Effectiveness and Efficiency of Imperfect

[Human Vaccines 3:6, 231-238, November/December 2007]; ©2007 Landes Bioscience

Research Paper

Effectiveness and Efficiency of Imperfect Therapeutic HSV-2 Vaccines Elissa J. Schwartz1,2,† Erin N. Bodine1,3,‡ Sally Blower1,3,*

Abstract

†Current

address: Department of Mathematics and School of Biological Sciences; Washington State University; Pullman, Washington 99164 USA ‡Current address: Department of Mathematics; University of Tennessee; Knoville,

Tennessee 37996 USA *Correspondence to: Sally Blower; 1100 Glendon Avenue; Penthouse 2; Los Angeles, California 90024 USA; Tel.: 310.794.8911; Fax: 310.794.8653; Email: [email protected]

Key words

SC

IEN

Herpes Simplex Virus type 2 (HSV-2), post-exposure vaccine, therapeutic vaccine, mathematical model, herpes epidemic, effectiveness, efficiency

CE

Previously published online as a Human Vaccines E-publication: http://www.landesbioscience.com/journals/vaccines/article/4529

.D

ON

Original manuscript submitted: 02/28/07 Manuscript accepted: 06/3/07

RIB

of Biostatistics; 3AIDS Institute, David Geffen School of Medicine; University of California, Los Angeles; Los Angeles, California USA

IST

2Department

OT D

1Disease Modeling Group; Semel Institute for Neuroscience and Human Behavior;

UT E

.

Background: Efforts are currently underway to develop therapeutic vaccines for Herpes Simplex Virus type 2 (HSV-2). Methods: We use a mathematical model to predict the potential public health impact of imperfect, therapeutic HSV-2 vaccines. We evaluate vaccine effectiveness and efficiency for the general population in the United States where HSV-2 prevalence is currently 22%. We assume that therapeutic vaccines will produce two therapeutic benefits in vaccinated infected-individuals: (1) the rate of viral reactivation will decrease (hence infected-individuals will experience fewer viral shedding episodes), and (2) the average length of the viral shedding episodes will be shortened. In addition, we assume that therapeutic vaccines will benefit uninfected individuals by reducing viral shedding in (and hence transmission from) vaccinated infected-individuals. Results: Our predictions show that therapeutic vaccines could substantially reduce HSV-2 epidemics by reducing new infections by 77% and preventing 0.84 new infections for each vaccinated individual. These vaccines could prevent 212,600 (median; IQR, 156,064–288,558) new infections after only one year. We show that increased effectiveness and efficiency are more strongly correlated with a vaccine-induced reduction in transmission probability than with either of the two therapeutic benefits that accrue directly to the infected individuals (specifically, the reduction in episode length and number of episodes). Conclusions: We suggest that current vaccine development efforts target mechanisms that reduce viral shedding (thereby reducing transmission) thus providing both a beneficial therapeutic and a beneficial epidemic-level impact. Our results also demonstrate that therapeutic vaccines would be substantially more useful than prophylactic vaccines for epidemic control.

Acknowledgements

Abbreviations HSV-1, Herpes Simplex Virus type 1; HSV-2, Herpes Simplex Virus type 2; HIV-1, Human Immunodeficiency Virus type 1; IQR, interquartile range; gB2, glycoprotein B2; gD2, glycoprotein D2; LHS, Latin hypercube sampling; PRCC, partial rank correlation coefficient; PDF, probability distribution function

©

20

07

LA

ND

ES

BIO

S.B. has received research support from, and consulted for, GlaxoSmithKline. This work was supported by NIH/NIAID R01 AI041935 (to S.B.) and GlaxoSmithKline.

www.landesbioscience.com

Introduction Herpes Simplex Virus type 2 (HSV-2) has infected 22% of the general population in the US1 and up to 60% of the general population in many developing countries.2-4 While there is great interest in developing prophylactic HSV-2 vaccines, of the two recent clinical trials5,6 only one showed a high vaccine efficacy level, and only in women who were seronegative for both HSV-1 and HSV-2.6 Therapeutic HSV-2 vaccines are being developed with the aim of preventing viral shedding recurrences (or at least minimizing their frequency, duration and severity), and reducing the probability of transmission.7-9 Several of the first therapeutic HSV-2 vaccine clinical trials (1920’s–1930’s) initially showed great therapeutic effects; however, these trials lacked placebo-controlled groups, and when later tested against control groups showed little to no difference from the placebo group.10-12 More recently, Lupidon H and Lupidon G trials, which administered whole heat-killed virus vaccines from HSV-1 and HSV-2, respectively, reported a significant reduction in the length and number of recurrences in vaccinated subjects with genital or facial herpes disease. However, results from this study are inconclusive as many treatments were being Human Vaccines

231

RIB

UT E

.

Imperfect Therapeutic HSV-2 Vaccines

However, current research has produced vaccines that only partially fulfill these objectives.8,19 Therefore we developed a new mathematical model of the HSV-2 epidemic that accounts for the potentially imperfect therapeutic effects of the vaccines. We use the model to determine the potential public health impact therapeutic HSV-2 vaccines could have on reducing HSV-2 incidence if given to HSV-2 infected individuals. We evaluate two measures: (1) vaccine effectiveness (the cumulative percent reduction in new infections) and (2) vaccine efficiency (the cumulative number of new infections prevented per 100 individuals vaccinated). We evaluate effectiveness and efficiency for the general population in the US, where HSV-2 prevalence is currently 22%.1 In addition, we identify which vaccine characteristic would have the greatest impact on increasing (or decreasing) vaccine effectiveness and efficiency.

©

20

07

LA

ND

ES

BIO

SC

IEN

CE

.D

ON

given during the study, and no clear method of how the outcome was evaluated was presented with the results.12-14 The Skinner vaccine, an inactivated subunit vaccine containing mixed HSV-2 glycoproteins, reduced severity of recurrences in men and lowered the number of recurrences in women.15 However, the vaccine showed no consistent pattern of efficacy or immunogenicity.12 The TA-HSV-2 vaccine, a disabled infectious single cycle (DISC) virus vaccine, uses an HSV-2 strain with glycoprotein H deleted and completes only a single cycle of replication, never producing infectious progeny. However, development of this vaccine was stopped when phase II trials showed no protection against recurrences.7,10,12 The most promising recent therapeutic vaccine trials have focused on development of recombinant glycoprotein vaccines and live-attenuated virus vaccines. The best recombinant glycoprotein vaccine results were seen in the gD2-alum vaccine trial which administered glycoprotein D2 with adjuvant alum. Vaccine recipients had 36% fewer culture-proven recurrences than the control group, and over the one year period of the trial had a median of two fewer total recurrences.12,16 In an attempt to improve upon this vaccine, gD2 was combined with gB2 and administered with the stronger adjuvant MF59. When evaluated, the gD2gB2-MF59 vaccine did not significantly reduce the monthly rate of recurrences or the duration of virus shedding; however it did significantly reduce symptom duration, new lesion formation and lesion duration.12,17 Nevertheless, the manufacturer stopped the trial due to the lack of reduction in recurrence rate. Promising live-attenuated virus vaccine results were seen in the ICP10DPK phase I/II clinical trials, which administered a deletion mutant virus lacking the PK domain of an essential viral protein. The vaccine completely prevented recurrences in 37.5% and 43.5% of the vaccinated patients in the first18 and second10 trials, respectively. The first trial also showed that the vaccinated patients who did have recurrences experienced them at a lower frequency compared with the placebo group.10,18 This vaccine has yet to undergo safety and efficacy trials, and thus, has not yet been approved for therapeutic use. Ideally, a therapeutic HSV-2 vaccine will completely prevent viral shedding recurrences as well as transmission to a sex partner.

OT D

IST

Figure 1. Flow diagram of the transmission dynamics of the HSV-2 epidemic with therapeutic vaccination. The model specifies the rate of change over time of individuals in five states: unvaccinated susceptibles (X), unvaccinated infected individualswho are shedding virus (and may be symptomatic or asymptomatic) (H), vaccinated infected individuals who are shedding virus (and may be symptomatic or asymptomatic) (Hv), unvaccinated infected individuals who are not shedding virus (Q), and vaccinated infected individuals who are not shedding virus (Qv). Model equations and parameter definitions are detailed in the Methods and the Appendix.

232

Methods Therapeutic HSV-2 vaccine model. We develop a new mathematical model of the HSV-2 epidemic which predicts and quantifies the effect of administering therapeutic vaccines to infected individuals; a flow diagram of our model is shown in Figure 1. As in our previous models,20-22 our new model contains the intrinsic dynamics of HSV-2 infection; thus infected individuals oscillate between a viral shedding stage (which includes both symptomatic and asymptomatic individuals) and a quiescent or latent stage. We assume that individuals can only transmit HSV-2 when they are shedding virus, but this transmission can occur from symptomatic and asymptomatic individuals. Within our modeling framework susceptible individuals enter the sexually active population at rate p and become infected with HSV-2 according to the contact rate c and the time-dependent per capita force of infection l, which is the product of the HSV-2 transmission probability (b for unvaccinated and bv for vaccinated individuals) and the probability of selecting an infectious partner; therefore l(t)=b(H(t)/N(t))+bv(Hv(t)/N(t)), where N(t) represents the population size of the sexually active community at a given time t, and H(t) and Hv(t) respectively represent the infectious unvaccinated and vaccinated population size

Human Vaccines

2007; Vol. 3 Issue 6


Table 1

Parameter values used in uncertainty and sensitivity analyses

Parameter Units Min a Max a Definition of Parameter b

0.1

0.5

Transmission probability per partnership

1/m

yrs

10

20

Average length of time acquiring new sex partners

NE

per yr

1

20

Number of infectious viral shedding episodes (symptomatic or asymptomatic)b

1/q

days

2

5

Average length of viral shedding episodes (symptomatic or asymptomatic)b

1/w

yrs

10

20

Average duration of vaccine-induced immunity

p

0.3

0.9

Proportion of infected group that receives the therapeutic vaccine

e

0.3

1.0

Proportion of vaccinated individuals in whom the vaccine takes

0

0.35

Transmission probability per partnership (in vaccinated infected-individuals)

NEv

per yr

0.1

14

Number of infectious viral shedding episodes (in vaccinated infected-individuals)b

1/qv

days

0.2

3.5

Average length of viral shedding episodes (in vaccinated infected-individuals)b

UT E

.

bv

probability distribution functions (pdfs) were used for all parameters except NE which used a triangular pdf with a peak value of 12 per year and 1/q which used a triangular pdf with a peak value of 3.5 days.bThe reactivation rates (r and r v) are derived from the average length of viral shedding episodes (1/q and 1/qv) and the average number of infectious viral shedding episodes per year (NE and NEv): 1/r = 1/NE–1/q and 1/r v = 1/NEv–1/qv for unvaccinated and vaccinated individuals, respectively.

OT D

IST

We conducted these uncertainty and sensitivity analyses for a moderate HSV-2 epidemic in which seroprevalence is 22%. We also performed the analyses for higher risk populations, in which HSV-2 seroprevalence is 60%. Parameter estimates. Parameter ranges are given in Table 1. HSV-2 infection parameters (b, NE, 1/q) are given ranges consistent with current data on HSV-2 pathogenesis.25-29 We assume that the average duration of therapeutic HSV-2 vaccines (1/w) could range from 10–20 years; this assumption is based upon the durations of current prophylactic vaccines for smallpox, polio and diphtheria.30,31 We vary therapeutic vaccine parameters fairly broadly in order to investigate the impact of a wide range of hypothetical therapeutic vaccines. Therefore, we vary the proportion of infected individuals that are vaccinated (p) from 0.3–0.9, and the proportion of vaccinated individuals in whom the vaccine takes (e) from 0.3–1.0. We investigate the impact of therapeutic vaccines that (1) reduce the probability of transmission by 30–100%, (2) reduce the number of viral shedding episodes per year by 30–90% and (3) reduce the average viral shedding episode length by 30–90%. The average length of time acquiring new sex partners (1/m) was varied from 10–20 years.23 The rate that new susceptibles enter the sexually active population (p) was varied from 10,000–20,000 individuals per year. By using LHS we are able to vary the values of all parameters at once, and thus investigate the potential public health impact of 1,000 different therapeutic vaccines.

©

20

07

LA

ND

ES

BIO

SC

IEN

CE

.D

ON

at a given time t. The average period of acquisition of new sex partners is 1/m. A certain proportion p of the infected individuals are vaccinated; the proportion e of vaccinated individuals in which the vaccine has a therapeutic effect (i.e. the vaccine “takes”) is considered “successfully vaccinated.” However, we assume that vaccine-induced therapeutic effects in the successfully vaccinated will eventually wane at rate w. Unvaccinated and vaccinated individuals will reactivate viral shedding at rates r and rv, respectively, which are calculated from the average length of a viral shedding episode (1/q and 1/qv for unvaccinated and vaccinated individuals, respectively), and the average number of viral shedding episodes per year (NE and NEv for unvaccinated and vaccinated individuals, respectively); specifically 1/r = 1/NE–1/q and 1/rv = 1/NEv–1/qv for unvaccinated and vaccinated individuals, respectively. The rate of reactivation increases with the average length of viral shedding episodes and the average number of episodes per year. We assume the therapeutic vaccine has three effects in successfully vaccinated individuals: the average length of their viral shedding episodes decreases (1/qv < 1/q), they have fewer viral shedding episodes per year (NEv < NE), and they have a lower probability of transmitting infection (bv < b) in comparison with unvaccinated infected individuals. Model equations are given in the Appendix. Uncertainty and sensitivity analyses. Using time-dependent uncertainty analyses, we predict the impact of imperfect therapeutic HSV-2 vaccines given to HSV-2 infected individuals over a 30-year period. To account for uncertainty in parameter values, we use a range of values, and then sample the parameter ranges using Latin hypercube sampling (LHS).20,21,23,24 Table 1 shows parameter ranges used for the uncertainty analysis. Simulations were run 1,000 times (each simulation using a unique parameter set). Effectiveness (cumulative percent reduction in new infections) and efficiency (cumulative number of new infections prevented per individual vaccinated) were calculated each year for each simulation. To identify key parameters in increasing (or decreasing) effectiveness and efficiency, time-dependent sensitivity analyses were carried out by calculating partial rank correlation coefficients (PRCCs), as previously described.20,24 PRCCs were calculated to determine the relative sensitivity of the outcome variables (effectiveness and efficiency) to variation in the input parameters. Parameters with positive PRCC values have the effect of increasing the outcome value as they increase; parameters with negative PRCC values have the effect of decreasing the outcome variable as they decrease.

RIB

aUniform


Results Predicted effectiveness and efficiency of therapeutic vaccines. Assuming parameter ranges given in Table 1, therapeutic HSV-2 vaccines could substantially reduce new HSV-2 infections over 30 years. Our predictions show the effectiveness (cumulative percent reduction in new infections) of a therapeutic vaccine would rise quickly over the first five years of vaccination, and then slowly level off over the next 25 years (Fig. 2A). After five years, 43% (median; IQR, 34–52%) of cumulative new infections would be prevented, and after 30 years, 77% (median; IQR, 70–83%) of cumulative new infections would be prevented (Fig. 2A). Our predictions also show that the efficiency (cumulative new infections prevented per individual vaccinated) of a therapeutic vaccine would rise steadily over time (Fig. 2B). After five years of vaccinating infected individuals, the vaccine would prevent 0.17

Human Vaccines

233

CE

.D

ON

OT D

IST

RIB

UT E

.


Figure 3. Key factors for increasing effectiveness and efficiency. Values of partial rank correlation coefficients (PRCCs) showing the sensitivity of (A) effectiveness, and (B) efficiency to: e, proportion of vaccinated individuals in whom the vaccine takes (grey circles); p, proportion of infected population that receives the vaccine (dotted black line); 1/m, average length of time acquiring new sex partners (dotted gray line); 1/w, average vaccine duration (dashed black line); a, the vaccine-induced reduction in transmission probability in vaccinated individuals (dashed gray line); g, the vaccine-induced reduction in length of a viral shedding episode in vaccinated individuals (solid black line); z, the vaccine-induced reduction in the number of viral shedding episodes per year in vaccinated individuals (solid gray line).

ES

BIO

SC

IEN

Figure 2. Cumulative Infections Prevented. (A) Effectiveness. Use of therapeutic HSV-2 vaccines could reduce cumulative new infections by 77% [median; interquartile range (IQR), 70–83%] after 30 years. (B) Efficiency. Therapeutic HSV-2 vaccines could prevent 0.84 [median; interquartile range (IQR), 0.70–1.05] cumulative new infections for every vaccination administered, after 30 years. Gray lines show median values of 1,000 simulations, boxes show interquartile ranges, and vertical lines show minimum and maximum values.

©

20

07

LA

ND

(median; IQR, 0.14–0.20) cumulative new infections per individual vaccinated (Fig. 2B). After 15 years of vaccination, efficiency would rise to 0.47 (median; IQR, 0.40–0.57) cumulative new infections prevented per individual vaccinated. After 30 years, almost one infection (0.84 median; IQR, 0.70–1.05) would be prevented for each individual vaccinated. Comparable results for effectiveness and efficiency were also found in an analysis of higher risk populations (such as in developing countries, men who have sex with men and those of African descent2-4) where HSV-2 seroprevalence is 60%. Time-dependent sensitivity analysis: predicted effectiveness and efficiency. Using time-dependent sensitivity analysis, we identify the key factors for increasing (and decreasing) the effectiveness and efficiency of the vaccine. The values of the PRCCs for effectiveness (cumulative percent reduction in new infections) and efficiency (cumulative number of new infections prevented per individual vaccinated) over 30 years of vaccination with a therapeutic vaccine are shown in Figure 3A and B respectively. The effectiveness of the vaccine is most sensitive to variation in the take (e) and the proportion of infected individuals vaccinated (p) (Fig. 3A). 234

In contrast, efficiency is most sensitive to the average length of time an individual spends acquiring new sex partners (1/m) (Fig. 3B); as the average sexual life span increases, efficiency decreases. The sensitivity of the vaccine effectiveness and efficiency to the average duration of vaccine-induced therapeutic effects in successfully vaccinated individuals (1/w) changed dramatically over time, having relatively no sensitivity to this parameter when vaccination begins, and having relatively high sensitivity to this parameter after 15 years of vaccination. After about ten years of vaccination, efficiency becomes more sensitive to 1/w than to the proportion vaccinated (p) or the proportion in which the vaccine takes (e). This phenomenon does not occur for vaccine effectiveness. Other important though relatively less influential parameters for both effectiveness and efficiency were the vaccine-induced reduction in the transmission probability (bv/b), the vaccine-induced reduction in episode length (q/qv), and the vaccine-induced reduction in number of episodes per year (NEv/NE ). Variations in the ranges of the remaining parameters were found to have relatively little effect

Human Vaccines



ND

ES

BIO

SC

IEN

CE

.D

ON

OT D

IST

RIB

UT E

.

in increasing (on decreasing) effectiveness and efficiency. Results were similar for high HSV-2 prevalence (60%) epidemics. The sensitivity of effectiveness and efficiency to key vaccine characteristics, shortly after the introduction of therapeutic vaccines, is shown in Figure 4. Figure 4A shows the effect of vaccine coverage (p) and vaccine take (e) on effectiveness after five years of vaccination. Reductions in cumulative new infections of 15–30%, 30–45%, 45–60% or 60–75% are shown (Fig. 4A). The greatest reduction in cumulative new infections after five years occurs when both coverage levels and vaccine take are maximized; for example, a therapeutic vaccine with a vaccine take of 85% administered to 85% of sexually-active HSV-2 infected individuals would have an effectiveness of 60–75%. However, a therapeutic vaccine with a vaccine take of 40% administered to only 40% of sexually-active HSV-2 infected individuals would only have an effectiveness of 15–30%. Figure 4B shows the effect of vaccine duration (1/w) and the vaccine-induced reduction in transmission probability (bv/b) on efficiency after 30 years. Prevention of 0.65–0.80, 0.80–0.95, and 0.95–1.10 cumulative infections per indvidual vaccinated are shown (Fig. 4B). These results quantify the projected effect of vaccine duration and vaccine-induced reduction in transmission probability on efficiency (Fig. 4B). For example, a therapeutic vaccine that reduces the probability of transmission by 90% in vaccinated infected-individuals (i.e., bv/b = 0.1) and that lasts (on average) for 20 years would prevent 0.95–1.1 cumulative new infections for each infected individual vaccinated. However, even a therapeutic vaccine that wanes (on average) in 10–12 years and reduces the probability of transmission by only 30% in vaccinated infected individuals (i.e., bv/b = 0.7) would prevent 0.65–0.80 cumulative new infections for each infected individual vaccinated. In Figure 4C the effect of the vaccine-induced reduction in the length of viral shedding episodes (g) and the vaccine-induced reduction in the number of viral shedding episodes per year (z) on efficiency after 30 years is shown. Prevention of 0.70–0.80, 0.80–0.90, and 0.90–1.00 cumulative infections per individual vaccinated are shown in Figure 4C. These results show that vaccine efficiency is greatest the more the vaccine reduces the length of viral shedding episodes or the number of episodes per year, i.e., when the relative length (g) or number (z) of viral shedding episodes are minimal.

©

20

07

LA

Figure 4. Sensitivity of effectiveness and efficiency to key factors. (A) The effect of vaccination coverage level (p) and vaccine “take” (e) on vaccine effectiveness. Results show the reduction in cumulative new infections (15–30%, dark gray circles; 30–45%, light gray triangles; 45–60%, black diamonds; 60–75%, open gray circles), five years after the introduction of a therapeutic vaccine. Median values for other parameters were used. (B) The effect of vaccine duration and vaccine-induced reduction in transmission probability in vaccinated individuals on vaccine efficiency. Results show the cumulative number of infections prevented per vaccinated individual (0.65–0.80, dark gray squares; 0.80–0.95, black circles; 0.95–1.10, light gray triangles), 30 years after the introduction of a therapeutic vaccine. Median values for other parameters were used. (C) The effect of vaccine-induced reduction in length of viral shedding episodes (γ) and vaccine-induced reduction in number of viral shedding episodes per year (z) on vaccine efficiency. Results show the cumulative number of infections prevented per vaccinated individual (0.70–0.80, black diamonds; 0.80–0.90, light gray circles; 0.90–1.00, dark gray squares), 30 years after the introduction of a therapeutic vaccine. Median values for other parameters were used.


Discussion Our results have shown that imperfect therapeutic vaccines could be extremely effective, reducing cumulative new infections by 77%, and preventing 0.84 cumulative new infections for each vaccinated individual after 30 years of vaccination. Effectiveness (cumulative percent reduction in new infections) will increase rapidly in the first years after the vaccine is introduced, and begin to level off after 15 years of vaccination. Efficiency (cumulative number of new infections prevented per individual vaccinated) will be fairly low initially, but over time, will steadily increase, making vaccination programs become more “efficient” with time. Our results have important implications for therapeutic HSV-2 vaccine development and vaccination strategies. We have shown that vaccine “take” and coverage are key characteristics for maximum effectiveness and efficiency, especially in the short term, and so it follows that high initial vaccination coverage levels are essential, as are adjuvants that induce a high vaccine take. After the first five years of vaccination, vaccine duration will be an important factor determining vaccine

Human Vaccines

235


OT D

IST

RIB

UT E

.

and shown to be high.46 Thus, HIV-1 infections can also be prevented if HIV-infected individuals are able to use future therapeutic HSV-2 vaccines. We have recently evaluated the potential public health impact of imperfect prophylactic HSV-2 vaccines.36 We found that their effectiveness and efficiency are likely to be modest, as compared to imperfect therapeutic HSV-2 vaccines. Previously,36 we determined that moderately effective prophylactic HSV-2 vaccines are likely to reduce incidence only by 20%, (as compared to our predictions of incidence reductions of 77% for therapeutic vaccines) and that prophylactic HSV-2 vaccines would prevent only 0.11 infections per individual vaccinated (as compared to our current predictions of preventing 0.84 infections per individual vaccinated). Thus, we suggest that therapeutic vaccines would be substantially more useful than prophylactic vaccines for controlling HSV-2 epidemics because a high proportion of the general population is already infected with HSV-2; 22% of the general population in the US is HSV-2 infected. Thus, this epidemic is not restricted to behavior core-groups, and therapeutic vaccines that target large numbers of infected individuals who transmit HSV-2 are essential to controlling the epidemic. Obviously, it is also necessary to develop prophylactic vaccines and target them to younger individuals who are initiating sexual activity; however, even if these vaccines protect these individuals from infection, they should not be expected to have a major effect on reducing the incidence of HSV-2. Targeting virological core groups for daily antiviral therapy could also prove to be an extremely novel and effective control strategy for reducing HSV-2 epidemics.23 However, therapeutic vaccines that lower the risk of HSV-2 transmission would be an improvement upon antiviral therapy, considering single administration of a vaccine, potentially followed by booster vaccinations.26 Thus prevention efforts may benefit from therapeutic HSV-2 vaccines given to virological core groups of individuals who reactivate and shed infectious virus frequently; this potential should be explored. Though it is likely that a combination of prophylactic and therapeutic vaccines and antivirals will be necessary to fully control HSV-2 epidemics, therapeutic HSV-2 vaccines alone will provide a great degree of both epidemic and individual level benefit.

©

20

07

LA

ND

ES

BIO

SC

IEN

CE

.D

ON

efficiency; booster vaccinations that increase duration may be needed to sustain high efficiency over time. The strong and consistent negative correlation of efficiency with the length of time individuals acquire new sex partners indicates the importance of behavioral intervention strategies that shorten the period of acquiring new partners. Finally, our results show that the epidemic-level measures of effectiveness and efficiency were more sensitive to reduction in transmission probability than to therapeutic benefits that accrue directly to the infected individual, specifically, reduction in the number and length of viral shedding26 episodes. Thus, to control the HSV-2 epidemic, current vaccine development efforts should target mechanisms that reduce asymptomatic and symptomatic viral shedding (thereby reducing transmission) in an effort to provide maximum individual-level therapeutic benefit and epidemic-level infection-prevention benefit. Such vaccines, which interfere with virus production, would consequently reduce the number of viral particles shed by cells per unit time. There is a precedent for this kind of therapeutic intervention; for example antiviral therapy reduces viral shedding. There are some potential limitations of our model that warrant discussion. Our model excludes aspects of HSV-2 epidemiology such as differences due to gender and age. In this study we do not include stratification by subgroups; instead our goal is to model novel therapeutic HSV-2 vaccines in order to predict the reduction in new infections for the entire population. Therefore our conclusions apply to the combined population, but they do not address subpopulations by age or gender. Differences in gender may not be substantial in the context of therapeutic vaccines. The gender difference found in phase III trials for a prophylactic vaccine involved protection against clinical disease and not infection: a subset of HSV-seronegative women showed a reduction in disease and a non-statistically significant trend toward reduction in infection.6 The current study is concerned with reduction in infection. Our study addressed differences in risk behavior and ethnicity by considering high risk populations in which HSV-2 incidence is 60%. We found equivalent results for such populations as for the general US population (where HSV-2 prevalence is 22%). Therefore the same vaccines, used in both moderate and high risk populations (such as in developing countries, men who have sex with men and those of African descent), will be expected to have approximately the same impact, independent of HSV-2 prevalence. The potential public health impact of imperfect vaccines has been previously evaluated for HIV-1 vaccines, TB vaccines, and pre-exposure HSV-2 vaccines.30,32-37 These studies have shown that even imperfect vaccines could have substantial public health impact, even with moderate vaccine efficacy and coverage levels;30,32-37 it is even possible that imperfect vaccines could eradicate HIV-1 epidemics.34,35 Our current results on therapeutic HSV-2 vaccines are in accord with these previous results for imperfect vaccines for other diseases. Our results can be used to generate predictions of the potential impact that imperfect therapeutic vaccines could have on controlling the HSV-2 epidemic. Considering the current rate of new infections in the United States,38 imperfect therapeutic HSV-2 vaccines could prevent 212,600 (median; IQR, 156,064–288,558) new infections after only one year. This reduction translates to 24 (median; IQR, 18–33) infections prevented per hour every day, after the first year of vaccination alone. HSV-2 is also a risk factor for both infection and transmission of HIV-1;39-45 the impact of HSV-2 epidemics increasing the incidence of HIV has recently been quantified 236

Human Vaccines



References

Appendix

1. Fleming DT, McQuillan GM, Johnson RE, Nahmias AJ, Aral SO, Lee FK, St Louis ME. Herpes simplex virus type 2 in the United States, 1976 to 1994. N Engl J Med 1997; 337:1105‑11. 2. Nahmias AJ, Lee FK, Beckman‑Nahmias S. Sero‑epidemiological and ‑sociological patterns of herpes simplex virus infection in the world. Scand J Infect Dis Suppl 1990; 69:19‑36. 3. Oberle MW, Rosero‑Bixby L, Lee FK, Sanchez‑Braverman M, Nahmias AJ, Guinan ME. Herpes simplex virus type 2 antibodies: High prevalence in monogamous women in Costa Rica. Am J Trop Med Hyg 1989; 41:224‑9. 4. Whitley RJ, Roizman B. Herpes simplex virus infections. �� Lancet 2001; 357:1513‑8. 5. Corey L, Langenberg AG, Ashley R, Sekulovich RE, Izu AE, Douglas JM Jr, Handsfield HH, Warren T, Marr L, Tyring S, DiCarlo R, Adimora AA, Leone P, Dekker CL, Burke RL, Leong WP, Straus SE. �� Recombinant glycoprotein vaccine for the prevention of genital HSV‑2 infection. Two randomized controlled trials: Chiron HSV Vaccine Study Group. JAMA 1999; 282:331‑40. 6. Stanberry LR, Spruance SL, Cunningham AL, Bernstein DI, Mindel A, Sacks S, Tyring S, Aoki FY, Slaoui M, Denis M, Vandepapeliere P, Dubin G; GlaxoSmithKline Herpes Vaccine Efficacy Study Group. Glycoprotein‑D‑adjuvant vaccine to prevent genital herpes. N Engl J Med 2002; 347:1652‑61. 7. Koelle DM. Vaccines for herpes simplex virus infections. �� Curr Opin Investig Drugs 2006; 7:136‑41. 8. Hosken NA. �� Development of a therapeutic vaccine for HSV‑2. Vaccine 2005; 23:2395‑8. 9. De Bruyn G, Vargas‑Cortez M, Warren T, Tyring SK, Fife KH, Lalezari J, Brady RC, Shahmanesh M, Kinghorn G, Beutner KR, Patel R, Drehobl MA, Horner P, Kurtz TO, McDermott S, Wald A, Corey L. Vaccine 2006; 24:914‑20. 10. Aurelian L. Herpes simplex virus type 2 vaccines: New ground for optimism? Clin Diagn Lab Immunol 2004; 11:437‑45. 11. Jones CA, Cunningham AL. Vaccination strategies to prevent genital herpes and neonatal herpes simplex virus (HSV) disease. Herpes 2004; 11:12‑7. 12. Stanberry LR. Clinical trials of prophylactic and therapeutic herpes simplex virus vaccines. Herpes 2004; 11:161‑9. 13. Mastrolorenzo A, Tiradritti L, Salimbeni L, Zuccati G. Multicentre clinical trial with herpes simplex virus vaccine in recurrent herpes infection. Int J STD AIDS 1995; 6:431‑5. 14. Weitgasser H. Controlled clinical study of the herpes antigens LUPIDON H and LUPIDON G. Z Hautkr 1977; 52:625‑8. 15. Skinner GR, Turyk ME, Benson CA, Wilbanks GD, Heseltine P, Galpin J, Kaufman R, Goldberg L, Hartley CE, Buchan A. The efficacy and safety of Skinner herpes simplex vaccine towards modulation of herpes genitalis; report of a prospective double‑blind placebo‑controlled trial. Med Microbiol Immunol (Berl) 1997; 186:31‑6. 16. Straus S, Corey L, Burke R, Savarese B, Barnum G, Krause P, Kost R, Meier J, Sekulovich R, Adair S, Dekker C. Placebo‑controlled trial of vaccination with recombinant glycoprotein D of herpes simplex virus type 2 for immunotherapy of genital herpes. Lancet 1994; 343:1460‑3. 17. Straus SE, Wald A, Kost RG, McKenzie R, Langenberg AG, Hohman P, Lekstrom J, Cox E, Nakamura M, Sekulovich R, Izu A, Dekker C, Corey L. Immunotherapy of recurrent genital herpes with recombinant herpes simplex virus type 2 glycoproteins D and B: Results of a placebo‑controlled vaccine trial. �� J Infect Dis 1997; 176:1129‑34. 18. Casanova G, Cancela R, Alonzo L, Benuto R, Magana Mdel C, Hurley DR, Fishbein E, Lara C, Gonzalez T, Ponce R, Burnett JW, Calton GJ. �� A double‑blind study of the efficacy and safety of the ICP10deltaPK vaccine against recurrent genital HSV‑2 infections. Cutis 2002; 70:235‑9. 19. Rouse BT, Kaistha SD. A tale of 2 alpha‑herpesviruses: Lessons for vaccinologists. Infect Dis 2006; 42:810‑7. 20. Blower SM, Porco TC, Darby G. Predicting and preventing the emergence of antiviral drug resistance in HSV‑2. Nat Med 1998; 4:673‑8. 21. Gershengorn HB, Blower SM. Impact of antivirals and emergence of drug resistance: HSV‑2 epidemic control. AIDS Patient Care STDS 2000; 14:133‑42. 22. Gershengorn HB, Darby G, Blower SM. Predicting the emergence of drug‑resistant HSV‑2: New predictions. BMC Infect Dis 2003; 3:1. 23. Blower S, Wald A, Gershengorn H, Wang F, Corey L. Targeting virological core groups: A new paradigm for controlling herpes simplex virus type 2 epidemics. J Infect Dis 2004; 190:1610‑7. 24. Blower SM, Dowlatabadi H. Sensitivity and uncertainty analysis of complex models of disease transmission: An HIV model, as an example. Int Stat Rev 1994; 2:229‑43. 25. Benedetti J, Corey L, Ashley R. Recurrence rates in genital herpes after symptomatic first‑episode infection. Ann Intern Med 1994; 121:847‑54. 26. Corey L, Wald A, Patel R, Sacks SL, Tyring SK, Warren T, Douglas JM Jr, Paavonen J, Morrow RA, Beutner KR, Stratchounsky LS, Mertz G, Keene ON, Watson HA, Tait D, Vargas-Cortes M; Valacyclovir HSV Transmission Study Group. Once‑daily valacyclovir to reduce the risk of transmission of genital herpes. N Engl J Med 2004; 350:11‑20. 27. Mertz GJ, Benedetti J, Ashley R, Selke SA, Corey L. Risk factors for the sexual transmission of genital herpes. Ann Intern Med 1992; 116:197‑202. 28. Mertz GJ, Coombs RW, Ashley R, Jourden J, Remington M, Winter C, Fahnlander A, Guinan M, Ducey H, Corey L. Transmission of genital herpes in couples with one symptomatic and one asymptomatic partner: A prospective study. J Infect Dis 1988; 157:1169‑77. 29. Wald A, Zeh J, Selke S, Ashley RL, Corey L. Virologic characteristics of subclinical and symptomatic genital herpes infections. N Engl J Med 1995; 333:770‑5.

ND

ES

BIO

SC

IST

OT D

IEN

CE

.D

ON

where X represents the population of susceptible individuals in the sexually active pool, H represents the population of HSV-2-infected individuals (asymptomatic and symptomatic) who are shedding virus and are therefore infectious, and Q represents those who are infected but are not shedding virus and hence are not infectious. Qv represents the population of vaccinated infected individuals who are not shedding virus (and hence are not infectious), and Hv represents vaccinated infected-individuals who are shedding virus and hence are infectious. Paramaters bv, qv, rv and NEv have the following relations: bv = ab, 1/qv = g (1/q), NEv = zNE, and 1/rv = 1/NEv – 1/qv where a, g, and z are independent parameters between 0 and 1. The size of the sexually active population is N = X + H + Hv + Q + Qv. The rate at which new susceptibles enter the sexually active population is p = mN. Pre-vaccine equilibrium values for X(t), H(t), and Q(t), denoted by X*, H*, and Q*, respectively, are given by

RIB

UT E

.

The new mathematical model is described by the following equations:

07

LA

The initial HSV-2 prevalence

©

20

The contact rate, c, is derived from the pre-vaccine equilibrium conditions and initial HSV-2 population prevalence (22% or 60%) and is therefore


Human Vaccines

237

UT E RIB IST OT D

©

20

07

LA

ND

ES

BIO

SC

IEN

CE

.D

ON

30. Blower S, Schwartz EJ, Mills J. Forecasting the future of HIV epidemics: The impact of antiretroviral therapies & imperfect vaccines. AIDS Rev 2003; 5:113‑25. 31. In: Plotkin S, Mortimer Jr E, eds. Vaccines. 2nd ed. Philadelphia: W.B. Saunders Co., 1994. 32. Blower SM, Koelle K, Kirschner DE, Mills J. Live attenuated HIV vaccines: Predicting the tradeoff between efficacy and safety. Proc Natl Acad Sci USA 2001; 98:3618‑23. 33. Blower SM, Koelle K, Mills J. Health policy modeling: Epidemic control, HIV vaccines and risky behavior. In: Eds K, Brookmeyer, eds. Quant Eval HIV Prevent Progr: Yale University Press 2002; 260‑89. 34. Blower SM, McLean AR. Prophylactic vaccines, risk behavior change, and the probability of eradicating HIV in San Francisco. Science 1994; 265:1451‑4. 35. McLean AR, Blower SM. Imperfect vaccines and herd immunity to HIV. Proc Biol Sci 1993; 253:9‑13. 36. Schwartz EJ, Blower S. Predicting the potential individual‑level and population‑level impact of imperfect HSV‑2 vaccines. J Infect Dis 2005; 191:1734‑46. 37. Ziv E, Daley CL, Blower S. Potential public health impact of new tuberculosis vaccines. Emerg Infect Dis 2004; 10:1529‑35. 38. Armstrong GL, Schillinger J, Markowitz L, Nahmias AJ, Johnson RE, McQuillan GM, St Louis ME. Incidence of herpes simplex virus type 2 infection in the United States. Am J Epidemiol 2001; 153:912‑20. 39. Nagot N, Ouedraogo A, Foulongne V, Konate I, Weiss HA, Vergne L, Defer MC, Djagbare D, Sanon A, Andonaba JB, Becquart P, Segondy M, Vallo R, Sawadogo A, Van de Perre P, Mayaud P. ANRS 1285 study group: Reduction of HIV‑1 RNA levels with therapy to suppress herpes simplex virus. N Engl J Med 2007; 356:790‑9. 40. Freeman EE, Weiss HA, Glynn JR, Cross PL, Whitworth JA, Hayes RJ. Herpes simplex virus 2 infection increases HIV acquisition in men and women: Systematic review and meta‑analysis of longitudinal studies. AIDS 2006; 20:73‑83. 41. DeJesus E, Wald A, Warren T, Schacker TW, Trottier S, Shahmanesh M, Hill JL, Brennan CA; Valacyclovir International HSV Study Group. Valacyclovir for the suppression of recurrent genital herpes in human immunodeficiency virus‑infected subjects. J Infect Dis 2003; 188:1009‑16. 42. Renzi C, Douglas JM Jr, Foster M, Critchlow CW, Ashley-Morrow R, Buchbinder SP, Koblin BA, McKirnan DJ, Mayer KH, Celum CL. Herpes simplex virus type 2 infection as a risk factor for human immunodeficiency virus acquisition in men who have sex with men. J Infect Dis 2003; 187:19‑25. 43. Reynolds SJ, Risbud AR, Shepherd ME, Zenilman JM, Brookmeyer RS, Paranjape RS, Divekar AD, Gangakhedkar RR, Ghate MV, Bollinger RC, Mehendale SM. Recent herpes simplex virus type 2 infection and the risk of human immunodeficiency virus type 1 acquisition in India. J Infect Dis 2003; 187:1513‑21. 44. Corey L, Wald A, Celum CL, Quinn TC. The effects of herpes simplex virus‑2 on HIV‑1 acquisition and transmission: A review of two overlapping epidemics. J Acquir Immune Defic Syndr 2004; 35:435‑45. 45. Wald A, Corey L. How does herpes simplex virus type 2 influence human immunodeficiency virus infection and pathogenesis? J Infect Dis 2003; 187:1509‑12. 46. Blower S, Ma L. Calculating the contribution of herpes simplex virus type 2 epidemics to increasing HIV incidence: Treatment implications. Clin Infect Dis 2004; 39:240‑7.

.


238

Human Vaccines
