key: cord-0758517-uy0zjk7z authors: Gallagher, Molly E.; Sieben, Andrew J.; Nelson, Kristin N.; Kraay, Alicia N. M.; Lopman, Ben; Handel, Andreas; Koelle, Katia title: Considering indirect benefits is critical when evaluating SARS-CoV-2 vaccine candidates date: 2020-08-11 journal: medRxiv DOI: 10.1101/2020.08.07.20170456 sha: e1564510ec3f98605e6ca641bb5ea239e8f16ced doc_id: 758517 cord_uid: uy0zjk7z Significant progress has already been made in development and testing of SARS-CoV-2 vaccines, and Phase III clinical trials have begun for 6 novel vaccine candidates to date. These Phase III trials seek to demonstrate direct benefits of a vaccine on vaccine recipients. However, vaccination is also known to bring about indirect benefits to a population through the reduction of virus circulation. The indirect effects of SARS-CoV-2 vaccination can play a key role in reducing case counts and COVID-19 deaths. To illustrate this point, we show through simulation that a vaccine with strong indirect effects has the potential to reduce SARS-CoV-2 circulation and COVID-19 deaths to a greater extent than an alternative vaccine with stronger direct effects but weaker indirect effects. Protection via indirect effects may be of particular importance in the context of this virus, because elderly individuals are at an elevated risk of death but are also less likely to be directly protected by vaccination due to immune senescence. We therefore encourage ongoing data collection and model development aimed at evaluating the indirect effects of forthcoming SARS-CoV-2 vaccines. have shorter durations of infection or lower viral loads that reduce their infectiousness. 48 Vaccination campaigns can significantly reduce the number of infections and deaths in subpop-49 ulations that remain unvaccinated, even when vaccination coverage is quite low [12] -a crucial 50 consideration, given that extensive vaccine coverage will be a formidable challenge. Vaccine doses, 51 and the public health infrastucture needed to administer them, will almost certainly be limited 52 in supply relative to demand, and given the current politically and emotionally charged climate, 53 vaccine refusal could pose an additional barrier [13, 14] . The vast majority of the global population 54 remains susceptible to the virus and we are likely well below the herd immunity threshold, despite two hypothetical vaccines. Vaccine 1 reduces the risk of clinical infection in vaccinated individuals to 30% of the original risk (a direct effect), and reduces the infectiousness of vaccinated individuals 66 to 70% of the original value (an indirect effect). Vaccine 2 reduces the risk of clinical infection 67 to 70% of the original and the infectiousness to 30% of the original. To evaluate the direct and 68 total (direct + indirect) effects of these vaccines, we can use a compartmental susceptible-exposed- CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 11, 2020. . https://doi.org/10.1101/2020.08.07.20170456 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 11, 2020. . https://doi.org/10.1101/2020.08.07.20170456 doi: medRxiv preprint Model Equations 125 We model the dynamics of SARS-CoV-2 using a set of deterministic ordinary differential equations, 126 with susceptible individuals S, exposed individuals E, infected individuals I, and recovered indi- 6 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 11, 2020. . https://doi.org/10.1101/2020.08.07.20170456 doi: medRxiv preprint Total population size for the simulations was fixed at N =100k and we assume 20% of the population 141 is already in the 'recovered' class R. The initial size of the exposed class E was set to 200 individuals, 142 and values for the I c and I sc classes were calculated under a fast dynamics assumption: The initial size of the susceptible class S = 0.8(1 − f )N and the initial susceptible vaccinated 7 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 11, 2020. Vaccination and herd immunity to infectious diseases Direct and indirect 192 effects of influenza vaccination Vaccine hesitancy, vaccine refusal and 194 the anti-vaccine movement: influence, impact and implications. Expert Review of Vaccines AP-NORC poll: Half of Americans would get a 197 COVID-19 vaccine The effect of control strategies to 203 reduce social mixing on outcomes of the COVID-19 epidemic in Wuhan, China: a modelling 204 study. The Lancet Public Health Optimizing influenza vaccine distribution Clinical relevance of age-related immune dysfunction The reproductive 213 number of COVID-19 is higher compared to SARS coronavirus Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus 217 (SARS-CoV-2) Case-fatality rate and characteristics 219 of patients dying in relation to COVID-19 in italy Jamie Grif-222 fin Estimates of the severity