Abstract
To analyze the effect of a splenectomy on pancytopenia after a peripheral blood stem cell transplantation (PBSCT) in chronic myeloid leukemia (CML) patients. Three CML patients diagnosed with splenomegaly in our department from 2002 to 2006 received a splenectomy after PBSCT. Patient 1, a 42-year-old male in chronic phase (CP), received an HLA-identical sibling allo-PBSCT. Patient 2, a 28-year-old female in aggressive phase (AP), received a PBSCT from her twin. Patient 3, a 26-year-old male in chronic phase (CP), received a PBSCT from an unrelated donor. The conditioning regimen included busulfan and cyclophosphamide (BU/CY2). Patients 1, 2, and 3 received splenectomies on days 168, 51, and 114, respectively. Bcr-abl transcripts were detected using the polymerase chain reaction (PCR). Chimerism was documented by PCR amplification of a variable number of tandem repeat (VNTR) polymorphrisms. Neither metrisone (2 mg/kg/day for 7 days), mycophenolic acid (MMF) (0.5 g twice daily for 1 month), high-dose γ-globulin (0.4 g/kg/day for 5 days), granulocyte colony-stimulating factor (G-CSF) therapy, nor erythropoietin (EPO) therapy had produced post-PBSCT hematopoiesis recovery. Patients 1 and 2 had 5% Ph-positive chromosomes while patient 3 exhibited graft-versus-host disease (GVHD). After receiving the splenectomy, all three had a rapid hematopoeitic recovery with no evidence of Ph-positive chromosomes, and patients 1 and 3 showed complete donor chimerism (CDC). Patient 1 developed chronic GVHD (cGVHD) on day 210 postsplenectomy that was treated successfully with prednisone. Patient 2 had acute GVHD on day 55 that was treated successfully with dexamethasone (10 mg), administered intravenously once a day for 3 days with good clinical response. To date, patients 1, 2, and 3 have survived postprocedure for 85, 49, and 25 months, respectively. Splenectomy is an effective option for the patients who have pancytopenia after PBSCT and the patients recovered a good graft function after splenectomy without procedure-related complication and with long-time survival. GVHD can develop in both allo-PBSCT and syngeneic PBSCT. A splenectomy after PBSCT may increase the risk of GVHD, enhance the effect of graft versus leukemia (GVL), promote CDC formation.
Keywords
Introduction
Hematopoietic stem cell transplantation (HSCT) is an established treatment for patients with several disorders including life-threatening hematological diseases, immune deficiencies, and genetic metabolic disorders (1,8). Despite advances in prevention and posttransplant immunosuppressive strategies, graft-versus-host disease (GVHD) continues to be a major life-threatening complication after allogeneic bone marrow transplantation (11,13). Many factors influence GVHD, such as the duration of time between diagnosis and transplantation, disease status at transplantation, conditioning regimen, marrow cell doses, and the donor–recipient ABO blood match (6).
Recently, many studies have assessed the role of the spleen in peripheral blood stem cell transplantation (PBSCT). A splenectomy performed before an allogenic PBSCT (allo-PBSCT) for the treatment of chronic myelogenous leukemia (CML) has been controversial. Banaji et al. have reported a faster neutrophil and platelet recovery and decreased platelet transfusion requirements when a splenectomy is performed prior to transplantation (2). In addition, Michallet et al. concluded that a splenectomy was the most important factor in predicting GVHD severity, since one possible function of the spleen is to filter activated T lymphocytes arising from the transplant (14). One study suggested that a splenomegaly is a poor prognostic feature in patients with CML undergoing a related donor marrow transplant (7). In contrast, two studies have reported that a splenectomy performed prior to an allo-PBSCT had no influence on the occurrence of GVHD (3,5). The effect of a splenectomy on graft function performed after a PBSCT has rarely been reported. Richard et al. reported that four patients diagnosed with CML that had poor graft function after receiving a non-T-cell-depleted bone marrow transplantation from an HLA-compatible sibling donor were successfully treated by a splenectomy (16). The study conclusions showed that CML patients with poor graft function could be successfully treated with a splenectomy. However, the effect of a splenectomy on GVHD performed after a PBSCT, analysis of complete donor chimerism (CDC) formation, and disease relapse have not been reported.
In this study, we analyzed the effect of a splenectomy performed after an allo-PBSCT and syngeneic PBSCT on GVHD and the formation of CDC in three CML patients from our clinical department between 2002 and 2006.
Materials and Methods
Subjects
Patient 1 was a 42-year-old man who presented with left abdominal bump in 2002. He had no relevant past history. Clinically, his spleen was enlarged 17 cm below the left costal margin. Full blood examination revealed hemoglobin 112 g/L, a white blood cell (WBC) count of 213 × 109/L, and a platelet count of 543 × 109/L. Bone marrow examination morphologically resembled chronic phase CML. Cellularity was markedly increased with an M/E ratio of 18:1. The differential was as follows: neutrophils 83%, eosinophils 4%, metamyelocytes 8%, myelocytes 12%, promyelocytes 4%, myeloblasts 1%, basophils 2%, lymphocytes 1%, and erythroblasts 5%. Bone marrow cytogenetics revealed Ph-positive cells in all metaphases analyzed. The patient received hydrea and interferon (IFN) therapy, which effectively controlled the WBC count prior to PBSCT.
Patient 2 was a 28-year-old woman who presented with fatigue in 2003. Her spleen was enlarged 4 cm below the left costal margin. Full blood examination revealed hemoglobin 108 g/L, a WBC count of 67.7 × 109/L, and a platelet count of 539 × 109/L. Bone marrow examination morphologically resembled chronic phase CML. After 2 years, bone marrow examination showed that cellularity was markedly increased with an M/E ratio of 19:1. The differential was as follows: neutrophils 62%, eosinophils 6%, metamyelocytes 8%, myelocytes 12%, promyelocytes 14%, myeloblasts 6%, basophils 6%, lymphocytes 1%, and erythroblasts 5%. Bone marrow examination morphologically resembled accelerated phase CML. Bone marrow cytogenetics revealed Ph-positive cells in all metaphases analyzed. The patient received hydrea and IFN therapy, which effectively controlled the WBC count prior to PBSCT.
Patient 3 was a 26-year-old man who presented with fatigue in 2006. Clinically, his spleen was enlarged 12 cm below the left costal margin. Full blood examination revealed hemoglobin 110 g/L, a WBC count of 380.4 × 109/L, and a platelet count of 587 × 109/L. Bone marrow examination morphologically resembled chronic phase CML. Cellularity was markedly increased with an M/E ratio of 244:1. The differential was as follows: neutrophils 92%, eosinophils 2%, metamyelocytes 12%, myelocytes 9%, promyelocytes 1%, myeloblasts 0%, basophils 3%, lymphocytes 2%, and erythroblasts 0.4%. Bone marrow cytogenetics revealed Ph-positive cells in all metaphases analyzed. The patient received hydrea and IFN therapy, which effectively controlled the WBC count prior to PBSCT.
Patient 1 received an HLA-identical sibling allo-PBSCT. Patient 2 received a PBSCT from her twin. Patient 3 received a PBSCT from an unrelated donor. The interval from diagnosis to PBSCT ranged from 4 to 24 months.
Three patients diagnosed with splenomegaly and CML were used in this study. The busulfan and cyclophosphamide (BU-CY2) conditioning regimen consisted of busulfan (1 mg/kg) administered every 6 h for 16 days and cyclophosphamide (CTX) (60 mg/kg) administered once a day for 2 days. Each donor received granulocyte colony-stimulating factor (G-CSF) (300 μg) once a day for 6 days for mobilization. A median of 12 L (3 times the total blood volume) of donor blood was processed and collected. Each recipient received 8 × 106 CD34+ cells/kg of body weight. All patients and their donors gave written informed consent, and this study was approved by Institutional Review Board and the Ethics Committee at Southeast University.
GVHD Prophylaxis
Cyclosporin A (CsA) and methotrexate (MTX) were administered as a GVHD prophylaxis treatment. The regimen consisted of 3 mg/kg of intravenous CsA with a target blood level of 200–400 ng/L and a short course treatment of MTX (15 mg on day 1 and 10 mg on days 3, 6, and 11, respectively). A full dose of CsA was administered through day 50. If GVHD was successfully treated at that point, the dose was gradually reduced, and ended on the sixth month. Anti-thymocyte globulin (ATG) was administered for 3 days at a dose of 1000 mg for the unrelated donor of patient 3. Chronic GVHD (cGVHD) was diagnosed if the disease occurred in patients that survived for more than 100 days posttransplantation. The diagnosis of acute and chronic GVHD was based on classic clinical and biological criteria and was confirmed by histological examination whenever possible.
Cytomeglovirus and Hepatic Vein Occlusion Disease Prophylaxis
Intravenous ganciclovir (5 mg/kg) was administered every day from 8 days prior to the procedure to 5 days after the procedure. Cytomegalovirus DNA (CMV-DNA) was monitored by PCR once a week. Prostaglandin E1 (PGE1) (20 μg) was administered prophylactically and intravenously once a day for 2 days for hepatic vein occlusion disease (VOD) starting 8 days prior to the procedure and continuing 28 days postprocedure.
Engraftment
The neutrophil engraftment start point was defined as the first day of 3 consecutive days when the absolute neutrophil count (ANC) exceeded 0.5 × 109/L. The platelet engraftment start point was defined as the first day of 5 consecutive days when the platelet count exceeded 50 × 109/L independently without a platelet transfusion. Engraftment was evaluated by blood cell counts, bone marrow (BM) examination, and cytogenetic and molecular analysis.
Supportive Care
Patients were cared for in single rooms equipped with a high-efficiency particle air filtration system until reaching neutrophil recovery. Trimethoprim-sulfamethoxazole was administered immediately in cases of neutropenic fever. Itraconazole was given for suspected fungal infections. Patients were transfused with irradiated blood products to maintain their hemoglobin level above 8.0 g/dl and platelet count above 20 × 109/L. Daily G-CSF was administered intravenously starting on the fifth day after the allo-PBSCT and continued until the ANC was greater than 0.5 × 109/L for 3 consecutive days.
Results
Clinical details of the three patients included in this study are provided in Table 1.
Clinical Details of Three Patients Treated With PBSCT
Patient 1
Engraftment
Patient 1 received a sustained myeloid and platelet engraftment on day 15 and had a WBC count of 1.2 × 109/L, an ANC of 0.9 × 109/L, and a platelet count greater than 20 × 109/L at day 46. The hemoglobin level increased to 86 g/L after transplantation. Bone marrow examination 1 month after transplantation revealed obvious proliferation with a granulocytic and erythrocytic percentage of 62% and 26.4%, respectively. No increase in erythroblasts was detected. Cytogenetic studies on bone marrow specimens showed that the Ph-positive cells decreased to 5%. Chromosome analysis of bone marrow cells each month showed mixed donor chimerism.
Pancytopenia
After day 90, the patient's hemoglobin decreased gradually to a level of 60 g/L with the platelet count and WBC count being 19 × 109/L and 1.8 × 109/L, respectively (Fig. 1). Metrisone (2 mg/kg) was administered each day for 7 days without an effect. Mycophenolic acid (MMF) (0.5 g) was administered twice each day for 1 month without any indication of hematopoiesis recovery. Subsequently, a high-dose of γ-globulin (0.4 g/kg) was administered once a day for 5 days with no effect. G-CSF (150 μg) was given once per day subcutaneously and erythropoietin (EPO) (10,000 units) was administered three times a week without clear benefits. The patient exhibited an increase in WBC count without any improvement in hemoglobin level or platelet count.
Hematological parameters after allo-PBSCT in patient 1.
Splenectomy and Hematopoietic Recovery
An abdominal ultrasound of patient 1 showed an enlarged spleen (22 × 16 × 6 cm) and the patient received a splenectomy on day 168. Pathological analysis of the spleen revealed hypersplenism. Prophylactic platelet transfusion was administered before the splenectomy. Hematological recovery occurred quickly after the splenectomy and a level of hemoglobin greater than 10 g/L was reached within 24 h of the procedure. Pathological analysis of the resected spleen confirmed hypersplenism. Histological examination of the spleen showed splenic white pulp disappear, trabecular thinning, medullary cord widened, sinusoidal hyperplasia with large numbers of neutrophils including some immature forms. Metamyelocytes and promyelocytes and late erythroblast are also scattered visible. Cytogenetic studies of bone marrow samples 2 months after the transplantation showed no Ph-positive cells. Chromosome analysis of bone marrow cells after the splenectomy showed complete donor chimerism.
GVHD
The patient had no signs of acute GVHD. On day 210, the patient developed cGVHD that was limited to the skin after the splenectomy. Administration of a high dose of CsA and prednisone resulted in a good clinical response.
Patient 2
Engraftment
On day 15, patient 2 had a WBC count of 1.8 × 109/L, ANC of 0.5 × 109/L, and platelet count of 15 × 109/L. Bone marrow examination 1 month after the transplantation revealed low cell proliferation with granulocytic and erythrocytic percentages of 34% and 47%, respectively. Cytogenetic analysis of bone marrow samples showed that the Ph-positive cells decreased to 5%.
Pancytopenia and Delayed Bone Marrow Recovery
On day 18, the platelet count decreased to 5 × 109/L. The WBC count and ANC were 1.7 × 109/L and 1.1 × 109/L, respectively. A platelet transfusion on day 46 increased the platelet count to 11 × 109/L. After the transfusion, the WBC count and ANC were 3.0 × 109/L and 2.2 × 109/L, respectively. A second bone marrow examination after the transplantation revealed a higher cell proliferation with granulocytic and erythrocytic percentages of 70.4% and 24.8%, respectively.
Splenectomy and Hematopoitic Recovery
An abdominal ultrasound showed an enlarged spleen (14 × 8 × 6 cm) in patient 2. Considering the possibility of hypersplenism and a delay in the recovery of bone marrow, the patient received a splenectomy on day 51 (Fig. 2). The platelet count increased quickly to 20 × 109/L after the splenectomy and the WBC count and hemoglobin level were 20.2 × 109/L and 112 g/L, respectively. Bone marrow examination 1 month after the splenectomy revealed cell proliferation with granulocytic and erythrocytic percentages of 56% and 33.2%, respectively. Two megakaryocytes were also observed. Pathological analysis of the resected spleen confirmed hypersplenism. Histological examination of the spleen showed splenic white pulp shrink, the medullary cord and sinus of red pulp with large numbers of neutrophils. Some immature forms and late erythroblast are scattered visible. A mild red pulp congestion was also observed. Cytogenetic analysis of the bone marrow samples performed 2 months after the transplantation showed no Ph-positive cells.
Hematological parameters after allo-PBSCT in patient 2.
GVHD
On day 55, patient 2 presented with erythra on her body. To determine the possibility of syngeneic GVHD (sGVHD), we performed a left clavicle skin biopsy. Pathological analysis of the skin biopsy confirmed sGVHD. Dexamethasone (10 mg) was administered once a day for 3 days and resulted in a good clinical response.
Patient 3
Engraftment
On day 12, patient 3 had a WBC count of 0.8 × 109/L, ANC of 0.6 × 109/L, and platelet count of 17 × 109/L. Bone marrow examination 1 month after the transplantation revealed obvious proliferation with granulocytic and erythrocytic percentages of 53% and 42%, respectively. Two megakaryocytes were also observed. Cytogenetic analysis of bone marrow samples showed no Ph-positive cells. Chromosome analysis of bone marrow cells each month showed complete donor chimerism.
Pancytopenia
Patient 3 was treated with G-CSF and achieved a maximal unsustained WBC count of 8.2 × 109/L that decreased to 1.1 × 109/L after treatment withdrawal (Fig. 3). The patient was transfusion dependent after the transplantation. The range of hemoglobin levels and platelet counts were 50–65 g/L and 17–33 × 109/L, respectively.
Hematological parameters after allo-PBSCT in patient 3.
Splenectomy and Hematopoietic Recovery
An abdominal ultrasound showed an enlarged spleen (21 × 14.5 × 4 cm) in the patient. On day 114, the patient received a splenenctomy. Spleen pathology revealed hypersplenism. Three days after the splenectomy, the WBC count, hemoglobin level, and platelet count were 6.3 × 109/L, 108 g/L, and 53 × 109/L, respectively. Pathological analysis of the resected spleen confirmed hypersplenism. This patient's spleen showed sinusoidal hyperplasia with large numbers of neutrophils including some immature forms, but the spleen showed no other abnormality. Bone marrow examination performed 1 month after the splenectomy revealed proliferation with granulocytic and erythroblast percentages of 62% and 32%, respectively. Four megakaryocytes were also observed. Cytogenetic studies performed 2 months after the transplantation on bone marrow samples showed no Ph-positive cells. Chromosome analysis of bone marrow cells after the splenectomy showed complete donor chimerism.
GVHD
Patient 3 presented with a rash on the upper and lower limbs on day 10. The patient received immunosuppressive treatment for GVHD at the time of diagnosis. Administration of dexamethasone (10 mg/day for 7 days) resulted in a good clinical outcome. On day 72, the patient presented with a rash on both palms and feet and had diarrhea. Additional treatment with dexamethasone (5 mg) and high-dose intravenous γ-globulin cleared the rash and diarrhea after 15 days.
Discussion
Pancytopenia that occurs after PBSCT is associated with GVHD, graft failure, myelofibrosis, viral infection, the need for administration of drugs, and autoimmune diseases. In this study, we have analyzed the effect of a splenectomy on pancytopenia. In all cases, the patients had enlarged spleens but no evidence of cytomegalovirus infection or severe acute GVHD at the time of the splenectomy. Severe pancytopenia refractory to G-CSF-associated and red blood cell (RBC)/platelet transfusion dependency was present after the bone marrow transplant (BMT). After surgery, all patients were transfusion independent and had sustained graft function with normal peripheral hematological parameters. It has been previously reported that four CML patients that showed poor graft function after BMT from an HLA-compatible sibling donor were successfully treated by a splenectomy (11). These data suggest that a poor graft function may be successfully corrected by a splenectomy. Hypersplenism should be considered in patients that present persistent or recurrent pancytopenia and splenomegaly posttransplantation (9). Others believe a splenectomy should be considered prior to a transplant in patients with significant splenomegaly and hypersplenism (15).
Although the role of the spleen in GVHD is controversial, it is considered to be a crucial secondary lymphoid organ in acute GVHD when the disease is initiated by donor T cells that recognize host alloantigens after an allo-BMT (5). Hägglund et al. analyzed 35 risk factors in acute GVHD, such as age, time of diagnosis, female donor to male recipient, relative response and donor-responding capacity in mixed lymphocyte culture (MLC), and MNS blood group antigen, and concluded that a splenectomy was not associated with acute GVHD (10). Another study has shown no significant influence of a splenectomy on acute GVHD, cGVHD, or overall patient survival (12). However, other studies have indicated that the spleen may modulate the severity and incidence of acute GVHD (19). One explanation for this could be the possible function of the spleen as a filter of activated T lymphocytes arising from the transplant (14). Chronic GVHD is also associated with hyposplenism (4,17). In one study, a 39-year-old female patient underwent HLA-identical sibling allogeneic BMT for the treatment of CML in the accelerated phase. A splenectomy was performed on day 225 and the pancytopenia was resolved; however, extensive cGVHD developed with associated hepatic cholestasis, diffuse scleroderma, and sicca-like syndrome. The authors concluded that hyposplenism may trigger or aggravate extensive cGVHD and could lead to increased morbidity and mortality (18). In the current study, patients 1 and 2 had no evidence of GVHD before the splenectomy was performed. However, both patients developed GVHD immediately after the splenectomy. Subsequently, after treatment all of the patients have remained disease free.
The effect of a splenectomy performed post-PBSCT on the formation of CDC has not previously been reported. In the current study, the chimerism of bone marrow from patient 1 before the splenectomy was mixed donor chimerism but changed to CDC immediately after the splenectomy. Patient 2 had syngeneic PBSCT and therefore we could not test the chimerism of the bone marrow samples. Patient 3 received an allogeneic PBSCT from an unrelated donor and we observed acute GVHD. The bone marrow chimerism after the PBSCT was CDC.
The present observations suggest that splenectomy may be associated with an increased risk of GVHD while promoting the formation of CDC, and eliminating minimal residual disease (MRD). However, our data did not reach statistical significance since there were no adjustments for other factors influencing the development of GVHD. Nevertheless, our report is the first to demonstrate effects of splenectomy on GVHD, GVL, formation of CDC, and reoccurrence of leukemia.
Future studies will examine larger numbers of splenectomized patients and retrospectively analyze the long-term effects of a splenectomy. The immunology and pathology of the resected spleens will also be studied in greater detail to confirm our hypothesis that the spleen is an essential organ for holding activated host cells and combating graft cells. Use of a splenectomy may provide a new method for overcoming relapse in patients who receive an allo-PBSCT, syngeneic PBSCT, or auto-PBSCT.
Footnotes
Acknowledgments
J.D. designed and performed the research, analyzed the data, and wrote the paper along with W.B.