Person:

Hitchings, Matt

Background: The 2014–6 West African Ebola epidemic highlights the need for rigorous, rapid clinical trial methods for vaccines. A challenge for trial design is making sample size calculations based on incidence within the trial, total vaccine effect, and intracluster correlation, when these parameters are uncertain in the presence of indirect effects of vaccination. Methods and findings We present a stochastic, compartmental model for a ring vaccination trial. After identification of an index case, a ring of contacts is recruited and either vaccinated immediately or after 21 days. The primary outcome of the trial is total vaccine effect, counting cases only from a pre-specified window in which the immediate arm is assumed to be fully protected and the delayed arm is not protected. Simulation results are used to calculate necessary sample size and estimated vaccine effect. Under baseline assumptions about vaccine properties, monthly incidence in unvaccinated rings and trial design, a standard sample-size calculation neglecting dynamic effects estimated that 7,100 participants would be needed to achieve 80% power to detect a difference in attack rate between arms, while incorporating dynamic considerations in the model increased the estimate to 8,900. This approach replaces assumptions about parameters at the ring level with assumptions about disease dynamics and vaccine characteristics at the individual level, so within this framework we were able to describe the sensitivity of the trial power and estimated effect to various parameters. We found that both of these quantities are sensitive to properties of the vaccine, to setting-specific parameters over which investigators have little control, and to parameters that are determined by the study design. Conclusions: Incorporating simulation into the trial design process can improve robustness of sample size calculations. For this specific trial design, vaccine effectiveness depends on properties of the ring vaccination design and on the measurement window, as well as the epidemiologic setting.

Power considerations for trials evaluating vaccines against infectious diseases are complicated by indirect protective effects of vaccination. While cluster-randomized controlled trials (cRCTs) are less statistically efficient than individually randomized controlled trials (iRCTs), a cRCT’s ability to measure direct and indirect vaccine effects may mitigate the loss of efficiency due to clustering. Within cRCTs, the number and size of clusters affects 3 determinants of power: the effect size being measured, disease incidence, and intracluster correlation. We simulated trials conducted in a collection of small communities to assess how indirect protection and clustering affected the power of cRCTs and iRCTs during an emerging epidemic. Across diverse parameters, we found that within the same trial population, cRCTs were never more powerful than iRCTs, although the difference can be small. We also identified 2 effects that attenuated the loss of cRCT power traditionally associated with increased cluster size. First, if enrollment of fewer, larger clusters was performed to achieve higher vaccine coverage within vaccinated communities, this increased the effect to be measured and, consequently, power. Second, the greater rate of imported transmission in larger communities may increase the attack rate and similarly mitigate loss of power relative to a trial in many, smaller communities.

Background/aims Network structure and individuals' level of exposure to a pathogen can impact results from efficacy evaluation studies of interventions against infectious diseases. Heterogeneity in infection risk can cause randomized groups to increasingly differ as a trial progresses and as more high-risk individuals become infected (described in prior work as the "frailty" phenomenon). Here, we show the impact this phenomenon can have on an individually randomized trial of a leaky vaccine in which all participants are exchangeable a priori. Methods We model a vaccine trial by generating a network of individuals grouped into communities, which are connected to a larger main population. We then simulate an epidemic, deterministically and with time-varying transmission rates in the main population and stochastically in the communities. The disease natural history follows a susceptible-exposed-infectious-recovered model. Simulation results are used to estimate vaccine efficacy [Formula: see text] with a Cox proportional hazards model. Results We find downward bias in [Formula: see text] associated with low connectivity between communities in the study population and high force of infection, even when all participants in the trial are exchangeable at the time of randomization. This phenomenon arises because the stochastic dynamics in such a setting randomly lead to community-level variation in the force of infection. Stratifying a Cox model by community alleviates this bias with no loss of power. Conclusion Understanding and accounting for the impact of heterogeneous hazard rates can allow for more accurate estimates of [Formula: see text] in epidemic settings.