Summary: In a post hoc analysis of three clinical trials spanning the glucose tolerance spectrum, fasting and 1-hour circulating Gremlin 2 (Grem2) levels were significantly lower in participants with type 2 diabetes than in those without (728±25 vs 649±31 pg/mL fasting, P=0.020; 631±26 vs 537±31 pg/mL at 1 hour, P=0.007). Fasting Grem2 was restored after antidiabetic treatment (P=0.019) and 1-hour Grem2 increased following calorie restriction (P for trend=0.002). The 1-hour Grem2 level was positively associated with beta-cell function assessed by the oral disposition index and HOMA-β.
PICO Summary
| Element | Detail |
|---|---|
| Population | Participants across the glucose tolerance spectrum from three trials: with versus without T2D (Trial 1, n=59 vs 119); T2D on oral antidiabetic drugs (Trial 2, n=67); prediabetes on calorie restriction (Trial 3, n=231). |
| Intervention | Measurement of circulating Grem2 at 0, 1 and 2 hours during oral glucose tolerance testing; observation across antidiabetic treatment and calorie restriction. |
| Comparison | T2D versus non-T2D; before versus after antidiabetic treatment; across calorie-restriction time points. |
| Outcome | Fasting and 1-h Grem2 lower in T2D (P=0.020; P=0.007). Fasting Grem2 restored after treatment (P=0.019); 1-h Grem2 rose with calorie restriction (P trend=0.002). 1-h Grem2 positively associated with beta-cell function (oral disposition index, HOMA-β). |
Clinical Context
Type 2 diabetes ultimately results from beta-cell failure, the inability of pancreatic beta cells to produce sufficient insulin to overcome insulin resistance. While insulin resistance initiates the disease process, beta-cell dysfunction determines progression from prediabetes to overt diabetes. Gremlin 2 is a secreted antagonist of bone morphogenetic proteins (BMPs); BMP signalling affects insulin sensitivity, adipose tissue function, and potentially beta-cell biology, providing biological plausibility for a metabolic role. This study examined whether circulating Grem2 differs across the glucose tolerance spectrum and whether it correlates with beta-cell function.
Clinical Pearls
- Grem2 tracks with glucose tolerance. Lower fasting and 1-hour levels in type 2 diabetes suggest a potential biomarker of metabolic health rather than a binary indicator.
- The correlation with beta-cell function is mechanistically interesting. Unlike glucose and HbA1c, which reflect the consequences of beta-cell failure, the positive association of 1-hour Grem2 with the oral disposition index and HOMA-β hints at a more direct link to beta-cell biology.
- Treatment-induced restoration suggests modifiability. Fasting Grem2 rose after antidiabetic therapy and 1-hour Grem2 increased with calorie restriction, implying responsiveness to metabolic improvement rather than fixed beta-cell loss.
- BMP signalling is an underexplored metabolic axis. BMPs and their antagonists have emerged as metabolic regulators beyond bone and development.
Practical Application
Clinical applicability is currently limited, as this is discovery-stage research and Grem2 is not a standard clinical test. The findings reinforce that beta-cell preservation is a worthwhile treatment goal and that prediabetes represents an intervention window, since Grem2 is already altered at that stage. Therapies that preserve or restore beta-cell function, together with early intensive glycaemic control and lifestyle modification, remain the practical levers while biomarkers like this mature.
Broader Evidence Context
The search for biomarkers predicting diabetes progression and beta-cell function is long-standing, with traditional markers such as fasting glucose, HbA1c, HOMA-IR and HOMA-β all carrying limitations. Novel candidates including betatrophin, fetuin-A and various adipokines have shown variable utility, and Grem2 adds to this investigational landscape. The observation that treatment improves biomarker levels aligns with evidence that early intensive glucose control can partially restore beta-cell function, though whether Grem2 mediates this recovery or simply reflects an improved metabolic state requires mechanistic study.
Study Limitations
This is an observational post hoc analysis, so associations do not prove causation, and whether low Grem2 causes beta-cell dysfunction or results from it is unknown. Confounding factors such as age, obesity and diabetes duration may influence both Grem2 and beta-cell function. Validation in larger and more diverse populations is needed before any clinical application.
Expert Commentary
I approach a new diabetes biomarker paper with a fixed question: will this ever change what I order or what I do, or is it interesting biology that stays in the laboratory? Grem2 is, for now, firmly the latter, and I want to be honest about that rather than dress a post hoc analysis up as a clinical advance. What earns my attention is not the headline association but the texture of it. The stepwise fall with worsening glucose tolerance, the positive link specifically with the oral disposition index, and the partial restoration after treatment together tell a more beta-cell-centred story than glucose or HbA1c can, because those only report the damage after the fact. The honest limitation is fundamental: this is observational and cannot tell me whether low Grem2 drives beta-cell failure or merely marks it, and pooled trial data carry their own confounding. Can I use this with my patients? Not yet, and I would not order it. Its value today is conceptual, reinforcing that beta-cell preservation deserves to be an explicit treatment goal and that prediabetes is the window worth acting in. I would like to see a prospective study testing whether Grem2 trajectory predicts progression.
References
Ni M, Chen Y, Gu W, et al. Association between circulating Gremlin 2 and β-cell function among participants with prediabetes and type 2 diabetes. J Diabetes. 2025;17(4):e70090. doi:10.1111/1753-0407.70090