Potential Benefits and Harms of Colon Cancer Screening; Sleep in Adolescents

TTHealthWatch is a weekly podcast from Texas Tech. In it, Elizabeth Tracey, director of electronic media for Johns Hopkins Medicine in Baltimore, and Rick Lange, MD, president of Texas Tech Health El Paso, look at the top medical stories of the week.

This week’s topics include sleep in adolescents, GLP-1 receptor agonists in different people, predicting Alzheimer’s onset, and screening for colorectal cancer (CRC) and potential benefits and harms.

Program notes:

0:44 Adolescents and sleep duration

1:44 Very short sleep duration

2:44 Increased among all subgroups

3:45 Structural factors more important

4:45 Moving school start later

5:02 GLP-1 agonists and various patient characteristics

6:02 Greater weight loss in women

7:06 A blood test to predict onset of symptomatic Alzheimer’s

8:06 Median error of 3-4 years

9:06 Window of 11.4 years for an 80-year-old

10:05 Screening for colorectal cancer

11:05 Colonoscopy and fecal immunochemical testing (FIT) diagnosed early

12:33 End

Transcript:

Elizabeth: What’s going on with sleep among adolescents in the U.S.?

Rick: Do the GLP-1 receptor agonists work differently in different patient populations?

Elizabeth: Can we predict the onset of symptomatic Alzheimer’s disease with a plasma test?

Rick: And routine screening for colorectal cancer versus usual care, advantages and adverse effects.

Elizabeth: That’s what we’re talking about this week on TTHealthWatch, your weekly look at the medical headlines from Texas Tech University Health Sciences Center in El Paso. I’m Elizabeth Tracey, a Baltimore-based medical journalist.

Rick: And I’m Rick Lange, president of Texas Tech Health El Paso.

Elizabeth: Rick, how about if we turn first to JAMA and take a look at this issue of, gosh, what is going on with adolescents in the United States with regard to their sleep?

This is a research letter. They’re examining data from 2007 to 2023 and looking at sleep, and whether any of these trends that they identify differ by demographic or behavioral risk subgroups. This data is from the 2007-2023 National Youth Risk Behavior Survey. This is a school-based survey that purports to provide a representative estimate of what’s going on with U.S. school students in different areas, not just with respect to sleep.

Their sample included 120,000+ students who answered a question about sleep duration on an average school night. Insufficient sleep was less than or equal to 7 hours per night based on the American Academy of Sleep Medicine guidelines. And very short sleep was defined as less than or equal to 5 hours per night. And let me just note parenthetically that there’s at least one person on this call who engages in that behavior on a regular basis, and that would not be me. Then they also looked at 15 health risk behaviors that were associated with insufficient sleep and they grouped those into five categories: mental health, victimization, physical or sedentary, weight-related behaviors, and substance use.

What they found was that the percentage of students who reported insufficient sleep increased from just shy of 69% in 2007 to just shy of 77% in 2023. This increase was driven by a rise in that very short sleep — let me remind you, less than or equal to 5 hours a night sleep duration — which increased from just shy of 16% to 23%.

Why are these kids sleeping less? It didn’t appear to be due to social media. It increased across all demographic subgroups by sex, grade, race, and ethnicity, occurring as much or more among adolescents without behavioral risk factors among those with them. This says, at least to the authors and the editorialist, that what we really need are societal and system changes in order to change this. It’s not going to be an individual-level intervention that’s going to help.

Rick: First of all, nailed, OK? I don’t get as much sleep as is recommended for these kids because at adolescence, they’re recommending 8 to 10 hours of sleep each night. OK, I don’t do that. But I’m no longer an adolescent.

We do know that sleep deprivation in adolescence causes worse cognitive performance, worse academic achievement, as well as an increased risk of depression and other mental health conditions. Do they actually assess use of social or digital media? I don’t recall seeing that.

Elizabeth: The fact that this insufficient sleep increased as much or more among students without behavioral risks, suggesting that structural and environmental factors affecting most adolescents — rather than specific behaviors such as electronic media use, substance use, or sedentary activity — are contributing to this widespread sleep loss.

Rick: They actually didn’t assess digital use. They assessed what they call behavioral risk, and what I would submit to you is that the use of digital and social media is increased across all types of adolescents. I don’t care what your risk is.

They say there are a couple things we can do. Delay school opening, because I didn’t realize that during puberty, it actually causes a shift in melatonin secretion and sleeping drive that lead to difficulty for these kids to fall asleep before 11 p.m. So if we can delay school for an hour, studies have shown kids have a longer sleep duration and a decrease in depressive symptoms. Limiting social and digital communication to at least 2 hours before sleep also improves sleep duration. It’s really disconcerting that almost 80% of kids sleep for less than 7 hours a night. That’s a problem.

Elizabeth: It is a problem. The editorialist, of course, advocates, just as you’ve already identified, for moving the start time back for schools, and then they identify, of course, all of the factors that make that difficult to do: parent work schedules and teachers working later. However, I think since we’re all vested in the health of our adolescents, this sounds like a strategy that’s well worth entertaining.

Would you like to go to JAMA Internal Medicine?

Rick: Do GLP-1 receptor agonists work equally well across different types of patients? These were first approved for the FDA use in 2005 for treatment of diabetes and subsequently we found out that there’s significant weight loss associated with these. They’re used at least as often or more often for weight control in nondiabetic patients. Some people respond well with a significant amount of weight loss and others don’t. If we look across the different characteristics — things like age, sex, race, ethnicity, baseline body mass index [BMI], baseline hemoglobin A1c — is there a difference with regard to how people respond to the GLP-1 receptor agonist?

They looked at all randomized controlled trials where they could get all this information. They found 41 different articles representing 64 randomized controlled trials. Weight loss was greater among women than men, about 11% decrease in weight in women versus about 7% in men. There was no significant heterogeneity with regard to age, race, baseline BMI, or baseline hemoglobin A1c. So we’re not going to be successful targeting GLP-1s to specific patient populations, with the exception of women having a better response than men.

Elizabeth: This is good information because if I’m prescribing and I can ask my patient if they have insurance coverage or whatever other factors might be involved with the decision-making process, that’s a good thing.

Rick: And I agree. It’s interesting why women may respond more than men. The authors hypothesize that it may be the ability of these GLP-1 receptor agonists to synergistically interact with estrogen.

Elizabeth: It’s going to be interesting to see what happens with all of this as things start to transition into the oral meds.

Rick: It is. In fairness, when they looked at the GLP-1 receptor agonists, they looked at all five, but most of the data were restricted to semaglutide [Ozempic, Wegovy] and dulaglutide [Trulicity]. Maybe when we get additional information by bigger studies with the other agents, and especially the oral agents, we may have different information.

Elizabeth: Turning from here then to Nature Medicine, something that it seems like it’s going to be a while before this develops any clinical utility, at least to me: predicting the onset of symptomatic Alzheimer’s disease with plasma p-tau217 clocks.

We can get a plasma sample from people and look at a number of different biomarkers, and specifically for Alzheimer’s disease, this phosphorylated tau. They have developed a longitudinal plasma %p-tau217. That’s the ratio of phosphorylated to nonphosphorylated tau at a certain position on the molecule. And they looked at this in two independent cohorts of folks who had Alzheimer’s disease.

What they found is that this estimated age at the time when this ratio shows up as being positive was associated with the age at onset of Alzheimer’s disease symptoms, they say with a median absolute error of 3 to almost 4 years. The other thing that they noticed was that that ratio and the time to symptom onset was markedly shorter in older individuals.

This is interesting, and it’s interesting from a, I guess, intellectual perspective that you might be able to say, all right, when I get this ratio and it indicates positivity that it looks like you’re going to end up with symptomatic Alzheimer’s disease within a certain window. But when you drill down and you look at that window, even in the older folks, they still are only able to say that symptom onset in those who were positive at age 80 had a median time of 11.4 years. And if you get this ratio to become positive at age 60, their median time until symptom onset, it was 20.5 years.

I have to say, a) I’m not surprised that somebody who’s 81 years old may become symptomatic for Alzheimer’s disease within a window of 11.4 years. I mean, I would predict that a priori, regardless of what I saw in plasma. So I unfortunately am not feeling terribly persuaded by this. The authors conclude, however, that this particular ratio might be useful in identifying people who might be better to enroll in clinical trials for interventions.

Rick: I agree, Elizabeth. It’s not ready for prime time, but I applaud the authors for what they’re trying to do. The most common cause of dementia is characterized by amyloid plaques, but those amyloid plaques increase for about 10 to 20 years in the preclinical phase of Alzheimer’s dementia. So by the time that someone has symptoms, we haven’t been able to show that treating the later stages has been beneficial. We have to be able to predict whether the medicine prolongs the time to when Alzheimer’s or prevents it. And how do we do that? We have to predict when it will occur. There’s still a lot more to be done on this particular topic.

Elizabeth: Remaining in Nature Medicine, then.

Rick: The two major ways that we screen for colorectal cancer, one is to do colonoscopy. Alternatively, people use fecal immunochemical testing. So we compare those two.

We actually have not compared those to usual care because there is some minor risk associated with getting colonoscopy. And this was done in Sweden where 75% of the population is involved in this trial where they randomized to colonoscopy, FIT, or usual care, and they followed them for 4 years. Was the colorectal cancer screening by either of those techniques better at identifying early cancer, that is stage I or stage II cancer, when it’s easier to care? Do those individuals suffer an increased risk of adverse events compared to usual care? Almost 280,000 individuals aged 60 years or older.

And here’s what they discovered. Both types of colorectal cancer screening were effective in detecting earlier cancer more than usual care. Colonoscopy, 38% increase opportunity to diagnose early colorectal cancer. And with FIT, about 20% increased risk. Interestingly enough, over that first year after screening, there was a slightly increased risk of gastrointestinal complications or cardiovascular complications. What the authors couldn’t tell us is were those slightly increased risks due to the procedures or not? We don’t know. But overall, I think it confirms that screening does allow earlier detection of colorectal cancers.

Elizabeth: In view of the fact that colorectal cancer is the fastest growing cancer among those 50 years of age and younger, I would be interested in seeing an extension of this study into that population.

Rick: Right. This study was started before those recommendations occurred. But I agree with you. Listeners should be aware, as we’re screening earlier, because there’s an increased risk of colorectal cancer in younger populations.

Elizabeth: Right. In fact, I think I just saw data that it might be the most common cancer in those younger than 50 years of age.

Rick: Wow.

Elizabeth: And unaccountable factors at this point that are driving that. We just don’t know.

Rick: Right. Since both are effective, you can choose whether you want to have colonoscopy or FIT testing.

Elizabeth: On that note then, that’s a look at this week’s medical headlines from Texas Tech. I’m Elizabeth Tracey.

Rick: And I’m Rick Lange. Y’all listen up and make healthy choices.