Figure 5.

Upper panel in (A) shows the bifurcation diagram for how the persistent (stable) dormant infection (lower blue curve), the persistent (stable) fatal infection (upper blue curve), and the infection burden at the threshold curve (separatrix, middle red curve) varies with the baseline production (α) of naïve T-cells. The lower panel shows the similarity for cancer instead of infection. Panel (B) shows a similar diagram as in panel (A) but instead, as a function of age assuming α decreases linearly with age. The distance between the dormant infection curve and the threshold curve (indicated by the black dotted double arrow at year 25 as an example) is inversely proportional to the probability of disease escape due to infection. Panel (C) illustrates how cancer and infection progress over time for different perturbation corresponding to the initial value at time 0. The dotted curves start with a high disease burden (near the persistent fatal disease steady state) while the full curves start with a low disease burden (near the persistent (stable) steady state). The red curves and axis show the relative cancer burden while the blue curves and axis show the infection burden. From upper left to lower right, α = 16, 17, 99, 100. Panel (D) shows a hypothetical treatment scenario for an elderly VP with disease progression toward fatal disease (low value of α). First, disease escape is seen (no treatment), thus treatment is applied by increasing α (shown as the green dotted curve). This results in a decrease in disease burden. When the VP is in the basin of attraction for the persistent (stable) dormant state, the VP is ‘cured’. However, stopping the treatment will likely let α approach its low value again and history repeats.