RED CELL DISTRIBUTION WIDTH (RDW) AS A MARKER OF DISEASE ACTIVITY IN PATIENTS WITH HAIRY CELL LEUKEMIA

Summary: Red cell distribution width (RDW) was examined in 18 patients with hairy cell leukemia (HCL) treated with 2-chlorodeoxyadenosine (2-CdA), in 5 patients treated with Interferon alpha (IFN-alpha) and in 9 patients subjected to splenectomy. Out of 18 patients treated with 2-CdA one patient was excluded of the study because of association of HCL with acquired sideroblastic anemia. In the remaining 17 patients the mean value of RDW before terapy was 18,8% (range 13,5% to 25,0%) and dropped after successful theapy after 6 to 12 months to the mean value of 13,6% (range 11,2% to 17,9%) and after 18 months to 13,4% (range 12,6% to 14,7%) (p = 0,00015 and p = 0,00049 respectivelly). The hemoglobin level increased from the mean value of 119 g/l (range 99g/l to 157 g/l) before therapy to the mean value of 145,9 g/l (range 127 g/l to 172 g/l) after 6 to 12 months and after 18 months to 147,8 g/l (range 132 g/l to 168 g/l) (p = 0,000017 and p = 0,00036 respectively). The same trend was observed in the group of patients treated with IFN-alfa. The RDW decreased from the mean value of 21,3% (range 18,8% to 28,7%) to the mean value of 15,3% (range 12,4% to 16,7%), (p = 0,031). The hemoglobin level increased in this group of patients from the mean value of 115 g/l (range 98 g/l to 127 g/l) to the mean value of 136 g/l (range 127 g/l to 146 g/l) (p=0,031). In 9 patients in complete hematologic remission 34 to 293 months after splenectomy the mean value of RDW was 13,9% (range 13,0% to 15,5%). Conclusion: Increased RDW in HCL is associated with active disease and is reversible after successful therapy. This phenomenon has not been reported in the lite-rature yet. Preliminary results show that the increase of RDW may be due to the dyserythropoiesis.


Introduction
Hairy cell leukemia (HCL) is a chronic lymphoproliferative disorder characterized by abnormal mononuclear cells of B lymphocyte origin infiltrating bone marrow and spleen. Patients often present with some combination of anemia, neutropenia, thrombocytopenia and splenomegaly (1,6,11). Anemia is very common and is observed in 75 to 80 percent of patients (3,9,12). The mechanism of anemia has been attributed to the splenic sequestration with a substantial increase in splenic volume (13) and to the reduced red-cell production due to a diffuse bone marrow infiltration by tumor cells and bone marrow fibrosis (4,15). Erythrokinetic studies revield that in addition to splenic sequestration and reduced red-cell production there is some reduction in red cell-survival (5). The anemia is usually normocytic or macrocytic with slight anisocytosis and poikilocytosis. The mean cell volume (MCV) is nearly always toward the upper limit of normal and often elevated (6). We have observed that red cell distribution width, which is indicative of anisocytosis, reflects in HCL the diseases activity (7).

Patients and Methods
The RDW values were assessed in 32 patients with HCL divided in three groups according to the therapy.
The diagnosis of HCL was based on the presence of morphologically characteristic cells in the peripheral blood and/or in the bone marrow, demonstration of the tartarat resistant acid phosphatase activity in the neoplastic cells (16), typical histologic pattern in bone marrow biopsies with infiltration by malignant cells characteristically surrounded by a rim of pale cytoplasm resulting in clearly separated nuclei (4). In all splenectomized patients the diagnosis was reconfirmed by the histologic finding in the spleen showing heavy infiltration of the red pulp by abnor-mal interdigitating mononuclear cells and the presence of blood-filled spaces lined by hairy cells, so called pseudosinuses (14).

Response criteria to the therapy
Complete hematologic remission (CR) required all of the following: 1) Complete absence of hairy cells in the peripheral blood. 2) Normalization of peripheral blood counts (hemoglobin level ≥120 g/l, white cell count ≥3x10 9 /l, absolute neutrophil count ≥1,5x10 9 /l, platelet count ≥100x10 9 /l, disappearance of retroperitoneal lymphadenopathy and hepatosplenomegaly by CT or US scan. Partial remission (PR) required 1) Failure of normalization in one of low peripheral blood counts. 2) Reduction grater than 50% of lymphadenopathy and/or hepatosplenomegaly. The Red Cell Width (RDW in %) was determined using Coulter JT3 or Coulter STKS. Normal value 11,5-14,5%.

Patients
Group 1 (see table 1): consisted of 18 patients treated by 2-chlorodeoxyadenosine (2-CdA). All patients received a single cycle of 2-CdA at a dose of 0,1 mg/kg/d by continous intravenous infusion for 7 days. Out of 18 patients there were 14 male and 4 female with an age range 39 to 82 years (median, 61,5 years). 11 patients were previously treated: 7 by splenectomy, 2 by splenectomy and IFN-alpha, 2 by repeated course of IFN-alpha because of relapse, in remaining 7 patients the 2-CdA therapy was the first treatment. All 18 patients responded to the therapy, in 16 complete hematologic remission was achieved, in one patient (M.W. No 11) the remission was partial because the platelet count increased after therapy only to 90x10 9 /l, but the hemoglobin level after therapy was within normal limits. Patient V.J. (No 15) reached the normal hemoglobin level after 18 months. Patient V.M. (No 6) was excluded from further assessement because of associated acquired sideroblastic anemia and will be considered separately.
Group 2 (see table 2) consisted of 5 patients treated by Interpheron-alpha. 3 patients were previously treated by splenectomy, 1 patient received the second course of IFNalpha because of relapse. Out of 5 patients there were 4 male and 1 female with an age range of 39 to 78 years (median, 58 years).
Group 3 (see table 3) consisted of 9 patients treated by splenectomy with the follow up period after splenectomy between 34 and 293 months (median, 193 months). Out of 9 patients there were 5 male and 4 female with an age range between 51 to 71 years (median, 60 years). Bone marrow biopsy revieled interstitial infiltration by hairy cells in all patients. All patients have been after splenectomy in complete hematologic remission.   In the first set of the data (17 and 14 subjects) when the result are compared after 6-12 months period and more than 18 months respectively it is again possible to reject the null hypothesis and accept the alternative one by means of Wilcoxon test. Values of RDW decreased significantly after the period of 6-12 month (p=0,00015) and after more than 18 months (p=0,00049).
In the case of Hb concentration it is possible to accept the null hypothesis as far as the normality of the differences is concerned and a paired t-test can be used. It allows to accept the alternative hypothesis and state that this variable increases x (p=0,000017 and p=0,00036 respectively).
It is not possible to verify the normality of the differences in the second set (data from 5 subjects) and Wilcoxon test allows to reject null hypothesis at 5% significance level. Therefore we can accept the alternative hypothesis which states that RDW is lower (p=0,031) and Hb concentration is higher after the treatment (p=0,031).
The hemoglobin level simultaneously increased from the mean value of 119g/l (range 99 g/l to 157 g/l) before therapy to the mean value of 145,9 g/l (range 127 g/l to 172 g/l) after 6 to 12 months and to 147,8 g/l (range 132 g/l to 168 g/l) after 18months (p=0,00049 and p=0,00036 respectively).
The hemoglobin level simultaneously increased from the main value of 115 g/l (range 98 g/l to 127 g/l) to the main value of 136 g/l (range 127 g/l to 146 g/l), (p= 0,031).
Group 3 (see table 3): In the group of 9 patients treated by splenectomy all patients were in complete hematologic remission. Mean value of RDW was 13,9%. The RDW was slightly above the upper limit of normal value in two patients (15,1% and 15,5%) despite normal levels of hemoglo-bin 135 g/l and 177 g/l respectively. The RDW values before splenectomy were not accessible. The examination before splenectomy was performed with an automated instrument not giving RDW value.

Discussion
Anisocytosis is an increase in the variability of erythrocyte size beyond that which is observed in a healthy subject. In automated instruments the calculated value of RDW reflects very accurately the degree of anisocytosis and makes the comparison of its changes possible. RDW has been found to be increased in iron deficiency anemia, megaloblastic anemias, congenital and acquired dyserythropoietic anemias (1,2,16,17). We have found that increased RDW in HCL was present in patients with an active disease and subsided after successful therapy with 2-CdA or IFN-alpha with a simultaneous rise of the hemoglobin levels. This phenomenon as far as we know has not been reported in the literature yet. Preliminary data show that the increased RDW may be due to the dyserythropoiesis (8). This finding is the subject of a further study and will be reported separately in more details.