2008 LANDES BIOSCIENCE. DO NOT DISTRIBUTE.

Transcription

1 [Human Vaccines 5:1, 33-40; January 2009]; 2009 Landes Bioscience Research Paper Phase 2 clinical trial of three formulations of tetravalent live-attenuated dengue vaccine in flavivirus-naïve adults Wellington Sun, 1, * Dennis Cunningham, 2 Steven S. Wasserman, 2 Judith Perry, 3 J. Robert Putnak, 1 Kenneth H. Eckels, 4 David W. Vaughn, 1 Stephen J. Thomas, 1 Niranjan Kanesa-Thasan, 1 Bruce L. Innis 5 and Robert Edelman 2 1 Department of Virus Diseases; Walter Reed Army Institute of Research; Silver Spring, Maryland USA; 2 Department of Medicine and the Center for Vaccine Development; University of Maryland School of Medicine; Baltimore, Maryland USA; 3 University Health Center; University of Maryland; College Park, Maryland, USA; 4 Department of Biologics Research; Walter Reed Army Institute of Research; Silver Spring, Maryland USA; 5 GlaxoSmithKline Biologicals; King of Prussia; Pennsylvania USA Abbreviations: ALT, alanine aminotransferase; ANC, absolute neutrophil count; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CBC, complete blood count; CEF, clinical evaluation form; DENV, dengue virus; GMT, geometric mean titer; LFT, liver function test; PDK, primary dog kidney; PRNT 50, 50% plaque reduction neutralization test; RI, reactogenicity index; SEM, standard error of the mean; TDV, live-attenuated tetravalent dengue vaccine; UMCP, university of maryland, college park Key words: dengue, tetravalent vaccine, Phase 2 clinical trial Sixteen dose formulations of our live-attenuated tetravalent dengue virus vaccines (TDV) were previously evaluated for safety and immunogenicity. Two of the sixteen candidate TDV formulations (Formulations 13 and 14) were selected for further evaluation. A new TDV formulation, Formulation 17, using a higher primary dog kidney (PDK) cell passage Dengue-1 virus (DENV-1) and a lower PDK cell passage DENV-4, was developed to optimize the neutralizing antibody response. All three formulations consist of combinations of 10exp3-5 pfu/dose of the four dengue vaccine virus serotypes. This double-blind, randomized trial in 71 healthy adult subjects evaluated vaccine safety, reactogenicity and immunogenicity. TDV s were given subcutaneously in the deltoid on Day 0 and 180 (6 months). Subjects were seen in clinic on Study Days 0, 10, 28, 180, 190 and 208 and filled out daily symptom diaries for 21 days after each vaccination. Formulation 13 was the most reactogenic, while both Formulations 14 and 17 were similar in reported reactions. Seventy-five percent, 31% and 31% of subjects were viremic on Day 10 after primary vaccination with Formulations 13, 14 and 17 respectively. Viremia was not detected in any subject following the second dose of vaccine. The immunogenicity endpoint was neutralizing antibody titer one month after the second vaccination. Thirty-six percent, 40% and 63% of vaccinated subjects developed tetravalent neutralizing antibodies after two doses of Formulations 13, 14 and 17, respectively. Formulation 17 was selected for further clinical evaluation based on this study. *Correspondence to: Wellington Sun; Dengue Branch; Centers for Disease Control and Prevention; 1324 Calle Canada; San Juan, Puerto Rico USA; Tel.: ; Fax: ; wks4@cdc.gov Submitted: 03/13/08; Revised: 05/18/08; Accepted: 05/27/08 Previously published online as a Human Vaccines E-publication: Introduction Dengue is the leading vector-borne viral disease causing morbidity and mortality in man with over 50 million infections occurring annually. Factors contributing to dengue as an emerging infectious disease include human population growth, poorly planned urbanization, increased travel and expansion of the geographic distribution of the primary mosquito vector, Aedes aegypti. Dengue prevention and control strategies have failed to contain the expansion of this disease. 1 An effective dengue vaccine will be an important part of any integrated dengue prevention program and has become a research and development priority. 1,2 A tetravalent vaccine should simultaneously immunize against all four dengue virus (DENV) serotypes in order to reduce the theoretical risk of dengue hemorrhagic fever from subsequent wild-type DENV infections. Several vaccine strategies are being evaluated, with the live-attenuated approach the most advanced in clinical development. 3-6 Our tetravalent vaccine is based on PDK cell passaged attenuated dengue viruses and its preclinical and clinical evaluations have been previously described In an earlier study, we used a factorial design clinical trial to evaluate sixteen tetravalent vaccine formulations made using all possible permutations of two dose levels of each serotype. 12 Two of the 16 formulations, Formulations 13 and 14 were considered suitable based on the best balance of reactogenicity and immunogenicity. The results from our previous studies also suggested that the DENV-1 PDK20 vaccine virus may be contributing disproportionately to the reactogenicity and that the DENV-4 PDK20 virus was suboptimally immunogenic. 10,12 Therefore, this study was designed (1) to further evaluate Formulations 13 and 14, and (2) to evaluate a new formulation not previously tested in human subjects, designated Formulation 17, made with a further passaged DENV-1 PDK27, previously tested in subjects, 13 and a lower passaged DENV-4 PDK6. Results Seventy-one subjects were enrolled into the study spanning two academic years. Of the 71 subjects 41 (58%) were men, 40 (56%) Human Vaccines 33

2 Table 1 Vaccine potency* by formulation DENV-1 45AZ5 DENV-1 45AZ5 DENV-2 S16803 DENV-3 CH53489 DENV DENV Total dose PDK20 PDK27 PDK50 PDK20 PDK6 PDK x x x x x 10 5 (0.8 x 10 5 ) (0.2 x 10 4 ) (0.4 x 10 4 ) (0.5 x 10 4 ) (0.2 x 10 5 ) x x x x x 10 5 (0.8 x 10 5 ) (0.6 x 10 5 ) (0.3 x 10 3 ) (0.5 x 10 4 ) (0.3 x 10 5 ) x x x x x 10 5 (0.5 x 10 5 ) (0.6 x 10 5 ) (0.4 x 10 4 ) (0.3 x 10 5 ) (0.3 x 10 5 ) *Potency expressed as mean pfu/ml titrated on Vero cells (SEM). were Caucasian, 19 (27%) African-American, 8 (11%) Asian, and 4 (6%) Hispanic. The mean age was 21.1 years (range years). Subject demographic characteristics are shown in Table 2. Volunteer retention. Fifty-two subjects (75%) received two doses of vaccine according to protocol, while nineteen received only one dose of vaccine. Eighteen subjects did not return after the Day 28 visit and one subject, CVD7-18, received only dose 1 but returned for the end-of-study visit on Day 221. The stated reasons for dropping out of the study were: Relocation (4), lack of time (4), recent death of a research subject at another research institution (2), and adverse reactions from first vaccination(3). Five were lost to follow-up after Day 28 despite multiple attempts at contact; their reasons for dropping out could not be ascertained. Clinical safety. Among the first two cohorts of 26 subjects, three subjects, CVD7-02, 07 and 18, reported at least three consecutive days of fever over 38.5 C associated with grade 3 symptoms, with RI s of 29, 43 and 42 respectively after the first dose of vaccine (Table 6). Subject CVD7-07 had peak temperature of 39.8 C, the highest in the study. These findings of three consecutive days of fever were reported to the Medical Monitor. The Medical Monitor unblinded the vaccine assignments of these three subjects, and found that all three had received Formulation 13. A fourth Formulation 13 subject, CVD7-13, also developed dengue-like symptoms lasting 5 days with peak temperature of 39.7 C. The Medical Monitor and the clinical investigators agreed that Formulation 13 would be eliminated from subsequent cohorts. The twelve subjects who had already received Formulation 13 all agreed to receive a second dose. Formulations 14 and 17 met the criteria for safety in all subjects who received them. Local reactions. Arm pain after the first dose was reported by 9 (75%), 28 (78%) and 20 (87%) of subjects vaccinated with Formulations 13, 14 and 17, respectively. One (8%), 4 (11%) and 1 (4%) subjects who received Formulations 13, 14 and 17, respectively, reported the pain as moderate; none reported severe arm pain. All arm pain was associated with only mild redness or swelling alone. After the second vaccination, arm pain was reported by 10 (91%), 19 (76%) and 12 (75%) of subjects. Again, none described the arm pain as severe and only 1, 1 and 2 subjects, respectively, reported it as moderate. No subjects reported any problems with use of the arm after either vaccination. There were no significant differences in reported local reactions among the three formulations, nor between the 1 st and 2 nd vaccinations within formulations. Systemic reactogenicity. Table 3 shows the occurrence of fever and generalized rash during the study. The incidence of prolonged high fever after primary vaccination was significantly higher in Table 2 Study subject demographics Formulation No. of subjects Mean Age (range) 23.9 (18 43) 20.1 (18 41) 20.3 (18 26) Gender Male 6 (50%) 20 (56%) 15(65%) Female 6 (50%) 16 (44%) 8(35%) Formulation13 compared to Formulations 14 or 17 (p < 0.03). The occurrence of any solicited grade 2 or 3 symptoms as reported from the 21-day post-vaccination Symptom Diaries is shown in Figure 1. Headache, mostly grade 2, was the most common symptom reported with all three formulations. With the single exception of headaches in Formulation 17 recipients, all reported symptoms were lower after the second vaccination (Fig. 1B). The most frequently elicited symptoms on the Symptom Diaries described as grade 3 (requiring bedrest) were fever/chills in 4/12 (33%), 5/36 (14%) and 3/23 (13%) for Formulations 13, 14 and 17 recipients, respectively, and headache/retro-orbital pain in 3/12 (25%), 4/36 (11%) and 4/23 (17%). In contrast, no subjects who received the second dose of Formulation 17 described any grade 3 symptoms. There was no difference among the formulations in reactogenicity from the second vaccination as adverse reactions from all three were minimal. The mean RI s were 19 (SEM 4.8), 12 (SEM 1.7) and 10 (SEM 2.1) for Formulations 13, 14 and 17, respectively, after the first dose of vaccine and 4, 4 and 5, respectively, after the second dose of vaccine. However, the differences in RI s between the Formulations were not statistically significant perhaps due to the small number of Formulation 13 subjects. The first dose RI s of the subjects who dropped out of the study did not differ from those that remained (p = 0.49). Clinical laboratory studies. There were no clinically significant abnormalities in hematologic parameters. Two subjects, CVD7-55 and CVD7-57, who received Formulations 14 and 17, respectively, developed transient neutropenia with absolute neutrophil counts (ANC s) of 891 and 832 on Day 15 after their first dose. The lowest platelet count recorded during the study was 124,000/mm Human Vaccines 2009; Vol. 5 Issue 1 Race Caucasian 8 (67%) 20 (56%) 12 (53%) African-American 4 (33%) 8 (22%) 7 (30%) Asian 0 5 (14%) 3 (13%) Hispanic 0 3 (8%) 1 (4%)

3 Table 3 Recorded fever, generalized rash and reactogenicity index by formulation and dose of vaccination Elevated AST or ALT s after vaccination were observed in 13 subjects during the study. Two of these 13 subjects, CVD7-07 and CVD7-40, both with unremarkable LFT s on screening but had abnormal LFT s on Day 0 prior to receiving vaccination. The other 11 developed elevated liver enzymes only after vaccination. An ALT of five times normal on 3 consecutive days was a safety signal according to the protocol. Two subjects, CVD7-7 and CVD7-40, who received Formulation 13 and 14, respectively, developed this level of ALT elevation but both had abnormal ALT at baseline. Subject CVD7-7 developed a peak ALT of 238 IU/ml on Day 15 (baseline 46 IU/ml) Formulation 13 Formulation 14 Formulation 17 Difference First Dose Fever F for 1 day 4/12 (33%) 4/36 (11%) 2/23 (9%) NS Fever F for 3 consecutive days 3/12 (25%) 0/36 (0%) 0/23 (0%) p < 0.03** Generalized rash 6/12 (50%) 7/36 (19%) 5/23 (22%) NS Mean Reactogenicity Index* NS ψ Second Dose Fever F for 1 day 0/11 (0%) 1/25 (4%) 0/16 (0%) NS Fever F for 3 consecutive days 0/11 (0%) 0/25 (0%) 0/16 (0%) NS Generalized rash 2/11 (18%) 0/25 (0%) 2/26 (8%) NS Mean Reactogenicity Index* NS ψ *Reactogenicity Index = Σ Symptom grade x duration. **Fisher s exact test. Ψ Non-parametric median test. Figure 1. (A and B) Solicited symptoms of any grade and recorded fevers from 21-days post-vaccination Symptom Diaries. (A) after first dose, (B) after second dose. Fever/chills (F/C), headache or eye pain (HA/Eye), myalgia or arthralgia (My/Arth), nausea, vomiting or abdominal pain (GI), anorexia (Anorex), photophobia (Phot), conjunctivitis (Conjunc), pruritis (Prur). and CVD7-40 had peak ALT of 266 IU/ml on Day 28 (baseline 97 IU/ml). Of the 13 subjects who developed elevation of AST or ALT after vaccination 3 received Formulation 13, 7 received Formulation 14 and 3 received Formulation 17. The elevations in liver function tests (LFT s) were seen after the first dose in 9 subjects, after the second dose in 2 subjects and after both vaccinations in 2 subjects. Two per-protocol subjects, CVD7-07 and CVD7-13, still had abnormal AST and ALT at the end of the study (AST s of 43 and 47 IU/ml and AST s of 72 and 80 IU/ml). Viremia. Nine of 12 (75%), 11/36 (31%) and 7/23 (31%) of the Formulation 13, 14 and 17 subjects, respectively, were viremic on Day 10 after their first dose of vaccine. In contrast, none of the 52 subjects had measurable viremia on Day 10 after their second vaccination. For each formulation, the mean RI of subjects with detectable viremia was higher than that of non-viremic subjects. However, some subjects were viremic without symptoms, as seen in Formulation 13 subjects CVD7-08 and CVD7-25 and Formulation 14 subject CVD7-05 (Table 6). Overall the mean RI was 19 (SEM 2.2) for viremic compared to 8 (SEM 1.5) for non-viremic subjects (p < 0.001). Serotype identification of viremia was not done for this study. But at Day 180, 6 months after the first vaccination and before the second, the serotype neutralizing antibody response of six of the eight viremic Formulation 13 subjects were to DENV-1 alone (Table 6). The same was not true of the antibody responses of the nine viremic Formulation 14 subjects 6 months after the first vaccination; only two of nine, CVD7-49 and CVD7-58, demonstrated monovalent responses to DENV-2 and DENV-1 respectively. The other seven subjects showed multivalent dengue neutralizing antibody responses (Table 6). Only four of the seven Formulation 17 subjects with viremia had Day 180 specimens for evaluation; two of the four had monovalent response to DENV-4, which may suggest selective DENV-4 replication. Immunogenicity. Serologic response to first dose. The first dose of Formulation 13 elicited 100% DENV-1 seroconversion rate with a disproportionately high GMT, concurrent with the lowest seroconversion to DENV-2 (17%). The titers to DENV-1 were at least ten-fold higher than for any other serotype in 7 of the 12 Formulation 13 subjects. Formulation 13 and 14 differed in concentrations of DENV-2 and DENV-3. After one dose of Formulation 14 the seroconversion rate and GMT were higher to DENV-2 (p < 0.007) and lower to DENV-3 (Table 4). Though the concentrations of DENV-1 and DENV-4 were the same for Formulations Human Vaccines 35

4 Table 4 Presence of neutralizing antibody by PRNT days after TDV vaccinations 13 and 14 the antibody response was ten-fold lower for both these serotypes in Formulation 14. The DENV-1 chosen in Formulation 17 was a more passaged, more attenuated strain than the DENV-1 in Formulations 13 and 14. Accordingly the DENV-1 seroconversion rate from Formulation 17 was significantly lower (p < 0.001) than that of Formulation 13 or 14. Similarly the DENV-1 GMT was seven-fold lower than Formulation 13 (Table 4 and Fig. 2). Other than the 100% seroconversion rate to DENV-1 with Formulation 13, the highest seroconversion rate to any serotype in any formulation was 67% after a single dose of vaccination. The tetravalent seroconversion rates after the first dose were 8%, 17% and 4% for Formulations 13, 14 and 17 respectively, which were not significantly different (Table 5). All Formulation 13 subjects developed at least bivalent antibody responses whereas 6/36 (17%) and 4/23 (17%) of Formulations 14 and 17 did not develop any neutralizing antibody response after the first dose of vaccination (Table 6). The relative ranking of serotype seroconversion rates were DENV-1, 4, 3, 2 for Formulation 13, DENV-1, 2, 3, 4 for Formulation 14 and DENV-4, 2, 3, 1 for Formulation 17. The relative ranking of serotype GMT s followed the seroconversion rates. Serologic response to second dose. The second dose of Formulation 13 elicited a different antibody response than Formulations 14 or 17. Two doses of Formulation 13 increased seroconversion to DENV-2 from 17% to 73% (p < 0.02) but not to the other three serotypes. Similarly, there was a GMT booster response of neutralizing antibody to DENV-2 alone. In contrast, there was a nearly four-fold decline in the DENV-1 GMT after revaccination. Revaccination with Formulation 14 increased the seroconversion rates for DENV-1 and DENV-3, with corresponding booster responses seen in the GMT s. Revaccination with Formulation 17 also increased seroconversion rates and boosted GMT s to DENV-1, 2 and 3 (Table 4). For Formulations 14 and 17 the GMT s to all four serotypes were increased by the second vaccination (Figs. 2 and 3). Formulation 17 elicited stronger antibody response to DENV-4 (94% seroconversion, GMT 145) than the other two formulations, but its DENV-1 response was weaker than Formulations 13 and 14. The tetravalent response rate after two doses of vaccine were 36%, 40% and 63% for Formulations 13, 14 and 17 respectively. The immunogenicity endpoint for the study did not differ between Formulations 13 and 14. Four of six first-dose non-responders (Subjects CVD7-09, 12, 23 and 66) developed tetravalent antibody responses after their Formulation 13 (n = 12) Formulation 14 (n = 36) Formulation 17 (n = 23) Difference ψ First Dose DENV-1 100% (638)* 64% (37) 17% (2) p < DENV-2 17% (3) 67% (24) 65% (17) p < DENV-3 67% (14) 44% (6) 30% (4) NS DENV-4 67% (17) 22% (3) 65% (56) p < Formulation 13 (n = 11) Formulation 14 (n = 25) Formulation 17 (n = 16) Second Dose DENV-1 100% (160) 96% (177) 69% (23) p < 0.02 DENV-2 73% (29) 84% (49) 100% (92) NS DENV-3 45% (12) 76% (28) 81% (28) NS DENV-4 55% (13) 48% (8) 94% (145) p < 0.01 *% subjects with PRNT 50 1:10; number in parenthesis represents the reciprocal GMT, calculated bases on all subjects, including those without antibody titers (<1:5) who were assigned a reciprocal titer of 1. ψ Fisher s exact test of proportions with neutralizing antibody. Figure 2. Twenty-eight days post-first dose PRNT 50 titers and geometric mean titer(longer bar) with 95% confidence interval on log 10 scale by serotype and formulation. Figure 3. Twenty-eight days post-second dose PRNT 50 titers and geometric mean titer(longer bar) with 95% confidence interval on log 10 scale by serotype and formulation. second vaccination. Overall, for the study immunologic endpoint, Formulation 17 elicited the highest seroconversion rates and GMT s for all serotypes except DENV-1. Discussion This study expanded our experience with PDK cell passaged tetravalent live-attenuated dengue virus vaccines in flavivirus-naïve adult human subjects. Previous studies have suggested that the 36 Human Vaccines 2009; Vol. 5 Issue 1

5 Table 5 Valency* of neutralizing antibody response by formulation Ab valency Formulation 13 (n = 12) Formulation 14 (n = 36) Formulation 17 (n = 23) First Dose No response 0 17% 17% Monovalent 0 25% 17% Bivalent 58% 19% 35% Trivalent 33% 22% 26% Tetravalent 8% 17% 4% Ab valency Formulation 13 (n = 11) Formulation 14 (n = 25) Formulation 17 (n = 16) Second Dose No response *% subjects with PRNT 50 1:10. Monovalent 27% 8% 6% Bivalent 18% 20% 6% Trivalent 18% 32% 25% Tetravalent 36% 40% 63% dose of each serotype component and the ratio of monovalent vaccine virus titers are important determinants of reactogenicity and immunogenicity for the tetravalent combinations. 12,14 The interactions among the four viruses in an outbred human population are complex; clinical and immunologic outcomes are not always predictable based on the experience with monovalent components. 14,15 For example, one serotype may dominate the immune response when given in a combination. 15,16 Our study was an expanded evaluation of two selected candidate formulations from a previous factorial-design trial. Two earlier trials suggested the DENV-1 PDK20 and DENV-4 PDK20 to be under and over attenuated, respectively. 10,12 Formulation 17 was designed to bring more balance to the antibody responses. Indeed this formulation induced a balanced tetravalent neutralizing antibody and cellmediated immunity response in cynomolgus macaques. 9 This study was further step in a series of dose-ranging studies in human subjects to arrive at an optimal formulation for our vaccine candidates. Our trial s 25% drop-out rate was high. Various reasons were stated for the drop-outs after one vaccination but anecdotally a well publicized death of a research subject in another study at a different local institution was probably more contributory than stated. Side effects from the vaccines accounted for very few of the drop-outs according to subject reports. In fact, the drop-out rate was lowest in the Formulation 13 group. Formulation 13 was eliminated from further evaluation because it caused dengue fever-like illness in three subjects. One subject reported grade 3 headaches and myalgia/arthralgias for four consecutive days. The differences between Formulations 13 and 14 were simply the lower dose of DENV-2 and higher dose of DENV-3 in Formulation 13. The higher systemic reactogenicity corroborated our previous finding that the higher dose of the DENV-2 component is associated with lower reactogenicity. 12 Grade 3 symptoms described with Formulations 14 and 17 were brief, usually lasting one day, consistent with the observation that the vaccine viruses are attenuated. There were no safety signals that preclude further evaluation of these two formulations. As consistent with the attenuation of the vaccine viruses there was no vaccine-related thrombocytopenia in any subject at any time during the study. The elevation of liver enzymes in 13 subjects after vaccination were asymptomatic but may be related to the vaccine. However, the two subjects with the highest elevations both had abnormal transaminases at baseline, before vaccination. Subject CVD7-40 had baseline AST and ALT of 263 IU/ml and 97 IU/ml respectively, which was suggestive of alcoholic hepatitis. Underlying liver disease may predispose vaccinees to develop hepatotoxicity. Elevated transaminases is also associated with more severe dengue; liver toxicity should continue to be monitored in future trials. We elected to measure viremia only on Day 10 after vaccination based on limited previous experience with viremia of the monovalent vaccines. In our study, viremias occurred only after the first dose and were associated with increased reactogenicity. We did not determine serotype of the viremias but the monovalent DENV-1 neutralzing antibody responses 6 months after dose one suggested that DENV-1 replication predominated in Formulation 13 subjects. This is consistent with our previous finding that the DENV-1 PDK20 virus is the most reactogenic virus in the formulation. 10 Increased DENV-4 replication in Formulation 17 subjects is indicated by a more robust antibody response and viremia. This was what we had hoped to achieve with the substitution of DENV PDK6 for PDK20, which was a poorly immunogenic virus. 10 The fact that there were no viremias after the second dose suggests several possibilities; either viremia kinetics differed for the second dose and did not occur on Day 10; or the first dose induced sufficient serotypespecific memory antibody or cellular immunity to prevent viremia at detectable levels for any of the vaccine viruses; or lastly that crossreactive immunity induced by some serotype virus replication was still present at the time of the second vaccination. In any case, while there was no detectable viremia at Day 10 after the second dose, there was still sufficient viral replication to boost the antibody response in most subjects. This study further supports our and others experience with live-attenuated TDVs that a second dose is safe, elicits minimal clinical adverse reactions and can boost the antibody response. 10,12,14,17 Due to the passage level and dose differences between Formulation 17 and the other two formulations we cannot be certain which difference(s) contributed most to the observed differences in DENV-1 and DENV-4 immunogenicity. The neutralizing antibody response to a second Formulation 13 vaccination was qualitatively different from that of the other Human Vaccines 37

6 Table 6 Individual subject demographics, reactogenicity, viremia and neutralizing antibody response by formulation two formulations; there was a slight decrease in the DENV-1, DENV-3 and DENV-4 GMT s or percentages with antibody or both. This may be due to the immuno-dominant effect of the DENV-1 component in Formulation 13 resulting in prevention of not only DENV-1, but also DENV-3 and DENV-4 replication from the second vaccination. In contrast, the second dose of Formulations 14 and 17 augmented the antibody responses to all four serotypes, indicating that the interference among the four serotypes can be overcome with revaccination. The immunogenicity endpoint for this study was the presence of neutralizing antibody after two doses of vaccine. The tetravalent response rates after two doses of vaccination were 36%, 40% and 63% for Formulations 13, 14 and 17 respectively. The serotype GMT s for Formulation 17 were well balanced. Formulation 17 was subsequently evaluated in 7 flavivirus-naïve Thai children and found to be well-tolerated with 100% tetravalent neutralizing antibody response rate after two doses. 18 For these reasons we selected Formulation 17 as our leading vaccine candidate for further evaluation in a broader age range of subjects in endemic areas. We based our selection on considerations of clinical safety and the neutralizing antibody response. In primate models our live-attenuated tetravalent vaccines afforded protection against live virus challenges. 8,9 The level of vaccine-induced antibody required for protection in humans is unknown. It remains for clinical efficacy trials to show whether this vaccine will protect against dengue disease. Subjects and Methods Vaccines. The origins of the four monovalent vaccine candidates have been described. 7 Tetravalent Formulations 13 and 14 consist of the same four vaccine viruses with differing concentrations of each of the four dengue virus serotypes, as shown in Table 1. Formulation 14 differed from Formulation 13 in that the concentration of DENV-2 was 32-fold higher and DENV-3 was 32-fold lower. Formulation 17 was made with DENV-1 45AZ5 PDK-27 instead of PDK-20, and with DENV PDK-6, instead of PDK-20 (Table 1). Vaccine administration and handling. The three tetravalent formulations were prepared on the day of vaccination from lyophilized monovalent vaccines reconstituted with sterile water for injection. Each monovalent component was diluted as required to the correct potency and then mixed with other monovalent components to achieve the correct tetravalent formulation. For each tetravalent formulation, 1-ml doses were drawn up into individual 1-ml tuberculin syringes and kept on ice before administration. Volunteers were inoculated subcutaneously over the deltoid. Unused vaccines were transported on ice back to the laboratory to determine residual viral titer. The final injected concentrations of tetravalent vaccine formulations in pfu/ml are shown in Table 1. Subjects. Seventy-one healthy male and female paid volunteers between the age of years were recruited at the University of Maryland, College Park (UMCP). Subjects were recruited by posted and published advertisements followed by verbal presentation of the study protocol. All subjects were screened for eligibility by a health 38 Human Vaccines 2009; Vol. 5 Issue 1

7 Table 6 Individual subject demographics, reactogenicity, viremia and neutralizing antibody response by formulation (continued) *Subject received only the first dose of vaccination because of drop-out. Subject CVD7-18 received only the first dose but returned for D178 follow-up blood collection. RI(1) Reactogenicity Index (Dose 1), Vir(1) viremia (Dose 1). C (Caucasian), Af (African-American), As (Asian), H(Hispanic), M (Male), F (Female), + = positive; neg = below limit of detection: 0.5 pfu/ml for DENV-1,2,4, 5 pfu/ml for DENV-3. A reciprocal PRNT 50 titer of <1:5 was assigned a value of 1. history including prior flavivirus vaccinations, directed physical examination and laboratory studies. Screening laboratory tests included a complete blood count (CBC) with differential, glucose, blood urea nitrogen (BUN), creatinine, alanine aminotransferase (ALT); aspartate aminotransferase (AST) and urinalysis. Subjects with clinically significant laboratory abnormalities at screening were excluded. Females had to have a negative urine pregnancy test within 48 hours of both vaccinations. All subjects were tested for hepatitis B surface antigen and antibodies to hepatitis C and HIV and excluded from participation if positive. Similarly, all subjects who tested positive prior to vaccination for antibodies against dengue viruses 1 4, St. Louis encephalitis virus, yellow fever virus and Japanese encephalitis virus by hemagglutination inhibition assays were excluded. Before enrollment all volunteers had to score 70% or better on a written examination designed to test their understanding of the clinical trial. The study protocol was approved by the Institutional Review Boards of the University of Maryland, Baltimore and the UMCP, as well as the Office of the Surgeon General, U.S. Army. Written, informed consent was obtained from all subjects. This study was conducted in Study design. This was a randomized double-blinded, outpatient Phase 1/2 trial to evaluate the clinical safety, reactogenicity and immunogenicity of three TDV formulations given by a two-dose schedule at time 0 and 180 days. There was no placebo group. The study was conducted in sequential cohorts of 12, 14, 15 and 30 subjects based on rates of subject accrual and limitations due to the students academic calendar. In the first and second cohorts the subjects were randomized to Formulations 13, 14 or 17 at a ratio of 5:5:2. Because of the relatively frequent occurrence of dengue feverlike symptoms associated with the first vaccination in the first two cohorts, the assignment code was broken by a study investigator not involved in subject clinical evaluations. The study continued with the two subsequent cohorts evaluating only Formulations 14 and 17. In the 4 th cohort the subjects were randomized to receive Formulations 14 or 17 at a 1:1 ratio. Safety and reactogenicity. The injection site was evaluated for pain, redness, swelling and effect on use of the arm. Clinical safety was defined as absence of all of the following: (1) oral temperature 38.5 C (101.3 F) for 4 determinations over 24 hours, a Human Vaccines 39

8 maximum daily oral temperature of 38.5 C on three successive days, or temperature exceeding 40 C (104.0 F) on any individual determination, (2) thrombocytopenia (fewer than 90,000 platelets/ mm 3 ) or neutropenia (absolute neutrophil count < 1000/mm 3 ) and (3) ALT of more than five times the upper limit of normal (normal range 0 45 U/L) on three or more successive days which is otherwise unexplained. Each subject filled out a daily Symptom Diary and recorded their oral temperature using a digital thermometer each morning and evening for 21 days after each vaccination. Each subject was evaluated at scheduled visits and solicited for symptoms which were recorded on a Clinical Evaluation Form (CEF) by an UMCP study physician on the days of vaccination (Study Days 0 and 180), 10 days following vaccination (Study Days 10 and 190), 15 days following vaccination (Study Days 15 and 195) and 28 days following vaccination (Study Days 28 and 208). For the diary as well as the CEF, each symptom was graded on a scale of 0 3 defined as: 0-none; 1-mild ( did not affect normal activity, no medication required); 2-moderate (required medication or change in activity); 3-severe (required bed rest, and/or unrelieved by medication). Fever was defined as an oral temperature >38 C. A Reactogenicity Index (RI), basically a summation of products of main symptom scores and their duration, was calculated for each subject. 10 In addition, study nurses checked on each subject 7 days after each vaccination by telephone. CBC with differential, AST and ALT were performed by an accredited commercial clinical laboratory on subjects blood obtained on each of the two vaccination days and on post-vaccination Days 15 and 28. Viremia. Sera obtained 10 days after each vaccination were tested for dengue virus by a delayed plaque method adapted from Yuill. 19 In this 14-day amplification assay the limit of detection of viremia was 0.5 pfu/ml, <0.5 pfu/ml, 5 pfu/ml and 0.5 pfu/ml for the four serotypes respectively. Immunogenicity. The study immunogenicity endpoint was the development of neutralizing antibodies against the four serotypes 28 days after the second vaccination as measured by the 50% plaque reduction neutralization test (PRNT 50 ). The PRNT 50 was adapted from Russell using a 6-well format with Vero cells. 20 All PRNT 50 s were performed in triplicate. The dose viruses for the PRNT assays were the vaccine parental strains with the exception of DENV-4, which used DENV-4 TVP-360. Sera were first tested at 1:5 dilution in a screening PRNT 50. If there was 50% plaque reduction at 1:5 then the specimen underwent end-titration starting at 1:10 dilution. The minimum level of neutralizing antibody that correlates with human protection against dengue is not known. By convention, a PRNT 50 titer of 1:10 was chosen as threshold for a positive test. Seroconversion was defined as a change from a negative to a positive PRNT 50. Geometric mean titer (GMT) is calculated based on all subjects; a titer of <1:5 was assigned a value of 1 in the calculation. A booster response was defined as at least a four-fold rise in the PRNT 50 titer. Analysis of data. The reactogenicity and immunogenicity data for each subject were pooled by formulation in the analysis. Relationships between categorical variables were examined using Fisher s exact test. Reactogenicity Indices of the three formulations were compared using the non-parametric median test. The Mann-Whitney test was used to compare RI s between two groups of subjects. Immunogenicity of the formulations was described by seroconversion rates and GMT with 95% confidence intervals based on PRNT 50. Two-sided null hypotheses were evaluated at 5% throughout unless otherwise stated. Data analysis was performed using GraphPad Prism v5. Acknowledgements The authors thank the participation of the voluntary study subjects at the University of Maryland, College Park and the skilled research coordination provided by Ms. Joanna Becker and Ms. Jeanie Harper-Jones. The authors also thank Dr. Carol O. Tacket of the University of Maryland School of Medicine for serving as the Medical Monitor of the study. The study was supported by the U.S. Army Medical Research Acquisition Activity, Ft. Detrick, Frederick, MD USA. References 1. Farrar J, Focks D, Gubler D, Barrera R, Guzman MG, Simmons C, Kalayanarooj S, Lum L, McCall PJ, Lloyd L, Horstick O, Dayal-Drager R, Nathan MB, Kroegar A. On behalf of the WHO/TDR Dengue Scientific Working Group. Editorial: Towards a global dengue research agenda. Trop Med Intl Health 2007; 12: World Health Assembly. Dengue fever and dengue hemorrhagic fever prevention and control (WHA55.17). (accessed 27 October 2007). 3. Edelman R. Dengue vaccines approach the finish line. Clin Inf Dis 2007; 45: Whitehead SS, Blaney JE, Durbin AP and Murphy BR. Prospects for a dengue vaccine. Nat Rev Microbiol 2007; 5: Hombach J. Vaccines against dengue: a review of current candidate vaccines at advanced development stages. Rev Panam Salud Publica 2007; 21: Guy B and Almond JW. Towards a dengue vaccine: Progress to date and remaining challenges. Comp Immunol Microbiol Infect Dis 2007; [Epub ahead of print]. 7. Eckels KH, Dubois DR, Putnak R, Vaughn DW, Innis BL, Henchal EA, Hoke CH Jr. Modification of dengue virus strains by passage in primary dog kidney cells: preparation of candidate vaccines and immunization of monkeys. Am J Trop Med Hyg 2003; 69: Sun W, Nisalak A, Gettayacamin M, Eckels KH, Putnak JR, Vaughn DW, Innis BL, Thomas SJ, Endy TP. Protection of Rhesus Monkeys against Dengue Virus Challenge following Tetravalent Live-attenuated Dengue Virus Vaccination. J Inf Dis 2006; 193: Koraka P, Benton S, van Amerongen G, Stittelaar KJ, Osterhaus ADME. Efficacy of a live attenuated tetravalent candidate dengue vaccine in naïve and previously infected cynomolgus macaques. Vaccine 2007; 25: Sun W, Edelman R, Kanesa-thasan N, Eckels KH, Putnak JR, King AD, Houng HS, Tang D, Scherer JM, Hoke CH Jr, Innis BL. Vaccination of human volunteers with monovalent and tetravalent live-attenuated dengue vaccine candidates. Am J Trop Med Hyg 2003; 69: Gwinn W, Sun W, Innis BL, Caudill J, King AD. Serotype-specific TH1 responses in recipients of two doses of candidate live-attenuated dengue virus vaccines. Am J Trop Med Hyg 2003; 69: Edelman RE, Wasserman SS, Bodison SA, Putnak RA, Eckels KH, Tang D, Kanesa-thasan N, Vaughn DW, Innis BL, Sun W. Phase 1 trial of 16 formulations of a tetravalent liveattenuated dengue vaccine. Am J Trop Med Hyg 2003; 69: Edelman R, Tacket CO, Wasserman SS, Vaughn DW, Eckels KH, Dubois DR, Summers PL, Hoke CH Jr. A live attenuated dengue-1 vaccine candidate (45AZ5) passaged in primary dog kidney cell culture is attenuated and immunogenic for humans. J Inf Dis 1994; 170: Sabchareon A, Lang J, Chanthavanich P, Yoksan S, Forrat R, Attanath P, Sirivichayakul C, Pengsaa K, Pojjaroen-anant C, Chokejindachai W, Jagsudee A, Saluzzo JF, Bhamarapravati N. Safety and immunogenicity of tetravalent live-attenuated dengue vaccines in Thai adult volunteers: Role of serotype concentration, ratio and multiple doses. Am J Trop Med Hyg 2002; 66: Kitchener S, Nissen M, Nasveld P, Forrat R, Yoksan S, Lang J, Saluzzo JF. Immunogenicity and safety of two live-attenuated tetravalent dengue vaccine formulations in healthy Australian adults. Vaccine 2006; 24: Kanesa-thasan N, SunW, Kim-Ahn G, Van Albert S, Putnak JR, King A, Raengsakulsrach B, Christ-Schmidt H, Gilson K, Zahradnik JM, Vaughn DW, Innis BL, Saluzzo JF, Hoke CH Jr. Safety and immunogenicity of attenuated dengue virus vaccines (Aventis Pasteur) in human volunteers. Vaccine 2001; 19: Sabchareon A, Lang J, Chanthavanich P, Yoksan S, Forrat R, Attanath P, Sirivichayakul C, Pengsaa K, Pojjaroen-Anant C, Chambonneau L, Saluzzo JF, Bhamarapravati N. Safety and immunogenicity of a three dose regimen of two tetravalent live-attenuated dengue vaccine in five- to twelve-year-old Thai children. Pediatr Infect Dis J 2004; 23: Simasathien S, Thomas SJ, Watanaveeradej V, Nisalak A, Barberousse C, Innis BL, Sun W, Putnak JR, Eckels KH, Hutagalung Y, Gibbons RV, Zhang C, De La Barrera R, Jarman RG, Chawachalasai W, Mammen MP Jr. Safety and immunogenicity of a tetravalent liveattenuated dengue vaccine in flavivirus naive children. Am J Trop Med Hyg. 2008;78: Yuill TM, Sukhavachana P, Nisalak A, Russell PK. Dengue-virus recovery by direct and delayed plaques in LLC-MK 2 cells. Am J Trop Med Hyg 1968; 17: Russell PK, Nisalak A, Sukhavachana P, Vivona S. A plaque reduction test for dengue virus neutralizing antibodies. J Immunol 1967; 99: Human Vaccines 2009; Vol. 5 Issue 1