Results
Figure 2 reports the overview of treatment effects by intervention time and disease metric. The major points from this overview are:
  1. Outcome improves for every disease metric with earlier intervention. This suggests that PrEP- and PEP provides the best opportunity for repurposed drugs with (potentially) low potency to impact disease.
  2. Treatment initiated after viral peak have little to no impact on viral load AUC. Treatments initiated 3 days after viral peak have no impact on Epithelial Cells Infected. Treatments initiated after viral peak still have potential to shorten the Duration of viral shedding.
  3. Combinations targeting multiple pathways can be as effective or more effective as targeting single pathways with equivalent summed treatment effects. The heterogeneity observed at each summed effect level suggests that some combinations are less effective than others.
Figure 3 compares all 1 and 2 target treatments by treatment initiation time (Peak -3, 0, +3 days), endpoint, and 1 and 2 log10 summed drug effect. Supplemental 2 reports results for all combinations. All endpoints are improved with earlier intervention. The major points from this presentation are:
Table 1 provides a qualitative ranking of target choices by metric of interest. Slowing transition, k, of infected epithelial cells from eclipse, I1 , to productive, I2 , is not effective relative to other target choices. Increasing turn-over, δ, and/or decreasing productivity, ρ, of infected epithelial cells, I2 , is predicted to have positive benefit for all metrics and should be considered a “backbone” of proposed combinations. Broadly speaking, targeting δ and ρ seek to disrupt the production machinery of SARS-CoV-2. Increasing virion kill, c, to deplete extracellular virions,V , is predicted to have positive benefit for all metrics. Inhibiting infection, β, is interchangeable with c for the Epithelial Cells Infected metric, but not the viral load AUC and Duration of viral shedding: c both removes virions and prevents infection but β only prevents infection.
Supplemental Figure 3 and Figure 4 report the predicted impact viral and infected epithelial cell kinetics assuming intervention six days before and six days after peak viral load, respectively.