Cardiovascular outcomes in a SELECT-like obesity population: real-world insights from the swedish AROS study

Background In the SELECT trial, semaglutide reduced major cardiovascular event (MACE) in individuals with overweight or obesity and established cardiovascular disease (CVD), but without diabetes with an estimated Number Needed to Treat (NNT) of 67 to avoid one MACE. However, data on real-world event rates in similar populations is needed to increase knowledge of potential preventive effects. Purpose To analyse event rates and NNT in an obese SELECT-like population and to explore event rates compared to the general Swedish population. Methods Primary and specialist care data from healthcare registries and electronic healthcare records in three Swedish regions from 2013 to 2023 was collected (representing approximately 40% of the Swedish population). Individuals aged ≥45 years with a body mass index (BMI) ≥30 kg/m² (lower BMI was not available in this database) and established CVD (in line with SELECT trial inclusion criteria) without diabetes were included. MACE and heart failure were collected from national registries. NNT for adverse CV outcomes was estimated by applying the relative effect from the SELECT trial to the absolute risk observed in our SELECT-like obesity population. Additionally, NNT was calculated separately for each BMI class. Since age and sex distributions varied between BMI classes, NNT calculations were weighted to match the distribution in BMI class 1. In addition, a matched (year of birth, sex and region of residence) general Swedish background population (n=52 140) was constructed in a 1:5 manner. Results The SELECT-like obesity population (n=10 428) had a mean age of 68.7 years and comprised 58.9% men. The mean BMI was 33.2 kg/m², with 79.5% classified as class 1 obesity, 15.4% as class 2 obesity, and 5.0% as class 3 obesity. The majority had a history of myocardial infarction (47.1%) or stroke (40.5%) and very few (8.3%) had previous peripheral artery disease. Compared to the SELECT trial, the real-world SELECT-like obesity population were older, more often women and more often had previous stroke. NNT to avoid one MACE was 34.9 (95% CI 22.3-79.4) in the SELECT-like obesity population (Figure). Over a mean follow-up time of 5.7 years the SELECT-like obesity population had an increased risk of MACE compared to the general population (HR 2.19; 95% CI 2.09-2.31) and risk was even higher for heart failure (2.77; 2.57-2.98). Conclusion In a real-world setting a SELECT like population with obesity were older, more often women and burdened with more previous stroke than the SELECT trial. Accordingly, NNT to avoid one MACE was substantially lower in real life, estimated to 35. These findings highlight that the potential preventive effect of semaglutide is even larger in a population with obesity and cardiovascular disease than what was seen in the original SELECT study. Furthermore, this real-life population had a significantly higher risk of an adverse cardiovascular outcome than the general population in Sweden.