Lloyd Damon, 1 Hope Rugo, 1 Sara Tolaney, 2 Willis Navarro, 1 Thomas Martin III, 1 Curt Ries, 1 Delvyn Case, 3 Kenneth Ault, 3 Charles Linker 1

Transcription

1 Biology of Blood and Marrow Transplantation 12: (2006) 2006 American Society for Blood and Marrow Transplantation /06/ $32.00/0 doi: /j.bbmt Cytoreduction of Lymphoid Malignancies and Mobilization of Blood Hematopoietic Progenitor Cells with High Doses of Cyclophosphamide and Etoposide Plus Filgrastim Lloyd Damon, 1 Hope Rugo, 1 Sara Tolaney, 2 Willis Navarro, 1 Thomas Martin III, 1 Curt Ries, 1 Delvyn Case, 3 Kenneth Ault, 3 Charles Linker 1 1 University of California, San Francisco, San Francisco, California; 2 Department of Medicine, Johns Hopkins Medical Center, Baltimore, Maryland; 3 Department of Medicine, Maine Cancer Center, Scarborough, Maine Correspondence and reprint requests: Lloyd Damon, MD, University of California, San Francisco, 400 Parnassus Ave., San Francisco, CA ( damonl@medicine.ucsf.edu). Received October 28, 2004; accepted October 24, 2005 ABSTRACT We evaluated the efficiency of high doses of cyclophosphamide (6 g/m 2 ) and etoposide (2 g/m 2 ) plus filgrastim (granulocyte colony-stimulating factor; G-CSF) to mobilize autologous hematopoietic progenitor cells in patients with non-hodgkin lymphoma, multiple myeloma, and Waldenström macroglobulinemia. We also evaluated the safety of this regimen and the engraftment kinetics after myeloablative chemotherapy. Seventynine patients with high-risk or relapsed/primary refractory non-hodgkin lymphoma, multiple myeloma, or Waldenström macroglobulinemia were treated. The mobilizing regimen was as follows: cyclophosphamide 600 mg/m 2 twice daily for 10 doses, etoposide 200 mg/m 2 twice daily for 10 doses (continuous; n 57) or 2 g/m 2 over 10 hours on day 5 of etoposide (bolus; n 22), and G-CSF 5 g/kg/d beginning day 14. Fifty-nine percent of patients achieved the primary end point (a CD34 cell dose of 5 million per kilogram with a single leukapheresis). More bolus etoposide patients achieved the primary end point (86%) compared with continuous etoposide patients (47%; P <.0001). The CD34 cell dose collected was greater in bolus etoposide patients (44 million per kilogram) than in continuous etoposide patients (10.9 million per kilogram; P <.0001). Patients took 3 weeks to recover >500/ L neutrophils and >20 000/ L platelets after cyclophosphamide and etoposide. The overall response rate was 69% for non-hodgkin lymphoma patients and 71% for multiple myeloma/ Waldenström macroglobulinemia patients. The treatment-related mortality was 2.5%. Sixteen percent of surviving patients experienced grade >3 nonhematologic toxicity. Patients receiving bolus etoposide had significantly less grade >2 oral mucositis, less use of total parenteral nutrition, and less need for red blood cell and platelet transfusions. Sixty-four patients (81%) underwent autologous hematopoietic progenitor cell transplantation, with prompt engraftment. Four patients (5%) did not undergo autologous hematopoietic progenitor cell transplantation because of toxicity from high-dose cyclophosphamide and etoposide. We conclude that high doses of cyclophosphamide and etoposide combined with G-CSF are an efficient and safe mobilizing regimen for the collection of hematopoietic progenitor cells during aggressive cytoreduction of tumor burden in patients with lymphoid malignancies American Society for Blood and Marrow Transplantation KEY WORDS Cyclophosphamide Etoposide Filgrastim Cytoreduction Lymphoid malignancies INTRODUCTION Hematopoietic progenitor cells can be mobilized from the bone marrow into the blood via the administration of myeloid growth factors or cytotoxic chemotherapy [1-4]. The efficiency of hematopoietic progenitor cell mobilization seems to be increased by cytotoxic chemotherapy followed by myeloid growth factors [5-7], albeit at a price: pancytopenia and nonhematologic toxicities from chemotherapy, opportunistic infections, and increased cost. Conversely, an additional benefit of combination-modality mobilization is further cytoreduction of the malignancy before 316

2 Cyclophosphamide and Etoposide to Cytoreduce Lymphoid Malignancies and Mobilize Stem Cells autologous hematopoietic progenitor cell transplantation. However, the incremental benefit of chemotherapy plus myeloid growth factor hematopoietic progenitor cell mobilization to disease control in the setting of autologous hematopoietic progenitor cell transplantation is uncertain. The most common combination-modality hematopoietic progenitor cell mobilization regimen is cyclophosphamide (4-7 g/m 2 ) plus either granulocyte colonystimulating factor (G-CSF; filgrastim) or granulocytemacrophage colony-stimulating factor (sargramostim) [8-10]. High-dose etoposide plus myeloid growth factor has also been used as a mobilization regimen and has been useful in individuals who did not adequately mobilize with other regimens [11,12]. Multidrug chemotherapy plus myeloid growth factor to mobilize hematopoietic progenitor cells is less common. For myeloid malignancies, the combinations of high-dose cytarabine and either etoposide or mitoxantrone plus G-CSF have proven to be highly effective mobilizers of hematopoietic progenitor cells [13-16]. The use of conventional-dose combination chemotherapy plus myeloid growth factors to mobilize hematopoietic progenitor cells has been studied in lymphoid malignancies [17-23]. The optimal mobilizing regimen in patients with lymphoid malignancies has not yet been determined. The administration of high doses of cyclophosphamide ( mg/kg) and etoposide ( g/m 2 ) has been established to be a safe and effective treatment for patients with chemotherapy-refractory myeloid leukemias [24]. These doses of cyclophosphamide and etoposide are not myeloablative, and bone marrow recovery kinetics are quite reasonable. We hypothesized that the combination of high doses of cyclophosphamide (6 g/m 2 ) and etoposide (2 g/m 2 ) plus G-CSF would achieve effective cytoreduction in patients with lymphoid malignancies with a heavy tumor burden before autologous hematopoietic progenitor cell transplantation and act as an efficient mobilizing regimen for the collection of autologous hematopoietic progenitor cells. Preliminary data demonstrated that the combination of high-dose cyclophosphamide (4 g/m 2 ) and conventional-dose etoposide (up to 600 mg/m 2 ) plus myeloid growth factor effectively mobilized hematopoietic progenitor cells [25-29]. At the time of the design of our trial, tandem autologous hematopoietic progenitor cell transplantation was first being explored for multiple myeloma, testicular cancer, and breast cancer [30-35]. Our strategy was to use this intensive cytoreductive treatment regimen as the first part of a 2-step autologous hematopoietic progenitor cell transplantation in place of a first tandem transplantation for lymphoid malignancies. The primary objective was to explore whether this regimen would also efficiently and safely mobilize hematopoietic progenitor cells. BB&MT MATERIALS AND METHODS Patient Eligibility Patients aged 16 to 65 years with an established histologic diagnosis of non-hodgkin lymphoma (NHL), multiple myeloma, or Waldenström macroglobulinemia were eligible for enrollment provided that a prospective plan to treat the patient with high-dose therapy and autologous hematopoietic progenitor cell transplantation was in place. Patients were required to have chemotherapy-responsive disease, defined as a partial response (PR) or better (defined below) to their most recent cytotoxic therapy regimen or less than a PR to their first chemotherapy regimen (primary induction failure). A bone marrow aspiration and biopsy performed within 1 month of enrollment needed to demonstrate 30% involvement with malignant cells. Adequate organ function was required as follows: serum creatinine 2 mg/dl, measured creatinine clearance 40 ml/min, left ventricular ejection fraction 50%, total bilirubin 2 mg/dl, aspartate aminotransferase 4 times the upper limit of normal, negative serology for the human immunodeficiency virus, and a Karnofsky performance status of 80%. This protocol and consent form were approved by the University of California, San Francisco (UCSF) Committee on Human Research. All patients gave informed, written consent for participation. All patients were hospitalized to receive protocol treatment. Treatment Cyclophosphamide was given as a 1-hour intravenous (IV) infusion at 600 mg/m 2 every 12 hours for 10 doses (days 1-5). Mesna was given as a 24-hour IV infusion at 1.2 g/m 2 daily for 6 doses (days 1-6). In the first 57 patients, etoposide was given as a 3-hour IV infusion at 200 mg/m 2 every 12 hours for 10 doses (days 1-5; continuous etoposide). In the remaining 22 patients, etoposide was given as a single 2 g/m hour IV infusion on day 5 (bolus etoposide). In both instances, the etoposide concentration in normal saline was 0.4 mg/ml. Standard antiemetics were given before all chemotherapy infusions. G-CSF was given at 5 g/kg subcutaneously daily beginning on day 14 and continuing until the completion of blood hematopoietic progenitor cell collections. The CD34 cell dose target was 5 million per kilogram. The CD34 cell dose was determined via published guidelines of the International Society of Hematotherapy and Graft Engineering [36]. Viability markers were not routinely performed during the CD34 cell flow cytometry. Daily leukapheresis began when the leukocyte count was 5000/ L. Each leukapheresis was performed via a COBE Spectra (Gambro BCT, Inc., Lakewood, CO) apheresis machine with 18 L of blood processed through a Quinton dialysis catheter. A colony-forming unit granulocyte-macrophage (CFU- 317

3 L. Damon et al. GM) assay was performed on aliquots from leukapheresis except when the desired CD34 cell dose was collected on a single leukapheresis [37]. For patients whose total CD34 cell dose was 5 million per kilogram, proceeding to autologous blood progenitor cell transplantation was permitted if the CFU-GM dose was /kg. Postthaw CFU-GM assays were performed on random samples quarterly for quality-assurance purposes. Viability assays were not routinely performed on individual hematopoietic progenitor cell collections. Supportive Care Platelet transfusions (6 units of random donor platelets or a single donor apheresis unit) were given when the morning platelet count was / L (low-risk patients) or / L (standard-risk patients [standard risk was defined as at least 1 of the following: fever, age 70 years, active mucositis, or uncontrolled hypertension]). Packed red blood cell transfusions were given when the morning hematocrit was 25%. Infection prophylaxis included ceftazidime and amphotericin B (0.3 mg/kg/d IV) when the absolute neutrophil count was 500/ L and was continued until neutrophil recovery. Vancomycin was added for neutropenic fever, at which time the ceftazidime was also switched to levofloxacin. Cytomegalovirus-seronegative patients received cytomegalovirus-negative blood products. Mouth care was performed 3 times a day with swish-and-spit salt and soda, Peridex, and amphotericin B (0.1 mg/ml). Outcome Measures The primary end point was to determine the proportion of patients achieving the CD34 cell dose target of 5 million per kilogram with a single apheresis collection. Secondary end points included (1) response rates from high-dose cyclophosphamide and etoposide; (2) bone marrow recovery kinetics after this treatment and engraftment kinetics of the blood hematopoietic progenitor cells collected during this treatment and later used for autologous transplantation; and (3) the nonhematologic toxicities of this intense chemotherapy regimen. Nonhematologic toxicities were evaluated daily during each subject s study hospitalization and scored according to the UCSF leukemia/bone marrow transplantation grading scale as previously published [38]. An additional objective emerged during the active phase of this study. We changed the etoposide schedule from 10 doses over 5 days (continuous etoposide) to a 10-hour bolus infusion once on day 5 (bolus etoposide) as a result of our observations in a concurrent randomized study of etoposide schedules in patients with acute leukemia who were also receiving cytarabine and idarubicin [38]. In this concurrent study, bolus etoposide produced significantly less oral mucositis and significantly less use of total parenteral nutrition and parenteral narcotics than continuous etoposide. We wished to reduce oral mucosal toxicity in our patients but needed to establish that the etoposide schedule change had no adverse effects on blood hematopoietic progenitor cell procurement efficiency, bone marrow recovery kinetics, or post autologous transplantation engraftment kinetics. Patients with NHL were staged within 1 month of study treatment with physical examination, computed tomographic scans of chest, abdomen, and pelvis, and a bilateral bone marrow aspiration and biopsy. Patients with multiple myeloma or Waldenström macroglobulinemia were staged within 1 month of study treatment with a unilateral bone marrow aspiration and biopsy, skeletal survey, serum quantitative immunoglobulin measurement, and serum and urine protein electrophoresis. Quantitation of a 24-hour urine paraprotein was performed when a urine paraprotein was present. Restaging of all patients took place (identical to that described previously, except that no bone marrow biopsy was performed in NHL patients with a negative pretreatment bone marrow biopsy) 1 month after blood hematopoietic progenitor cell procurement. Response Criteria The response criteria for NHL were as follows: complete response (CR), no detectable disease on physical examination, computed tomographic scan, or bone marrow biopsy; and PR, a 50% reduction in the sum of bidimensional products of the perpendicular diameters of measurable lymphoma masses and improvement in the bone marrow infiltration with lymphoma compared with baseline (if the marrow was involved at baseline). Progressive disease (PD) for NHL was defined as a 25% increase (from nadir) in the sum of the bidimensional perpendicular diameter products of measurable lymphoma masses and/or a 25% increase in the percentage of bone marrow involvement with lymphoma and/or any new lymphoma masses. Response criteria for multiple myeloma and Waldenström macroglobulinemia were as follows: CR, no detectable serum or urine paraproteins based on protein electrophoresis, no new skeletal lesions on plain radiographs, and bone marrow aspiration and biopsy with 5% plasma cells or 5% lymphocytes. A PR was a 50% decrease in serum paraprotein (as well as a 90% decrease in 24-hour urine paraprotein if present at baseline), no new skeletal lesions on plain radiographs, and improvement in the bone marrow infiltration of malignant cells compared with baseline. PD for multiple myeloma or Waldenström macroglobulinemia was a 25% increase (from nadir) in serum and/or urine quantitative paraprotein and/or a 25% increase in the percentage of bone marrow involvement and/or new skeletal lytic 318

4 Cyclophosphamide and Etoposide to Cytoreduce Lymphoid Malignancies and Mobilize Stem Cells Table 1. Patient Demographics Patient Features All Patients Continuous Etoposide Bolus Etoposide Patients (n) Age (y) 49.1 (21-70) 49.7 (21-70) 47.5 (31-64) Sex Male 53 (67%) 42 (74%) 11 (50%) Female 26 (33%) 15 (26%) 11 (50%) Diagnosis NHL 36 (46%) 27 (47%) 9 (41%) Indolent Aggressive Multiple myeloma 39 (49%) 26 (46%) 13 (59%) Waldenström macroglobulinemia 4 (5%) 4 (7%) 0 (0%) Disease status (n) NHL CR PR PIF CR2 or greater PR2 or greater SD Unknown Myeloma/Waldenström macroglobulinemia CR PR SD PR2 or greater Data are median (range) or n (%). CR indicates complete remission; PR, partial remission; PIF, primary induction failure; SD, stable disease; NHL, non-hodgkin lymphoma. lesions on plain radiographs. For all study patients, stable disease was that which did not meet criteria for CR, PR, or PD. Treatment-related mortality was defined as any noncancer death during the study hospitalization. The accrual goal was 50 patients to adequately assess the efficiency of blood hematopoietic progenitor cell collection, nonhematologic toxicities, and bone marrow recovery kinetics after high-dose cyclophosphamide and etoposide. An additional 22 patients were enrolled to assess the effect of switching from continuous etoposide to bolus etoposide on these outcomes. The Student t test was used to compare continuous variables between continuous and bolus etoposide groups, and contingency table analysis by using the Fisher exact test was used for categorical variables. Kaplan-Meier probabilities were used to determine event-free survival (EFS) and overall survival (OS) of study participants [39]. RESULTS Patient Characteristics Between January 1992 and July 1999, 80 patients were enrolled on this protocol (Table 1). One patient did not receive protocol treatment per physician discretion and was excluded from analysis. Nearly half of the patients had NHL, and the other half had multiple myeloma or Waldenström macroglobulinemia. Eleven patients (14%) were treated in first CR (scheduled for autologous hematopoietic progenitor cell transplantation because of high-risk features [NHL] or the diagnosis of multiple myeloma or Waldenström macroglobulinemia). Hematopoietic Progenitor Cell Procurement Forty-one (59%) of 69 evaluable patients achieved the CD34 cell dose target of 5 million per kilogram with a single leukapheresis procedure (Table 2). A significantly higher proportion (19/22; 86%) achieved the target with a single leukapheresis after bolus etoposide compared with continuous etoposide (22/57, 47%; P.0001). The median number of leukapheresis procedures to achieve the CD34 cell target was 1 (range, 1-12) and was not significantly different between the continuous and bolus etoposide groups, although there was a trend that favored bolus etoposide (Table 2). If the CD34 cell dose target was 2.5 or 3.5 million per kilogram (more typical targets at the time of this trial design), 82% and 78% of patients, respectively, would have achieved this target with a single apheresis (range for both, 1-8 aphereses). Six (8%) of 79 patients never achieved the CD34 cell target dose of 5 million per kilogram. The median CD34 cell dose collected was 18.3 million per kilogram (range, ). Four patients Table 2. Blood Hematopoietic Progenitor Cell Procurement Etoposide Schedule Parameter All Patients Continuous (n 57) Bolus (n 22) P Value Single collection to reach target 41/69 (59%) 22/47 (47%) 19/22 (86%) <.0001 Total collections 1 (1-12) 2 (1-12) 1 (1-7).052 Day of therapy of first leukapheresis 24 (19-75) 25 (20-39) 22 (19-75).98 CD34 cell dose ( 10 6 /kg) 18.3 ( ) 10.9 (1.3-93) 44.2 ( ) <.0001 CFU-GM dose ( 10 4 /kg) 109 ( ) 109 ( ) 165 (54-860).81 Data are n (%) or median (range). CFU-GM indicates colony-forming unit granulocyte-macrophage. BB&MT 319

5 L. Damon et al. Table 3. Blood Hematopoietic Progenitor Cell Engraftment Etoposide Schedule Parameter All Patients Continuous (n 57) Bolus (n 22) P Value Number undergoing autologous transplantation 64 (81%) 42 (74%) 22 (100%).2 CD34 cell dose infused ( 10 6 /kg) 10.6 (1.3-93) 8.9 (1.3-93) 16.3 (3.4-48).4 CFU-GM dose infused ( 10 4 /kg) 58.4 (4-889) 60.6 (4-889) 54 (9-256).6 Neutrophils >500/ L (day of transplantation) 11 (8-29) 11 (8-27) 11 (9-29).5 Platelets >20 000/ L (day of transplantation) 15 (8-122) 13 (8-27) 17 (8-122).008 Platelet transfusions 4.5 (1-29) 5.5 (1-29) 4.0 (1-21).3 Red blood cell transfusions 4.5 (0-23) 4.5 (0-17) 5.0 (2-23).7 Data are n (%) or median (range). CFU-GM indicates colony-forming unit granulocyte-macrophage. (5%) achieved a CD34 cell dose between 2 and 5 million per kilogram. Two patients (2.5%) achieved a CD34 cell dose of 2 million per kilogram. Four (8%) of 49 patients achieved a CFU-GM dose /kg (2 of these 4 had CD34 cell doses 2 million per kilogram). The total CD34 cell dose was greater after bolus etoposide compared with continuous etoposide (P.0001; Table 2), but the quality of these CD34 cells, as measured by CFU-GM assay, was not different (Table 2). The quality of hematopoietic progenitor cells, as measured by the CFU-GM/CD34 cell dose plating efficiency, was also no different (mean: 12.4% 10.3% for continuous etoposide versus 15.1% 11.9% for bolus etoposide; P.56 [Student t test] and P.33 [Mann-Whitney U test]). The plating efficiency analysis could be performed only in patients with paired CFU-GMs and CD34 cell doses (n 42 for continuous etoposide; n 6 for bolus etoposide). Hematopoietic Progenitor Cell Engraftment Sixty-four patients (81%) underwent autologous hematopoietic progenitor cell transplantation. Fifteen patients (19%) did not undergo the planned autologous hematopoietic progenitor cell transplantation because of disease progression (n 7), refusal (n 1), death before transplantation (n 2), inadequate quantity of collected hematopoietic progenitor cells (n 2), use of a syngeneic donor (n 1), or poor performance status after nonspecific interstitial pneumonitis (n 1) or pulmonary embolism (n 1). Thus, 4 patients (5%) did not undergo autologous hematopoietic progenitor cell transplantation because of toxicity from high-dose cyclophosphamide and etoposide. One patient underwent autologous hematopoietic progenitor cell transplantation with a suboptimal dose of progenitor cells (CD34 cell dose of 1.29 million per kilogram and CFU-GM /kg) and had rapid and sustained engraftment. In patients who underwent transplantation, the conditioning regimen was thiotepa/etoposide/cyclophosphamide in 23 [40], total body irradiation and melphalan in 31, cyclophosphamide/carmustine/etoposide in 7 [41], and other in 4. Patients with large CD34 cell collections did not receive all of their hematopoietic progenitor cells back at the time of transplantation, but, rather, some were kept in reserve. Each patient s physician determined the CD34 cell dose infused when the dose was above that particular patient s transplantation protocol specified target dose. The median CD34 cell dose infused was 10.6 million per kilogram (1.3-93), and this did not differ between the continuous and bolus etoposide groups (Table 3). Comparing continuous etoposide patients with bolus etoposide patients, the time to 500/ L neutrophils was the same, but it took a median of 4 days longer to achieve / L platelets in bolus etoposide patients (P.008). There was no difference between etoposide groups with regard to the number of platelet and red blood cell transfusions needed after transplantation (Table 3). Response to High-Dose Cyclophosphamide and Etoposide Sixty-one patients (77%) were evaluable for response. Eighteen patients were not evaluable for response because of treatment-related mortality (n 2), already in CR (n 11), or lack of data (n 5). In patients with NHL, the responses from high-dose cyclophosphamide and etoposide were as follows: CR, 12 (46%) of 26; PR, 6 (23%) of 26; stable disease, 3 (12%) of 26; and PD, 5 (19%) of 26. In multiple myeloma and Waldenström macroglobulinemia, the responses were as follows: CR, 6 (17%) of 35; PR, 19 (54%) of 35; stable disease, 8 (23%) of 35; and PD, 2 (6%) of 35. All 11 patients in CR at study entry remained in CR after protocol treatment. Hematopoietic Recovery after High-Dose Cyclophosphamide and Etoposide Hematopoietic recovery (absolute neutrophils 500/ L and platelets / L) occurred a me- 320

6 Cyclophosphamide and Etoposide to Cytoreduce Lymphoid Malignancies and Mobilize Stem Cells Table 4. Toxicities after High-Dose Cyclophosphamide and Etoposide Etoposide Schedule Parameter All Patients (n 79) Continuous (n 57) Bolus (n 22) P Value Neutrophils >500/ L (day of therapy) 21 (11-32) 22 (11-32) 20.5 (17-25) <.01 Platelets >20 000/ L (day of therapy) 23 (12-47) 24 (12-47) 20.5 (13-27) <.002 No. platelet transfusions 5 (0-32) 6 (1-32) 2 (0-16) <.0001 No. RBC transfusions 6 (0-18) 6 (0-18) 4 (0-13).006 Treatment-related mortality 2 (2.5%) 2 (3.5%) 0 >0.9 Hospital discharge (day of therapy) 26 (21-69) 26 (21-69) 24 (21-54).5 Mucositis Grade 2 13 (16%) 12 (21%) 1 (4.5%) <.002 Grade (1.3%) 1 (1.8%) 0 >0.9 Total parental nutrition, d, mean (range) 5.2 (0-29) 7.4 (0-29) 1.0 (0-11).003 Parental narcotics (d) 0 (0-46) 0 (0-32) 0 (0-46).4 Maximum total bilirubin grade >2 2 (2.5%) 2 (3.5%) 0 >0.9 Maximum alkaline phosphatase grade >2 12 (15%) 10 (17.5%) 2 (9%).3 Maximum gastrointestinal toxicity grade > >0.9 Maximum cutaneous toxicity grade >2 3 (3.8%) 2 (3.5%) 1 (4.5%) >0.9 Data are n (%) or median (range) unless otherwise noted. RBC indicates red blood cell. dian of 3 weeks after the beginning of high-dose cyclophosphamide and etoposide (Table 4). Neutrophil recovery was 1 day faster, and platelet recovery was 3 days faster when comparing bolus etoposide with continuous etoposide. Patients required fewer platelet and red blood cell transfusions after bolus etoposide compared with continuous etoposide (Table 4). Nonhematologic Toxicity from High-Dose Cyclophosphamide and Etoposide There were 2 treatment-related deaths, both in patients receiving continuous etoposide. One was due to respiratory syncytial virus (RSV) pneumonitis, and the other was due to nonspecific adult respiratory distress syndrome. Other grade 3 nonhematologic toxicities included pulmonary embolism (n 1), Candida albicans sepsis (n 1), herpes simplex esophagitis (n 1), subarachnoid/subdural hemorrhage (n 2), atrial fibrillation (n 3), nonfatal RSV pneumonitis (n 2), Bell s palsy (n 1), gastrointestinal hemorrhage (n 1), pericarditis/tamponade (n 2), non Q wave myocardial infarction (n 1), respiratory failure due to etoposide-induced metabolic acidosis (n 1), and cardiomyopathy (n 1). These events occurred in 13 patients (16%). Patients were discharged from the hospital after 26 days (21-69 days), and no difference was seen between continuous etoposide and bolus etoposide patients (Table 4). Continuous etoposide produced more oral mucositis than bolus etoposide, as evidenced by a significantly greater proportion of patients who experienced grade 2 oral mucositis (23% versus 4.5%; P.002) and a longer average duration of total parenteral nutrition (P.003; Table 4). Hepatic, gastrointestinal, and cutaneous toxicity was mild, with no difference seen in patients receiving continuous versus bolus etoposide (Table 4). BB&MT Survival after High-Dose Cyclophosphamide and Etoposide For all enrolled patients, the median EFS and OS was 23.7 and 46.1 months, respectively. For all patients, the 1- and 5-year EFS was 65.5% 10.8% and 25.5% 10.9%, respectively. For all patients, the 1- and 5-year OS was 74.3% 10% and 43.2% 12.3%, respectively. For NHL patients, the median EFS and OS were 19.6 and 37.6 months, respectively (Figure 1). For multiple myeloma/waldenström macroglobulinemia patients, the median EFS and OS were 27.1 and 58.8 months, respectively (Figure 1). DISCUSSION Our data demonstrate that high-dose cyclophosphamide combined with high-dose etoposide and G- CSF is a safe and effective mobilizing regimen for hematopoietic progenitor cells in patients with lymphoid malignancies. Fifty-nine percent of patients achieved the CD34 cell dose target of 5 million per kilogram with a single leukapheresis procedure. For less stringent and more typical CD34 cell dose targets of 2.5 and 3.5 million per kilogram, 82% and 78% of patients, respectively, would have achieved these targets with a single apheresis. Only 8% of patients did not achieve the CD34 protocol-specified cell dose target with repeated leukapheresis procedures. The median CD34 cell dose collected was high (18.3 million per kilogram), and patients engrafted promptly when their hematopoietic progenitor cells were used for subsequent autologous hematopoietic progenitor cell transplantation. Seventy percent of evaluable patients had a CR or PR to this mobilizing therapy; this demonstrates the cytoreductive capacity of this com- 321

7 L. Damon et al. survival Event-Free Survival Overall A B Non-Hodgkin's Lymphoma Myeloma/Waldenstrom's Time from Enrollment (years) Non-Hodgkin's Lymphoma Myeloma/Waldentrom's Time from Enrollment (years) Figure 1. Event-free survival (A) and overall survival (B) of non- Hodgkin lymphoma patients and multiple myeloma/waldenström macroglobulinemia patients receiving high-dose cyclophosphamide and etoposide plus filgrastim from the time of study enrollment. Patients who did not experience an event were censored at the date of last contact. The curves represent Kaplan-Meier probabilities bination chemotherapy. Despite the intensity of this treatment, 82% of patients proceeded to hematopoietic progenitor cell transplantation, and the treatment-related mortality was only 2.5%. Only 4 patients did not proceed to autologous hematopoietic progenitor cell transplantation as a result of toxicity from this mobilizing regimen. This hematopoietic progenitor cell mobilizing regimen requires substantial supportive care and resource allocation. Patients spent a median of 26 days in the hospital. Patients needed a median of 6 units of red blood cells transfused and 5 platelet transfusions. Eighteen percent of patients experienced UCSF grade 2 oral mucositis that necessitated 5 days of total parenteral nutrition. The toxicity to the liver, gastrointestinal tract, and integument, however, was mild. The only treatment-related deaths were 2 cases of respiratory failure, 1 due to RSV pneumonitis during a winter season of virulent viral outbreaks. Overall, the toxicity profile of this treatment is similar to that for an autologous hematopoietic progenitor cell transplantation. In an effort to reduce oral mucositis, we modified our regimen after the first 57 patients from continuous etoposide to bolus etoposide (with the same total etoposide dose). We were able to confirm our previous observation that bolus etoposide results in less grade 2 oral mucositis and less use of total parenteral nutrition [38]. Bolus etoposide also resulted in the need for significantly fewer red blood cell and platelet transfusions. Surprisingly, bolus etoposide resulted in a greater number of collected CD34 cells compared with continuous etoposide, and more bolus etoposide patients achieved their CD34 cell dose target with a single leukapheresis procedure (86% versus 39%, respectively; P.0001). Even though more CD34 cells were collected after bolus etoposide, this did not translate into the expected greater CFU-GM cell dose being collected after bolus etoposide (Table 2). Furthermore, it took 4 days longer for the autologous hematopoietic progenitor cells collected after bolus etoposide to engraft platelets compared with those collected after continuous etoposide (Table 3), although bolus etoposide patients required fewer platelet transfusions. This suggests, but does not prove, that the autologous hematopoietic progenitor cells collected after bolus etoposide might have experienced some damage from the bolus dose of etoposide, especially when it is considered that twice as many CD34 cells were infused into bolus etoposide patients compared with continuous etoposide patients. However, the quality of hematopoietic progenitor cells collected, as measured by the ratio of CFU-GM to CD34 cell dose (plating efficiency), was no different in a comparison of continuous and bolus etoposide patients, and this does not support the hypothesis of hematopoietic progenitor cell damage from bolus etoposide. Our lack of CD34 cell viability data at the time of hematopoietic progenitor cell collection permits the possibility of differences in CD34 cell functionality at the time of collection between continuous and bolus etoposide patients that we were otherwise unable to detect. Alternate explanations are that this was a chance finding in a retrospective subgroup analysis and/or that this was a statistical aberration in that more continuous etoposide patients (n 64) underwent autologous hematopoietic progenitor cell transplantation than bolus etoposide patients (n 22). It is not clear from this study whether high-dose cyclophosphamide and high-dose etoposide combined with G-CSF is superior to other mobilizing regimens for lymphoid malignancies. The length of hospitalization and the incidence of significant oral mucositis limits the utility of this regimen compared with simpler mobilizing regimens. Whether the increased cost, 322

8 Cyclophosphamide and Etoposide to Cytoreduce Lymphoid Malignancies and Mobilize Stem Cells toxicity, and resource utilization are balanced by increased overall antitumor effectiveness of this 2-step procedure remains to be determined. We conclude that 6 g/m 2 cyclophosphamide combined with 2 g/m 2 etoposide and G-CSF is a safe and efficient mobilizing regimen of hematopoietic progenitor cells in patients with lymphoid malignancies. Limitations of this regimen are 3 weeks of hospitalization and a significant proportion of patients experiencing grade 2 oral mucositis. The use of bolus etoposide rather than continuous infusion etoposide increases hematopoietic progenitor cell mobilization efficiency and reduces significant oral mucositis. ACKNOWLEDGMENTS The authors wish to thank Alexandra Tredinnick for her assistance in the preparation of this manuscript. REFERENCES 1. Socinski MA, Cannistra SA, Elias A, et al. Granulocyte-macrophage colony stimulating factor expands the circulating hematopoietic progenitor cell compartment in man. Lancet. 1988; 1: Gianni AM, Siena S, Bregni M, et al. Granulocyte-macrophage colony stimulating factor to harvest circulating hematopoietic progenitor cells for autotransplantation. Lancet. 1989;2: Nademanee A, Sniecinski I, Schmidt GM, et al. High-dose therapy followed by autologous peripheral-blood progenitorcell transplantation for patients with Hodgkin s disease and non-hodgkin s lymphoma using unprimed and granulocyte colony-stimulating factor-mobilized peripheral-blood progenitor cells. J Clin Oncol. 1994;12: To LB, Sheppard KM, Haylock DN, et al. Single high doses of cyclophosphamide enable the collection of high numbers of hemopoietic progenitor cells from the peripheral blood. Exp Hematol. 1990;18: Demirer T, Buckner CD, Bensinger WI. Optimization of peripheral blood progenitor cell mobilization. Progenitor Cell. 1996;14: Milone G, Leotta S, Indelicato F, et al. G-CSF alone vs cyclophosphamide plus G-CSF in PBPC mobilization of patients with lymphoma: results depend on degree of previous pretreatment. Bone Marrow Transplant. 2003;31: Linn YC, Heng KK, Rohimah S. Peripheral blood progenitor cell mobilization in three groups of subjects: a comparison of leukapheresis yield and timing. J Clin Apheresis. 2000;15: Marques JF, Vigorito AC, Aranha FJ, et al. Early total white blood cell recovery is a predictor of low number of apheresis and good CD34( ) cell yield. Transfus Sci. 2000;23: Vela-Ojeda J, Tripp-Villanueva F, Montiel-Cervantes L, et al. Prospective randomized clinical trial comparing high-dose ifosfamide GM-CSF vs high-dose cyclophosphamide GM- CSF for blood progenitor cell mobilization. Bone Marrow Transplant. 2000;25: Campos L, Bastion Y, Roubi N. Peripheral blood progenitor cells harvested after chemotherapy and GM-CSF for treatment intensification in patients with advanced lymphoproliferative diseases. Leukemia. 1993;7: Junghanss C, Leithauser M, Wilhelm S, et al. High-dose etoposide phosphate and G-CSF mobilizes peripheral blood progenitor cells in patients that previously failed to mobilize. Ann Hematol. 2001;80: Scheid C, Reiser M, Draube A, et al. Mobilization with etoposide and granulocyte colony-stimulating factor can replace bone marrow harvesting in patients with malignant lymphoma who previously failed to mobilize sufficient progenitor cells with cyclophosphamide and G-CSF. J Hematother Stem Cell Res. 2000;9: Linker CA, Ries C, Damon LE, et al. Autologous progenitor cell transplantation for acute myeloid leukemia in first remission. Biol Blood Marrow Transplant. 2000;6: Schiller G, Lee M, Miller T, et al. Transplantation of autologous peripheral blood progenitor cells procured after high-dose cytarabine-based consolidation chemotherapy for adults with acute myelogenous leukemia in first remission. Leukemia. 1997; 11: Shimazaki C, Yamagata N, Tatsumi T, et al. Mobilization of peripheral blood progenitor cells by high-dose Ara C, VP-16 and recombinant human granulocyte colony-stimulating factor: factors affecting progenitor cell yields. Bone Marrow Transplant. 1995;15: Visani G, Lemoli R, Tosi P, et al. Use of peripheral blood progenitor cells for autologous transplantation in acute myeloid leukemia patients allows faster engraftment and equivalent disease-free survival compared with bone marrow cells. Bone Marrow Transplant. 1999;24: Pettengell R, Testa NG, Swindell R, et al. Transplantation potential of hematopoietic cells released into the circulation during routine chemotherapy for non-hodgkin s lymphoma. Blood. 1993;82: Hohaus S, Martin H, Wassmann B, et al. Recombinant human granulocyte and granulocyte-macrophage colony-stimulating factor (G-CSF and GM-CSF) administrated following cytotoxic chemotherapy have a similar ability to mobilize peripheral blood progenitor cells. Bone Marrow Transplant. 1998;22: Kroger N, Zeller W, Fehse N, et al. Mobilizing peripheral blood progenitor cells with high-dose G-CSF alone is as effective as with Dexa-BEAM plus G-CSF in lymphoma patients. Br J Haematol. 1998;102: Vogel W, Kunert C, Blumenstengel K, et al. Correlation between granulocyte/macrophage-colony-stimulating-forming units and CD34 cells in apheresis products from patients treated with different chemotherapy regimens and granulocyte-colonystimulating factor to mobilize peripheral blood progenitor cells. J Cancer Res Clin Oncol. 1998;124: Aurlien E, Holte H, Pharo A, et al. Combination chemotherapy with mitoguazon, ifosfamide, MTX, etoposide (MIME) and G-CSF can efficiently mobilize PBPC in patients with Hodgkin s and non-hodgkin s lymphoma. Bone Marrow Transplant. 1998; 21: Mayer J, Koristek Z, Vasova I, et al. Ifosfamide and etoposidebased chemotherapy as salvage and mobilizing regimens for poor prognosis lymphoma. Bone Marrow Transplant. 1999;23: Zelenetz AD, Hamlin P, Kewalramani T, et al. Ifosfamide, carboplatin, etoposide (ICE)-based second-line chemotherapy BB&MT 323

9 L. Damon et al. for the management of relapsed and refractory aggressive non-hodgkin s lymphoma. Ann Oncol. 2003;14(suppl 1):i5-i Brown RA, Herzig RH, Wolff SN, et al. High-dose etoposide and cyclophosphamide without bone marrow transplantation for resistant hematologic malignancy. Blood. 1990;76: Weaver CH, Schwartzberg LS, Birch R, et al. Collection of peripheral blood progenitor cells after the administration of cyclophosphamide, etoposide, and granulocyte-colony-stimulating factor: an analysis of 497 patients. Transfusion. 1997;37: Weaver CH, Zhen B, Schwartzberg L, et al. A randomized trial of mobilization of peripheral blood progenitor cells with cyclophosphamide, etoposide, and granulocyte colony-stimulating factor with or without cisplatin in patients with malignant lymphoma receiving high-dose chemotherapy. Am J Clin Oncol. 1998;21: Weaver CH, Schulman KA, Buckner CD. Mobilization of peripheral blood progenitor cells following myelosuppressive chemotherapy: a randomized comparison of filgrastim, sargramostim, or sequential sargramostim and filgrastim. Bone Marrow Transplant. 2001;27(suppl 2):S23-S Weaver CH, Schulman KA, Wilson-Relyea B, et al. Randomized trial of filgrastim, sargramostim, or sequential sargramostim and filgrastim after myelosuppressive chemotherapy for the harvesting of peripheral-blood progenitor cells. J Clin Oncol. 2000;18: Mollee P, Pereira D, Nagy T, et al. Cyclophosphamide, etoposide, and G-CSF to mobilize peripheral blood progenitor cells for autologous progenitor cell transplantation in patients with lymphoma. Bone Marrow Transplant. 2002;30: Attal M, Harousseau JL, Facon T, et al. Single versus double autologous progenitor-cell transplantation for multiple myeloma. N Engl J Med. 2003;349: Fassus AB, Barlogie B, Ward S, et al. Survival after relapse following tandem autotransplants in multiple myeloma patients: the University of Arkansas total therapy I experience. Br J Haematol. 2003;123: Broun ER, Nichols CR, Gize G, et al. Tandem high dose chemotherapy with autologous bone marrow transplantation for initial relapse of testicular germ cell cancer. Cancer. 1997;79: Ayash LJ, Clarke M, Silver SM, et al. Double dose-intensive chemotherapy with autologous progenitor cell support for relapsed and refractory testicular cancer: the University of Michigan experience and literature review. Bone Marrow Transplant. 2001;27: Wallerstein R Jr, Spitzer G, Dunphy F, et al. A phase II study of mitoxantrone, etoposide, and thiotepa with autologous marrow support for patients with relapsed breast cancer. J Clin Oncol. 1990;8: Ayash LJ, Elias A, Wheeler C, et al. Double dose-intensive chemotherapy with autologous marrow and peripheral-blood progenitor-cell support for metastatic breast cancer: a feasibility study. J Clin Oncol. 1994;12: Sutherland DR, Anderson L, Keeney M, Nayar R, Chin-Yee I. The ISHAGE guidelines for CD34 cell determination by flow cytometry. International Society of Hematotherapy and Graft Engineering. J Hematother. 1996;5: StemCell Quality Control Technical Manual. Version Vancouver, BC, Canada: StemCell Technologies; Damon LE, Johnson LJ, Ries C, et al. Treatment of acute leukemia with idarubicin, etoposide and cytarabine (IDEA). A randomized study of etoposide schedule. Cancer Chemother Pharmacol. 2004;53: Cheson BD, Horning SJ, Coiffer B, et al. Report of an international workshop to standardize response criteria for non- Hodgkin s lymphomas. J Clin Oncol. 1999;17: Damon L, Wolf J, Rugo H, et al. Autologous bone marrow transplant (ABMT) for non-hodgkins (NHL) and Hodgkins (HD) lymphoma using a thiotepa-based preparative regimen. Blood. 1994;84(suppl 1):706a. 41. Stockerl-Goldstein KE, Horning SJ, Negrin RS, et al. Influence of preparatory regimen and source of hematopoietic cells on outcome of autotransplantation for non-hodgkin s lymphoma. Biol Blood Marrow Transplant. 1996;2: