An overview of conditioning regimens for haploidentical stem cell transplantation with post-transplantation cyclophosphamide

CRITICAL REVIEW An overview of conditioning regimens for haploidentical stem cell transplantation with post-transplantation cyclophosphamide AJH Munira Shabbir-Moosajee, 1 Lindsey Lombardi, 2 and Stefan O. Ciurea 2 * Haploidentical related donors are an attractive alternative source of stem cells for allogeneic stem cell transplantation due to widespread availability and ease of stem cell procurement. Historically, haploidentical stem cell transplantation (HaploSCT) with extensive T-cell depletion has been associated with high rates of infectious complications and nonrelapse mortality (NRM). Post-transplantation cyclophosphamide (PTCy) has been shown to induce immune tolerance, effectively control graft-versus-host-disease (GVHD), and is associated with lower NRM, making it a preferred option for patients undergoing HaploSCT. Over the last decade, several groups investigated PTCy for GVHD prevention in HaploSCT; it is now successfully utilized with both myeloablative and nonmyeloablative conditioning regimens with survival comparable to HLAmatched transplantation. Future directions will focus on optimizing conditioning regimens by diagnosis, improving donor selection, and enhancing graft-versus-leukemia effect. Am. J. Hematol. 00:000 000, 2015. VC 2015 Wiley Periodicals, Inc. Introduction Hematopoietic stem cell transplantation is often the only curative option for many patients with malignant and benign hematological stem cell disorders [1]. An HLA-matched sibling (MSD) is the preferred donor; however, the probability of having such a donor is approximately 30%. According to the National Marrow Donor Program, up to 75% of the registered donors are Caucasian; the probability of finding an 8/8 matched unrelated donor (MUD) is approximately 75% in this population, with approximately half of such patients ultimately proceeding to a MUD transplant [2]. The effect of race on HLA matching is more pronounced in non-caucasians, with a probability of finding a match approximately 30 to 40% in the Hispanic population, less than 20% for African-Americans and Asian population [2,3]. Even though, the probability of finding a MUD has increased over the last couple of years due to worldwide expansion of donor registries, some major obstacles remain such as, underrepresentation of the ethnic minorities in the registries, significant genetic variability for some races, expansion of the number of mixed race individuals, and significant delays in obtaining stem cells of an average of four months from initiation of the donor search to transplantation [4,5]. Moreover, cost of the acquisition of unrelated donor cells remains prohibitive for the great majority if not all of underdeveloped countries. Therefore, there is a considerable need to develop alternate donor stem cell sources. Unrelated umbilical cord blood (UCB) offers several advantages: the donor search may be more expeditious and time to transplant is in general shorter as compared with MUDs, there is no risk to the donor, and a greater degree of mismatch is acceptable in order to proceed to a transplant [6]. However, the delayed engraftment and immaturity of the transplanted immune system are significant problems, particularly in adults where there can be a discrepancy between the patient s body weight and the number of hematopoietic cells in the cord blood unit [6]. This can partially be abrogated with the use of double cord blood units in adult patients in order to decrease the risk of graft failure [6]. Moreover, there are significant expenses in maintaining cord blood banks around the world acquisition and expansion of the cord is costly, and there is no possibility to collect additional cells for cellular therapy, in case this is needed. Transplantation from a full haplotype mismatch family donor has been studied for several decades, initially with T-cell depletion (TCD); however, due to suboptimal results, this approach has fallen out of favor over the years. Recently, utilization of haploidentical transplantation has been revived primarily due to the use of post-transplantation cyclophosphamide (PTCy) as well as the development of novel methods of selective depletion of T-cell subsets, such as the use of ab TCD [7 9]. Due to improved safety, ease in procurement, and perhaps lower costs, haploidentical donors are now considered to be a viable alternative donor source [10]. In addition, HaploSCT offers several key advantages: almost all patients 1 Department of Hematology/Oncology, Aga Khan University Hospital, Karachi, Pakistan; 2 Department of Stem Cell Transplantation and Cellular Therapy, The University of Texas, MD Anderson Cancer Center, Houston, Texas Conflict of interest: The authors have no potential conflict of interest to declare. *Correspondence to: Stefan O. Ciurea, Department of Stem Cell Transplantation and Cellular Therapy, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 423, Houston, TX 77030. E-mail: sciurea@mdanderson.org Received for publication: 18 January 2015; Revised: 24 February 2015; Accepted: 25 February 2015 Am. J. Hematol. 00:00 00, 2015. Published online: 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ajh.23995 VC 2015 Wiley Periodicals, Inc. doi:10.1002/ajh.23995 American Journal of Hematology, Vol. 00, No. 00, Month 2015 1

Shabbir-Moosajee et al. have a haploidentical first-degree related donor from the family, the donors are readily available and highly motivated to donate, and the patients can proceed to transplant very quickly (in less than 3 weeks) [5]. This is of particular importance for advanced/poor-risk hematological malignancies, when the goal is to transplant in remission or with minimal disease burden and there is high risk of disease progression in a short timeframe. Finally, there is generally the availability of donor-derived cellular therapy, such as donor lymphocyte infusion (DLI) or cellular subsets which can be used post-transplant, if needed. The major drawback, historically at least, has been strong alloreactivity in the graft-versus-host and host-versus-graft directions due to major HLA class I and II mismatches [11]. Thus, in essence, controlling alloreactivity in this setting has been the holy grail of transplantation as it will allow to perform this procedure for virtually every patient in need. Over the last three decades, most of the success in HaploSCT has come with improved understanding of the HLA barriers, determinants of T-cell alloreactivity, and immunologic reconstitution [5]. Other work has focused on refining the degree of myeloablation and immune suppressive effect of the conditioning regimen, primarily using either ex vivo or in vivo T-cell depletion of the graft [5]. More recently, much success has been achieved by using PTCy as the primary method of prevention of graft-versus-host disease (GVHD) in T-cell replete HaploSCT [9,12,13]. With better control of alloreactive reactions generated by transplants using major HLA-mismatched donors, GVHD and graft failure rates have been successfully controlled, with significant improvements in NRM, making outcomes of HaploSCT similar to matched transplants [9,14 17]. In this review we will focus on recent advances in HaploSCT with the use of PTCy and present available data with regards to conditioning regimens employed for both myeloid and lymphoid hematological malignancies. The lower cost in stem cell obtainment and ease of approach of haploidentical transplantation will likely favor expansion of this form of transplantation worldwide, especially in developing nations, which may not have an unrelated donor bank or access to unrelated donors may be cost-prohibitive. Furthermore, newer approaches will likely need to be compared in the future with this method, both in terms of efficacy and cost, as PTCy is establishing itself as the new gold-standard for performing HaploSCT. Post-Transplant Cyclophosphamide for GVHD Prevention CRITICAL REVIEW One of the earliest evidences of the effect of cyclophosphamide on allograft alloreactive responses was reported by Barenbum and Brown in the early 1960s [18]. Mice were injected with intraperitoneal cyclophosphamide 200 mg/kg either before or after the placement of a major histocompatibility complex (MHC) mismatched skin graft. When compared with the controls, the mice injected with cyclophosphamide exhibited delayed graft rejection by approximately 4 days. Interestingly, the maximal benefit of graft survival was seen when the cyclophosphamide was administered 3 to 4 days after the skin graft. This study suggested that the effect of cyclophosphamide on graft survival is likely a result of global as well as time dependent immune suppression [18]. This concept was translated to hematopoietic stem transplantation by Santos and Owens from Johns Hopkins. They found that mice treated with cyclophosphamide administered after day 2 after cell transfer had lower incidence of skin GVHD and reduced mortality and felt that this is well suited for the investigation of therapy directed toward modifying graft-versus-host disease [19]. Luznik et al. successfully demonstrated in mouse models that MHC-mismatched bone marrow cells can engraft stably after nonmyeloablative conditioning consisting of fludarabine or cyclophosphamide, and low-dose total body irradiation (TBI) followed by PTCy on day 13. Additionally, they proved that long-term mixed hematopoietic chimerism, clonal deletion of donor-reactive T cells, and bidirectional cytotoxic T-cell tolerance can be achieved, resulting in a decrease in graft-versus-host (GVH) reactions caused by major and minor histocompatibility antigen-incompatible T cells. Moreover, recipient mice demonstrated a robust graft-versus-leukemia (GVL) effect in response to infusion of DLI after receiving leukemia cells without causing lethal GVHD [12]. Cyclophosphamide-induced tolerance develops in a complex multisystem sequential manner [13]. Firstly, there is selective destruction of proliferating allo-antigen stimulated T cells. In this process, both donor anti-host and recipient anti-donor T cells are eliminated. These T cells have replicative DNA and therefore are exquisitely sensitive to cyclophosphamide; this occurs in the first few days of cyclophosphamide administration [13]. The approach is very effective in HaploSCT where there is a large proliferation of allo-reactive T cells in the early post-transplant period due to major HLA-mismatch. Additionally, after cyclophosphamide administration, the donor memory T-cells are relatively spared, which can provide the recipient with donorderived immunity against a variety of infections and contribute to immune reconstitution post-transplant [13]. Clear evidence was provided by our group; we have shown that a significantly higher proportion of patients who received a TCD haploidentical graft with CD341 selection had infectious complications (viral and fungal) than the patients treated with T-cell replete HaploSCT with PTCy, using the same conditioning regimen [14]. The second step involves the development of peripheral tolerance through clonal deletion, anergy, and T-reg mediated suppression [9,13,20 22]. The third step in the development of cyclophosphamide-induced tolerance involves deletion of anti-host T cells derived from donor hematopoietic stem cells in the thymus. This contributes to establishing both immediate and long-term tolerance of donor cells after allografting [12,13,22,23]. Based on promising preclinical studies, O Donnell et al. reported the first Phase I clinical trial of 13 patients who underwent HaploSCT using a nonmyeloablative conditioning regimen with fludarabine 30 mg/m 2 on day 26 to 22, 2 Gy TBI on day 21, and PTCy (50 mg/kg) on day 13. Additional immunosuppressive therapy included mycophenolate mofetil (MMF) (day 14 to day 135) and tacrolimus (day 14 to day 150). After graft failure in the first two out of three patients, cyclophosphamide (14.5 mg/kg day 26, 25) was added to the remaining patients and 8 of 10 had sustained donor engraftment at a median of 15 days. The incidence of severe (Grade III IV) agvhd was only 23%. At a median follow-up of 200 days, six patients were alive and disease-free [24]. Based on these positive results, additional reports of successful HaploSCT with PostCy had been reported [9,16,25]. Due to these early successes, there has been considerable interest in developing this approach for broader clinical practice. In 2011, Bone Marrow Transplant Clinical Trials Network (BMT-CTN) jointly published results of two parallel Phase II studies of haploidentical and cord blood transplants [15]. Both studies were conducted in high-risk hematological malignancies; BMT-CTN 0603 was investigated in patients undergoing HaploSCT with PTCy and a bone marrow graft, while BMT-CTN 0604 studied double umbilical cord blood transplantation (UCBT). All patients were treated with a nonmyeloablative protocol using a version of Flu/Cy/TBI regimen [15]. Patients undergoing HaploSCT additionally received PTCy along with tacrolimus and MMF; UCBT received cyclosporine and MMF as GVHD prophylaxis. The characteristics of these patients were strikingly similar between the two groups. There was no difference in the cumulative incidence of Grade II IV acute GVHD (agvhd) between the UCBT and haploidentical groups at day 100 (40% vs. 32%, respectively). Remarkably, there were no reported cases of severe (Grade III IV) agvhd in the HaploSCTgroup. Nonrelapse mortality 2 American Journal of Hematology, Vol. 00, No. 00, Month 2015 doi:10.1002/ajh.23995

CRITICAL REVIEW (NRM) was impressively low in the HaploSCT study (7% vs. 24% in the UCBT group); however, the relapse rate appeared higher (45% in the HaploSCT vs. 31%). While the 6-month and 1-year progressionfree survival (PFS) were similar, the causes of treatment failure appeared to differ, primarily NRM in the UCBT and relapse in the HaploSCT [15]. It is critical to note that these studies were not randomized and patient selection may have influenced outcomes; therefore a randomized BMT-CTN study #1101 (NCT 01597778) is underway to help clarify these differences. Other reports had similar findings, such as the one reported by Munchel et al. which presented outcomes of 210 patients with advanced hematological malignancies receiving PTCy after nonmyeloablative conditioning for HaploSCT [16]. The rate of agvhd and chronic GVHD (cgvhd) were low at 27% and 15%, respectively. NRM was 18% and the predominant cause of treatment failure was again relapse which was seen in 55% of the patients. These results were encouraging as most of the patients were heavily pretreated prior to transplant [16]. However, they also suggested that a more intense conditioning could be applied, especially for patients with acute leukemia. Delayed immune reconstitution is a major concern with intensification of post-transplantation immune suppression. In a recent analysis presented at the American Society of Bone Marrow Transplantation (ASBMT), we studied the kinetics of lymphocyte reconstitution after matched and mismatched transplants [26]. The recovery of lymphocyte subsets (CD4 1, CD8 1, naive and memory T cells, NK cells, B cells) were evaluated at 1, 3, 6, and 12 months post-transplant in 100 patients (MSD 5 25, MUD 5 38, and haploidentical 5 37). T-cell recovery was remarkably similar between haploidentical and matched transplants, although higher CD4 1 and CD8 1 T-cells were evident at day 30 post-transplant in the MSD group. However, the recovery of T-cell replete haploidentical (28 of 37 total HaploSCT) and MUD transplants displayed a virtually identical pattern of T-cell recovery despite the use of PTCy in the HaploSCT group. Our data suggests the use of PTCy does not impair or delay the recovery of immune function post-transplant. Reconstitution of CD3 1, CD4 1, and CD8 1 T-cell subsets to normal levels usually occurs by 6 months post-transplantation [26]. This analysis as well as several other studies suggest that the T-cell recovery for HaploSCT treated with PTCy is similar to matched donor transplants, which could, in part, explain comparable outcomes [26 28]. Haploidentical Transplantation for Myeloid Malignancies Outcomes of patients with relapsed or refractory acute myeloid leukemia (AML) without transplant are very poor with survival of 20% or less. Additionally, poor-risk cytogenetics and FLT3 mutated AML patients benefit from an allogeneic stem cell transplant in first complete remission (CR1). Performing a transplant using any donor in this case may be lifesaving, and with improved HaploSCT outcomes, lack of an HLA-matched donor should not limit access to transplantation of such patients. In addition, performing a transplant expeditiously could be more important than waiting for an HLAmatched unrelated donor. Haploidentical transplantation with complete T-cell depletion has been associated with higher NRM; up to 66% in patients with more advanced disease has been reported [29]. Our experience with a TCD haploidentical graft has also been associated with an unacceptably high TRM [14,30,31]. By changing to a fully unmanipulated bone marrow graft and PTCy, tacrolimus, and MMF for GVHD prevention, the transplant outcomes have improved significantly compared with TCD HaploSCT due to a lower TRM and infectious complications in myeloid malignancies [14]. Conditioning for HaploSCT with postcy Transplantation for myeloid malignancies with a matched donor has been traditionally performed with a myeloablative conditioning regimen for younger patients with minimal comorbidities. A more intense, myeloablative conditioning regimen for HaploSCT has been studied by several groups, including ours. Solomon et al. from Atlanta, and the Italian group led by Bacigalupo have used a busulfan-based conditioning regimen with fludarabine and cyclophosphamide (Bu/Flu/Cy) or with thiotepa and fludarabine (Thio/Bu/Flu), respectively, with very good results [32,33] (Table I). Solomon et al. reported on 20 patients of whom nine had relapsed-refractory leukemia. The myeloablative conditioning was comprised of fludarabine, busulfan (440 mg/m 2, dose subsequently reduced due to mucositis), and cyclophosphamide [32]. On day 0, patients received an unmanipulated peripheral blood T-cell replete allograft. The cumulative incidence of severe agvhd and cgvhd were low at 10% and 5%, respectively with a day 100 NRM of only 10%. The 1-year overall survival (OS) and relapse rates were encouraging at 69% and 40% respectively, better for standard-risk patients (88% and 33%) [32]. However, due to significant mucositis and higher incidence of BK virus cystitis this group is now exploring a TBI-based regimen with encouraging early results (Table I) [36]. Raiola et al. reported results of a Phase II retrospective trial, where 50 patients were treated with a myeloablative protocol consisting of thiotepa 10 mg/kg, busulfan 9 mg/kg, and fludarabine 150 mg/m 2 or total body irradiation (TBI) with fludarabine 120 mg/m 2 [33] (Table I). Nearly two thirds of the patients had high-risk myeloid malignancies [AML, chronic myeloid leukemia (CML), or myeloproliferative disorder (MPD)]. The patients subsequently underwent a T-cell replete haploidentical bone marrow transplant followed by PTCy on days 13 and 1 5, cyclosporine, and MMF for GVHD prophylaxis. The cumulative incidence of Grade II IV agvhd was 12% with moderate cgvhd occurring in 10%. With a median follow-up for surviving patients of approximately 8 months, the cumulative incidence of NRM was 18% with relapse occurring in 22%. These results were particularly encouraging as 27 patients had active disease at time of transplantation, 10 had a previous allogeneic transplant, only 9 of 25 of the patients with AML were in CR1, and 72% of the patients with AML were alive and in remission at the time of reporting; these data helps to confirm the hypothesis that increasing the intensity of the conditioning platform can decrease the risk of relapse without significantly affecting NRM and GVHD in HaploSCT performed with PTCy [33]. Our results for myeloid malignancies were recently presented at the 2014 ASBMT/CIBMTR Tandem Meeting [35]. All patients received a uniform conditioning regimen of melphalan 140 mg/m 2 (dose-reduced to 100 mg/m 2 in patients older than 55 years), fludarabine 160 mg/m 2 with or without thiotepa 5 to 10 mg/kg (Table I). GVHD prophylaxis consisted of PTCy, tacrolimus, and MMF. Of the first 100 patients treated on this protocol, 66 had myeloid malignancies [AML, myelodysplastic syndrome (MDS), CML] and 26% of the patients were not in CR at time of the transplant. The 3-year PFS, 1- year TRM, and cumulative incidence of relapse were 56.5%, 11.8%, and 30.1%, respectively. The incidence of Grade II IV agvhd was only 25%. Patients who underwent transplantation in morphologic CR or had low risk cytogenetics had a statistically significant improvement in PFS compared with other groups. Patients with intermediate or low-risk cytogenetics had a 3-year PFS of approximately 70% [35]. As opposed to a nonmyeloablative strategy, increased intensity of the conditioning platform appears to be more effective for patients with CML to reduce rejection rates. Outcomes for patients with CML treated with a T-cell replete graft from a haploidentical donor source and intensified GVHD prevention (without PTCy) were initially reported by the Chinese group [41]. We recently reported in abstract format our experience on a small number of advanced CML patients doi:10.1002/ajh.23995 American Journal of Hematology, Vol. 00, No. 00, Month 2015 3

4 American Journal of Hematology, Vol. 00, No. 00, Month 2015 doi:10.1002/ajh.23995 TABLE I. Preparative Regimens used in Unmanipulated Haploidentical Transplantation with Post-Transplantation Cyclophosphamide Reference Conditioning regimen Diseases Myeloid malignancies Solomon et al. [32] Flu/Bu/Cy regimen (n =5) Fludarabine 30 mg/m 2 on days 27 to 22(total dose 180 mg/m 2 ) Busulfan 130 mg/m 2 on days 27 to 24 (total dose 520 mg/m 2 ) Cyclophosphamide 14.5 mg/kg on days 23 and 22 (total dose 29 mg/kg) Flu/Bu/Cy regimen (n =15) (dose reduction due to mucositis) Fludarabine 25mg/m 2 on days 26 to 22(total dose 125 mg/ m 2 ) Busulfan 110 mg/m 2 on days 27 to 24 (total dose 440 mg/m 2 ) Cyclophosphamide 14.5 mg/kg on days 23 and 22 (total dose 29 mg/kg) Raiola et al. [33] Thio/Bu/Flu regimen (n =35; 8/35 received reduced dose busulfan) Thiotepa 5 mg/kg on days 26 and 25 (total 10 mg/kg) Busulfan 3.2 mg/kg IV on days 24 to 22 (total 9.6 mg/kg) Fludarabine 50 mg/m 2 on days 24 to 22 (total 150 mg/m 2 ) Flu/TBI regimen (n =15) TBI 3.3 Gy on days 28 to 26 (total 9.9 Gy); Fludarabine 30 mg/m 2 on days 25 to 22 (total 120 mg/ m 2 ) Bashey et al. [34] Bu/Flu/Cy regimen (n =18) Fludarabine 25 mg/m 2 on days 26 to 22 (total 125 mg/m 2 ) Busulfan 110 130 mg/m 2 /day IV on days 27 to 24 Cyclophosphamide 14.5 mg/kg on days 23 and 22 (total 29 mg/m 2 ) Flu/Cy/TBI (n =35) Fludarabine 30 mg/m 2 on days 26 to 22 Cyclophosphamide 14.5 mg/kg on days 26 and 25 (total 29 mg/m 2 ) Total body irradiation 2 Gy on day 21 Pingali et al. [35] Flu/Mel/Thio regimen Fludarabine 40 mg/m 2 on days 25 to 22 (total dose 160 mg/m 2 ) Mephalan 100 140 mg/m 2 on day 26 (total dose 100 140 mg/m 2 ) Thiotepa a 5 mg/kg on day 27 (total dose 5 mg/kg) (older patients/comorbidities received reduced doses of melphalan) Solomon et al. Flu/TBI regimen Fludarabine 25 mg/m 2 on days 27 to 25 [36] c (total dose 75 mg/m 2 ) TBI 150 cgy BID on days 24 to 21 (total dose 12Gy) Lymphoid malignancies Burroughs et al. [25] Raiola et al. [37] Castagna et al. [38] Kanakry et al. [34] Flu/Cy/TBI regimen Fludarabine 30 mg/m 2 /d on days 26 to 22(total dose 150 mg/m 2 ) Cyclophosphamide 14.5 mg/kg/ day on days 26 and 25 (total dose 29 mg/kg) 2 Gy TBI on day 21 (total dose 2 Gy) Flu/Cy/TBI regimen Fludarabine 30 mg/m 2 /d IV daily on days 26 to 22 (total dose 150 mg/m 2 ) Cyclophosphamide 14.5 mg/kg IV on days 26 and 25 (total dose 29 mg/kg) 2 Gy TBI on day 21 (total dose 2 Gy) Flu/Cy/TBI regimen Fludarabine 30 mg/m 2 /d IV daily on days 26 to 22 (total dose 150 mg/m 2 ) Cyclophosphamide 14.5 mg/kg IV on days 26 and 25 (total dose 29 mg/kg) 2 Gy TBI on day 21 (total dose 2 Gy) Flu/Cy/TBI regimen Fludarabine 30mg/m 2 for 5 days (total dose 150 mg/m 2 ) Cyclophosphamide 14.5 mg/kg for 2 days (total dose 29 mg/kg) 2 Gy TBI on one day (total dose 2 Gy) (only reduced intensity haploidentical included in table) Myeloid (30%- relapsed refractory) AML/CML (45%) Lymphoid (25%) AML 50% ALL 25% MPD 16% AML 32% MDS/MPD 15% ALL 19% HaploSCT (n) Graft agvhd (II IV) NRM Relapse rate PFS or DFS 20 PB 30% 10% at 1yr 40% at 1yr 50% at 1yr (DFS) 50 BM (all) 12% 18% at 6 mo 22% at 18 mo 51% at 18 mo (DFS) 53 PBSC (n =18) BM (n =35) 30% 7% at 2yr 33% at 2yrs 60% at 2yrs (DFS) AML/MDS 66% 66 BM (94%) 25% 11.8% at 3yr 30.1% at 3yrs 56.5% at 3yrs b (PFS) AML 70% ALL 10% CML 15% 30 PB 44% 5% at 2yrs 19% at 2yrs 76% at 2yrs HD 100% 28 BM 43% 9% at 2yrs 40% at 2yrs 51% at 2yrs (PFS) HD 100% 26 BM (100%) 24% 4% 31% at 18 mo 63% at 3 yrs (DFS) HD 55% NHL 39% 49 BM (80%) 26% 16% at 2yrs 19% at 2yrs 63% at 2yrs (PFS) PTCL 100% 18 BM 16% 11% at 1 yr 34% at 1 yr 37% at 2 yrs (PFS)- includes all reduced intensity in study Shabbir-Moosajee et al. CRITICAL REVIEW

CRITICAL REVIEW Conditioning for HaploSCT with postcy TABLE I. Continued HaploSCT (n) Graft agvhd (II IV) NRM Relapse rate PFS or DFS Reference Conditioning regimen Diseases NHL 75% HD 25% 151 BM 32% 16% at 1yr 31% at 1yr 40% at 3yrs (DFS) 19 BM 44% 11% at 2 yrs 26% at 2yrs 52% at 22 mo (PFS) (70% with FM100) HD 37% NHL 37% CLL/PLL 26% Kasamon et al. Flu/Cy/TBI regimen Fludarabine 30 mg/m 2 on days 26 to 22 [39] c (total dose 150 mg/m 2 ) Cyclophosphamide 14.5 mg/kg on days 26 and 25 (total dose 29 mg/kg) 2 Gy TBI on day 21 (total dose 2Gy) Brammer et al. Flu/Mel/Thio of Flu/Mel/TBI regimen Fludarabine 40 mg/m 2 [40] c day on days 25 to 22 (total dose 160 mg/m 2 ) Mephalan 100 140 mg/m 2 on day 26 (total dose 100 140 mg/m 2 ) Thiotepa a 5 mg/kg on day 27 (total dose 5 mg/kg) or 2 Gy TBI Dose reduction to preferred regimen of Flu/Mel100 mg/m 2 / TBI a Thiotepa replaced with 2 Gy TBI except for patients with prior CNS disease. b Outcomes for patients in remission at transplant. c Abstract. Bu busulfan, Cy cyclophosphamide, TBI total body irradiation, Mel melphalan, Flu fludarabine, Thio thiotepa, AML acute myeloid leukemia, MDS myelodysplastic syndromes, CML chronic myeloid leukemia, ALL acute lymphoblastic leukemia, HD Hodgkin s disease, NHL non-hodgkin s lymphoma, CLL chronic lymphocytic leukemia, PLL prolymphocytic leukemia, PTCL peripheral T cell lymphoma, BM bone marrow, PB peripheral blood, (all progressed to accelerated or blast phase) who underwent HaploSCT with PTCy, tacrolimus, and MMF for GVHD prevention. All patients received conditioning with our institutional protocol with fludarabine 160 mg/m 2, melphalan 140 mg/m 2 6 thiotepa 5 10 mg/kg. All patients engrafted the neutrophils at a median of 18 days. There were no deaths due to transplant related toxicities. The incidences of agvhd and cgvhd were 30% and 37.5% respectively. At a median follow-up of 22 months, six patients (60%) were alive and five remain in molecular remission [42]. These results are very encouraging, suggesting that patients with CML with a transplant indication and no matched donor can proceed safely to a haploidentical transplant, and that outcomes of patients with CML also may be similar with matched transplants. Haploidentical Transplantation for Lymphoid Malignancies Initially, a nonmyeloablative conditioning regimen has been used for patients with lymphoma undergoing HaploSCT. This approach gained favor as it was found to have acceptable toxicities with limited NRM and encouraging survival rates. One of the first reports of efficacy of haploidentical transplantation in lymphoma was published by Burroughs et al. [25]. Ninety patients with relapsed and refractory, heavily pre-treated Hodgkin s lymphoma (HL) received MSD (n =38), MUD (n =24), and haploidentical transplants (n =28). All patients received a nonmyeloablative conditioning regimen; the HaploSCT group received Flu/Cy/TBI with PTCy, tacrolimus, and MMF. In multivariate analyses, there was no significant difference in OS between the three groups at two years. HaploSCT had significantly improved PFS compared with the other two groups and significant decrease in NRM compared with MSD. No significant differences were noted in agvhd grades III-IV or cgvhd between the groups. The 2-year OS and PFS was similar between the three groups. Additionally, there was no statistically significant difference in the rate of Grade II IV agvhd between the haploidentical and matched transplants. Patients undergoing a haploidentical transplants tended to have a lower NRM as well as a lower relapse rate and improved survival when compared with the other two groups of matched transplants, likely related to increase degree of mismatch in the HaploSCT group. This data suggested that a nonmyeloablative approach has significant clinical efficacy in relapsed and refractory lymphoma, and importantly, HaploSCT can be suitable alternate donor source with at least similar treatment outcomes [25]. These initial good results have been subsequently confirmed at other centers in the past several years [37,38]. Raiola et al. reported results of 26 patients undergoing T-cell replete HaploSCT for advanced HL [37]. These patients received a nonmyeloablative conditioning protocol consisting of Flu/Cy/TBI and PTCy, tacrolimus or cyclosporine, and MMF for GVHD prophylaxis. All the patients had previously received an autograft and 65% had active disease at the time of transplant. The cumulative incidence of TRM and relapse was 4% and 31%, respectively. The actuarial 3-year OS and PFS were 77% and 63%, respectively [37]. Additionally a retrospective review of 151 lymphoma patients undergoing a haploidentical transplant was presented at the American Society of Hematology meeting in 2013 [39]. All patients received Flu/Cy/TBI nonmyeloablative conditioning followed by a T-cell replete bone marrow graft. GVHD prophylaxis consisted of PTCy, MMF, and tacrolimus. Seventy-five percent of patients had non-hodgkins lymphoma (NHL) (both B and T cell) and 25% had Hodgkin s lymphoma (HL). The rate of severe acute GVHD was low (5%) and the 1-year NRM and relapse rate were 16% and 31%, respectively. The 3-year probabilities of PFS and OS were 40% and 46%. The disease-specific cumulative incidence of OS were as follows: 67% for HL, 38% for aggressive NHL, 28% for mantle cell lymphoma, and 48% for indolent NHL and doi:10.1002/ajh.23995 American Journal of Hematology, Vol. 00, No. 00, Month 2015 5

Shabbir-Moosajee et al. chronic lymphocytic leukemia (CLL) [39]. These results in a large number of patients show that transplant outcomes with a haploidentical donor are likely at least as good as outcomes with matched transplants. We also explored a melphalan-based conditioning for lymphoma patients as this drug has proven efficacy against lymphomas. Our group recently presented the data on the first 19 patients with HL, NHL, and CLL [40]. GVHD prophylaxis consisted of PTCy, tacrolimus, and MMF (continued until day 1100). Seventeen of these patients had either relapsed or refractory NHL/HL or CLL, two thirds were not in remission at the time of transplant. The 2-year OS and PFS in this group was 63% and 52%, respectively, with a 2-year TRM of 11%, and relapse rate of only 25%. Patients with lymphoid malignancies who received fludarabine/melphalan 100 mg/m 2 (FM100) with either 2Gy TBI or thiotepa 5 mg/kg regimen were found to have a very encouraging 2-year PFS of 70% compared with 31% in the fludarabine/melphalan 140 mg/m 2 (FM140) group. Thus far, our data has shown that an intensified, melphalan-based conditioning platform can safely be used for patients with lymphoid malignancies treated with haploidentical transplants and PTCy with a corresponding reduction in relapse rates. The FM140 regimen appeared to be associated with higher NRM and we are currently using only the FM100 with 2Gy TBI regimen, which appears to be associated with lower TRM and at least as effective as the more intensive regimen (Table I) [40]. Comparable to matched transplants, haploidentical transplants with PTCy have been performed in T-cell lymphoma with similarly good outcomes [34]. Kanakry et al. reported results of 44 patients who underwent an allogeneic transplant for peripheral T-cell lymphoma, of which 18 patients had a haploidentical donor with reduced-intensity conditioning (RIC); only four patients received a myeloablative HaploSCT in this group. This was a notably poor-risk group with 75% of patients with either chemo-refractory or active disease. Sixty percent of patients had greater than two prior lines of therapy. Patients received a myeloablative conditioning protocol (busulfan/cyclophosphamide, Cy/TBI, or busulfan/fludarabine) or a RIC regimen of Flu/Cy/TBI or Flu/TBI. Patients who received a haploidentical donor were more likely to receive the RIC. PTCy was incorporated as part of the GVHD prophylaxis for most patients irrespective of donor source. Despite the considerable treatment variability between the donor groups, the estimated 1-year TRM and cumulative incidence of relapse were similar between HLA-identical and haploidentical donors [34] (Table I). As illustrated above, haploidentical transplants can result in improved OS and potential for cure in patients with advanced lymphoma with both a nonmyeloablative and more intensive conditioning platform with several groups reporting acceptable NRM and very good early outcomes, although the number of treated patients is still small. As mentioned above, very good results have been obtained using either the Flu/Cy/TBI regimen or our reduced-intensity conditioning regimen with FM100 and 2GyTBI (Table I). It is likely that a reduced-intensity/nonmyeloablative conditioning regimen is sufficient for these patients as most are heavily pretreated including prior autologous transplantation and it is unlikely they will tolerate very intense conditioning. In addition, newer targeted therapies in the conditioning regimen or in the maintenance post-transplant setting will likely improve the survival of these patients. Comparative Outcomes with Matched Transplants As outcomes of haploidentical transplantation have improved, several retrospective studies including our early data demonstrate comparable outcomes with matched transplants, for patients with hematologic malignancies in general, as well as for several diseases in CRITICAL REVIEW particular [25,34,37,43]. Bashey et al. retrospectively compared outcomes of 271 patients undergoing an allogeneic transplant for highrisk hematological malignancies. One hundred seventeen patients received a MSD, 101 patients had a MUD, and 53 patients underwent transplant using a haploidentical donor. GVHD prophylaxis consisted of PTCy, tacrolimus, and MMF for HaploSCT while matched transplants received conventional GVHD prophylaxis. For MSD, MUD, and haploidentical donors, the 2-year cumulative incidence of NRM was 13%, 16%, and 7%, respectively; relapse was similar between the groups (34%, 34%, and 33%); and 2-year DFS was similar 53%, 52%, and 60%, respectively. Patients undergoing HaploSCT were found to have a similar incidence of agvhd and less severe cgvhd (11%, 12%, and 4%, respectively) [43]. Raiola et al. also compared outcomes of 459 patients treated with different donor sources haploidentical, matched related and unrelated donor, mismatched unrelated and cord blood grafts [37]. Haploidentical transplants had a lower incidence of Grade II IV agvhd compared with MSD and the lowest NRM at 1,000 days (18% vs. 24 35% for all others), comparable relapse rates and 4 year OS (52% haploidentical, 45% MSD, 43% MUD, 40% mismatched MUD and 34% cord blood) [17]. We analyzed outcomes of a uniform cohort of 227 AML/MDS patients treated with the same conditioning regimen (fludarabine 120 160 mg/m 2, melphalan 100 140 mg/m 2, 1 thiotepa 5 10 mg/kg for HaploSCT). Eighty-seven patients received a MSD, 108 received a MUD, and 18 patients received a haploidentical donor. GVHD prophylaxis in matched donors consisted of tacrolimus and minimethotrexate with thymoglobulin in MUDs; HaploSCT received PTCy, tacrolimus, and MMF. Older patients and patients with major comorbidities received the reduced doses described above. More patients were in remission in the HaploSCT group and a separate analysis was hence performed to address patients in remission. For MSD, MUD, and HaploSCT, there was a similar incidence of Grade II-IV agvhd (24%, 19%, 26%) for patient s in remission; severe agvhd occurred in 4%, 4%, and 0%, respectively. Extensive cgvhd between the three groups was also similar (29%, 23%, and 17%), respectively. Also, there was no significant difference in PFS at 3 years for those in remission (57%, 45%, 41%, respectively). However, there was a non-significant trend towards better PFS at 3 years in patients undergoing a MSD transplant (57%) For those not in remission a significant differences was not evident (27%, 21%, and 10%, respectively) [44]. These very good results with nearly identical survival between haploidentical and matched transplants both for myeloid and lymphoid malignancies raise the question whether patients with more advanced disease who achieve a transient remission should proceed urgently to transplant using a haploidentical donor rather than wait several months for unrelated donor cells. During this search and procurement time, disease may progress and the opportunity to perform a transplant maybe lost; this question mandates prospective studies which will answer this question. Peripheral Stem Cells as Alternative Graft Source for Haploidentical Transplants Control of alloreactivity and decreasing the incidence of GVHD is of paramount importance in successful haploidentical transplantation. The traditional source of stem cells has been from bone marrow (BM) which are less mature and have been shown to induce less GVHD than peripheral blood stem cells (PBSC) [45]. However, acquisition of BM stem cells is cumbersome, involves the use of the operating room, and presents an increased risk of complications to 6 American Journal of Hematology, Vol. 00, No. 00, Month 2015 doi:10.1002/ajh.23995

CRITICAL REVIEW the donor. Also, the engraftment is usually delayed by approximately a week, and the cell doses are often lower, which can lead to increased infection risk and prolonged hospitalization for the patient. Therefore, there is considerable interest in developing peripheral blood stem cells as an acceptable graft source in HaploSCT. Recently, results of a multicenter collaborative study of haploidentical transplant patients treated with a peripheral blood graft have been reported [46]. Fifty-five patients with high-risk hematological malignancies underwent a T-cell replete HaploSCT with PBSC as source of the donor graft. Approximately half the patients had myeloid and half lymphoid malignancies. The mean CD341 cell dose was 6.4 3 10 6 /kg cells and CD31 was 2.0 3 10 8 /kg. Grade II and III agvhd occurred in 53% and 8% of patients at 1 year while Grade IV GVHD was not observed. The cumulative incidence of cgvhd at 2 years was 18%. With a median follow-up of approximately 500 days, 2-year OS and EFS were 48% and 51%, respectively with a 2- year cumulative incidence of NRM and relapse of 23% and 28% [46]. Castagna et al. retrospectively compared outcomes of patients who underwent a T-cell replete nonmyeloablative haploidentical transplant with PBSC versus BM. The median CD341 cells doses were 3 3 10 6 cells/kg (range: 0.8 7) and 5.1 cells 3 10 6 /kg (range: 3.2 14), and CD31 cell doses were 34 3 10 6 cells/kg (range: 17 92) and 273 3 10 6 cells/kg (range: 119 530) for the bone marrow and PBSC graft, respectively; both CD341 and CD31 cell doses were significantly different. No differences were evident in neutrophil and platelet recovery between the groups. When compared with patients who received a BM graft, PB had a higher incidence of Grade II IV agvhd (33% vs. 25%) with a nonsignificant difference in Grade III IV (14% vs. Conditioning for HaploSCT with postcy 3%). A comparable incidence of cgvhd (13% vs. 13%) was noted. There was no difference in NRM, relapse, or survival between the 2 groups [47]. From these early reports it may be concluded that transplantation of PBSC compared with BM can result in a higher rate of Grade II-III agvhd; however, the OS and EFS appear similar. Most of these studies reported thus far have been retrospective, therefore limited by patient selection and bias. Further prospective studies will be needed to adequately compare these two donor sources [47]. Conclusions and Future Directions Over the last decade, haploidentical donors have become an attractive option for allogeneic stem cell transplantation, in particularly for ethnic minorities and in underdeveloped countries where procurement of unrelated donor stem cells is difficult or cost prohibitive. PTCy has proven to be a low cost and highly effective strategy in the armamentarium of GVHD prophylaxis for HaploSCT. Collectively, studies to date have shown that PTCy decreases the rate of severe agvhd and cgvhd and is associated with low NRM and improved survival. A nonmyeloablative or reduced-intensity conditioning regimen appears very effective for lymphoid malignanices, while myeloid malignancies may require more intensive preparative regimens, especially for select patients who can tolerate it. Early data from multiple groups is highly suggestive that outcomes of T-cell replete HaploSCT performed with PTCy, tacrolimus, and MMF for GVHD prevention are comparable with outcomes of matched transplants. Future directions will optimize conditioning regimens for different diseases and focus on prevention of disease relapse post-transplant. References 1. Copelan EA. Hematopoietic stem-cell transplantation. New Engl J Med 2006;354:1813 1826. 2. Gragert L, Eapen M, Williams E, et al. HLA match likelihoods for hematopoietic stem-cell grafts in the U.S. registry. N Engl J Med 2014; 371:339 348. 3. Dehn J, Arora M, Spellman S, et al. Unrelated donor hematopoietic cell transplantation: Factors associated with a better HLA match. Biol Blood Marrow Transplant 2008;14:1334 1340. 4. Huang XJ. Current status of haploidentical stem cell transplantation for leukemia. J Hematol Oncol 2008;1:27 5. Bayraktar UD, Champlin RE, Ciurea SO. Progress in haploidentical stem cell transplantation. Biol Blood Marrow Transplant 2012;18:372 380. 6. Munoz J, Shah N, Rezvani K, et al. Concise review: Umbilical cord blood transplantation: Past, present, and future. Stem Cells Translat Med 2014;3:1435 1443. 7. Locatelli F, Bauquet A, Palumbo G, et al. Negative depletion of alpha/beta1 T cells and of CD191 B lymphocytes: A novel frontier to optimize the effect of innate immunity in HLAmismatched hematopoietic stem cell transplantation. Immunol Lett 2013;155:21 23. 8. Schumm M, Lang P, Bethge W, et al. Depletion of T-cell receptor alpha/beta and CD19 positive cells from apheresis products with the Clini- MACS device. Cytotherapy 2013;15:1253 1258. 9. Luznik L, O Donnell PV, Symons HJ, et al. HLA-haploidentical bone marrow transplantation for hematologic malignancies using nonmyeloablative conditioning and high-dose, posttransplantation cyclophosphamide. Biol Blood Marrow Transplant 2008;14:641 650. 10. Roth JA, Bensink ME, O Donnell PV, et al. Design of a cost-effectiveness analysis alongside a randomized trial of transplantation using umbilical cord blood versus HLA-haploidentical related bone marrow in advanced hematologic cancer. J Comp Effect Res 2014;3:135 144. 11. Powles RL, Morgenstern GR, Kay HE, et al. Mismatched family donors for bone-marrow transplantation as treatment for acute leukaemia. Lancet 1983;1:612 615. 12. Luznik L, Jalla S, Engstrom LW, et al. Durable engraftment of major histocompatibility complex-incompatible cells after nonmyeloablative conditioning with fludarabine, low-dose total body irradiation, and posttransplantation cyclophosphamide. Blood 2001;98:3456 3464. 13. Luznik L, O Donnell PV, Fuchs EJ. Post-transplantation cyclophosphamide for tolerance induction in HLA-haploidentical bone marrow transplantation. Semin Oncol 2012;39:683 693. 14. Ciurea SO, Mulanovich V, Saliba RM, et al. Improved early outcomes using a T cell replete graft compared with T cell depleted haploidentical hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2012;18:1835 1844. 15. Brunstein CG, Fuchs EJ, Carter SL, et al. Alternative donor transplantation after reduced intensity conditioning: Results of parallel phase 2 trials using partially HLA-mismatched related bone marrow or unrelated double umbilical cord blood grafts. Blood 2011;118:282 288. 16. Munchel AT, Kasamon YL, Fuchs EJ. Treatment of hematological malignancies with nonmyeloablative, HLA-haploidentical bone marrow transplantation and high dose, post-transplantation cyclophosphamide. Best Pract Res Clin Haematol 2011;24:359 368. 17. Raiola AM, Dominietto A, di Grazia C, et al. Unmanipulated haploidentical transplants compared with other alternative donors and matched sibling grafts. Biol Blood Marrow Transplant 2014; 20:1573 1579. 18. Berenbaum MC, Brown IN. Prolongation of homograft survival in mice with single doses of cyclophosphamide. Nature 1963;200:84 19. Santos GW, Owens AH. Production of graftversus-host disease in the rat and its treatment with cytotoxic agents. Nature 1966; 210:139 140. 20. Velasquez-Lopera MM, Eaton VL, Lerret NM, et al. Induction of transplantation tolerance by allogeneic donor-derived CD4(1)CD25(1)Foxp3(1) regulatory T cells. Transplant Immunol 2008;19:127 135. 21. Strauss G, Osen W, Debatin KM. Induction of apoptosis and modulation of activation and effector function in T cells by immunosuppressive drugs. Clin Exp Immunol 2002;128:255 266. 22. Mayumi H, Himeno K, Shin T, et al. Druginduced tolerance to allografts in mice. VI. Tolerance induction in H-2-haplotype-identical strain combinations in mice. Transplantation 1985;40:188 194. 23. Mapara MY, Pelot M, Zhao G, et al. Induction of stable long-term mixed hematopoietic chimerism following nonmyeloablative conditioning with T cell-depleting antibodies, cyclophosphamide, and thymic irradiation leads to donor-specific in vitro and in vivo tolerance. Biol Blood Marrow Transplant 2001;7:646 655. 24. O Donnell PV, Luznik L, Jones RJ, et al. Nonmyeloablative bone marrow transplantation from partially HLA-mismatched related donors using posttransplantation cyclophosphamide. Biol Blood Marrow Transplant 2002;8:377 386. 25. Burroughs LM, O Donnell PV, Sandmaier BM, et al. Comparison of outcomes of HLA-matched related, unrelated, or HLA-haploidentical related hematopoietic cell transplantation following nonmyeloablative conditioning for relapsed or refractory Hodgkin lymphoma. Biol Blood Marrow Transplant 2008;14:1279 1287. 26. Antonio di Stasi MP, Amir Hamdi, Hila S, et al. Ciurea Reconstitution of Lymphocyte Subsets and Outcomes After Matched and Mismatched Hematopoietic Stem-Cell Transplantation. Grapevine, TX: American Society of Bone Marrow Transplantation; 2014. 27. Ding L, Dong L, Zheng XL, et al. [analysis of T lymphocyte absolute number and function in the early phase after haploidentical hematopoietic stem cell transplantation]. Zhongguo Shi Yan Xue Ye Xue Za Zhi/Zhongguo Bing Li Sheng Li Xue Hui [J Exp Hematol Chin Assoc Pathophysiol] 2013;21:702 706. 28. Chang YJ, Zhao XY, Huang XJ. Immune reconstitution after haploidentical hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2014;20:440 449. 29. Ciceri F, Labopin M, Aversa F, et al. A survey of fully haploidentical hematopoietic stem cell transplantation in adults with high-risk acute leukemia: A risk factor analysis of outcomes for doi:10.1002/ajh.23995 American Journal of Hematology, Vol. 00, No. 00, Month 2015 7

Shabbir-Moosajee et al. CRITICAL REVIEW patients in remission at transplantation. Blood 2008;112:3574 3581. 30. Ciurea SO, Saliba R, Rondon G, et al. Reducedintensity conditioning using fludarabine, melphalan and thiotepa for adult patients undergoing haploidentical SCT. Bone Marrow Transplant 2010;45:429 436. 31. Mulanovich VE, Jiang Y, de Lima M, et al. Infectious complications in cord blood and T- cell depleted haploidentical stem cell transplantation. Am J Blood Res 2011;1:98 105. 32. Solomon SR, Sizemore CA, Sanacore M, et al. Haploidentical transplantation using T cell replete peripheral blood stem cells and myeloablative conditioning in patients with high-risk hematologic malignancies who lack conventional donors is well tolerated and produces excellent relapse-free survival: results of a prospective phase II trial. Biol Blood Marrow Transplant 2012;18:1859 1866. 33. Raiola AM, Dominietto A, Ghiso A, et al. Unmanipulated haploidentical bone marrow transplantation and posttransplantation cyclophosphamide for hematologic malignancies after myeloablative conditioning. Biol Blood Marrow Transplant 2013;19:117 122. 34. Kanakry JA, Kasamon YL, Gocke CD, et al. Outcomes of related donor HLA-identical or HLA-haploidentical allogeneic blood or marrow transplantation for peripheral T cell lymphoma. Biol Blood Marrow Transplant 2013;19:602 606. 35. Pingali SR, Milton D, di Stasi A, et al. Haploidentical transplantation for advanced hematologic malignancies using melphalan-based conditioning - Mature results from a single center. Biol Blood Marrow Transplant 2014;20:S40 S41. 36. Solomon SR, Sizemore C, Zhang X, et al. TBI- Based Myeloablative Haploidentical Stem Cell Transplantation Is a Safe and Effective Alternative to Unrelated Donor Transplantation in Patients without Matched Sibling Donors. Blood 2014. pp 426 426. 37. Raiola A, Dominietto A, Varaldo R, et al. Unmanipulated haploidentical BMT following nonmyeloablative conditioning and posttransplantation CY for advanced Hodgkin s lymphoma. Bone Marrow Transplant 2014;49: 190 194. 38. Castagna L, Bramanti S, Furst S, et al. Nonmyeloablative conditioning, unmanipulated haploidentical SCT and post-infusion CY for advanced lymphomas. Bone Marrow Transplant 2014;49: 1552 39. Kasamon YL, Bolanos-Meade J, Gladstone D, et al. Outcomes of nonmyeloablative (NMA) haploidentical blood or marrow transplantation (haplobmt) with high-dose posttransplantation cyclophosphamide (PT/Cy) for lymphoma. Blood 2013;122. 40. Brammer J KI, Gaballa S, Ledesma C, et al. Safety and Efficacy of Nonmyeloablative Melphalan-Based Conditioning for Haploidentical Allogeneic Stem Cell Transplantation in Patients with Advanced Lymphoma. BMT Tandem Meetings. Supplement 2015, 21, S195 S196. 41. Xiao-Jun H, Lan-Ping X, Kai-Yan L, et al. HLAmismatched/haploidentical hematopoietic stem cell transplantation without in vitro T cell depletion for chronic myeloid leukemia: Improved outcomes in patients in accelerated phase and blast crisis phase. Ann Med 2008;40: 444 455. 42. Kehinde Adekola M, Antonio di S, Roberto F, et al. Safety and Efficacy of Haploidentical Stem Cell Transplantation for Advanced Chronic Myeloid Leukemia. American Society of Bone Marrow Transplantation. Grapevine, TX: Biology of Blood and Marrow Transplantation; 2014. pp S213-S214. 43. Bashey A, Zhang X, Sizemore CA, et al. T. cellreplete HLA-haploidentical hematopoietic transplantation for hematologic malignancies using post-transplantation cyclophosphamide results in outcomes equivalent to those of contemporaneous HLA-matched related and unrelated donor transplantation. J Clin Oncol 2013;31: 1310 1316. 44. Di Stasi A, Milton DR, Poon LM, et al. Similar transplantation outcomes for acute myeloid leukemia and myelodysplastic syndrome patients with haploidentical versus 10/10 human leukocyte Antigen-matched unrelated and related donors. Biol Blood Marrow Transplant 2014;20: 1975 1981. 45. Holtick U, Albrecht M,, Chemnitz JM, et al. Bone marrow versus peripheral blood allogeneic haematopoietic stem cell transplantation for haematological malignancies in adults. Cochrane Database Syst Rev 2014;4:CD010189[WorldC at] 46. Raj K, Pagliuca A, Bradstock K, et al. Peripheral blood hematopoietic stem cells for transplantation of hematological diseases from related, haploidentical donors after reduced-intensity conditioning. Biol Blood Marrow Transplant 2014;20:890 895. 47. Castagna L, Crocchiolo R, Furst S, et al. Bone marrow compared with peripheral blood stem cells for haploidentical transplantation with a nonmyeloablative conditioning regimen and post-transplantation cyclophosphamide. Biol Blood Marrow Transplant 2014;20:724 729. 8 American Journal of Hematology, Vol. 00, No. 00, Month 2015 doi:10.1002/ajh.23995