Different risk factors of acute and chronic graft-versus-host disease with conventional prophylaxis and posttransplantation cyclophosphamide in matched related and unrelated donor transplantations

Introduction

Graft -versus-host disease (GVHD) is one the most life-threatening complications in allogeneic stem cell transplantation (HSCT). With more the 40000 transplants per year, more than 10 thousand patients have this complication [1, 2]. Clinically signifi cant forms of GVHD occur in 20-50% of HSCT recipients, and it is associated with signifi cant mortality and morbidity reaching 30% in severe cases [3, 4]. Only a few large studies have been published with analysis of GVHD risk factors.
Flowers et al. in the cohort of 2941 related and unrelated graft recipients has demonstrated that unrelated donor, HLA mismatch, female donor in male recipient and donor age were the risk factors for both acute and chronic GVHD, while intensity of the conditioning was a predictor of acute GVHD, whereas patient age and peripheral blood stem cell (PBSC) graft ing were the predictors of the chronic GVHD. Also GVHD was less frequent in CML than in acute leukemia. Th e GVHD prophylaxis with antithymocyte immunoglobulin (ATG) did not reach statistical signifi cance [5]. High cellularity of the graft and high prevalence of CD3-positive cells in the graft was another predictor of acute GVHD with conventional prophylaxis, especially in PBSC recipients [6]. In another large study, the use of PBSC compared to bone marrow (BM) was the risk factor for both acute and chronic GVHD [7].
The Center for International Blood and Marrow Transplant Research (CIBMTR) study with data from 226 centers has identifi ed total body irradiation PBSC, ethnicity, poor performance study and positive cytomegalovirus status of donor and recipient as the signifi cant risk factors of acute GVHD [7]. Th e study also showed that ABO incompatibility was not a signifi cant factor, and the incidence of GVHD in CML is higher probably due to transplantation techniques.
The abovementioned studies were conducted relatively long ago, and were based on population of patients receiving predominantly cyclosporine and methotrexate (MTX) as prophylaxis, and ATG in unrelated donors. However, transplantation technologies have signifi cantly evolved over time. Novel prophylaxis regimens have been introduced, like mTOR inhibitors [8], posttransplant cyclophosphamide (PTCy) [9], TCR alpha/beta cell depletion [10]. No studies have been published on risk factors of GVHD with these novel approaches. In the present study, we searched for risk factors in two large cohorts of patients, one with conventional prophylaxis based on calcineurin inhibitors (CNIs) with MTX/mycophenolate mofetil (MMF) and other, with PTCy prophylaxis. Th e purpose of this study was to evaluate whether GVHD prophylaxis does change the pattern of risk
factors.

Patients and methods

One thousand thirteen adult patients transplanted at the First State I. Pavlov Medical University from 2006 to 2017 were included into the study. All the patients were graft ed either from matched related donor (32%) or unrelated donor (68%). In this group, 470 patients received prophylaxis with PTCy, and 543 were subjected to conventional prophylaxis (Table 1). Only patients who successfully engraft ed were included in the analysis.
GVHD prophylaxis under conventional regimen included tacrolimus with target concentration of 5 to 15 ng/ml, starting from day-1 until day+120, or cyclosporine A with target concentrations of 150 to 350 ng/ml, starting from day-1 until day+120. Th e second agent was either MMF 30 mg/kg (day -1 to day+30), or methotrexate 15 mg/m2 (day+1, 10 mg/m2; day +3, 6). Th e recipients of unrelated graft s did also receive ATG (ATGAM, Pfi zer, Inc.), at 20 mg/kg from day -3 until day -1. In the PTCy group, the prophylaxis consisted of single-agent cyclophosphamide (50 mg/kg) on days +3,+4 for matched related or unrelated bone marrow. In recipients of PBSCs, we used cyclophosphamide (50 mg/kg) on days +3,+4 followed by tacrolimus and MMF (30mg/kg) starting on day +5. In the mismatched graft s, the dose of MMF was increased to 45 mg/kg. Myeloablative conditioning (MAC) in conventional prophylaxis group was performed with oral busulfan (16 mg/kg), and cyclophosphamide (120 mg/kg). In the PTCy group, the majority of patients received MAC containing fl udarabine 180 mg/m2 and busulfan 14 mg/kg. Reduced-intensity conditioning (RIC) was performed with oral busulfan (8 mg/kg) and fl udarabine (180 mg/m2). Minority of patients received conditioning with melphalan (140 mg/m2) and fl udarabine (150 mg/m2. RIC was performed in patients, who were either older than 40 years, had HSCT-specifi c co-morbidity index (HCT-CI)≥2, or exhibited, at least, grade 3 hepatotoxicity during the induction therapy. Supportive care did not diff er for the two prophylaxis arms.

Statistical analysis

Th e Consensus Conference criteria were used for acute GVHD grading [11] and National Institutes of Health criteria were used for chronic GVHD grading [12]. Diagnosis of skin GVHD was established either clinically or histologically, the diagnosis of liver GVHD was assessed clinically, whereas gastrointestinal GVHD was specifi ed by pathological examination. Incidence of acute and chronic GVHD was evaluated with cumulative incidence estimates. Time frame for acute GVHD was 125 days, for chronic GVHD, 2 years. Evaluation of risk factors was performed by means of Gray test. Early discontinuation of immunosuppression due to relapse or minimal residual disease was considered a competing risk for aGVHD. Donor lymphocyte infusion was considered a competing risk for cGVHD. Multivariate evaluation and analysis of continuous variables were done using Fine and Grey regression. Th e variables were selected for the multivariate analysis in case of signifi cance <0.15 obtained in the univariate mode. Th e cutoff levels for continuous variables were determined in ROC analysis with maximal sum of sensitivity and specifi city as a criterion. Th e analyses were conducted in SAS 9.3 (SAS Institute, Inc.).

Results

The conventional prophylaxis group comprised 199 recipients of matched related graft s and 344 subjects were transplanted from unrelated donors. The PTCy group consisted of 104 matched related transplants, and 27 matched unrelated HSCTs, with single-agent PTCy prophylaxis. 338 patients received combined prophylaxis with PTCy, tacrolimus and MMF. Among the evaluated patients, 93.4% has engraft ed. Among the engraft ed patients, 436 received PTCy prophylaxis and 485, conventional GVHD prophylaxis, with a fi veyear survival of 47%. Incidence of acute GVHD in the whole group was 43.9%. Of them, 18.6% had grade I; 9.8%, grade II; 12.6%, grade III, and 3% had grade IV GVHD. Th e incidence of chronic GVHD was 31.6%, including 12.6% with mild; 9%, moderate degree, and 10% showed severe GVHD according to NIH criteria. Th e most common organs involved in chronic GVHD were skin, mucosa, eyes, gastrointestinal tract (GIT) and liver (Fig. 1). Incidence of acute GVHD in the PTCy group was 35%; grade II-IV acute GVHD, 15%; chronic GVHD, 29%. Moderate and severe chronic GVHD was registered in 18% of the cases. Similar incidence rates (resp., 53%, 35%, 35% and 30%) were noted for conventional prophylaxis.

Table 1. Patients’ characteristics in the study group

AML= acute myeloblastic leukemia; ALL =acute lymphoblastic leukemia; CML= chronic myeloid leukemia; HL= Hodgkin lymphoma; MDS= myelodysplastic syndrome; АА=aplastic anemia; NHL=non-Hodgkin lymphoma MF=myelofi brosis; MPN=myeloproliferative neoplasm; CLL=chronic lymphoid leukemia; MAC=myeloablative regimen; RIC =reduced-intensity conditioning.

Figure 1. The incidence and severity of acute (A) and chronic GVHD (B). The incidence of organ involvement is calculated only for the patients who developed GVHD

Figure 2. Risk factors of acute GVHD grade II-IV after conventional prophylaxis (A) and posttransplant cyclophosphamide (B). Number of CD34+ cells in the graft, age and BMI are continuous variables, all the others are logistic

For the conventional prophylaxis group, the following factors were revealed in the univariate analysis with signifi cance >0.15 for acute GVHD grade II-IV development: unrelated donor (p<0.0001), salvage group (p=0.0014), number of HSCT (p=0.1064), ABO incompatibility (p=0.0709), cytomegalovirus (CMV) serostatus (p=0.1487), graft source (p=0.0004 conditioning intensity (p=0.0003), alkylating agents in the conditioning (0.1377), age of the recipient (0.0002), CD34 cell number in the graft (p<0.0001), engraft - ment time (p<0.0001), diagnosis (p=0.0379), body mass index (BMI) (p=0.0220). Th ese parameters were included into the multivariate model where only unrelated donor (HR 1.86, 95%CI 1.11-3.19, p=0.0219), salvage disease status at transplant (HR 0.50, 95%CI 0.30-0.79), use of RIC (HR 0.58, 95%CI 0.40-0.85), older age (HR 0.0442, 95%CI 0.96-0.99), higher BMI (HR 0.97, 95%CI 0.97-1.00) and engraft ment before day +15 (HR 1.55, 95%CI 1.08-2.22) signifi cantly affected the incidence of acute GVHD grade II-IV (Figure 2A). Th e BMI cut-off value was 28 kg/m2 indicating that obese patients had less acute GVHD. Despite common risk factor of graft source, the fast engraft ment was more signifi cant than the graft source factor (p=0.48), despite the fact that fast engraft ment occurred more oft en in the PBSC recipients (32% vs 17%, p<0.0001).
In the univariate analysis of PTCy group, only unrelated donor (p=0.0170), HLA matching (p=0.0347), number of HSCT (p=0.0592), recipient gender (p=0.0592), CMV serostatus (p=0.0592), female donor for male recipient (p=0.0706), time of engraft ment (p=0.0037), and time from diagnosis to transplant were shown to be signifi cant factors (p<0.0001). Since the recipient gender was signifi cant because of female-male combination, only that factor was included in the multivariate analysis. Also in ROC analysis, the cut-off for engraft ment time was diff erent: 20 days instead of 15. Th ese parameters were added to the multivariate model where CMV serostatus was the only signifi cant factor (HR 0.71, 95%CI 0.54-0.95, p=0.0251). Th e highest incidence was in the -/- pair of donor / recipient (32%), lower in the +/- pair (20%) and the lowest in the CMV-positive recipients (13% with +/+ pair vs 15% with -/+ pair). Th e univariate analysis in conventional prophylaxis group revealed unrelated donor (p=0.0002), salvage group (p=0.0635), cytomegalovirus serostatus (p=0.0248), graft source (p<0.0001), age (p=0.1360), number of CD34 cells in the graft (p=0.0047), engraft ment at <15 days (p=0.0013), time from diagnosis to transplantation (p=0.1151), diagnosis (p=0.0205) and previous acute GVHD (p<.0001) as signifi cant factors for moderate and severe chronic GVHD. In the PTCy group, the univariate analysis revealed unrelated donor (p=0.0980), donor gender (p=0.0138), graft source (p=0.0805), male recipient with female donor (p=0.1082), number of CD34+ cells in the graft (p=0.0272), time to engraft ment (p=0.0302) and previous acute GVHD (p=0.0929), as predisposing factors for chronic GVHD. In the multivariate model with conventional GVHD prophylaxis, only PBSC graft (HR 2.26, 95%CI 1.28-4.11) and previous acute GVHD (HR 3.76, 95%CI 2.32-6.37) were signifi cant risk factors for moderate and severe chronic GVHD (Figure 3A).

Figure 3. Risk factors for moderate and severe chronic GVHD with conventional prophylaxis (A) and posttransplantation cyclophosphamide (B)

Modeling in the PTCy group demonstrated only a weak statistical signifi cance for previous acute GVHD (HR 1.59, 95%CI 0.99-2.54), while all the other factors were non-significant (Figure 3B).
Regarding mild acute GVHD (grade I-II) which is usually favorable for prognosis in the conventional prophylaxis group, only the CD34 cell dose increased the probability of this condition (HR 1.08, 95%CI 1.01-1.150, р=0.0133). The other variables were not signifi cantly different. With PTCy prophylaxis, the unrelated donorship was associated with increased probability of grade I-II acute GVHD (HR 3.26, 95%CI 1.26-8.39, p=0.0145). Also a combination of a female donor/ male recipient had week statistical signifi cance (HR 1.87, 0.94-3.72, p=0.0761). No predictors were determined for mild chronic GVHD in conventional prophylaxis patients, whereas, with PTCy, the CMV-positive recipient serostatus was protective against this condition (HR 0.675, 0.496-0.918, p=0.0123). Mild chronic GVHD was lowest in +/+ CMV positive donor/ recipient (6%), being highest in -/+ (15%) and +/- (19%) combinations. In the both CMV-negative pairs, mild chronic GVHD rate was also substantial (14%).

Discussion

In this relatively large study, we have confi rmed that the novel prophylaxis regimens may dramatically change the landscape of risk factors which was not demonstrated before. Previous registry studies mostly documented only evolutional changes in the risk factors due to other aspects of HSCT. In the era of only BM transplantation from matched siblings with cyclosporine and methotrexate as prophylaxis, the predominant risk factors were female donor for male recipient, pregnancy history and older recipient age [13]. Th e subsequent CIBMTR study identifi ed the risk factors of PBSC use, ethnicity, TBI versus busulfan-based conditioning, and positive CMV serostatus [7]. Aft er broad introduction of unrelated transplants, it became obvious that GVHD incidence is higher than aft er sibling transplants [14]. Additional risks of GVHD are associated with partial HLA mismatches [15] and non-HLA allele mismatches [16]. Furthermore, the donor age was also identifi ed as risk factor in unrelated HSCT [17]. Nonetheless the recent mathematical analysis indicates that the predictive potential of clinical parameters is relatively low [18]. Thus, the risk factors of GVHD were slowly evolving, due to implementation of novel cell sources and donor types. We, however, confi rm that the use of PTCy completely abolished the previously signifi cant risk factors. It has been previously published that HLA matching is not a signifi cant factor with PTCy prophylaxis [19], and this was confirmed in the current study. Nonetheless, the diff erence between matched sibling and unrelated donor had a tendency to signifi cance, which was also confirmed in our group of patients. What was not established earlier is the preventive
role of CMV-positive serology in recipient, despite a weak statistical trend in the EBMT study [20]. Th e probable reason for that is diff erent prevalence of CMV seropositivity in Russia, Europe and the USA. In Russia, the CMV seroprevalence is above 85% [21, 22, 23]. Th e CMV seropositivity is unlikely to represent the reason for diff erences, but, rather, it may be a consequence of changes in immune system that, probably, led to decreased GVHD incidence. It was demonstrated that CMV causes expansion of T-regulatory cells [24] and upregulation of IL-33 pathway, which protects against lethal GVHD in animal models [25, 26]. Th e signifi cance of this factor with no such evidence for conventional prophylaxis [7] indicates the presence of different immunological mechanisms behind PTCy prophylaxis that still should be elucidated.
The risk factors identifi ed for moderate and severe chronic GVHD with conventional prophylaxis were similar to the ones previously reported [27, 28]. Th e history of severe acute GVHD was the most predominant risk factor. However, no risk factors were identifi ed for PTCy, probably due to low incidence of this complication and low incidence of preceding acute GVHD in the study cohort. Contrary to this data, the European Registry Study defi ned recipient age, use of PBSC and combination prophylaxis as the risk factors [20]. Th e diff erences might be due to diff erent PTCy schedule (day +3, +5), use of cyclosporine instead of tacrolimus, duration
of immunosuppression [29]. Th e absence of diff erences between PBSC and BM is explained by single-agent PTCy prophylaxis in the matched bone marrow group and combination with tacrolimus and MMF in the PBSC group, which alleviated the diff erences.
In conclusion, this study identified the changing pattern of GVHD risk factors with introduction of novel prophylaxis regimens in related and unrelated HSCT graft s. Further studies are required to elucidate the biological mechanisms behind these changes.

Conflicts of interest

No conflicts of interest are reported by the authors.

References

1. Passweg JR, Baldomero H, Bader P, Basak GW, Bonini C, Duarte R, Dufour C, Kroeger N, Kuball J, Lankester A, Montoto S, Nagler A, Snowden JA, Styczynski J, Mohty M. Is the use of unrelated donor transplantation leveling off in Europe? Th e 2016 European Society for Blood and Marrow Transplant activity survey report. Bone Marrow Transplant. 2018. doi: 10.1038/s41409-018-0153-1 [Epub ahead of print].

2. D'Souza A, Fretham C. Current uses and outcomes of hematopoietic cell transplantation (HCT): CIBMTR summary slides, 2017. Available at: www.cibmtr.org, as of 28/07/18.

3. Koc S, Leisenring W, Flowers ME, Anasetti C, Deeg HJ, Nash RA, Sanders JE, Witherspoon RP, Storb R, Appelbaum FR, Martin PJ. Th erapy for chronic graft -versus-host disease: a randomized trial comparing cyclosporine plus prednisone versus prednisone alone. Blood. 2002;100(1):48-51.

4. Saliba RM, Couriel DR, Giralt S, Rondon G, Okoroji GJ, Rashid A, Champlin RE, Alousi AM. Prognostic value of response aft er upfront therapy for acute GVHD. Bone Marrow Transplant. 2012;47(1):125-131.

5. Flowers ME, Inamoto Y, Carpenter PA, Lee SJ, Kiem HP, Petersdorf EW, Pereira SE, Nash RA, Mielcarek M, Fero ML, Warren EH, Sanders JE, Storb RF, Appelbaum FR, Storer BE, Martin PJ. Comparative analysis of risk factors for acute graft -versus-host disease and for chronic graft -versus-host disease according to National Institutes of Health consensus criteria. Blood. 2011;117(11):3214-3219.

6. Czerw T, Labopin M, Schmid C, Cornelissen JJ, Chevallier P, Blaise D, Kuball J, Vigouroux S, Garban F, Lioure B, Fegueux N, Clement L, Sandstedt A, Maertens J, Guillerm G, Bordessoule D, Mohty M, Nagler A. High CD3+ and CD34+ peripheral blood stem cell graft s content is associated with increased risk of graft -versus-host disease without benefi cial eff ect on disease control aft er reduced-intensity conditioning allogeneic transplantation from matched unrelated donors for acute myeloid leukemia - an analysis from the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation. Oncotarget. 2016;7(19):27255-27266.

7. Hahn T, McCarthy PL Jr, Zhang MJ, Wang D, Arora M, Frangoul H, Gale RP, Hale GA, Horan J, Isola L, Maziarz RT, van Rood JJ, Gupta V, Halter J, Reddy V, Tiberghien P, Litzow M, Anasetti C, Pavletic S, Ringden O. Risk factors for acute graft -versus-host disease aft er human leukocyte antigen- identical sibling transplants for adults with leukemia. J Clin Oncol. 2008;26(35):5728-34.

8. Cutler C, Antin JH. Sirolimus for GVHD prophylaxis in allogeneic stem cell transplantation. Bone Marrow Transplant. 2004;34(6):471-476.

9. Luznik L, Bolanos-Meade J, Zahurak M, Chen AR, Smith BD, Brodsky R, Huff CA, Borrello I, Matsui W, Powell JD, Kasamon Y, Goodman SN, Hess A, Levitsky HI, Ambinder RF, Jones RJ, Fuchs EJ. High-dose cyclophosphamide as single-agent, short-course prophylaxis of graft -versus-host disease. Blood. 2010;115(16):3224-3230.

10. Chaleff S, Otto M, Barfi eld RC, Leimig T, Iyengar R, Martin J, Holiday M, Houston J, Geiger T, Huppert V, Handgretinger R. A large-scale method for the selective depletion of alphabeta T lymphocytes from PBSC for allogeneic transplantation. Cytotherapy. 2007;9(8):746-754.

11. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J, Th omas ED. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant. 1995;15:825-828.

12. Filipovich AH, Weisdorf D, Pavletic S, Socie G, Wingard JR, Lee SJ, Martin P, Chien J, Przepiorka D, Couriel D, Cowen EW, Dinndorf P, Farrell A, Hartzman R, Henslee-Downey J, Jacobsohn D, McDonald G, Mittleman B, Rizzo JD, Robinson M, Schubert M, Schultz K, Shulman H, Turner M, Vogelsang G, Flowers ME. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft -versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant. 2005;11:945-956.

13. Gale RP, Bortin MM, van Bekkum DW, Biggs JC, Dicke KA, Gluckman E, Good RA, Hoff mann RG, Kay HE, Kersey JH, et al. Risk factors for acute graft -versus-host disease. Br J Haematol. 1987;67(4):397-406.

14. Bradley BA, Hows JM, Gore SM, Bidwell JL, Clay T, Downie TR, Gluckman E, Howard MR, Laundy GJ. Current status of unrelated-donor bone marrow transplantation. Th e International Marrow Unrelated Search and Transplant (IMUST) Study. Clin Transpl. 1992:91-107.

15. Morishima Y, Sasazuki T, Inoko H, Juji T, Akaza T, Yamamoto K, Ishikawa Y, Kato S, Sao H, Sakamaki H, Kawa K, Hamajima N, Asano S, Kodera Y. Th e clinical signifi cance of human leukocyte antigen (HLA) allele compatibility in patients receiving a marrow transplant from serologically HLA-A, HLA-B, and HLA-DR matched unrelated donors. Blood. 2002; 99(11):4200-4216.

16. Chukhlovin AB. Beyond HLA system: non-HLA gene alleles of donor origin may infl uence risk of immune allo- HSCT complications. Cell Th er Transplant. 2017; 6(2): 36-51.

17. Kollman C, Howe CW, Anasetti C, Antin JH, Davies SM, Filipovich AH, Hegland J, Kamani N, Kernan NA, King R, Ratanatharathorn V, Weisdorf D, Confer DL. Donor characteristics as risk factors in recipients aft er transplantation of bone marrow from unrelated donors: the eff ect of donor age. Blood. 2001;98(7):2043-2051.

18. Lee C, Haneuse S, Wang HL, Rose S, Spellman SR, Verneris M, Hsu KC, Fleischhauer K, Lee SJ, Abdi R. Prediction of absolute risk of acute graft -versus-host disease following hematopoietic cell transplantation. PLoS One. 2018;13(1):e0190610.

19. Kasamon YL, Luznik L, Leff ell MS, Kowalski J, Tsai HL, Bolaños-Meade J, Morris LE, Crilley PA, O'Donnell PV, Rossiter N, Huff CA, Brodsky RA, Matsui WH, Swinnen LJ, Borrello I, Powell JD, Ambinder RF, Jones RJ, Fuchs EJ. Nonmyeloablative HLA-haploidentical bone marrow transplantation with high-dose posttransplantation cyclophosphamide: eff ect of HLA disparity on outcome. Biol Blood Marrow Transplant. 2010;16(4):482-489.

20. Ruggeri A, Labopin M, Bacigalupo A, Afanasyev B, Cornelissen JJ, Elmaagacli A, Itala-Remes M, Blaise D, Meijer E, Koc Y, Milpied N, Schouten HC, Kroeger N, Mohty M, Nagler A. Post-transplant cyclophosphamide for graft -versushost disease prophylaxis in HLA matched sibling or matched unrelated donor transplant for patients with acute leukemia, on behalf of ALWP-EBMT. J Hematol Oncol. 2018;11(1):40. doi: 10.1186/s13045-018-0586-4.

21. Lantos PM, Hoff man K, Permar SR,, Jackson P, Hughes BL, Swamy GK. Geographic disparities in cytomegalovirus infection during pregnancy. J Pediatric Infect Dis Soc. 2017;6(3):e55-e61.

22. Lachmann R, Loenenbach A,, Waterboer T, Brenner N, Pawlita M, Michel A, Th amm M, Poethko-Müller C, Wichmann O, Wiese-Posselt M. Cytomegalovirus (CMV) seroprevalence in the adult population of Germany. PLoS One. 2018;13(7):e0200267.

23. Kisteneva LB. Clinical and laboratory features of cytomegalovirus and NS viral infection in pregnant women and newborns: Development of therapeutic and prophylactic measures. 2001. Th e Ph.D. Th esis, Moscow (In Russian).

24. Almanan M, Raynor J, Sholl A, Wang M, Chougnet C, Cardin RD, Hildeman DA. Tissue-specific control of latent CMV reactivation by regulatory T cells. PLoS Pathog. 2017;13(8):e1006507.

25. Popovic B, Golemac M, Podlech J, Zeleznjak J, Bilic-Zulle L, Lukic ML, Cicin-Sain L,, Reddehase MJ, Sparwasser T, Krmpotic A, Jonjic S. IL-33/ST2 pathway drives regulatory T cell-dependent suppression of liver damage upon cytomegalovirus infection. PLoS Pathog. 2017;13(4):e1006345.

26. Matta BM, Reichenbach DK, Zhang X, Mathews L, Koehn BH, Dwyer GK, Lott JM, Uhl FM, Pfeifer D, Feser CJ, Smith MJ, Liu Q, Zeiser R, Blazar BR, Turnquist HR. Peri-alloHCT IL-33 administration expands recipient T-regulatory cells that protect mice against acute GVHD. Blood. 2016;128(3):427-439.

27. Grube M, Holler E, Weber D, Holler B, Herr W, Wolff D.Risk factors and outcome of chronic graft -versus-host disease after allogeneic stem cell transplantation: results from a single-center observational study. Biol Blood Marrow Transplant. 2016;22(10):1781-1791.

28. Remberger M, Kumlien G, Aschan J, Barkholt L, Hentschke P, Ljungman P, Mattsson J, Svennilson J, Ringden O. Risk factors for moderate-to-severe chronic graft -versushost disease aft er allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2002;8(12):674-682.

29. Chiusolo P, Bug G, Olivieri A, Brune M, Mordini N, Alessandrino PE, Dominietto A, Raiola AM, Di Grazia C, Gualandi F, Van Lint MT, Ferrara F, Finizio O, Angelucci E, Bacigalupo A. A Modifi ed post-transplant cyclophosphamide regimen, for unmanipulated haploidentical marrow transplantation, in acute myeloid leukemia: a multicenter study. Biol Blood Marrow Transplant. 2018;24(6):1243-1249.