Is it time to stop PSA testing? … and other research

Back to the drawing board on prostate cancer screening

Prostate cancer screening is currently based on prostate-specific antigen (PSA) testing followed by a detailed multiparametric magnetic resonance imaging (mpMRI) if the PSA is raised. But we know that PSA testing misses some cancers and throws up lots of false alarms, so is it time to ditch it?

This decision model analysis used lifetime data from US men aged over 55 years with no prior screening or diagnosis of prostate cancer to compare biennial PSA testing (with or without second-line mpMRI if PSA >4 µg/L) with biparametric MRI (bpMRI) as a screening test (with MRI guided biopsy or ultrasound as a secondary test). Biparametric MRI is quicker than mpMRI and doesn’t involve an injection of dye which some people react badly to.

Compared with PSA testing, bpMRI prevented 2-3 deaths from prostate cancer and added 4-11 days of life. The disadvantage of bpMRI was a marked increase in the number of biopsies (1506 to 4174) and overdiagnosis (38 to 124). The current practice of first-line PSA testing therefore remains the more cost effective option.

Ann Intern Med doi:10.7326/M23-1504

Rheumatic heart disease in low and middle income countries

One consequence of our cavalier overprescribing of antibiotics for sore throats is that acute rheumatic fever and rheumatic heart disease, which result from an autoimmune reaction to streptococcal infection, have virtually disappeared from high income countries. However, rheumatic heart disease remains an important problem in low and middle income countries (LMIC) as a result of poorer living conditions and less access to prompt penicillin treatment.

This large prospective observational study of nearly 14 000 patients from 24 LMICs highlights the seriousness of rheumatic heart disease. Nearly 15% of patients died within three years, mostly due to heart failure or sudden death. Valve surgery substantially reduced the risk of dying, but only 5% of the cohort were operated on in the three year study period, despite most of them having symptoms.

Antibiotic prophylaxis, anticoagulation, and good access to treatment of valve disease should all improve outcomes in established rheumatic heart disease. The Holy Grail remains prevention of acute rheumatic fever and rheumatic heart disease by improving living conditions and appropriate use of penicillin for cases of strep throat.

JAMA doi:10.1001/jama.2024.8258

Shedding light on rare diseases

In common diseases, whole exome sequencing is useful because, even though the exome only covers 60 million base pairs (2% of the whole genome), it contains the DNA regions that encode for proteins and the variants that contribute to diseases. But genetic variants responsible for rare disorders are harder to find. Could whole genome sequencing—which covers 3 billion base pairs including the exons (coding regions) and less studied introns (non-coding regions)—offer more information?

This study used whole genome sequencing in 744 families with suspected monogenic disease in whom previous genetic testing hadn’t revealed a diagnosis. A molecular diagnosis was made in 218 families (29.3%), mainly by re-examining the exome with a few extra tweaks and just over a quarter after whole genome sequencing. Similar results were obtained in research and clinical settings. Overall, the diagnostic yield of whole genome sequencing in this population, in whom a genetic cause of a rare disease was suspected but hadn’t been found, was 8%.

There’s justification for using whole genome sequencing as a first-line test in the diagnosis of rare diseases. But for now it's unlikely to be cost effective when exome sequencing doesn’t yield an answer in more common scenarios.

N Engl J Med doi:10.1056/NEJMoa2314761

Bariatric surgery bears fruit

Polycystic ovary syndrome (PCOS) is the commonest reason that women can’t get pregnant because they’re not ovulating regularly (anovulatory infertility). Obesity contributes to the problem, but women with PCOS find it particularly hard to lose weight, so just telling them to lose weight is unhelpful.

This small randomised controlled study found that treating obese women who have PCOS and irregular periods with bariatric surgery (sleeve gastrectomy) resulted in 2.5 times more spontaneous ovulations compared with drug treatment (metformin or orlistat) and behavioural therapy over the one year study period. The surgical group lost a significant amount of weight (−29% body weight), whereas the medical group didn’t lose weight at all. The surgical group also had a greater reduction in cardiometabolic risk factors and reported significantly better quality of life and psychological health. There were more adverse events in the surgical group (66.7% v 30%), but no treatment related deaths and no long term serious ill effects.

Numbers were small, the new “wonder drug” semaglutide wasn’t tested, and impact on fertility in terms of actual pregnancies wasn’t measured. But the results do suggest that bariatric surgery may offer hope to women with PCOS and obesity who are struggling to get pregnant.

Lancet doi:10.1016/S0140-6736(24)00538-5

Antidepressant discontinuation symptoms

Many people stop taking their antidepressants abruptly despite our advice not to. Most seem to cope, and I’ve often wondered how common and serious discontinuation symptoms are.

This useful meta-analysis of 6095 studies including 21 000 patients found that around one in three patients have discontinuation symptoms after stopping an antidepressant (while one in six report discontinuation-like symptoms after stopping placebo). Subjective factors may play a part, and it can be hard to tell whether symptoms are due to discontinuation or recurrence of the original depression. Approximately one in 30 patients have severe antidepressant discontinuation symptoms. Certain antidepressants (venlafaxine, imipramine, paroxetine) are associated with a higher incidence and severity of reported antidepressant discontinuation symptoms. The bottom line is that we need to monitor and support our patients when they come off the drugs at least as carefully as we do when they’re on them.

Lancet doi:10.1016/S2215-0366(24)00133-0