Bone Marrow Transplantation (2006) 37, 1037–1042 & 2006 Nature Publishing Group All rights reserved 0268-3369/06 $30.00 www.nature.com/bmt ORIGINAL ARTICLE Early lymphocyte recovery predicts longer survival after autologous peripheral blood stem cell transplantation in multiple myeloma H Kim1, H-J Sohn2, S Kim2, J-S Lee2, W-K Kim2 and C Suh2 1Division of Hematology-Oncology, Ulsan University Hospital, University of Ulsan College of Medicine, Ulsan, Korea and 2Division of Oncology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea To understand the prognostic value of lymphocyte Although the use of allogeneic hematopoietic stem cell recovery after autologous peripheral blood stem cell transplantation has been increasing, autologous stem cell transplantation (APBSCT), we performed a retrospective transplantation (ASCT) remains the standard therapeutic study of 59 newly diagnosed multiple myeloma (MM) modality for MM.1–3 The main drawback of ASCT is its patients who underwent frontline APBSCT. Conditioning high relapse rate. Because immune-mediated tumor eradi- regimens were melphalan 100 mg/m2 for 2 days. Follow- cation may lower the relapse rate associated with ASCT, ing APBSCT, all patients showed complete or partial there is increasing interest in immune reconstitution response. Median follow-up time was 29.57 months and following ASCT.4–9 In support of this, a relationship median recovery of absolute lymphocyte count (ALC) between lymphocyte recovery after ASCT and relapse rate X1000/mm3 was 23 days. Univariate analysis revealed or survival has been observed in many diseases, suggesting that significant predictors of overall survival (OS) that early lymphocyte recovery associated with immune included bone marrow (BM) plasma cells p40% at reconstitution can act against residual disease progres- diagnosis (P ¼ 0.0243) and recovery of ALC X1000/mm3 sion.10–17 by day þ 23 (P ¼ 0.0156). Positive predictors for Despite advances in treatment, MM is still an incurable progression-free survival (PFS) were BM plasma cells disease. After ASCT had been proven to prolong disease- p40% at diagnosis (P ¼ 0.0134) and recovery of ALC free survival in MM patients, ASCT became an essential X1000/mm3 by day þ 23 (P ¼ 0.0243). Absolute neutro- treatment modality for MM patients. Absolute lymphocyte phil count X1000/mm3 on day þ 12 was marginally count (ALC) on day þ 15 after ASCT has been reported to significant for OS and PFS (P ¼ 0.0821 and P ¼ 0.1153, be an independent prognostic factor for MM, and the respectively). Multivariate analysis showed that ALC infused dose of autografted lymphocytes has been found X1000/mm3 on day þ 23 independently predicted OS to have a significant impact on clinical outcome after (P ¼ 0.031) and prolonged PFS (P ¼ 0.011), and that ASCT.11,15 These findings, however, were based on results serum b2-microglobulin was marginally significant for using bone marrow (BM) and peripheral blood (PB) stem prolonged OS (P ¼ 0.066). In conclusion, ALC recovery cells taking into consideration that most ASCTs are was an independent predictor of both OS and PFS performed using PB as a hematopoietic stem cell source. in MM. Moreover, there are no confirmatory data to support the Bone Marrow Transplantation (2006) 37, 1037–1042. previous report. The survival of late lymphocyte recovery doi:10.1038/sj.bmt.1705373 can be influenced by early treatment-related mortality Keywords: lymphocyte; multiple myeloma; survival; (TRM) without full lymphocyte recovery. Therefore, the peripheral blood stem cell transplantation poorer survival in a group with late lymphocyte recovery can be misinterpreted when early TRM is incorporated. To evaluate the prognostic significance of ALC recovery after autologous PB stem cell transplantation (APBSCT), we have undertaken a retrospective analysis of patients with Introduction MM who underwent this procedure. Multiple myeloma (MM) is characterized by the clonal pro- liferation of neoplastic plasma cells producing M-protein. Materials and methods Patients Correspondence: Professor C Suh, Department of Internal Medicine, From March 1993 to December 2003, 70 newly diagnosed ASAN Medical Center, 388-1 Pungnap-2dong, Songpa-gu, Seoul MM patients underwent APBSCT at our institution. All 138-736, Korea. E-mail: [email protected] patient data were collected prospectively and stored in Received 14 September 2005; revised 13 March 2006; accepted 14 March a computerized database. Absolute lymphocyte count 2006 was reviewed retrospectively by retrieving computerized Lymphocyte recovery after APBSCT predicts survival in MM H Kim et al 1038 medical records. All patients were confirmed by national Statistical analysis medical insurance organization to fit the guidelines for The start date for all time variables was the day of APBSCT. All patients gave written informed consent. APBSCT. The Kaplan–Meier method was used to assess Inclusion criteria included age less than 70 years, Eastern overall survival (OS) and progression-free survival (PFS), Cooperative Oncology Group (ECOG) performance status and two-tailed log-rank tests were used to compare OS and of 0–2 at the time of transplantation, life expectancy X8 PFS curves. Prognostic factors tested for MM were age; weeks, and no evidence of serious organ dysfunction stage; performance; infused MNC dose and CD34 þ cells; including renal, hepatic, cardiac, pulmonary, central total HPGF dose; ALC, ANC and platelet recovery days, nervous system and metabolic functions. Exclusion criteria and percentage of BM plasma cells; and serum concentra- included malignancies associated with human immunode- tions of B2MG, hemoglobin, serum albumin and C-reactive ficiency virus or human T-cell lymphoma virus-1-associated protein (CRP). Factors with a P-value of less than 0.2 in malignancy, history of other malignant disease in the univariate analysis were included in multivariate analysis, previous 5 years, except for squamous cell or basal cell which was performed using the Cox regression hazard carcinoma of the skin or stage I uterine cervical carcinoma model. For quantitative variables, data are reported as or cervical carcinoma in situ or tandem APBSCT. We also medians (range) and means7standard errors. For qualita- excluded patients who died of TRM without engraftment. tive variables, data are reported as number (percentage) of patients unless specified otherwise. Disease Stage was originally classified by the Durie–Salmon staging Results system and reclassified by the international staging system.18 The new staging system is focused on two Patient characteristics common measures with prognostic importance in MM: A total of 70 newly diagnosed MM patients were enrolled serum b2-microglobulin (B2MG) and serum albumin in the study. We excluded patients who did not survive concentrations. After three or more cycles of VAD more than 30 days after APBSCT with full recovery of (vincristine, adriamycin, dexamethasone) chemotherapy, ANC. Of the 70 enrolled patients, three were ineligible all eligible MM patients underwent APBSCT. owing to early death caused by TRM, without full lymphocyte recovery; and eight were ineligible due to double transplants. Thus, 59 patients, 32 (54.2%) males Mobilization and collection of autologous hematopoietic and 27 (45.8%) females, all of whom had a complete or stem cells partial response after APBSCT, were included in this study. Mobilization regimens mainly consisted of cyclophospha- Their median age was 51 years (range 20–68 years). Of the mide. Hematopoietic growth factor (HPGF), also termed 59 patients, 38 (80.9%) had stage III disease at diagnosis. granulocyte-colony stimulating factor (10 mg/kg/day) was Their median serum M-protein concentration at diagnosis infused intravenously until completion of autologous was 2.67 g/dl (range 0–10 g/dl). Their median BM plasma hematopoietic stem cell collection. Mononuclear cells cells were 40.3% (range 0.3–95.6%). Median number (MNCs) were collected using a blood cell separator CS- of pretransplant chemotherapy cycles was four (range, 3000s Plus (Baxtor, USA). At least 5 Â 106 CD34 þ cells/ 1–28 cycles). Performance status at APBSCT was ECOG kg were collected, but neither CD34-positive selection nor 0–1 in 54 (91.5%). Median infused MNC dose during purging was applied. APBSCT was 3.8 Â 108 cells/kg (range 0.3–23.5 Â 108 cells/ kg), whereas median infused CD34 þ cells during APBSCT 6 6 Conditioning regimen was 7.4 Â 10 cells/kg (range 2.4–60.1 Â 10 cells/kg). Med- Conditioning consisted of melphalan (100 mg/m2) for 2 ian value of total dose of HPGF during APBSCT was days. Hematopoietic growth factor (filgrastim 300 mg/day) 3250 mg (range 1200–12 400 mg). Median days of recovery 3 3 3 was given from day þ 1 until hematological engraftment, to ANC X500/mm , ALC X500/mm , ALC X1000/mm 3 3 with the latter defined as the first day of absolute neutrophil and platelets X20 Â 10 /mm were 12 days (range, count (ANC) X500/mm3 for 3 consecutive days. 10–25 days), 19 days (range, 10–57 days), 23 days (range, 11–68 days) and 16 days (range, 8–45 days), respectively (Table 1). Absolute lymphocyte count and absolute lymphocyte count Following stratification of the patients according to ALC recovery days o or X1000/mm3 by day þ 23 after APBSCT, we observed White blood cell (WBC) counts were performed using XE- no differences in age, sex, prognostic factors at diagnosis, 2100t (SYSMEX Co., Japan), and the percentage of the number of pretransplantation chemotherapy cycles, lymphocytes was determined microscopically using the disease status at APBSCT, infused MNC dose, infused Wright stain. The ALC was calculated as the product of CD34 þ cells dose, patients with platelet X20 Â 103/mm3 on total WBC count and percentage of lymphocytes. Complete day þ 23 and patients with ANC X500/mm3 on day þ 23. blood counts were measured every day until hematological Absolute lymphocyte count recovery weakly correlated engraftment, and the ALC was measured daily until it with ANC recovery (r ¼ 0.282, P ¼ 0.032), but infused reached 500 cells/ml, after which it was measured until ALC (r ¼ 0.111, P ¼ 0.406) and infused mononuclear cell recovery to pretreatment values.