Outcomes of unrelated cord blood transplantation in patients with multiple myeloma: a survey on behalf of Eurocord, the Cord Blood Committee of Cellular Therapy and Immunobiology Working Party, and the Chronic Leukemia Working Party of the EBMT

Abstract

Although allogeneic stem cell transplantation is not a standard therapy for multiple myeloma, some patients can benefit from this intense therapy. There are few reports on outcomes after umbilical cord blood transplantation in multiple myeloma, and investigation of this procedure is warranted. We retrospectively analyzed 95 patients, 85 with multiple myeloma and 10 with plasma cell leukemia, receiving single or double umbilical cord blood transplantation from 2001 to 2013. Median follow up was 41 months. The majority of patients received a reduced intensity conditioning. The cumulative incidence of neutrophil engraftment was 97%±3% at 60 days, and that of 100-day acute graft-versus-host disease grade II-IV was 41%±5%. Chronic graft-versus-host disease at two years was 22%±4%. Relapse and non-relapse mortality was 47%±5% and 29%±5% at three years, respectively. Three-year progression-free survival and overall survival were 24%±5% and 40%±5%, respectively. Anti-thymocyte globulin was associated with decreased incidence of acute graft-versus-host disease, higher non-relapse mortality, decreased overall and progression-free survival. Patients with high cytogenetic risk had higher relapse, and worse overall and progression-free survival. In conclusion, umbilical cord blood transplantation is feasible for multiple myeloma patients.

Introduction

The current standard of care for patients with multiple myeloma (MM) is the use of drugs such as bortezomib, thalidomide and lenalidomide followed by autologous stem cell transplantation (ASCT).21 Although this combined treatment has markedly improved prognosis, disease recurrence remains high in MM patients. The Intergroupe Francophone du Myelome3 was the first to demonstrate, in a large randomized trial, the long-term benefits in survival of double ASCT in comparison with single transplants. Since then, other randomized trials4 and registry studies65 have shown less recurrence and long-term disease control in patients treated with double ASCT.74 However, relapse remains the main reason for treatment failure after ASCT, due to the presence of non-detectable residual disease in the patient, the graft, or both.8 Conversely, allogeneic stem cell transplantation (alloSCT) is a potentially curative alternative, offering a tumor-free graft along with the benefit of a graft-versus-MM (GvM) effect.9 Nevertheless, the role of alloSCT in MM patients is still controversial.1210 Several studies have shown elevated rates of molecular remission after myeloablative conditioning regimen (MAC) alloSC.13 This regimen is characterized by high morbidity and mortality, especially in elderly patients with co-morbidities due to previous treatments, for whom less intensive conditioning regimens are preferable. The use of reduced intensity conditioning (RIC) extends the indication of hematopoietic stem cell transplantation to a higher number of patients who may benefit from an aggressive, but less toxic therapy than MAC, while maintaining the GvM effect. However, the efficacy of RIC on MM outcome remains uncertain. Different studies comparing tandem ASCT to alloSCT/RIC have given discordant results, with some of them failing to demonstrate the benefit of alloSCT/RIC.151410 There is, therefore, a general hesitancy in recommending up-front alloSCT. Consequently, most of the time alloSCT is offered as a salvage therapy post-ASCT relapse or for refractory disease. Outcomes after matched unrelated donor (MUD) transplantations for MM have been reported to be similar to those with HLA identical siblings; however, outcomes after MUD appear to be associated with higher non-relapse mortality (NRM).16 Little is known of the use of other alternative donors, such as haploidentical or cord blood in patients with MM.1917 We performed a registry-based study to evaluate risk factors and outcomes of patients undergoing umbilical cord blood transplantation (UCBT) with the aim of analyzing the role of this stem cell source in patients with plasma cell disorders.

Methods

Study design, inclusion criteria and data collection

This is a retrospective observational registry-based study using Eurocord/EBMT data.

Patients over 18 years of age and diagnosed with MM or plasma cell leukemia (PCL) receiving single or double UCBT (dUCBT) between 2001 and 2013 were included. Exclusion criteria were: previous alloSCT, primary amyloidosis without MM, manipulated cord blood, intra-bone injection of cord blood cells or cord blood transplants associated with another stem cell source.

All patients gave informed consent for research. The study was conducted in accordance with the Declaration of Helsinki. The Internal Review Board of Eurocord-EBMT approved the study.

End points and definitions

The primary end point was progression-free survival (PFS) defined as time from UCBT to progression, relapse or death from any cause, whichever occurred first. Secondary end points were neutrophil and platelet recovery, acute and chronic graft-versus-host disease (GvHD), NRM, relapse incidence (RI) and overall survival (OS). OS was defined as time from transplant to death from any cause. Neutrophil (PMN) engraftment was defined as the first of three consecutive days with an absolute neutrophil count of 0.5×10/L or over, without evidence of autologous reconstitution. Platelet (PLT) engraftment was defined as the first date at which an unsupported platelet count of 20×10/L or over for seven consecutive days was achieved. MAC regimen was defined as a regimen containing total body irradiation (TBI) with a dose of more than 6 Gy or a dose of more than 8 mg/kg oral or more than 6.4 mg/kg intravenous busulfan or chemotherapy combination containing more than 10 mg/kg thiotepa. Response to treatment was defined according to standard criteria.20 Chemo-refractory myeloma was defined as progression or non-response within 60 days of last therapy. GvHD was evaluated based on standard criteria.2221 For dUCBT, human leukocyte antigen (HLA) degree of matching was defined considering the UCB unit with the higher number of disparities with the recipient. High-risk cytogenetic abnormalities included at least one of the following: del17p, t(4;14) or t(14;16) performed by fluorescence in situ hybridization (FISH) or conventional metaphase cytogenetics, according to the policy of each center.

Statistical analysis

The probabilities of PFS and OS were estimated using the Kaplan-Meier method and compared with the log rank test. In the case of no event, observations were censored at the time of last follow up. Cumulative incidence (CI) was calculated in a competing risk setting. Death without an event was treated as a competing risk to calculate probabilities of neutrophil and platelet engraftment, acute and chronic GvHD. Death without progression or relapse was considered as competing risk for RI and relapse was the competing event for NRM. P<0.05 was considered statistically significant. All variables found to have P<0.10 in the univariate analysis were included in a Cox model for PFS and OS, or in a Fine and Gray proportional hazard regression model for engraftment, GvHD, NRM and relapse. Analysis was performed with SPSS 19 and SPLUS software.

Results

Patients’ and transplant characteristics are summarized in Tables 1 and 2. A total of 95 patients with a median follow up of 41.3 (range 3.7–96) months met the inclusion criteria for the study. Median age at UCBT was 53.3 years (range 24.1–69.6) and median body weight was 70 kg (range 48–110). Median time from diagnosis to UCBT was 41.6 months (range 4.6–235.6). Diagnosis was MM for 85 (90%) and PCL for 10 (10%) patients. The immunoglobulin (Ig) subtype was IgG in 39 (46%), IgA in 23 (27%) and IgD in one case. Light chain myeloma accounted for 24% of patients, non-secretory for 2%, and the isotype was unknown in 9 patients. Twelve patients (17%) had chemo-refractory disease. Nearly all patients (96%) received at least one ASCT before UCBT: 26 (30%) received a planned tandem auto-auto and 18 (20%) a tandem procedure which included the current UCBT transplantation. The remaining 45 patients received one or more ASCT, but not as part of a planned tandem procedure. Only 4 patients did not receive a previous ASCT: 3 of them had PCL and received UCBT as first-line therapy in a median time of 5.5 months (range 4–6) from diagnosis; one had MM and received a UCBT after relapse at five years from diagnosis. Cytogenetic analysis was performed and available in 45 patients and was abnormal in 32 of them. The most frequent alteration was del13q (n=17). High-risk abnormalities [del17p or t(4;14)] were present in 11 patients. Ten patients were in first complete remission (CR) at UCBT, 10 in second CR, 20 in very good partial response (VGPR), 37 in partial response (PR), 14 in stable or progressive disease, and data were missing for the remaining 4 patients. Eighty-two patients received proteasome inhibitors or immunomodulatory drugs before UCBT. Among 43 patients with available information on HCTI-CI, 21 were reported as HCTI-CI 0, 4 HCTI-CI 1, 13 HCTI-CI 2 and 5 HCTI-CI 3.

Table 1.Patients’ characteristics.

Table 2.Transplant characteristics.

Fifty-nine patients (62%) received a dUCBT. The majority of patients were conditioned with an RIC regimen (n=77, 82%). The most common conditioning regimen was cyclophosphamide+fludarabine+TBI (2–6 Gy) (64%) and antithymocyte globulin (ATG) was given to 24% of the patients. Cyclosporine A (CSA)+mycophenolate mofetil (MMF) was the most frequent GvHD prophylaxis (80%).

The median number of total nucleated cells (TNC) was 4.24×10/kg (range 2.2–7.8) at cryopreservation, and 3.3×10/kg (range 0.8–7.8) at infusion. The median number of cryopreserved and infused CD34 cells was 1.78×10/kg (range 0.5–6.6) and 1.25×10/kg (range 0.1–4.5), respectively. The majority of patients (68%) received a graft with 2 HLA mismatches. Among 63 patients with available information on maintenance therapy, 3 were treated with lenalidomide after UCBT.

Summaries of the univariate and multivariate analyses for major outcomes are shown in Tables 3 and 4, respectively.

Table 3.Univariate analysis of main transplant outcomes.

Table 4.Multivariate analysis.

Engraftment and GvHD

The CI of 60-day PMN and 180-day PLT engraftment were 97%±3% and 72%±5%, respectively. The median time of PMN and PLT engraftment were 20 (range 7–53) and 33 (range 8–98) days, respectively. Seven patients failed to achieve PMN engraftment; of these, 4 died within a median time of 21 months after UCBT. Three patients who experienced graft failure were alive at last follow up, 2 after an autologous rescue.

The CI of 100-day acute GvHD (aGvHD) grade II-IV and grade III-IV were 41%±5% and 16%±4%, respectively. aGvHD was lower in patients who received ATG (18% vs. 48%; P=0.02) and in those who did not receive TBI (15% vs. 49%; P<0.001). The use of ATG was associated with significant lower incidence of aGvHD in multivariate analysis (HR 0.25, 95%CI: 0.08–0.80; P=0.020).

The CI of chronic GvHD (cGvHD) at two years was 22%±4%, with a median time of onset of 188 days. Among the 23 patients who experienced cGvHD, 11 were alive at last follow up and 9 were disease free. Extensive cGvHD was observed in 5 patients (5 of 23). Patients who underwent dUCBT had a higher incidence of cGvHD than those receiving sUCBT in univariate analysis (30% vs. 9%; P=0.015).

Non-relapse mortality and relapse incidence

The CI of NRM at three years was 29%±5% (Figure 1). Overall, 63 patients died: 30 of relapse and 33 of transplant-related causes (infections, n=16; GvHD, n=5; other causes, n=12). In univariate analysis, ATG use (52% vs. 22%; P=0.004), MAC conditioning (54% vs. 23%; P=0.01) and TBI (51% vs. 22%; P=0.005) were associated with higher incidence of NRM. The use of ATG was independently associated with higher NRM in the multivariate analysis (HR 3.35, 95%CI: 1.44–7.81; P=0.005).

Figure 1.3-year non-relapse mortality and relapse incidence. Solid line represents relapse incidence; dashed line represents non-relapse mortality incidence.

The CI of relapse at three years was 47%±5% (Figure 1). The RI was higher in chemo-refractory MM (75% vs. 45%; P=0.05). Moreover, in multivariate analysis patients with high cytogenetic risk had higher RI (HR 3.83, 95%CI: 1.26–11.61; P=0.018).

Overall survival and progression-free survival

The median follow up for survivors was 41 months (range 3.7–96). The 3-year probability of PFS and OS was 24%±5% and 40%±5%, respectively (Figures 2 and 3). In univariate analysis, RIC regimen and CsA+MMF as GvHD prophylaxis were associated with improved OS (43% vs. 30%, P=0.05, and 45% vs. 14%, P<0.001, respectively). Conversely, the use of ATG was associated with a decreased survival (10% vs. 46%; P<0.001). The effect of ATG use retained significance in multivariate analysis, with decreased OS (HR 4.03, 95%CI: 2.13–7.64; P<0.001) and PFS (HR=2.73, 95%CI: 1.48–5.05; P=0.001). Moreover, in multivariate analysis patients with high-risk cytogenetic had poorer OS (HR 2.99, 95%CI: 1.31–6.83; P=0.009) and PFS (HR 2.88, 95%CI: 1.26–6.57; P=0.012).

Figure 2.3-year overall survival.

Figure 3.3-year progression-free survival.

Discussion

We conducted a registry-based study with the objective of defining the role of UCBT in patients with plasma cell disorders. AlloSCT is not a standard treatment for patients with MM, and transplantation with alternative stem cell sources, such as UCBT, is even less common. The results of the current study suggest that UCBT is a feasible alternative for MM patients, and that high-risk cytogenetics and the use of ATG are independently associated with worse survival.

In the recent guidelines from the American Society for Blood and Marrow Transplantation on the indications for ASCT and alloSCT, the former was considered “standard of care” for MM patients in initial response or in sensitive relapse, and is considered “standard of care with clinical evidence” in refractory MM and PCL. On the other hand, alloSCT is still considered “developmental” for MM in initial response, but for patients in other disease stages or PCL, it is accepted as “standard of care with clinical evidence”.23 These recommendations do not take into account other factors such as age, comorbidities, donor source, and HLA incompatibilities. Overall, results of alloSCT are poor because of the high transplant-related mortality and high risk of relapse. The EBMT reported 3-year OS and PFS of 41% and 21%, respectively, in 229 MM patients who received RIC alloSCT from related and unrelated donors.24 To date, only a few studies on the use of UCBT in MM have been published, and they were mostly isolated cases.1817 Recently, a more comprehensive survey was reported by the Japanese registry19 in 86 patients with MM, showing 6-year OS and PFS of 15.2% and 13%, respectively. However, it has been shown in several publications that results are not always similar for Japanese and Western populations.25 Moreover, our study differs from the previous publication because it includes both single and double UCBT and uses a different classification for high-risk cytogenetics.

In our series, in which 82% of patients received RIC regimen, OS and PFS were 40% and 24% at three years, respectively. Although the number of previous therapy lines is not available, 90% of patients were transplanted beyond CR1, indicating that UCBT was not the first-line therapy for these patients. Our results are comparable to those observed in MM patients undergoing RIC-alloSCT with other stem cell sources, not only for PFS and OS, but also NRM (29%).24

We observed a detrimental impact of adverse karyotype in PFS and OS in multivariate analysis. Other authors have previously shown the negative impact of high-risk abnormalities, such as t(4;14), t(14;16), t(14;20) and del17p, on survival outcomes in patients with newly diagnosed MM.2726 Similar findings were demonstrated in patients receiving front-line ASCT, in which high-risk cytogenetics was associated with worse outcomes28 and unsustained CR at one year.29 The prognostic impact of adverse cytogenetics on alloSCT outcome in MM is not well established. Schilling et al. showed that alloSCT can be beneficial for patients with t(4;14), but not for those with del17p.30 On the contrary, Roos-Weil et al. demonstrated that the increased risk associated with either of these mutations could be overcome with alloSCT.31 More recently, the benefit of alloSCT for patients with MM harboring both t(4;14) and del17p was confirmed in a prospective tandem auto/RIC-alloSCT protocol.32 Contrary to the current results and other previous publications,27 a recent study on UCBT19 showed no association between high-risk cytogenetic and poor outcomes. A possible explanation for these findings might be the different risk group classification of patients harboring del13q.

In our series, ATG use was associated with lower OS, PFS and aGvHD, and with higher NRM. This was also reported, recently, in a larger series of patients with hematologic malignancies undergoing UCBT after RIC regimen.33 As described in previous studies, immunosuppression with ATG is associated with a high incidence of infections.34 In our series, infection was the primary cause of transplant-related deaths among patients who received ATG (n=22). We were unable to identify any significant association between the impact of disease status at UCBT and planned tandem transplantation on MM outcomes, as suggested by the Japanese group.19 The lack of association of these factors on UCBT outcomes may be related to the low number of patients included in the categories of some specific variables or to actual differences between the populations in the different studies.

In our series, RI was high (47%), but comparable to results reported with other stem cell sources (bone marrow and peripheral blood stem cell) in previous alloSCT for MM studies.

Strategies to prevent relapse after UCBT may include post-transplant consolidation and/or maintenance with immunomodulatory drugs or proteasome inhibitors.2

Thalidomide has been investigated as salvage therapy in 31 patients35 and at low doses with DLI in 18 patients,36 both after alloSCT. Lenalidomide is already used after ASCT, but its use is still controversial after alloHSCT.37 However, it may be beneficial, especially for high-risk MM,38 as it has been demonstrated to improve response rate and to increase T-cell activity.39 Bortezomib has been used in relapsed MM after RIC alloSCT (n=18) with a certain level of toxicity40 and in patients not responding to DLI.41 However, the application of novel agents in the UCBT setting and their potential in intensifying the GvMM effect after transplant ought to be further explored. In fact, there are several ongoing prospective studies (clinicaltrials.gov identifiers: 02440464, 020308280, 01460420, 01131169, 02447055) including anti-myeloma drugs as maintenance therapy early after alloSCT that may improve outcomes of MM patients.42 One study in particular (clinicaltrials.gov identifier: 02440464) will investigate the use of a 2-generation anti-myeloma drug (ixazomib) in association with immunosuppressive therapy after alloSCT.

Unfortunately, only 3 patients in this series were reported to have received maintenance therapy, therefore we were unable to evaluate such strategies. Despite some limitations intrinsic to the retrospective nature of our study, we have demonstrated that UCBT is a feasible option for MM patients needing alloSCT.

Furthermore, the clinical applications for UCBT are still evolving. Several methods, such as the combination of cord blood and CD34 selected haploidentical graft, the addition of mesenchymal stem cells, cord blood intrabone infusion and ex vivo expansion techniques are under investigation to improve engraftment.4543 However, further studies are needed to determine the potential benefit of these innovative strategies. Also, the use of haploidentical transplantation may deserve to be investigated to determine its applicability in this setting.

The place of alloSCT, including UCBT, is still unclear, but progress may be expected with a better identification of high-risk criteria, and a co-ordinated sequential approach with new drugs and transplant strategies.

Footnotes

• ↵* AP and EX, and NF and MM contributed equally to this work.
• Check the online version for the most updated information on this article, online supplements, and information on authorship & disclosures: www.haematologica.org/content/101/9/1120
• FundingMM was supported by educational grants from the “Association for Training, Education and Research in Hematology, Immunology and Transplantation” (ATERHIT). M Mohty would like to thank Prof. Junia V. Melo (University of Adelaide, Australia, and Imperial College, London) for critical reading of this manuscript.

• Received November 13, 2015.
• Accepted May 24, 2016.