Bone Marrow Transplantation (2001) 28, 835–839  2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Myeloma DCEP (dexamethasone, cyclophosphamide, etoposide, and cisplatin) is an effective regimen for peripheral blood stem cell collection in multiple myeloma

M Lazzarino1, A Corso1, L Barbarano2, EP Alessandrino1, R Cairoli2, G Pinotti3, G Ucci4, L Uziel5, F Rodeghiero6, S Fava7, D Ferrari8, M Fiumano`9, G Frigerio10, L Isa11, A Luraschi12, S Montanara12, S Morandi13, D Perego14, A Santagostino15, M Savare`16, A Vismara17 and E Morra2

1Division of Hematology, IRCCS Policlinico S Matteo, University of Pavia, Italy; 2Division of Hematology, Ospedale Niguarda, Milano, Italy; 3Division of Oncology, Varese, Italy; 4Division of Oncology, Lecco, Italy; 5Division of Oncology, Milano S Paolo, Italy; 6Division of Hematology, Vicenza, Italy; 7Division of Internal Medicine 2, Legnano, Italy; 8Division of Internal Medicine, Abbiategrasso, Italy; 9Division of Oncology, Sondrio, Italy; 10Division of Internal Medicine, Como, Valduce, Italy; 11Division of Oncology, Gorgonzola, Italy; 12Division of Oncology, Verbania, Italy; 13Division of Hematology, Cremona, Italy; 14Division of Internal Medicine, Desio, Italy; 15Division of Oncology, Vercelli, Italy; 16Division of Internal Medicine, Magenta, Italy; and 17Division of Internal Medicine 2, Rho, Italy

Summary: grams for multiple myeloma. Bone Marrow Transplan- tation (2001) 28, 835–839. DCEP (dexamethasone, cyclophosphamide, etoposide, Keywords: peripheral stem cell; mobilization; myeloma; and cisplatin) has proved to be an effective salvage ther- chemotherapy; purging in vivo apy for refractory-relapsed MM patients. Little is known, however, about its potential as mobilizing ther- apy. The aim of this study was to evaluate the efficacy of DCEP in mobilizing PBSC and to define its toxicity. Peripheral blood stem cell (PBSC) transplantation follow- Fifty-five MM patients received DCEP followed by G- ing high-dose chemotherapy represents an integral part of CSF as part of high-dose programs including autolog- front-line therapy in patients with multiple myeloma.1,2 ous transplantation. At the time of mobilization, 40 Many studies have tried to define the optimal number of patients had previously received VAD only, and 15 CD34+ stem cells for rapid hematopoietic reconstitution alkylating agents. Mobilization was successful after PBSC transplantation. A number ranging between 2.0 minimum number of CD34+ cells 2 ؋ 106/kg) in 48/55 and 5.0 ϫ 106/kg has been reported to be sufficient for) patients (87%), and 41/55 patients (75%) collected Ͼ4 engraftment.3–5 Unfortunately, not all patients mobilize an ؋ 106/kg CD34+ cells. Of the seven patients who did adequate number of PBSC to receive intensive conditioning not mobilize stem cells, five (71%) had been previously regimens safely. The combination of high-dose cyclophos- exposed to alkylating agents. The median number of phamide (HDCTX) with hematopoietic growth factors is -CD34+ cells harvested was 5.8 ؋ 106/kg (range 2.1–22.4). considered the standard mobilizing protocol to collect per There was no treatment-related mortality. The side- ipheral blood stem cells.6–8 HDCTX, however, is associated effects of DCEP were always tolerable. No neutropenia with several side–effects such as nausea, emesis, neutro- Ͻ1000/␮l nor thrombocytopenia Ͻ50000/␮l were penic fever, sepsis, cardiac dysfunction, and hemorrhagic observed. No patient required transfusion as a conse- cystitis, with mortality rates of 1% to 2%.7,8 In addition, quence of therapy, or hospitalization for septic HDCTX has limited activity against the myeloma clone complications. In conclusion, DCEP, in addition to its with little improvement of response after three or four demonstrated anti-tumor activity, is an effective regi- cycles of VAD in newly diagnosed patients.2 men for mobilizing peripheral blood progenitor cells in Tumor contamination of the graft is another crucial prob- myeloma patients, with little or no side-effects. These lem. As collection and reinfusion of monoclonal plasma properties render DCEP a useful regimen for the cells might be associated with a shorter disease-free sur- debulking and mobilization phase of high-dose pro- vival, a reduction of tumor contamination of the autograft through positive selection of CD34+ cells has been pro- posed.9–12 Whether the CD34 antigen is present on MM progenitor cells is still debated and, so far, there is no con- Correspondence: Prof M. Lazzarino, Division of Hematology, IRCCS clusive evidence that purging of graft reduces the risk of Policlinico San Matteo, Viale Golgi 19, 27100 Pavia, Italy relapse or progression in multiple myeloma patients. On the Received 7 June 2001; accepted 13 August 2001 other hand, it is known that in vitro purging techniques DCEP in multiple myeloma M Lazzarino et al 836 cause stem cell loss and depletion of T and B lymphocytes Becton Dickinson, San Jose, CA, USA) for 15 min at 20°C, from the graft, with consequent delayed immune recovery then lysed with a lyse-no-wash standard assay and finally and increased infectious complications in transplanted incubated for 10 min at 4°C. Cells were processed with patients.9,13,14 To reduce tumor contamination of the graft, FACSort analyzer (Becton Dickinson). Data acquisition an alternative is to improve in vivo debulking using more was performed with Cellquest software (Becton intensive chemotherapy programs during the mobilizing Dickinson). phase.15 We used a combination of dexamethasone, cyclo- phosphamide, etoposide and cisplatin (DCEP) with the aim Cryopreservation of treating myeloma and at the same time mobilizing PBSC. The rationale was: (1) DCEP has proved to be an Cells, diluted in dimethylsulfoxide (DMSO) solution at a effective and well tolerated regimen for refractory MM final DMSO concentration of 10%, were frozen to −160°C patients; (2) the side-effects of this combination are accept- by means of a KRYO II Series freezer (Planer R203) at able even in the fragile setting of refractory patients; (3) a controlled rate and then stored in the liquid phase of schemes including cisplatin, in association with G-CSF, nitrogen. have been reported as good mobilizing protocols in cancer patients. To our knowledge, this is the first study on DCEP Statistical analysis as a mobilizing regimen in multiple myeloma patients. The aim of this study was to evaluate its efficacy in mobilizing Continuous variables are reported as median and range, and stem cells and to define its toxicity. categorical variables as count and relative frequency. For each dichotomized clinical parameter, analysis of variance (ANOVA) techniques were used to test for a difference in Materials and methods the mean numbers of collected CD34+ cells between the two groups of patients defined according to the clinical Patients parameter. Because of the small sample size, no multivari- Between January 2000 and February 2001, 55 patients (30 ate analysis was carried out. Fisher’s exact test was used males and 25 females) with myeloma received DCEP to assess the association between dichotomous clinical + (dexamethasone, cyclophosphamide, etoposide, and parameters and a yield of CD34 cells below the threshold cisplatin) in 4-day infusions followed by G-CSF as part of of 4 ϫ 106/kg. The dichotomous parameters significantly + high-dose programs including autologous transplantation. associated with a yield of CD34 cells below the threshold + The median age was 54 years (range 30–68). Of 55 patients, were then considered as risk factors for a low CD34 col- 40 had previously received VAD only, and 15 alkylating lection. Finally, the number of risk factors present in each agents. Patients were restaged at the start of DCEP. The patient was used to build a scoring system to predict an + stages by Durie and Salmon criteria were as follows: 36 inadequate yield of CD34 cells. All tests were two-tailed, patients stage I; six stage II; 13 stage III. With the aim of with a significance level P р 0.05. identifying factors correlated with the number of CD34+ cells mobilized, the therapeutic history of the patients and the following clinical and laboratory parameters at the start Results of DCEP were evaluated: stage, interval from first treatment to DCEP, previous exposure to alkylating agents, type of M The clinical and laboratory characteristics of patients and component, Bence–Jones proteinuria, skeletal involvement, prior chemotherapy history are reported in Table 1. Of the age, WBC and platelet counts, hemoglobin level, bone mar- 55 patients, 15 (27%) had been previously exposed to row plasmacytosis, creatinine, urea. WBC and platelet counts, hemoglobin level, creatinine, and urea were evalu- ated as categorical variables with respect to the normal Table 1 Characteristics of patients at the start of DCEP ranges. We used a cut-off of 50% for bone marrow plasma- cytosis, and of 6 months as interval from first treatment Number of patients 55 to DCEP. Male/Female 30/25 Age 54 (30–68) Patients received DCEP (dexamethasone 40 mg for 4 Type of MC days, and 4-day continuous infusion of daily doses of IgG 38 cyclophosphamide 400 mg/m2, etoposide 40 mg/m2, cispla- IgA 10 tin 10 mg/m2) as mobilizing therapy, followed by hemopo- BJ 6 ␮ NS 1 ietic growth factor (G-CSF 5 g/kg/day) starting 48 h after Hb g/dl 11.6 ± 1.75 the end of chemotherapy. PBSC were collected with a con- WBC/109/l 6.2 ± 2.7 tinuous-flow blood cell separator (Spectra, COBE BCT, Platelets/109/l (mean ± s.d.) 251 ± 94.8 Lakewood, CO, USA), processing a total volume per Bone marrow involvement Ͼ50% 15 (27%) leukapheresis of two to three blood mass volumes. Prior exposure to alkylating agents 15 (27%) Disease stage at the start of DCEP I 36 (65%) Flow cytometry II 6 (11%) III 13 (24%) Aliquots of the leukapheresis products were incubated with Months from first treatment to DCEP 4 (2–54) phycoerythrin-conjugated monoclonal anti-CD34 (HPCA2,

Bone Marrow Transplantation DCEP in multiple myeloma M Lazzarino et al 837 alkylating agents. The median interval from the start of first interval from first treatment to DCEP Ͼ6 months (P = treatment to DCEP was 4 months (range 2–54). Peripheral 0.003), were found statistically to correlate with a low num- blood counts are reported as means Ϯ standard deviation. ber of CD34+ cells, considered as a continuous variable. Since, however, our target for transplantation was 4 ϫ 106 + Progenitor cell collections CD34 cells/kg, we also analyzed these variables with respect to CD34+ cells harvested considered in a dichot- Table 2 summarizes the results of PBSC and apheresis. A omous fashion. So, platelets Ͻ150 ϫ 109/l (P = 0.004), Hb total of 102 procedures were performed after 55 mobilizing Ͻ10 g/dl (P = 0.033), prior exposure to alkylating agents procedures, with the aim of collecting at least 2 ϫ 106 (P = 0.011), and a pre-DCEP interval Ͼ6 months (P = CD34+ cells/kg and with a target of 4 ϫ 106 CD34+ cells/kg 0.018) were confirmed risk factors for collections below the when possible. The first collection was performed when the cut-off of 4 ϫ 106 CD34+ cells/kg. On the basis of the four number of peripheral blood CD34+ cells reached 20/␮l. Of factors negatively associated with CD34+ cell collection, 55 patients, 48 (87%) had more than 2 ϫ 106 CD34+ patients were scored as having from 0 to 4 factors. The cells/kg collected, and 41 (75%) more than 4 ϫ 106 CD34+ analysis was performed considering the number of CD34+ cells/kg. The median number of stem cells harvested was cells as a continuous, as well as a dichotomous variable. 5.82 ϫ 106 CD34+ cells/kg (range 2.1–22.4) with a median Patients presenting with more than one risk factor showed of two leukaphereses (range 1–4), performed after a median a high probability of collecting lower numbers of stem cells time of 13 days from the start of DCEP (range 11–15). (P = 0.002), in particular, less than 4 ϫ 106/kg (P = 0.003). Seven patients (13%) failed collection. Of these, five (71%) Figure 1 shows the likelihood of mobilizing Ͼ4 × 106/kg had been previously treated with melphalan. Ten other CD34+ cells according to the number (0, 1, 2, 3) of adverse patients, in spite of a long previous exposure to melphalan, factors. Only one patient presented all four risk factors and yielded enough CD34+ cells. then it was not possible to apply the logistic regression.

Toxicity of DCEP Discussion Toxicities following DCEP were always mild and tolerable. Granulocyte counts and Hb levels never dropped below 1.0 This is the first report investigating the efficacy of a DCEP ϫ 109/l and 10 g/dl, respectively, as a consequence of treat- protocol in combination with G-CSF as mobilizing therapy ment. The only two patients who needed packed red cells in multiple myeloma patients. High-dose therapy followed transfusion had low hemoglobin levels before starting by autologous stem cell transplantation is the therapy of DCEP. No patient had thrombocytopenia Ͻ50000 ϫ 109/l. choice for patients with multiple myeloma. Most high-dose No patient needed hospitalization for septic complications. protocols include an induction phase with VAD-based regi- The most frequent extra-hematologic toxicity was nausea mens, because this combination has no negative effects on related to the administration of cisplatin, which was mild PBSC collection. The most commonly used mobilizing or moderate and always easily controlled with antiemetic regimen is HDCTX which has proved to be effective in therapy. harvesting adequate numbers of stem cells. However, because of its limited anti-myeloma activity, HDCTX has Factors affecting the yield of peripheral stem cells little capacity to reduce the tumor burden further. More intensive chemotherapy regimens have been A statistical analysis was performed of pre-DCEP para- reported to have a better capacity of purging in vivo in MM meters to evaluate their influence on stem cell collections. patients. In particular, Cook et al16 demonstrated a higher In univariate analysis, a WBC count Ͻ4.0 ϫ 109/l (P = efficacy of DHAP (dexamethasone, aracytin and cisplatin) 0.038), platelet count Ͻ150 ϫ 109/l (P = 0.012), Hb Ͻ10 than protocols including intermediate- or high-dose cyclo- g/dl (P = 0.029), marrow plasmacytosis Ͼ50% (P = 0.036), phosphamide. Demirer et al17,18 compared the combination previous exposure to alkylating agents (P = 0.011), and an of cyclophosphamide 4 g/m2 and etoposide 600 mg/m2, with or without cisplatin 105 mg/m2, with cyclophospham-

Table 2 Characteristics of PBSC collections 1.0 0.93 0.9 No. of patients 55 0.8 No. of apheresis procedures 102 0.7 0.67 No. of procedures per patient 2 (1–4) 0.6 0.50 Patients failing PBSC collection 7 (13%) 0.5 + P No. of patients collecting Ͼ2 × 106/kg CD34 cells 48 (87%) 0.4 with one apheresis 48 (87%) 0.3 0.20 with two aphereses 8 (17%) 0.2 with Ͼtwo aphereses 26 (54%) 0.1

14 (29%) Probability and 95% Cl 0.0 No. of patients with collection Ͼ4 × 106/kg CD34+ 41 (75%) 01 23 cells Score Days from start of DCEP to first apheresis 13 (11–15) + × 6 No. of CD34 cells 10 /kg, median (range) 5.82 (2.1–22.4) Figure 1 Probability of collecting Ͼ4 × 106/kg CD34 cells according to the number of 0, 1, 2, and 3 adverse factors.

Bone Marrow Transplantation DCEP in multiple myeloma M Lazzarino et al 838 ide 4 g/m2 alone, finding the triple combination superior in References mobilizing stem cells. Stewart and colleagues15 reported the superiority of a similar scheme in mobilizing blood stem 1 Attal M, Harousseau J-L, Stoppa AM et al. Prospective, ran- cells when compared with less intensive chemotherapy domised trial of autologous bone marrow transplantation and regimens. All these intensive protocols, however, created chemotherapy in multiple myeloma. New Engl J Med 1996; significant hematologic and extrahematologic problems 335:91–97. 2 Barlogie B, Jagannath S, Vesole D et al. Superiority of tandem such as transfusion requirement, septic complications, and autologous transplantation over standard therapy for pre- need for hospitalization. viously untreated multiple myeloma. Blood 1997; 89: 789– The antitumor efficacy of DCEP has been proved in the 793. difficult setting of refractory patients.19 This combination 3 Bensinger W, Appelbaum F, Rowley S et al. Factors that is also included in ongoing trials as front line,20 or consoli- influence collection and engraftment of autologous peripheral dation therapy after double transplant in MM patients.21 In blood stem cells. J Clin Oncol 1995; 13: 2547–2555. 4 Scheid C, Draube A, Reiser M et al. Using at least 5 ϫ 106/kg our hands, preliminary analysis confirmed the efficacy of + DCEP in reducing tumor burden (data not shown). This CD34 cells for autologous stem cell transplantation signifi- study, focused on the mobilization activity of the combi- cantly reduces febrile complications and use of antibiotics nation, clearly demonstrates that DCEP is an effective mob- after transplantation. Bone Marrow Transplant 1999; 23: 1177–1181. ilizing regimen for transplant-eligible patients with MM. In 5 Tricot G, Jagannath S, Vesole D et al. Peripheral blood stem fact, in the entire population of MM patients studied, 87% cell transplant for multiple myeloma: identification of favor- 6 + of cases could yield Ͼ2 ϫ 10 CD34 cells/kg, and 75% able variables for rapid engraftment in 225 patients. Blood + yielded more than 4 ϫ 106 CD34 cells/kg. Notably, 10 1995; 85: 588–596. out of 15 patients (67%) previously exposed to alkylating 6 Goldschmidt H, Hegenbart U, Haas R, Hunstein W. Mobiliz- agents mobilized enough CD34+ cells to support high-dose ation of peripheral blood progenitor cells with high-dose melphalan-based autologous transplantation. These results cyclophosphamide (4 or 7 g/m2) and granulocyte colony-sti- compare favorably with those reported with CTX at 4 or 7 mulating factor in patients with multiple myeloma. Bone Mar- g/m2,6,16,18,22 and are similar to those obtained with more row Transplant 1996; 17: 691–697. intensive combination protocols.15,16 These latter studies, 7 Kotasek D, Shepherd KM, Sage RE et al. Factors affecting blood stem cells collections following high-dose cyclophos- however, showed considerable toxicity with prolonged neu- phamide mobilization in lymphoma, myeloma and solid tropenia which entailed in the majority of patients septic tumors. Bone Marrow Transplant 1992; 9:11–17. complications, marked anemia, and thrombocytopenia with 8 To LB, Shepperd KM, Haylock DN et al. Single high doses need for transfusions and hospitalization. By contrast, of cyclophosphamide enable the collection of high numbers DCEP was always well tolerated with very mild side- of haemopoietic stem cells from the peripheral blood. Exp effects. Actually, no patient needed to be hospitalized for Hematol 1990; 18: 442–447. 9 Bensinger W, Buckner CD, Shannon DK et al. Transplan- septic complications, and only two patients, who were + anemic at the start of DCEP, received packed red cell trans- tation of allogeneic CD34 peripheral blood stem cells in fusions. No instances of neutropenia Ͻ1.0 ϫ 109/l or patients with advanced hematologic malignancy. Blood 1996; thrombocytopenia Ͻ50 ϫ 109/l were observed. All patients 88: 4132–4138. 10 Gorin NC, Lopez M, Laporte JP et al. Preparation and suc- were managed in an outpatient setting apart from the 4 days cessful engraftment of purified CD34+ bone marrow progeni- of DCEP infusion. tor cells in patients with non-Hodgkin lymphoma. Blood 1995; The evaluation of the factors affecting the yield of 85: 1647–1654. + CD34 cells showed in univariate analysis that WBC count, 11 Schiller G, Vescio R, Freytes C et al. Transplantation of platelet count, Hb level, plasmacytosis, previous exposure CD34+ peripheral blood progenitor cells after high-dose to alkylating agents, and an interval from the first treatment chemotherapy for patients with advanced multiple myeloma. to DCEP Ͼ6 months, were statistically correlated with the Blood 1995; 86: 390–397. collection of low numbers of CD34+ cells. Dichotomizing 12 Lemoli RM, Martinelli G, Zamagni E et al. Engraftment, clini- + cal and molecular follow-up of patients with multiple myel- the number of CD34 cells collected as more or less than + 4 ϫ 106, only platelets Ͻ150 ϫ 109/l, Hb Ͻ10 g/dl, prior oma who were reinfused with highly purified CD34 cells to Ͼ support single or tandem high-dose chemotherapy. Blood exposure to alkylating agents, and a 6 months interval 2000; 95: 2234–2239. maintained their negative influence on stem cell yield. We 13 Stockerl-Goldstein KE, Brown JM, O’Brian RM et al. have previously demonstrated the utility of a scoring sys- Increased transplant-related mortality following high-dose tem to identify patients at a high risk of yielding insufficient sequential chemotherapy and autologous hematopoietic cell numbers of CD34+ cells.22 In this report, we also show that transplantation using the Ceprate SC stem cell concentration the presence of more than one of the above-identified risk system for multiple myeloma. Blood 1999; 94 (Suppl. 1): factors implies a high probability of collecting lower num- 2704a. bers of stem cells, in particular less than 4 ϫ 106/kg, which 14 Bensinger WI. Should we purge? Bone Marrow Transplant is considered the standard target for a rapid recovery after 1998; 21: 113–115. PBSC-supported high-dose therapy in MM. 15 Stewart DA, Guo D, Morris D et al. Superior autologous blood stem cell mobilization from dose-intensive cyclophos- In conclusion, DCEP is an effective regimen for mobiliz- pamide, etoposide, cisplatin plus G-CSF than from less inten- ing peripheral blood progenitor cells in transplant-eligible sive chemotherapy regimens. Bone Marrow Transplant 1999; patients with multiple myeloma. This combination is 23: 111–117. always well tolerated with very mild side-effects, allowing 16 Cook G, Marinaki P, Farrell E et al. Peripheral blood progeni- easy management of the mobilization phase. tor cell mobilization in patients with multiple myeloma fol-

Bone Marrow Transplantation DCEP in multiple myeloma M Lazzarino et al 839 lowing oral idarubicin and dexamethasone (Z-Dex) induction 20 Munshi N, Desikan R, Anaissie E et al. Peripheral blood stem therapy. Leukemia 1997; 11 (Suppl. 5): 35–40. cell collection after CAD+ G-CSF as part of total therapy II in 17 Demirer T, Bukner CD, Storer B et al. The effect of different newly diagnosed multiple myeloma: influence of thalidomide chemotherapy regimens on peripheral blood stem cell mobiliz- administration. Blood 1999; 94 (Suppl. 1): 2577a. ation yield. Proc Ann Meet Am Soc Oncol 1996; 15: 1007a. 21 Desikan R, Munshi N, Zangari M et al. DCEP consolidation 18 Demirer T, Bukner CD, Gooley T et al. Factors influencing chemotherapy after 2 cycles of melphalan-based high-dose collection of peripheral blood stem cells in patients with mul- therapy. High incidence of CR and superior outcome in com- tiple myeloma. Bone Marrow Transplant 1996; 17: 937–941. parison with matched historical controls. Blood 1999; 94 19 Desikan KR, Jagannath S, Siegel D et al. Dexamethasone, (Suppl. 1): 1411a. cyclophosphamide, etoposide and cis-platinum (DCEP), an 22 Corso A, Caberlon S, Pagnucco G et al. Blood stem cell col- effective regimen for relapse after high-dose chemotherapy lection in multiple myeloma: definition of a scoring system. and autologous transplantation. Blood 1996; 88 (Suppl. 1): Bone Marrow Transplant 2000; 26: 283–286. 2331a.

Bone Marrow Transplantation