Model analysis of the impact of thymic dysfunction or chronic immune hyperactivation on T-cell numbers. (A) Cartoon representing the model as explained in “Patients, materials, and methods.” Briefly, the size of the naive T-cell pool (N) increases with thymic production of naive T cells (σ(t)) and naive T-cell proliferation (p_n), and decreases by cell death (d_n) and activation of naive T cells that acquire a memory phenotype (a). The memory T-cell pool size (M) increases by T-cell activation (r × a; r is the clonal size resulting from activation of a single naive T cell) and proliferation of memory T cells (p_m) and decreases by memory T-cell death (d_m). (B, left panels) The age-related changes in total body naive and memory T-cell and TREC dynamics in healthy children as described by the model accurately imitate changes observed experimentally (see Figures 1 and 2 for comparison). Parameters describing naive T-cell dynamics were adopted from Hazenberg et al¹¹  and are based on experimental data: d_n = 0.001 per day, and p_0n = 0.1 per day. The number of TRECs per mature thymocyte was set to c = 0.2, to reach realistic total TREC numbers. On the basis of experimental estimates,¹  the rate of thymic involution was set to 4% (ie, v = 0.0001 per day). By setting σ₀ = 10⁸ cells/day, the model reveals a thymic output of 4 × 10⁷ cells per day for a 25-year-old individual, which seems appropriate.^18,19  Memory cells were assumed to proliferate and die several fold faster than naive T cells^18,19 : p_0m = 0.2 per day, d_m = 0.02 per day, a = 0.0001 per day, and r = 100 cells. The parameter h was set to 2 × 10⁹ cells, to scale the total peripheral T-cell pool to 2 to 3 × 10¹¹ cells.¹⁸  The effect of HIV infection was simulated by reducing thymic output 100-fold (ie, σ₀ = 10⁶ cells/day; middle panels) or by increasing T-cell activation and death rates (p_0n = 0.2, d_n = 0.005, p_0m = 0.4, d_m = 0.1, and a = 0.0002 per day; right panels). In the upper panels, thick lines represent naive T cells and thin lines the memory T cells. Modeling immune hyperactivation led to a better description of the experimentally observed reduced expansion of the naive and memory T-cell pools during HIV infection than thymic impairment.