# Personalized p-THYROSIM model for thyroid hormone dynamics, hypothyroidism treatment & implementation in an iOS version for wide distribution

## TL;DR

A personalized thyroid hormone simulation tool, p-THYROSIM, was improved and made available as an iOS app and Python version to optimize hypothyroidism treatment.

## Contribution

A refined, personalized p-THYROSIM model with iOS and Python implementations for simulating thyroid hormone dynamics and treatment.

## Key findings

- Combination therapy with low LT3 doses (5–7.5 ug) was more effective in restoring euthyroid levels than mono-therapy.
- The iOS and Python versions of p-THYROSIM were validated in clinical and simulated disease scenarios.
- Personalization using height and weight, rather than BMI, improved model accuracy for hormone replacement dosing.

## Abstract

To: (1) update and refine the predictive abilities of original p-THYROSIM, a uniquely personalized simulation tool that mathematically mimics the thyroid hormone regulation system in humans, to optimize replacement LT4 and LT4+LT3 dosing for hypothyroid patients, based on individual hormone levels, heights (H), weights (W) and sex. Refinements include dependence on H and W separately, rather than composite BMI=W/H2. And to: (2) present a new and user-friendly software tool, an iOS p-THYROSIM app for the iPhone and iPad, to accomplish these goals, as well as a Python version for longer term disease progression simulations.

Original p-THYROSIM was refined and updated, first by refitting male and female data for establishing blood volume Vb as a function of Ws and Hs of males and females separately. A superfluous parameter was also removed and FT4 and FT3 output plotting were slightly adjusted to align them with current assay ranges. We also developed two software packages for simulating the model: (1) a 100-day iOS implementation for the iPhone and iPad using Apple developer tools; and (2) a 1000-day version in Python, with time units converted from hours to days to render it more practical for research use with clinical diseases that evolve over months and years.

The iOS app was implemented, exercised and tested in several clinical applications. Most notably, simulation results are shown and compared for hemi-thyroidectomy and for optimal dosing for mono- and combination hormone replacement therapies. With the Python version, graphic hormone responses for 240 days of evolving mono- and combination replacement therapies were illustrated for a simulated Hashimoto’s disease patient. With both versions, simulated combination therapy was shown to be more effective in achieving normal range FT4, FT3 and TSH concentrations in plasma.

p-THYROSIM can predictively provide accurate mono- and combination LT4+LT3 replacement hormone therapies for male and female hypothyroid patients, personalized with their Hs and Ws. Where combination therapy is warranted, our results predict that not much LT3 (typically 5 – 7.5 ug) is needed in addition to LT4 to restore euthyroid levels, with larger LT3 doses rarely needed, suggesting opportunities for further research exploring safe and effective combination therapy with lower T3 doses and slow-releasing T3 formulations.