This week in Endocrinology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Endocrinology. This week we're covering 10 notable papers spanning adrenal and neuroendocrine disease, the heterogeneity and cardiometabolic burden of diabetes, and the question of when to start, stop, or simplify hormone therapy. Let's dive in.

We begin with the adrenal gland and its neighbours, where three papers address conditions that are either rare, underrecognized, or short on good treatment data. The largest is an international retrospective cohort in The Lancet Diabetes and Endocrinology from Chacko and colleagues, drawing on 49 high-volume referral centres across five continents and nearly 1,500 patients with ectopic Cushing syndrome [1]. This is a sick population: more than half of patients had metastatic disease, roughly nine in ten had severe hypercortisolism, and close to four in five had hypokalaemia. Over a median follow-up of almost five years, about one in eight patients died of causes adjudicated as Cushing-related, and around a third of patients developed an infection. The central finding is that achieving complete biochemical remission within six months of diagnosis was independently associated with roughly halving both the risk of late complications and Cushing-related death, while showing no association with deaths from other causes. That specificity is reassuring, although this remains retrospective data, and patients who reach remission quickly may differ in ways that adjustment cannot fully capture. The authors argue that the findings support framing ectopic Cushing syndrome as a time-critical endocrine emergency. Staying with cortisol, Endocrine Reviews carries a scientific statement from Carvajal and colleagues on non-classical apparent mineralocorticoid excess [2]. The panel formally defines it as partial insufficiency of 11-beta-hydroxysteroid dehydrogenase type 2 without biallelic gene variants, marked by an elevated cortisol-to-cortisone ratio, low cortisone, low renin, and normal-to-low aldosterone. Given that roughly a third of people with hypertension have a low-renin phenotype, the panel positions this as a common but overlooked cause. The statement proposes a mass spectrometry diagnostic algorithm alongside a resource-adapted pathway and endorses mineralocorticoid receptor antagonists as mechanism-directed first-line therapy, targeting normalization of both blood pressure and renin. This is consensus rather than trial evidence, and the authors themselves call for threshold standardization and multicentre validation before it enters hypertension guidelines. Finally, in the Journal of Clinical Endocrinology and Metabolism, Román-González and colleagues pooled three prospective trials covering 114 patients treated with sunitinib for metastatic pheochromocytoma and paraganglioma [3]. About a fifth of patients had a partial response, with no complete responses, and about half of patients achieved stable disease; median progression-free survival was just under a year. The cost is substantial, with severe adverse events in more than half of patients. The authors rate the certainty as low to very low, concluding that sunitinib offers meaningful disease control in a subset of patients but modest response rates.

Our second theme is diabetes, both how we classify it and how widely its cardiovascular consequences reach. In The Lancet Diabetes and Endocrinology, Prystupa and colleagues applied unsupervised machine learning to 600 adults from the German Diabetes Study with recent-onset type 1 diabetes, then reproduced the structure in a separate registry of nearly 500 patients [4]. They found three subtypes: a classic antibody-positive group that was leaner, insulin sensitive, and had severe beta-cell dysfunction; a mixed antibody-positive group with higher body mass index, insulin resistance, and better preserved beta-cell function; and an antibody-negative group with lower genetic type 1 risk. The two antibody-positive groups did not differ in autoimmunity or type 1 polygenic risk, and modelling placed the mixed group about six years earlier along the beta-cell decline trajectory. Retinopathy risk was roughly five times higher in the mixed group than the classic group. The authors interpret insulin resistance as bringing autoimmune diabetes to clinical attention earlier, opening a potential window for disease-modifying therapy, though this is a hypothesis to be tested in trials. In the Journal of Clinical Endocrinology and Metabolism, Liu and colleagues meta-analysed nine randomized trials and one cohort, covering more than 81,000 women, comparing lower versus higher glucose thresholds for diagnosing gestational diabetes [5]. Lower criteria doubled diagnoses and pharmacotherapy, and increased neonatal hypoglycaemia by about 18 additional cases per thousand births and maternal-neonatal separation by about 6 per thousand, with little to no effect on other short-term outcomes. Longer-term data came from a single trial suggesting less maternal hypertension and child neurodevelopmental impairment at five years, but more subsequent gestational diabetes. That leaves a genuine tension between short-term harms of overdiagnosis and possible long-term benefits that rest on one trial. Broadening the lens, a narrative review in Nature Reviews Endocrinology from Gong and colleagues looks beyond atherosclerosis and heart failure, summarizing evidence that diabetes is a risk factor for atrial fibrillation, infective endocarditis, hypertensive heart disease, intracerebral haemorrhage and aortic stenosis, and reviewing what is known about incretin-based drugs and SGLT2 inhibitors for each [6].

Two papers address cardiometabolic risk reduction more directly. In PLOS Medicine, Blais and colleagues followed more than 185,000 new statin users without cardiovascular disease across Hong Kong, New Zealand and the United Kingdom [7]. Compared with an achieved LDL cholesterol of 2.6 millimoles per litre or higher, reaching 1.8 to 2.5 was associated with about a fifth fewer major cardiovascular events, with a similar picture for non-HDL cholesterol. The lowest category, below 1.4, did not reach statistical significance for cardiovascular events, and very low levels were not associated with higher all-cause mortality after adjustment. As an observational study, confounding by indication is a real concern, which the authors acknowledge. Their conclusion is that achieving non-HDL cholesterol below at least 3.4 millimoles per litre after statin initiation could yield further event reductions in primary prevention. On obesity, the ATTAIN-J trial led by Yokote, in The Lancet Diabetes and Endocrinology, randomized 238 Japanese adults with obesity disease to oral orforglipron at 6, 12, or 36 milligrams or placebo for 72 weeks [8]. Weight fell by roughly six to nine percent across doses versus under one percent with placebo, and about six in ten participants on the two higher doses lost at least five percent of body weight. Notably, the 36 milligram dose was not more effective than 12 milligrams and had more nausea-related discontinuations. The trial is modest in size and specific to Japanese indications, but it adds to the evidence for an oral, non-peptide option in this population.

Our last theme concerns simplifying long-term hormone therapy. In Thyroid, Toloza and colleagues updated their meta-analysis of supervised thyroid hormone discontinuation, pooling 23 studies, nearly all observational [9]. Overall, about 37 percent of participants remained euthyroid after stopping, rising to around 41 percent of those originally treated for subclinical hypothyroidism but only about 11 percent of those with overt disease. Around 60 percent of patients ultimately restarted therapy, and positive antibodies, heterogeneous ultrasound appearance and higher baseline TSH predicted reinitiation. Symptoms were generally unchanged and adverse events uncommon. Heterogeneity was very high, but the authors conclude the data support considering structured deprescribing in selected patients who lacked a clear original indication. In paediatrics, the REAL8 and REAL9 trials reported by Maniatis and colleagues in the Journal of Clinical Endocrinology and Metabolism examined once-weekly somapacitan in children born small for gestational age [10]. Height velocity stayed well above baseline through two years, including after switching from daily growth hormone, safety was comparable, and most children who switched preferred the weekly injection. These are open-label data, so preference findings carry some expected bias.

If you only have time for one paper this week, make it the ectopic Cushing syndrome cohort from Chacko and colleagues [1]. It brings the largest outcome dataset yet to a rare condition and reframes it around a concrete, time-bound target, the first six months, that ties early biochemical control specifically to Cushing-related harm.

That's your roundup for This Week in Endocrinology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if you have a paper you have been meaning to read, upload the PDF, or paste any link, at audioscholar dot C C. We will turn it into audio like this one, in any of thirty-one languages.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.