Transplantation, (1997) 20, 375–380  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Factors influencing recovery after cell transplantation in multiple myeloma

MA Gertz1, MQ Lacy1, DJ Inwards1, AA Pineda2, MG Chen3, DA Gastineau1, A Tefferi1, RA Kyle1 and MR Litzow1

1Division of Hematology and Internal Medicine, 2Division of Transfusion Medicine, and 3Division of Radiation Oncology, Mayo Clinic and Mayo Foundation, Rochester, MN, USA

Summary: cells result in durable engraftment after transplantation.2 Hematopoietic recovery can be accelerated by the use of We sought to determine factors that impact on the blood stem cells mobilized with growth factors or chemo- recovery of after transplantation in therapy (or both). Hematopoietic recovery is more rapid patients with multiple myeloma. We performed retro- after mobilized peripheral blood stem cell transplantation spective analyses in 51 patients undergoing blood cell than after autologous bone marrow transplantation, and transplantation for multiple myeloma. The pro- hematopoietic recovery has been shown to be sustained, portional-hazards model was applied to determine sig- indicating that accessory cells found in the bone marrow nificant risk factors. Of 51 transplants, 14 patients failed are not required for durable engraftment.3 The more rapid to achieve a platelet count of 50 × 109/l. Median time to engraftment with mobilized stem cells is likely due to an a neutrophil count of 0.5 × 109/l was 10.5 days. Median increase in committed progenitors. The addition of bone time to achieve a platelet count of 50 × 109/l was 32 days. marrow to blood stem cells appears to be unnecessary.4–6 Multivariate analysis revealed that cyclophosphamide In myeloma, tumor contamination appears to be less fre- and G-CSF priming before collection of hematopoietic quent in peripheral blood stem cells compared with bone precursors (P Ͻ 0.001) was a positive predictor of rapid marrow in the same patient.7 In most but not all cases, engraftment and prior exposure to melphalan given unmanipulated peripheral blood stem cell products are pref- orally (P = 0.02) was a negative predictor of subsequent erable to bone marrow harvests as a method of reducing platelet engraftment. The number of mononuclear cells myeloma autograft tumor contamination.8 For these collected, the patient’s disease status at the time of reasons, mobilized blood stem cell transplants have largely transplant and the presence of circulating plasma cells replaced autologous bone marrow transplants for multiple in the harvested product did not have a significant myeloma. impact on time to platelet engraftment. We conclude The median time to achieve sustained platelet counts that cyclophosphamide and G-CSF priming shortened greater than 50 × 109/l has been reported to be as short as the time to achieve platelet engraftment compared with 15 days after peripheral blood stem cell transplantation.9 G-CSF alone. Prior exposure to melphalan delayed Others have demonstrated a reduced ability to mobilize pro- platelet engraftment and can lead to complete failure of genitors in patients who have myeloma compared with platelet recovery. Stem cells should be collected before those who have lymphoma.10 Moreover, complete failure melphalan administration in patients with multiple of platelet recovery has been reported, and some patients myeloma who are candidates for possible blood cell have taken in excess of 50 days to achieve a platelet count transplantation. of 20 × 109/l.11 Keywords: hematopoietic stem cell; leukapheresis; We also have encountered patients undergoing peripheral melphalan; multiple myeloma; stem cell transplantation blood stem cell transplantation for multiple myeloma who experienced delayed or inadequate platelet engraftment. Failure of platelet recovery is rarely seen in our larger group of patients undergoing peripheral blood stem cell Myeloablative chemotherapy with infusion of hematopo- transplantation for recurrent malignant lymphoma. We ietic precursors, capable of reconstituting marrow function, undertook the current study to determine if there are clinical is applied increasingly in the primary management of predictors for the risk of poor platelet recovery in patients patients with multiple myeloma. Initially, autologous bone who receive transplants for multiple myeloma. marrow was reinfused with up to 30% plasma cells with excellent reconstitution of hematopoietic function.1 It was subsequently demonstrated that there are circulating hema- Materials and methods topoietic progenitor cells that can be collected and these Patient population Correspondence: Dr MA Gertz, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA The patients included in this study all had multiple mye- Received 15 December 1996; accepted 5 May 1997 loma and underwent blood cell transplantation between Platelet recovery in myeloma transplant MA Gertz et al 376 May 1989 and May 1996. All patients gave written infor- cells/kg. In four patients, the goal was 6 × 108 mononuclear med consent. Consent was in accord with the Declaration cells/kg, and this goal was achieved in all four. of Helsinki. The consent form was approved by the Insti- Eleven patients received conditioning with melphalan tutional Review Board of Mayo Foundation. All patients 140 mg/m2, plus total body radiation 12 Gy, and cyclophos- were followed until 1 August 1996, and none were lost to phamide 60 mg/kg daily for 2 days. Thirty-five received follow-up. conditioning with melphalan 140 mg/m2, plus total body radiation 12 Gy. Two received cyclophosphamide 60 mg/kg daily for 2 days and total body radiation of 12 Gy. Two Stem cell procurement and conditioning received standard-dose busulfan and cyclophosphamide, Thirty-one patients received G-CSF priming (Table 1), 5 and one received melphalan 200 mg/m2. Forty-eight of the ␮g/kg subcutaneously daily, until the completion of periph- 51 received radiation and three received conditioning with eral blood stem cell collection. For patients receiving G- non-radiation-containing regimens. CSF alone, stem cell apheresis was begun on the fifth day following G-CSF initiation. Definition of engraftment Cyclophosphamide priming was added in June 1994.12,13 The 12 patients who received cyclophosphamide priming All patients had the following measurements taken on a were given 1.5 g/m2 over 60 min daily for 2 consecutive daily basis for the first 30 days after stem cell infusion: days. The dose of cyclophosphamide was selected to obvi- count, absolute neutrophil count, and plate- ate the need for uroprophylaxis. An attempt was made to let count. From day +30 to day +100, the counts were meas- have the entire collection procedure be performed as an ured at a minimum of once weekly. The definition of gra- outpatient procedure. To minimize the period of severe nulocyte engraftment was the first day of 3 consecutive granulocytopenia, a dose of 1.5 g/m2 × 2 days was selected. days in which the absolute neutrophil count exceeded Following cyclophosphamide, G-CSF was begun on day 3. 0.5 × 109/l. Platelet engraftment was considered the first Stem cell apheresis began when the total white cell count day of 3 consecutive days in which the platelet count recovered to greater than 500/␮l. Of the 12 patients, five untransfused exceeded 50 × 109/l. required platelets during collection. Ten liters of blood were processed with each leukapher- Statistical analysis esis collection. The target progenitor cell number was 8 × 108 mononuclear cells/kg in 47 of the patients. This All patients were assessed at day +100 for graft function. goal was achieved in 42 of the 47, with the remaining five At day +100, patients who did not fulfill these criteria were collecting 7.7, 7.4, 6.9, 6.8 and 3.8 × 108 mononuclear censored as graft failures. All patients who died before day

Table 1 Patient characteristics

Variable Number or range Median

Age, years 32–69 (53) Sex 31 M/20 F Myeloma bone disease 48 Creatinine, ␮mol/l 50–210 (80) ␤2-Microglobulin, mg/l 0.9–9.5 (3.2) % Marrow plasma cells at transplantation 2–95 (34) Bone marrow labeling index, % 0–7.4 (0.9) No mobilization 8 G-CSF primed collection 31 CTX- and G-CSF-primed collection 12 Disease response to prior chemotherapy 42 Chemotherapy resistant at time of conditioning 34 Post-BMT complete response 18 Pheresis day 1: count, × 109/l Leukocyte 2.0–22.6 (11.5) Lymphocyte 0.04–3.16 (0.86) Monocyte 0.04–1.14 (0.35) Total MNC infused × 108/kg 3.8–18.5 (9.1) Prior exposure Melphalan 37 CTX 22 BCNU 22 Durie–Salmon stage IIA 6 IIB 1 IIIA 36 IIIB 8

CTX = cyclophosphamide; MNC = mononuclear cells. Platelet recovery in myeloma transplant MA Gertz et al 377 +100 who failed to achieve engraftment as defined above above as well as the two patients who had were censored on the date of death. No patients were engraftment after day +100. Of the nine remaining patients, excluded from analysis. Graft function analysis was perfor- three represent early deaths before platelet engraftment but med using the Kaplan–Meier method. Multivariate analysis after granulocyte engraftment and six were true failures of was performed using the proportional-hazards model (Cox). platelet recovery in spite of survival beyond day +100. Fig- All P values were calculated by log-rank (Mantel-Cox) test. ure 1 demonstrates the time course of platelet recovery in all 51 patients. There is a steeply sloped portion of the curve from day +7 until day +18 that reflects a subset of Results patients with rapid platelet engraftment. After day +20, the curve levels out, indicating delayed platelet recovery for A total of 51 patients had transplants, and their character- roughly half of the patients. istics are given in Table 1. At the time of pretransplant In vitro measurement of granulocyte–macrophage col- evaluation, the median percentage of bone marrow plasma ony-forming units (GM-CFU) was performed in 35 of the cells was 34%, with 54% of the population having greater 51 patients. The number of GM-CFU/kg body weight was than 30% bone marrow plasma cells. The bone marrow associated with rapid platelet recovery (P Ͻ 0.001). labeling index ranged from 0 to 7.4%, with a median of Because of the large number of missing observations, the 0.9%; 56% had a labeling index greater than 0.8%. Forty- number of GM-CFU was not included in the multivariate six of the patients underwent melphalan total body analysis of clinical factors used to predict platelet recovery. irradiation (TBI)-based conditioning, two patients received A univariate analysis was undertaken using the factors cyclophosphamide and TBI, and three patients who had in Table 2 to assess their impact on graft function. Five extensive prior radiation treatment received non-radiation- factors were significant in predicting time to platelet containing regimens. engraftment in the univariate model. When these factors Of the 51 patients, nine received transplants after fewer were placed into a proportional-hazards multivariate model, than 8 × 108 mononuclear cells/kg were collected. Two of only two factors remained predictive of time to platelet these patients were considered graft failures, not achieving engraftment. Only cyclophosphamide and G-CSF-primed 50 × 109 platelets/l by day 100. The median time to platelet harvests (P Ͻ 0.001) and melphalan given before stem cell recovery in these nine patients was 28 days. All patients collection (P = 0.02) were statistically significant predictors received G-CSF, 5 ␮g/kg per day subcutaneously, begin- of graft function. ning on day +1. The median time to achieve an absolute Although melphalan was statistically significant as a con- neutrophil count greater than 0.5 × 109/l was 10.5 days. tinuous variable, it was also assessed as a categorical vari- Three patients died not having achieved an absolute neutro- able to determine whether a threshold existed for melphalan phil count of 0.5 × 109/l. One patient died on day −1of uncontrolled sepsis, one died on day +12 of veno-occlusive disease of the liver, and one patient died on day +49 of Table 2 Univariate analysis of impact of variables on platelet adult respiratory distress syndrome and pulmonary asper- engraftment gillosis. There were two patients who achieved an absolute neutrophil count greater than 0.5 × 109/l after day +100 Variable P value (day 174 and day 199). Recovery of platelet counts was more protracted than Sex 0.03 Age 0.64 granulocyte recovery. The median time to achieve 20 × 109 Stage (LI/␤2M) 0.19 and 50 × 109 platelets/l for all patients was 17 and 32 days, Creatinine 0.48 respectively (Figure 1). Twelve and 14 patients, respect- C-Reactive protein 0.73 ively, failed to achieve 20 × 109 and 50 × 109 platelets/l LDH 0.14 Bone marrow plasma cells 0.66 (27%). This includes the three early deaths mentioned Circulating plasma cells 0.96 G-CSF-stimulated harvest 0.01 0 Cyclophosphamide-stimulated harvest 0.0001 Bone disease 0.88 Prior response to chemotherapy 0.20 20 Complete response post-BMT 0.78 Chemotherapy resistant at BMT 0.34 Melphalan dose preharvest 0.001 40 Preharvest WBC 0.001 MNC collected 0.76 60 Cyclophosphamide dose preharvest 0.17 BCNU dose preharvest 0.10 Engrafted (%) ␤2-Microglobulin 0.78 80 Labeling index, plasma cells 0.91 Lymphocyte count preharvest 0.60 Monocyte count preharvest 0.41 100 0 20 40 60 80 100 120 GM-CFU (n = 35) 0.0001 Days = = ␤2M ␤2-microglobulin; GM-CFU granulocyte–macrophage colony- Figure 1 Kaplan–Meier curve demonstrates platelet engraftment forming units; LDH = lactate dehydrogenase; LI = labeling index; MNC (Ͼ50 × 109/l) in all transplant patients (n = 51). = mononuclear cells. Platelet recovery in myeloma transplant MA Gertz et al 378 0 number of GM-CFU (35 patients) and the total cumulative melphalan dose before stem cell collection. There was a 20 correlation with Spearman rho of −0.57 (P = 0.001). P < 0.001 40 Prior melphalan Discussion 60 Our data indicate that cyclophosphamide plus G-CSF prim- Engrafted (%) ing before collection of circulating progenitors and avoid- 80 No prior melphalan ance of melphalan are important to collect adequate pro- genitors to prevent delay or failure of platelet engraftment 100 0 20 40 60 80 100 120 after transplantation in multiple myeloma patients. In pre- Days viously untreated patients who survive the early post-trans- plant period, the autograft invariably provides satisfactory Figure 2 Kaplan–Meier curve demonstrates time to platelet engraftment and sustained hematopoietic reconstitution.14 It has been (Ͼ50 × 109/l) in those patients receiving no prior melphalan (n = 14) com- pared with those receiving any prior melphalan (n = 37). suggested that higher progenitor cell yields occur with higher priming doses of growth factor. G-CSF at 10 ␮g/kg provided a higher yield than a combination of cyclophos- exposure. Figure 2 illustrates time to platelet engraftment phamide priming with a growth factor dose of 5 ␮g/kg.15 in those patients who had received no melphalan compared However, because none of our patients received a dose of with those who had received any exposure to melphalan G-CSF greater than 5 ␮g/kg, it is impossible to make a before stem cell collection. The median time to platelet direct comparison. Others have suggested that combi- engraftment was 14 days vs 50 days (P Ͻ 0.001). All nations of growth factors, including G-CSF and GM-CSF, patients who had not previously received melphalan had enhance the mobilization of hematopoietic precursors with- engraftment by day +63, whereas more than one-third of out pulsed chemotherapy.16 In one study, higher peak the patients who had melphalan exposure still had not values for CD34+ were seen in patients receiving 7 g/m2 achieved 50 × 109 platelets/l by day +100. An identical of cyclophosphamide compared with 4 g/m2 of cyclophos- analysis was performed on those patients receiving Ͻ200 phamide.17 Even under these circumstances, mobilization mg total dose melphalan compared with those receiving in myeloma patients was unreliable; poor yields were seen Ͼ200 mg (Figure 3). The median time to platelet in patients with the heaviest bone marrow infiltration with engraftment was significantly different: 15 days vs 100 days plasma cells.12 There appears to be no difference between (P Ͻ 0.001). In the group of patients given Ͼ200 mg of GM-CSF and G-CSF in terms of the ability to mobilize melphalan, only one-third had achieved 50 × 109 platelets/l sufficient numbers of hematopoietic precursors.18 It is gen- by day +100. Analysis was also undertaken for a melphalan erally agreed, however, that growth factors are needed to dose of 100 mg. There were 32 patients who received Ͼ100 reduce the number of collections required to perform safe mg melphalan. Median time to 50 × 109 platelets/l was 60 autografting.13,19,20 days. Nineteen patients received Ͻ100 mg of melphalan Hematopoietic recovery has been correlated with the before a peripheral blood stem cell collection, with median dose of GM-CFU or CD34+ cells.21 A correlation between time for platelet engraftment of 14.5 days (P = 0.0003). the number of CD34+ cells and GM-CFU has been reported, The risk of graft failure appears to be minimal in patients suggesting one can serve as a surrogate measure of the whose prior melphalan exposure did not exceed 200 mg. other.22 Patients receiving fewer than 2 × 106 CD34+ A Spearman rank correlation was performed on the cells/kg have been reported to have significantly prolonged platelet recovery.23 The absolute number of CD34 cells in the leukapheresis product has previously11 been correlated 0 P < 0.001 with exposure to prior treatment. Measures of GM-CFU are not available in many clinical laboratories that perform 20 blood cell transplants. Standardization between laboratories Melphalan >200 has not been perfected, and there is a time delay to allow 40 for colony growth, so that immediate results are not available. 60 Most clinical centers prospectively determine the

Engrafted (%) adequacy of collection based on the number of collected 80 mononuclear cells and not on CD34 or GM-CFU counts. Melphalan <200 In our study we could not conclude that the number of mononuclear cells collected correlated with engraftment. 100 0 20 40 60 80 100 120 However, since there was a minimum threshold set for each Days patient, and this threshold was achieved in 90%, it would be meaningless to conclude that no correlation exists. The Figure 3 Kaplan–Meier curve demonstrates time to platelet engraftment (Ͼ50 × 109/l) in those patients whose prior melphalan exposure was Ͻ200 clinical correlations we describe should be useful in patient 24 mg (n = 29) compared with those whose prior melphalan exposure was and collection technique selection. Although graft func- + Ͼ200 mg (n = 22). tion correlates well with the total number of CD34 cells or Platelet recovery in myeloma transplant MA Gertz et al 379 GM-CFU infused,25,26 difficulty with standardizing assays Prince et al42 assessed myeloma patients undergoing makes it hard to establish an appropriate threshold. More- stem cell transplant and found that extensive prior mel- over, some authors have found no correlation between phalan therapy resulted in a high incidence of delayed CD34+ content and GM-CFU.27 There are also those who engraftment. We continue to see patients referred for con- have not been able to demonstrate a correlation between sideration of transplant after 6 to 12 months of melphalan- GM-CFU infused and the time to achieve a platelet count of based therapy, suggesting that nontransplant oncologists are 50 × 109/l.28 Without a good in vitro measure for predicting not fully aware of the toxic effect of melphalan on stem platelet recovery, clinical factors associated with durable cells. engraftment become important. In conclusion, platelet engraftment is problematic in Why does melphalan therapy appear to have such a pro- patients with multiple myeloma following blood cell trans- found effect on graft function? The leukemogenic effect of plantation. The kinetics of the late engraftment suggest melphalan in multiple myeloma and ovarian cancer is well inadequate collection of progenitor cells with standard leu- recognized, reflecting its damaging effect on hematopoietic kapheresis techniques. There is a need for clinical predic- stem cells.29,30 Lymphoma patients exposed to the related tors of engraftment because of delays involved in acquiring alkylator cyclophosphamide as part of cyclophosphamide, GM-CFU quantification and the lack of standardization hydroxydaunomycin, vincristine (Oncovin), and prednisone among laboratories. Our clinical analysis suggests that mel- (CHOP) chemotherapy have a low risk of graft failure. One phalan delays engraftment. It likely decreases the number possibility is that the oral mode of administration of mel- of viable progenitors collected. Stem cells are best acquired phalan results in greater cumulative toxic effect on the stem before any exposure to melphalan-based chemotherapy, and cells. Melphalan, when administered intravenously, has a priming with both cyclophosphamide and G-CSF enhances relatively short half-life (17–75 min) and a brief area under the likelihood of subsequent rapid engraftment. the curve.31 Kinetics of melphalan given orally are far dif- ferent, with most patients receiving four daily doses each month or seven daily doses every 6 weeks.32 This results Acknowledgements in a much greater cumulative area under the curve.33 It also appears that the area under the curve increases with suc- This work was supported by Program Project Grant No. cessive courses.34 The area under the curve correlates best CA62242, National Cancer Institute (National Institutes of with the total exposure of stem cells to melphalan and Health). might suggest why the oral formulation is far more toxic to stem cells than parenteral formulations of cytotoxic agents. Most patients with lymphoma and breast cancer References who subsequently undergo stem cell transplant have never been exposed to daily oral administration of cytotoxic 1 Barlogie B, Jagannath S, Vesole D, Tricot G. Hematopoietic agents.35 stem cell autografts in support of myeloablative therapy for Melphalan is selectively more toxic to early developing multiple myeloma. J Hematother 1994; 3: 149–153. hematopoietic precursors than busulfan, cyclophosphamide, 2 Siena S, Bregni M, Di Nicola M et al. Durability of hematopo- ietic reconstitution after transplantation of peripheral blood or BCNU.36 This can result in insufficient mobilization of 37 progenitors primed by cancer chemotherapy and rhGM-CSF peripheral blood progenitors before transplantation. It or rhG-CSF (abstract). Prog Proc Annu Meet Am Soc Clin appears that the toxic effect of melphalan is not due to Oncol 1993; 12: A1547. damage to the bone marrow microenvironment but to the 3 Watts MJ, Jones HM, Sullivan AM et al. Accessory cells do stem cells themselves.38 We have transplanted cells to not contribute to G-CSF or IL-6 production nor to rapid patients who had peripheral blood stem cells collected haematological recovery following peripheral blood stem cell before receiving a full year of melphalan therapy, and these transplantation. Br J Haematol 1995; 91: 767–772. patients demonstrated prompt engraftment at subsequent 4 Dimopoulos MA, Alexanian R, Przepiorka D et al. Thiotepa, transplant. We found that stem cell engraftment failure was busulfan, and cyclophosphamide: a new preparative regimen not rescued by backup marrow, suggesting damage to stem for autologous marrow or blood stem cell transplantation in high-risk multiple myeloma. Blood 1993; 82: 2324–2328. cells in both marrow and blood. 39 5 Langenmayer I, Weaver C, Buckner CD et al. Engraftment of Tricot et al noted that more than 6 months of exposure patients with lymphoid malignancies transplanted with auto- to alkylating agents significantly delayed hematopoietic logous bone marrow, peripheral blood stem cells or both. Bone + recovery. Sufficient numbers of CD34 cells could be Marrow Transplant 1995; 15: 241–246. obtained in only 28% of patients who had received 2 years 6 Tarella C, Benedetti G, Caracciolo D et al. Both early and of prior chemotherapy. Bensinger et al40,41 reported that the committed haemopoietic progenitors are more frequent in per- recovery of peripheral blood cell counts was more rapid in ipheral blood than in bone marrow during mobilization patients receiving peripheral blood stem cells who had induced by high-dose chemotherapy + G-CSF. Br J Haematol fewer than eight cycles of prior therapy. The French mye- 1995; 91: 535–543. loma group reported that length of previous chemotherapy 7 Henry JM, Sykes PJ, Brisco MJ et al. Comparison of myeloma cell contamination of bone marrow and peripheral blood stem has an effect on overall and event-free survival. They found cell harvests. Br J Haematol 1996; 92: 614–619. good graft recovery in all surviving patients, but they also 8 Vescio RA, Han EJ, Schiller GJ et al. Quantitative comparison reported seven early deaths, and it is unclear whether these of multiple myeloma tumor contamination in bone marrow deaths were due to chemotherapy-related toxicity or graft harvest and leukapheresis autografts. Bone Marrow Trans- failure.14 plant 1996; 18: 103–110. Platelet recovery in myeloma transplant MA Gertz et al 380 9 Berenson JR, Schiller G, Vescio R et al. CD34 peripheral stem 25 Siena S, Bregni M, Brando B et al. Flow cytometry for clinical cell transplantation in multiple myeloma (abstract). Cancer estimation of circulating hematopoietic progenitors for Invest 1995; 13 (Suppl. 1): 8–9. autologous transplantation in cancer patients. Blood 1991; 77: 10 Haas R, Mo¨hle R, Murea S et al. Characterization of periph- 400–409. eral blood progenitor cells mobilized by cytotoxic chemo- 26 Haas R, Witt B, Mo¨hle R et al. Sustained long-term hemato- therapy and recombinant human granulocyte colony-stimulat- poiesis after myeloablative therapy with peripheral blood pro- ing factor. J Hematother 1994: 3: 323–330. genitor cell support. Blood 1995; 85: 3754–3761. 11 Mahe´ B, Milpied N, Hermouet S et al. G-CSF alone mobilizes 27 Healy LE, Nirsimloo N, Scott M et al. In vitro proliferation sufficient peripheral blood CD34+ cells for positive selection by cells mobilized into the peripheral blood for collection and in newly diagnosed patients with myeloma. Br J Haematol autologous transplantation. Exp Hematol 1994; 22: 1278– 1996; 92: 263–268. 1282. 12 Tarella C, Boccadoro M, Omede P et al. Role of chemo- 28 Hirao A, Kawano Y, Takaue Y et al. Engraftment potential therapy and GM-CSF on hemopoietic progenitor cell mobiliz- of peripheral and cord blood stem cells evaluated by a long- ation in multiple myeloma. Bone Marrow Transplant 1993; term culture system. Exp Hematol 1994; 22: 521–526. 11: 271–277. 29 Mittelman M, Lewinski UH, Weiss H et al. Secondary myelo- 13 Lamy T, Drenou B, Grulois I et al. Improvement of hematol- dysplastic syndrome in multiple myeloma – a study of nine ogic recovery after high-dose intensification using peripheral patients with an attempt to detect myeloma patients at risk. blood progenitor cells (PBPC) mobilized by chemotherapy Haematologia 1994; 26: 67–74. and GM-CSF. Ann Hematol 1994; 69: 297–302. 30 Walton LA, Yadusky A, Rubinstein L et al. Stage II carci- 14 Fermand JP, Chevret S, Ravaud P et al. High-dose chemoradi- noma of the ovary: an analysis of survival after comprehensive otherapy and autologous blood stem cell transplantation in surgical staging and adjuvant therapy. Gynecol Oncol 1992; multiple myeloma: results of a phase II trial involving 63 44: 55–60. patients. Blood 1993; 82: 2005–2009. 31 Samuels BL, Bitran JD. High-dose intravenous melphalan: a 15 Feremans W, Le Moine F, Ravoet C et al. Optimal blood stem review. J Clin Oncol 1995; 13: 1786–1799. cell mobilization using 10 micrograms/kg granulocyte colony- 32 Alberts DS, Chang SY, Chen HS et al. Oral melphalan kin- stimulating factor (G-CSF) alone for high-dose melphalan etics. Clin Pharmacol Ther 1979; 26: 737–745. intensification in multiple myeloma: an intrapatient controlled 33 Reece PA, Kotasek D, Morris RG et al. The effect of food study. Am J Hematol 1994; 47: 135–138. on oral melphalan absorption. Cancer Chemother Pharmacol 16 Winter JN, Lazarus HM, Rademaker A et al. Phase I/II study 1986; 16: 194–197. of combined granulocyte colony-stimulating factor and gra- 34 Loos U, Musch E, Engel M et al. The pharmacokinetics of nulocyte-macrophage colony-stimulating factor administration melphalan during intermittent therapy of multiple myeloma. for the mobilization of hematopoietic progenitor cells. J Clin Eur J Clin Pharmacol 1988; 35: 187–193. Oncol 1996; 14: 277–286. 35 Ehrsson H, Eksborg S, Osterborg A et al. Oral melphalan 17 Goldschmidt H, Hegenbart U, Haas R, Hunstein W. Mobiliz- pharmacokinetics – relation to dose in patients with multiple ation of peripheral blood progenitor cells with high-dose myeloma. Med Oncol Tumor Pharmacother 1989; 6: 151–154. cyclophosphamide (4 or 7 g/m2) and granulocyte colony- 36 Down JD, Boudewijn A, Dillingh JH et al. Relationships stimulating factor in patients with multiple myeloma. Bone between ablation of distinct haematopoietic cell subsets and Marrow Transplant 1996; 17: 691–697. the development of donor bone marrow engraftment following 18 Demuynck H, Delforge M, Verhoef G et al. Comparative recipient pretreatment with different alkylating drugs. Br J study of peripheral blood progenitor cell collection in patients Cancer 1994; 70: 611–616. with multiple myeloma after single-dose cyclophosphamide 37 Lokhorst HM, Sonneveld P, Wijermans PW et al. Intermedi- combined with rhGM-CSF or rhG-CSF. Br J Haematol 1995; ate-dose melphalan (IDM) combined with G-CSF (filgrastim) 90: 384–392. is an effective and safe induction therapy for autologous stem 19 Majolino I, Marceno` R, Buscemi F et al. Mobilization of cir- cell transplantation in multiple myeloma. Br J Haematol 1996; culating progenitor cells in multiple myeloma during VCAD 92: 44–48. therapy with or without rhG-CSF. Haematologica 1995; 80: 38 Barbot C, Rice A, Vane`sIet al. Quality and functional 108–114. capacity of the bone marrow microenvironment of autologous 20 Boiron JM, Marit G, Faberes C et al. Collection of peripheral blood stem cell transplantation (ABSCT) recipients. Nouv Rev blood stem cells in multiple myeloma following single high- Fr Hematol 1994; 36: 325–331. dose cyclophosphamide with and without recombinant human 39 Tricot G, Jagannath S, Vesole D et al. Peripheral blood stem granulocyte–macrophage colony-stimulating factor (rhGM- cell transplants for multiple myeloma: identification of favor- CSF). Bone Marrow Transplant 1993; 12: 49–55. able variables for rapid engraftment in 225 patients. Blood 21 Bender JG, To LB, Williams S, Schwartzberg LS. Defining a 1995; 85: 588–596. therapeutic dose of peripheral blood stem cells. J Hematother 40 Bensinger WI, Rowley SD, Demirer T et al. High-dose ther- 1992; 1: 329–341. apy followed by autologous hematopoietic stem-cell infusion 22 Scott MA, Ager S, Apperley JF et al. Peripheral blood pro- for patients with multiple myeloma. J Clin Oncol 1996; 14: genitor cell harvesting in multiple myeloma and malignant 1447–1456. lymphoma. Leuk Lymphoma 1995; 19: 479–484. 41 Bensinger W, Appelbaum F, Rowley S et al. Factors that 23 Berenson J, Schiller G, Vescio R et al. Transplantation of influence collection and engraftment of autologous peripheral- CD34-positive peripheral blood progenitor cells following blood stem cells. J Clin Oncol 1995; 13: 2547–2555. high-dose chemotherapy for patients with advanced multiple 42 Prince HM, Imrie K, Sutherland DR et al. Peripheral blood myeloma. J Cell Biochem 1995; (Suppl. 21A): 361 (Abstr.). progenitor cell collections in multiple myeloma: predictors 24 Roth P, Maples J, Hall J, Dailey T. Use of control cells to and management of inadequate collections. Br J Haematol standardize enumeration of CD34+ stem cells. Ann NY Acad 1996; 93: 142–145. Sci 1995; 770: 370–372. Platelet recovery in myeloma transplant MA Gertz et al 381