A recent Endocrinology digest,
as sent to subscribers

Study Type: Phase 3 randomised, placebo-controlled trial (with open-label extension)

Key Question: Is HTD1801 monotherapy safe and effective for glycaemic control in patients with type 2 diabetes inadequately managed by diet and exercise alone?

Key Findings:

• HTD1801 1,000 mg BID reduced HbA1c by -1.3% vs -0.6% with placebo at 24 weeks (LS mean difference -0.7%; 95% CI -0.8 to -0.5; P<0.0001), from a baseline of 8.5% in both arms.
• Glycaemic improvements were durable to 52 weeks; secondary cardiometabolic and inflammatory markers also improved.
• Diarrhoea was the most common adverse event (9.6% vs 0.7% placebo), mostly at treatment initiation; 56-week safety was consistent with the double-blind phase.

Clinical Relevance: HTD1801 may offer NHS clinicians a novel, generally well-tolerated oral option for early type 2 diabetes management, with potential added cardiometabolic benefits beyond glycaemic control.

Limitations: The trial population and comparator design (placebo rather than active standard-of-care agents) limit direct comparison with existing first-line therapies used in UK practice.

Study Type: RCT with nested cohort analysis (follow-up study)

Key Question: Does third-trimester energy restriction or spontaneous weight loss in women with gestational diabetes and BMI ≥25 kg/m² adversely affect maternal and infant body composition/growth up to 12 months postnatally?

Key Findings:

• Randomisation to a 1,200 vs 2,000 kcal/day diet did not alter antenatal or postnatal infant growth; infants remained normal weight for age throughout year one.
• In cohort analysis, maternal weight loss (~3 kg) was associated with lower maternal weight at 3 months (not 12 months), reduced risk of large-for-gestational-age birth, and lower infant fat mass at 3 months.
• No evidence of growth faltering, excessive catch-up/catch-down growth, or increased prevalence of low infant weight (<5% with weight-for-age <-2 SD).

Clinical Relevance: Supports the safety of dietary energy restriction/weight loss advice in obese women with gestational diabetes, reassuring UK clinicians that such interventions—increasingly considered in NHS antenatal diabetes pathways—do not compromise infant growth in the first year.

Limitations: Cohort (non-randomised) weight-loss analysis is subject to residual confounding, and longer-term child outcomes (e.g., obesity at 3 years) remain unassessed.

Study Type: Cohort study (prospective, dual-cohort: UK Biobank and All of Us)

Key Question: In people with diabetes, does cardiovascular health (LE8) modify risk of MCI and dementia, and does this interact with genetic risk (APOE ε4, PRS)?

Key Findings:

• Moderate/high CVH vs low CVH was associated with lower MCI risk (HR 0.82, 95% CI 0.71–0.95) but showed no significant association with all-cause dementia (HR 0.97, 95% CI 0.84–1.11).
• Among high genetic-risk individuals, better CVH still lowered MCI risk (HR 0.78, 95% CI 0.62–0.98; significant additive interaction), though the protective effect appeared attenuated compared with low/moderate genetic risk groups.
• Findings were consistent across UK Biobank and All of Us cohorts.

Clinical Relevance: Supports proactive cardiovascular risk optimisation (diet, activity, glycaemic/lipid/BP control) as a modifiable strategy to reduce MCI risk in diabetic patients seen in NHS diabetes and memory services, regardless of genetic risk profile.

Limitations: Observational design limits causal inference, and dementia diagnosis reliance on routine records may underestimate incidence/misclassify outcomes.

Study Type: Secondary analysis of RCT data (EASE 2/3 trial repository) using regression and machine-learning modelling.

Key Question: What is the minimum frequency of well-day capillary ketone testing required to predict 1-month DKA/severe ketosis risk in type 1 diabetes without losing predictive accuracy?

Key Findings:

• Weekly well-day ketone testing matched twice-weekly testing for predicting DKA/severe ketosis risk (AUC 0.692 vs 0.678, P=0.19 for maximum-ketone models; 0.719 vs 0.711, P=0.38 for gradient-boosted tree models).
• Reducing testing frequency below weekly was not evaluated as maintaining equivalent accuracy.

Clinical Relevance: Supports a simplified, less burdensome ketone self-monitoring schedule (once weekly) for risk-stratifying DKA in type 1 diabetes patients, particularly relevant for those on SGLT2 inhibitors or intensive insulin regimens within NHS diabetes services.

Limitations: Derived from SGLT2 inhibitor trial populations (EASE 2/3), so generalisability to broader type 1 diabetes populations and real-world adherence patterns is uncertain.

Study Type: RCT (controlled feeding trial, secondary analysis)

Key Question: Does a DASH-style diet adapted for diabetes (DASH4D) improve glycaemic biomarkers compared with a typical American diet in adults with type 2 diabetes?

Key Findings:

• DASH4D significantly reduced fructosamine (adjusted difference −5.6 μmol/L, P=0.002), fasting glucose (−4.5 mg/dL, P=0.02), and HbA1c (−0.09%, P=0.04) versus comparison diet.
• Effects were consistent across higher and lower sodium versions of each diet.
• Cohort was older (mean 67 years), predominantly female (65%) and Black (87%).

Clinical Relevance: Supports dietary pattern advice (DASH-style, low sodium) as an adjunct glycaemic management strategy in NHS diabetes dietetic services, particularly relevant given UK guidance promoting dietary approaches alongside pharmacotherapy.

Limitations: Short 5-week feeding periods and controlled feeding design limit generalisability to real-world dietary adherence and long-term outcomes.

Study Type: Cross-sectional multi-cohort analysis (harmonised individual-level data)

Key Question: Do complication profiles differ between lean and overweight/obese African adults with type 2 diabetes?

Key Findings:

• Lean patients (BMI <25 kg/m²) had higher pooled prevalence of retinopathy (PR 1.36, 95% CI 1.13–1.63) and stroke (PR 1.41, 95% CI 1.01–1.99), but lower hypertension (PR 0.77, 95% CI 0.71–0.85) and lower 10-year cardiovascular risk (PR 0.85, 95% CI 0.74–0.97) vs overweight/obese.
• No difference in chronic kidney disease prevalence between groups.
• Body fat percentage mediated up to 92% of observed differences; lean phenotype linked to reduced beta-cell function and lower insulin levels, supporting a distinct pathophysiology.

Clinical Relevance: With ~40% of African adults with type 2 diabetes being lean, and increasing UK diaspora populations of African descent, clinicians should consider phenotype-specific risk stratification and avoid assuming insulin-resistance-driven pathology in lean patients when guiding screening and treatment choices.

Limitations: Cross-sectional design precludes causal inference regarding complication trajectories.

Study Type: Narrative review

Key Question: How have cohort studies shaped current understanding and clinical guidelines for type 2 diabetes over the past 50 years, and where might this research direction go next?

Key Findings:

• Global cohorts (population-based, disease-based, intervention-based, registry-based) have informed diagnostic criteria, screening strategies, complication risk, and pathophysiological understanding of type 2 diabetes.
• Observational cohort data complement, rather than replace, RCTs and animal studies, offering real-world insight into disease progression not always captured experimentally.
• Future directions highlighted include integration of artificial intelligence, precision health approaches, and exposome-based research to refine risk prediction and phenotyping.

Clinical Relevance: UK clinicians rely heavily on cohort-derived evidence (e.g., UKPDS legacy, biobank data) underpinning NICE and international diabetes guidelines, and this review contextualises how such evidence continues to evolve alongside emerging analytic tools.

Limitations: As a narrative (non-systematic) review, it lacks a reproducible search strategy or quality appraisal, limiting objectivity in evidence selection.

Study Type: Population-based cohort study using an interrupted time series (natural experiment) design.

Key Question: Does infant rotavirus vaccination alter the incidence of type 1 diabetes in early childhood?

Key Findings:

• Among 740,744 Norwegian children (2007–2019), 846 developed type 1 diabetes between 6 months and 5 years.
• Primary interrupted time series analysis showed a slightly increased incidence trend post-vaccination introduction (relative HR 1.11, 95% CI 1.03–1.20), but this lost significance after excluding transitional birth cohorts (relative HR 1.05, 95% CI 0.93–1.18).
• Secondary analysis comparing vaccinated vs unvaccinated children showed no significant association (adjusted HR 1.17, 95% CI 0.78–1.77).

Clinical Relevance: Reassures UK clinicians and immunisation policymakers that rotavirus vaccination (part of the routine NHS infant schedule) does not appear to reduce—or increase—type 1 diabetes risk, countering earlier hypothesis-generating claims of a protective effect.

Limitations: Observational design with potential residual confounding from herd immunity effects and modest case numbers limiting precision of subgroup estimates.