Influence of applied CD34 + cell dose on the survival of Hodgkin's lymphoma and multiple myeloma patients following autologous stem cell transplants

Background/Aim. Autologous stem cell transplants (ASCTs) improve the rate of overall survival (OS) in patients with hematological malignancies such as multiple myeloma (MM) after induction chemotherapy, aggressive non-Hodgkin's lymphomas (NHL), and relapsed, chemo-therapy-sensitive Hodgkin's lymphoma the most important parameter that influenced both EFS and OS after ASCT. Conclusion . Data obtained in this study undoubtedly confirmed that CD34 + cell dose applied is an independent factor that may contribute to superior clinical outcome and OS of HL and MM patients following ASCT.


Introduction
Autologous stem cell transplants (ASCTs) improve the rate of overall survival (OS) in patients with hematological malignancies such as multiple myeloma (MM) after induction chemotherapy, aggressive non-Hodgkin's lymphomas (NHL) 1 , and relapsed, chemotherapy-sensitive Hodgkin's lymphoma (HL) 2 . As a result, ASCT has become the standard therapeutic option for these malignancies 3,4 . In order to identify patients benefiting from ASCT, several clinical parameters were reported to be of prognostic importance in HL 5 , and MM 6 . Moreover, some ASCT parameters may also influence OS of transplanted patients including early lymphocyte, neutrophil and platelet recovery, infused lymphocyte dose, and the number of infused CD34 + cells 7 . Of particular importance is the number of CD34 + cells received by patients, which is a common predictor of the potential engraftment 8 . Moreover, there may be a correlation between the number of given CD34 + cells, and disease relapse, transplant-related mortality and OS. However, the role of an infused autograft CD34 + cell dose and early lymphocyte, neutrophil, and platelet recovery following ASCT has not been firmly established as standard procedure 1,2,7 .
The present study aimed to evaluate the influence of applied CD34 + cell dose and various clinical parameters that might influence early post-ASCT and OS of HL or MM patients following transplants.

Methods
This retrospective study included a total of 210 patients who underwent ASCT between November of 2005 and January of 2017. Ninety patients were diagnosed with HL and 120 with MM.
Each patient with HL went through an initial standard staging according to the Ann Arbor classification evaluation before treatment 8 , with calculation of the International Prognostic Score (IPS) for risk stratification 9 .
MM patients were, after initial evaluation, staged according to the Durie and Salmon clinical staging system, and risk groups were determined according to the International Scoring System (ISS) 10 . Chromosomal abnormalities were revealed using interphase fluorescence in situ hybridization (iFISH) 11 .
All HL patients were initially treated according to ABVD protocol (adriamycin, bleomycin, vinblastine and dacarbazine) and were evaluated according to current response criteria 12 . Platinum-based salvage chemotherapy was given at relapse.
Stem cell (SC) mobilization was completed by granulocyte-colony stimulating factor (G-CSF) at standard dose of 10-16 μg per kg of body mass (kgbm) in all patients with previously application of chemotherapy [salvage regimen in HL and cyclophosphamide, adriamycin and dexamethasone (CAD) or high dose (HD)-cyclophosphamide in MM)].
Finally, cells were cryopreserved using our original controlled-rate freezing procedure by optimized dimethyl sulfoxide (10% DMSO) and stored at -140 ± 5 °C (mechanical freezer) or at -196 °C (liquid nitrogen) and thawed immediately prior clinical use in a water bath at 37 ± 3 °C, as described previously 13,14 . The CD34 + cell quantity in harvest was determined with a flow cytometer (Beckman-Coulter, USA). Cell viability (i.e. the ratio of "non-apoptotic" CD34 + cells) was also estimated on the basis of the 7-aminoactinomycin D (7-AAD) flow cytometry assay (Immunotech, France), as earlier described 15 .
The BEAM [total dose (TD) -carmustine 300 mg/m 2 , etoposide 800 mg/m 2 , cytarabine 1600 mg/m 2 and melphalan 140 mg/m 2 ] conditioning-protocol was given in 76 HL patients (84.4%), while 14 patients (15.6%) received the CBV (TD-cyclophosphamide 6,000 mg/m 2 , carmustine 300 mg/m 2 , etoposide 750 mg/m 2 ). G-CSF was administered after autologous SCs infusion and was continued until the absolute neutrophil count (ANC) was at least 1.0 × 10 9 /L on two consecutive days. Platelet (PLT) transfusions were administered empirically for patients with PLT counts of 20 × 10 9 /L or lower, or in patients who experienced bleeding. Mediastinal radiation was applied after ASCT on initially bulky mediastinal mass, if post ASCT positron emission tomography / computed tomography (PET/CT) was positive. Within posttransplant relapse, five patients received brentuximabvedotin (anti-CD30 antibody), and two more cases, as postransplant consolidation due to high risk of relapse.
Regarding MM patients, a historical VAD regimen, as initial treatment was given in 36 patients (30.0%), Thalidomide-based combinations in 80 (66.7%) patients, and bortezomib-based regimes in 4 (3.3%) patients. Peripheral blood SCs were collected during 1-2 consecutive aphereses following mobilization protocol CAD. In poor mobilizers (6 patients), second mobilization was conducted with addition of Plerixafor with a sufficient number of CD34 + cells for transplant (≥ 4 × 10 6 /kgbm). In 5 patients, who underwent "tandem" ASCT, a target CD34 + cell dose of 8.0 × 10 6 /kgbm was collected. The conditioning regiment consisted of high dose melphalan, as a single agent at a dose of 200 mg/m 2 , and at reduced dose of 100 or 140 mg/m 2 for patients with reduced creatinine clearance (30-60 mL/min) or with a high comorbidity index. Patient therapeutic response was evaluated according to criteria of the International Myeloma Working Group 16 . Relapsed patients were treated with bortezomibbased combinations if they did not receive bortezomib initially.
The study was performed according to the guidelines of the Declaration of Helsinki and was approved by the local Ethics board.
Following ASCT, the OS was measured from the date of ASCT until the last follow-up or until death from any cause, while event free survival (EFS) was measured from the date of ASCT until the disease progression/relapse or the last follow-up. OS functions were calculated using the Kaplan-Mayer approach, while a log-rank test was used to compare statistical differences between curves. The cutoff points for recovery of absolute lymphocyte count (ALC) of 500 × 10 6 /L or greater (ALC500), ANC ≥ 500 × 10 6 /L (ANC500), and PLT count ≥ 20 × 10 9 /L (PLT20), by Day +20, Day +11, and Day +13, respectively, were calculated according to previously published data 7 . The Spearmanʼs correlation coeffi-cient was used to analyze correlations among variables of interest. In order to predict OS after ASCT, cutoff values of CD34 + cells for HL and MM, were determined as 25th and 75th percentile values of its distribution, respectively. Statistical analyses were done using IBM SPSS statistical package (Version 21). All statistical tests were two-sided. The level of significance (alpha level) in all analyses was set at p < 0.05.

Patient characteristics and cellular research
The clinical and laboratory characteristics of HL and MM patients are summarized in Tables 1 and 2. A total of 90 patients with HL, and 120 patients with MM were analyzed.

Analysis of patients' survival
The median follow-up time for patients with HL was 67 months (range 12-192 months). Median EFS after ASCT was 20 months (range 1-119 months), and median OS after ASCT was 38 months (3-119 months). OS after ASCT wasn't influenced by gender, presence of B symptoms, bulky disease and Ann Arbor clinical stage at diagnosis (p > 0.05). Initial IPS influenced EFS (p = 0.015), but not OS (p = 0.062). Pre-ASCT Eastern Cooperative Oncology Group Performance Status (ECOG PS) influenced both EFS and OS (p < 0.0001). Favorable pre-ASCT treatment response, as well as after-ASCT, strongly influenced EFS and OS (p < 0.0001). OS of the patients with an unfavorable treatment response (PD, SD) was very poor with a median survival of less than 12 months.

Fig. 1 -Event free survival (A) and overall survival (B) following autologous stem cell transplant (ASCT)
according to applied CD34 + cell dose in Hodgkin's lymphoma patients.

Discussion
Previous reports showed that many clinical and laboratory variables after ASCT in hematological malignancies were associated with better OS 7,[17][18][19] . However, there is no firm evidence as to which parameter represents the best OS predictor.
Early lymphocyte, neutrophil and PLT recovery were reported to influence OS and EFS after ASCT in patients with HL 2 , NHL 7 , and MM 17 . Our results suggest that a delayed recovery of ALC500 after Day +20, and PLT after Day +13 were associated with inferior OS and EFS in HL, while prolonged PLT20 recovery in MM patients correlated with inferior OS.
Although CD34 + cell dose was investigated as a potential factor that might affect early recovery of ALC500, ANC500 and PLT20 after ASCT 7, 20 , this was not the case as determined by our study. The absence of any strong correlation between blood cell recovery and CD34 + cell dose in our study might be the result of additional variables such as different pre-transplant conditioning protocols that were administered to patients. Our data support the results of some previous studies that have suggested there might be an OS benefit from receiving higher CD34 + cell dose 7,21,22 . The patients with HL who received lower CD34 + cell dose had shorter EFS as well as OS. The administration of CD34 + cell dose remained an independent predictive factor of OS in multivariate models, which is in accordance with the study of Gordan et al. 23 , who have suggested the predictive role of CD34 + cell dose on OS in a mixed population of patients with HL and NHL undergoing ASCT. Additionally, it was demonstrated that ALC by Day +15 was an independent prognostic marker for the progression free survival (but not prognostic for OS), indicating faster overall recovery caused by CD34 + cell dose. Delayed ALC recovery and lower CD34 + cell dose may allow minimal residual disease to outgrow and overcome immunologic activity 24 . Furthermore, current investigations in this field suggest the potential role of lymphocyte subsets that contribute to early immune reconstitution, and may have a protective role against residual disease progression, as well as possibility to better and safer mobilize lymphocyte subsets 25,26 .
Since in both HL and MM patients the higher CD34 + cell doses correlated with an improved chance for OS, it is possible that receiving higher CD34 + dose indicates a "healthy" marrow which could mobilize more CD34 + cells, since better mobilization is not only represented by number of collected SCs 7 . Furthermore, in the present study, MM patients had lower median collected and infused CD34 + cell dose compared to HL ones, which might be the consequence of age-related factors and poorer mobilization potential, since MM patients are more older compared to HL patients. Moreover, the bone marrow microenvironment likely has an additional, still unknown stimulating role in engraftment, especially in younger patients.
In HL patients, not only did CD34 + cell dose independently influenced OS, but pre-transplant disease status showed prognostic significance on EFS and OS after ASCT. This may suggest that high dose chemotherapy followed by ASCT improves treatment response 5 . Of particular interest in MM patients is the presence of renal impairment, which was of borderline significance regarding OS, and correlated with unfavorable EFS, possibly due to disease aggressiveness and reduced-dose melphalan for conditioning. However, some previous studies reported that MM patients with impaired renal function may have outcomes comparable to those with normal renal function, despite the use of conditioning dose reduction 27 . This is mainly due to the usage of proteasome inhibitors in induction treatment, whose proportion in our study is rather small.
Although the current study has a few limitations including its retrospective nature, the fact that different conditioning regimens were used, the relatively limited number of patients and the inability to determine lymphocyte subsets, it points out the prognostic role of CD34 + cell dose as an easy detectable parameter that correlate with OS after ASCT.

Conclusion
Although SC transplant represents standard procedure in relapsed/refractory HL and MM patients, there is no variable that might help in identifying high-risk patients who underwent ASCT. The results obtained in this study confirm that advanced response through pre-ASCT treatment, early recovery of ALC500 and PLT20 (HL patients), as well as PLT20 (MM patients) could influence the patients' OS. Also, superior CD34 + cell dose could be a useful predictive factor for treatment efficacy. More precise evaluation of overall treatment effectiveness by ASCT required prospective CD34 + cell and some lymphocyte subsets investigations using randomized, controlled and larger clinical studies.