A promising blood test aimed at early detection of cancer failed to reduce late-stage cancer diagnoses in a major clinical trial, the test’s maker, Grail, announced Thursday.

The results cast doubt on a developing field of screening for diseases that has generated enormous hopes and investment. The goal of the blood tests is to save and extend lives by detecting cancers when they can be more easily and successfully treated.

Grail’s test, known as Galleri, looks for minuscule shards of cancer DNA in the blood and for more than 50 types of cancer, according to the company. The test was briefly featured in a Super Bowl ad this month.

Galleri has been sold in the United States since 2021 and costs $949. Even though the blood test is not yet approved by the Food and Drug Administration, the company markets it under a regulatory loophole that permits sales of certain tests without review. Very few insurers cover the test, so most people pay for it out of pocket. Grail said it had sold nearly half a million of the tests so far, including more than 185,000 last year.

The company lobbied heavily for the U.S. government to cover the test under Medicare. The spending package passed by Congress and signed into law by President Donald Trump earlier this month authorized Medicare to cover cancer detection tests, and Galleri could be among those that qualify to receive the government’s approval.

Demand for tests to detect cancer and other diseases has grown in recent years, reflected by a surge of interest among companies and consumers. Exact Sciences offers a test, Cancerguard, that is similar to Galleri. And medical centers are promoting whole body scans with MRI or CT imaging to look for tumors.

Cancer researchers had been closely watching the Grail study, which the company ran in partnership with the National Health Service of Britain. The study’s size and design promised to rigorously help answer the question of whether such tests were effective at catching cancer early.

But the study results show that Galleri is “not a very good test,” said Dr. Richard Houlston, who studies cancer genetics at the Institute of Cancer Research in Britain. “This doesn’t support rollout within the American health care system,” he said.

Grail, which has its headquarters in Silicon Valley, announced the results of the study in a news release. The company plans to present more detailed results from the study at a cancer conference in the spring.

The study enrolled 142,000 healthy adults ages 50 to 77 in Britain and followed them for three years. People had their blood drawn three times, each spaced about a year apart. One group of participants’ blood samples were run through the Galleri test, and those people were referred for medical care if they tested positive.

The company compared the group that underwent Galleri screening with another group that didn’t get the test. The hope was that with Galleri, people would receive cancer diagnoses earlier, in Stages 1 or 2.

But recipients of the Galleri test did not show a significant reduction in the total number of cancers diagnosed at Stage 3, when the disease would have grown or spread near its original site, or Stage 4, when the cancer would have spread to other parts of the body, according to the company.

In a call with analysts Thursday, Grail executives said they were encouraged by other findings that were not the primary goal of the study, citing a decrease in Stage 4 cancers and an increase in Stage 3. That result could indicate that cancers were being found slightly earlier.

“It’s absolutely right to say we didn’t hit the primary endpoint,” Harpal Kumar, Grail’s chief scientific officer, said. “But what we did see was a very compelling clinical benefit here.”

Still, researchers said even if some cancers were found at Stage 3 instead of Stage 4, that would not necessarily mean treatment would be less onerous or that the patients would fare better.

Dr. Adewole Adamson, a cancer researcher at the University of Texas at Austin’s Dell Medical School, was adamant about the conclusion that should be drawn from the results.

The study failed, he said. “End of story.”

“If there was a slam dunk result, we would have seen it,” given how large the trial was, he said.

The study’s goal was to show a 20% reduction in advanced cancers among those who got the test. That did not happen.