Alternative Donor Graft Sources for Adults with Hematologic Malignancies: A Donor for All Patients in 2017!

Hematopoietic stem cell transplant is potentially curative for a wide variety of malignant diseases; however, choice of a stem cell donor is dependent on donor availability, donor compatibility and health, recipient disease type, and recipient condition. Given the variety of different donor stem cell sources available today, nearly every patient who needs an allogeneic hematopoietic stem cell transplant has a potential donor in 2017.

Abstract

Hematopoietic stem cell transplant (HSCT) is potentially curative for a wide variety of malignant diseases, including acute and leukemias, lymphoma, and myelodysplasia. Choice of a stem cell donor is dependent on donor availability, donor compatibility and health, recipient disease type, and recipient condition. Current sources of stem cell donation for HSCT are matched sibling donors (MSDs), matched unrelated donors (MUDs), 1‐antigen mismatched unrelated donors (MMUDs), haploidentical donors (haplo), and umbilical cord blood (UCB) units. Historically, preferred donors for HSCT have been human leukocyte antigen (HLA)‐matched sibling donors; however, only about 30% of U.S. patients will have a MSD available. The majority of patients referred for HSCT will require an alternative donor graft: MUD, MMUD, UCB, or haplo. The likelihood of finding a MUD varies depending on the ethnicity of the recipient. White Caucasians of European descent have the greatest chance of finding a MUD. Chances of finding a MUD are significantly less for African‐American or Hispanic recipients due to HLA polymorphisms. Therefore, MMUD, UCB, and haplo donor graft sources expand the donor pool for recipients who do not have a MSD or MUD available. Given the variety of different donor stem cell sources available today, nearly every patient who needs an allogeneic HSCT has a potential donor in 2017. All transplant‐eligible patients with hematologic malignancies should be evaluated by a transplant center to determine if HSCT is a viable treatment option for their underlying disease process.

Implications for Practice.

The goal of this review is to increase the awareness of oncology practitioners to the availability of alternative donor stem cell transplants for patients with hematologic malignancies. Despite new agents, stem cell transplant remains the only curative therapy for many patients with acute and chronic leukemia, myelodysplasia, and lymphoma. Given the variety of different donor stem cell sources available today, nearly every patient who needs an allogeneic stem cell transplant will have a donor.

Introduction

Allogeneic hematopoietic stem cell transplant (HSCT) is potentially curative for a wide variety of hematologic malignancies, including acute and chronic leukemias, Hodgkin lymphoma, non‐Hodgkin lymphoma, and myelodysplastic syndrome (MDS) [1], [2]. Choice of allogeneic stem cell donor is dependent on donor availability, donor compatibility and health, recipient disease, and recipient condition. Successful matched sibling donor (MSD) transplants have been performed for almost 50 years [3]; however, the number of MSD transplants has remained relatively stable over the past 10 years [1]. More recently, volunteer or matched unrelated donors (MUDs), 1‐antigen mismatched unrelated donors (MMUDs), umbilical cord blood (UCB), and haploidentical donors (haplo) have emerged as alternative donors for recipients who do not have a MSD. The number of alternative donor transplants performed in the U.S. in the past decade has dramatically increased [1].

In order to determine the degree of human leukocyte antigen (HLA) match for a donor‐recipient pair, a blood sample or buccal swab must be obtained from the donor and the recipient for comparison. Human leukocyte antigen loci are located on chromosome 6 and inherited as haplotypes, one half from the mother and one half from the father. Human leukocyte antigen typing is divided into Class I antigens (HLA‐A, HLA‐B, HLA‐C) and Class II antigens (HLA‐DR, HLA‐DQ, HLA‐DP). An individual's immune system develops tolerance to its own HLA type and is intolerant of HLA types from other individuals. A fully matched donor usually refers to either an 8 out of 8 (A, B, C, DR) or 10 out of 10 (A, B, C, DR, DQ) match. A haplo donor is at a minimum a half match, 4 out of 8 alleles.

Since only 30% of U.S. patients will have a MSD, an alternative donor search is often performed for those recipients without a MSD. The likelihood of finding a fully matched (8 out of 8 alleles) unrelated donor varies depending on the ethnicity of the recipient [4], [5]. Caucasians of European descent have the greatest chance of finding a MUD, whereas the chances of finding a donor significantly decrease for African‐American or Hispanic recipients. Haplo and UCB units are viable transplant donors when a MSD or MUD are unavailable. Potential advantages of UCB are decreased relapse and graft versus host disease (GVHD); potential disadvantages are cost, and delayed immune recovery leading to increased infection (Table 1). In theory, haplo transplants are easier to perform and have lower upfront costs, but they may have a higher relapse rate, although this has not been proven in a randomized trial. Matched unrelated donors and MMUD transplants traditionally have had more GVHD. With the number of different donor graft sources possible today, nearly everyone who needs an allogeneic HSCT has a potential donor. The availability of alternative donors and advances in supportive care, infection prophylaxis and treatment, and reduced intensity conditioning allows transplants to be performed safely and effectively in many more patients than what was possible a decade ago.

Table 1. Advantages and disadvantages of graft sources.

Abbreviations: GVHD, graft versus host disease; HLA, human leukocyte antigen; TRM, transplant related mortality.

MUD and MMUD

Historical Issues

More than 70% of patients will not have a MSD. When a MSD is not available, transplant programs will often search a donor registry to select an unrelated donor. Over 25 million volunteer donors have joined the “Be the Match” donor registry since its inception in 1987 [6]. Historically, GVHD and graft failure have limited the success of MUD and MMUD transplants.

Challenges and Successes

MMUDs are usually matched at 7 out of 8 HLA loci, which increases the availability of adonor, particularly for non‐Caucasian patients [7]. Caucasians of European descent have a 75% chance of finding an 8 out of 8 MUD compared with African‐American or Hispanic patients, who have a 19%–40% chance [4]. If a 7 out of 8 MMUD is considered, the match rates increase to 82%–86% [7]. Better HLA matching techniques have improved outcomes over the last 10 years [8]. Graft versus host disease has been a major limitation of MMUD transplants [9]. Increasing genetic disparity, particularly for HLA loci, has been associated with greater risks of acute GVHD, decreased overall survival, and higher transplant‐related mortality (TRM) [10], [11], [12], [13]. Increased mortality has been seen with any mismatch at A, B, C, or DR, with similar rates of mortality no matter which loci had the mismatch [14].

Recent Results and Comparisons

Overall survival, disease‐free survival, TRM, and relapse are now similar in MSD and MUD HSCT recipients [9], [15]. A calcineurin inhibitor plus methotrexate combination for GVHD prophylaxis has decreased the rates of acute and chronic GVHD in patients who have had a MUD HSCT [16], [17]. When antithymocyte globulin was added to cyclosporine and methotrexate GVHD prophylaxis, a reduction in acute and chronic GVHD was seen, as well as decreased immunosuppression use at 1 year without an increase in morbidity or mortality [18], [19]. Additionally, a bone marrow graft source may decrease the risk of chronic GVHD in recipients who have received a MUD transplant [20]. Increased donor age has been correlated with decreased immune recovery [21] as well as an increased risk of age‐related clonal hematopoiesis [22], making a younger MUD donor potentially superior to an older MSD [23]. Controversial issues include the optimal GVHD prophylaxis and the optimal donor, including locus of mismatch, age, and gender.

Future Research

Current approaches to decrease the rates of acute GVHD and TRM in MMUD HSCT have included adding additional agents, including antithymocyte globulin [24], sirolimus [25], and bortezomib [26] to the standard calcineurin inhibitor plus methotrexate GVHD prophylaxis combination. The use of post‐transplant cyclophosphamide [27] as GVHD prophylaxis in the MUD or MMUD setting is under investigation, and appears promising. Graft versus host disease incidence and severity is reduced with post‐transplant cyclophosphamide regardless of whether a myeloablative MUD or reduced intensity conditioning (RIC) haplo transplant is performed [28]. While MUD HSCT is often the next preferred graft source after a MSD, unfortunately, many patients, particularly non‐white patients, will not have a MUD donor. MMUD HSCTs have been considered in the past when a MUD has not been available; however, given the encouraging outcomes seen with haplo HSCT with post‐transplant cyclophosphamide, and the increased morbidity associated with MMUD HSCT, MMUD transplant numbers are declining in the U.S.

While MUD HSCT is often the next preferred graft source after a MSD, unfortunately, many patients, particularly non‐white patients, will not have a MUD donor. MMUD HSCTs have been considered in the past when a MUD has not been available; however, given the encouraging outcomes seen with haplo HSCT with post‐transplant cyclophosphamide, and the increased morbidity associated with MMUD HSCT, MMUD transplant numbers are declining in the U.S.

UCB

Historical Issues

The first case report of a UCB transplant occurring in a patient with Fanconi's anemia was published in 1989 [29]. Currently, UCB transplants in adult recipients account for approximately 10% of the alternative donor transplants performed in the U.S. [1]. Umbilical cord blood units should be selected both by HLA match and by cell dose [30]. Umbilical cord blood units are readily available, have potentially decreased rates of chronic GVHD compared with other graft sources, and require less strict HLA matching than MUDs (Table 1). Disadvantages of UCB HSCT include increased risk of infectious complications, delayed engraftment, higher rates of graft rejection, increased TRM, and increased cost of unit procurement.

Challenges and Successes

The risk of graft failure in adult recipients has been partially overcome by the use of two UCB units for transplant [30], [31]. However, in pediatric transplant, there is no advantage to double unit UCB transplant over single unit UCB transplant [32]. Better supportive care and prophylaxis for viral infections have contributed to improved outcomes [33]. Additionally, the use of allele level typing has been associated with improved survival [34], [35]. Controversial issues include the use of double versus single UCB unit transplant for adults, the addition of antithymocyte globulin to the conditioning regimen, and the importance of matching at the HLA‐C locus.

Recent Results and Comparisons

Comparative trials of UCB with other donor sources for patients with hematologic malignancies are summarized in Table 2. In several studies, UCB transplants were compared with MUD and MMUD transplants in patients with acute leukemia or MDS receiving myeloablative or RIC [36], [37], [38], [39]. No difference in overall survival was seen between UCB and MUD transplants [36], [37]. Chronic GVHD rates were similar or better in UCB transplant recipients when compared with unrelated donor transplant recipients [36], [37], [38], [39]. Three studies evaluated unique subsets of patients undergoing UCB HSCT, including those patients with minimal residual disease prior to transplant [36] and elderly patients over age 50 [38, 39]. Umbilical cord blood recipients with leukemia who had minimal residual disease had comparable survival to patients with minimal residual disease who received MUD transplants and better survival than MMUD transplants [36]. When elderly patients who received a UCB HSCT were compared with patients who received a MUD or MMUD HSCT, chronic GVHD was lowest with UCB [38], [39]. Acute GVHD and TRM rates were similar or better in the UCB HSCT group compared with the MMUD HSCT group. Additionally, overall survival was similar with UCB and MMUD HSCTs, suggesting that in situations where there is not a MUD, UCB can be used safely as a graft source in patients over the age of 50 with less chronic GVHD.

Table 2. Retrospective comparative trials of umbilical cord blood transplant with matched sibling donor, matched unrelated donor, and mismatched unrelated donor (2012–2016).

^aSevere acute GVHD at 100 days.

^bAdditional UCB group with other conditioning regimens not included in table (n = 40).

Abbreviations: —, no data; @, at; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; BM, bone marrow; CML, chronic myeloid leukemia; GVHD, graft versus host disease; MAC, myeloablative conditioning; MDS, myelodysplastic syndrome; MM, multiple myeloma; MMUD, mismatched unrelated donor; MPD, myeloproliferative disorder; MSD, matched sibling donor; MUD, matched unrelated donor; NS, not significant; OS, overall survival; PB, peripheral blood; RIC, reduced intensity conditioning; TRM, transplant related mortality; UCB, umbilical cord blood; yr, year(s).

There have been several retrospective comparison studies comparing RIC UCB HSCT with other donor sources in patients with hematologic malignancies, but no prospective studies have been performed to date [40], [41], [42], [43]. Chronic GVHD rates were decreased in the UCB HSCT groups compared with the other donors. Overall survival and disease‐free survival for UCB HSCT groups were similar to MUD HSCT, because relapse and GVHD were higher in MUD transplant group and early TRM was higher in the UCB transplant group [42]. Umbilical cord blood HSCT outcomes were similar to MUD and MMUD HSCT outcomes in most of the studies presented. Hematopoietic and immunologic recovery was significantly longer in the UCB group [40], [41], [42] compared with other donor groups, increasing the length of stay for these patients [44]. These results were consistent despite conditioning regimen, type of hematologic malignancy, or age of the recipients. Umbilical cord blood transplant is a reasonable transplant alternative and should be considered when a MSD or MUD are not available.

Future Research

Several groups are studying strategies to improve engraftment (Table 3). These include expansion studies using mesenchymal stem cells [45], Notch ligand [46], and nicotinamide [47], among others. A phase III registration trial (NCT02730299) is currently in progress. Homing strategies have used novel agents, such as prostaglandin [48], intra‐marrow administration of UCB [49], selectin fucosylation [50], and oral hypoglycemic agents [51]. An interesting approach is the use of hyperbaric oxygen [52] to improve engraftment. Because UCB transplants have an increased risk of infectious complications, Heslop and colleagues are investigating trivirus‐specific cytotoxic T lymphocytes to treat viral infections [55]. Furthermore, other centers have used rituximab to decrease Epstein Barr virus (EBV) reactivation [56].

Table 3. Strategies to improve UCB engraftment.

Abbreviations: UCB, umbilical cord blood; UCBT, umbilical cord blood transplant.

Haplo

Historical Issues

Haplo donors can be from parents, children, or siblings of the transplant recipient, and, more recently, from second‐degree relatives. Advantages of haplo donors are their ready availability, less strict HLA matching, and decreased cost of graft acquisition (Table 1). Due to the HLA disparity, there may be a higher risk of GVHD and graft failure. Traditional methods to reduce GVHD have included intensive T‐cell depletion techniques that were difficult to accomplish in smaller centers [57].

Challenges and Successes

Post‐transplant cyclophosphamide, pioneered by investigators at Johns Hopkins University, produces immunologic tolerance in recipients after they receive unmanipulated haplo bone marrow transplants [27]. This advancement has decreased the risks of GVHD and graft rejection, allowing haplo transplants to become more widely utilized and leading to a marked increase in the number of haplo transplants performed in the U.S. and Europe [1]. The strategy can be done in any transplant center, unlike the older, complex T‐cell depletion techniques. A significant concern regarding haplo HSCT utilizing post‐transplant cyclophosphamide is the theoretical increase in risk of relapse. Recent data indicates when haplo HSCT recipients were risk stratified by the disease risk index, haplo HSCT relapse rates were similar to matched transplants [58]. Controversial issues in haplo HSCT include whether bone marrow or peripheral blood stem cells produce equivalent outcomes, selection of the optimal haplo donor, and choice of myeloablative regimens. Long‐term outcomes, including risks of second malignancy, are still under investigation.

Recent data indicates when haplo HSCT recipients were risk stratified by the disease risk index, haplo HSCT relapse rates were similar to matched transplants.

Recent Results and Comparisons

Several studies evaluated the outcomes of MUD transplants and haplo transplants in patients with hematologic malignancies (Table 4) [59], [60], [61], [62], [63], [64], [65]. No differences were seen between the haplo HSCT and the MUD HSCT in overall survival, disease‐free survival, TRM, and relapse in the majority of studies [59], [61], [62], [63], [64], [65]. Only one RIC study had lower TRM and higher relapse risk with haplo HSCT compared with MUD HSCT; however, this was not seen with myeloablative haplo transplant [60].

Table 4. Retrospective comparative trials of haploidentical transplant trials with matched related donor and matched unrelated donor transplant (2013–2016).

Abbreviations: —, no data; @, at; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; BM, bone marrow; CML, chronic myeloid leukemia; CLL, chronic lymphocytic leukemia; GVHD, graft versus host disease; Haplo, haploidentical transplant; MAC, myeloablative conditioning; MDS, myelodysplastic syndrome; MM, multiple myeloma; mo, months; MPD, myeloproliferative disorder; MSD, matched sibling donor; MUD, matched unrelated donor; OS, overall survival; PB, peripheral blood; RIC, reduced intensity conditioning; TRM, transplant related mortality; yr, year(s).

Studies evaluating haplo transplant and UCB transplant are summarized in Table 5; two of the studies were retrospective and one study was prospective [66], [67], [68]. A retrospective multi‐center European study evaluated transplant outcomes after haplo and UCB HSCT in acute myelogenous leukemia (AML) and acute lymphoblastic leukemia patients [66]. Umbilical cord blood transplant patients were slower to engraft and had a higher rate of graft failure. Chronic GVHD incidence was lower in the UCB transplant group compared with the haplo transplant group. No difference was seen between the two graft sources in grade II–IV acute GVHD, relapse, disease‐free survival, and TRM. In contrast, a single institution study from Italy evaluated five different types of donor transplants—MSD, MUD, MMUD, haplo, and UCB [67]. Transplant‐related mortality was lower in the haplo HSCT arm, with higher relapse rates than UCB HSCT.

Table 5. Haploidentical transplant versus umbilical cord blood transplant trials.

Abbreviations: —, no data; @, at; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; BM, bone marrow; CI, confidence interval; GVHD, graft versus host disease; Haplo, haploidentical transplant; HR, hazard ratio; MAC, myeloablative conditioning; MDS, myelodysplastic syndrome; MMUD, mismatched unrelated donor; MPD, myeloproliferative disorder; MSD, matched sibling donor; MUD, matched unrelated donor; OS, overall survival; PB, peripheral blood; RIC, reduced intensity conditioning; TRM, transplant related mortality; UCB, umbilical cord blood; yr, year(s).

Two parallel prospective independent studies of 50 patients each, one study with haplo transplants and the other with UCB transplants, were performed in the U.S. [68]. Both studies used RIC and had similar eligibility criteria. Cumulative incidence of chronic GVHD at 1 year was 25% and 13% for UCB transplants and haplo transplants, respectively [68]. Transplant‐related mortality and relapse was 24% and 31% in the UCB transplant trial and 7% and 45% in the haplo transplant trial, respectively [68]. One year disease‐free survival was similar between the two groups: 46% UCB versus 48% haplo [68]. Due to the small numbers of patients enrolled on each study, direct comparison between the two studies is impossible.

Future Research

A randomized multi‐center phase III trial is currently in progress to compare haplo versus UCB transplant (BMT CTN 1101; NCT01597778). This is a high‐priority study nationwide and will complete accrual in 2018. An additional cost effectiveness study will compare the cost and quality of life outcomes of UCB and haplo HSCT. Further efforts to decrease relapse after transplant include the use of post‐transplant maintenance, such as FLT‐3 inhibitors in AML patients with the FLT3‐ITD mutation [69]. Other strategies to decrease relapse are to study post‐transplant maintenance for other leukemias, lymphoma, and myeloma [70].

Discussion

Patients with high‐risk hematologic malignancies who have an adequate performance status should be referred to an HSCT program for a transplant evaluation. In many cases, HSCT is the only curative option. Many allogeneic HSCTs utilize RIC, expanding the possibility for curative therapy for patients into their 70s. Moreover, patients who would not have been eligible for HSCT a decade ago due to lack of MSD or MUD have alternative donors available today. Graft versus host disease remains a major toxicity of HSCT, regardless of donor source, but new treatment modalities may decrease the severity of this complication [71], [72].

How does one choose the best donor for the recipient with all the alternative donor options? Ideally, if a patient has a healthy suitable MSD available, using that donor would be the “gold standard.” However, only 30% of transplant patients will have a suitable MSD available. Often, when a MSD is unavailable, a MUD search is performed through the Be the Match donor registry. Unfortunately, patients who are not Caucasian of European descent are less likely to find a donor through the registry [4], [5]. Alternative donors, including UCB units and haplo donors, should also be considered. More research is needed to determine the ideal situations in which to use an alternative donor, especially in situations of an elderly MSD or minor haplo donor. It is important to perform these donor searches concurrently and soon after diagnosis to avoid the possibility of a delay in HSCT and potential relapse.

Conclusion

Haplo and UCB HSCT are excellent alternative donor graft sources, and current retrospective data has indicated comparable survivals. The 2‐antigen MMUD transplants have inferior outcomes compared with other donor sources due to increased GVHD and TRM, and should be avoided. Advantages of UCB grafts are rapid availability and the lower risk of chronic GVHD. Moreover, UCB may be preferable for patients with minimal residual disease. Disadvantages of using UCB are the slower engraftment, increased risk of graft failure, and the cost of obtaining two UCB units for adult recipients. Near universal and rapid availability of haplo donors make this type of alternative donor HSCT appealing. However, haplo transplants may have a higher risk of relapse. A prospective comparison between UCB and haplo donor transplants is in progress. Over the past 10–15 years, new advances in reduced intensity conditioning, infection prophylaxis and treatment, supportive care, and alternative donors have allowed patients to proceed to potentially curative HSCT, which would not have been possible just a decade ago.

Author Contributions

Conception/design: Tamila L. Kindwall‐Keller, Karen K. Ballen

Collection and/or assembly of data: Tamila L. Kindwall‐Keller, Karen K. Ballen

Manuscript writing: Tamila L. Kindwall‐Keller, Karen K. Ballen

Final approval of manuscript: Tamila L. Kindwall‐Keller, Karen K. Ballen

Disclosures

The authors indicated no financial relationships.