番茄社区

Normally, I like titles of articles to tell me what the conclusion is. But in this case, I'd like the audience to come up with their own ending. The issue at hand is the recent guidance by the American College of Physicians (ACP) that proposes colorectal cancer (CRC) screening starting at age 50 and using fecal occult blood-based testing every 2 years, and flexible sigmoidoscopy every 10 years with interval fecal immunochemical testing or colonoscopy every 10 years.

For background, I'll start the story in 2016, when the U.S. Preventive Services Task Force (USPSTF) recommended screening in average-risk adults ages 50 to 75 using one of several modalities. Two years later, the American Cancer Society came out with their guideline to initiate screening at , citing the epidemiologic trends for a relative increase in CRC incidence in those younger than 50 and based on modelling studies with the emerging data. Much public debate ensued, and experts weighed in on pros and cons of initiating screening at age 45 versus 50. Fast forward to 2021, all major guidelines (, American College of Gastroenterology, and the U.S. Multisociety Task Force) updated their recommendations and aligned at age 45 to initiate screening. With Medicare, Medicaid, and commercial payer coverage for screening at age 45, the last 2 years have been directed at spreading awareness, improving screening rates, and conducting studies to understand prevalence of advanced neoplasia and CRC in those under age 50.

This brings us to present day, with surprising guidance from ACP that seems to fly in the face of this neatly folded issue and challenges the consensus. The most remarkable aspect is that all guideline bodies look at the same evidence, yet come up with very different recommendations. How is this possible? To answer this question, let's first briefly review the two sides of the debate.

There is strong evidence that screening reduces the risk of colorectal cancer. The incidence rates of CRC in individuals 50 and older have continued to decline while cancer cases in individuals younger than 50 have been rising. In 2023, a total of of CRC (13% of all CRC cases for the year) are projected to occur in individuals younger than 50. There is also showing that these cancers are detected at advanced stages. Initiating screening at age 45 is to potentially reduce cancer diagnosis and deaths by detecting precancerous lesions or cancer that may have been growing for years, but otherwise may not be detected until CRC screening at age 50. Extending the benefit of screening to 45- to 49-year-olds is by epidemiologic data and modelling studies.

The other side of the issue is that there are no randomized controlled trials demonstrating the benefit of screening in 45- to 49-year-olds, and epidemiologic data and modelling studies are indirect and too weak to drive such a big guideline shift. Age being the strongest risk factor for colon cancer, the absolute number of colon cancers is for 50- to 54-year-olds compared with 45- to 49-year-olds (59.5 vs 33.1 per 100,000, respectively), and keeps increasing after age 55. And there are always harms to consider, from invasive tests, or downstream sequelae of over-testing and over-diagnosis. And, of course, it is costly. While other guideline bodies may be agnostic to cost, the ACP feels a responsibility for the high price tag associated with resource burden for screening an additional approximately 20 million younger individuals. Modelling that do factor in cost agree that from a societal perspective we get much more bang for our buck by improving screening rates in those 60 and older, and through increasing efforts to encourage colonoscopy after an abnormal fecal immunochemical test.

So, how do guideline experts come up with different conclusions when looking at the same evidence? The answer lies in the framework or context of the guideline application. For example, what resources are we willing to take up to achieve this, and how do we compare it to other competing efforts? Certainly, a preventable cancer in a younger individual, no matter how rare, is tragic and we would like to avoid it as much as possible -- but at what cost? Guideline bodies can certainly have different opinions on this, which can influence the ultimate recommendations.

Another question at the heart of the debate is this: Does lack of evidence constitute evidence of lack of benefit in this case? We don't have evidence from randomized clinical trials or an equivalent for most things we do in medicine. And we have also reversed course on many practices once held as dogma, that were later found not helpful or even harmful.

So, what is the outcome when guidelines collide?

One option to complete the opening sentence is: When guidelines collide...everyone wins. Highlighting different viewpoints helps us make better informed decisions that take into account our healthcare system's resources and our population's risk and competing priorities.

Another option is: When guidelines collide...everyone loses. The conflicting messages are confusing for patients and providers, and reduce trust in medical societies and "experts." This in turn undermines efforts to provide a unified message that will increase the uptake of CRC screening (which is ultimately a shared goal).

One thing all the current guidelines agree on is that the majority of new colorectal cancer cases occur in those ages 50 and older, and that this population should remain a focus of our screening efforts. Where they diverge is whether foregoing screening among 45- to 49-year-olds is a missed opportunity to reduce the CRC burden and its associated suffering among a productive group, or if it's a waste of resources and money that is associated with potential harms. I'll let the jury (you) weigh in below in the comments.

is the Robert M. and Mary H. Glickman Professor of Medicine and Gastroenterology and the director of outcomes research in the Division of Gastroenterology & Hepatology at NYU Langone in New York City. She is a practicing gastroenterologist and researcher in colon cancer screening and quality indicators of colonoscopy.

Comment