Do selenium supplements raise cancer risk? A 15-year trial follow-up weighs in

The Selenium and Vitamin E Cancer Prevention Trial, known as SELECT, was the largest randomized trial ever run on whether a mineral supplement could prevent cancer. It enrolled 34,887 men age 55 and older in North America, randomized to 200 micrograms a day of selenium, 400 IU of vitamin E, both, or placebo, and it was stopped early, around 2008, for futility and concern about possible harm. A new analysis published this year asked a question the original trial was never big enough, or long enough, to answer well on its own: did selenium's long shadow include cancers and neurodegenerative disease that only show up after many more years.

The study

The paper, led by Marco Vinceti and colleagues, was published August 12, 2026 in the American Journal of Epidemiology. It is a post-intervention analysis, meaning it goes back to the men randomized in the original trial and tracks what happened to them long after the pills stopped, using Medicare claims through 2019 and the National Death Index through 2024, for 341,697 person-years of follow-up. Because assignment to selenium or not was randomized at the start, comparisons between the groups still carry much of the protection against confounding that randomization provides, even though the follow-up itself is observational.

The reason to look specifically at Hodgkin lymphoma, multiple myeloma and ALS is that non-experimental studies had previously hinted selenium overexposure might raise the risk of each.

What the numbers actually show

This is where the finding gets more careful than the headline version. For the three conditions under suspicion, the point estimates leaned toward higher risk with selenium, but the confidence intervals were wide and included 1.0, the value that means no effect:

• ALS: hazard ratio 1.20 (95% CI 0.52–2.77), based on 13 cases in the selenium group versus 10 in the reference group.
• Multiple myeloma: hazard ratio 1.28 (95% CI 0.73–2.26).
• Hodgkin lymphoma: hazard ratio 2.01 (95% CI 0.60–6.70), the largest point estimate of the three and also the least precise, built on 8 cases in the selenium group versus 6 in the reference group.
• Hematopoietic malignancies overall: hazard ratio 1.07 (95% CI 0.86–1.33).

None of those cross the bar for statistical significance. A hazard ratio above 1.0 means more cases showed up in the selenium group than the reference group, but when the confidence interval spans 1.0, the data cannot rule out that the true effect is zero, or even protective. With only single-digit to low-double-digit case counts for lymphoma and ALS specifically, that imprecision is expected, not a flaw in the analysis.

Two other results in the same paper were more clear-cut. Melanoma risk was significantly lower in the selenium group, hazard ratio 0.71 (95% CI 0.54–0.95). Colorectal cancer trended lower too, hazard ratio 0.86 (95% CI 0.67–1.10), though that interval also touches 1.0. Lung cancer showed essentially no change, hazard ratio 1.07 (95% CI 0.87–1.31).

Put plainly: 15 years on, this trial does not confirm that selenium supplementation raises the risk of Hodgkin lymphoma, multiple myeloma or ALS. It also does not rule it out for those specific, rare outcomes, because the trial was not statistically powered to catch effects that small in conditions that uncommon. What it does show with more confidence is a lower melanoma risk in the group that took selenium.

Getting enough selenium

Selenium is widespread in seafood, organ meat and grains, and content tracks the selenium level of the soil or water where the food was raised, which is why the same food can vary a great deal by source. Amounts below are per realistic serving, ordered by density:

• Halibut, cooked, 3 oz (85 g): 92 mcg
• Brazil nuts, 1 whole nut (about 5 g): 68 to 91 mcg, and this one varies enough by soil that a single nut can occasionally exceed a full day's need
• Tuna, canned in oil, drained, 3 oz (85 g): 63 mcg
• Sardines, canned, 1 can (about 3.75 oz): 45 mcg
• Shrimp, cooked, 3 oz (85 g): 42 mcg
• Shiitake mushrooms, cooked, 1 cup: 36 mcg
• Beef, cooked, 3.5 oz (100 g): 35 mcg
• Chicken breast, cooked, 3 oz (85 g): 35 mcg, 40 mcg with the skin
• Turkey breast, cooked, 3 oz (85 g): 26 mcg
• Cottage cheese, 1% milkfat, 1 cup: 20 mcg
• Egg, large: 15 mcg
• Brown rice, cooked, 1 cup: 12 mcg
• Oatmeal, instant, fortified, cooked, 1 cup: 12 mcg
• Whole wheat bread, 1 slice: 6 mcg
• Lentils, cooked, 1 cup: 6 mcg
• Spinach, frozen, cooked, ½ cup: 5 mcg
• Garlic, 1 clove: 1.2 mcg

How much you need

The US RDA for selenium, from the National Academies, does not differ by sex:

• Adults, 14 and up: 55 mcg
• Pregnancy: 60 mcg
• Breastfeeding: 70 mcg
• Children 1 to 13: roughly 20 to 40 mcg, rising with age
• Infants: a much smaller adequate intake, on the order of 10 mcg a day in the first months, met by breast milk or formula

The tolerable upper intake level for adults is 400 mcg a day from food and supplements combined. The SELECT trial's dose, 200 micrograms a day on top of diet, sat well under that limit and still produced the uncertain signals described above, which is part of why the findings matter: the ceiling most people are told to stay under does not, by itself, guarantee an absence of long-term risk for rare conditions.

Too little, and too much

Deficiency is uncommon in the US, where food is imported from many regions and soil selenium content varies less in the aggregate diet than it does in a single-source subsistence diet. Where it does happen, historically in parts of China with selenium-poor soil, it has been tied to Keshan disease, a form of heart muscle damage that develops alongside a specific virus, and Kashin-Beck disease, a joint disorder in children. Milder deficiency shows up as muscle weakness or pain, fatigue, and a weakened immune response.

Too much, called selenosis, starts with a garlic-like odor on the breath and a metallic taste, then nausea, diarrhea, fatigue, irritability and joint pain, which together affect the large majority of people who develop it. Continued excess brings hair loss, brittle or deformed nails, skin rashes, and in severe cases numbness or tingling. The most common real-world route to overshooting the limit is not diet but supplements, especially selenium yeast products and an over-reliance on Brazil nuts, whose selenium content is both very high and very inconsistent from nut to nut.

Tracking it in AIM

AIM does not track selenium as standard, so add it as a custom metric: give it a name, a unit (mcg) and a daily target of 55 mcg (60 mcg during pregnancy, 70 mcg while breastfeeding), and it is estimated on every meal from then on, pinned to your cards or left behind "more".

Estimates from a photograph carry real uncertainty for micronutrients, because content varies with soil, cultivar, storage and cooking in ways no image contains. Treat the week as meaningful and any single day as approximate.