Controlling long-term SARS-CoV-2 infections can slow viral evolution and reduce the risk of treatment failure

Abstract

The rapid emergence and expansion of novel SARS-CoV-2 variants threatens our ability to achieve herd immunity for COVID-19. These novel SARS-CoV-2 variants often harbor multiple point mutations, conferring one or more evolutionarily advantageous traits, such as increased transmissibility, immune evasion and longer infection duration. In a number of cases, variant emergence has been linked to long-term infections in individuals who were either immunocompromised or treated with convalescent plasma. In this paper, we used a stochastic evolutionary modeling framework to explore the emergence of fitter variants of SARS-CoV-2 during long-term infections. We found that increased viral load and infection duration favor emergence of such variants. While the overall probability of emergence and subsequent transmission from any given infection is low, on a population level these events occur fairly frequently. Targeting these low-probability stochastic events that lead to the establishment of novel advantageous viral variants might allow us to slow the rate at which they emerge in the patient population, and prevent them from spreading deterministically due to natural selection. Our work thus suggests practical ways to achieve control of long-term SARS-CoV-2 infections, which will be critical for slowing the rate of viral evolution.

Figure 1

Selection within individuals with COVID-19 leads to selection of more fit viral variants. (A) Schematic of viral replication model used to simulate SARS-CoV-2 evolutionary dynamics. Colors denote the descendants created in generation N_i+1 by the replication burst of each individual ancestor in generation N_i. (B) Sputum viral load curve for a typical COVID-19 infection. The x-axis represents the time starting from the initial transmission event that caused the infection. (C) Mean frequency of variants with point mutations within individuals with COVID-19 for different mutation fitness effects (colors). (D) Probability of a specific single mutation to be present in at least one virion transmitted if transmission occurs within the first 3–7 days of infection (lighter red curves) or anytime during infection (darker red). For (C) and (D), shaded areas represent ± SEM, n = 1000 simulations per condition. (E) Number of total new single mutant infections generated per day that establish a surviving variant lineage, assuming all infections are of standard length with viral load profile given in (B) and that transmission occurs within the first 7 days of infection. Unless otherwise specified, simulation parameter values are those given in Table 1.

Figure 2

Within-host selection is stronger for infections that last longer or have higher viral loads. (A) (Left panel) example of increased peak viral load period length. (Right) Probability that at least one transmitted virion has a specific advantageous single mutation, for different overall lengths of infection. Infection lengths were adjusted by increasing the length of the peak viral load period (left panel schematic). (B) (Left panel) example of decreased viral load. (Right) Probability that at least one transmitted virion has a specific single mutation, for different viral loads. Viral loads were adjusted by reducing the viral load by a constant factor over the entire course of infection (left panel schematic). For right panels, shaded areas represent ± SEM, n = 1000 simulations per condition, with a selection coefficient of 0.2 for the single mutant. Unless otherwise specified, simulation parameter values are those given in Table 1.

Figure 3

SARS-CoV-2 can acquire multiple mutations during infections with sustained viral replication. (A) Fitness valley crossing model for acquisition of multiple mutations. Intermediate states with fewer mutations (light grey) have lower fitness than the WT virus within individuals, while variants with a specific combination of two or three mutations (blue and red, respectively) have higher fitness. (B) Mean frequency of variants with a beneficial combination of two (blue) or three (red) mutations within individuals with long-term SARS-CoV-2 infection. (C) Probability of a beneficial combination of two mutations (blue line) to be present in at least one virion transmitted if transmission occurs anytime during infection. For (B) and (C), shaded areas represent ± SEM, n = 1000 simulations per condition. (D) Number of de novo double mutant infections that establish a surviving lineage when some COVID-19 patients shed live virus for more than 30 days after developing symptoms. Deleterious intermediates had a fitness cost of 0.05 and beneficial mutation combinations had a selective advantage of 0.2. Unless otherwise specified, simulation parameter values are those given in Table 1.

Figure 4

Interventions targeting stochastic events during infection or transmission reduce the efficiency of SARS-CoV-2 evolution. (A) Schematic showing events necessary for the generation and population-level establishment of a new SARS-CoV-2 variant. (B) Reduction in generation rate of new double mutant lineages that establish in the entire population caused by interventions targeting different events required to create a new surviving viral variant (n = 1000 simulations per condition). Each intervention reduces the stated parameter value by 90%. Parameter value changes are relative to the control parameter set given in Table 1, with 0.1% of new infections lasting > 30 days after symptom onset. The “reduced transmissibility” condition refers to a 90% reduction in the transmission advantage of the variant over wild-type under steady-state transmission conditions, leading to a lower reproductive number R₀ of 1.05.