& 2003 Nature Publishing Group All rights reserved /03 $

Transcription

1 (2003) 31, & 2003 Nature Publishing Group All rights reserved /03 $ Progenitor cell mobilization : safety profile and variables affecting peripheral blood progenitor cell collection J Sevilla 1, M González-Vicent 1, L Madero 1, F García-Sánchez 2 and M Angel Díaz 1 1 Hospital Infantil Universitario Nin o Jesu s, Avd Mene ndez Pelayo 65, Madrid, Spain; and 2 Centro de Tranfusio n de la Comunidad de Madrid, Avd Mene ndez Pelayo 65, Madrid, Spain Summary: Large volume leukapheresis (LVL) has been proposed as a simplified single-apheresis approach to collect the target number of CD34 + cells. We retrospectively analyzed results of LVL in cytokine-mobilized patients weighing less than 20 kg to evaluate adverse events and variables affecting the yield. The only major adverse event recorded was transient and reversible systolic hypotension (three episodes). All the other adverse events were mild and did not require treatment. In multivariate analysis leukocyte count (P ¼ 0.001) and younger age (P ¼ 0.009) affected the CD34 + cell number in the peripheral blood before apheresis. The number of CD34 + cells before the apheresis was the only variable affecting CD34 + cell yield in multivariate analysis (P ¼ ). In all, 77% of patients achieved the target CD34 + cell dose of / kg in their first apheresis. Recruitment was seen in 72% of the procedures, and this was related to the total blood volume processed (P ¼ ). (2003) 31, doi: /sj.bmt Keywords: large-volume leukapheresis; LVL; apheresis; children; adverse events; recruitment In children, peripheral blood progenitor cells (PBPCs) have largely replaced bone marrow as source of autologous stem cells in part because of their relatively easy collection. However, there is a concern regarding medical, psychological, social and technical difficulties in small patients. 1 3 Large volume leukapheresis (LVL) defined as the processing of more than three volumes of blood in a single session 4 has been proposed as a simplified singleapheresis approach allowing us to obtain a CD34 + target cell dose with fewer apheresis procedures. We report a safety study on LVL in small patients. We have also analyzed the variables affecting PBPC collection by this procedure. Correspondence: Dr J Sevilla, Hospital Infantil Universitario Nin o Jesu s, Avd Mene ndez Pelayo 65, Madrid 28009, Spain Received 11 July 2002; accepted 11 September 2002 Patients and methods In all, 89 progenitor cell collections by LVL were performed in 66 consecutive patients (male 42, female 24) weighing less than 20 kg (median 15 kg; range 7 20) between December 1993and December 2001 for autologous transplantation. Median age was 40 months (10 79). In total, 46 patients had solid tumors including neuroblastoma (18), central nervous system tumor (18), rabdomyosarcoma (5), primitive neuroectodermal tumor (2), Wilm s tumor (2) and germinal tumor (1). In all, 19 had hematologic malignancies including acute myeloid leukemia (8), acute lymphoid leukemia (8) and non-hodgkin s lymphoma (3). One patient had a non-malignant disease (osteopetrosis). Patients underwent 78 mobilization courses with cytokines in different schedules. A total of 11 procedures were performed on the second or third day after the first apheresis day. Total blood volume (TBV) was estimated multiplying body weight by 80, as previously reported. 5,6 Parental consent was obtained in every patient. Mobilization Four different mobilization schedules were used over this period of time: G-CSF alone at a dose of 12 mg/kg/ day (n ¼ 53), or 10 mg/kg/day(n ¼ 1), or 12 mg/kg/12 h (n ¼ 26), or G-CSF+GM-CSF at a dose of 12 mg/kg/day of G-CSF and GM-CSF at a dose of 5 mg/kg/day(n ¼ 9), given subcutaneously for 4 5 days before starting PBCP apheresis. Vascular access LVL were performed via central venous catheter (Arrow- Howes, Reading, PA, USA) with offset lumens. Catheters were placed as previously reported, 7 but in this series mainly femoral access was elected. An attempt was made to use the same catheter for harvesting and transplantation. Peripheral blood progenitor cell (PBPC) collection LVL were performed in the Pediatric Oncology Department, in a quiet room where the patients stayed with their families. Patients were not routinely sedated. PBPC were collected by a COBE Spectra cell separator (COBE, Denver, CO, USA) as described elsewhere. 8 Leukapheresis

2 264 was performed on the 5th day of mobilization. A median of one (range 1 3) apheresis was performed per mobilization regimen. Prior treatment Each different course of chemotherapy the patient received before every apheresis was counted as a chemotherapy cycle. In total, 11 procedures were performed in patients that did not receive chemotherapy before the apheresis, 20 procedures in patients that had received up to 5 cycles of chemotherapy, and 58 procedures in patients who had received more than five cycles. In all, 66 apheresis were performed on patients who had not received previous radiotherapy treatment. In total, 20 apheresis were performed on patients who had prior radiotherapy treatment, and in three cases information regarding prior radiotherapy treatment was not available. Priming The cell separator was primed with a unit of irradiated and leukocyte-depleted packed red blood cells. Anticoagulant Initially acid citrate dextrose (ACD-A) was used as anticoagulant with the inlet : anticoagulant ratio being fixed automatically by the cell separator. Since June 1997, a solution of 500 ml ACD and 5000 IU of preservative-free heparin was infused at a whole blood to anticoagulant ratio of 30 : 1 for anticoagulation. Platelet clumping was prevented by adding heparin to the collection bag. Anticoagulation schedule was performed with ACD-A in 46 procedures, and with ACD-A and heparin in 43procedures. CD34+ cells determination Leukapheresis products and peripheral blood were analyzed for expression of the CD34 antigen as previously reported. 2,8 CD34 + cell quantification in peripheral blood was performed inmediately before the procedure. Efficiency and recruitment Collection efficiency (CE) was calculated according to Ford et al (10) formula, and the recruitment in each procedure according to Torrabadella et al. 5 To determine if there was recruitment of CD34 + cells during LVL in our series, we followed the procedure described by Cull et al. 11 Briefly, they calculated the absolute CD34 + cells in peripheral blood before apheresis and compared it with the total CD34 + cells collected in their series. Safety Adverse events related with mobilization were evaluated retrospectively until December Since January 2000, data related with mobilization were recorded prospectively. Adverse effects related with leukapheresis were evaluated retrospectively until December 1999, and prospectively since January These data were obtained following the Hemapheresis Committee of the American Association of Blood Banks questionnaire. 12 In our series, we have included some adverse events considered physiologic effects by these authors, as mild paresthesias, perioral tingling or mild vasovagal events. Changes in peripheral blood counts were also analyzed, and the influence of time or volume processed during LVL evaluated. Statistical analysis A software program (Statview 4.0, Abacus Concept, Berkeley, CA, USA) was used for statistical analysis. Correlations were determined using linear regression. The influence of variables that could affect the CD34 + cell count preapheresis and CD34+ cell yield was examinated by univariate and multivariate analysis using a stepwise regression model. Results were considered significant if the P value was o0.05. Results Safety Only three adverse events were recorded related with mobilization (low fever, headache and myalgia). Vasovagal events were the most frequent adverse events related with leukapheresis (Table 1). Mild decrease in blood pressure and diaphoresis was found in nine cases, seven of them with nausea. Three cases of systolic hypotension were recorded as severe adverse events during the procedure. These patients (with no special characteristics) rapidly responded to plasma volume expanders, and the procedure did not need to be interrupted. Three patients suffered pain or hematoma at the venipuncture site. Minor bleeding, not requiring treatment, after removal of central venous catheter was observed in three patients. Four patients on anticoagulation treatment only with ACD-A suffered perioral tingling although these patients usually had prophylactic treatment with calcium gluconate requiring additional calcium treatment. One patient also had nausea probably related with citrate toxicity. Treatment with calcium was necessary in all these patients. Table 1 Incidence of adverse events Adverse events Number Citrate effects Nausea and/or vomiting 1 Perioral tingling 4 Vasovagal effects Pallor and/or diaphoresis 9 Hypotension 9 Systolic blood pressure o80 3 Pulse > Venipuncture Pain 3 Palpable hematoma 1

3 There were significant changes in peripheral blood counts following each procedure. Median platelet loss was 52%, but platelet transfusion was not required in any patient. Platelet loss was related to the blood volumes processed (P ¼ ), but not to the time employed for the procedure. Even with red blood cells priming, there was difference between pre-lvl and post-lvl hematocrit (32% vs 29%; P ¼ ). There was also difference between pre-lvl and post-lvl leukocyte number/ml (34.5 vs 25.0; P ¼ ) Evaluation of leukapheresis parameters Total blood volume processed (median 5611 range ) and blood volumes processed (median 4.6 range 3 8.4) defined this series as LVL since more than three blood volumes were processed in all procedures. LVL results are described in Table 2. Of 34 patients mobilized in their first mobilization course with G-CSF at a dose of 12 mg/kg/day, 24 achieved the target dose in a single procedure (58.8%) and 10 patients required at least a second procedure. However, those patients mobilized with 12 mg/kg/12 h achieved the target dose by one leukapheresis in 95.2% of procedures (20 of 21 LVL). Finally, six of seven patients mobilized with G-CSF+GM-CSF achieved the target dose in their first leukapheresis (85%). Recruitment Recruitment was evaluated in 50 procedures since CD34 + cell count before the apheresis was available in these cases. Most of these procedures had recruitment in our series (72%) with a median of 1.2 (range ). According to Cull et al, 11 a mean absolute CD34 + cell count of was calculated before apheresis. The mean number of total CD34 + cells collected was (range ). Therefore, a mean of 1.17-fold greater CD34 + cells was collected than were present in peripheral blood before the procedure (P ¼ 0.04). The only variable affecting the recruitment was total blood volume processed (P ¼ ). Collection efficiency Median CE was 27.5 ( ). No variable affected CE (total blood volume processed, time, leukocyte count before the procedure, hematocrit or age). Variables affecting CD34 + cell number in peripheral blood before leukapheresis Median CD34 + cell/ml in peripheral blood before each apheresis was 59 (range 2 319) (n ¼ 50). There were several variables affecting CD34 + cell number before apheresis (Table 3). Higher platelet and leukocyte count before apheresis and younger age were related in the univariate 265 Table 2 Large volume leukapheresis results Characteristic Large volume leukapheresis procedures (n=88) Number of apheresis performed after each mobilization course Median 1 Range 1 3 Total blood volume Median 1200 Range Total blood volume processed (ml) Median 5611 Range Blood volume processed (ml/kg) Median 367 Range Blood volumes processed Median 4.6 Range Inlet/flow (ml/min) Median 20 Range 6 50 Time (min) Median 280 Range CD34+cells 10 6 /kg Median 5.5 Range Table 3 Variable Variables affecting CD34+ cell count preapheresis Median (range)- CD34+ cells/ml- Age (months) a Leukocytes preapheresis/ml a Platelets preapheresis/ml a Diagnosis group b Hematologic malignancy 32 (2 292) Solid neoplasia 80 (4 319) Diagnosis b Acute lymphoid leukemia 32 (2 84) Acute myeloid leukemia 93(29 292) Non-Hodgkin s lymphoma 33 (19 58) Neuroblastoma 36 (4 136) Wilm s tumor 62 (38 86) Germinal tumor 88 (61 115) Primitive neuroectodermal 146 (94 199) tumor Central nervous system 114 (26 319) tumor Rabdomyosarcoma 49 (30 154) Prior chemotherapy b None 94 (26 319) o 6 cycles 125 (29 292) X 6 cycles 40 (2 192) Radiotherapy c No 80 (7 292) Yes 12 (2 96) Mobilization schedule b G-CSF 12 mg/kg/day 40 (2 134) G-CSF 24 mg/kg/day 86 (4 319) GM-CSF+G-CSF 40 (5 94) a Analyzed by linear regression analysis. b Analyzed by Kruskal Wallis test. c Analyzed by Mann Whitney test. r 0.3; P=0.05 r 0.4; P=0.004 r 0.4; P=0.01 P=0.03 P=0.01 P=0.004 P=0.008 P=0.04

4 266 Table 4 Variable Variables affecting CD34+ cell yield Median (range)- CD34+ cells 10 6 /kg- Age (months) a Sex b Mobilization course b First 8.2 ( ) Second 3.9 ( ) Mobilization schedule c Leukocytes preapheresis/ml a Platelets preapheresis/ml a CD34+ cell preapheresis/ml a Diagnosis group c Hematologic malignancy 6 (0.2 35) Solid neoplasia 8 ( ) Diagnosis c Status c Prior chemotherapy c None 9.29 ( ) o6 cycles 7.6 (0.6 35) >6 cycles 5.5 ( ) Radiotherapy b No 5.8 ( ) Yes 2.6 ( ) a Analyzed by linear regression analysis. b Analyzed by Mann Whitney test. c Analyzed by Kruskal Wallis test. analysis with higher CD34 + cell values. CD34 + cell count before apheresis in patients with solid tumors was higher than in patients diagnosed with hematologic malignancies (80 vs 32) being the neuroblastoma patients those whose yield was lower between those diagnosed with solid tumors (Table 3). Mobilization schedule was also related with CD34 + cell count. Those patients mobilized with higher doses of G-CSF had higher values than those mobilized with lower doses (12 mg/kg/day) or the combination of G- CSF and GM-CSF (86 vs 40 vs 40, respectively). Finally, the previous treatment was related with CD34 + cell count. Median CD34 + cell count in patients with prior radiotherapy was 12 vs 80 in those patients who had not received prior radiotherapy. There were also differences between those patients who received more or less than six cycles of chemotherapy (median 125 vs 40, respectively). In the multivariate analysis the leukocyte count before apheresis (P ¼ 0.001) and younger age (P ¼ 0.009) are the only variables related with higher CD34 + cell values in peripheral blood. Variables affecting CD34 + cell yield r 0.8; P= P=0.04 P=0.02 P=0.01 Median CD34 + cell yield was /kg (range ). Several variables had an influence on the yield of CD34 + cells in the univariate studies (Table 4). Diagnosis influenced the CD34 + cell yield in those patients diagnosed with solid tumors having higher yield than those patients diagnosed with hematologic malignancies (median 8 vs cells/kg). Those patients who received chemotherapy or radiotherapy before harvesting yielded less CD34 + cells than those who were not treated (Table 4). In the multivariate analysis, all these variables were included and CD34 + cell preapheresis was the only variable related with CD34 + cell yield (P ¼ ). Discussion In this study LVL has been found to be a safe procedure with only three patients suffering transient and reversible hypotension. All the other events were mild and did not require treatment. No serious complications have been recorded related with the vascular access, and no procedure was interrupted as the result of any physical difficulty. These complications were similar to our previous experience, 2 and that previously published by others. 1,3 Hematologic toxicities such as platelet and hematocrit drop are recognized complications of leukapheresis. 3,5,13 Changes in peripheral blood counts following LVL were found in our series with a median platelet loss of 52% being the main hematological toxicity. This result is similar to previous LVL series. Kanold et al 3 reported a 37% platelet depletion, and Torrabadella et al 5 a 39%. In this latter study, the authors referred a median platelet loss of 29% in their standard leukapheresis group with no significant differences when comparing with the LVL patients. However, in our series, platelet loss was strongly related with the blood volumes processed. Nevertheless, no serious bleeding has been found in our patients in accordance with the literature. 3,5 Recruitment, defined as a higher number of CD34 + cells collected than estimated in peripheral blood immediately before the procedure, was found in our study. In our series, 72% of the procedures showed recruitment. The mean amount of harvested CD34 + cells was 1.17-fold higher compared with the number of circulation CD34 + cells in peripheral blood before apheresis. We could not find any clinical or demographic characteristic that could determine which patients might improve their yield by recruiting progenitor cells. We only found total blood volume processed affecting the recruitment, with higher recruitment in those procedures with more blood volume processed. It will lead us to process more blood volume in those patients with lower CD34 + cells preapheresis in order to increase the yield. Some authors have defined the recruitment as a time-dependent or blood-processed phenomenon, 14 but we consider it merely as a higher collection than expected by the CD34 + cell quantification in peripheral blood before the procedure by processing more blood volume than in a standard apheresis. Whether the collection is much lower or not, after a determinate time or blood volume processed, is controversial. Gorlin et al 15 reported a significant decrease in the CD34 + cells collected after the first hour of procedure in younger patients. However, Abe et al 16 found a continuous supply of CD34 + cells during the procedure, with no decrease in their number. Recently, Cassens et al 14 found a continuous decrease in the CD34 + cells collected after a maximum peak during the second processed blood volume. We did not study the CD34 + cell kinetics during the procedure, but the recruitment is highly correlated with the total blood volume processed, and not with the time employed in the apheresis. Therefore, processing more blood volume allows us to collect more CD34 + cells from the peripheral blood, than previously estimated to be in the patient. LVL has a similar CE in this study to that previously reported. 5,10 However, hematocrit and leukocyte count

5 before apheresis are not related with CE as previously reported by others. 10 Several variables influence the CD34 + cell count preapheresis in the univariate study (Table 3). There is still some controversy related with the best dose of cytokines used for mobilization. In our series, patients mobilized with higher doses of G-CSF needed only one apheresis to achieved the CD34 + target cell dose in 95.2% of patients vs 59% in those mobilized with lower doses. We found this variable to be related to the CD34 + cell count preapheresis (P ¼ 0.04), but not with the CD34 + cell yield. This discrepancy may be because of the fact that we processed more blood volume in those patients that had lower CD34 + cell count preapheresis trying to achieved the target dose. It could lead to similar CD34 + cell yield. Some other variables were related to the CD34 + cell count preapheresis and not with the yield, and the explanation may be again in the blood volumes processed. In the multivariate analysis the leukocyte count before apheresis and younger age are the only variables related in our series with higher CD34 + cell values in peripheral blood. Higher leukocyte counts, with higher CD34 + cell numbers, may reflect a better hematopoietic response to cytokines in younger bone marrows. Several variables also influence the CD34 + cell yield in the univariate study. Diagnosis group is related to CD34 + cell yield with hematologic malignancies collecting less CD34 + cells. It may reflect some marrow disturbance by the disease, or the treatment employed. However, in the multivariate study, CD34 + cell count preapheresis is the only variable related with the yield. It has been repeatedly reported as the main variable related to the yield. Other complications studied in this paper are those related with cytokine mobilization and vascular access. Cytokine mobilization has been found to be safe although some adverse events have been reported in the past: splenic rupture, atypical dermatological lesions, acute iritis and acute gouty arthritis. 17 In this series no serious adverse event was found related to mobilization. Regarding vascular access we found pain as the only variable related to vascular access insertion. In general, no special vascular access is necessary for leukapheresis. Either apheresis catheters 2 as cuffed indwelling catheters, 18 or a special arterial access 19 have been used for collection without significant complications. In conclusion, LVL is a safe and efficient procedure, also in very small patients. Several variables are related with the number of CD34 + cells collected, but the CD34 + cell count preapheresis is the main variable that predicts the yield. Processing larger amounts of blood volume might lead to increase in the yield by recruiting progenitor cells. References 1 Gorlin JB, Humphreys D, Kent P et al. Pediatric large volume PBPC collections from patients under 25 kg: a primer. J Clin Apheresis 1996; 11: Diaz MA, Alegre A, Benito A et al. PBPC collection by largevolume leukapheresis in low-weight children. J Hematother 1998; 7: Kanold J, Halle P, Berger M et al. Large-volume leukapheresis procedure for PBPC collection in children weighing 15 kg or less: efficacy and safety evaluation. Med Pediatr Oncol 1999; 32: Alegre A, Dı az MA, Madero L. Large-volume leukapheresis for peripheral blood stem cell collection in children: a simplified single-apheresis approach. Bone Marrow Transplant. 1996; 17: Torrabadella M, Olive T, Ortega JJ et al. Enhanced HPC recruitment in children using LVL and a new automated apheresis system. Transfusion 2000; 40: Jeter EK, Rogers RL. Pediatric apheresis. In: Apheresis: Principles and Practice, McLeod BC, Price TH, Drew MJ (eds). AABB Editorial Board: Maryland, 1997, pp Madero L, Ruano D, Villa M et al. Non-tunneled catheters in children undergoing bone marrow transplantation. Bone Marrow Transplant 1996; 17: Diaz MA, Garcia-Sa nchez F, Lillo R et al. Large-volume leukapheresis in pediatric patients: pre-apheresis peripheral blood CD34+ cell count predicts progenitor cell yield. Haematologica 1999; 84: Diaz MA, Villa M, Madero L. Analysis of engraftment kinetics in pediatric patients undergoing autologous PBPC transplantation. J Hematother 1998; 7: Ford CD, Pace N, Lehman C. Factors affecting the efficiency of collection of CD34-positive peripheral blood cells by a blood cell separator. Transfusion 1998; 38: Cull G, Ivey J, Chase P et al. Collection and recruitment of CD34+ cells during large-volume leukapheresis. J Hematother 1997; 6: Mc Leod BC, Price TH, Owen H et al. Frequency of immediate adverse effects associated with apheresis donation. Transfusion 1998; 38: Takaue Y, Kawano Y, Abe T et al. Collection and transplantation of peripheral blood stem cells in very small children weighing 20 kg or less. Blood 1995; 86: Cassens U, Momkvist PH, Zuehlsdorf M et al. Kinetics of standardized large volume leukpheresis (LVL) in patients do not show a recruitment phenomenon of peripheral blood progenitor cells (PBPC). Bone Marrow Transplant 2001; 28: Gorlin JB, Vamvakas EC, Cooke E et al. LVL in pediatric patients: processing more blood diminishes the apparent magnitude of intra-apheresis recruitment. Transfusion 1996; 36: Abe T, Makimoto A, Kawano Y et al. Intra-apheresis recruitment of blood progenitor cells in children. Transfusion 1998; 38: Gutierrez-Delgado F, Bensinger W. Safety of granulocyte colony-stimulating factor in normal donors. Curr Opin Hematol 2001; 8: Kletzel M, Longino R, Rademaker AW et al. Peripheral blood stem cell transplantation in young children: experience with harvesting, mobilization and engraftment. Pediatr Transplant 1998; 2: Watanabe T, Kawano Y, Watanabe A et al. Autologous and allogeneic transplantation with peripheral blood CD34 + cells: a pediatric experience. Haematologica 1999; 84: