This Week in General Medicine — Sep 28, 2026

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

Welcome to This Week in General Medicine. This week we're covering 7 notable papers spanning sex-specific risk and cardiovascular medicine, randomised trials that either delivered or fell flat, and the newer frontiers of immunotherapy toxicity and global disease elimination. Let's dive in.

We start with women's health, where a large registry study in The BMJ sharpens a question clinicians field every week: how safe is contemporary menopausal hormone therapy? Berggreen and colleagues used nationwide Danish registries to build a nested case-control study within a cohort of women aged 50 to 69, identifying nearly ten thousand women with venous thromboembolism, more than eighteen thousand with ischaemic stroke, and close to twelve thousand with myocardial infarction, each matched by birth year to thrombosis-free controls, with women who had prior thrombosis, cancer, thrombophilia and several other conditions excluded up front [1]. Compared with no current use, current oral oestrogen therapy, whether alone or combined with a progestin, was associated with roughly a sixty percent higher rate of venous thromboembolism, and more modestly raised rates of ischaemic stroke and myocardial infarction. The absolute numbers matter here, and the authors give them: about one extra venous thromboembolism for every thousand women treated for a year, and a considerably smaller excess for stroke and for infarction. Risk concentrated in higher-dose, longer-duration oral oestradiol, above one milligram a day for more than five years, where stroke and infarction rates were close to double. Transdermal therapy showed no increase in thrombotic rates, with the single exception of combined cyclic transdermal preparations and myocardial infarction, an estimate based on small numbers. This is observational and susceptible to prescribing by indication, but it is nationwide, contemporary, and consistent with the route-of-administration story that has been building for a decade, and so it strengthens the evidence base favouring transdermal delivery in women who want hormone therapy and carry thrombotic concern.

That theme of sex as a biological variable is argued more broadly in a Perspective in Nature Medicine, where Dal Canto and colleagues make the case that cardiovascular medicine remains built on male-pattern disease [2]. They point to spontaneous coronary artery dissection, Takotsubo syndrome and coronary vasomotor disorders as conditions that disproportionately affect women yet were long unrecognised as distinct entities, and therefore remain thin on mechanism and thinner on evidence-based therapy. This is opinion rather than new data, but it is a useful frame for reading a paper like the Danish hormone therapy analysis, where the exposure is female-specific and the outcomes have historically been studied in men.

Turning to randomised trials, this week gives us one clear positive, one clear negative, and one that sits uncomfortably in between. In The Lancet, the ROAM slash EORTC-1308 trial led by Jenkinson addressed adjuvant radiotherapy after complete resection of WHO grade 2 atypical meningioma, randomising 157 patients across 58 centres in 11 countries to sixty gray in thirty fractions or to observation [3]. At a median follow-up of just over five years, recurrence occurred in 14 percent of the radiotherapy group versus 30 percent under observation, and five-year disease-free survival was about 80 percent with radiotherapy compared with about 64 percent with observation, roughly halving the hazard of recurrence or death. Grade 2 or 3 radiation-related serious adverse events occurred in about eight percent of those treated, with no treatment-related deaths. The trial recruited far fewer patients than screened, so precision is limited, and the authors themselves frame the result as supporting a shared decision in which recurrence prevention is weighed against potential late toxicity.

Also in The Lancet, the REFIT2 trial is squarely negative. Johnsen and colleagues randomised 450 adults with moderate-to-severe irritable bowel syndrome across five Norwegian hospitals, two to one, to a single rectal enema of donor faecal microbiota or to an autologous placebo enema, with everyone masked [4]. At ninety days, about 40 percent of the donor group and about 38 percent of the placebo group had achieved the prespecified 75-point drop on the severity score, an absolute difference of around two percentage points that was not statistically significant, with similar adverse event rates. This is a properly powered phase 3 trial in a field built largely on small positive studies, and the authors conclude that microbiota modulation alone is likely insufficient for symptom improvement in irritable bowel syndrome. Notably, nearly four in ten patients improved on placebo, a reminder of how much of the apparent benefit in earlier open trials may have been non-specific.

The ambiguous case is ASPEN-06 in Nature Medicine, where Shitara and colleagues added evorpacept, a CD47 blocker that enhances antibody-dependent phagocytosis, to trastuzumab, ramucirumab and paclitaxel in 127 patients with pretreated HER2-overexpressing gastric or gastro-oesophageal junction cancer [5]. Investigator-assessed response was about 40 percent with evorpacept versus about 27 percent with the backbone alone, and in the subgroup with HER2 positivity confirmed on a fresh post-trastuzumab biopsy, about 55 percent versus 23 percent. Those differences exceeded the prespecified improvement thresholds, meeting one of two primary objectives, but neither response rate cleared the statistical bar against the historical 30 percent benchmark. Haematologic toxicity was more common. The authors describe the efficacy as encouraging, and it is phase 2 data awaiting the phase 3 portion.

Two final papers extend the week in different directions. Also in Nature Medicine, Kethidi and colleagues characterised enterocolitis after ciltacabtagene autoleucel, a B cell maturation antigen-directed CAR-T therapy for myeloma, using single-cell transcriptomics and imaging of intestinal biopsies from ten affected patients, seven treated controls without enterocolitis, and 26 healthy volunteers [6]. They found mucosal B cell and plasma cell depletion, persistent highly cytotoxic CAR-T cells in the mucosa, and interferon and JAK-STAT reprogramming across stromal, endothelial and epithelial compartments, with two patients improving clinically, endoscopically and histologically on the oral JAK1 inhibitor upadacitinib. That is mechanistic work plus two cases, hypothesis-generating rather than practice-defining, but relevant as B cell-targeted therapies spread into autoimmune disease. And in The Lancet Global Health, Leta and colleagues surveyed more than 121,000 Ethiopian children at over four thousand sites and built one-kilometre-resolution prevalence maps after a decade of preventive chemotherapy [7]. Combined prevalence was about five percent, heavy-intensity infection was about two in a thousand, below the World Health Organization elimination threshold of one percent, and the population living in areas needing annual mass treatment fell from roughly 26 million to about 11 million, with around 11 percent of subdistricts still above the annual treatment threshold.

If you only have time for one paper this week, make it the Danish nested case-control study of menopausal hormone therapy in The BMJ [1]. It is the paper most likely to come up in a general medicine clinic this month, and it puts contemporary numbers, by route and by dose, on a risk conversation that has been running on twenty-year-old trial data.

Here is what this week's evidence adds up to in general medicine. First, the thrombotic risk of menopausal hormone therapy appears to be largely a property of the oral route and of higher-dose, longer-duration oestradiol, with absolute excess risk on the order of one event per thousand women per year; the data are observational but nationwide and contemporary. Second, adjuvant radiotherapy after complete resection of atypical meningioma reduces recurrence in a single modest-sized randomised trial, and the authors frame it as a value-dependent decision rather than a mandate. Third, faecal microbiota transplantation did not outperform placebo in properly powered phase 3 testing in irritable bowel syndrome, which should temper enthusiasm for microbiome-only strategies in functional gut disease. Fourth, CD47 blockade in HER2-positive gastric cancer produced higher response rates but did not clear its statistical benchmark, so it remains investigational. And fifth, the wider agenda items, sex-specific cardiovascular biology and schistosomiasis elimination, are both about evidence infrastructure: knowing where disease actually is, and in whom.

That's your roundup for This Week in General Medicine. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

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