Abstract / Summary
Men on testosterone therapy often judge how well they respond to a dose by comparing their measured total testosterone with the weekly milligrams they administer. That ratio depends heavily on the shape of the protocol rather than on the person: the same man on the same delivered dose can appear to respond very differently depending on whether blood was drawn at peak or trough, how often he doses, and which ester he uses. This note describes the dose-response model used in the TRT Plus app, which classifies response from a single blood draw after removing these protocol effects. On the level side, a peak reading is converted to the trough of the user's own protocol using that protocol's modeled steady-state fluctuation, and the result is re-expressed as the trough the same average level would produce on a fixed reference protocol. On the dose side, the labeled dose is converted to testosterone base by molecular weight and then to a reference-ester equivalent; cream doses use a previously published injectable-equivalent model. The ratio of the two gives a response index, which is mapped to six ordered responder types, from Super-Responder to Non-Responder. We describe the pipeline, the ester parameters, the published dose-response evidence the classification is anchored to, implementation conventions and limitations. The reference protocol, band boundaries and calibration are proprietary and withheld. The model classifies how someone responds; it is not dosing guidance, has not been validated against clinical outcomes, and is published as a methods description.