EClinicalMedicine 33 (2021) 100787 Contents lists available at ScienceDirect EClinicalMedicine journal homepage: https://www.journals.elsevier.com/eclinicalmedicine Research Paper A randomised, placebo-controlled phase 3 study to evaluate the efficacy and safety of ASP0113, a DNA-based CMV vaccine, in seropositive allogeneic haematopoietic cell transplant recipients Per Ljungmana,*, Arancha Bermudezb, Aaron C. Loganc, Mohamed A. Kharfan-Dabajad,y, Patrice Chevalliere, Rodrigo Martinof, Gerald Wulfg, Dominik Selleslagh, Kazuhiko Kakihanai, Amelia Langstonj, Dong-Gun Leek, Carlos Solanol, Shinichiro Okamotom, Larry R. Smithn,z, Michael Boeckho, John R. Wingardp, Beth Cywinq, Christine Fredericksq, Christopher Lademacherq, Xuegong Wangq, James Youngq, Johan Maertensr a Department of Cellular Therapy and Allogeneic Stem Cell Transplantation, Karolinska University Hospital and Division of Hematology, Department of Medicine Huddinge, Karolinska Institutet, SE-14186, Stockholm, Sweden b Department of Hematology, Hospital Universitario Marques de Valdecilla, Instituto de Formacio n e Investigacio n Marques de Valdecilla, 39008, Santander, Spain c Division of Hematology and Blood and Marrow Transplantation, Department of Medicine, University of California, San Francisco, 94143, San Francisco, United States d Department of Blood and Marrow Transplantation, H. Lee Moffitt Cancer Center and Research Institute and University of South Florida, 33612, Tampa, United States e Service d’Hematologie Clinique, CHU de Nantes, 44093, Nantes, France and CRCINA / INSERM UMR1232 / CNRS ERL6001 IRS UN  8 Quai Moncousu  BP 70721, 44007 Nantes cedex 1, France f Division of Clinical Hematology, Hospital de la Santa Creu I Sant Pau, 08041, Barcelona, Spain g Department of Hematology and Medical Oncology, University Medical Center Go €ttingen, D-37075, Go€ttingen, Germany h Department of Hematology, AZ Sint-Jan Brugge-Oostende, 8000, Brugge, Belgium i Hematology Division, Tokyo Metropolitan Komagome Hospital, 113-8677, Tokyo, Japan j Winship Cancer Institute, Emory University, 30322, Atlanta, United States k Division of Infectious Diseases, Department of Internal Medicine, Catholic Hematology Hospital and Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 06591, Seoul, Republic of Korea l Hematology Service, Hospital Clínico Universitario, Institute of Research INCLIVA and University of Valencia, 46010, Valencia, Spain m Division of Hematology, Department of Medicine, Kelo University School of Medicine, 160-8582, Tokyo, Japan n Vical Inc., 92121, San Diego, United States o Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center and University of Washington, 98109, Seattle, United States p Division of Hematology & Oncology, Department of Medicine, University of Florida, 32610, Gainesville, United States q Astellas Pharma Global Development, Inc., 60062, Northbrook, United States r Department of Microbiology and Immunology, K.U. Leuven and Department of Hematology, UZ Leuven, Leuven, Belgium A R T I C L E I N F O A B S T R A C T Article History: Background: Cytomegalovirus (CMV) is a complication of allogeneic haematopoietic cell transplantation (allo- Received 30 June 2020 HCT). ASP0113, a DNA-based vaccine, contains two plasmids encoding human CMV glycoprotein B and phos- Revised 19 February 2021 phoprotein 65 (pp65). We assessed ASP0113 in CMV-seropositive allo-HCT recipients. Accepted 22 February 2021 Methods: In this phase 3, randomised, placebo-controlled study, CMV-seropositive allo-HCT recipients were Available online 19 March 2021 randomly assigned (1:1) via interactive response technology to receive five injections of 1 mL of 5 mg/mL ASP0113 or placebo. The pharmacist and designated staff were unblinded. Masked syringes maintained the blind for patients and study personnel. Efficacy and safety analyses included patients who received 1 dose of ASP0113/placebo. The primary efficacy endpoint was the proportion of allo-HCT recipients with composite all-cause mortality and adjudicated CMV end-organ disease (EOD) by 1 year post-transplant. ClinicalTrials. gov: NCT01877655 (not recruiting). Findings: Patients were recruited between Sept 11, 2013 and Sept 21, 2016. Overall, 501 patients received 1 dose of ASP0113 (n = 246) or placebo (n = 255). The proportion of patients with composite all-cause mortality * Corresponding author. E-mail address: per.ljungman@ki.se (P. Ljungman). y Dr Kharfan-Dabaja was an employee at H. Lee Moffitt Cancer center and Research Institute and University of South Florida during the conduct of this study, and is now an employee at the Division of Hematology-Oncology and Blood and Marrow Transplantation Program, Mayo Clinic, 32224, Jacksonville, United States. z Dr Smith was an employee at Vical Inc. during the conduct of this study, and is now a consultant. https://doi.org/10.1016/j.eclinm.2021.100787 2589-5370/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) 2 P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 and adjudicated CMV EOD by 1 year post-transplant was 35.4% (n = 87) with ASP0113 and 302% (n = 77) with placebo (odds ratio 1.27; 95% confidence interval: 0.87 to 1.85; p = 0.205). Incidence of injection site- related treatment-emergent adverse events (TEAEs) was higher with ASP0113 than placebo. Overall inci- dence and severity of other TEAEs was similar between groups. T-cell response to pp65 increased over time and was greater with placebo than ASP0113 (p = 0.027). Interpretation: ASP0113 did not reduce overall mortality or CMV EOD by 1 year post-transplant. Safety find- ings were similar between groups. Funding: Astellas Pharma Global Development, Inc . © 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) CMV disease, should be carefully considered. Letermovir, a CMV Research in Context DNA terminase complex inhibitor, has recently been approved by the US Food and Drug Administration and the European Medicines Evidence before this study Agency for the prophylaxis of CMV infection and disease in adult CMV-seropositive allo-HCT recipients [68]. In a phase 3, double- There is an unmet need to develop treatments with novel blind, placebo-controlled study, letermovir significantly reduced mechanisms of action that are associated with favourable safety the number of patients with clinically significant CMV infection by profiles compared with antiviral therapies for the prevention of week 24 post-transplant (p<0.001), while maintaining a similar cytomegalovirus (CMV) end-organ disease in allogeneic haema- safety profile to placebo [6]. Furthermore, letermovir seemed to be topoietic cell transplant (allo-HCT) recipients. ASP0113 is a associated with decreased all-cause mortality at week 24; however, DNA-based vaccine containing two plasmids encoding glyco- late CMV reactivation occurred after discontinuation of letermovir, protein B and phosphoprotein 65 from human CMV. In a phase indicating a need for novel strategies to stimulate anti-CMV immu- 2 study of allo-HCT recipients, ASP0113 significantly reduced nity. [6] the occurrence and recurrence of CMV viraemia compared with Despite recent advances and developments in the field of CMV placebo; it was well tolerated but was not significantly more infection and disease prophylaxis, an unmet need remains for thera- immunogenic than placebo. pies with novel mechanisms of action that are not associated with safety concerns. [9] A number of vaccines have been investigated for Added value of this study the prevention of CMV infection and/or disease in solid-organ trans- The results of this phase 3 study show that ASP0113 did not plant recipients. reduce overall mortality and CMV end-organ disease through 1 A study of the immunogenicity of a CMV glycoprotein B (gB) vac- year post-transplant. These findings reiterate the need to con- cine combined with MF59 adjuvant, administered pre-transplant in firm promising phase 2 data with larger phase 3 studies, and kidney or liver transplant recipients, demonstrated significantly highlight the difficulty of inducing immune responses against increased gB antibody titres compared with placebo in both CMV- CMV in the setting of highly immunosuppressed allo-HCT seronegative (geometric mean titre: 12,537 [95% confidence interval recipients. (CI): 6593 to 23,840] versus 86 [95% CI: 63 to 118]; p < 0.0001) and CMV-seropositive (geometric mean titre: 118,395 [95% CI: 64,503 to Implications of all the available evidence 217,272] versus 24,682 [95% CI: 17,909 to 34,017]; p < 0.0001) recipi- ents of the vaccine [10]. This study also demonstrated a significantly ASP0113 was not effective for inducing immune responses to decreased duration of viraemia (p = 0.048) and reduced number of CMV when administered pre- and post-allo-HCT. Given that days of ganciclovir treatment (p = 0.0287) in CMV-seronegative recip- letermovir has demonstrated efficacy for the prophylaxis of ients [10]. The modified vaccinia virus Ankara vector, currently being CMV infection with limited toxicity, the development of vac- investigated, has been demonstrated to elicit potent humoral and cel- cines with improved immunogenicity and novel vaccination lular immune responses against multiple immunodominant CMV strategies targeting the prevention of late or recurrent CMV antigens in preclinical and in vitro studies [11]. These data support infection/reactivation may be appropriate. the development of a multi-antigenic vaccine candidate for control- ling CMV in transplant recipients. ASP0113, a DNA-based therapeutic vaccine, contains two plas- 1. Introduction mids, VCL-6365 and VCL-6368, encoding human CMV gB and phos- phoprotein 65 (pp65), respectively, and is formulated with a Severely immunocompromised patients, including patients CRL1005 poloxamer and benzalkonium chloride delivery system undergoing allogeneic haematopoietic cell transplantation (allo- designed to enhance plasmid expression [12]. In a phase 1 study of HCT), are particularly vulnerable to cytomegalovirus (CMV) infection healthy individuals, ASP0113 was well tolerated and immunogenic or reactivation [1]. CMV seropositivity (or prior exposure) is a known [13]. In a phase 2 study of allo-HCT recipients, ASP0113 was well tol- risk factor for CMV end-organ disease (EOD), bacterial and fungal erated and significantly reduced the occurrence and recurrence of infections, and increased mortality [2]. CMV viraemia compared with placebo, but was not statistically sig- Pre-emptive therapies with ganciclovir or its valine ester valgan- nificantly more immunogenic than placebo [14]. In another phase 2 ciclovir are effective methods of preventing the development of study, despite not demonstrating superior efficacy or immunogenic- CMV disease in allo-HCT recipients [2], although both are associated ity in the prevention of CMV viraemia in CMV-seronegative trans- with toxicity. Treatment with ganciclovir or valganciclovir is associ- plant recipients receiving a kidney from a CMV-seropositive donor, ated with myelosuppression, as well as bacterial and fungal infec- ASP0113 did demonstrate a safety profile that was generally similar tions [35]. In view of these toxicities, the riskbenefit of to placebo [15]. administering potentially toxic agents as a prophylactic or pre-emp- Our study aimed to evaluate the efficacy, safety, and immunoge- tive treatment to patients who may be at low risk of developing nicity of ASP0113 in CMV-seropositive allo-HCT recipients. P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 3 2. Methods to 40, 60 § 5, 90 § 10, and 180 § 10. CMV viral load plasma samples were collected weekly (§2 days) from Day 0 to Day 100, every other 2.1. Study design and participants week (§5 days) from Day 101 to Day 180, and every 30 days (§5 days) from Day 181 to Day 365. All CMV viral load plasma testing was This was a randomised, double-blind, placebo-controlled, multi- performed by the central laboratory (Abbott Molecular Inc., Des centre phase 3 study conducted at 95 sites in 11 countries (Australia Plaines, IL, USA; although they could be tested at a local laboratory if [4], Belgium [4], Canada [4], France [5], Germany [13], Japan [10], approved by the sponsor and at the discretion of the investigator). South Korea [4], Spain [11], Sweden [5], Taiwan [3] and the USA [32]; Pre-emptive therapy could then be started based on the central or appendix 1). Eligible patients were CMV-seropositive allo-HCT recipi- local laboratory assessment and the patient’s clinical condition. ents aged 18 years who were planning to undergo one of the fol- Allo-HCT recipients were followed for 1 year post-transplant for lowing: a sibling donor transplant with a 7/8 human leucocyte overall mortality and adjudicated CMV EOD, CMV viraemia, and adju- antigen (HLA)-A, -B, -C, -DRb1 match utilising high-resolution typing, dicated CMV-specific antiviral therapy use. Allo-HCT recipients were or an 8/8 HLA-A, -B, -C, -DRb1 match utilising low- or high-resolution evaluated for local and systemic reactogenicity 15 min and 60 min typing, or an unrelated donor transplant with a 7/8 or 8/8 HLA-A, -B, (§10 min) after each study drug injection and for 7 days after each -C, -DRb1 match utilising high-resolution typing, as a minimum study drug injection by patient reporting in a diary. requirement. Patients were excluded if they had active CMV infection Treatment-emergent adverse events (TEAEs) and serious adverse or disease, had received treatment for active CMV infection or disease events (SAEs) were collected for 30 days after the last study drug within 3 months before transplant, had planned CMV prophylactic injection. All events requiring adjudication, grade 3 TEAEs, and antiviral therapy or CMV-specific immunoglobulins from randomisa- SAEs were collected from 31 days after the last dose of study drug tion through primary study period completion (day 365), or had a through to 1 year post-transplant. modified HCT-specific comorbidity index score 4 [16]. Patients Immunogenicity was measured by T-cell responses to pp65- and were also excluded if they were known to be positive for human gB-specific antibody levels through 1 year after the first study drug immunodeficiency virus, hepatitis B virus, or hepatitis C virus; had injection. T-cell responses to pp65 were measured in peripheral unknown CMV serostatus; had received alemtuzumab within 60 blood mononuclear cells using an ex vivo pp65-specific IFN-g days before transplant; had T-cell depletion of a donor cell product; enzyme-linked ImmunoSpotÒ assay. Antibody responses to gB were had received a CMV vaccine (including any prior exposure to measured in serum samples using a gB-specific serum immunoglobu- ASP0113); had received an allo-HCT within 1 year before transplant; lin G-binding enzyme-linked immunosorbent assay-based platform. had residual chronic graft-versus-host disease (GVHD) from a previ- For both T-cell responses to pp65 and antibody responses to gB, six ous HCT; or had a platelet count of <50,000/mm3 within 3 days measurements were taken during the study on Day 14 to 3, Day before randomisation. Use of anti-thymocyte globulin (ATG) during 14 to 40, Day 70 § 5 to 74 § 5, Day 90 § 10, Day 180 § 10 and Day transplant conditioning was permitted as per institutional prefer- 365 + 14. Samples for T-cell assays were not collected if the absolute ence. lymphocyte count was confirmed to be <500 mm3 by a local or cen- An Independent Ethics Committee (IEC) or Institutional Review tral laboratory. Board (IRB) reviewed the ethical appropriateness of the study before it was conducted, and their approval of the study protocol was obtained before authorisation of a drug shipment to a study site. An 2.4. Outcomes IEC or IRB provided approval of all written informed consent as per national regulations. The study was conducted in accordance with The primary efficacy endpoint was the proportion of allo-HCT the study protocol, Good Clinical Practice requirements, International recipients with composite all-cause mortality and adjudicated CMV Council for Harmonisation of Technical Requirements for Human Use EOD by 1 year post-transplant. Secondary efficacy endpoints were: guidelines, and other applicable regulations and guidelines for con- time to first protocol-defined CMV viraemia (CMV plasma viral load ducting clinical studies and ethical principles that have their origin in 1000 IU/mL) through 1 year post-transplant; time to first use of the Declaration of Helsinki. adjudicated CMV-specific antiviral therapy after the first study drug injection through 1 year post-transplant; the proportion of patients 2.2. Randomisation and masking with a composite endpoint of protocol-defined CMV viraemia and CMV-specific antiviral therapy use per the adjudication committee; Patients were randomly assigned in a 1:1 ratio to receive ASP0113 time to first occurrence of either use of adjudicated CMV-specific or placebo. Randomisation was performed via interactive response antiviral therapy or adjudicated CMV EOD rate; and all-cause mortal- technology after the completion of the screening period and at least ity at 1 year post-transplant. Exploratory endpoints included, but 314 days before the anticipated transplant day/donor cell infusion were not limited to, the proportion of all-cause mortality and adjudi- (Day 0). Randomisation was stratified by donorrecipient related- cated CMV EOD at 6 months, the proportion of adjudicated CMV EOD, ness and by donor CMV serostatus. Patients were assigned to a dou- the incidence of grade 34 acute GVHD (aGVHD), the proportion of ble-blinded treatment group in the order in which they were entered graft rejection, and the incidence of engraftment. A subgroup analysis into the interactive response technology system. Only the pharmacist of the primary endpoint stratified by whether ATG was included in and designated staff, including the person(s) administering the vac- the patients’ conditioning regimen and whether the patients received cine, were unblinded to treatment. The syringes were masked before a myeloablative conditioning regimen was conducted. dosing to maintain the blind for patients and other study personnel. Safety endpoints included adverse events classified by the Medi- Randomisation was blinded, so allocation to treatment group could cal Dictionary for Regulatory Activities (MedDRA) version 16.0 [17] not influence the likelihood of receiving treatment. and reported using the National Cancer Institute’s Common Termi- nology Criteria for Adverse Events version 4.03 grading scale, [18] 2.3. Procedures vital signs, physical examinations, local reactogenicity signs, and clin- ical laboratory assessments. A data monitoring committee monitored Patients were randomly assigned to receive 1 mL of 5 mg/mL the safety of patients. ASP0113 (manufactured by Vical, San Diego, CA, USA) or placebo via The immunogenicity endpoints through 1 year after the first study intramuscular injection (appendix 2). Patients received five intra- drug injection were T-cell responses to pp65 and gB-specific antibody muscular doses of either ASP0113 or placebo on Days 14 to 3, 14 levels. 4 P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 2.5. Statistical analyses The sample size was estimated based on a phase 2 study of the efficacy of ASP0113 in HCT recipients; [14] to detect the estimated difference in mortality and CMV EOD from the phase 2 study (35.3% vs 22.5%), the study needed a sample size of at least 424 (212 per arm) to have 80% power at the two-sided significance level of 0.050. A sample size of 500 (250 per arm) was expected to have 86% power for the composite endpoint. A comparison between ASP0113 and pla- cebo with respect to the primary endpoint included data from the time of the first dose of the study drug through 1 year post-transplant in the follow-up period. The primary analysis was performed using the CochranMantelHaenszel test, which was stratified by factors for randomisation (donorrecipient relatedness and donor CMV serosta- tus) and a supportive analysis of time to first occurrence of death or CMV EOD using a Cox proportional hazards model, [19] which included treatment and stratification factors for randomisation. The key secondary endpoints of time to first protocol-defined CMV viraemia through 1 year post-transplant and time to first use of adjudicated CMV-specific antiviral therapy use through 1 year post- transplant were both compared between treatments using a Cox pro- portional subdistributional hazards model [19] with factors for treat- ment, donorrecipient relatedness, and donor CMV serostatus. This model accounts for the competing risk of death, after which these key secondary endpoints can no longer be observed. Efficacy analyses were conducted on the full analysis set (FAS), which consisted of all randomised patients who received at least one dose of the study drug. The safety analysis set (SAF) consisted of all randomised patients who received at least one dose of the study drug. The immunogenicity analysis set consisted of all patients who received at least one dose of the study drug and for whom at least one post-transplant immunogenicity measurement was available. All Fig. 1. Trial profile *Randomisation was blinded, so allocation to treatment group statistical analyses were conducted using SASÒ version 9.4 or higher could not influence the likelihood of receiving treatment. yThe FAS consisted of all randomised patients who received at least one dose of randomly assigned study drug. on Unix. zThe SAF consisted of all randomised patients who received at least one dose of study This study is registered on ClinicalTrials.gov (NCT01877655). drug. xThe IAS included all patients who received at least one dose of study drug and for whom at least one post-transplant immunogenicity measurement was available. 2.6. Role of the funding source AE=adverse event; FAS=full analysis set; IAS=immunogenicity analysis set; SAF=safety analysis set. The study sponsor, Astellas Pharma Global Development, Inc., was involved in the design or conception of the study, the analysis and was 35.4% (n = 87) in the ASP0113 group and 30.2% (n = 77) in the pla- interpretation of the data and drafting and critically reviewing the cebo group (odds ratio [OR] 1.27; 95% CI: 0.87 to 1.85; p = 0.205; publication. The authors who are employees of the study sponsor Table 3). There was also no statistically significant difference for the were Beth Cywin, Christine Fredericks, Christopher Lademacher, Xue- time to occurrence of the primary endpoint between the ASP0113 gong Wang and James Young. The corresponding author had full and placebo groups (hazard ratio 1.20; 95% CI: 0.87 to 1.67; p = 0.263; access to all the data in the study and had final responsibility for the Fig. 2). decision to submit this final version for publication. No statistically significant differences were found between ASP0113 and placebo for any of the secondary endpoints (Table 4) or 3. Results exploratory efficacy endpoints (Table 5). There were no statistically significant differences for the time to first protocol-defined CMV vir- Patients were recruited between Sept 11, 2013 and Sept 21, 2016. aemia (1-hazard ratio [1-HR] 0.04; 95% CI: 0.22 to 0.24; p = 0.748) Overall, 612 patients were assessed for eligibility. Of 514 randomly or for the time to first use of adjudicated CMV-specific antiviral ther- assigned patients, 501 received at least one dose of ASP0113 apy (1-HR 0.02; 95% CI: 0.29 to 0.20; p = 0.888) between the (n = 246) or placebo (n = 255) and were included in the FAS and the ASP0113 and placebo groups (Fig. 3). In subgroup analyses of the pri- SAF (Fig. 1). The number of patients who discontinued treatment for mary endpoint, there were no statistically significant differences any reason was 93 in the ASP0113 group (18.6%) and 94 in the pla- between the ASP0113 and placebo groups when patients were strati- cebo group (18.8%). Most patients were white (71.3%) and male fied by whether ATG was included in their conditioning regimen (57.9%), with a median age (interquartile range [IQR]) of 54 years (p = 0.382) or whether they received a myeloablative conditioning (4462). For donors, more were white (42.9%) than any other ethnic- regimen (p = 0.563; Table 6). ity; most were male (61.5%), and the median age (IQR) was 38 years All patients in the SAF experienced at least one TEAE (Table 7 and (2751; Table 1). More donors were CMV seropositive than CMV supplementary table S1). Minimal differences were observed seronegative in both the ASP0113 (58.9%) and placebo (61.2%) between the safety profiles of the ASP0113 and placebo groups, with groups. The number of patients who received each of the five study the exception of a higher proportion of patients in the ASP0113 group drug injections is provided in Table 2. experiencing drug-related TEAEs versus the placebo group (78.9% The proportion of allo-HCT recipients with composite of all-cause [n = 194] vs 29.0% [n = 74]). The difference observed was mostly due mortality and adjudicated CMV EOD through 1 year post-transplant to a higher number of mild to moderate injection-site-related TEAEs P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 5 Table 1 Table 3 Demographics and baseline characteristics (full analysis set). Primary efficacy composite endpoint of all-cause mortality and adjudicated cytomega- lovirus end-organ disease (full analysis set). ASP0113 Placebo Total (n = 246) (n = 255) (N = 501) Parameter, n (%) ASP0113 Placebo Odds ratio* (n = 246) (n = 255) (95% CI) p-value Recipient characteristics Male sex 137 (55.7) 153 (60.0) 290 (57.9) Total composite 87 (35.4) 77 (30.2) 1.27 (0.87 to 1.85) Median age (IQR), years 54 (4361) 54 (4462) 54 (4462) p = 0.205 Race All-cause mortality 78 (31.7) 72 (28.2) White 174 (70.7) 183 (71.8) 357 (71.3) Known deathsy 65 (26.4) 65 (25.5) Asian 50 (20.3) 45 (17.6) 95 (19.0) Unknown survival status 13 (5.3) 7 (2.7) Black or African American 7 (2.8) 6 (2.4) 13 (2.6) due to withdrawal of consent Other 15 (6.1) 21 (8.2) 36 (7.2) Adjudicated CMV EOD 15 (6.1) 9 (3.5) Strata * ASP0113 vs placebo. Related seropositive donor 72 (29.3) 63 (24.7) 135 (26.9) y Deaths listed here are from transplant through Day 365. Analysis was performed Related seronegative donor 26 (10.6) 28 (11.0) 54 (10.8) using the CochranMantelHaenszel test stratified by factors for randomisation Unrelated seropositive donor 73 (29.7) 93 (36.5) 166 (33.1) (donorrecipient relatedness and donor CMV serostatus). CI=confidence interval; Unrelated seronegative donor 75 (30.5) 71 (27.8) 146 (29.1) CMV=cytomegalovirus; EOD=end-organ disease. Most frequent indications for transplant Acute myelogenous leukaemia 96 (39.0) 123 (48.2) 219 (43.7) Myelodysplastic syndrome 49 (19.9) 45 (17.6) 94 (18.8) Acute lymphocytic leukaemia 34 (13.8) 31 (12.2) 65 (13.0) Lymphoma 31 (12.6) 32 (12.5) 63 (12.6) The mean T-cell response to pp65 increased over time; the Conditioning regimen increase was greater with placebo than with ASP0113 (p = 0.027; Fig. Myeloablative 120 (48.8) 122 (47.8) 242 (48.3) 4). The mean gB-specific antibody response increased over time dur- Non-myeloablative 124 (50.4) 131 (51.4) 255 (50.9) ing the study (Fig. 5). Although there was no statistically significant Missing 2 (0.8) 2 (0.8) 4 (0.8) difference in mean gB-specific antibody response between ASP0113 ATG Yes 43 (17.5) 45 (17.6) 88 (17.6) and placebo overall (p = 0.112), the mean gB-specific antibody No 203 (82.5) 210 (82.4) 413 (82.4) response was significantly greater in the ASP0113 group compared Donor characteristics with the placebo group at Month 12 (p = 0.036). Male 147 (59.8) 161 (63.1) 308 (61.5) Median age (IQR), years 39 (2751) 37 (2750) 38 (2751) Race 4. Discussion White 103 (41.9) 112 (43.9) 215 (42.9) Asian 49 (19.9) 41 (16.1) 90 (18.0) This phase 3 study investigating ASP0113 in allo-HCT recipients Black or African American 5 (2.0) 5 (2.0) 10 (2.0) did not achieve its primary endpoint, which was largely unexpected American Indian or Alaska native 1 (0.4) 1 (0.4) 2 (0.4) given the results from the phase 2 study but illustrates the need for Other 71 (28.9) 86 (33.7) 157 (31.3) Unknown 17 (6.9) 10 (3.9) 27 (5.4) phase 3 studies to validate new treatments in large cohorts of patients. [14] Several possible reasons can be given to explain the Data are n (%) unless otherwise specified. ATG=anti-thymocyte globulin; IQR, inter- quartile range. results of this study. The primary endpoint was chosen for this study as it would capture the ability of ASP0113 to reduce both direct (CMV disease) and indirect effects of CMV as well as possible adverse effects associated with the use of pre-emptive therapy, both of which may Table 2 result in a reduction of mortality. The primary endpoint may have Number of patients who received each study drug injection (full analysis set). been unsuitable for determining the efficacy of ASP0113 in this Study drug ASP0113 (n = 246) Placebo (n = 255) Total (n = 501) patient population, however. Instead, one might speculate that an injection, n (%) endpoint such as clinically significant CMV infections may have been more appropriate; this has recently been used in two studies of pro- First injection 246 (100) 254 (99.6) 500 (99.8) Second injection 207 (84.1) 213 (83.5) 420 (83.8) phylactic antiviral therapies. [6,20] In the analysis of the secondary Third injection 182 (74.0) 192 (75.3) 374 (74.7) endpoints in this study, however, this alternative endpoint would Fourth injection 173 (70.3) 189 (74.1) 362 (72.3) not have altered the outcome of the study. Interestingly, ASP0113 did Fifth injection 148 (60.2) 155 (60.8) 303 (60.5) not induce a potent CMV-specific immune response, and the immune Patients received intramuscular doses of either ASP0113 or placebo on Days 14 response to pp65 in the ASP0113 group was inferior to that with pla- to 3, 14 to 40, 60§5, 90§10, and 180§10. cebo, which contrasts with the results of the phase 2 study and other studies investigating vaccines for the prevention of CMV infection. [14,21] The explanation for this finding is unknown as both our study (eg, injection-site pain, injection-site erythema, injection-site indura- and the phase 2 study used ASP0113 that was produced via the same tion, and injection-site swelling) in the ASP0113 group compared manufacturing process and met the same release criteria. In addition, with the placebo group (Table 8). all laboratories included in our study adhered to standard quality SAEs were reported in 221 (89.8%) patients in the ASP0113 group control measures, so it is unlikely that there were differences in test- and 221 (86.7%) patients in the placebo group. TEAEs leading to death ing methodologies between the central and local laboratories. The that were considered by investigators to be study drug-related patient populations in this phase 3 study and the phase 2 study met occurred in four (1.6%) patients in the ASP0113 group (haemolytic similar inclusion and exclusion criteria; the main difference between anaemia, organising pneumonia, recurrent acute myeloid leukaemia, the studies was that this phase 3 study enrolled patients from numer- and aGVHD in the intestine and liver) and one (0.4%) patient in the ous countries, whereas the phase 2 study enrolled patients from 16 placebo group (acute respiratory distress syndrome). TEAEs resulting US centres. [14] In addition, the pre-transplant dose of ASP0113 was in the permanent discontinuation of the study drug were reported in administered earlier in the phase 2 study (Days 21 to 2) compared 48 (19.5%) and 45 (17.6%) patients in the ASP0113 and placebo with this phase 3 study (Days 14 to 3), and this phase 3 study con- groups, respectively. Overall, 135 (26.9%) patients died during the tained an additional dose of ASP0113 at Day 60§5. study (67 [27.2%] in the ASP0113 group and 68 [26.7%] in the placebo In theory, ASP0113 could have limited the risk of late CMV infec- group). tion, but these data were not collected, which is a limitation of the 6 P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 Fig. 2. Time to composite of all-cause mortality and adjudicated CMV EOD (full analysis set) *Parameter estimate from a Cox proportional hazards model with treatment and ran- domisation strata. p-value based on Cox proportional hazards model parameter estimate for the treatment effect. Circles indicate censored observations. CMV=cytomegalovirus; EOD=end-organ disease; HR=hazard ratio. Table 4 Secondary endpoints (full analysis set). ASP0113 (n = 246) Placebo (n = 255) HR/1-HR/OR (95% CI) p-value CMV viraemia rate, % (95% CI) 56.7 (50.1 to 62.8) 58.6 (52.0 to 64.6) HR 0.96 (0.76 to 1.22) p = 0.748 Adjudicated CMV-specific antiviral therapy rate, % (95% CI) 54.6 (48.1 to 60.6) 53.2 (46.8 to 59.1) HR 1.02 (0.80 to 1.29) p = 0.888 A composite of protocol-defined CMV viraemia and adjudicated CMV-specific 150 (61.0) 155 (60.8) OR 1.05 (0.73 to 1.51) p = 0.802 antiviral therapy use, n (%) First occurrence of adjudicated CMV-specific antiviral therapy or adjudicated 55.4 (48.9 to 61.5) 54.4 (48.0 to 60.3) 1-HR 0.01 (0.28 to 0.20) p = 0.928 CMV EOD rate, % (95% CI) Mortality, n (%) 78 (31.7) 72 (28.2) OR 1.18 (0.81 to 1.73) p = 0.393 1-HR=1-hazard ratio; CI=confidence interval; CMV=cytomegalovirus; EOD=end-organ disease; HR=hazard ratio; OR=odds ratio. Table 5 Exploratory efficacy endpoints (full analysis set). n (%) ASP0113 (n = 246) Placebo (n = 255) Odds ratio (95% CI) p-value Composite endpoint of all-cause mortality and adjudicated CMV EOD at 6 months 54 (22.0) 45 (17.6) 1.36 (0.87 to 2.13) p = 0.184 post-transplant, n (%) Incidence of adjudicated CMV EOD, n (%) 15 (6.1) 9 (3.5) 1.81 (0.78 to 4.19) p = 0.162 Incidence of grade 34 aGVHD, n (%) 26/239 (10.9) 30/252 (11.9) 0.89 (0.51 to 1.55) p = NR Incidence of graft rejection/poor graft function, n (%) 5/239 (2.1) 7/252 (2.8) 0.75 (0.23 to 2.42) p = NR Incidence of engraftment, n (%) 238/239 (99.6) 248/252 (98.4) 3.82 (0.41 to 35.39) p = NR aGVHD=acute graft-versus-host disease; CI=confidence interval; CMV=cytomegalovirus; EOD=end-organ disease; NR=not reported. study. However, considering the absence of a potent CMV-specific mechanisms of action and it remains to be seen if these are more immune response, this possibility is unlikely. The development of an effective for the prevention of CMV EOD. effective CMV vaccine was viewed as a promising therapeutic option Although this was a global, multicentre study of CMV-seropositive to prevent CMV EOD, but the failure of this study has underlined the allo-HCT recipients, we recognise that the proportion of African and/ difficulty of producing an effective vaccine for this purpose. Other or Afro-American allo-HCT recipients was low. However, we do not vaccines against CMV are in development using different believe that the results of our study would have been substantially P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 7 Fig. 3. (A) Time to first protocol-defined CMV viraemia through 1 year post-transplant. (B) Time to first use of adjudicated CMV-specific antiviral therapy after first study vaccine injection through 1 year post-transplant *Parameter estimate from a Cox proportional hazards model with treatment and randomisation strata, adjusting for death as a competing risk. p-value based on Cox proportional hazards model parameter estimate for the treatment effect. Circles indicate censored observations. CMV=cytomegalovirus; HR=hazard ratio. 8 P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 Table 6 Subgroup analyses of the primary endpoint (full analysis set). n/N (%) ASP0113 Placebo Odds ratio (95% CI) p-value ATG included in conditioning regimen 16/43 (37.2) 11/45 (24.4) 1.83 (0.73 to 4.59) p = 0.382 ATG not included in conditioning regimen 71/203 (35.0) 66/210 (31.4) 1.17 (0.78 to 1.77) Myeloablative conditioning regimen 35/120 (29.2) 34/122 (27.9) 1.07 (0.61 to 1.86) p = 0.563 Non-myeloablative conditioning regimen 50/124 (40.3) 43/131 (32.8) 1.38 (0.83 to 2.31) ATG=anti-thymocyte globulin; CI=confidence interval. Table 7 5. Contributors Summary of treatment-emergent adverse events and deaths (safety analysis set). Parameter, n (%) ASP0113 (n = 246) Placebo (n = 255) All authors met the International Committee of Medical Journal Editors authorship criteria. All authors were equally involved in the TEAEs 246 (100) 255 (100) design or conception of the study, the analysis and interpretation of Drug-related* TEAEs 194 (78.9) 74 (29.0) SAEs 221 (89.8) 221 (86.7) the data and drafting and critically reviewing the publication. All Drug-related* serious TEAEs 22 (8.9) 13 (5.1) authors provided written approval for the manuscript to be submit- TEAEs leading to death 61 (24.8) 63 (24.7) ted to EClinicalMedicine. Dr Ljungman, Dr Bermudez, Dr Logan, Dr Drug-related* TEAEs leading to 4 (1.6) 1 (0.4) Kharfan-Dabaja, Dr Chevallier, Dr Martino, Dr Wulf, Dr Selleslag, Dr death TEAEs leading to permanent dis- 48 (19.5) 45 (17.6) Kakihana, Dr Langston, Dr Lee, Dr Solano, Dr Okamoto, Dr Boeckh, Dr continuation of the study drug Wingard and Dr Maertens were equally involved in the acquisition of Drug-related TEAEs leading to per- 5 (2.0) 4 (1.6) the data as investigators. manent discontinuation of the study drug Deathsy 67 (27.2) 68 (26.7) Data sharing * Possible or probable, as assessed by the investigator, or records where rela- tionship is missing. Access to anonymised individual participant level data collected y Deaths listed here are those from transplant through to the day after Day 380. during the trial, in addition to supporting clinical documentation, is AE=adverse event; SAE=serious adverse event; TEAE=treatment-emergent adverse planned for trials conducted with approved product indications and event. formulations, as well as compounds terminated during develop- ment. Conditions and exceptions are described under the Sponsor different if there was a larger African and/or Afro-American popula- Specific Details for Astellas on www.clinicalstudydatarequest.com. tion. In addition, no site contributed more than 5.4% of the total num- Study-related supporting documentation is redacted and provided ber of patients in our study. Overall, 46% of patients were if available, such as the protocol and amendments, statistical analy- randomised at EU sites, 33% at US sites, 18% at Asian sites and 3% at sis plan, and clinical study report. Access to participant-level data is Canadian sites. These data suggest that clustering in each of the par- offered to researchers after publication of the primary manuscript ticipating countries was not a limitation of our study. (if applicable) and is available as long as Astellas has legal authority Overall, safety findings were similar between the ASP0113 and to provide the data. Researchers must submit a proposal to conduct placebo groups, with the exception of a higher percentage of drug- a scientifically relevant analysis of the study data. The research pro- related TEAEs in the ASP0113 group, which were mostly due to mild posal is reviewed by an Independent Research Panel. If the proposal to moderate injection-site reactions that are commonly associated is approved, access to the study data is provided in a secure data- with vaccines [10,14]. sharing environment after receipt of a signed Data Sharing Agree- In conclusion, ASP0113 was not efficacious in the reduction of ment. overall mortality and CMV EOD, and did not appear to be effective at reducing CMV viraemia through 1 year post-transplant in allo-HCT Funding recipients. In addition, ASP0113 did not elicit a potent T-cell or B-cell response compared with placebo, but was associated with a similar The study sponsor, Astellas Pharma Global Development Inc., was safety profile to that of placebo with the exception of injection-site- involved in the design or conception of the study, the analysis and related TEAEs, which were more frequent in the ASP0113 group. interpretation of the data and drafting and critically reviewing the Owing to a lack of efficacy, the development of ASP0113 has been ter- publication. The corresponding author had full access to all the data minated. Participants in this study will continue to be followed up for in the study and had final responsibility for the decision to submit 5.5 years post-transplant for long-term safety assessments. this final version for publication. Table 8 Drug-related treatment-emergent adverse events experienced by >5% of patients (safety analysis set). Drug-related TEAE occurring in >5% of patients, n (%) ASP0113 (n = 246) Placebo (n = 255) Total Mild Moderate Severe Total Mild Moderate Severe Injection-site pain 183 (74.4) 120 (48.8) 62 (25.2) 1 (0.4) 39 (15.3) 38 (14.9) 1 (0.4) 0 Injection-site erythema 45 (18.3) 39 (15.9) 6 (2.4) 0 4 (1.6) 4 (1.6) 0 0 Injection-site induration 36 (14.6) 32 (13.0) 4 (1.6) 0 2 (0.8) 2 (0.8) 0 0 Injection-site swelling 24 (9.8) 20 (8.1) 4 (1.6) 0 4 (1.6) 4 (1.6) 0 0 TEAE severity grading was according to MedDRA version 16.0, mild: grade 01; moderate: grade 2; and severe: grade 3. MedDRA=Medical Dictionary for Regula- tory Activities; TEAE=treatment-emergent adverse event. P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 9 Fig. 4. T-cell response to pp65 (immunogenicity analysis set) Data are mean and standard deviation. Interaction and overall p-value are based on the repeated measure model with post-baseline T-cell response to pp65 values on the 10 based log scale as the outcome variable, randomisation stratum, treatment, time, and treatment by time interaction as the explanatory variables. The self-consistent sandwich correlation structure was used. As zero values are possible these were replaced by 1 before taking the log transform of the pp65 values. *p-value is based on Wilcoxon rank sum test at each visit. Circles indicate censored observations. Time intervals on the x-axis are not equal. ELISpot=enzyme-linked immune absorbent spot; PBMC=peripheral blood mononuclear cell; pp65=phosphoprotein 65; SFU=spot-forming units. Fig. 5. gB antibody levels (immunogenicity analysis set) Data are mean and standard deviation. Interaction and overall p-value are based on the repeated measure model with post- baseline gB antibody level values on the 10 based log scale as the outcome variable, randomisation stratum, treatment, time, and treatment by time interaction as the explanatory variables. The self-consistent sandwich correlation structure was used. Time intervals on the x-axis are not equal. *p-value is based on Wilcoxon rank sum test at each visit. Circles indicate censored observations. gB=glycoprotein B; RU=relative units. 10 P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 Appendix Dr Fredericks reports personal fees from Astellas, during the con- Appendix 1. Study protocol duct of the study; personal fees from Astellas, outside the submitted work; and is an employee of Astellas. Appendix 2. Study design Dr Lademacher reports personal fees from Astellas, during the CMV=cytomegalovirus; MA=myeloablative conditioning; conduct of the study, and is an employee of Astellas. pp65=phosphoprotein 65; RIC=reduced-intensity conditioning; Dr Wang reports other from Astellas, during the conduct of the rtPCR=reverse transcription polymerase chain reaction; SOC=stan- study; other from Astellas Pharma global Development, outside the dard of care. submitted work; and employment by Astellas. Dr Young reports personal fees from Astellas, during the conduct of the study, and is an employee of Astellas. Declaration of Competing Interest Dr Maertens reports personal fees and non-financial support from Astellas, grants, personal fees and non-financial support from Merck, Dr Ljungman reports grants, personal fees and non-financial sup- personal fees and non-financial support from Pfizer, personal fees port from Astellas personal fees from Vical, during the conduct of the and non-financial support from Basilea, personal fees and non-finan- study; personal fees from AiCuris, grants from Merck, grants from cial support from F2G, personal fees and non-financial support from Shire, outside the submitted work. Cidara, grants, personal fees and non-financial support from Gilead, Dr Bermudez reports other from Astellas, during the conduct of outside the submitted work. the study. Dr Logan reports other from Astellas, during the conduct of the Acknowledgements study; other from Novartis, outside the submitted work. Dr Kharfan-Dabaja reports other from Seattle Genetics, other from Medical writing support in the development of this publication Alexion Pharmaceuticals, other from Incyte, personal fees from Daii- was provided by Matthew Reynolds of OPEN Health Medical Commu- chi Sankyo, personal fees from Pharmacyclics, outside the submitted nications, London, UK, which was funded by Astellas Pharma Global work. Development, Inc. Dr Chevallier has nothing to disclose. Dr Martino reports other from Astellas, during the conduct of the study. Supplementary materials Dr Wulf reports personal fees from Astellas, during the conduct of the study. Supplementary material associated with this article can be found, Dr Selleslag reports personal fees from Astellas Pharma Global in the online version, at doi:10.1016/j.eclinm.2021.100787. Development, Inc., personal fees from GlaxoSmithKline, personal fees from MSD, personal fees from Pfizer, during the conduct of the study. References Dr Kakihana reports personal fees from Chugai Pharmaceutical Co. Ltd, personal fees from Kyowa Hakko Kirin, personal fees from Bris- [1] Kharfan-Dabaja MA, Nishihori T. Vaccine therapy for cytomegalovirus in the set- tol-Myers Squibb, personal fees from Takeda Pharmaceutical Co., per- ting of allogeneic hematopoietic cell transplantation. Expert Rev Vaccines 2015;14(3):341–50. sonal fees from Dainippon Sumitomo Pharma, outside the submitted [2] Boeckh M, Ljungman P. How we treat cytomegalovirus in hematopoietic cell work. transplant recipients. Blood 2009;113(23):5711–9. Dr Langston has nothing to disclose. [3] Boeckh M, Gooley TA, Myerson D, Cunningham T, Schoch G, Bowden RA. Cyto- megalovirus pp65 antigenemia-guided early treatment with ganciclovir versus Dr Lee reports grants from Astellas, during the conduct of the ganciclovir at engraftment after allogeneic marrow transplantation: a random- study; grants from GSK, personal fees from MSD, personal fees from ized double-blind study. Blood 1996;88(10):4063–71. Pfizer, personal fees from Gilead, personal fees from SL Vaxigen, out- [4] Salzberger B, Bowden RA, Hackman RC, Davis C, Boeckh M. Neutropenia in alloge- neic marrow transplant recipients receiving ganciclovir for prevention of cyto- side the submitted work. megalovirus disease: risk factors and outcome. Blood 1997;90(6):2502–8. Dr Solano reports grants from Astellas Pharma Global Develop- [5] Goodrich JM, Bowden RA, Fisher L, Keller C, Schoch G, Meyers JD. Ganciclovir pro- ment, Inc., during the conduct of the study; personal fees from Gilead, phylaxis to prevent cytomegalovirus disease after allogeneic marrow transplant. Ann Intern Med 1993;118(3):173–8. personal fees from Mitsubishi Tanabe Pharma, outside the submitted [6] Marty FM, Ljungman P, Chemaly RF, et al. Letermovir prophylaxis for cytomegalo- work. virus in hematopoietic-cell transplantation. N Engl J Med 2017;377(25):2433–44. Dr Okamoto reports grants, personal fees and other from Astellas [7] FDA. Letermovir prescribing information. URL: https://www.accessdata.fda.gov/ Pharma, during the conduct of the study. drugsatfda_docs/label/2017/209939Orig1s000,209940Orig1s000lbl.pdf. Accessed September 2018. Dr Smith reports personal fees from Astellas Pharma, during the [8] EMA. Letermovir summary of product characteristics. URL: https://ec.europa.eu/ conduct of the study. health/documents/community-register/2018/20180108139609/anx_139609_en. Dr Boeckh reports grants from Astellas Pharma Global Develop- pdf. Accessed December 2018. [9] Mori T, Kanda Y, Takenaka K, et al. Safety of ASP0113, a cytomegalovirus DNA vac- ment Inc, during the conduct of the study; grants and personal fees cine, in recipients undergoing allogeneic hematopoietic cell transplantation: an from Gilead, grants and personal fees from Merck, grants and per- open-label phase 2 trial. Int J Hematol 2017;105(2):206–12. sonal fees from Takeda, grants and personal fees from Vir Bio, per- [10] Griffiths PD, Stanton A, McCarrell E, et al. Cytomegalovirus glycoprotein-B vaccine with MF59 adjuvant in transplant recipients: a phase 2 randomised placebo-con- sonal fees from Allovir, personal fees from GlaxoSmithKline, personal trolled trial. Lancet 2011;377(9773):1256–63. fees from Moderna, personal fees from Oxford Immunotec, personal [11] Chiuppesi F, Nguyen J, Park S, et al. Multi-antigenic modified vaccinia ankara vac- fees from Evrys Bio, personal fees from Helocyte, grants from Lophius cine vectors to elicit potent humoral and cellular immune reponses against human cytomegalovirus in mice. J Virol 2018. doi: 10.1128/JVI.01012-18. Biosciences, outside the submitted work. [12] Smith LR, Wloch MK, Chaplin JA, Gerber M, Rolland AP. Clinical development of a Dr Wingard reports personal fees from Astellas Pharma Global cytomegalovirus DNA vaccine: from product concept to pivotal phase 3 trial. Vac- Development, Inc., during the conduct of the study; personal fees cines (Basel) 2013;1(4):398–414. [13] Wloch MK. Safety and immunogenicity of a bivalent CMV DNA vaccine in healthy from Allovir, personal fees from Ansun, personal fees from Celgene, adult subjects. J Infect Dis 2008;197(12):1634–42. personal fees from Cidara, personal fees from Merck, personal fees [14] Kharfan-Dabaja MA, Boeckh M, Wilck MB, et al. A novel therapeutic cytomegalo- from Pluristem, personal fees from Shire, outside the submitted virus DNA vaccine in allogeneic haemopoietic stem-cell transplantation: a rando- work. mised, double-blind, placebo-controlled, phase 2 trial. Lancet Infect Dis 2012;12 (4):290–9. Dr Cywin reports personal fees from Astellas, during the conduct [15] Vincenti F, Budde K, Merville P, et al. A randomized, phase 2 study of ASP0113, a of the study, and is an employee of Astellas. DNA-based vaccine, for the prevention of CMV in CMV-seronegative kidney P. Ljungman et al. / EClinicalMedicine 33 (2021) 100787 11 transplant recipients receiving a kidney from a CMV-seropositive donor. Am J [19] Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a com- Transplant 2018;18:2945–54. peting risk. J Am Stat Assoc 1999;94(446):496–509. [16] Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)-spe- [20] Marty FM, Winston DJ, Chemaly RF, et al. A randomized, double-blind, placebo- cific comorbidity index: a new tool for risk assessment before allogeneic HCT. controlled phase 3 trial of oral brincidofovir for cytomegalovirus prophylaxis in Blood 2005;106(8):2912–9. allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant [17] Medical Dictionary for Regulatory Activities (MedDRA) version 16.0. URL: https:// 2018;25(2):369–81. www.meddra.org/. Accessed September 2013. [21] Nakamura R, La Rosa C, Longmate J, et al. Viraemia, immunogenicity, and survival [18] National Cancer Institute and National Institutes of Health. Common Terminology outcomes of cytomegalovirus chimeric epitope vaccine supplemented with Criteria for Adverse Events (CTCAE) version 4.03. URL: https://www.eortc.be/ PF03512676 (CMVPepVax) in allogeneic haemopoietic stem-cell transplantation: services/doc/ctc/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf. Accessed randomised phase 1b trial. The Lancet Haematol 2016;3(2):e87–98. August 2020.