- Email: [email protected]
Cytomegalovirus Status and the Outcome of T Cell–Replete Reduced-Intensity Allogeneic Hematopoietic Stem Cell Transplantation
Biol Blood Marrow Transplant 22 (2016) 1883–1887
Biology of Blood and Marrow Transplantation j o u r n a l h o m e p a g e : w w w. b b m t . o r g
Cytomegalovirus Status and the Outcome of T Cell–Replete Reduced-Intensity Allogeneic Hematopoietic Stem Cell Transplantation Frans M. Verduyn Lunel 1,*, Reinier Raymakers 2, Anette van Dijk 2, Lotte van der Wagen 2,3, Monique C. Minnema 2,3, Jurgen Kuball 2,3 1
Department of Medical Microbiology, Eijkman Winkler Institute for Microbiology, Infectious Diseases and Inflammation, Utrecht, The Netherlands Department of Hematology, Utrecht University Medical Center, Utrecht, The Netherlands 3 Laboratory of Translational Immunology, Utrecht University Medical Center, Utrecht, The Netherlands 2
Article history: Received 27 May 2016 Accepted 13 July 2016 Keywords: Cytomegalovirus Reduced intensity Stem cell transplantation
A B S T R A C T Cytomegalovirus (CMV) serostatus of donor and recipient are frequently used in algorithms of donor selection, whereas the impact of CMV reactivation on transplantation-related mortality, leukemia control, and overall survival (OS) remains controversial. Therefore, we retrospectively studied the impact of latent or active CMV infections on the outcome and occurrence of graft-versus-host disease (GVHD) after reduced-intensity conditioning (RIC) allogeneic hematopoietic stem cell transplantation (SCT) in 294 patients during the period from 2004 to 2010. CMV viral load was routinely monitored in plasma using a quantitative PCR. Preemptive antiviral therapy was initiated when the viral load in plasma exceeded a predefined threshold. In a proportional hazards model, a seropositive recipient was significantly associated with increased occurrence of acute GVHD. However the CMV serostatus of both recipient and donor and the presence of active CMV infection was not associated with the occurrence of relapses, chronic GVHD, or OS. We conclude that in the presence of viral load monitoring and preemptive treatment, latent or active CMV infection does not substantially affect the OS after T cell–replete RIC allogeneic SCT. © 2016 American Society for Blood and Marrow Transplantation.
INTRODUCTION Infections with cytomegalovirus (CMV) have been reported to impact donor engraftment [1] and to cause severe diseases after allogeneic hematopoietic stem cell transplantation (SCT), including pneumonia, hepatitis, gastroenteritis, retinitis, and encephalitis [2]. CMV has also been associated with an increased risk for acute graft-versus-host disease (GVHD) in patients receiving T cell–depleted grafts and for chronic GVHD [3]. Because of this detrimental impact of CMV disease on the prognosis of patients after SCT, preventive measures against active CMV infection after hematopoietic SCT were introduced to allow early interventions and to reduce mortality and morbidity mediated through CMV-associated diseases. Early interventions include the preemptive antiviral therapy
Financial disclosure: See Acknowledgments on page 1886. * Correspondence and reprint requests: Frans M. Verduyn Lunel, MD, PhD, Department of Virology, Eijkman Winkler Institute for Microbiology, Infectious Diseases and Inflammation, University Medical Center, G04.614, P.O. Box 85500, 3508 GA Utrecht, The Netherlands. E-mail address: [email protected] (F.M. Verduyn Lunel).
http://dx.doi.org/10.1016/j.bbmt.2016.07.009 1083-8791/© 2016 American Society for Blood and Marrow Transplantation.
with, for example, ganciclovir and valganciclovir as soon as the viral load in plasma or whole blood exceeds a predefined cut-off value during close monitoring [3]. Despite intervention strategies, a number of studies still reported on a negative impact of a positive CMV serostatus or active CMV infections on the outcome [4-6]. However, methodology including conditioning regimens and posttransplant patient management differ extensively between studies. Furthermore, the use of different techniques for the detection of CMV varies considerably. PCR techniques for detecting CMV viral load in plasma allows a more sensitive monitoring of the burden of CMV infection compared with the use of the pp65 antigenemia assay [7]. This difference may result in an earlier start of preemptive therapy in centers using PCR as gold standard. We therefore undertook a retrospective study in a very homogenous transplantation cohort undergoing exclusively reduced-intensity conditioning (RIC) while being collectively monitored by quantitative PCR techniques and treated with defined anti-CMV interventions. This cohort allowed addressing whether latent or active CMV infection in the recipient and the CMV serostatus of the donor or other
1884
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
factors contributed to the outcome after a defined type of transplantation and assessing whether CMV status of the donor is correctly placed early in donor selection algorithms. METHODS All patients 17 years and older who underwent a RIC allogeneic SCT during the period from 2004 to 2010 were included for analysis. All treatments and monitoring were standard of care, and all patients provided informed consent in line with JACIE (Joint Accreditation Committee ISCT & EBMT) guidelines to the transplantation as well as transplantation-related assessments and in accordance with the regulations of the medical ethics committee of University Medical Center Utrecht. Data were retrieved from patient record files and the electronic database from the department of Medical Microbiology. Allogeneic SCT included transplantation of grafts from matched unrelated donors (MUDs) and siblings. RIC regimen included fludarabine, antithymocyte globulin, and low-dose total body irradiation with 2 Gy. In case of allogeneic SCT for multiple myeloma after high dose melphalan and autologous SCT, a total body irradiation regimen with only .2 Gy was used. After allogeneic stem cell transplantation, GVHD prophylaxis consisted of cyclosporine A and/or mycophenolate mofetil. Patients were subjected to a routine screening protocol, which included the collection of plasma samples twice weekly after the transplantation during admission at the department of hematology followed by once weekly collection thereafter during follow-up visits at the outpatient clinic. End of follow-up of patients was due to death, or 120 days after SCT and increased to 180 days when active CMV infection was diagnosed before 120 days after SCT, or when antithymocyte globulin had been used in the conditioning regimen. Followup was continued thereafter during the treatment of acute or chronic GVHD. Plasma samples were stored at −80°C until testing. IgG antibodies against CMV were detected using the Enzygnost CMV/IgG (Siemens Healthcare Diagnostic Products, Marburg, Germany). CMV viral load was determined in EDTA-plasma samples using a quantitative PCR based on a real-time TaqMan CMV-DNA PCR assay as described previously [8]. Active CMV infection was defined as a viral load > 100 cps/mL, because artifacts in viral load determination could not be excluded when the viral load was below 100 cps/ mL; also, a reliable quantification was not possible below this threshold. Viral loads beyond that threshold were in the linear range of the assay used and could reliably and reproducibly be quantified. Primary infection and reactivation were defined as an active CMV infection occurring in an initial seronegative or seropositive recipient, respectively. Ganciclovir i.v. 5 mg/ kg, oral valganciclovir 900 mg orally twice daily, or foscarnet i.v. 60 mg/kg twice daily was given preemptively when the CMV viral load in plasma exceeded the threshold of 500 cps/mL. Dosages were adjusted when renal function was impaired. Statistical analysis was performed using the Mann-Whitney U test for comparing continuous variables and the chi-square test for noncontinuous variables. A Cox proportional hazards model, with active CMV as a timedependent variable because active CMV occurred at different time points during the follow up period after SCT, was used to analyze significant associations between variables and survival, the occurrence of acute or chronic GVHD, or relapses of disease. Variables with a Type I error rate < .2 in the univariate analysis were included in the multivariate model. Kaplan-Meier curves were constructed for analysis of overall survival (OS) during followup. The statistical software package R [9] was used to perform the calculations.
RESULTS Patient Characteristics Data from the entire follow-up period were available for all 294 patients who underwent an allogeneic hematopoietic SCT during the study period. Underlying diseases are summarized in Table 1. The median age was 55 years (range, 17 to 69). The median follow-up period was 1278 days (range, 27 to 3519) after SCT. IgG antibodies against CMV were detected in 176 patients (60%) before transplantation. According to their serostatus, patients and their donors could be subdivided into 4 groups: seronegative recipient and donor (R−/ D−, n = 64), seronegative recipient and seropositive donor (R−/ D+, n = 54), seropositive recipient and a seronegative donor (R+/D−, n = 46), and a seropositive recipient and donor (R+/ D+, n = 130). In total, 5571 plasma samples were collected from these patients for determination of the CMV viral load. Active CMV infection was observed in 102 patients (35%), including 9 (9%) with primary infections. In 77 of these patients (75%), viral
Table 1 Patient Characteristics (N = 294) Characteristic Median age, yr (range) Diagnosis type MM AML NHL MDS CLL ALL CML SAA Hodgkin disease Miscellaneous* Myelofibrosis Recipient CMV seropositive Active CMV infection R/D pairs R−/D− R−/D+ R+/D− R+/D+ Antithymocyte globulin conditioning Acute GVHD Grade I Grade II Grade III Grade IV Present, grade unknown Chronic GVHD Cord blood graft Donortype Related Unrelated Relapse disease Death
Value 55 (17-69) 77 (26) 65 (22) 49 (17) 27 (9) 20 (7) 15 (5) 12 (4) 12 (4) 8 (3) 6 (2) 3 (1) 176 (60) 102 (35) 65 (22) 54 (18) 46 (16) 130 (44) 155 (53) 41 (14) 58 (20) 23 (8) 10 (3) 1 (.3) 114 (39) 6 (2) 129 (44) 165 (56) 74 (25) 166 (56)
Values are number of cases with percents in parentheses, unless otherwise noted. MM indicates multiple myeloma; NHL, non-Hodgkin lymphoma; MDS, myelodysplastic syndrome; CLL, chronic lymphatic leukemia; ALL, acute lymphatic leukemia; CML, chronic myeloid leukemia; SAA, severe aplastic anemia. * Miscellaneous includes chronic promyelocytic leukemia (n = 1), hypereosinophilic syndrome (n = 2), chronic Epstein-Barr virus infection with secondary aplasia (n = 2), and undifferentiated leukemia, amyloidosis (n = 1).
load increased beyond the predefined threshold for preemptive treatment, whereas in 83 seropositive patients without active CMV infection, CMV DNA was not detected in plasma, not even in values below 100 copies/mL. The highest observed median viral load was 1338 copies/mL (range, 115 to 2,180,000). The difference between mean peak viral load during primary infections and reactivations, 1230 copies/ mL and 1394 copies/mL, respectively, was not significant (P = .80). Patients with an active CMV infection were slightly but not significantly older than patients without an active CMV infection, with a mean age of 53 years and 52 years, respectively. Initiation of active CMV infection was observed between 1 and 985 days (median, 40 days) after SCT. Risk Factors for OS Overall mortality during the entire follow-up period was 166 of 294 patients (56%). Mortality was similar in both patients with and without an active CMV infection, where 60 of 102 patients (59%) and 106 of 192 patients (55%), respectively, eventually died (chi-square test: P = .66). In contrast to previous reports [6,10,11], when constructing KaplanMeier curves, 5-year OS was not significantly different between recipients who were seropositive at the time of SCT when compared with recipients who were seronegative (P = .12) and between patients with or without an active CMV
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
1885
Figure 1. Survival curves. (A) Serostatus of the recipient. (B) Active CMV or no active CMV after SCT. (C) Survival according to different recipient and donor serostatus combinations.
infection (P = .96). Also, no significant differences in 5-year OS were observed between the different subgroups R−/D−, R−/ D+, R+/D−, and R+/D+ (P = .39, Figure 1). In a Cox proportional hazards model with active CMV as a time-dependent variable, chronic GVHD was significantly associated with improved OS (hazard ratio [HR], .7; 95% confidence interval [CI], .5 to .9), whereas the use of a graft from a MUD was significantly associated with a reduced OS (HR, 1.9; 95% CI, 1.3 to 2.8) (Table 2). Impact of CMV Serostatus and Active Infection on Relapse or Nonrelapse Mortality Equivalent OS of all subgroups could be a consequence of improved leukemia control in combination with increased mortality after active CMV infection, thus neutralizing the net effect. Therefore, relapse of underlying disease and nonrelapse mortality were analyzed separately. Relapse of underlying disease occurred in 74 patients (25%). Occurrence of relapse could not be retrieved in 11 patients who were subsequently excluded from the analysis. However, neither serostatus nor active CMV infection had an impact
Table 2 Significant Variables with the Conditions and Their Hazard Ratios Using a Cox Proportional Hazards Model with Active CMV Infection as a TimeDependent Variable Condition
Variable
HR
95% CI
Overall mortality
Chronic GVHD Unrelated donor Chronic GVHD Unrelated donor Recipient CMV seropositive Unrelated donor Antithymocyte globulin Chronic GVHD Chronic GVHD
.7 1.9 .7 2.2 1.7 2.2 .5 .3 .5
.5-.9 1.3-2.8 .5-1.0 1.4-3.3 1.2-2.5 1.4-3.4 .3-.7 .2-.6 .2-1.1
NRM Acute GVHD Chronic GVHD Relapse Relapse of AML
on the occurrence of relapse in a Cox proportional hazards model. Only the occurrence of chronic GVHD was significantly associated with a decreased occurrence of relapses (HR, .3; 95% CI, .2 to .6) (Table 2), most likely because chronic GVHD in patients with defined National Institutes of Health scoring (in 86% of patients) was in most patients only mild (28%) to moderate (41%) and only in some patients severe (30%) and did therefore not substantially affect OS but did allow a graftversus-leukemia effect. Because impact of active CMV infections on the occurrence of relapses was mainly observed by others for patients with acute myeloid leukemia (AML) [12-14], we therefore performed a separate analysis on this subgroup of 62 assessable AML patients. However, again, only chronic GVHD was associated with a decreased number of relapses, although the level of significance was not reached in this subgroup of patients (HR, .5; 95% CI, .2 to 1.1). In line with the observation that CMV-related factors did not contribute to leukemia relapse, no significant contribution to nonrelapse mortality was observed. Chronic GVHD was associated with a significant decrease (HR, .7; 95% CI, .5 to 1) whereas the use of a graft from a MUD was associated with an increase (HR, 2.2; 95% CI, 1.4 to 3.3) in nonrelapse mortality in a Cox proportional hazards model (Table 2). CMV Infection and GVHD Although OS was similar in all CMV serogroups and also CMV infection did not impact OS, quality of life could be heavily impaired because of the induction of, predominantly chronic, GVHD by CMV infection [12,15]. Associations between CMV and the occurrence of GVHD have been observed in previous studies [6,13,14]. Therefore, we analyzed the impact of latent or active CMV infection on acute and chronic GVHD. Acute GVHD was observed in 133 patients (45%), including 92 (69%) with grade II or higher. Active CMV infections were observed in 54 patients. Acute GVHD was reported for
1886
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
19 patients after development of active CMV infections (range, 1 to 106 days), thus suggesting that reactivation is not always elicited simply by immunosuppressive treatment of acute GVHD. In a Cox proportional hazards model, significant associations were only observed for a positive serostatus of the recipient (HR, 1.7; 95% CI, 1.2 to 2.5) (Table 2). One hundred fourteen patients (39%) suffered from chronic GVHD. Use of a graft from a MUD gave significantly higher rates of chronic GVHD (HR, 2.2; 95% CI, 1.4 to 3.4), whereas antithymocyte globulin in the conditioning regimen was protective (HR, .5; 95% CI, .3 to .7) (Table 2). Thus, only latency CMV infection, defined as a positive serostatus of the recipient, was associated with a significant higher occurrence of acute GVHD, whereas all other factors did not impact acute or chronic GVHD.
DISCUSSION Main findings of our study are that in contrast to previous reports, CMV status of both recipient and donor and active CMV infection do not necessarily associate with OS. One possible explanation for equivalent transplantation mortality in patients with and without CMV infection in our current cohort as compared with previous reports could arise from novel techniques for CMV monitoring. Although historically pp65 antigenemia was used as the gold standard for diagnosis of active CMV infections and treatment, our cohort has been monitored for active CMV infections by PCR. This usually results in both earlier diagnosis and start of preemptive therapy of patients. An alternative explanation could be that most analyses on the impact of CMV have mainly been performed in cohorts with myeloablative conditioning [16,17]. Our patients have been exclusively transplanted with nonmyeloablative conditioning. The very severely immunecompromised host after myeloablative therapy is most likely prone to more severe course of disease. The relatively modest sample size of our cohort poses another limitation, because huge data cohorts have been required for inclusion in the European Group for Blood and Marrow Transplantation database to suggest a link between CMV infection and OS. In the present study we were also not able to detect a direct association between CMV infection and the occurrence of relapses. The protective role of active CMV infections against relapses is controversial. This relation has been suggested in publications using both pp65 antigenemia and quantitative PCR to monitor CMV infections [18-21]. However, a recently published retrospective study using the Center for International Blood and Marrow Transplant Research database, was not able to show a reduced incidence of relapse in AML patients, even though most patients were transplanted after myeloablative conditioning [22]. In pediatric cohorts, using quantitative PCR for CMV monitoring also showed no altered risk for relapses when active infections occurred [23]. However, in another pediatric cohort using pp65 antigenemia for monitoring, a decreased incidence of relapses was reported [24]. Again, differences could be explained by variations in methodology and patient management, including myeloablative conditioning, between these and our cohort. Type of conditioning seems to be a crucial factor. In SCT patients with AML using PCR for monitoring CMV viral load, the association between reactivation of CMV during the first 100 days after transplantation and the incidence of relapse has been shown in myeloablative conditioned patients but not in patients who were transplanted after RIC regimens [25].
A possible biologic explanation for the absence of associations between CMV infection and relapses after RIC SCT may be that reconstitution of adaptive immunity occurs earlier in RIC when compared with myeloablative conditioning. This results in earlier control of opportunistic viral infections, which in turn may result in less prominent expansion of virusdriven natural killer cells [26] and γδ T cells from the young immune system with cross-reactive immunity against leukemic cells [27]. Today, many different aspects need to be considered when choosing the best possible donor, and CMV donor status still has quite a substantial impact in many different algorithms. Our data suggest that CMV donor status can be under certain circumstances neglected. In particular, in T cell–replete transplantation with a PCR-based early intervention strategy, active CMV infection does no harm. Vice versa, it is interesting to speculate that allowing “controlled” CMV infections could be a novel strategy to induce a more potent antileukemia activity. Therefore, a higher threshold before initiating CMV treatment might actually be beneficial for the patient. In particular, the use of low toxicity myeloablative regimens might provide an alternative window for an early innate immune recovery during a more controlled CMV infection, with even higher thresholds to be allowed before interventions are initiated. In summary, CMV status of the recipient and donor did not contribute to overall outcome in T cell–replete transplantations performed with a RIC protocol. CMV serostatus of the recipient or the presence or absence of active CMV infection does not have a significant impact on OS or occurrence of disease relapse after SCT when preemptive therapy is initiated early. This supports alternative donor selection algorithms for this type of transplants and opens also the avenue toward controlled later antiviral interventions to allow a well-balanced and potentially protective immune response cross-reactive toward CMV and leukemia.
ACKNOWLEDGMENTS Financial disclosure: The authors have nothing to disclose. Conflict of interest statement: J.K. is the co-founder and chief scientific officer of Gadeta (http://www.gadeta.nl/).
REFERENCES 1. Steffens H-P, Podlech J, Kurz S, Angele P, Dreis D, Reddehase MJ. Cytomegalovirus inhibits the engraftment of donor bone marrow cells by downregulation of hemopoietin gene expression in recipient stroma. J Virol. 1998;72:5006-5015. 2. Peggs KS, Mackinnon S. Cytomegalovirus: the role of CMV posthaematopoietic stem cell transplantation. Int J Biochem Cell Biol. 2004;36:695-701. 3. Boeckh M, Ljungman P. How we treat cytomegalovirus in hematopoietic cell transplant recipients. Blood. 2009;113:5711-5719. 4. Boeckh M, Nichols WG. The impact of cytomegalovirus serostatus of donor and recipient before hematopoietic stem cell transplantation in the era of antiviral prophylaxis and preemptive therapy. Blood. 2004;103:2003-2008. 5. Meijer E, Dekker AW, Rozenberg-Arska M, Weersink AJL, Verdonck LF. Influence of cytomegalovirus seropositivity on outcome after T celldepleted bone marrow transplantation: contrasting results between recipients of grafts from related and unrelated donors. Clin Infect Dis. 2002;35:703-712. 6. Broers AEC, van der Holt R, van Esser JWJ, et al. Increased transplantrelated morbidity and mortality in CMV-seropositive patients despite highly effective prevention of CMV disease after allogeneic T-cell–depleted stem cell transplantation. Blood. 2000;95:22402245. 7. Cardeñoso L, Pinsky BA, Lautenschlager I, et al. CMV antigenemia and quantitative viral load assessments in hematopoietic stem cell transplant recipients. J Clin Virol. 2013;56:108-112.
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
8. van Doornum GJ, Guldemeester J, Osterhaus ADME, Niesters HGM. Diagnosing herpesvirus infections by real-time amplification and rapid culture. J Clin Microbiol. 2003;41:576-580. 9. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. Available at: http:// www.R-project.org/. Accessed May 27, 2016. 10. Nichols WG, Corey L, Gooley T, Davis C, Boeckh M. High risk of death due to bacterial and fungal infection among cytomegalovirus (CMV)—seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection. J Infect Dis. 2002;185:273-282. 11. Craddock C, Szydlo RM, Dazzi F, et al. Cytomegalovirus seropositivity adversely influences outcome after T-depleted unrelated donor transplant in patients with chronic myeloid leukaemia: the case for tailored graft-versus-host disease prophylaxis. Br J Haematol. 2001; 112:228-236. 12. Baker KS, Gurney JG, Ness KK, et al. Late effects in survivors of chronic myeloid leukemia treated with hematopoietic cell transplantation: results from the Bone Marrow Transplant Survivor Study. Blood. 2004;104:1898-1906. 13. Cornelissen JJ, Carston M, Kollman C, et al. Unrelated marrow transplantation for adult patients with poor-risk acute lymphoblastic leukemia: strong graft-versus-leukemia effect and risk factors determining outcome. Blood. 2001;97:1572-1577. 14. George B, Pati N, Gilroy N, et al. Pre-transplant cytomegalovirus (CMV) serostatus remains the most important determinant of CMV reactivation after allogeneic hematopoietic stem cell transplantation in the era of surveillance and preemptive therapy. Transpl Infect Dis. 2010;12:322-329. 15. Wong FL, Francisco L, Togawa K, et al. Long-term recovery after hematopoietic cell transplantation: predictors of quality-of-life concerns. Blood. 2010;115:2508-2519. 16. Schmidt-Hieber M, Labopin M, Beelen D, et al. CMV serostatus still has an important prognostic impact in de novo acute leukemia patients after allogeneic stem cell transplantation: a report from the Acute Leukemia Working Party of EBMT. Blood. 2013;122:3359-3364. 17. Ljungman P, Brand R, Hoek J, et al. Donor cytomegalovirus status influences the outcome of allogeneic stem cell transplant: a study by the European Group for Blood and Marrow Transplantation. Clin Infect Dis. 2014;59:473-481.
1887
18. Ito S, Pophali P, Co W, et al. CMV reactivation is associated with a lower incidence of relapse after allo-SCT for CML. Bone Marrow Transplant. 2013;48:1-4. 19. Elmaagacli AH, Steckel NK, Koldehoff M, et al. Early human cytomegalovirus replication after transplantation is associated with a decreased relapse risk: evidence for a putative virus-versus-leukemia effect in acute myeloid leukemia patients. Blood. 2011;118:1402-1412. 20. Green ML, Leisenring WM, Xie H, et al. CMV reactivation after allogeneic HCT and relapse risk: evidence for early protection in acute myeloid leukemia. Blood. 2013;122:1316-1324. 21. Takenaka K, Nishida T, Asano-Mori Y, et al. Cytomegalovirus reactivation after allogeneic hematopoietic stem cell transplantation is associated with a reduced risk of relapse in patients with acute myeloid leukemia who survived to day 100 after transplantation: the Japan Society for Hematopoietic Cell Transplantation Transplantation-related Complication Working Group. Biol Blood Marrow Transplant. 2015;21:2008-2016. 22. Teira P, Battiwalla M, Ramanathan M, et al. Early cytomegalovirus reactivation remains associated with increased transplant related mortality in the current era: a CIBMTR analysis. Blood. 2016;127:24272438. doi:10.1182/blood-2015-11-679639. 23. Jeljeli M, Guérin-El Khourouj V, Porcher R, et al. Relationship between cytomegalovirus (CMV) reactivation, CMV-driven immunity, overall immune recovery and graft-versus-leukaemia effect in children. Br J Haematol. 2014;166:229-239. 24. Inagaki J, Noguchi M, Kurauchi K, Tanioka S, Fukano R, Okamura J. Effect of cytomegalovirus reactivation on relapse after allogeneic hematopoietic stem cell transplantation in pediatric acute leukemia. Biol Blood Marrow Transplant. 2016;22:300-306. 25. Manjappa S, Bhamidipati PK, Stokerl-Goldstein KE, et al. Protective effect of cytomegalovirus reactivation on relapse after allogeneic hematopoietic cell transplantation in acute myeloid leukemia patients is influenced by conditioning regimen. Biol Blood Marrow Transplant. 2014;20:46-52. 26. Muccio L, Bertaina A, Falco M, et al. Analysis of memory-like natural killer cells in human cytomegalovirus-infected children undergoing αβ+T and B cell-depleted hematopoietic stem cell transplantation for hematological malignancies. Haematologica. 2016;101:371-381. 27. Scheper W, Gründer C, Straetemans T, Sebestyen Z, Kuball J. Hunting for clinical translation with innate-like immune cells and their receptors. Leukemia. 2014;28:1181-1190.
Biology of Blood and Marrow Transplantation j o u r n a l h o m e p a g e : w w w. b b m t . o r g
Cytomegalovirus Status and the Outcome of T Cell–Replete Reduced-Intensity Allogeneic Hematopoietic Stem Cell Transplantation Frans M. Verduyn Lunel 1,*, Reinier Raymakers 2, Anette van Dijk 2, Lotte van der Wagen 2,3, Monique C. Minnema 2,3, Jurgen Kuball 2,3 1
Department of Medical Microbiology, Eijkman Winkler Institute for Microbiology, Infectious Diseases and Inflammation, Utrecht, The Netherlands Department of Hematology, Utrecht University Medical Center, Utrecht, The Netherlands 3 Laboratory of Translational Immunology, Utrecht University Medical Center, Utrecht, The Netherlands 2
Article history: Received 27 May 2016 Accepted 13 July 2016 Keywords: Cytomegalovirus Reduced intensity Stem cell transplantation
A B S T R A C T Cytomegalovirus (CMV) serostatus of donor and recipient are frequently used in algorithms of donor selection, whereas the impact of CMV reactivation on transplantation-related mortality, leukemia control, and overall survival (OS) remains controversial. Therefore, we retrospectively studied the impact of latent or active CMV infections on the outcome and occurrence of graft-versus-host disease (GVHD) after reduced-intensity conditioning (RIC) allogeneic hematopoietic stem cell transplantation (SCT) in 294 patients during the period from 2004 to 2010. CMV viral load was routinely monitored in plasma using a quantitative PCR. Preemptive antiviral therapy was initiated when the viral load in plasma exceeded a predefined threshold. In a proportional hazards model, a seropositive recipient was significantly associated with increased occurrence of acute GVHD. However the CMV serostatus of both recipient and donor and the presence of active CMV infection was not associated with the occurrence of relapses, chronic GVHD, or OS. We conclude that in the presence of viral load monitoring and preemptive treatment, latent or active CMV infection does not substantially affect the OS after T cell–replete RIC allogeneic SCT. © 2016 American Society for Blood and Marrow Transplantation.
INTRODUCTION Infections with cytomegalovirus (CMV) have been reported to impact donor engraftment [1] and to cause severe diseases after allogeneic hematopoietic stem cell transplantation (SCT), including pneumonia, hepatitis, gastroenteritis, retinitis, and encephalitis [2]. CMV has also been associated with an increased risk for acute graft-versus-host disease (GVHD) in patients receiving T cell–depleted grafts and for chronic GVHD [3]. Because of this detrimental impact of CMV disease on the prognosis of patients after SCT, preventive measures against active CMV infection after hematopoietic SCT were introduced to allow early interventions and to reduce mortality and morbidity mediated through CMV-associated diseases. Early interventions include the preemptive antiviral therapy
Financial disclosure: See Acknowledgments on page 1886. * Correspondence and reprint requests: Frans M. Verduyn Lunel, MD, PhD, Department of Virology, Eijkman Winkler Institute for Microbiology, Infectious Diseases and Inflammation, University Medical Center, G04.614, P.O. Box 85500, 3508 GA Utrecht, The Netherlands. E-mail address: [email protected] (F.M. Verduyn Lunel).
http://dx.doi.org/10.1016/j.bbmt.2016.07.009 1083-8791/© 2016 American Society for Blood and Marrow Transplantation.
with, for example, ganciclovir and valganciclovir as soon as the viral load in plasma or whole blood exceeds a predefined cut-off value during close monitoring [3]. Despite intervention strategies, a number of studies still reported on a negative impact of a positive CMV serostatus or active CMV infections on the outcome [4-6]. However, methodology including conditioning regimens and posttransplant patient management differ extensively between studies. Furthermore, the use of different techniques for the detection of CMV varies considerably. PCR techniques for detecting CMV viral load in plasma allows a more sensitive monitoring of the burden of CMV infection compared with the use of the pp65 antigenemia assay [7]. This difference may result in an earlier start of preemptive therapy in centers using PCR as gold standard. We therefore undertook a retrospective study in a very homogenous transplantation cohort undergoing exclusively reduced-intensity conditioning (RIC) while being collectively monitored by quantitative PCR techniques and treated with defined anti-CMV interventions. This cohort allowed addressing whether latent or active CMV infection in the recipient and the CMV serostatus of the donor or other
1884
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
factors contributed to the outcome after a defined type of transplantation and assessing whether CMV status of the donor is correctly placed early in donor selection algorithms. METHODS All patients 17 years and older who underwent a RIC allogeneic SCT during the period from 2004 to 2010 were included for analysis. All treatments and monitoring were standard of care, and all patients provided informed consent in line with JACIE (Joint Accreditation Committee ISCT & EBMT) guidelines to the transplantation as well as transplantation-related assessments and in accordance with the regulations of the medical ethics committee of University Medical Center Utrecht. Data were retrieved from patient record files and the electronic database from the department of Medical Microbiology. Allogeneic SCT included transplantation of grafts from matched unrelated donors (MUDs) and siblings. RIC regimen included fludarabine, antithymocyte globulin, and low-dose total body irradiation with 2 Gy. In case of allogeneic SCT for multiple myeloma after high dose melphalan and autologous SCT, a total body irradiation regimen with only .2 Gy was used. After allogeneic stem cell transplantation, GVHD prophylaxis consisted of cyclosporine A and/or mycophenolate mofetil. Patients were subjected to a routine screening protocol, which included the collection of plasma samples twice weekly after the transplantation during admission at the department of hematology followed by once weekly collection thereafter during follow-up visits at the outpatient clinic. End of follow-up of patients was due to death, or 120 days after SCT and increased to 180 days when active CMV infection was diagnosed before 120 days after SCT, or when antithymocyte globulin had been used in the conditioning regimen. Followup was continued thereafter during the treatment of acute or chronic GVHD. Plasma samples were stored at −80°C until testing. IgG antibodies against CMV were detected using the Enzygnost CMV/IgG (Siemens Healthcare Diagnostic Products, Marburg, Germany). CMV viral load was determined in EDTA-plasma samples using a quantitative PCR based on a real-time TaqMan CMV-DNA PCR assay as described previously [8]. Active CMV infection was defined as a viral load > 100 cps/mL, because artifacts in viral load determination could not be excluded when the viral load was below 100 cps/ mL; also, a reliable quantification was not possible below this threshold. Viral loads beyond that threshold were in the linear range of the assay used and could reliably and reproducibly be quantified. Primary infection and reactivation were defined as an active CMV infection occurring in an initial seronegative or seropositive recipient, respectively. Ganciclovir i.v. 5 mg/ kg, oral valganciclovir 900 mg orally twice daily, or foscarnet i.v. 60 mg/kg twice daily was given preemptively when the CMV viral load in plasma exceeded the threshold of 500 cps/mL. Dosages were adjusted when renal function was impaired. Statistical analysis was performed using the Mann-Whitney U test for comparing continuous variables and the chi-square test for noncontinuous variables. A Cox proportional hazards model, with active CMV as a timedependent variable because active CMV occurred at different time points during the follow up period after SCT, was used to analyze significant associations between variables and survival, the occurrence of acute or chronic GVHD, or relapses of disease. Variables with a Type I error rate < .2 in the univariate analysis were included in the multivariate model. Kaplan-Meier curves were constructed for analysis of overall survival (OS) during followup. The statistical software package R [9] was used to perform the calculations.
RESULTS Patient Characteristics Data from the entire follow-up period were available for all 294 patients who underwent an allogeneic hematopoietic SCT during the study period. Underlying diseases are summarized in Table 1. The median age was 55 years (range, 17 to 69). The median follow-up period was 1278 days (range, 27 to 3519) after SCT. IgG antibodies against CMV were detected in 176 patients (60%) before transplantation. According to their serostatus, patients and their donors could be subdivided into 4 groups: seronegative recipient and donor (R−/ D−, n = 64), seronegative recipient and seropositive donor (R−/ D+, n = 54), seropositive recipient and a seronegative donor (R+/D−, n = 46), and a seropositive recipient and donor (R+/ D+, n = 130). In total, 5571 plasma samples were collected from these patients for determination of the CMV viral load. Active CMV infection was observed in 102 patients (35%), including 9 (9%) with primary infections. In 77 of these patients (75%), viral
Table 1 Patient Characteristics (N = 294) Characteristic Median age, yr (range) Diagnosis type MM AML NHL MDS CLL ALL CML SAA Hodgkin disease Miscellaneous* Myelofibrosis Recipient CMV seropositive Active CMV infection R/D pairs R−/D− R−/D+ R+/D− R+/D+ Antithymocyte globulin conditioning Acute GVHD Grade I Grade II Grade III Grade IV Present, grade unknown Chronic GVHD Cord blood graft Donortype Related Unrelated Relapse disease Death
Value 55 (17-69) 77 (26) 65 (22) 49 (17) 27 (9) 20 (7) 15 (5) 12 (4) 12 (4) 8 (3) 6 (2) 3 (1) 176 (60) 102 (35) 65 (22) 54 (18) 46 (16) 130 (44) 155 (53) 41 (14) 58 (20) 23 (8) 10 (3) 1 (.3) 114 (39) 6 (2) 129 (44) 165 (56) 74 (25) 166 (56)
Values are number of cases with percents in parentheses, unless otherwise noted. MM indicates multiple myeloma; NHL, non-Hodgkin lymphoma; MDS, myelodysplastic syndrome; CLL, chronic lymphatic leukemia; ALL, acute lymphatic leukemia; CML, chronic myeloid leukemia; SAA, severe aplastic anemia. * Miscellaneous includes chronic promyelocytic leukemia (n = 1), hypereosinophilic syndrome (n = 2), chronic Epstein-Barr virus infection with secondary aplasia (n = 2), and undifferentiated leukemia, amyloidosis (n = 1).
load increased beyond the predefined threshold for preemptive treatment, whereas in 83 seropositive patients without active CMV infection, CMV DNA was not detected in plasma, not even in values below 100 copies/mL. The highest observed median viral load was 1338 copies/mL (range, 115 to 2,180,000). The difference between mean peak viral load during primary infections and reactivations, 1230 copies/ mL and 1394 copies/mL, respectively, was not significant (P = .80). Patients with an active CMV infection were slightly but not significantly older than patients without an active CMV infection, with a mean age of 53 years and 52 years, respectively. Initiation of active CMV infection was observed between 1 and 985 days (median, 40 days) after SCT. Risk Factors for OS Overall mortality during the entire follow-up period was 166 of 294 patients (56%). Mortality was similar in both patients with and without an active CMV infection, where 60 of 102 patients (59%) and 106 of 192 patients (55%), respectively, eventually died (chi-square test: P = .66). In contrast to previous reports [6,10,11], when constructing KaplanMeier curves, 5-year OS was not significantly different between recipients who were seropositive at the time of SCT when compared with recipients who were seronegative (P = .12) and between patients with or without an active CMV
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
1885
Figure 1. Survival curves. (A) Serostatus of the recipient. (B) Active CMV or no active CMV after SCT. (C) Survival according to different recipient and donor serostatus combinations.
infection (P = .96). Also, no significant differences in 5-year OS were observed between the different subgroups R−/D−, R−/ D+, R+/D−, and R+/D+ (P = .39, Figure 1). In a Cox proportional hazards model with active CMV as a time-dependent variable, chronic GVHD was significantly associated with improved OS (hazard ratio [HR], .7; 95% confidence interval [CI], .5 to .9), whereas the use of a graft from a MUD was significantly associated with a reduced OS (HR, 1.9; 95% CI, 1.3 to 2.8) (Table 2). Impact of CMV Serostatus and Active Infection on Relapse or Nonrelapse Mortality Equivalent OS of all subgroups could be a consequence of improved leukemia control in combination with increased mortality after active CMV infection, thus neutralizing the net effect. Therefore, relapse of underlying disease and nonrelapse mortality were analyzed separately. Relapse of underlying disease occurred in 74 patients (25%). Occurrence of relapse could not be retrieved in 11 patients who were subsequently excluded from the analysis. However, neither serostatus nor active CMV infection had an impact
Table 2 Significant Variables with the Conditions and Their Hazard Ratios Using a Cox Proportional Hazards Model with Active CMV Infection as a TimeDependent Variable Condition
Variable
HR
95% CI
Overall mortality
Chronic GVHD Unrelated donor Chronic GVHD Unrelated donor Recipient CMV seropositive Unrelated donor Antithymocyte globulin Chronic GVHD Chronic GVHD
.7 1.9 .7 2.2 1.7 2.2 .5 .3 .5
.5-.9 1.3-2.8 .5-1.0 1.4-3.3 1.2-2.5 1.4-3.4 .3-.7 .2-.6 .2-1.1
NRM Acute GVHD Chronic GVHD Relapse Relapse of AML
on the occurrence of relapse in a Cox proportional hazards model. Only the occurrence of chronic GVHD was significantly associated with a decreased occurrence of relapses (HR, .3; 95% CI, .2 to .6) (Table 2), most likely because chronic GVHD in patients with defined National Institutes of Health scoring (in 86% of patients) was in most patients only mild (28%) to moderate (41%) and only in some patients severe (30%) and did therefore not substantially affect OS but did allow a graftversus-leukemia effect. Because impact of active CMV infections on the occurrence of relapses was mainly observed by others for patients with acute myeloid leukemia (AML) [12-14], we therefore performed a separate analysis on this subgroup of 62 assessable AML patients. However, again, only chronic GVHD was associated with a decreased number of relapses, although the level of significance was not reached in this subgroup of patients (HR, .5; 95% CI, .2 to 1.1). In line with the observation that CMV-related factors did not contribute to leukemia relapse, no significant contribution to nonrelapse mortality was observed. Chronic GVHD was associated with a significant decrease (HR, .7; 95% CI, .5 to 1) whereas the use of a graft from a MUD was associated with an increase (HR, 2.2; 95% CI, 1.4 to 3.3) in nonrelapse mortality in a Cox proportional hazards model (Table 2). CMV Infection and GVHD Although OS was similar in all CMV serogroups and also CMV infection did not impact OS, quality of life could be heavily impaired because of the induction of, predominantly chronic, GVHD by CMV infection [12,15]. Associations between CMV and the occurrence of GVHD have been observed in previous studies [6,13,14]. Therefore, we analyzed the impact of latent or active CMV infection on acute and chronic GVHD. Acute GVHD was observed in 133 patients (45%), including 92 (69%) with grade II or higher. Active CMV infections were observed in 54 patients. Acute GVHD was reported for
1886
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
19 patients after development of active CMV infections (range, 1 to 106 days), thus suggesting that reactivation is not always elicited simply by immunosuppressive treatment of acute GVHD. In a Cox proportional hazards model, significant associations were only observed for a positive serostatus of the recipient (HR, 1.7; 95% CI, 1.2 to 2.5) (Table 2). One hundred fourteen patients (39%) suffered from chronic GVHD. Use of a graft from a MUD gave significantly higher rates of chronic GVHD (HR, 2.2; 95% CI, 1.4 to 3.4), whereas antithymocyte globulin in the conditioning regimen was protective (HR, .5; 95% CI, .3 to .7) (Table 2). Thus, only latency CMV infection, defined as a positive serostatus of the recipient, was associated with a significant higher occurrence of acute GVHD, whereas all other factors did not impact acute or chronic GVHD.
DISCUSSION Main findings of our study are that in contrast to previous reports, CMV status of both recipient and donor and active CMV infection do not necessarily associate with OS. One possible explanation for equivalent transplantation mortality in patients with and without CMV infection in our current cohort as compared with previous reports could arise from novel techniques for CMV monitoring. Although historically pp65 antigenemia was used as the gold standard for diagnosis of active CMV infections and treatment, our cohort has been monitored for active CMV infections by PCR. This usually results in both earlier diagnosis and start of preemptive therapy of patients. An alternative explanation could be that most analyses on the impact of CMV have mainly been performed in cohorts with myeloablative conditioning [16,17]. Our patients have been exclusively transplanted with nonmyeloablative conditioning. The very severely immunecompromised host after myeloablative therapy is most likely prone to more severe course of disease. The relatively modest sample size of our cohort poses another limitation, because huge data cohorts have been required for inclusion in the European Group for Blood and Marrow Transplantation database to suggest a link between CMV infection and OS. In the present study we were also not able to detect a direct association between CMV infection and the occurrence of relapses. The protective role of active CMV infections against relapses is controversial. This relation has been suggested in publications using both pp65 antigenemia and quantitative PCR to monitor CMV infections [18-21]. However, a recently published retrospective study using the Center for International Blood and Marrow Transplant Research database, was not able to show a reduced incidence of relapse in AML patients, even though most patients were transplanted after myeloablative conditioning [22]. In pediatric cohorts, using quantitative PCR for CMV monitoring also showed no altered risk for relapses when active infections occurred [23]. However, in another pediatric cohort using pp65 antigenemia for monitoring, a decreased incidence of relapses was reported [24]. Again, differences could be explained by variations in methodology and patient management, including myeloablative conditioning, between these and our cohort. Type of conditioning seems to be a crucial factor. In SCT patients with AML using PCR for monitoring CMV viral load, the association between reactivation of CMV during the first 100 days after transplantation and the incidence of relapse has been shown in myeloablative conditioned patients but not in patients who were transplanted after RIC regimens [25].
A possible biologic explanation for the absence of associations between CMV infection and relapses after RIC SCT may be that reconstitution of adaptive immunity occurs earlier in RIC when compared with myeloablative conditioning. This results in earlier control of opportunistic viral infections, which in turn may result in less prominent expansion of virusdriven natural killer cells [26] and γδ T cells from the young immune system with cross-reactive immunity against leukemic cells [27]. Today, many different aspects need to be considered when choosing the best possible donor, and CMV donor status still has quite a substantial impact in many different algorithms. Our data suggest that CMV donor status can be under certain circumstances neglected. In particular, in T cell–replete transplantation with a PCR-based early intervention strategy, active CMV infection does no harm. Vice versa, it is interesting to speculate that allowing “controlled” CMV infections could be a novel strategy to induce a more potent antileukemia activity. Therefore, a higher threshold before initiating CMV treatment might actually be beneficial for the patient. In particular, the use of low toxicity myeloablative regimens might provide an alternative window for an early innate immune recovery during a more controlled CMV infection, with even higher thresholds to be allowed before interventions are initiated. In summary, CMV status of the recipient and donor did not contribute to overall outcome in T cell–replete transplantations performed with a RIC protocol. CMV serostatus of the recipient or the presence or absence of active CMV infection does not have a significant impact on OS or occurrence of disease relapse after SCT when preemptive therapy is initiated early. This supports alternative donor selection algorithms for this type of transplants and opens also the avenue toward controlled later antiviral interventions to allow a well-balanced and potentially protective immune response cross-reactive toward CMV and leukemia.
ACKNOWLEDGMENTS Financial disclosure: The authors have nothing to disclose. Conflict of interest statement: J.K. is the co-founder and chief scientific officer of Gadeta (http://www.gadeta.nl/).
REFERENCES 1. Steffens H-P, Podlech J, Kurz S, Angele P, Dreis D, Reddehase MJ. Cytomegalovirus inhibits the engraftment of donor bone marrow cells by downregulation of hemopoietin gene expression in recipient stroma. J Virol. 1998;72:5006-5015. 2. Peggs KS, Mackinnon S. Cytomegalovirus: the role of CMV posthaematopoietic stem cell transplantation. Int J Biochem Cell Biol. 2004;36:695-701. 3. Boeckh M, Ljungman P. How we treat cytomegalovirus in hematopoietic cell transplant recipients. Blood. 2009;113:5711-5719. 4. Boeckh M, Nichols WG. The impact of cytomegalovirus serostatus of donor and recipient before hematopoietic stem cell transplantation in the era of antiviral prophylaxis and preemptive therapy. Blood. 2004;103:2003-2008. 5. Meijer E, Dekker AW, Rozenberg-Arska M, Weersink AJL, Verdonck LF. Influence of cytomegalovirus seropositivity on outcome after T celldepleted bone marrow transplantation: contrasting results between recipients of grafts from related and unrelated donors. Clin Infect Dis. 2002;35:703-712. 6. Broers AEC, van der Holt R, van Esser JWJ, et al. Increased transplantrelated morbidity and mortality in CMV-seropositive patients despite highly effective prevention of CMV disease after allogeneic T-cell–depleted stem cell transplantation. Blood. 2000;95:22402245. 7. Cardeñoso L, Pinsky BA, Lautenschlager I, et al. CMV antigenemia and quantitative viral load assessments in hematopoietic stem cell transplant recipients. J Clin Virol. 2013;56:108-112.
F.M. Verduyn Lunel et al. / Biol Blood Marrow Transplant 22 (2016) 1883–1887
8. van Doornum GJ, Guldemeester J, Osterhaus ADME, Niesters HGM. Diagnosing herpesvirus infections by real-time amplification and rapid culture. J Clin Microbiol. 2003;41:576-580. 9. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. Available at: http:// www.R-project.org/. Accessed May 27, 2016. 10. Nichols WG, Corey L, Gooley T, Davis C, Boeckh M. High risk of death due to bacterial and fungal infection among cytomegalovirus (CMV)—seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection. J Infect Dis. 2002;185:273-282. 11. Craddock C, Szydlo RM, Dazzi F, et al. Cytomegalovirus seropositivity adversely influences outcome after T-depleted unrelated donor transplant in patients with chronic myeloid leukaemia: the case for tailored graft-versus-host disease prophylaxis. Br J Haematol. 2001; 112:228-236. 12. Baker KS, Gurney JG, Ness KK, et al. Late effects in survivors of chronic myeloid leukemia treated with hematopoietic cell transplantation: results from the Bone Marrow Transplant Survivor Study. Blood. 2004;104:1898-1906. 13. Cornelissen JJ, Carston M, Kollman C, et al. Unrelated marrow transplantation for adult patients with poor-risk acute lymphoblastic leukemia: strong graft-versus-leukemia effect and risk factors determining outcome. Blood. 2001;97:1572-1577. 14. George B, Pati N, Gilroy N, et al. Pre-transplant cytomegalovirus (CMV) serostatus remains the most important determinant of CMV reactivation after allogeneic hematopoietic stem cell transplantation in the era of surveillance and preemptive therapy. Transpl Infect Dis. 2010;12:322-329. 15. Wong FL, Francisco L, Togawa K, et al. Long-term recovery after hematopoietic cell transplantation: predictors of quality-of-life concerns. Blood. 2010;115:2508-2519. 16. Schmidt-Hieber M, Labopin M, Beelen D, et al. CMV serostatus still has an important prognostic impact in de novo acute leukemia patients after allogeneic stem cell transplantation: a report from the Acute Leukemia Working Party of EBMT. Blood. 2013;122:3359-3364. 17. Ljungman P, Brand R, Hoek J, et al. Donor cytomegalovirus status influences the outcome of allogeneic stem cell transplant: a study by the European Group for Blood and Marrow Transplantation. Clin Infect Dis. 2014;59:473-481.
1887
18. Ito S, Pophali P, Co W, et al. CMV reactivation is associated with a lower incidence of relapse after allo-SCT for CML. Bone Marrow Transplant. 2013;48:1-4. 19. Elmaagacli AH, Steckel NK, Koldehoff M, et al. Early human cytomegalovirus replication after transplantation is associated with a decreased relapse risk: evidence for a putative virus-versus-leukemia effect in acute myeloid leukemia patients. Blood. 2011;118:1402-1412. 20. Green ML, Leisenring WM, Xie H, et al. CMV reactivation after allogeneic HCT and relapse risk: evidence for early protection in acute myeloid leukemia. Blood. 2013;122:1316-1324. 21. Takenaka K, Nishida T, Asano-Mori Y, et al. Cytomegalovirus reactivation after allogeneic hematopoietic stem cell transplantation is associated with a reduced risk of relapse in patients with acute myeloid leukemia who survived to day 100 after transplantation: the Japan Society for Hematopoietic Cell Transplantation Transplantation-related Complication Working Group. Biol Blood Marrow Transplant. 2015;21:2008-2016. 22. Teira P, Battiwalla M, Ramanathan M, et al. Early cytomegalovirus reactivation remains associated with increased transplant related mortality in the current era: a CIBMTR analysis. Blood. 2016;127:24272438. doi:10.1182/blood-2015-11-679639. 23. Jeljeli M, Guérin-El Khourouj V, Porcher R, et al. Relationship between cytomegalovirus (CMV) reactivation, CMV-driven immunity, overall immune recovery and graft-versus-leukaemia effect in children. Br J Haematol. 2014;166:229-239. 24. Inagaki J, Noguchi M, Kurauchi K, Tanioka S, Fukano R, Okamura J. Effect of cytomegalovirus reactivation on relapse after allogeneic hematopoietic stem cell transplantation in pediatric acute leukemia. Biol Blood Marrow Transplant. 2016;22:300-306. 25. Manjappa S, Bhamidipati PK, Stokerl-Goldstein KE, et al. Protective effect of cytomegalovirus reactivation on relapse after allogeneic hematopoietic cell transplantation in acute myeloid leukemia patients is influenced by conditioning regimen. Biol Blood Marrow Transplant. 2014;20:46-52. 26. Muccio L, Bertaina A, Falco M, et al. Analysis of memory-like natural killer cells in human cytomegalovirus-infected children undergoing αβ+T and B cell-depleted hematopoietic stem cell transplantation for hematological malignancies. Haematologica. 2016;101:371-381. 27. Scheper W, Gründer C, Straetemans T, Sebestyen Z, Kuball J. Hunting for clinical translation with innate-like immune cells and their receptors. Leukemia. 2014;28:1181-1190.