Biol Blood Marrow Transplant 25 (2019) 17031712 Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org Immunotherapy Response to Novel Drugs before and after Allogeneic Stem Cell Transplantation in Patients with Relapsed Multiple Myeloma pez-Corral1,*, Teresa Caballero-Vela Lucia Lo zquez2, Oriana Lo pez-Godino3, Laura Rosin~ ol4, Sabina Perez-Vicente , Francesc Fernandez-Avile 2 s , Isabel Krsnik , Daniel Morillo , 4 5 5 Inmaculada Heras3, Mireia Morgades6, Jose J. Rifon7, Antonia Sampol8, Francisca Iniesta9, Enrique-María Ocio1, Jesu  s Martin2, Montserrat Rovira4, Martín Cabero5, Cristina Castilla-Llorente10, Josep-María Ribera , Marta Torres-Juan8, Jose María Moraleda9, Carmen Martinez4, 6 Alejandro Vazquez5, Gonzalo Gutierrez4, Dolores Caballero1, Jesu  s F. San Miguel7, 1 María-Victoria Mateos , Jose Antonio Pe rez-Simon 2,11 1 Hematology Department, Complejo Asistencial Universitario de Salamanca-IBSAL, Centro de Investigacio ncer-IBMCC, Spain n del Ca 2 Hematology Department, Instituto de Biomedicina de Sevilla, IBiS/Hospital Universitario Virgen del Rocío/CSIC/CIBERON/Universidad de Sevilla, Spain 3 Hematology Department, Hospital Universitario Morales Meseguer y Centro Regional de Hemodonacio n, IMIB, Universidad de Murcia, Spain 4 Department of Hematology, Amyloidosis and Myeloma Unit, Hospital Clínic, IDIBAPS, Barcelona, Spain 5 n Puerta de Hierro Majadahonda, Madrid, Spain Hematology Department, Instituto de Investigacio 6 Department of Hematology, ICO Badalona-Hospital Germans Trias i Pujol, Josep Carreras Leukemia Research Institute, Universitat Auto noma de Barcelona, Badalona, Spain 7 Hematology Service, Clínica Universidad de Navarra, CIMA, IDISNA, CIBERONC, Pamplona, Spain 8 Hematology Department, Hospital Son Espases, Palma de Mallorca, Spain 9 Hematology Department, University Hospital of Virgen de la Arrixaca, Murcia, Spain 10 Hematology Department, Gustave Roussy Cancer Center, Villejuif, France Article history: A B S T R A C T Received 30 November 2018 Multiple myeloma (MM) remains as an incurable disease and, although allogeneic hematopoietic stem cell trans- Accepted 30 April 2019 plantation (allo-HSCT) is a potentially curative approach, most patients ultimately relapse, and their treatment remains challenging. Because allo-HSCT can modify not only the biology of the disease, but also the immune sys- Key Words: tem and the microenvironment, it can potentially enhance the response to rescue therapies. Information on the Multiple myeloma efficacy and safety of novel drugs in patients relapsing after allo-HSCT is lacking, however. The objectives of this Allogeneic hematopoietic stem study were to evaluate the efficacy and toxicity of rescue therapies in patients with MM who relapsed after allo- cell transplantation HSCT, as well as to compare their efficacy before and after allo-HSCT. This retrospective multicenter study Allo-HSCT included 126 consecutive patients with MM who underwent allo-HSCT between 2000 and 2013 at 8 Spanish cen- Proteasome inhibitor ters. All patients engrafted. The incidence of grade II-IV acute graft-versus-host disease (GVHD) was 47%, and non- Immunomodulatory drug relapse mortality within the first 100 days post-transplantation was 13%. After a median follow-up of 92 months, overall survival (OS) was 51% at 2 years and 43% at 5 years. The median progression-free survival after allo-HSCT was 7 months, whereas the median OS after relapse was 33 months. Patients relapsing in the first 6 months after transplantation had a dismal prognosis compared with those who relapsed later (median OS, 11 months versus 120 months; P < .001). The absence of chronic GVHD was associated with reduced OS after relapse (hazard ratio, 3.44; P < .001). Most patients responded to rescue therapies, including proteasome inhibitors (PIs; 62%) and immunomodulatory drugs (IMiDs; 77%), with a good toxicity profile. An in-depth evaluation, including the type and intensity of PI- and IMiD-based combinations used before and after allo-HSCT, showed that the overall response rate and duration of response after allo-HSCT were similar to those seen in the pretransplantation period. Patients with MM who relapse after allo-HSCT should be considered candidates for therapy with new drugs, which can achieve similar response rates with similar durability as seen in the pretransplantation period. This pattern does not follow the usual course of the disease outside the transplantation setting, where response rates and time to progression decreases with each consecutive line of treatment. © 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. * Correspondence and reprint requests: Lucía Lo pez Corral, MD, PhD, Hospital Universitario de Salamanca, Paseo San Vicente 58-182, 37007 Salamanca, Spain. pez-Corral). E-mail address: lucialopezcorral@usal.es (L. Lo 11 CIBERONC CB16/12/00480, on behalf of the GETH (Grupo Espan ~ ol de Trasplante Hematopoyetico) and GEM (Grupo Espan ~ ol de Mieloma) https://doi.org/10.1016/j.bbmt.2019.04.026 1083-8791/© 2019 American Society for Transplantation and Cellular Therapy. Published by Elsevier Inc. 1704 pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo INTRODUCTION The present study evaluated the outcomes of a series of 126 Multiple myeloma (MM) is a neoplastic plasma cell disorder patients with MM who underwent allo-HSCT. After a median characterized by clonal proliferation of malignant plasma cells follow-up of 92 months, overall survival (OS) was 51% at in the bone marrow and the presence of a monoclonal protein 2 years and 43% at 5 years, with 10% of patients alive at longer in the blood and/or urine, resulting in myeloma-defining than 10 years beyond transplantation. The median OS after events. The outcomes of patients with MM have improved sig- relapse was 33 months. Most patients responded to rescue nificantly due to the introduction of novel agents in both the therapy including PIs (62%) and IMiDs (77%) with good tolera- relapse and upfront settings; however, the disease remains bility. Moreover, this is the first study to compare the efficacy incurable for most patients. The role of allogeneic hematopoi- and toxicity of novel drug/novel combinations before and after etic stem cell transplantation (allo-HSCT) in treating MM allo-HSCT, and we show that they can be safely used and that remains challenging, because although it is a potentially cura- the response rate and duration of response are similar to those tive approach [14], the associated high toxicity and relapse observed in the pretransplantation setting, indicating that rate are important concerns. In this regard, the International allo-HSCT can potentially sensitize myelomatous plasma cells, Myeloma Working Group together with the Blood and Marrow modifying the usual course of MM. Transplant Clinical Trials Network, the American Society of Blood and Marrow Transplantation, and the European Society METHODS of Blood and Marrow Transplantation (EBMT) have agreed that Patients allo-HSCT should be considered an appropriate therapy for all We report a retrospective multicenter analysis of 126 consecutive eligible patients with early relapse (occurring within patients with MM who underwent allo-HSCT between January 2000 and December 2013 at 8 Spanish centers. The institutional Ethics Committees of 24 months) after primary therapy that included autologous all participating centers approved the study, and all patients provided written stem cell transplantation (ASCT) and/or high-risk features, informed consent before entering the study in accordance with the Declara- while acknowledging that prospective randomized trials are tion of Helsinki. All but 2 procedures were done as routine clinical practice needed to define the role of allo-HSCT in patients who relapse outside of an investigational clinical trial. after primary therapy for MM [5]. Allo-HSCT offers a graft-versus-myeloma effect that may be Disease Response and Definitions agnostic to cytogenetic risk. The unacceptably high morbidity Response to treatment and progression were determined according to and nonrelapse mortality (NRM), related at least in part to the the criteria formulated by the International Myeloma Working Group [25]. Complete response (CR) was defined as negative immunofixation in serum use of myeloablative conditioning regimens [69], have and urine and <5% plasma cells in bone marrow aspirate, including the disap- resulted in a shift toward reduced-intensity conditioning (allo- pearance of any soft tissue plasmacytoma. Progression-free survival (PFS) was RIC) regimens [1016]. Despite this, the decision to proceed to defined as the time from allo-HSCT to the date of progression or death from allo-HSCT is increasingly challenging with the advent of new any cause or last follow-up. Patients alive without progression at their last therapies. However, although the role of allo-HSCT is a subject follow-up were censored. OS was defined as the time from allo-HSCT to the date of death or last follow-up. Patients alive at their last follow-up were cen- of intense debate, survey studies by the EBMT have shown an sored. OS after relapse was defined as the time from relapse after allo-HSCT increasing number of patients undergoing allo-HSCT for MM to the date of death or last follow-up. NRM was defined as death without pre- [17], and in fact, the use of novel drugs does not appear to vious occurrence of relapse or progression. Relapse incidence (RI) was calcu- have impacted the rate of transplantation [18]. lated from the date of allo-HSCT to the date of relapse or progression. NRM and RI were considered competing events. Despite the high response rate and the reduced NRM Deaths due to infection (bacterial, fungal, or viral) in the context of active observed in the years after the advent of allo-RIC regimens, graft-versus-host disease (GVHD) were considered related to GVHD. Acute most patients ultimately relapse. Thus, relapse and progression GVHD (aGVHD) was defined as grade II-IV according to the Seattle criteria are the main causes of treatment failure. Treatment of patients [26]. Chronic GVHD (cGHVD) was defined as mild, moderate, or severe according to National Institutes of Health 2005 criteria [27]. Patients who had who relapse after allo-HSCT remains a challenge. Until evidence of engraftment were evaluable for aGVHD, whereas patients who recently, these patients have not been considered candidates engrafted and survived longer than 100 days after allo-HSCT were evaluable for treatment with experimental drugs, based mainly on con- for cGVHD, with death or progression/relapse without cGVHD as competing cerns regarding the toxicity profile of these drugs in the trans- events. plantation setting. Moreover, most of the recent trials that have led to the approval of new drugs excluded allo-HSCT Statistical Analysis recipients, and thus these drugs’ efficacy has not been appro- The co-primary endpoints of this study were (1) to evaluate the efficacy priately evaluated in this setting. Several small studies have and safety of rescue regimens based on PIs and IMiDs in patients with MM been reported describing the efficacy of proteasome inhibitors who relapsed after allo-HSCT and (2) to compare the efficacy after allo-HSCT with that observed before allo-HSCT for the same patients treated with simi- (PIs) or immunomodulatory drugs (IMiDs) with or without lar combinations before and after transplantation. Secondary endpoints concomitant donor lymphocyte infusion (DLI) as salvage ther- included the efficacy of allo-HSCT in terms of response and RI, PFS, and OS. apy after allo-HSCT [1923], but none of these has compared The incidences of aGVHD and cGVHD were also analyzed. the efficacy achieved after allo-HSCT with that obtained pre- All nominal and continuous characteristics were described the usual tables and indexes; comparisons were done using standard nonparametric transplantation with similar combinations. It is well known tests, including the chi-square or Fishers exact test for categorical variables that the outcomes of patients with MM follows a pattern in and the Mann-Whitney Utest for continuous variables. Survival curves were which the response rate decreases and the median duration of estimated using the Kaplan-Meier method, with group comparison using the response decreases after each consecutive line of therapy, a log-rank test. Prognostic factors for OS after relapse were analyzed for statis- tical significance using the Cox proportional hazards model. Factors that process that ultimately leads to treatment failure and death showed a significance of P < .10 were included in a multivariate Cox regres- [24]. Whether or not this concept applies to relapse after allo- sion model: age, number of lines before allo-HSCT, extramedullary disease at HSCT has not been evaluated, and considering the changes to allo-HSCT, extramedullary disease at relapse, aGVHD, and cGVHD. The proba- the immune system and microenvironment induced by the bilities of disease progression and GVHD were calculated using cumulative incidence estimates. NRM and RI were analyzed and compared using Gray’s allograft, this scenario might represent a good opportunity to test. Cumulative incidence was computed with the cmprsk package for R ver- evaluate the efficacy of novel agents before and after trans- sion 2.14.0 (R Institute for Statistical Computing, Vienna, Austria), and other plantation. analyses were performed using SPSS 19.0 (IBM, Armonk, NY). pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo 1705 RESULTS 92 months (range, 22 to 197 months). In the pretransplantion Patient Characteristics, Conditioning Regimens, and GVHD period, 71 patients (56%) had been treated with a PI-containing Prophylaxis regimen and 48 (38%) had received an ImiD-containing regi- Baseline and transplantation characteristics of the 126 men. Thirty-six patients (28.5%) received both agents. Details patients with MM who underwent allo-HSCT between January of the PI- and ImiD-based regimens are provided in Table 2. 2000 and December 2013 are summarized in Table 1. The median duration of follow-up for surviving patients was Engraftment, GVHD, and NRM All 126 patients engrafted. The incidence of grade II-IV aGVHD was 47% (grade III-IV, 22%) and the 3-year cumulative Table 1 Baseline Patient and Transplantation Characteristics (N = 126) incidence of cGVHD was 51% (moderate, 14%; severe, 12%). NRM within the first 100 days post-transplantation was 13%, Characteristic Value whereas the global NRM at any time was 31.7%. More informa- Age, yr, median (range) 52 (22-66) tion is provided in Table 3. Male sex, n (%) 78 (62) IgG/IgA/Bence Jones/nonsecretory, 68 (54)/23 (18)/31 (25)/4 (3) Overall Outcomes after Transplantation n (%) The overall response rate (ORR) after allo-HSCT was 74%, ISS I/II/III, n (%)* 20 (28)/26 (37)/25 (35) with 56% of patients achieving CR. At day +100 after transplan- High cytogenetic risk: (t(4;14), t 19 (29) tation, 82 patients (84.5%) either remained stable or showed (14;16), or 17p del), n (%)y improved disease status compared with their status at trans- Number of treatment lines before 3 (1-9) plantation, including 13 patients with progressive or stable allo-HSCT, median (range) disease at transplantation. Seventy-five patients (59.5%) 1, n (%) 33 (26.2) relapsed after allo-HSCT, 57 of them with extramedullary 2-3, n (%) 62 (49.2) involvement, with a median time to relapse or progression of 8 >3, n (%) 31 (24.6) months post-transplantation (range, 1 to 141 months). The PIs, n (%) 71 (56) cumulative incidences of NRM and relapse after allo-HCT are IMiDs, n (%) 48 (38) shown in Figure 1. Previous ASCT, n (%) 105 (83) After a median follow-up of 92 months for surviving Time between diagnosis and allo- 30 (8-130) patients (range, 22 to 197 months), 85 patients (67.4%) died, HSCT 44 due to relapse or progression and 40 due to NRM. PFS was Allo-HSCT indication, n (%) 24% at 2 years and 18% at 5 years, with a median PFS of 7 First response 15 (12) months (Figure 2). Sensitive relapse 66 (52.3) Remarkably, even among patients who relapsed after allo- Refractory disease 45 (35.7) HSCT, time to relapse discriminates subgroups of patients with Disease status at allo-HSCT, n (%)z very different outcomes: patients who relapse <6 months, 6 to CR 16 (13) 24 months, and >24 months after transplant have a median OS VGPR/PR 86 (68) of 11 months, 68 months and not reached, respectively Stable/progressive disease 22 (19) (P < .001). Moreover, 83% of patients who relapsed at 12 Extramedullary disease at allo-HSCT, 35 (28) months after transplantation were still alive at 5 years after n (%) allo-HSCT, with a median OS not reached. Reduced-intensity conditioning, 111 (89) n (%) Evaluation of New Drug Efficacy before and after Allo-HSCT in Conditioning regimen, n (%) All Patients Fludarabine 150 mg/m2/melphalan 60 (48) Seventy-one patients (56%) received threatment with PIs in 140 mg/m2 the pretransplantation period, and 35 (27.7%) did so in the Fludarabine 90 mg/m2/TBI 2 Gy 16 (13) post-transplantation period. The ORR after treatment with a Fludarabine/melphalan/ 11 (9) PI-based scheme was superior in the pretransplantation period bortezomib than in the post-transplantation period (76% versus 65.7%; T cell depletion 20 (16) P > .05), with a longer time to progression (TTP) (13 months Others 19 (14) versus 7 months; P > .05), although these differences did not Unrelated donor/HLA-matched 24 (19)/11 (91) reach statistical significance. However, an in-depth evaluation (10/10) (Table 2) shows that the PI-based schemes used in the pre- Peripheral blood stem cell source 111 (89) transplantation period were more intensive than those used in CD34+ cells infused per kg, £ 106, 4.8 (.5-14.2) the post-transplantation period, including ASCT with or with- median (range) out maintenance in 21% of patients (n = 15) and high-intensity GVHD prophylaxis, n (%) combination chemotherapy regimens (VDLPACE/VDTPACE) in Calcineurin 65 (56) 15% (n = 11), whereas the majority of patients treated with PIs inhibitor + methotrexate post-HSCT received PI in monotherapy with or without ste- Cyclosporine A + mycophenolate 28 (23) roids (n = 21; 60%). Comparing response to PIs including equiv- mofetil alent schemes in terms of intensity (Table 2), the ORR to Others 33 (26) bortezomib with or without steroids was higher in the pre- Follow-up post-allo-HSCT, mo, 92 (22-197) transplantation period than in the post-transplantation period median (range) (79% versus 62%; P > .05), but similar when mono/polychemo- ISS indicates Injury Severity Score; VGPR, very good partial response. * ISS data available for 71 patients. therapy was added (60% versus 63%; P> .05). In addition, 4 out y Fluorescein in situ hybridization data available for 65 patients. of 5 patients responded to bortezomib plus panobinostat, and z Disease status at allo-HSCT available for 124 patients. 1 patient responded to carfilzomib in the post allo-HSCT. 1706 pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo Table 2 New Drugs before and after Allo-HSCT Drug Pre-HSCT Post-HSCT PIs (any regimen) N = 71 N = 35 ORR, % (n/N) 76 (52/71) 65.7 (23/35) CR 21 19 PR 55 44 .05) (Table 2). sity combination chemotherapy regimens (VDTPACE/ OS was 60% at 1 year after relapse and 53% at 2 years after VDLPACE) in 23% (n = 11) (Table 2). Comparing responses to relapse, with a median OS after relapse of 33 months (range, IMiDs within equivalent schemes (Table 2), response rates 13 to 53 months) (Figure 3). In a multivariable model, the were not significantly different for thalidomide combinations, absence of cGVHD was the most significant factor associated although the ORR to lenalidomide with or without steroids with reduced OS after relapse (hazard ratio [HR], 3.48; 95% was higher in the pretransplantation period compared with confidence interval [CI], 1.6 to 7.1; P = .001). In addition, less the post-transplantation period (91% versus 72%; P > .05). Of pretreated patients (ie, number of lines before allo-HSCT, 1 note, all 4 patients who received post-allo-HSCT pomalidomide versus 2/3 versus 4) had better OS after relapse (HR, .41; 95% with or without steroids responded, as did the 2 patients who CI, .08 to .68), just like absence of grade III-IV aGVHD had a received IMiDs plus elotuzumab or panobinostat. Interestingly, (HR, .29; 95% CI, 01-06; P < .01). pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo 1707 Table 3 both the pretransplantation and post-transplantation periods. Engraftment, GVHD, and NRM (N = 126) Fifteen of these 18 patients achieved at least a partial response Variable Value (PR) (CR, n = 6) before transplantation, and 11 of these 15 Engraftment 126 (100) patients responded again when they received PIs after allo- Time to neutrophils >.5 £ 109/L, d, 16 (6-27) HSCT (CR, n = 3). Moreover, 2 of the 3 patients who were median (range) refractory to PIs before allo-HSCT responded when PIs were Time to platelets >20 £ 109/L, d, 11 (0-18) used as salvage therapy after HSCT. Finally, 5 patients who did median (range) not respond to PIs in the post-transplantation period achieved aGVHD, n (%) 68 (57) stable disease, with TTP between 4 and 8 months. Grade II-IV 59 (47) In contrast, 12 out of 75 patients who relapsed after allo- Grade III-IV 28 (22) HSCT received IMiDs in both the pretransplantation and post- cGVHD at 3 yr, n (%) 64 (51) transplantation periods. Ten of these 12 patients achieved at Mild, moderate, severe, unknown 11 (9)/13 (10)/11 (9)/29 (23) least PR pretransplantation (CR, n = 3) and, when they were NRM, n (%) retreated post-allo-HSCT, 7 responded again to IMiDs (CR, At day +100 17 (13) n = 2). Moreover, 2 patients who were refractory before allo- HSCT responded to IMiDs post-transplantation. Finally, 2 of 3 Global 40 (31.8) patients who were refractory to IMiDs in the post-transplanta- Causes of death, n (%) tion period achieved stable disease, with a TTP of 8 months in GVHD 17 (42.5) one and 13 months in the other. Infections not related to GVHD 15 (37.5) Remarkably, in the group of patients who responded to Post-transplantation lymphoproli- 2 (5) either PIs or IMiDs in both the pretransplantation and post- ferative disorder transplantation periods, the time to response and TTP were Pulmonary toxicity 1 (2.5) equivalent, even though the regimens were significantly more Pulmonary hemorrhage 1 (2.5) intensive in the pretransplantation period. Outcomes and Sinusoidal obstruction syndrome 1 (2.5) treatment responses of patients who received PIs and IMiDs in Other causes 3 (7.5) both the pretransplantation and post-transplantation periods are summarized in Table 4. Toxicity of PIs and IMiDs after Allo-HSCT Evaluation of New Drug Efficacy before and after Allo-HCST in Regarding salvage treatment toxicity, 14 of 33 patients Patients Receiving the Same Drug before and after Allo-HSCT treated with IMiDs had a previous history of aGVHD, and 15 had To further compare drug sensitivity before and after allo- a previous history of cGVHD. GVHD reactivation was reported HSCT, we next focused only on those patients receiving PIs in 2 patients, both of whom were treated with a lenalidomide- and/or IMiDs before allo-HSCT and the same drug after trans- based regimen. One patient with a previous history of aGVHD plantation and compared the responses in the same patient and cGHVD developed late aGVHD after 3 cycles of lenalidomide before and after allo-HSCT. Eighteen out of 75 patients who plus prednisone with skin involvement, and another patient relapsed after allo-HSCT had received a PI-based regimen in with a previous history of cGVHD developed a new episode of 1.0 RELAPSE MRT 0.8 Cumulative Incidence 0.6 0.4 0.2 0.0 0 20 40 60 80 100 120 months Figure 1. Cumulative incidences of NRM and relapse after allo-HSCT. 1708 pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo Figure 2. PFS and OS. moderate cGHVD with lichenoid mouth and ocular involvement after transplantation, but also after relapse. In this regard, the after 2 cycles of lenalidomide plus prednisone. In contrast, 19 of immune profile of patients with MM has been characterized 35 patients treated with PIs had a history of previous aGVHD, in detail, and immune exhaustion, including such abnormali- and 17 had a history of previous cGVHD. No GVHD flares were ties as a significant increment of TCR-gd+ T lymphocytes and reported. In addition, 2 patients developed grade 2 liver toxicity, Tregs, altered distribution of BDCA-1+ myeloid dendritic cells and 4 patients developed grade 3-4 peripheral neuropathy; only and tissue macrophages, and down-regulation of activation 1 patient needed to discontinue bortezomib. markers in T lymphocytes, have been noted, all of which have Hematologic toxicity was reported in 6 patients and was been correlated with a poorer outcomes outside of the trans- related to low neutrophil and platelet counts (thrombocytopenia plantation setting [31]. and neutropenia in 4 patients and grade 2 thrombocytopenia in 2 Accordingly, even after disease relapse, allo-HSCT might patients. represent an appropriate platform for the use of rescue treat- ments based on novel drugs. Unfortunately, most of the recent DISCUSSION trials that have led to the approval of new drugs excluded allo- Allo-HSCT is usually relegated to advanced lines of treat- HSCT recipients, and thus their efficacy has not been appropri- ment based on the assumption that once relapse has occurred ately evaluated in this setting. Moreover, neither the EBMT nor after transplantation, no other rescue therapy is appropriate the Center of International Blood and Marrow Transplant for these patients, due to either ineffectiveness or toxicity. Research retrospective study, in which the effects of new drugs Nevertheless, the present study, as well as several previous as salvage therapy after allo-HSCT were analyzed, compared studies, have challenged this concept. Based on findings the efficacy of these drugs with that seen in the pretransplan- reported by the EBMT [13], OS from the time of first relapse/ tation setting. Thus, Coman et al [19] evaluated the efficacy of progression was superior in patients treated with ASCT/allo- lenalidomide with or without dexamethasone in 52 patients RIC compared with patients treated with ASCT alone (50% who relapsed after allo-RIC and reported an ORR of 83%, versus 27% at 60 months; P = .003). The same finding was including 29% in CR. Of note, the development of GVHD was recently reported by the Center of International Blood and associated with the response rate [19]. Similarly, Minnema et Marrow Transplant Research; although both cohorts (ASCT/ al [20] administered lenalidomide with or without dexametha- ASCT and ASCT/allo-HSCT) had a similar risk of death in the sone to 13 patients who relapsed after allo-HSCT. The ORR was first year after relapse (HR, .72; P= .12), the ASCT/allo-HSCT 46% for lenalidomide alone and was increased to 87.5% with a group had superior OS (HR for death in ASCT/ASCT, 1.55; combination of lenalidomide and dexamethasone. Bensinger P= .0052) beyond 12 months after relapse [28]. This observa- et al [23] evaluated the response to and tolerability of lenalido- tion might reflect the fact that allo-HSCT can modify the biol- mide monotherapy in 18 patients with MM who progressed or ogy of the disease, directly targeting the myelomatous relapsed after allo-HSCT. In their study, lenalidomide resulted plasma cells as well as the microenvironment, and the well- in extended disease control (>12 months) in 50% of patients. documented graft-versus-myeloma effect induced by reactive Montefusco et al [21] reported a PFS of 31% and OS of 73% allogeneic T cells may persist after relapse and contribute to using the combination of bortezomib/dexamethasone plus DLI an enhanced disease response [29,30]. In line with this obser- in patients with relapsed MM. El Cheikh et al [22] treated 37 vation, our analysis showed that the presence of cGVHD was patients with progressive or residual disease after allo-RIC the most important favorable factor predicting OS not only using bortezomib with (n = 26) or without (n = 11) pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo 1709 Figure 3. OS after relapse. dexamethasone and achieved am ORR of 73% with 65% OS at [22] reported no worsening of GVHD symptoms in patients 18 months [22]. treated with bortezomib. In our series, none of the patients In our study, most patients responded to rescue therapies who received PIs post-HSCT developed GVHD, and only 1 including IMiDs (76%) and PIs (65.7%), in agreement with pub- out of the 33 patients with MM discontinued bortezomib lished data. In addition, we had the opportunity to evaluate treatment due to grade 3 neuropathy. Moreover, overall, the efficacy of different drug combinations used in the same our study shows that both the hematologic and nonhema- patient before and after allo-HSCT, representing a unique tologic toxicity profile observed with PIs and/or IMiDs post- opportunity to evaluate whether or not allo-HSCT modifies HSCT is not significantly different from that reported out- drug sensitivity in relapsed patients and, to the best of our side the allo-HSCT setting. The retrospective nature of the knowledge, the first study to evaluate drug sensitivity to the present study should be emphasized to carefully consider same drug in the same patient both before and after allo-HSCT. this toxicity profile, and prospective studies are needed to Interestingly, we observed that efficacy, in terms of response confirm these data. rate and duration of response, was at least similar to that Although the role of allo-HSCT is controversial in MM, and observed in the pretransplantation setting. Moreover, 4 its use in the current era remains a topic of debate, it is worth patients who were refractory to PIs (n = 2) or IMiDs (n = 2) mentioning that even in this series of patients who relapsed before allo-HSCT responded when PIs or IMiDs were used after allo-HSCT, long-term survival could be achieved. This was again as salvage therapy after allo-HSCT, and 7 of 9 patients especially true for patients who relapsed between 6 and 24 who did not respond to these agents after allo-HSCT achieved months after transplantation or at >24 months after transplan- stable disease, with TTP between 4 and 13 months. tation, in whom the median OS was 68 months and not In addition, information is scanty regarding the safety of reached, respectively. Moreover, 83% of patients who relapsed novel drugs used as rescue therapy after allo-HSCT. In this at 12 months after transplantation were still alive at 5 years regard, Coman et al [19] reported that 31% of patients after allo-HSCT, with a median OS not reached. Thus, for very developed new or exacerbated aGVHD after lenalidomide high-risk patients, allo-HCT could be considered as a platform treatment as salvage therapy post-HSCT. In line with this for additional therapeutic strategies. observation, Minnema et al [20] reported aGVHD in 5 out Our study is the first to compare the efficacy and toxicity of of 13 patients who received lenalidomide as a single agent novel drugs before and after allo-HSCT, and our findings show after allo-HSCT, and Bensinger et al [23] reported that 2 that they can be used safely; moreover, the new immune sys- out of 18 patients who received lenalidomide monotherapy tem may enhance the activity of these agents. Therefore, died due to GVHD. In our retrospective study, only 2 out of patients with MM who relapse after allo-HSCT should be con- the 24 patients treated with a lenalidomide-based regimen sidered candidates for receipt of novel agents, which are safe developed GVHD. Dose, timing, and concomitant medica- and allow for response rates in a similar proportion of patients tions should be considered to explain these differences. On and with similar durability as in the pretransplantation period. the other hand, PIs seem to be safe for patients with MM This pattern does not reproduce the usual course of the disease, relapsing or progressing after allo-HSCT, without unex- in which response rates and TTP decrease with consecutive lines pected organ toxicities in a prospective phase II trial of treatment, suggesting that the new immune system may con- reported by Monefusco et al [21], while El-Cheickh et al tribute to the enhanced efficacy of novel drugs. 1710 Table 4 Outcomes of Patients Treated with PIs and/or IMiDs in Both in the Pre- and Post-Allo-HSCT Settings (N = 23) Pre-Allo-HSCT Post-Allo-HSCT Patient PIs/IMiDs Response TTR, mo TTP, mo IPs/IMiDs Response TTR, mo TTP, mo Comments Schemes (Line of Treatment) Schemes 1* Bz + polychemotherapy x 4 (2’) PD - - Bz + Len + Pred x 10 (1’) Stringent CR 6 60 No grade 3-4 toxicity. Severe cGVHD after DLI. Dead 5 yr after VDTPACE x 2 (4’) PR7 2 - relapse 2* Bz/CC x 6 + ASCT (1’) VGPR 3 16 Thal x 8 (1°) VGPR 2 10 No grade 3-4 toxicity. Dead 4 yrs after relapse Len + Dx x 9 (2’) PR 4 11 Thal + CFM + Dx x 7 (2°) CR 4 12 Bz + LBH + Dx x 4 (3’) SD - 6 Bz x 4 (3°) SD - 6 VDLPACE x 3 (5°) PR 3 - Bz + Poma + Pred x 6 (4°) SD - 8 3 Bz + Dx x 3 (2’) SD - 2 Thal + CFM + Pred x 7 (1°) SD - 13 Grade 3 thrombocytopenia. Alive 44 months after relapse L. Lo PAD x 2 (3’) SD - 2 Ben + Bz + Dx x 6 (2’) PR 4 12 pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 Len + Dx x 4 (4’) PR 4 2 Thal + CFM + Dx x 8 (5’) PR 5 - 4* Thal + Dx x 12 (1’) CR 6 24 Bz x 7 (1’) CR 6 10 No grade 3-4 toxicity. Dead 4 years after relapse Bz + Dx x 5 (3’) VGPR 4 - Len + Pred x 8 (2’) PR 2 10 5* Thal/Dx x 5 + ASCT CR 7 16 Bz + Dx x 7 PR 3 7 No grade 3-4 toxicity. Alive 38 months after relapse Bz/Dx x 6 + ASCT CR 4 6 Len + Dx x 8 PR 4 12 Len + Dx x 10 CR 9 - Len + CFM x 9 PR 6 11 Bz + Poma + Dx PR 9 - 6* Bz/Dx x 6 + ASCT (1’) CR 3 24 Bz + Len + Dx x 11 (1’) PR 7 - No grade 3-4 toxicity. Alive 54 months after relapse 7* Bz + Dx x 5 (2’) CR 4 NA Bz + CFM x 6 (1’) CR 3 7 No grade 3-4 toxicity. Alive 12 months after relapse Len + Dx x 6 (2’) PR 5 - 8* Bz + Dx x 6 (2’) VGPR 3 NA Bz + Dx x 7 (1’) PR 3 8 No grade 3-4 toxicity. Dead 3 years after relapse Len + Dx x 11 (2’) PR 6 12 9* Bz + CFM + Dx x 7 (2’) PR 2 NA Bz x 6 (1’) SD 6 Grade 3 thrombocytopenia. Dead 2.5 years after relapse Thal x 6 (2’) PR 2 4 Thal + CFM + Dx x 5 (3’) PR 2 4 10* Bz + CFM + Dx x 4 (2’) PR 2 NA Bz +Thal+Dx x 2 (3’) SD - 4 Grade 3 thrombocytopenia. Dead 5 months after relapse 11* Bz/Dx x 6 + ASCT (1’) CR 4 108 Len + Dx x 9 (2’) PR 8 16 No grade 3-4 toxicity. Mild cGVHD after Len. Alive 28 months Bz + Dx x 7 (2’) PR 4 - Bz + CFM x 8 (3’) PR 3 7 after relapse 12* Thal + CFM + Dx (2’) PR 4 7 Bz (2’) PR 4 7 No grade 3-4 toxicity. Poor performance status at relapse. Bz + Dx (3’) CR 3 18 Died 1 year after relapse Len + Dx (4’) PR 4 2 PAD (5’) PR 3 - 13* Bz/Dx + ASCT (1’) CR 3 18 Bz + chemo (1’) PD - 2 No grade 3-4 toxicity. Very poor performance status at trans- Len + Dx (2’) PR 4 3 plant. Dead 4 months after transplant Bz + Len + Dx (4°) SD - 5 VDTPACE (5’) PD - 1 14 Bz + Dx (1’) PR 2 5 Bz + Mel + Pred (2’) PR 2 8 No grade 3-4 toxicity. Dead 18 months after relapse (continued) Table 4 (Continued) Pre-Allo-HSCT Post-Allo-HSCT Patient PIs/IMiDs Response TTR, mo TTP, mo IPs/IMiDs Response TTR, mo TTP, mo Comments Schemes (Line of Treatment) Schemes - 15* Bz + Dx x 4 (2’) PR 5 - Bz (1’) SD - 6 No grade 3-4 toxicity. Dead 144 months after relapse Bz + LBH + Dx (3’) SD - 4 Thal + CFM + Dx (4’) SD - 4 L. Lo pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 Len + Dx (6’) SD - 8 Thal + Elo + Dx (7’) SD 4 16* Thal + CFM + Dx x 6 (2’) CR 3 NA Bz x 7 (2’) PR 3 8 No grade 3-4 toxicity. Dead 144 months after relapse Bz + Dx x 2 (3’) PD - 2 Thal + CFM + Pred x 11 (4’) PR 4 9 17* VDTPACE x 2 (2’) PR 2 - Thal + CFM x 2 (4’) PD - - No grade 3-4 toxicity. Dead 5 months after relapse 18* Thal + CFM + Dx x 13 (4’) PR 8 27 Len + Dx (2’) PR 3 18 Mild GVHD reactivation with Len. Alive 36 months after Len + Dx x 3 (5’) PD - - relapse VDTPACE x 2 (6’) PR 2 - 19* Thal + polychemo (2’) PR 2 6 Len + Dx (2’) SD - 8 No grade 3-4 toxicity. Dead 12 months after relapse 20* Len + Dx x 6 (2’) VGPR 3 - Len + LBH + Dx x 9 (2’) VGPR 6 10 Grade 3 astenia. Alive 65 months after relapse 21* Bz + Len + Dx x 2 (2’) SD - 2 Poma + Dx x 9 (3’) PR 6 12 Thal + Elo + Dx x 3 (3’) PD - 4 No grade 3-4 toxicity. Dead 15 months after relapse VDLPACE x 3 (4’) PR 2 - 22* Bz/Dx x 6+ ASCT (1’) PR NA 8 Len + Carfilz + Dx x 3 (1’) SD NA 4 No grade 3-4 toxicity. Alive 54 months after transplant Len + Dx (2’) PR NA - Poma + Dx x 8 (2’) PR 3 8 23* Bz/Dx + ASCT (1’) CR 4 30 Bz x 1 (1’) - - - Poor performance status at transplantation. Dead after 1 Thal + CFM + Dx (2’) PR 4 - month of monotherapy with Bz Bz indicates bortezomib; PD, progressive disease; Len, lenalidomide; Pred, prednisone; Dx, dexamethasone; LBH, panobinostat; SD, stable disease; Thal, thalidomide; Elo, elotuzumab; CFM, cyclophosphamide; Poma, pomalidomide; PAD, bortezomib + adryamicin + dexamethasone; Ben, bendamustine; NA, not available; Mel, melphalan; Carfilz, carfilzomib; * Patients who received previous ASCT. 1711 1712 pez-Corral et al. / Biol Blood Marrow Transplant 25 (2019) 17031712 L. Lo REFERENCES 16. Patriarca F, Einsele H, Spina F, et al. Allogeneic stem cell transplantation in 1. Lokhorst H, Einsele H, Vesole D, et al. International Myeloma Working multiple myeloma relapsed after autograft: a multicenter retrospective Group consensus statement regarding the current status of allogeneic study based on donor availability. Biol Blood Marrow Transplant. 2012;18: stem-cell transplantation for multiple myeloma. J Clin Oncol. 2010;28: 617–626. 4521–4530. 17. Sobh M, Michallet M, Gahrton G, et al. Allogeneic hematopoietic cell trans- 2. Barlogie B, Shaughnessy J, Tricot G, et al. Treatment of multiple myeloma. plantation for multiple myeloma in Europe: trends and outcomes over Blood. 2004;103:20–32. 25 years. A study by the EBMT Chronic Malignancies Working Party. Leu- 3. Hunter HM, Peggs K, Powles R, et al. Analysis of outcome following allo- kemia. 2016;30:2047–2054. geneic haemopoietic stem cell transplantation for myeloma using mye- 18. Passweg JR, Baldomero H, Bader P, et al. Impact of drug development on loablative conditioning—evidence for a superior outcome using the use of stem cell transplantation: a report by the European Society for melphalan combined with total body irradiation. Br J Haematol. Blood and Marrow Transplantation (EBMT). Bone Marrow Transplant. 2005;128:496–502. 2017;52:191–196. 4. Kro€ ger N, Badbaran A, Zabelina T, et al. Impact of high-risk cytogenetics 19. Coman T, Bachy E, Michallet M, et al. Lenalidomide as salvage treatment and achievement of molecular remission on long-term freedom from dis- for multiple myeloma relapsing after allogeneic hematopoietic stem cell ease after autologous-allogeneic tandem transplantation in patients with transplantation: a report from the French Society of Bone Marrow and multiple myeloma. Biol Blood Marrow Transplant. 2013;19:398–404. Cellular Therapy. Haematologica. 2013;98:776–783. 5. Giralt S, Garderet L, Durie B, et al. American Society of Blood and Marrow 20. Minnema MC, van der Veer MS, Aarts T, Emmelot M, Mutis T, Lokhorst Transplantation, European Society of Blood and Marrow Transplantation, HM. Lenalidomide alone or in combination with dexamethasone is highly Blood and Marrow Transplant Clinical Trials Network, and International effective in patients with relapsed multiple myeloma following allogeneic Myeloma Working Group consensus conference on salvage hematopoietic stem cell transplantation and increases the frequency of CD4+Foxp3+ cell transplantation in patients with relapsed multiple myeloma. Biol T cells. Leukemia. 2009;23:605–607. Blood Marrow Transplant. 2015;21:2039–2051. 21. Montefusco V, Spina F, Patriarca F, et al. Bortezomib plus dexamethasone 6. Gahrton G, Tura S, Ljungman P, et al. Prognostic factors in allogeneic bone mar- followed by escalating donor lymphocyte infusions for patients with mul- row transplantation for multiple myeloma. J Clin Oncol. 1995;13:1312–1322. tiple myeloma relapsing or progressing after allogeneic stem cell trans- 7. Bensinger WI, Buckner CD, Anasetti C, et al. Allogeneic marrow transplan- plantation. Biol Blood Marrow Transplant. 2013;19:424–428. tation for multiple myeloma: an analysis of risk factors on outcome. Blood. 22. El-Cheikh J, Michallet M, Nagler A, et al. High response rate and improved 1996;88:2787–2793. graft-versus-host disease following bortezomib as salvage therapy after 8. Barlogie B, Kyle RA, Anderson KC, et al. Standard chemotherapy compared reduced-intensity conditioning allogeneic stem cell transplantation for with high-dose chemoradiotherapy for multiple myeloma: final results of multiple myeloma. Haematologica. 2008;93:455–458. phase III US Intergroup Trial S9321. J Clin Oncol. 2006;24:929–936. 23. Bensinger WI, Green DJ, Burwick N, Becker PS. A prospective study of lena- 9. Lokhorst HM, Segeren CM, Verdonck LF, et al. Partially T-cell-depleted lidomide monotherapy for relapse after Allo-SCT for multiple myeloma. allogeneic stem-cell transplantation for first-line treatment of multiple Bone Marrow Transplant. 2014;49:492–495. myeloma: a prospective evaluation of patients treated in the phase III 24. Kumar SK, Therneau TM, Gertz MA, et al. Clinical course of patients with study HOVON 24 MM. J Clin Oncol. 2003;21:1728–1733. relapsed multiple myeloma. Mayo Clin Proc. 2004;79:867–874. 10. Bruno B, Rotta M, Patriarca F, et al. A comparison of allografting with 25. Durie BG, Harousseau JL, Miguel JS, et al. International uniform response autografting for newly diagnosed myeloma. N Engl J Med. 2007;356: criteria for multiple myeloma. Leukemia. 2006;20:1467–1473. 1110–1120. 26. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft- 11. Garban F, Attal M, Michallet M, et al. Prospective comparison of autologous versus-host disease in human recipients of marrow from HL-A-matched stem cell transplantation followed by dose-reduced allograft (IFM99-03 sibling donors. Transplantation. 1974;18:295–304. trial) with tandem autologous stem cell transplantation (IFM99-04 trial) in 27. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health high-risk de novo multiple myeloma. Blood. 2006;107:3474–3480. consensus development project on criteria for clinical trials in chronic 12. Rosin ~ ol L, Pe rez-Simo  n JA, Sureda A, et al. A prospective PETHEMA study graft-versus-host disease, I: Diagnosis and Staging Working Group report. of tandem autologous transplantation versus autograft followed by Biol Blood Marrow Transplant. 2005;11:945–956. reduced-intensity conditioning allogeneic transplantation in newly diag- 28. Htut M, D'Souza A, Krishnan A, et al. Autologous/allogeneic hematopoietic nosed multiple myeloma. Blood. 2008;112:3591–3593. cell transplantation versus tandem autologous transplantation for multi- 13. Gahrton G, Iacobelli S, Bjo € rkstrand B, et al. Autologous/reduced-intensity ple myeloma: comparison of long term postrelapse survival. Biol Blood allogeneic stem cell transplantation vs autologous transplantation in mul- Marrow Transplant. 2018;24:478–485. tiple myeloma: long-term results of the EBMT-NMAM2000 study. Blood. 29. McSweeney PA, Niederwieser D, Shizuru JA, et al. Hematopoietic cell 2013;121:5055–5063. transplantation in older patients with hematologic malignancies: replac- 14. Krishnan A, Pasquini MC, Logan B, et al. Autologous haemopoietic stem-cell ing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood. transplantation followed by allogeneic or autologous haemopoietic stem- 2001;97:3390–3400. cell transplantation in patients with multiple myeloma (BMT CTN 0102): a 30. Maloney DG, Molina AJ, Sahebi F, et al. Allografting with nonmyeloablative phase 3 biological assignment trial. Lancet Oncol. 2011;12:1195–1203. conditioning following cytoreductive autografts for the treatment of 15. Freytes CO, Toro JJ, Yeh RF, et al. Safety and efficacy of targeted-dose busul- patients with multiple myeloma. Blood. 2003;102:3447–3454. fan and bortezomib as a conditioning regimen for patients with relapsed 31. Paiva B, Mateos MV, Sanchez-Abarca LI, et al. Immune status of high-risk multiple myeloma undergoing a second autologous blood progenitor cell smoldering multiple myeloma patients and its therapeutic modulation transplantation. Biol Blood Marrow Transplant. 2014;20:1949–1957. under LenDex: a longitudinal analysis. Blood. 2016;127:1151–1162.