A Phase 1 Study of 4 Live, Recombinant Human Cytomegalovirus Towne/Toledo Chimeric Vaccines

Transcription

1 MAJOR ARTICLE A Phase 1 Study of 4 Live, Recombinant Human Cytomegalovirus Towne/Toledo Chimeric Vaccines Thomas C. Heineman, 1 Mark Schleiss, 2,a David I. Bernstein, 2 Richard R. Spaete, 3 Lihan Yan, 5 Greg Duke, 3 Mark Prichard, 3 Zhaoti Wang, 3 Qing Yan, 3 Margaret A. Sharp, 4 Nicola Klein, 4 Ann M. Arvin, 4 and George Kemble 3 1 Division of Infectious Diseases and Immunology, Saint Louis University School of Medicine, St. Louis, Missouri; 2 Division of Pediatric Infectious Diseases, Children s Hospital Medical Center, University of Cincinnati, Cincinnati, Ohio; 3 MedImmune Vaccines, Mountain View, and 4 Department of Pediatrics, Stanford University School of Medicine, Stanford, California; 5 EMMES, Rockville, Maryland Background. Human cytomegalovirus (HCMV) infection acquired in utero often results in severe consequences, including mental retardation and deafness. Although not evaluated for this indication, live attenuated HCMV vaccines based on the Towne strain are well-tolerated and have demonstrated moderate efficacy in other clinical settings. Methods. To produce live HCMV vaccine candidates that retain the excellent safety profile of the Towne strain but are more immunogenic, the genomes of the Towne strain and the unattenuated HCMV Toledo strain were recombined to yield 4 independent chimeric vaccine candidates. These vaccine candidates were evaluated in 20 HCMV-seropositive persons, in a phase 1, double-blinded, placebo-controlled trial. Participants received a single dose of vaccine or placebo, and the safety and tolerability of the vaccine candidates were evaluated. Results. There was no difference in systemic symptoms between the vaccine and placebo recipients. As a group, vaccine recipients experienced more injection-site reactions than did placebo recipients; however, these were generally minor and short-lived. Vaccine virus could not be detected in blood, urine, or saliva samples obtained from any vaccine recipient. Conclusions. The Towne/Toledo chimeric vaccine candidates were well tolerated and did not cause systemic infection. Additional human trials are warranted to further evaluate the potential of these vaccine candidates as live virus vaccines. Human cytomegalovirus (HCMV) is the leading infectious cause of congenital mental retardation and deafness. It is estimated that in utero infection with HCMV causes death or serious sequelae in children each year in the United States [1]. Because of the profound effects of HCMV on health, and because of the financial burden that the virus imposes on the health care system in the United States, the Institute of Medicine recently Received 22 July 2005; accepted 9 December 2005; electronically published 12 April Potential conflicts of interest: A.M.A. has a consulting relationship with MedImmune Vaccines. G.K., R.R.S., G.D., M.P., Z.W., and Q.Y. are employed by MedImmune Vaccines. All other authors: no conflicts. Financial support: National Institute of Allergy and Infectious Diseases (contract N01-AI 25464). a Present affiliation: Division of Pediatric Infectious Diseases, University of Minnesota School of Medicine, Minneapolis. Reprints or correspondence: Dr. Thomas C. Heineman, Div. of Infectious Diseases and Immunology, Saint Louis University School of Medicine, 3635 Vista Ave., FDT- 8N, St. Louis, MO (heinemtc@slu.edu). The Journal of Infectious Diseases 2006; 193: by the Infectious Diseases Society of America. All rights reserved /2006/ $15.00 included HCMV among the highest-priority targets for vaccine development [2]. Efforts to develop an HCMV vaccine began 130 years ago, with the generation of live attenuated HCMV strains [3, 4]. The best studied of these HCMV strains, the Towne strain, is well tolerated and has an excellent safety record on the basis of the findings of numerous clinical trials; however, long-term evaluations of this strain have not been performed. The Towne strain has never been isolated from blood, urine, or saliva samples obtained from vaccinated persons; this finding suggests that the Towne strain fails to establish systemic infection [5 10]. Efficacy testing of the Towne vaccine has been performed in immunocompetent and immunocompromised persons [5 7, 10 23]. Renal transplant recipients vaccinated with the Towne strain were partially protected against severe HCMV disease; however, they acquired HCMV infection at the same rate as did unvaccinated control subjects [13, 19, 20]. Similarly, vaccination with the Towne strain, unlike natural HCMV infection, failed to protect immunocompetent women 1350 JID 2006:193 (15 May) Heineman et al.

2 Figure 1. Structural features and similarity of the Toledo and Towne genomes. Dot plot depicting the nucleotide identity between the Toledo and Towne genomes. Windows of 10 nucleotides were compared and were scored as positive if all 10 were identical. The diagonal line denotes the high level of collinear identity. The UL/b region of the Toledo strain is denoted by the gap in the diagonal line. against primary HCMV infection [12]. Finally, in a small challenge study, Towne vaccine recipients were partially protected against HCMV disease and infection, but they were less protected than were naturally HCMV-seropositive individuals [21]. Together, these data suggest that the Towne vaccine affords only modest protection against HCMV disease and that it has little, if any, ability to prevent HCMV infection. The limited protection afforded by the Towne vaccine may be caused by overattenuation resulting from extensive passage in cultured cells [24]. To produce a live HCMV vaccine that is more immunogenic than the Towne vaccine, we constructed genetic recombinants in which regions from the unattenuated Toledo strain of HCMV were substituted for the corresponding regions of the Towne genome. Because it is unknown which substitutions would produce an attenuated phenotype, we constructed 4 different chimeras. In the present report, we report the safety, tolerability, and attenuation of these 4 HCMV vaccine candidates when administered to healthy HCMV-seropositive study volunteers. MATERIALS AND METHODS Study sites. Study participants were enrolled and followed at the Saint Louis University Center for Vaccine Development (St. Louis, MO) or Cincinnati Children s Hospital Medical Center (Cincinnati, OH). Vaccines. Libraries of overlapping cosmid clones were prepared from genomic DNA of the HCMV TownevarRIT and Toledo strains [25, 26]. For each strain, 8 cosmids were selected that spanned the entire HCMV genome and, thus, had the capacity to regenerate infectious virus after cotransfection. The vaccine candidates (chimeras 1 4) were constructed by cotransfection of selected Towne and Toledo cosmids [25]. In addition, the Toledo unique long (UL)/b region was incorporated into each vaccine virus. This region serves as a genetic marker for the vaccine viruses, because its orientation in the chimeras is opposite that in clinical HCMV isolates. To generate the clinical trial materials, genomic DNA from the vaccine candidates was used to transfect MRC-5 cells from a highly characterized bank. The resulting vaccine viruses were amplified by infection of MRC-5 cells and were harvested from the supernatants. The placebo consisted of the vaccine diluent alone. Volunteers. Twenty-nine volunteers were enrolled in the study from June 2000 through May Twenty-five of these study volunteers were enrolled as vaccine or placebo recipients, and 4 were enrolled as close contacts. Close contacts were defined as (1) persons who come in direct contact with the bodily fluids of the vaccinated person through sexual relations or kissing, or (2) household members who share personal items, such as toothbrushes, razors, or eating utensils, without washing the items prior to using them. All volunteers, including those enrolled in the study as close contacts, were required to be years of age and HCMV seropositive. Female vaccine recipients were required to be surgically sterile or postmenopausal. Prospective vaccine recipients with close contacts were not enrolled unless their close contacts were also enrolled as part of a separate group that did not receive vaccine or placebo. Persons were excluded from study participation if they were immunodeficient or had a serious disease (e.g., cancer, HIV infection, autoimmune disease, chronic renal failure, or diabetes), or if they were hepatitis C antibody positive or hepatitis B surface antigen positive. They were also excluded if their complete blood cell counts, liver function test results, creatinine or glucose levels, or coagulation times were outside the normal laboratory value range. Persons were excluded as vaccine or placebo recipients if they had children!12 years of age in their home; were child care providers; or were health care providers who routinely came in contact with pregnant women or immunologically compromised persons, or who participated directly in surgical procedures. The present study was approved by the institutional review boards of both participating institutions. Vaccination and study design. The study was doubleblinded, randomized, and placebo controlled. The 25 volunteers who were enrolled as vaccine or placebo recipients were randomized to 1 of 5 groups that included 5 persons per group. Persons in 4 of the groups (groups 1 4) received 1 of the 4 investigational vaccines administered as a dose of 1000 pfu; persons in the fifth group (group 5) received placebo. The investigational vaccines or placebo was administered subcutaneously over the deltoid muscle as a single 0.5-mL dose. The study volunteers had clinical evaluations performed at weeks 1 8, 16, 24, HCMV Towne/Toledo Chimeric Vaccines JID 2006:193 (15 May) 1351

3 Table 1. Contiguous nucleotide sequences (of 110 nucleotides) with!95% identity between the Toledo and Towne strains of human cytomegalovirus. Toledo strain nucleotide coordinates Sequence length, no. of nucleotides Identity between strains, % a 3249 b b b b c b a The b /UL region. b One of the 5 largest contiguous regions with sequence identity of!95%. c The UL/b region. and 52 after inoculation. Close contacts were evaluated at weeks 4, 8, and 12 after inoculation of the vaccine recipient with whom they had contact. Additional evaluations of close contacts beyond week 12 were planned in the event that the vaccinated person with whom they had contact was found to be shedding HCMV. Clinical assessment. Participants completed daily diary cards to describe local reactions for 4 weeks after immunization and systemic reactions for 8 weeks after immunization. Solicited local reactions included pain, erythema, induration, and tenderness. Solicited systemic symptoms included fever (oral temperature, 38.0 C), rash, malaise, myalgias, sore throat, headache, and nausea. The symptoms recorded on the diary cards were reported as mild (causing transient or mild discomfort), moderate (causing mild-to-moderate limitation of normal daily activity), or severe (causing marked limitation of normal daily activity). Clinical evaluations consisted of an interval medical history and physical examination. Specific symptoms and signs that were prospectively identified as likely sequelae of HCMV infection were graded on a scale from 1 (least severe) to 4 (most severe), according to preestablished criteria. These symptoms and signs included fever, headache, arthralgia and/or arthritis, myalgias, rash, and nausea and/or vomiting. In addition, the presence of lymphadenopathy or splenomegaly was specifically noted. Clinical laboratories. Laboratory tests to assess vaccine safety were completed at every study visit. These tests included complete blood cell counts with differential and platelet counts, liver function tests (for the assessment of alanine aminotransferase [ALT], aspartate aminotransferase [AST], alkaline phosphatase, and total bilirubin levels), determination of coagulation times, and measurement of the creatinine level. Abnormal laboratory findings that were prospectively identified as being particularly likely to denote the presence of HCMV infection (abnormal absolute neutrophil count and platelet count and abnormal hemoglobin, creatinine, ALT, AST, alkaline phosphatase, and total bilirubin levels) were graded on a scale from 1 (least severe) to 4 (most severe), according to preestablished criteria. Virus detection assays. At every study visit, blood, urine, and saliva samples were collected for HCMV culture performed using routine and shell vial methods [27 29]. In addition, all samples were analyzed by polymerase chain reaction (PCR) for the presence of HCMV DNA. In brief, nucleic acids were extracted from the samples and were amplified by PCR (Qiagen Hot Start kit), by use of primers that bound to UL54. Different primer pairs that identified the orientation of the UL/b region were used to distinguish vaccine strains from field isolates. Immunologic assays. Total IgG and neutralizing anti-hcmv serum antibody titers were determined for blood samples collected before vaccination (at week 0) and at weeks 4, 8, 24, and 52 after vaccination. Total titers of HCMV IgG to whole HCMV 1352 JID 2006:193 (15 May) Heineman et al.

4 Figure 2. Schematic representation of chimeric strains generated by recombination of the Towne and Toledo parental strains. A, Structure of the human cytomegalovirus Towne and Toledo genomes. Shown for the TowneAV variant is the sequence arrangement of the 235-kb genome with the unique L (UL) and S (US) components in the prototypic orientation bracketed by inverted repeats ab-b a and a c -ca. Gray lines denote the Towne genomic sequence, and black lines denote the Toledo genomic sequence contribution to the chimeras. Boxes denote repeat sequences on both genomes drawn roughly to scale. The unique sequences present in the Toledo genome are depicted as hatched segments, and their orientation is denoted by the arrow in the Toledo genome. B, One-step replication curves of the TowneAV variant, the Toledo strain, and the 4 vaccine chimeras on MRC-5 cells. C, Analysis of the structure of the parental and chimeric genomes by restriction enzyme fingerprinting. EcoRI digests of the genomes are shown as ethidium bromide stained fragments. antigen were quantitated by ELISA, and HCMV neutralizing antibody titers were determined using a fluorescence-based microneutralization assay, with the Towne and Toledo strains used separately as targets [30]. Cellular immune responses were evaluated by lymphoproliferation, enzyme-linked immunospot assay, and intracellular cytokine assays, by use of lymphocytes isolated from blood samples collected at weeks 0, 4, 8, 24, and 52 [31 33]. Statistical methods. Clinical safety was determined by evaluation of local and systemic reactions, adverse events, and clinical laboratory results. Sample sizes were calculated to have a high probability of detecting major safety-related events when the underlying true event rate was high ( 50%). Sample sizes were, however, insufficient to detect differences between vaccine candidates. Although the primary focus of this trial was the estimation of the safety and immunogenicity of the individual vaccine candidates, data from all vaccine candidates were pooled to allow for initial determination of whether the vaccine HCMV Towne/Toledo Chimeric Vaccines JID 2006:193 (15 May) 1353

5 Table 2. Demographic characteristics of study participants. Characteristic Group 1 Towne/Toledo chimeric vaccine candidate recipients Group 2 Group 3 Group 4 All (n p 20) Placebo recipients Close contacts a (n p 4) Age, years, mean (range) 42 (33 55) 39 (30 44) 38 (24 47) 37 (27 55) 39 (24 55) 45 (31 56) 38 (19 74) Sex Male (65) 4 (80) 2 (50) Female (35) 1 (20) 2 (50) Race/ethnicity Black (35) 2 (40) 2 (50) Hispanic (5) 0 (0) 0 (0) Non-Hispanic white (60) 3 (60) 2 (50) NOTE. Data are the no. or no. (%) of study participants, unless indicated otherwise. a Close contacts were defined as (1) persons who come in direct contact with the bodily fluids of the vaccinated person through sexual relations or kissing, or (2) household members who share personal items, such as toothbrushes, razors, or eating utensils, without washing the items prior to using them. candidates have the negative safety profile of a Toledo-like virus. Therefore, descriptive statistics are provided. Fisher s exact test was used to compare rates of adverse events, where appropriate. A binomial distribution was used to calculate the probability of having observed a number of events given the historical expectation. RESULTS Construction and characterization of the vaccine candidates. Human trials have shown that the Towne strain is highly attenuated and suboptimally immunogenic, whereas the Toledo strain, similar to wild-type HCMV, induces robust immune responses [12, 13, 19 21, 24]. These biological differences are predicted to result from genomic variations between the Towne and Toledo strains [22]; therefore, the sequences of the 2 viruses were compared. Aside from the previously described 13-kb UL/b region unique to the Toledo strain [22, 34], the nucleotide sequences of these 2 strains were 195% identical (figure 1). However, there were many regions in which sequence identity between the Towne and Toledo strains was lower (table 1). To construct the vaccine candidates, cosmid clones containing Towne and Toledo sequences were cotransfected into human fibroblasts. In each of the resulting chimeric viruses, portions of the Towne genome were replaced by colinear regions of the Toledo genome (figure 2A). The Towne/Toledo chimeras grew at rates similar to each other and at rates that were intermediate to the growth of the parental viruses (figure 2B). Because regions with low sequence identity are distributed throughout the Towne and Toledo genomes, each vaccine candidate is predicted to contain Towne-specific genes that are likely to confer attenuation and Toledo-specific genes that may enhance immunogenicity. Genomic DNA from each of the chimeric viruses had distinct and predictable restriction enzyme profiles that were different from those of either parent strain (figure 2C). The genetic determinants of the Towne strain s attenuation are unknown; thus, all 4 Towne/Toledo chimeric vaccine candidates were tested in this initial study to determine which, if any, are well tolerated and attenuated relative to the Toledo strain. Volunteers. Seventeen men and 8 women were enrolled as vaccine or placebo recipients, and 2 men and 2 women were enrolled as close contacts (table 2). The mean age of vaccine recipients in groups 1 5 (38.9 years) was somewhat lower than that of placebo recipients (45 years). Most vaccine recipients (60%) were non-hispanic white, and most of the remainder (35%) were black. Two vaccine recipients and 1 placebo recipient withdrew before completion of the study (at weeks 6, 8, and 24); in no instance was withdrawal from the study related to adverse events. Clinical assessments of vaccine and placebo recipients. The combined incidences of injection-site pain, redness, and induration during the 4 weeks after inoculation were higher for the investigational vaccine recipients (groups 1 4) than for the placebo recipients (60% vs. 0%; P p.04, by Fisher s exact test) (table 3). Local reactions occurred in each group of investigational vaccine recipients. Two of 12 persons reporting local reactions had grade 2 events; all others persons had grade 1 events. Nine of 12 local reactions resolved within 4 days; only 1 local reaction lasted 19 days. During the 8 weeks after vaccination, systemic symptoms were reported by 14 (70%) of 20 investigational vaccine recipients and by 2 (40%) of 5 placebo recipients ( P p.31) (table 3). The small sample sizes limited the ability to detect differences between the vaccine and placebo recipients, and they also precluded the ability to compare vaccines. However, the occurrence of systemic symptoms was not evenly distributed among study groups, varying from 5 occurrences of symptoms among the 5 subjects in group 1 to 2 occurrences among the 5 subjects in group 4. Most participants with systemic symp JID 2006:193 (15 May) Heineman et al.

6 Table 3. Clinical adverse events reported by study participants during weeks 1 8 after vaccination. Towne/Toledo chimeric vaccine candidate recipients All, no. (%) (n p 20) Placebo recipients, no. (%) Group 1 Group 2 Group 3 Group 4 Clinical adverse event Symptom a General Myalgias (55) 1 (20) Malaise (50) 1 (20) Sore throat (50) 1 (20) b Headache (50) 0 (0) Nausea and/or vomiting 4 b (35) 0 (0) Rash (15) 0 (0) Any c (70) 2 (40) At the infection site Redness (25) 0 (0) Induration (15) 0 (0) Pain (55) 0 (0) Any c (60) 0 (0) Sign Fever 0 1 d (5) 1 (20) Splenomegaly (0) 0 (0) Adenopathy 0 1 e (5) 0 (0) Any (10) 1 (20) NOTE. Data are no. of study participants, unless indicated otherwise. a Data on solicited symptoms were collected on a diary card. b The placebo recipient with a sore throat and 1 subject in group 1 who had nausea and/or vomiting had adverse events of grade 3 severity; all other adverse events were of grade 1 (i.e., mild) or grade 2 (i.e., moderate) severity. c Some participants had 11 symptom or sign. d Subject 105 had fever on study days (peak temperature, 38.2 C). e Subject 217 had adenopathy (cervical, on the right only) during week 4. toms initially reported the symptoms during the first 4 weeks after vaccination (12 of 20 vaccine recipients and 2 of 5 placebo recipients). Only 2 participants reported having a fever during the 8 weeks after vaccination. On days after vaccination, a study volunteer in group 2 had a fever (maximum temperature, 38.2 C) that was associated with diarrhea and an elevated bilirubin level (2.2 mg/dl). The other occurrence of fever was reported by a placebo recipient. No vaccine-related adverse events occurred during weeks 8 52 after vaccination. During the course of the study, 4 vaccine or placebo recipients had serious adverse events. None of these events was related to vaccination. Laboratory evaluation of vaccine and placebo recipients. Thirteen participants (9 vaccine recipients and 4 placebo recipients) had abnormal results of blood tests to evaluate vaccine safety during the 8 weeks after vaccination (table 4). Two of 3 volunteers with elevated transaminase levels were placebo recipients. The other elevated transaminase level was noted in a vaccine recipient in group 2 (peak AST level, 66 U/L); however, this participant also had elevated transaminase levels on day 0 before vaccination (AST level, 98 U/L). A single participant (from group 4) had atypical lymphocytosis (9% atypical lymphocytes in a blood sample) noted at the 8-week postvaccination visit. Isolation of virus from vaccine and placebo recipients. To maximize the sensitivity of HCMV detection, we performed routine cultures, shell vial assays, and PCR for all samples. One vaccine recipient (from group 1) had positive results of HCMV cultures of saliva samples at weeks 7, 16, and 24. In each case, these organisms were shown to be wild-type HCMV isolates, on the basis of the orientation of their UL/b regions. (This method did not exclude the unlikely possibility that the cultured virus represented recombinants between vaccine and wild-type strains of HCMV.) Another subject (from group 1) had positive results of shell vial cultures of blood, urine, and saliva samples obtained at week 5. HCMV could not be isolated by routine culture of these samples, and the results of PCR analyses were negative. Therefore, the presence of virus could not be confirmed, and its origin (vaccine vs. wild-type) could not be determined. This study volunteer had no symptoms, signs, or laboratory findings that were suggestive of HCMV disease, and all samples collected at subsequent visits were found to be HCMV Towne/Toledo Chimeric Vaccines JID 2006:193 (15 May) 1355

7 Table 4. Laboratory abnormalities reported by study participants during weeks 1 8 after vaccination or receipt of placebo. Laboratory abnormality Towne/Toledo chimera vaccine candidate recipients Group 1 Group 2 Group 3 Group 4 All, no. (%) (n p 20) Placebo recipients, no. (%) Elevated level Alkaline phosphatase (0) 0 (0) Bilirubin (10) 0 (0) Transaminase 0 1 a (5) 2 (40) Atypical lymphocytosis b 1 (5) 0 (0) Anemia (15) 1 (20) Elevated WBC count (20) 1 (20) Low ANC (5) 1 (20) Any c (45) 4 (80) NOTE. Data are no. of study participants, unless indicated otherwise. ANC, absolute neutrophil count; WBC, white blood cell. a This study volunteer had elevated aspartate aminotransferase and alanine aminotransferase levels before vaccination. b Atypical lymphocytosis (9% atypical lymphocytes in a blood sample) was noted at week 8 after vaccination only. c Some participants had 11 laboratory abnormality. HCMV negative. Together, these data suggest that the vaccine viruses do not cause systemic infection. Humoral and cellular immune assays. Vaccine recipients and close contacts were evaluated for HCMV immune responses before vaccination and at weeks 4, 8, 24, and 52. There were no differences in the pre- and post-vaccination total or neutralizing antibody titers between the investigational vaccine and placebo recipients at any of the study visits (table 5). Cellmediated responses, analyzed by a responder cell frequency assay for HCMV antigen-specific CD4 + and CD8 + cells, failed to demonstrate a vaccination effect at any time point studied (table 5). Also, there were no significant differences in the number of interferon-g secreting cells, which were quantitated using an enzyme-linked immunospot assay, between vaccine recipients and placebo recipients at any time point. Assessment of close contacts. Four close contacts, all of whom were contacts of vaccine recipients, were monitored for 12 weeks after vaccination of the person with whom they had contact. None experienced fever or other symptoms that were suggestive of HCMV disease, and none shed HCMV. DISCUSSION The present report describes the first human trial of genetically engineered, live HCMV vaccine candidates. The 4 chimeric HCMV Towne/Toledo vaccine candidates were well tolerated, and no serious adverse events were associated with vaccination. As a group, the investigational vaccines caused more local reactions than did placebo; these reactions, however, were generally mild and of brief duration. Although a somewhat higher proportion of investigational vaccine recipients experienced systemic reactions, compared with placebo recipients, the difference was not significant. This trial was not, however, powered to discern differences in the safety or tolerability of the 4 different investigational vaccines. The Towne/Toledo chimeric vaccine candidates were developed with the goal of producing an HCMV vaccine that retains the excellent safety profile of Towne but induces more-robust immune responses through incorporating sequences from the unattenuated Toledo strain of HCMV. An important study objective, therefore, was to determine whether the chimeric vaccine candidates remain attenuated relative to the Toledo strain, despite the presence of Toledo sequences in their genomes. A direct comparison was not possible because of safety concerns arising from the known potential of the Toledo strain to cause disease in both HCMV-seronegative and -seropositive study volunteers [21, 22]. However, by comparing the laboratory safety data derived from the current study with data derived from earlier Toledo challenge studies, it was possible to assess the pathogenicity of the chimeric vaccine candidates relative to that of the Toledo strain. It has previously been shown that healthy HCMV-seropositive persons who were inoculated with 1000 pfu of the Toledo strain uniformly had laboratory abnormalities suggestive of HCMV infection namely, elevated transaminase levels, atypical lymphocytosis, and/or positive HCMV culture results [22]. Therefore, an HCMV-seropositive person who is given a nonattenuated vaccine (a vaccine that behaves like the Toledo strain) would have a 155% chance of developing an elevated AST level or atypical lymphocytosis or of having a positive HCMV culture result. Accordingly, there would be a 198% chance of observing 1 such event in a group of 5 vaccine 1356 JID 2006:193 (15 May) Heineman et al.

8 Table 5. Immunogenicity of the Towne/Toledo chimeric vaccines. Postvaccination period Study group, laboratory finding Prevaccination period 4 Weeks 8 Weeks a 24 Weeks 52 Weeks Placebo recipients Total IgG, mu/ml, GMT (95% CI) ( ) ( ) ( ) ND ND Neutralizing antibodies, 1/log 2 antibody titer (95% CI) Against Toledo HCMV strain 6.15 ( ) ND 6.17 ( ) ND ND Against Towne HCMV strain 3.68 ( ) ND 3.67 ( ) ND ND Responder cell frequency, % (range) CD ( ) ND 0.76 ( ) 0.19 ( ) 0.34 ( ) CD ( ) ND 0.50 ( ) 0.13 ( ) 0.07 ( ) IFN-g response, median pg/ml (95% CI) 695 ( ) ND 365 ( ) 230 (3 6626) 78 (6 445) Proliferation, median stimulation index (95% CI) 27 ( ) ND 12 (68 272) 40 ( ) 7 ( ) Chimera 1 4 recipients Total IgG, mu/ml, GMT (95% CI) ( ) ( ) ( ) ND ND Neutralizing antibodies, 1/log 2 antibody titer (95% CI) Against Toledo HCMV strain 6.74 ( ) ND 6.90 ( ) ND ND Against Towne HCMV strain 4.22 ( ) ND 4.13 ( ) ND ND Responder cell frequency, % (range) CD ( ) ND 0.38 ( ) 0.33 ( ) 0.23 ( ) CD ( ) ND 0.11 ( ) 0.12 ( ) 0.12 ( ) IFN-g response, median pg/ml (95% CI) 476 ( ) ND 344 (66 397) 591 ( ) 145 (46 334) Proliferation, median stimulation index (95% CI) 12 (7 25) ND 21 (9.6 30) 16 (9.3 24) 15 (9.6 29) a NOTE. CI, confidence interval; GMT, geometric mean titer; HCMV, human cytomegalovirus; IFN, interferon; ND, not determined. At this visit, there were 4 study participants in the placebo group.

9 Table 6. Comparison of laboratory findings between recipients of the Towne/Toledo chimeric candidate vaccines and historical control subjects who received challenge with the Toledo strain of human cytomegalovirus (HCMV) [22]. Vaccine received Laboratory abnormality a Virus detected b Chimera candidate 1 0/5 0/4 c 2 1/5 d 0/5 3 0/5 0/5 4 1/5 e 0/5 All 2/20 0/19 HCMV Toledo f 5/5 2/5 NOTE Data are no. of study participants with the finding/no. of study participants evaluated. a An elevated transaminase level or atypical lymphocytosis. b Detected by routine culture, shell vial assay, or polymerase chain reaction (PCR) in the present study; detected by routine culture only in the study of HCMV Toledo vaccine [22]. c Results were not evaluable for 1 subject (although the results of shell vial assay were positive, the results of PCR and routine culture were negative; therefore, the source of the virus [wild-type vs. vaccine] could not be determined). d Elevated transaminase levels occurred in a volunteer who had elevated transaminase levels before vaccination. e Atypical lymphocytosis (9% atypical lymphocytes in the participant s blood sample at week 8 after vaccination only). f Data are from [22]. recipients. In the current study, however, only 2 of 20 persons who were inoculated with the investigational vaccine candidates had comparable laboratory abnormalities (table 6). One of these persons, a vaccine recipient in group 2 who had elevated transaminase levels after vaccination, had similarly elevated values before vaccination. Thus, it is unlikely that this abnormality was due to vaccine virus, particularly because this volunteer had no other evidence of HCMV infection. It seems more plausible that the other abnormality (atypical lymphocytosis noted 8 weeks after vaccination in a participant in group 4) was related to vaccination. However, of the 5 persons evaluated in the earlier study [22], the 4 who developed atypical lymphocytosis after Toledo challenge all developed this abnormality earlier (during weeks 5 7), and 3 of these 4 individuals had other clinical or laboratory evidence of HCMV infection that was not present in the volunteer in the present study. These data strongly support the conclusion that the Towne/Toledo chimeric vaccine candidates are attenuated relative to the Toledo strain. An important consideration in evaluating live attenuated vaccines is their potential to be transmitted by vaccinated individuals to susceptible individuals with whom they have contact. This issue was addressed in the present study in 2 ways. First, exhaustive efforts were made to detect vaccine virus in the blood, urine, or saliva samples of vaccine recipients. These efforts included the use of techniques (i.e., shell vial culture and PCR, in addition to routine culture) that were more sensitive than the techniques that had been employed in earlier studies [21, 22]. Despite these efforts, vaccine virus was not detected in any recipients of the Towne/Toledo chimeras. By contrast, in the earlier challenge study, HCMV was isolated, by routine culture, from 2 of 5 naturally HCMV-seropositive persons and from 4 of 7 Towne vaccine recipients after inoculation with the Toledo strain [21, 22]. These data suggest that, like the Towne vaccine, the Towne/ Toledo chimeric vaccine candidates do not establish systemic infection, and they further support the conclusion that they are attenuated relative to the Toledo vaccine. Second, no close contacts of vaccine recipients exhibited any symptoms or virological findings suggestive of HCMV infection. Together, these data suggest that the strains in the Towne/Toledo chimeric vaccine candidates are unlikely to be transmitted by vaccinated HCMVseropositive individuals. Numerous differences in the genomic sequences of the HCMV Towne and Toledo strains have been identified [24, 34]. Although it seems likely that some of these differences are responsible for the attenuation of the Towne strain, the precise genetic determinants of HCMV virulence have not been defined. All of the chimeric vaccine candidates tested in the current study appear to be attenuated relative to the Toledo strain, despite the fact that each contains different regions of the Towne genome. Thus, multiple gene products encoded in widely separated regions of its genome are likely to contribute to the attenuation of the Towne strain. Unfortunately, the lack of an animal model for HCMV disease makes it difficult to define the specific genetic determinants of HCMV virulence. The Toledo UL/b region was incorporated into each vaccine candidate and, by virtue of its orientation, provided a means to distinguish the vaccine viruses from wild-type isolates. It was also hoped that inclusion of this region would increase the overall immunogenicity of the vaccine candidates, because it encodes targets of humoral immune responses [24, 34, 35]. UL/ b -encoded proteins also influence HCMV cellular tropism and host responses to infection [26, 36, 37 40]. These effects may enhance the immunogenicity of the vaccine candidates, but they also have the potential to increase their virulence. This concern notwithstanding, the vaccine viruses evaluated in the present study remained attenuated relative to the Toledo strain, despite the presence of the UL/b region. In addition to congenital HCMV disease, it has been proposed that HCMV infection may promote the development of atherosclerosis [41], malignancy [42], and immunosenescence [43]. These possibilities raise concerns that live HCMV vaccines may engender various long-term health risks. Conversely, however, an effective HCMV vaccine may confer to recipients benefits beyond those associated with the prevention of fetal infection. Surprisingly, vaccination with the Towne/Toledo chimeras failed to boost humoral or cellular immune responses to HCMV JID 2006:193 (15 May) Heineman et al.

10 These results contrast with results of vaccination with live attenuated varicella-zoster virus, which, when administered to elderly varicella-zoster virus positive adults, results in increased varicella-zoster virus specific cellular immunity [44]. Because reinfection with new strains of HCMV can occur in HCMV-seropositive individuals, and because these strains may be passed to the fetus with attendant neurologic sequelae [45], strategies to augment HCMV immune responses in seropositive individuals may enhance protection against congenital transmission. An important limitation of the present study was that only seropositive persons were enrolled. This was done to maximize the safety of the participants in this first trial of these live vaccine candidates. In addition, small group sizes were used to limit the number of persons subjected to potential risk from the vaccine candidates. Although the group sizes were sufficient to provide an initial assessment of the safety, tolerability, and attenuation of the vaccine candidates, they did not provide sufficient power to distinguish differences between the 4 different Towne/Toledo chimeras. In summary, the present study demonstrated that Towne/ Toledo vaccine candidates are well tolerated in HCMV-seropositive persons and are attenuated relative to the Toledo strain. On the basis of these results, future studies are warranted to ascertain the safety and immunogenicity of these vaccine candidates in HCMV-seronegative persons, to determine whether significant differences exist between the 4 Towne/Toledo chimeras, and to establish the optimal dosages of these vaccine candidates. References 1. Pass RF. Cytomegalovirus. In: Knipe DM, Howley PM, eds. Philadelphia: Lippincott Williams & Wilkins, 1996: Stratton KR, Durch J, Lawrence RS, eds. Vaccines for the 21st century: a tool for decision making. Washington, DC: National Academy Press, Elek SD, Stern H. Development of a vaccine against mental retardation cased by cytomegalovirus infection in utero. Lancet 1974; 1: Plotkin SA, Furukawa T, Zygraich N, Huygelen C. Candidate cytomegalovirus strain for human vaccination. Infect Immun 1975; 12: Fleisher GR, Starr SE, Friedman HM, Plotkin SA. Vaccination of pediatric nurses with live attenuated cytomegalovirus. Am J Dis Child 1982; 136: Glazer JP, Friedman HM, Grossman RA, et al. Live cytomegalovirus vaccination of renal transplant candidates: a preliminary trial. Ann Intern Med 1979; 91: Just M, Buergin-Wolff A, Emoedi G, Hernandez R. Immunisation trials with live attenuated cytomegalovirus TOWNE 125. Infection 1975;3: Pass RF, Burke RL. Development of cytomegalovirus vaccines: prospects for prevention of congenital CMV infection. Semin Pediatr Infect Dis 2002; 13: Plotkin SA. Vaccination against cytomegalovirus, the changeling demon. Pediatr Infect Dis J 1999; 18: Plotkin SA, Smiley ML, Friedman HM, et al. Towne-vaccine induced prevention of cytomegalovirus disease after renal transplants. Lancet 1984; 1: Adler SP, Hempfling SH, Starr SE, Plotkin SA, Riddell S. Safety and immunogenicity of the Towne strain cytomegalovirus vaccine. Pediatr Infect Dis J 1998; 17: Adler SP, Starr SE, Plotkin SA, et al. Immunity induced by primary human cytomegalovirus infection protects against secondary infection among women of childbearing age. J Infect Dis 1995; 171: Balfour HH Jr. Prevention of cytomegalovirus disease in renal allograft recipients. Scand J Infect Dis Suppl 1991; 80: Brayman KL, Dafoe DC, Smythe WR, et al. Prophylaxis of serious cytomegalovirus infection in renal transplant candidates using live human cytomegalovirus vaccine: interim results of a randomized controlled trial. Arch Surg 1988; 123: Gehrz RC, Christianson WR, Linner KM, Groth KE, Balfour HH Jr. Cytomegalovirus vaccine: specific humoral and cellular immune responses in human volunteers. Arch Intern Med 1980; 140: Gonczol E, Ianacone J, Furlini G, Ho WQ, Plotkin SA. Humoral immune response to cytomegalovirus Towne vaccine strain and to Toledo low-passage strain. J Infect Dis 1989; 159: Marker SC, Simmons RL, Balfour HH Jr. Cytomegalovirus vaccine in renal allograft recipients. Transplant Proc 1981; 13: Plotkin SA, Farquhar J, Horberger E. Clinical trials of immunization with the Towne 125 strain of human cytomegalovirus. J Infect Dis 1976; 134: Plotkin SA, Higgins R, Kurtz JB. Multicenter trial of Towne strain attenuated virus vaccine in seronegative renal transplant recipients. Transplantation 1994; 58: Plotkin SA, Starr SE, Friedman HM, et al. Effect of Towne live virus vaccine on cytomegalovirus disease after renal transplant: a controlled trial. Ann Intern Med 1991; 114: Plotkin SA, Starr SE, Friedman HM, Gonczol E, Weibel RE. Protective effects of Towne cytomegalovirus vaccine against low-passage cytomegalovirus administered as a challenge. J Infect Dis 1989; 159: Quinnan GV Jr, Delery MRA, Rook AH, et al. Comparative virulence and immunogenicity of the Towne strain and a nonattenuated strain of cytomegalovirus. Ann Intern Med 1984; 101: Starr SE, Glazer JP, Friedman HM, Farquhar JD, Plotkin SA. Specific cellular and humoral immunity after immunization with live Towne strain cytomegalovirus vaccine. J Infect Dis 1981; 143: Prichard MN, Penfold ME, Duke GM, Spaete RR, Kemble GW. A review of genetic differences between limited and extensively passaged human cytomegalovirus strains. Rev Med Virol 2001; 11: Kemble G, Duke G, Winter R, Spaete R. Defined large-scale alteration of the human cytomegalovirus genome constructed by cotransfection of overlapping cosmids. J Virol 1996; 70: Penfold ME, Dairaghi DJ, Duke GM, et al. Cytomegalovirus encodes a potent alpha chemokine. Proc Natl Acad Sci USA 1999; 96: Clarke L. Selection, maintenance, and observation of uninoculated monolayer cell cultures. In: Eisenberg HD, ed. Clinical microbiology procedures handbook. 2nd ed. Washington, DC: ASM Press, 2004: Clarke L. Selection, maintenance, and observation of uninoculated monolayer cell cultures. In: Eisenberg HD, ed. Clinical microbiology procedures handbook. 2nd ed. Washington, DC: ASM Press, 2004: Gleaves CA, Smith TF, Shuster EA, Pearson GR. Comparison of standard tube and shell vial cell culture techniques for the detection of cytomegalovirus in clinical specimens. J Clin Microbiol 1985; 21: Wang Z, Mo C, Kemble G, Duke G. Development of an efficient fluorescence-based microneutralization assay using recombinant human cytomegalovirus strains expressing green fluorescent protein. J Virol Methods 2004; 120: Chen SF, Tu WW, Sharp MA, et al. Antiviral CD8 T cells in the control of primary human cytomegalovirus infection in early childhood. J Infect Dis 2004; 189: Jacobson MA, Schrier R, McCune JM, et al. Cytomegalovirus HCMV Towne/Toledo Chimeric Vaccines JID 2006:193 (15 May) 1359

11 (CMV) specific CD4 + T lymphocyte immune function in long-term survivors of AIDS-related CMV end-organ disease who are receiving potent antiretroviral therapy. J Infect Dis 2001; 183: Tu W, Chen S, Sharp M, et al. Persistent and selective deficiency of CD4 + T cell immunity to cytomegalovirus in immunocompetent young children. J Immunol 2004; 172: Cha TA, Tom E, Kemble GW, Duke GM, Mocarski ES, Spaete RR. Human cytomegalovirus clinical isolates carry at least 19 genes not found in laboratory strains. J Virol 1996; 70: Benedict CA, Butrovich KD, Lurain NS, et al. Cutting edge: a novel viral TNF receptor superfamily member in virulent strains of human cytomegalovirus. J Immunol 1999; 162: Hahn G, Revello MG, Patrone M, et al. Human cytomegalovirus UL genes are indispensable for virus growth in endothelial cells and virus transfer to leukocytes. J Virol 2004; 78: Cheung TC, Humphreys IR, Potter KG, et al. Evolutionarily divergent herpesviruses modulate T cell activation by targeting the herpesvirus entry mediator cosignaling pathway. Proc Natl Acad Sci USA 2005; 102: Gerna, G, Percivalle E, Lilleri D, et al. Dendritic-cell infection by human cytomegalovirus is restricted to strains carrying functional UL genes and mediates efficient viral antigen presentation to CD8 + T cells. J Gen Virol 2005; 86: Tomasec P, Wang EC, Davison AJ, et al. Downregulation of natural killer cell activating ligand CD155 by human cytomegalovirus UL141. Nat Immunol 2005; 6: Wang D, Shenk T. Human cytomegalovirus UL131 open reading frame is required for epithelial cell tropism. J Virol 2005; 79: Chen R, Xiong S, Yang Y, Fu W, Wang Y, Ge J. The relationship between human cytomegalovirus infection and atherosclerosis development. Mol Cell Biochem 2003; 249: Sabatier J, Uro-Coste E, Pommepuy I, et al. Detection of human cytomegalovirus genome and gene products in central nervous system tumours. Br J Cancer 2005; 92: Pawelec G, Akbar A, Caruso C, Solana R, Grubeck-Loebenstein B, Wikby A. Human immunosenescence: is it infectious? Immunol Rev 2005; 205: Levin JMJ, Smith JG, Kaufhold RM, et al. Decline in varicella-zoster virus (VZV) specific cell-mediated immunity with increasing age and boosting with a high-hose VZV vaccine. J Infect Dis 2003; 188: Boppana SB, Rivera LB, Fowler KB, Mach M, Britt WJ. Intrauterine transmission of cytomegalovirus to infants of women with preconceptional immunity. N Engl J Med 2001; 344: JID 2006:193 (15 May) Heineman et al.