The Weekly Briefing
The Everhealth Weekly
Evidence-based intelligence for preventing chronic disease and extending healthspan.
Good morning,
This week’s most-shared health stories fall neatly into the three pillars we care about most: your heart and metabolism, your brain, and cancer prevention. Two stories from each, read by a physician.
From the heart: what happens when people stop a GLP-1 drug like Ozempic or Mounjaro, and how hidden inflammation may be quietly reshaping the heart. From the brain: a 95,000-person study linking REM sleep to lower dementia risk, and evidence that the right blood thinner may protect memory in people with atrial fibrillation. From oncology: an FDA advisory panel vote on the Galleri multi-cancer blood test, and a 25-year Danish study showing the cancer gap in type 2 diabetes is getting wider.
As promised last week, this issue’s Long View takes on multi-cancer blood tests: what they can tell you, and what they can’t. That’s Section 3.
— The Everhealth Institute team
This Week in the Evidence
1. Stopping a GLP-1 drug may erase its heart protection, and restarting only partly restores it
Washington University and VA St. Louis researchers led by Dr. Ziyad Al-Aly compared 132,551 veterans on GLP-1 drugs (semaglutide, tirzepatide and others) with 201,136 on sulfonylureas, an older class of diabetes pills. They re-checked every six months to see who stayed on treatment, who took a break, and who stopped.
Staying on a GLP-1 continuously was associated with an 18% lower risk of major cardiovascular events (heart attack, stroke, or cardiovascular death) than taking a sulfonylurea. That protection faded after stopping. Compared with people who stayed on the drug, risk was 14% higher after one year off and 22% higher after two years off. People who took a six-month break and then restarted were 4–8% worse off than continuous users. About 26% of participants stopped entirely.
Source: Xie Y, Choi T, Al-Aly Z. Glucagon-like peptide 1 receptor agonist discontinuation and risks of major adverse cardiovascular events in adults with type 2 diabetes: target trial emulation. BMJ Med. 2026;5(1):e002150. Free full text → · WashU Medicine release → · ScienceDaily summary →
2. Hidden, low-grade inflammation may be quietly reshaping the heart years before symptoms
Imperial College London and MRC researchers led by Prof. Declan O’Regan measured glycoprotein acetyls (GlycA), a blood marker that reflects chronic, low-level inflammation. They then linked it to cardiovascular events, heart MRI scans, and genetics. Higher inflammation was strongly associated with smoking, excess body fat, psychological distress and socioeconomic disadvantage.
People in the top fifth for inflammation had a 43% higher risk of heart attack and stroke than those in the bottom fifth. The same 43% excess risk applied to people whose inflammation stayed high over time, even with no prior heart disease. On MRI, higher inflammation tracked with thicker heart walls, smaller chambers and poorer filling, the kind of silent remodeling that can progress toward heart failure. Genetics mattered too: some people appear naturally more resilient to inflammatory damage, and others more susceptible.
Source: Corianò M, … O’Regan DP. Gene–environment interactions shape cytokine-mediated inflammation and cardiovascular risk. Eur J Prev Cardiol. 2026. doi:10.1093/eurjpc/zwag435 → · ScienceDaily summary → · Medical Xpress coverage →
3. More REM sleep linked to lower risk of 83 diseases, including a 46% lower risk of dementia
Instead of asking people how they slept, researchers had participants wear a wrist accelerometer for a week. A deep-learning algorithm estimated REM, deep sleep, total duration and irregularity, which were then compared with more than 1,000 later diagnoses.
REM stood out. For every additional ~48 minutes of REM per night, risk was lower for 83 diseases, including dementia (HR 0.54), Alzheimer’s disease (HR 0.69), parkinsonism (HR 0.20) and heart failure (HR 0.74). More deep sleep tracked with lower risk of type 2 diabetes and depression. For total sleep, the lowest-risk window for most conditions was 6–8 hours. Sleeping under 5 hours was associated with higher risk of 37 conditions.
Source: Luo J, Liu R, Yin J, Cao W, Sun S, Chen R. Accelerometer-derived real-world sleep stages and risk of incident diseases: a UK Biobank cohort study and phenome-wide association analysis. PLOS Med. 2026. Free full text → · ScienceDaily summary → · Medical News Today coverage →
4. Newer blood thinners linked to slower memory decline in people with atrial fibrillation and Alzheimer’s
Atrial fibrillation (AFib), an irregular heart rhythm, raises the risk of stroke, including tiny “silent” strokes that chip away at memory over time. Karolinska Institutet researchers led by Prof. Maria Eriksdotter compared three matched groups of people living with both AFib and Alzheimer’s: those on newer oral anticoagulants (NOACs, such as apixaban or rivaroxaban), those on warfarin, and those on no blood thinner.
The NOAC group’s memory scores declined about 0.2 points per year more slowly on the standard MMSE test than either comparison group. NOAC users also had lower risks of death, stroke, blood clots and fractures than untreated patients. Warfarin reduced death, stroke and clots too, but carried a higher risk of major bleeding.
Source: Zhu N, Xu H, Garcia-Ptacek S, Mitra S, Eriksdotter M, et al. Oral anticoagulants, cognition, and clinical outcomes in atrial fibrillation and Alzheimer’s disease: a Swedish nationwide study. Eur Heart J. 2026. Journal article → · Karolinska Institutet release → · ScienceDaily summary →
5. FDA advisers vote in favor of the Galleri multi-cancer blood test, as PATHFINDER 2 is published
On September 23, an FDA advisory panel voted 7–2 (one abstention) that Galleri’s benefits outweigh its risks for adults 50 and older, 10–0 on safety, and a narrower 6–4 on effectiveness. The FDA is expected to make its final decision in the coming months. The day before, the PATHFINDER 2 results were published. Adding Galleri to standard screening (mammography, colonoscopy or stool tests, Pap/HPV, and lung CT) found about 6.5 times more cancers than standard screening alone.
Of 287 positive tests, 173 were confirmed cancers, a positive predictive value of 60.3%. The false-positive rate was 0.36%, and the test correctly predicted where in the body the cancer was about 91% of the time. About 70% of the cancers it found have no recommended screening test (pancreas, ovary, stomach, and others), and roughly half were stage I–II. The caveat: the randomized NHS-Galleri trial of about 142,000 people in England missed its primary endpoint. Combined stage III–IV cancers did not fall significantly, although stage IV diagnoses were lower in later screening rounds.
Sources: Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study. Nat Med. 2026. Nature Medicine → · OHSU Knight Cancer Institute release → · FDA advisory meeting page → · CancerNetwork vote coverage → · NHS-Galleri trial (NEJM) →
6. Cancer risk is higher in type 2 diabetes, and the gap has widened over 25 years
Researchers at Steno Diabetes Center Aarhus, led by Dr. Tinne Laurberg, tracked cancer across Denmark’s entire adult population for 25 years. In 2010–2022, adults with type 2 diabetes had 12% (men) and 22% (women) higher overall cancer incidence than those without diabetes, and the gap had widened compared with earlier years.
The excess was largest for obesity-related cancers: 45% higher in men and 30% higher in women. Colorectal cancer was 17% higher in men and 14% higher in women, lung cancer 23–25% higher, and postmenopausal breast cancer 8% higher. Prostate cancer was 19% lower, a pattern seen in earlier research as well.
The authors’ conclusion is direct: “highlight the need to prioritise cancer prevention and early detection as part of long-term diabetes care.”
Source: Laurberg T, et al. Cancer incidence trends in people with and without type 2 diabetes in Denmark, 1996–2022. EASD Annual Meeting 2026 (abstract). EASD news release → · Medical Xpress coverage → · News-Medical summary →
Headline vs. Evidence
Before you forward a health headline, ask three questions: Was it a randomized trial or an observation? Did it measure a real outcome (a diagnosis, a heart attack, a death) or a number on a lab report? And who was studied? Does that include someone like you? Here’s how this week’s six stories score.
- GLP-1 stoppingHeadline: “Stopping Ozempic raises heart attack risk.” Evidence: In veterans with type 2 diabetes, stopping was associated with losing cardiovascular protection. Verdict: Credible signal. Plan any change; don’t stop abruptly.
- InflammationHeadline: “Stress is reshaping your heart.” Evidence: High inflammation markers tracked with 43% higher heart attack and stroke risk and changes on heart MRI. Verdict: Strong association; inflammation is measurable and modifiable.
- REM sleepHeadline: “REM sleep lowers risk of 83 diseases.” Evidence: A week of wrist-tracked sleep predicted later diagnoses; reverse causation is possible. Verdict: Promising, not proven. Protecting sleep is still the right call.
- Blood thinnersHeadline: “Blood thinners benefit Alzheimer’s patients.” Evidence: Only in people who also have AFib; a modest 0.2-point-per-year MMSE difference. Verdict: Not a dementia treatment. It is a reason to take AFib seriously.
- GalleriHeadline: “One blood test for 50 cancers nears FDA approval.” Evidence: A favorable panel vote; strong detection data; no mortality benefit shown yet. Verdict: A meaningful add-on to screening, not a replacement.
- Diabetes & cancerHeadline: “Diabetes raises your cancer risk.” Evidence: 12–22% higher incidence in a national registry; not yet peer-reviewed. Verdict: Consistent with prior research. Screening belongs in diabetes care.
The Long View
A blood test for 50 cancers, and what it can’t tell you
For most of the last century, cancer screening has meant looking for one cancer at a time, in one organ at a time. We screen the breast, the colon, the cervix, and, for people who smoked, the lungs. Those four programs save lives. But they leave out the cancers that cause most cancer deaths: pancreas, ovary, liver, stomach, esophagus. Most of them are found only after they cause symptoms, and by then often too late.
Multi-cancer early detection tests take a different approach. Tumors shed fragments of DNA into the bloodstream, and those fragments carry chemical “tags” (methylation patterns) that differ from healthy cells and hint at where they came from. One blood draw, dozens of cancers. This week, an FDA advisory panel judged that for one of these tests the benefits outweigh the risks.
Here is what a result can tell you. A positive result means cancer is likely, but not certain. In PATHFINDER 2, about 6 in 10 positives were confirmed, and the test usually pointed to the right organ, which guides the next scan. That is far better odds than most screening tests achieve on their own.
Here is what a result cannot tell you. A negative result does not mean you don’t have cancer. These tests miss a meaningful share of cancers, especially small, early ones that shed little DNA. We don’t yet know whether finding cancers this way helps people live longer; the large English trial did not reduce later-stage diagnoses on its primary measure. And every screening test carries the risk of finding things that would never have harmed you.
So where does it fit? On top of the fundamentals, never instead of them. In our practice, an MCED test makes the most sense for people whose risk is already elevated: adults over 50, people with type 2 diabetes, a strong family history, or a smoking history. It works best when it sits beside hereditary cancer genetics, a full biomarker picture, and a physician who knows what to do with either answer.
That is the thread running through all six of this week’s stories. A drug works only while you take it. Inflammation damages only if nobody measures it. A test helps only if someone reads it in the context of you. The technology is arriving. The interpretation is still the medicine.
Do This Week
- 1On a GLP-1? Make an exit plan before you ever need one. If cost, side effects, or reaching your goal has you thinking about stopping, talk to your prescriber first. Ask about tapering, a maintenance dose, and which labs to re-check. Meanwhile, lift weights at least twice a week and prioritize protein; muscle is the part of weight loss worth keeping.
- 2Ask for an inflammation number, not just a cholesterol number. A high-sensitivity CRP (hs-CRP) is inexpensive and widely available. If it’s elevated on two checks, treat it as a risk factor. Then work the levers that move it: smoking, visceral fat, sleep, activity, and chronic stress. See what our biomarker panels measure.
- 3Protect the last two hours of your night. REM sleep is concentrated late in the night, so a 5:30 a.m. alarm after a midnight bedtime cuts it first. Keep a consistent wake time, avoid alcohol within three hours of bed, and if you snore or wake unrefreshed, ask about a sleep apnea evaluation.
- 4Check your rhythm. Take your pulse for 30 seconds, or use your smartwatch’s ECG feature if it has one. A pulse that is irregularly irregular deserves an evaluation for atrial fibrillation, a silent and treatable threat to both the heart and the brain.
- 5If you have type 2 diabetes, add cancer screening to your diabetes checklist. Confirm you’re current on colorectal screening (from age 45), mammography, and lung CT if you have a smoking history. Then ask whether your personal risk justifies going further with early cancer detection.
Questions We Got This Week
Is it dangerous to stop Ozempic, Wegovy, or Mounjaro?
Stopping is not dangerous in the short term, but the benefits may fade. A BMJ Medicine study of 333,687 US veterans with type 2 diabetes found that continuous GLP-1 use was associated with an 18% lower risk of heart attack, stroke, or cardiovascular death than an older diabetes drug. Compared with staying on treatment, risk was 14% higher after one year off and 22% higher after two years off. Restarting restored protection only partly. The study was observational and included only people with diabetes. The safest approach is to plan any change with your prescriber, maintain muscle and activity, and re-check weight, blood pressure, lipids and glucose afterward.
My cholesterol is normal. Should I get tested for inflammation?
For many adults, it’s reasonable. A 2026 UK Biobank study of nearly 480,000 people found that those in the highest fifth for a blood inflammation marker (GlycA) had a 43% higher risk of heart attack and stroke than those in the lowest fifth, along with measurable changes in heart structure on MRI. The most common test, high-sensitivity CRP, is inexpensive. Inflammation is influenced by smoking, excess body fat, poor sleep, inactivity, and chronic stress, and it improves when those change. Your physician can interpret it alongside your lipids, Lp(a), and other risk factors.
Does REM sleep really lower dementia risk?
It’s associated with lower risk, but causation isn’t proven. In a PLOS Medicine study of 95,559 UK adults whose sleep was measured with wrist devices, each additional ~48 minutes of REM sleep per night was associated with a 46% lower risk of dementia and a 31% lower risk of Alzheimer’s disease over about nine years. Because early brain changes can disrupt REM, part of the link may run in reverse. The practical steps are the same either way: get enough total sleep (most risks were lowest at 6–8 hours), keep a regular schedule, limit evening alcohol, and get evaluated for sleep apnea if you snore.
Do blood thinners prevent Alzheimer’s disease?
No. They are not a dementia treatment for the general population. A Swedish registry study of 7,308 people who had both atrial fibrillation and Alzheimer’s disease found that those taking newer oral anticoagulants (NOACs) declined about 0.2 MMSE points per year more slowly than those on warfarin or no anticoagulant, and had fewer strokes, clots, and deaths. The likely explanation is prevention of small strokes caused by AFib. Blood thinners carry real bleeding risk and should be used only when a physician determines that stroke risk justifies them.
Is the Galleri multi-cancer blood test FDA approved?
Not yet. On September 23, 2026, an FDA advisory panel voted 7–2, with one abstention, that Galleri’s benefits outweigh its risks for adults 50 and older. The FDA makes the final decision, expected in the coming months. In the PATHFINDER 2 study of 35,878 adults, adding Galleri to standard screening detected about 6.5 times more cancers, with a 60.3% positive predictive value and a 0.36% false-positive rate. The test does not replace standard screening, a negative result does not rule out cancer, and it has not yet been shown to reduce cancer deaths.
Does type 2 diabetes increase cancer risk?
Yes, modestly, and the gap may be growing. A 25-year nationwide Danish study, presented at EASD 2026, found that in 2010–2022 people with type 2 diabetes had 12% (men) and 22% (women) higher overall cancer incidence than people without diabetes. Obesity-related cancers were 30–45% higher, and colorectal, lung, and postmenopausal breast cancers were also more common. Shared risk factors such as excess body fat, insulin resistance, and smoking likely explain much of the link. People with type 2 diabetes should stay current on colorectal, breast, and (if eligible) lung cancer screening.
Saw a headline that worried you?
Bring it to your next Everhealth visit. We’ll translate the study into what it means for your biology, your medications, and your plan.
Turn this week’s headlines into your personal plan
Cardiometabolic, brain, and cancer-prevention testing, interpreted by a physician in one integrated, numbered action plan.
Schedule an AssessmentKnow someone on a GLP-1, living with diabetes, or caring for a parent with AFib? Forward them this issue. · Next week: the EASD 2026 diabetes meeting, with new heart-outcome data on oral GLP-1s and the next generation of weight-loss drugs. · Missed Issue 09? Read it here.