Decoding Inflammation: How Hybrid Diabetes Phenotypes Are Rewriting the Rules of Diagnosis and Treatment

Diabetes has long been divided into two seemingly distinct worlds: type 1, an autoimmune disease in which the immune system destroys insulin-producing beta cells, and type 2, a metabolic disorder driven by insulin resistance and lifestyle factors. But a new review published in Frontiers in Endocrinology argues that this binary classification is increasingly out of step with clinical reality. As global rates of overweight and obesity continue to climb, researchers led by Francesca D’Addio of the University of Milan describe an emerging convergence between the two diseases, giving rise to hybrid phenotypes such as double diabetes and latent autoimmune diabetes in adults (LADA) that blur the boundaries of traditional diagnosis and demand entirely new therapeutic strategies.

The numbers behind this convergence are striking. The DCCT/EDIC study, one of the longest-running investigations of type 1 diabetes, documented an increase in obesity prevalence among patients from just 2 percent at baseline in the 1980s to 28 percent after twelve years of follow-up. The international SWEET registry reported that nearly one in three children and adolescents with type 1 diabetes is now overweight or obese. Meanwhile, observational cohort studies encompassing approximately 1.7 million individuals have found that people with overweight or obesity face a pooled risk ratio of 1.87 for developing type 1 diabetes, a figure that rises to 2.55 when confirmed by islet autoantibody testing. These findings suggest that the relationship between metabolic dysfunction and autoimmune destruction is not coincidental but bidirectional and mechanistically intertwined.

At the heart of this convergence lies a paradox of modern type 1 diabetes treatment. Intensive insulin therapy, while dramatically reducing long-term complications, is itself a major contributor to weight gain and insulin resistance. The review explains that subcutaneous insulin delivery bypasses the liver’s first-pass metabolism, creating peripheral hyperinsulinemia and hepatic hypoinsulinemia. Chronically elevated peripheral insulin levels promote oxidative stress, inflammation, and dysregulation of insulin receptor expression, worsening the very insulin resistance that drives further weight gain. As glycemic control improves, calories that would once have been lost through glucosuria are retained, and alterations in growth hormone and IGF-1 signaling pathways further disrupt normal body composition regulation. The result is a vicious cycle in which higher insulin doses beget more weight gain, which in turn demands even more insulin.

But the influence of obesity on autoimmune diabetes extends far beyond the mechanical effects of insulin therapy. The review details how adipose tissue itself functions as an active immune organ. In obesity, the balance between pro-inflammatory Th17 cells and regulatory T cells is disrupted, while populations of regulatory B cells and invariant natural killer cells in adipose tissue dwindle. Increased lipolysis releases free fatty acids that recruit pro-inflammatory M1 macrophages and activate the NLRP3 inflammasome, driving secretion of the interleukins IL-1β and IL-18, cytokines deeply implicated in autoimmune pathology. Perhaps most provocatively, dying adipocytes can release antigens previously hidden from immune surveillance, potentially triggering autoreactive responses. Two complementary frameworks, the accelerator hypothesis and the overload hypothesis, propose that obesity-driven insulin resistance forces beta cells to work harder, accelerating their apoptosis and increasing their immunogenic visibility to the immune system.

This mechanistic overlap gives rise to two distinct hybrid phenotypes. Double diabetes refers to patients with type 1 diabetes who develop overweight, insulin resistance, and metabolic abnormalities characteristic of type 2 diabetes, often with a family history of the latter. LADA, by contrast, is a slowly progressive autoimmune diabetes that emerges in adults who frequently present with clinical and metabolic features resembling type 2 diabetes at onset. Both conditions carry a substantially elevated risk of complications. The FinnDiane study found that individuals with double diabetes had a hazard ratio of 2.52 for diabetes-related mortality compared to those with type 1 diabetes without metabolic syndrome. The review attributes this heightened risk to the combination of insulin resistance, altered cardiovascular risk factors, and a chronic low-grade inflammatory milieu that promotes endothelial dysfunction and oxidation of atherogenic LDL cholesterol.

Distinguishing these hybrid phenotypes from classical diabetes subtypes requires tools that go far beyond clinical observation. The review synthesizes evidence from genomics, miRNomics, metabolomics, and microbiomics to construct a multi-omics picture of disease heterogeneity. Genetic studies reveal that LADA shares substantial etiological overlap with type 1 diabetes, particularly at the major histocompatibility complex region and at genes including PTPN22, SH2B3, and the insulin gene, while showing only modest contributions from type 2 diabetes susceptibility variants aside from a notable signal at HNF1A. Notably, the class I MHC effects characteristic of childhood-onset type 1 diabetes are absent in LADA, providing a genetic discriminator between the two autoimmune subtypes. In double diabetes, Mendelian randomization analyses support a causal role for childhood adiposity in elevating disease risk, while polymorphisms in the KLF14 locus modulate cardiometabolic and inflammatory profiles in young patients with type 1 diabetes.

Circulating non-coding RNAs have emerged as particularly promising biomarkers for differential diagnosis. Plasma-derived exosomal miRNA profiling identified 75 dysregulated species in LADA, with three candidates, miR-146a-5p, miR-223-3p, and miR-21-5p, demonstrating strong discriminatory performance between LADA and type 2 diabetes. Circular RNA analysis revealed extensive dysregulation implicating TNF signaling, FoxO activity, and cellular senescence pathways. At the metabolomic level, LADA does not exhibit a unique metabolite signature but rather occupies an intermediate position between type 1 and type 2 diabetes, with patients whose profiles overlap more closely with type 1 diabetes progressing more rapidly to insulin therapy. Gut microbiome studies have identified disease-specific alterations in beneficial taxa such as Bifidobacterium species and their correlations with branched-chain amino acids and short-chain fatty acids, suggesting that dietary interventions targeting the microbiota could modulate both metabolic and immune pathways.

On the therapeutic front, the review highlights a growing arsenal of drugs originally developed for type 2 diabetes that show promise in hybrid phenotypes, largely because of their anti-inflammatory properties alongside their glucose-lowering effects. Metformin, SGLT2 inhibitors, DPP-4 inhibitors, and GLP-1 receptor agonists have all demonstrated improvements in insulin sensitivity, body weight, and cardiometabolic parameters when used as adjuncts to insulin therapy. The ADJUNCT trials showed that liraglutide added to insulin in type 1 diabetes reduced HbA1c, body weight, and daily insulin requirements, though with increased risk of hypoglycemia and ketosis. More recently, semaglutide has shown impressive results: the ADJUST-T1D trial demonstrated that semaglutide combined with automated insulin delivery significantly increased the proportion of adults achieving time in range above 70 percent alongside meaningful weight loss, while the ENDIS study showed improvements in endothelial function and reductions in high-sensitivity C-reactive protein, a key inflammatory marker. The dual GIP/GLP-1 receptor agonist tirzepatide has shown substantial reductions in HbA1c and body weight in GADA-positive patients previously diagnosed with type 2 diabetes, with associated improvements in beta-cell function markers.

Perhaps the most clinically actionable insight from the review concerns the role of C-peptide, a byproduct of insulin production that serves as a proxy for residual beta-cell function. Experts suggest that patients with fasting C-peptide levels below 0.3 nmol/L should be treated with insulin alone, those between 0.3 and 0.7 nmol/L may benefit from glucose-lowering drugs combined with basal insulin, and those above 0.7 nmol/L can be managed with algorithms typically reserved for type 2 diabetes. This stratification is supported by evidence that dulaglutide was less effective in LADA patients with low C-peptide levels, while semaglutide as add-on therapy significantly improved glycemic control, reduced body weight and insulin requirements, and increased C-peptide levels particularly in those with preserved beta-cell function. The review also proposes a novel concept: a standardized multi-omics and genetic approach could identify what the authors term a double diabetes risk inflammatory signature, or DRIS, to enhance patient stratification for tailored therapeutic approaches.

The challenges ahead are substantial. The molecular phenotype of double diabetes remains poorly characterized, limiting recognition and distinction from other hybrid forms. No randomized clinical trials have specifically evaluated immunomodulatory therapies such as teplizumab, which has shown efficacy in preserving beta-cell function in early type 1 diabetes, in patients with LADA or double diabetes. The review emphasizes that larger, longitudinal studies are urgently needed to unravel the mechanisms by which hybrid forms develop and to determine whether certain therapies offer superior benefit in specific subgroups. What is clear is that the era of treating diabetes as a binary condition is drawing to a close. As the boundaries between autoimmune and metabolic disease continue to dissolve, precision medicine approaches that integrate genetic susceptibility, inflammatory biomarkers, and multi-omics signatures will be essential to identify the right treatment for the right patient at the right time.

Subject of Research: Inflammatory mechanisms and multi-omics biomarkers in hybrid diabetes phenotypes including double diabetes and LADA

Article Title: Decoding inflammation to target hybrid diabetes phenotypes

Article References: Rossi, G., Terranegra, A., Lunati, M. E., Pastore, I., Pozzi Canova, L., Bucciarelli, L., Cimino, V., Loretelli, C., Ben Nasr, M., Fiorina, P., & D’Addio, F. (2026). Decoding inflammation to target hybrid diabetes phenotypes. Frontiers in Endocrinology, 17, Article 1922969. https://doi.org/10.3389/fendo.2026.1922969

Image Credits: AI Generated

DOI: 10.3389/fendo.2026.1922969

Keywords: diabetes, inflammation, obesity, LADA, double diabetes, immunology, precision medicine, endocrinology, GLP-1 receptor agonists, multi-omics, Decoding, target