Deciphering the relationship between type 2 diabetes and fracture risk: the genetic and observational evidences

Reviewer #1 (Public Review):

Summary:
The manuscript of Zhao et al. aimed at investigating the relationships between type 2 diabetes, bone mineral density (BMD) and fracture risk using Mendelian Randomization (MR) approach.
The authors found that genetically predicted T2D was associated with higher BMD and lower risk of fracture, and suggested a mediated effect of RSPO3 level. Moreover, when stratified by the risk factors secondary to T2D, they observed that the effect of T2D on the risk of fracture decreased when the number of risk factors secondary to T2D decreased.

Strengths:
- Important question
- Manuscript is overall clear and well-written
- MR analyses have been conducted properly, which include the usage of various MR methods and sensitivity analyses, and likely meet the criteria of the MR-strobe checklist to report MR results.

Weaknesses:
- Previous MR studies on that topic have not been discussed
- Multivariable MR could have been used to better assessed the mediative effect of BMI or RSPO3 on the relationships between T2D and fracture risk.

Reviewer #2 (Public Review):

The authors employed the Mendelian Randomization method to analyze the association between type 2 diabetes (T2D) and fracture using the UK Biobank data. They found that "genetically predicted T2D was associated with higher BMD and lower risk of fracture". Additionally, they identified 10 loci that were associated with both T2D and fracture risk, with the SNP rs4580892 showing the highest signal. While the negative relationship between T2D and fracture has been previously observed, the discovery of these 10 loci adds an intriguing dimension to the findings, although the clinical implications remain uncertain.