Pharmaceutical treatments can slow bone degradation, thus reducing the fracture risk inherent in osteoporosis. Antiresorptive treatments block the over-activation of osteoclasts vs osteoblasts, but the resulting decrease in bone remodeling frequency may weaken bone structure over time, with no gain in bone volume. Anabolic treatments, however, induce gain in bone volume. The quantitative results from existing studies on the effects of treatments over time are general and nonpatient-specific, while numerical models simulating evolution of patient-specific bone microarchitecture consider a spatially random distribution of the remodeling process. Here, we propose a new approach to simulate the remodeling over decades of an individual patient's bone microarchitecture, based on the hypothesis that the oldest sites, which are hypermineralized and more brittle, are remodeled first. Taking these older sites as prime targets of remodeling, simulations show that severe osteoporosis profoundly degrades the mechanical properties of the bone structure, which can be restored and even improved by anabolic, more than by antiresorptive, therapies.