Mind & body, in brief

Colorectal cancer became the leading cause of cancer death among Americans under 50 in 2023, up from fifth place in 1990, the American Cancer Society reported in January. [1]

Cases in under-50s have climbed about 2% a year since the 1990s, and 60% to 70% are found at an advanced stage. [1]

In March the society said cases in 20-to-49-year-olds are rising 3% a year, and that 45% of this year's diagnoses will be in people under 65, up from 27% in 1995. [2]

In England, 11 cancers are rising in people in their 20s, 30s and 40s, while smoking, drinking, exercise, red meat and fibre all got better or stayed flat over the same years. [3]

The only trend that matched the rise was excess weight, and it explains about 20 of every 100 extra bowel cancers; the other 80 are unexplained. [3]

Across three European cohorts with 1,369 early cases, higher body weight raised men's risk by a third per 5 BMI points, smoking by a quarter and alcohol by 15%, while exercise cut it by 29%. [5]

Rates have risen with each generation born since the 1960s, which points to something in daily life that changed around then. [5]

Up to a fifth of patients carry an inherited fault such as Lynch syndrome; about 80% of cases have no genetic explanation. [4]

Doctors now suspect the gut's bacteria; one study linked colibactin, a DNA-damaging toxin made by some E. coli, to colon cancer in younger patients. [4]

A polyp normally takes about ten years to become a cancer, so a 22-year-old's tumour would have started at about age 10. [1][7]

US screening now starts at 45 instead of 50; most countries start somewhere in the 40s or early 50s. [1][8]

People below the screening age often see several doctors before a diagnosis, and bleeding gets put down to piles. [7]

Why it matters. Watch whether the gut-bacteria and childhood-exposure studies now under way can explain the 80 extra cases in every 100 that excess weight does not. [3][4]

On 19 August Merck and Moderna said their personalised melanoma vaccine, intismeran, met the main goal of a final-stage trial. [14][11] The trial enrolled more than 1,000 people whose high-risk melanoma had been cut out; all got the immune drug Keytruda and some also got the vaccine. [13][11] Those who got both went longer without the cancer returning and were slower to see it spread to other organs. [12][13] Each dose is made from the DNA of that patient's own tumour, so no two patients get the same vaccine. [12][11] It is the first mRNA cancer treatment to succeed in a late-stage trial. [19]

Why it matters. Melanoma is the deadliest skin cancer; once it spreads, five-year survival can fall to between 16% and 35%. [11] The same companies are testing the same idea in lung, bladder and kidney cancer. [13] Worldwide about 330,000 people were diagnosed with melanoma in 2022 and nearly 60,000 died of it. [18][20]

Scientists take a piece of the removed tumour and sequence its DNA to find mutations that exist only in that cancer. [13][11] Software then picks the mutated proteins, called neoantigens, and writes them into an mRNA strand the immune system can learn from. [16][12] Earlier cancer vaccines aimed at proteins that healthy cells also carry, and the body's tolerance for its own proteins blunted them. [16] Of 442 melanoma vaccine trials counted up to August 2025, only 6.1% reached phase 3 and 22.9% were stopped early. [17]

Why it matters. Neoantigens skip the tolerance problem because they are not in any healthy cell. [16] Twenty years of vaccines aimed at shared tumour proteins produced few real responses in patients with advanced cancer. [16]

The companies announced the result in a press release and have not published the data. [15][13] The UK coordinating doctor, Heather Shaw of UCLH, said the size of the benefit is not yet known and that full results come at the ESMO cancer congress in October. [12] Cancer Research UK said the interim data still needs to be fully reviewed. [13] The vaccine still needs approval from regulators, and in May the US regulator refused a different melanoma treatment that had promising trial results. [13][11]

Why it matters. Shares in Moderna and Merck surged on the announcement, before anyone outside the companies had seen the numbers. [13]

The Carter Center said at the end of January that ten people caught Guinea worm disease in 2025, in South Sudan, Chad and Ethiopia, down from 15 in 2024. [23] In 1986 about 3.5 million people a year in 21 countries were infected. [23][22] The worm is swallowed as a larva in pond water, grows for about a year, then burns its way out through the skin. [22] There is no drug and no vaccine; the fall came from cloth filters, treating ponds and keeping cases away from water. [22][21] Only smallpox has been eradicated in people before. [23]

Why it matters. About 100 million infections are estimated to have been prevented. [23] The World Health Organization has certified 200 countries and territories free of the worm. [21]

Eradication has stalled on animals: 664 infected animals were reported in 2024, mostly dogs in Cameroon and Chad, and about 700 in 2025. [21][22] A dog or a person carries the worm for 10 to 14 months with no sign before it emerges. [24] Villages now tether dogs during the transmission season and pay cash rewards for reported worms. [21] Researchers testing a blood test found one antigen, DUF148, picked up 76.6% of hidden infections in Chadian dogs, but it also reacted to other worms. [24] Since 2023 the world counts as eradicated only when no worm emerges in people or animals for three years. [22]

Why it matters. Cameroon was certified worm-free in 2007 and got the worm back from Chad in 2019, so a country is only safe while its neighbours' dogs are. [21]

A trial in Israel randomly assigned 995 university students with psychological distress to 12 weeks of an AI conversation platform, face-to-face group therapy, or a waiting list. [25] Published in April 2026, it found the chatbot group's anxiety fell more than the group-therapy group's, by about 2 points on a 21-point scale, and depression fell more than in the waiting-list group. [25] Post-traumatic stress symptoms did not differ between the groups. [25] Students who felt a bond with the chatbot used it more and improved more. [25]

Why it matters. About one in eight people worldwide has a mental disorder and only a quarter of those who need care get it, the authors write. [25]

A meta-analysis in npj Digital Medicine pooled 39 randomised trials of mental-health chatbots published from 2017 to October 2025. [26] Across 7,401 people the chatbots cut depressive symptoms by a small-to-moderate amount, an effect size of 0.31, and anxiety by 0.28. [26] The effect on depression was larger in people who were clinically depressed than in healthy volunteers. [26] A separate review of 31 trials in 15-to-39-year-olds, covering 29,637 people, found similar sizes and judged the newest generative chatbots' results inconclusive. [27]

Why it matters. In Asia, eight trials with 921 people found a benefit for depression but none for anxiety or stress. [29]

In a review of nine chatbot studies in college students, up to 61% of participants dropped out. [30] Many commercial mental-health apps keep fewer than 4% of users active after the first month, the Israeli trial's authors note. [25] Repetitive content and technical problems were the main reasons young people quit. [27] The chatbots that worked best used cognitive behavioural therapy, daily check-ins and a friendly tone. [28][30]

Why it matters. Most of these trials lasted two to eight weeks, so nobody knows whether the benefit holds. [28]

The Korea Disease Control and Prevention Agency reported in January that 30.3% of people who had a sudden cardiac arrest outside hospital in 2024 got CPR from a bystander, up from 12.9% in 2014. [36] That is 2.3 times the rate of ten years earlier. [36] When a bystander did CPR, 14.4% of patients survived. [36] Bystander means anyone who is not an on-duty paramedic or medical professional. [36]

Why it matters. Survival after a cardiac arrest depends on the minutes before an ambulance arrives, and only a bystander is there for them. [37]

Qatar's national registry followed 1,238 adults whose hearts stopped outside hospital and whom paramedics tried to revive. [37] 17.8% survived to leave hospital and 13.5% did so with their brain largely intact, against 8.1% and 5.4% in the country's previous count. [37] Most arrests happened at home, 61.8% were witnessed and 42.4% got bystander CPR. [37] Being witnessed, getting a shock before hospital, and having a blocked artery opened within 24 hours each predicted survival. [37]

Why it matters. Survival to discharge after cardiac arrest is 7% to 11% in most places and has barely moved in 30 to 40 years. [37]

Among 9,081 adults treated for cardiac arrest at 42 Japanese emergency centres from 2019 to 2021, 61.9% were men. [38] Bystander CPR went with higher 30-day survival in both sexes: 11.0% against 4.5% in men and 5.4% against 3.4% in women. [38] After adjustment the link stayed strong in men but was not statistically significant in women, and the difference between the sexes was itself significant. [38] The authors say the reasons need investigating. [38]

Why it matters. The authors say the difference needs explaining, and that bystander CPR should be delivered equally to both sexes in the meantime. [38]

Germany's resuscitation registry holds 68,719 cardiac arrests from 2013 to 2023; 316 were caused by drowning, half of them in June, July and August. [39] Drowning patients were younger, median 50 against 72, and colder, with a body temperature of 31.1 degrees C against 35.8 at the scene. [39] Overall survival was no different from other arrests. [39] But children under 11 who drowned survived 44.7% of the time, against 16.3% for children whose hearts stopped for other reasons. [39]

Why it matters. Drowning is one of the five leading causes of death for 1-to-14-year-olds in Europe, and most of it is preventable. [39]

Singapore has trained 17,800 volunteers and given them a credit-card-sized device that shows the depth and speed of their chest compressions. [40] Even with it, only 36% delivered good-quality CPR in a simulation, and many said they feared pressing hard enough to do harm. [40] A registered randomised trial will now stream the card's data and a video call to the ambulance dispatcher, who can coach the volunteer as it happens. [40] The control group gets the usual voice-only instructions. [40]

Why it matters. Good CPR raises survival 3.9 times over poor CPR, the protocol says, and over half of volunteers in one survey had ignored an alert. [40]

A review of cohort studies finds that four or more hours a week of moderate-to-vigorous activity, or good measured fitness, goes with a 40% to 50% lower risk of sudden cardiac death. [41] Sudden cardiac death is an unexpected death from the heart within about an hour of symptoms, and it causes 15% to 20% of all deaths worldwide. [41] In Kuopio, Finland, 1,089 men measured in the 1980s were followed for a median of 31.2 years. [42] Fitter men died of heart disease less often, and fitness divided by waist size predicted death better than either measure alone. [42]

Why it matters. No randomised trial can assign people to be unfit for decades, so the link rests on counting, and the reviewers say the causal question is still open. [41]

A meta-analysis of 51 studies of passive heating, in saunas and hot baths, found a single session lowered average blood pressure by about 5 mmHg. [31] Repeated sessions lowered the top number by about 5 and the bottom number by about 3. [31] Markers of inflammation and artery stiffness did not change. [31] A second meta-analysis of 20 hot-water immersion studies found one session raised the heart rate by 28 beats a minute and cut the bottom number by 5 mmHg, but called the overall benefit inconclusive. [33]

Why it matters. The reviewers say the size of the effect was much the same whether people were healthy or ill, young or old. [31]

A Finnish sauna runs at 80 to 100 degrees C and pushes the heart rate to 120 to 150 beats a minute, the level of moderate exercise. [34] Blood flow to the skin can rise to 7 to 8 litres a minute. [35] A 2025 review of sauna studies in heart disease found the observational studies consistent and the randomised trials mixed, and asked for standard protocols and longer trials. [32] In 28 healthy young women, three ten-minute sauna rounds raised the top blood-pressure number in the first round and lowered it across the rounds. [35]

Why it matters. The review says the strongest case is for people who cannot exercise, and that the trials do not yet prove it. [32]

Scabies is a mite that burrows into the skin; about 207 million people have it at any time. [43] In 28 French hospitals, 289 households with a confirmed case were randomly given either ivermectin pills or permethrin cream for everyone in the home, on day 0 and day 10. [43] At day 28, 88.5% of cream households were cured against 71.8% of pill households. [43] A second French trial gave 132 adults with severe scabies a double dose of ivermectin or the standard dose, both with cream; cure was 75% and 82%, so the higher dose did no better. [44]

Why it matters. Ivermectin is easier to give to a whole village, which is why mass treatment programmes use it; these trials say the cream, applied head to toe, still cures more. [43][46]

Turkish doctors followed 230 children whose scabies had survived at least one course of permethrin cream. [45] Families chose, after cost counselling, between adding two doses of ivermectin or continuing the cream alone. [45] Adding the pills cut the time to halve the itch from 9.8 days to 7.2, raised the four-week cure rate from 66.0% to 83.8% and halved recurrence at three months. [45] Until March 2023 ivermectin was sold only in 20-tablet import packs costing 28% to 37% of a month's minimum wage; a generic cut that to 5% to 7%. [45]

Why it matters. The same four tablets cost under 1% of the minimum wage in Australia and under 0.3% in India, so which drug a child gets depends on where the child lives. [45]

A meta-analysis of ten randomised trials with 1,105 patients compared oral ivermectin with benzyl benzoate lotion, an old and cheap treatment. [46] Cure rates were the same at one week, at two to four weeks and after four weeks. [46] The lotion caused far more burning and stinging; ivermectin had about a quarter as many side effects. [46] Scabies was added to the WHO's list of neglected tropical diseases in 2017, and permethrin, the usual first choice, is failing more often. [46]

Why it matters. Where permethrin fails or is unaffordable, the cheap lotion works as well as the pill, at the price of stinging. [46]

Restless legs syndrome is an urge to move the legs in the evening that spoils sleep; it affects 2% to 5% of people badly enough to matter. [49] For years the first drugs given were dopamine agonists, the class used in Parkinson's disease. [47][49] In 2025 the American Academy of Sleep Medicine downgraded their long-term use because of augmentation, in which the drug makes symptoms start earlier in the day, get stronger and spread to other parts of the body. [47][63] Gabapentin-type drugs are now the mainstay, and most patients on the old drugs end up escalating doses and stopping all medication within ten years. [47][48]

Why it matters. A drug that worked at first and then made the illness worse was the standard treatment. [47]

Iron shortage is found in 25% to 40% of people with restless legs, and topping it up is the first-line treatment. [48] Guidelines say to give iron when ferritin, the blood's iron-store reading, is 75 micrograms per litre or below, and moderate evidence supports iron by drip. [53] A 2025 meta-analysis of 31 studies found iron by drip beat iron by mouth in kidney disease, bowel disease and cancer, but found no difference for restless legs. [50] Where iron is not the cause, treatment depends on severity and on whether another illness, such as kidney failure, is behind it. [54]

Why it matters. More than 160 genetic risk spots have been found and no single gene explains the illness. [49]

A meta-analysis found restless legs in about 15% of people with inflammatory bowel disease, against about 3% of adults in general. [51][63] Bowel disease causes iron loss and long-running inflammation, both linked to the syndrome. [51] In Latin America, a survey of 69,830 Mexican women found 15.6% had the syndrome, 54.5% of them with frequent symptoms, and only 18.1% were treated. [52] Iron testing is limited in many countries and the newer first-line drugs are hard to get, so dopamine agonists and sedatives are still widely used. [52]

Why it matters. The treatment a person gets depends on which drugs a country's pharmacies stock. [52]

Researchers tattooed the footpads of mice with black, red and green ink and watched it drain through the lymph vessels within ten minutes. [55] Within a day it had piled up in the nearest lymph node, and after two months the node was still inflamed. [55] Human lymph-node biopsies from tattooed patients showed the same pattern of ink. [55] In the mice, ink at the injection site lowered the antibody response to an mRNA COVID vaccine and raised it to an inactivated flu vaccine. [55] This is a mouse result; nobody has tested vaccine responses in tattooed people. [55]

Why it matters. About 29% of Swedish adults have a tattoo, and most got the first one between 18 and 35, so the ink sits in the body for most of a life. [61][58]

A Swedish case-control study of people diagnosed at 20 to 60 found tattooed people had a 29% higher rate of melanoma, and it was not higher for bigger tattoos. [58] A Dutch nationwide registry found 94 melanomas on tattoos among 142,492 melanomas from 1991 to 2023 and concluded tattoos are unlikely to raise the risk. [57] For lymphoma, a meta-analysis of four studies with 17,941 people found no link, an odds ratio of 1.01. [56] An earlier Swedish study had found a link with lymphoma, and a Danish twin study a link with skin cancer. [60]