Back

Modeling disease progression in newly diagnosed type 2 diabetes

Hao, K.; cao, y.

2020-05-05 pharmacology and toxicology
10.1101/2020.05.04.076133 bioRxiv
Show abstract

Type 2 diabetes (T2DM) is a progressive disease, which is primarily characterized by a decline in {beta}-cell function and worsening of insulin resistance. Unfortunately, most interventions (lifestyle, diet, and therapeutic agents) for T2DM only provide a transient restoration of {beta}-cell function and the progression is inevitable once it starts. To understand the natural progression of T2DM, a mechanistic model was developed to quantitatively characterize the dynamic interactions among {beta}-cell function, plasma fasting glucose (PFG), fasting insulin (FI), and the degree of insulin resistance, starting from an early stage of T2DM over up to 8 years. The model was validated using clinical data to optimize the disease parameters. The restoration and deterioration rates of {beta}-cell function were both predicted as 84.5 %/year and 1.10 /year for early stages of T2DM. The model predicted a positive correlation between the initial level of {beta}-cell function at diagnosis and its maximum restoration potential, underscoring the importance of early diagnosis and intervention. After the treatment, {beta}-cell function could be temporarily restored within several months, which has a long-term benefit in glycemic control. The maximal tolerated PFG level that permits {beta}-cell function restoration was predicted to be around 8.33 nM; and the temporal restoration of {beta}-cell function would be unlikely at a PFG level above this threshold. The intrinsic deterioration rates of {beta}-cell function and insulin resistance were both critical factors for long-term glycemic control. In conclusion, our model provides a quantitative analysis of the natural disease progression in T2DM and yields insights into factors that are critical for long-term glycemic control.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.