Do We All Really Need Those Recommended Cancer Screenings?

There’s been some promising news about an effective approach for targeting cancer screening at the people who need it the most. Before I tell you about the new study, let’s first take a look at how — and why — cancer screening needs to change.

Most of the cancer screening that’s performed today is recommended for huge groups of people. Women are encouraged to get annual mammograms to check for breast cancer starting at the age of 40. Virtually everyone 45 and older is advised to get a colonoscopy, repeated at regular intervals, to check for colorectal cancer. And so on.

These approaches can only be considered “targeted” in the broadest possible sense: they target people by age group. But cancer is not equally likely in all people, even within those age groups. And with such broad recommendations, too many people who may be at risk of cancer wind up skipping their screenings because they’re uncomfortable or inconvenient, or they don’t really seem necessary. When screenings can be directed to those who need them most, people may be more likely to bite the bullet and get checked out. And if it were somehow possible to identify the lowest-risk people and allow them to safely skip their screenings, that would help free up resources in a strained healthcare system.

That’s why this new study offers so much potential. In a randomized clinical trial of more than 28,000 women conducted at the University of California and other institutions, researchers compared standard annual screening for breast cancer with risk-based screening. Patients came from across the United States, ranged in age from 40 years to 74 years old, and had no prior history of breast cancer. Half of patients underwent annual screening during the seven-year study, and the other half had their risk evaluated based on genomic and clinical data. For the risk-based arm of the study, women at the highest risk were screened every six months; those at the lowest risk received no screening until they reached the age of 50. Participants were followed for at least two and a half more years to check for outcomes.

According to the researchers, the upshot of the study was clear: women fared just as well in the risk-based model as they did in the standard screening model. Women in the lowest-risk group who skipped their annual screenings did so safely, while women in higher-risk groups benefited from more aggressive screening. There was no meaningful difference in the rate of breast biopsies between the groups.

This entire approach is based on accurately categorizing each person’s risk, and the researchers spent some time discussing that in the paper describing the study. They used a combination of genetic testing and traditional clinical risk scoring to classify each participant’s risk; this method appears to be more accurate that either technique on its own. They also noted that relying on reported family history as a proxy for risk does not adequately reflect each person’s risk — a full 30% of the study participants who had genetic variants associated with cancer risk did not report any family history of breast cancer. The researchers contend that population-scale genetic testing and analysis is now feasible and useful to help target cancer screening at people with the highest risk of developing cancer.

While this study focused on breast cancer, the same approach could be just as useful for screening many other types of cancer. Colonoscopies, for example, might be used too broadly right now, while lung cancer screenings are probably used too narrowly. Addressing these issues with risk-based screening models could have a profound effect not just on detecting cancer but also on allocating healthcare resources more efficiently.

Salisbury’s Take will be taking a couple of weeks off for the upcoming holidays. I’ll be back in January with the latest news about science and medicine. Happy holidays, everyone!

Catch up on stories you may have missed:

• News Roundup: ‘Prebunking’ Fake News and More

• For Pediatric Cancer, Genomic Profiling Helps Guide Patient Care

• Bird Flu Update: H5N1 Virus Still Widely Circulating

• $100 Million Grant Awarded for the Prevention of Infectious Disease Outbreaks

• Beating the Odds: How People Stay Healthy When Their Genomes Should Be Causing Disease

• Brilliant or Bumbling? AI Tools Aren’t Quite Ready for Genetic Diagnosis