Last, we evaluate how time-to-treatment impacts host dynamics. In contrast to the previous assumptions, where drug concentrations are at full efficacious level when viral replication begins, Figure 4 shows how delaying treatment up to 48h (A-C) and 24 h (D-F) before peak viral load reduces the observed delay (to 3 and 0 days respectively, compared to 5 in the main analysis), while the increased distribution of cell infection over times remains similar to earlier treatment.