LAUSR

The UK Biobank is a large prospective study that was recently established to examine genetic and lifestyle risk factors for a variety of chronic diseases affecting middle-aged and older individuals living in the United Kingdom.1 Between 2006 and 2010, >9 million individuals 40 to 69 years who were registered in the UK National Health Service and lived within 25 miles of one of the 22 assessment centers in England, Wales, and Scotland were invited to enter the UK Biobank study. A total of 5.5% (≈500 000) volunteers consented and were enrolled in the UK Biobank study after answering baseline questionnaires, completing physical examinations, and providing baseline blood and other specimens.1 Participants are followed up prospectively for incident events through linkage to the electronic health care record. Although population based, the UK Biobank study is not representative of the general UK population in that it enrolled healthier volunteers who were older and were more likely to be women and white (95%), more likely to have a higher socioeconomic status, and less likely to have prevalent cardiovascular disease (CVD) or cardiovascular risk factors (eg, lower rates of smoking, lower body mass index) compared with the general UK population of the same age. Indeed, age-adjusted all-cause mortality rates over the 6-year follow-up in the UK Biobank are about half those seen in the general UK population.1 In the present study conducted among the UK Biobank study population, Welsh et al2 examined the associations of standard lipids (total, low-density lipoprotein [LDL], high-density lipoprotein [HDL], non-HDL cholesterol) and compared them with apolipoprotein (apo) B and A in relation to incident CVD. In the primary prevention subset of the UK Biobank (ie, those with no self reported previous CVD and not taking statin therapy), the authors found the expected associations of higher atherosclerotic CVD (ASCVD) risk with higher baseline levels of LDL cholesterol, non-HDL cholesterol, or apo B, which were mostly comparable in magnitude to each other and similar to the associations with ASCVD risk that previous studies had found. Furthermore, the results confirm results from previous studies in the United States such as the Framingham study,3 the Women’s Health Study,4 and the international individual participant-level meta-analysis results from the Emerging Risk Factor Collaboration,5 all of which did not show substantial improvement in risk prediction with apo B or apo A compared with standard lipids for overall CVD risk prediction. The study by Welsh et al2 in this issue of Circulation also provides several important insights. First, should clinicians measure apo B if overall risk prediction is similar for apo B compared with the standard lipids, in particular total and HDL cholesterol? Clinically useful biomarkers are ones that change patient management through more accurate classification of risk. From a clinical standpoint, the © 2019 American Heart Association, Inc.