Antihypertensive drugs are so 80s and 90s (well, except for pulmonary hypertension) to the point that I cannot recall when I last read research in the NEJM about a new blood pressure drug. One of the reason is that we have so many existing drugs (most of them cheap generics) but despite this, a fraction of hypertensive patients are refractory and cannot get to their blood pressure goal despite using multiple agents. This is where baxdrostat (CinCor Pharma) comes in; it is an oral drug blocking the synthesis of aldosterone, a key hormone driving salt retention by the kidney. In a 12-week phase 2 trial, it achieved a very meaningful lowering of systolic blood pressure of 20 mmHg (10 mmHg vs. placebo) in patients who were already on 3+ agents and still hypertensive. Further trials will be needed to confirm clinical effect, but to me the renal aspect of the mechanism of action raises an interesting possibility that like ACE inhibitors, baxdrostat might have a role in mitigating the progression of chronic kidney disease. Phase 2 Trial of Baxdrostat for Treatment-Resistant Hypertension
How sticky is gene therapy for hemophilia A?
The top line results of the phase 3 study for Roctavian (Biomarin) are very encouraging: the vast majority of patients experienced no bleeds and did not need exogenous factor. However, looking at the details (especially in the Supplementary Appendix that accompanies the paper) the picture is somewhat less sunny given an overall trend of gradual decline in factor VIII levels over time. In the study, 6 out 134 patients switched back to prophylactic therapy; what will it be like at 5 years? We may be in a situation where for many patients, instead of a cure, gene therapy becomes a drug holiday from the burden of prophylaxis. It also cries out for performance-based arrangements for reimbursement. Two-Year Outcomes of Valoctocogene Roxaparvovec Therapy for Hemophilia A; Hemophilia A Gene Therapy — Some Answers, More Questions
Algorithm vs. clinical judgment
Heart failure exacerbation is a common presentation to the emergency department and while in many cases it is clear that an admission to the hospital is warranted (or not), a lot of the time the decision is not straightforward and hinges on the somewhat subjective interpretation of a constellation of data elements by a clinician, and less advertised, on other ancillary considerations (e.g. hospital bed availability, the patient’s home support system, and, gasp… the patient’s insurance). What if one used an algorithm instead? This is what the COACH trial aimed to explore with 5452 patients enrolled in a cluster randomized trial (2972 managed by clinical judgment, 2480 by algorithm). Outcomes (death or hospitalization) slightly favored the algorithm. ED docs, the NPs are coming for your jobs. Trial of an Intervention to Improve Acute Heart Failure Outcomes
AIs for medical information and care
A special article describing the capabilities of GPT-4 in medical care in a series of extremely informative use-cases. The first about GPT-4 answering a simple patient-driven request for medical information about a drug (metformin), the second about GPT-4 using the transcript of a physician-patient encounter to generate a medical note, the third about GPT-4 answering a question from the US Medical Licensing Examination, and the fourth about GPT-4 making recommendations for the management of a patient with an exacerbation of COPD. It’s well worth reading, and shows how, at the very least, AIs will eventually become powerful productivity tools for health care professionals, though their tendency to occasionally confabulate (the famous issue of hallucinations) precludes for now their use as quasi-autonomous agents. Medical scribes, I am afraid your days are numbered. Benefits, Limits, and Risks of GPT-4 as an AI Chatbot for Medicine
The New England Journal of Medicine is a premier weekly medical journal covering many topics of interest to the health sector. In this series we offer an opinionated perspective on selected highlights that might be of interest to our clients and others.