statistical analysis All statistic calculations were performed using IBM SPSS Statistics 21 for Windows (Armonk, NY). Student’s t-test, Chi square test or Fisher test was performed to evaluate the observed differences among the groups. Survival analysis was computed using the
19
Table 1 Clinical features of the 71 patients diagnosed with MDS with or without concurrent LGL.
Patients (n) Age (median, range) Gender Diagnosis (n) RCUD, RA, RARS RCMD RAEB 5q-syndrome MDS, unclassifiable IPSS (n) Low Intermediate-1 Intermediate-2 High Cytogenetics (n) Good Intermediate Poor
MDS with LGL
MDS without LGL
35 71 (42–83) M = 29, F = 6
36 68.3 (31–85) M = 27, F = 9
3 19 9 1 3
6 13 13 0 4
9 19 3 4
16 9 4 7
23 10 2
26 8 2
Kaplan-Meier method and differences in survival curves were analyzed with the Breslow test. A p value less than 0.05 was considered statistically significant. 3. Results 3.1. Clinical features Diagnosis of MDS was based on peripheral blood complete blood count (CBC), bone marrow aspirate and biopsy examination, and was classified according to the World Health Organization (WHO) classification [11]. The 71 cases of MDS had a median age of 70 years, ranging from 31 to 85 years, and the male to female ratio was 3.5. According to the International Prognostic Scoring System (IPSS) [12], the patients were categorized into different risk groups (low, intermediate I, intermediate II and high risk, Table 1) and cytogenetics is classified into good, intermediate and poor karyotype. The general clinical features of the patients diagnosed with MDS, with or without concurrent LGL, are summarized in Table 1. The distribution of different IPSS risk scores between the two groups showed no statistically significant difference (p = 0.087, Fisher test). When compared the peripheral blood values including hemoglobin, white blood cell count, and platelet count, the two groups appeared to be similar. Overall, mild anemia was present 45.8% vs. 50% of the cases, and moderate anemia was present in 54.2% vs. 50% of the cases (p > 0.01). Leukopenia was present in 97% vs. 94% of the cases; thrombocytopenia was present in 54% vs. 67% of the cases. 3.2. Features of T-LGL cell proliferation in MDS patients Flow cytometry was performed on the peripheral blood of 32 control subjects and 71 MDS patients. Generally, in control subjects there were a small subset of cell with co-expression of CD3 and CD57 and aberrant loss or gain of other markers. These LGLs represent a low percentage of the overall white blood cells (2.5 ± 2.3%), with low absolute cell counts (173 ± 141/L). In contrast, in MDS patients, a distinct population of T-LGLs with co-expression of CD3 and CD57, along with CD8, HLA-DR, and frequently with dim expression or partial loss of CD5 and/or CD7 was observed. A clonal T-LGL proliferation was diagnosed based on identification of an overt aberrant T cell population with characteristic immunophenotype of CD3+ /CD57+ /CD7dim+ /CD5dim+ /CD8+ , often coexpressing CD11b, CD11c, CD16 and CD56, and clonal T cell receptor beta and/or gamma gene rearrangement. Out of 71 MDS patients, T-LGL proliferation was diagnosed in 35 cases. Of them, 30 cases showed
20
X. Zhang et al. / Leukemia Research 43 (2016) 18–23
clonal T cell receptor gene beta or/and gamma rearrangement by PCR. In five remaining cases (17%), negative results were reported, but flow cytometry detected distinct aberrant T cell population with the phenotype suggestive of T-LGL proliferation. As a result, the TCR or/and TCR␥ gene rearrangements were positive in 30 of the 35 MDS cases with T-LGL proliferation (85.7%), in contrast to 10 of the 31 (32.2%) tested cases of MDS without T-LGL proliferation. Percentage and absolute counts of CD3+/CD57+ cells were not used as one of the criteria, as the required numbers for diagnosis is debatable [13,14]. The immunophenotype of the T-LGL was typically CD3+ /CD57+ /CD7dim+ /CD5dim+ /CD8+ with variable CD11b, CD11c, CD16, CD56 and HLA-DR (Fig. 1). A frequent variant in these MDS patients was CD11b− , CD11c− , CD16+/− , CD56+/− , HLADR− and CD62L+ . When compared with MDS without T-LGL, the CD3+ /CD57+ and CD3+ /CD8+ /CD57+ absolute cell counts were increased (0.245 ± 0.153 × 109 /L vs. 0.074 ± 0.072 × 109 /L, p < 0.001) in the patients having MDS and T-LGL proliferation, although lymphocyte counts showed no significant difference (1.147 ± 0.624 × 109 /L vs. 1.188 ± 0.683 × 109 /L, p = 0.74). Notably, the T-LGL cell counts in MDS lower than 0.3 × 109 /L, a median normal range for control group, were not unusual. MDS patients usually presented with cytopenias and their absolute lymphocyte counts were already significantly lower than control subjects, which could cause the difference on T-LGL cell counts between MDS patients without LGL proliferation and the control subjects. The peripheral blood lymphocyte and CD3+ /CD57+ and CD3+ /CD8+ /CD57+ T-LGL cell counts are summarized in Table 2. When compared with 17 cases of T-LGL leukemia with no morphologic dysplasia, clonal TCR or/and TCR␥ gene rearrangements were present in both MDS with LGL proliferation (30/35, 85.7%) and classic T-LGL leukemia (17/17, 100%). The CD3+/CD57+ T-LGL cell counts were significantly higher in T-LGL leukemia patients than in MDS patients with concurrent T-LGL proliferation (0.245 ± 0.153 × 109 /L vs. 0.719 ± 0.720 × 109 /L, p < 0.001). The CD3+/CD57+ T-LGL counts were greater than 0.3 × 109 /L in 10 of 35 MDS patients with LGL proliferation, 0 of 36 MDS only patients and 12 of 17 T-LGL leukemia only patients. 3.3. Bone marrow status of MDS patients with T-LGL proliferation Since the presence of T-LGL proliferation may impair bone marrow hematopoiesis, we examined if there are bone marrow cytomorphological differences between MDS with T-LGL proliferation and MDS without T-LGL proliferation. Concurrent bone marrow biopsies from each patient at the time of flow cytometric analysis were reviewed. All the bone marrows were obtained at diagnosis or between chemotherapy cycles. Bone marrow cellularity, lineage hypoplasia (M:E >5:1 or 6 months, without other identifiable etiologies, is required for the diagnosis of T-LGL leukemia. Reactive T-LGL lymphocytosis is generally less than 0.5 × 109 /L [11]. However, this is still a subject of discussion [13,14], since in clinical practice many patients with symptomatic T-LGL leukemia present with
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
T-cell large granular lymphocyte proliferation in myelodysplastic syndromes: Clinicopathological features and prognostic significance Xiaohui Zhang a , Lubomir Sokol b , John M. Bennett c , Lynn C. Moscinski a , Alan List b , Ling Zhang a,∗ a
Department of Hematopathology and Laboratory Medicine, United States Department of Malignant Hematology, H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL, United States c Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, New York, United States b
a r t i c l e
i n f o
Article history: Received 1 October 2015 Received in revised form 5 February 2016 Accepted 14 February 2016 Available online 17 February 2016 Keywords: Myelodysplastic syndromes T-cell large granular lymphocyte proliferation Prognosis
a b s t r a c t Inflammatory and immune dysregulation are crucial in the initiation and development of myelodysplastic syndromes (MDS). It is noted that clonal T-cell large granular lymphocyte (T-LGL) proliferation associated with MDS is not uncommon. However, clinicopathological features, and prognostic and predictive value of presence of T-LGL proliferation in MDS patients is not very clear. This study compared 35 MDS patients with T-LGL proliferation with 36 MDS patients without T-LGL proliferation and summarized clinicopathologic features, including peripheral blood LGL cell counts, immunophenotype, T cell receptor gene rearrangement, bone marrow hematopoietic status, and adjuvant immunosuppressive therapy. The peripheral blood CD3+/CD57+ cell counts were significantly different (p < 0.01) between the two groups. Notably, on examination of the bone marrow, MDS patients with T-LGL proliferation showed more frequent hypocellularity and/or lineage hypoplasia, particularly erythroid hypoplasia. On survival analysis, no overall difference was noted between MDS patients with T-LGL proliferation and those without T-LGL proliferation, and between the patients who received therapy for LGL and those who did not receive adjuvant therapy for LGL in the same risk group. In conclusion, T-LGL proliferation present in MDS patients can be associated with bone marrow hypocellularity and lineage hypoplasia. Although immunosuppressive therapy to eliminate T-LGL cells is potentially beneficial to the MDS patients with associated T-LGL proliferation, there is no overall survival benefit to the patients who received such treatment. © 2016 Elsevier Ltd. All rights reserved.
1. Introduction Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematologic disorders of hematopoietic stem cell characterized by peripheral blood cytopenias due to ineffective hematopoiesis in bone marrow, often accompanied by a variable increase in myeloblasts (1–19%). Intrinsic genetic and epigenetic alterations and extrinsic abnormalities of bone marrow microenvironment are both critical in MDS evolution that may involve dysregulation of hematopoietic stem cell proliferation, differentiation and apoptosis. There is increasing evidence suggesting that
Abbreviations: MDS, myelodysplastic syndromes; T-LGL, T-cell large granular lymphocyte; IPSS, International Prognostic Scoring System. ∗ Corresponding author at: Department of Hematopathology and Laboratory Medicine, H. Lee Moffitt Cancer Center, MCC-LAB, 12902 USF Magnolia Dr., Tampa, FL 33612, United States. E-mail address: ling.zhang@moffitt.org (L. Zhang). http://dx.doi.org/10.1016/j.leukres.2016.02.006 0145-2126/© 2016 Elsevier Ltd. All rights reserved.
altered body immunity affects bone marrow microenvironment and contributes to the pathogenesis of MDS [1]. It has been noted that some MDS patients have expansion of cytotoxic (CD8+) T cells. The expanded and dysregulated CD8 + cells can be detrimental to hematopoiesis[2,3]. In accordance with these observations, immunosuppressive treatment with cyclosporine or antithymocyte globulin (ATG) significantly improves anemia and neutropenia in a subgroup of MDS patients [4,5]. T-cell large granular lymphocytes (T-LGL) are post-thymic antigen-primed, cytotoxic CD8+ lymphocytes, morphologically characterized by moderate amounts of cytoplasm with azurophilic, cytotoxic granulation. T-LGL comprise 10–15% of total peripheral blood mononuclear cells in normal adults [6]. The proliferation of TLGL can be observed in a spectrum of clinical settings, and can range from mild lymphocytosis to asymptomatic clonal T-LGL proliferation or even clinically overt leukemic process [7]. Per literature, clonal LGL cell expansion was detected in as many as 50% of MDS bone marrows [8,9]. Up to date, there are limited data on the clinicopathological features of T-LGL cell proliferation in MDS patients,
X. Zhang et al. / Leukemia Research 43 (2016) 18–23
and mechanism of the interference between T-LGL cells and the bone marrow hematopoietic and stromal niches. The main objective of present study is to examine the T-LGL proliferation in MDS patients and to explore clinical and pathological features of MDS patients with T-LGL proliferation, comparing them with control group with MDS and absence of T-LGL proliferation. 2. Patients and methods 2.1. Patients and controls We searched Moffitt Cancer Center hematopathology database and identified 71 patients with a diagnosis of MDS between 1/2006 and 12/2011. Flow cytometry on peripheral blood mononuclear cells with LGL monoclonal antibody panel was available from these patients. Clinical and pathologic data from these patients were retrieved and analyzed. Peripheral blood samples were obtained at routine patient visits. The patients who had a history of autoimmune disorders and presented with sustained cytopenia without sufficient evidence to diagnose MDS were excluded. Sampling and handling were in accordance with the guidelines of the Institutional Review Board at University of South Florida College of Medicine. In parallel, peripheral blood samples from 32 control healthy subjects were analyzed following the same protocol. Flow cytometry data obtained from 17 cases with T-LGL leukemia without simultaneous MDS were also analyzed. 2.2. T cell receptor gene rearrangement analysis Total cellular DNA was extracted and PCR amplification performed in five multiplex PCR tubes with ASR Biomed-2 primers (Invivoscribe Technologies, San Diego, California, USA) targeting TCR V, D, J, V␥, and J␥ regions [10]. The products were separated and detected by capillary gel electrophoresis on the ABI Prism 3130xl genetic analyser (Applied Biosystems, Warrington, UK). 2.3. Flow cytometry analysis Peripheral blood cell counts were determined by an XE-2100 automated hemocytometer (Sysmex, Hyogo, Japan). Aliquots of 50 uL whole blood in sodium heparin were incubated in the dark for 15 min at room temperature with combinations of four monoclonal antibodies conjugated to fluorescein isothiocyanate (FITC), phycoerythrin (PE), peridinin chlorophyll protein-Cy 5.5 (PerCP) or allophycocyanin (APC). Red blood cells were lysed with BD FACS lysing solution (BD Biosciences, San Jose, California) and nucleated cells were resuspended in phosphate buffered saline containing 2% paraformaldehyde. At least 200,000 events were acquired on a FACSCalibur flow cytometer with CellQuest software (Becton Dickinson, San Jose, CA). One panel using the four-color antibody combinations was available at MCC: A ‘T cell lymphoma panel’ was available since 2002 and consisted of an isotype control tube plus 8 tubes (CD14/CD7/CD5/CD45, CD3/CD4/CD5/CD8, CD25/HLADR/CD5/CD8, CD56/CD16/CD3/CD5, CD57/CD3/CD5/CD8, CD38/CD62L/CD5/CD19, CD44/CD11b/CD5/CD11c, TCR␣/TCR␥␦/CD3/CD5). All antibodies were obtained from BD Biosciences. Lymphoid populations were identified and categorized. Isotype controls were performed in all cases. 2.4. Statistical analysis All statistic calculations were performed using IBM SPSS Statistics 21 for Windows (Armonk, NY). Student’s t-test, Chi square test or Fisher test was performed to evaluate the observed differences among the groups. Survival analysis was computed using the
19
Table 1 Clinical features of the 71 patients diagnosed with MDS with or without concurrent LGL.
Patients (n) Age (median, range) Gender Diagnosis (n) RCUD, RA, RARS RCMD RAEB 5q-syndrome MDS, unclassifiable IPSS (n) Low Intermediate-1 Intermediate-2 High Cytogenetics (n) Good Intermediate Poor
MDS with LGL
MDS without LGL
35 71 (42–83) M = 29, F = 6
36 68.3 (31–85) M = 27, F = 9
3 19 9 1 3
6 13 13 0 4
9 19 3 4
16 9 4 7
23 10 2
26 8 2
Kaplan-Meier method and differences in survival curves were analyzed with the Breslow test. A p value less than 0.05 was considered statistically significant. 3. Results 3.1. Clinical features Diagnosis of MDS was based on peripheral blood complete blood count (CBC), bone marrow aspirate and biopsy examination, and was classified according to the World Health Organization (WHO) classification [11]. The 71 cases of MDS had a median age of 70 years, ranging from 31 to 85 years, and the male to female ratio was 3.5. According to the International Prognostic Scoring System (IPSS) [12], the patients were categorized into different risk groups (low, intermediate I, intermediate II and high risk, Table 1) and cytogenetics is classified into good, intermediate and poor karyotype. The general clinical features of the patients diagnosed with MDS, with or without concurrent LGL, are summarized in Table 1. The distribution of different IPSS risk scores between the two groups showed no statistically significant difference (p = 0.087, Fisher test). When compared the peripheral blood values including hemoglobin, white blood cell count, and platelet count, the two groups appeared to be similar. Overall, mild anemia was present 45.8% vs. 50% of the cases, and moderate anemia was present in 54.2% vs. 50% of the cases (p > 0.01). Leukopenia was present in 97% vs. 94% of the cases; thrombocytopenia was present in 54% vs. 67% of the cases. 3.2. Features of T-LGL cell proliferation in MDS patients Flow cytometry was performed on the peripheral blood of 32 control subjects and 71 MDS patients. Generally, in control subjects there were a small subset of cell with co-expression of CD3 and CD57 and aberrant loss or gain of other markers. These LGLs represent a low percentage of the overall white blood cells (2.5 ± 2.3%), with low absolute cell counts (173 ± 141/L). In contrast, in MDS patients, a distinct population of T-LGLs with co-expression of CD3 and CD57, along with CD8, HLA-DR, and frequently with dim expression or partial loss of CD5 and/or CD7 was observed. A clonal T-LGL proliferation was diagnosed based on identification of an overt aberrant T cell population with characteristic immunophenotype of CD3+ /CD57+ /CD7dim+ /CD5dim+ /CD8+ , often coexpressing CD11b, CD11c, CD16 and CD56, and clonal T cell receptor beta and/or gamma gene rearrangement. Out of 71 MDS patients, T-LGL proliferation was diagnosed in 35 cases. Of them, 30 cases showed
20
X. Zhang et al. / Leukemia Research 43 (2016) 18–23
clonal T cell receptor gene beta or/and gamma rearrangement by PCR. In five remaining cases (17%), negative results were reported, but flow cytometry detected distinct aberrant T cell population with the phenotype suggestive of T-LGL proliferation. As a result, the TCR or/and TCR␥ gene rearrangements were positive in 30 of the 35 MDS cases with T-LGL proliferation (85.7%), in contrast to 10 of the 31 (32.2%) tested cases of MDS without T-LGL proliferation. Percentage and absolute counts of CD3+/CD57+ cells were not used as one of the criteria, as the required numbers for diagnosis is debatable [13,14]. The immunophenotype of the T-LGL was typically CD3+ /CD57+ /CD7dim+ /CD5dim+ /CD8+ with variable CD11b, CD11c, CD16, CD56 and HLA-DR (Fig. 1). A frequent variant in these MDS patients was CD11b− , CD11c− , CD16+/− , CD56+/− , HLADR− and CD62L+ . When compared with MDS without T-LGL, the CD3+ /CD57+ and CD3+ /CD8+ /CD57+ absolute cell counts were increased (0.245 ± 0.153 × 109 /L vs. 0.074 ± 0.072 × 109 /L, p < 0.001) in the patients having MDS and T-LGL proliferation, although lymphocyte counts showed no significant difference (1.147 ± 0.624 × 109 /L vs. 1.188 ± 0.683 × 109 /L, p = 0.74). Notably, the T-LGL cell counts in MDS lower than 0.3 × 109 /L, a median normal range for control group, were not unusual. MDS patients usually presented with cytopenias and their absolute lymphocyte counts were already significantly lower than control subjects, which could cause the difference on T-LGL cell counts between MDS patients without LGL proliferation and the control subjects. The peripheral blood lymphocyte and CD3+ /CD57+ and CD3+ /CD8+ /CD57+ T-LGL cell counts are summarized in Table 2. When compared with 17 cases of T-LGL leukemia with no morphologic dysplasia, clonal TCR or/and TCR␥ gene rearrangements were present in both MDS with LGL proliferation (30/35, 85.7%) and classic T-LGL leukemia (17/17, 100%). The CD3+/CD57+ T-LGL cell counts were significantly higher in T-LGL leukemia patients than in MDS patients with concurrent T-LGL proliferation (0.245 ± 0.153 × 109 /L vs. 0.719 ± 0.720 × 109 /L, p < 0.001). The CD3+/CD57+ T-LGL counts were greater than 0.3 × 109 /L in 10 of 35 MDS patients with LGL proliferation, 0 of 36 MDS only patients and 12 of 17 T-LGL leukemia only patients. 3.3. Bone marrow status of MDS patients with T-LGL proliferation Since the presence of T-LGL proliferation may impair bone marrow hematopoiesis, we examined if there are bone marrow cytomorphological differences between MDS with T-LGL proliferation and MDS without T-LGL proliferation. Concurrent bone marrow biopsies from each patient at the time of flow cytometric analysis were reviewed. All the bone marrows were obtained at diagnosis or between chemotherapy cycles. Bone marrow cellularity, lineage hypoplasia (M:E >5:1 or 6 months, without other identifiable etiologies, is required for the diagnosis of T-LGL leukemia. Reactive T-LGL lymphocytosis is generally less than 0.5 × 109 /L [11]. However, this is still a subject of discussion [13,14], since in clinical practice many patients with symptomatic T-LGL leukemia present with
