The gold standards for cancer detection still have many flaws. Cancer researcher Jimmy Lin has developed a blood test that promises faster, less invasive, and more specific results. Research in cancer patients could lead to trials on healthy people. The best way people can maximize their chances of surviving most cancers is to catch them as early as possible, and Jimmy Lin thinks he’s found the new gold standard in detection: a blood test.

Lin is a cancer genomics researcher, which means he looks at how cancer works on a genetic level. In April, he gave a TED talk that shed light on the shortcomings of the three leading cancer detection methods.

The first, medical procedures, are highly invasive. The second, testing for protein biomarkers, can produce a lot of false positives. And the third, imaging, can expose people to harmful levels of radiation, and it doesn’t always work for everyone.

For the past few years, Lin has been interested in the idea of a non-invasive, highly specific, and widely applicable test that could find cancer early on.

“Does such a technology exist?” Lin asked the TED crowd. “Well, I wouldn’t be up here giving a talk if it didn’t.”

Detecting trace DNA

The solution currently being developed by Lin and his team at the biotech company Natera is a blood test that can screen for all types of cancer. They’ve already found promising results. In his talk, he described a study in which doctors monitoring lung cancer patients in remission were able to use the blood test to catch the cancer’s reccurence 100 days before traditional methods like a CT scan. That 100-day window, Lin said, could mean the difference between life and death.

The week after he gave his TED talk, Lin and his colleagues published the full study in the journal Nature. (Since the study hadn’t been published, he couldn’t talk about it in full at TED.)

The study showed that out of 24 cancer patients studied over time, the blood test picked up their cancer DNA 93% of the time and produced no false positives. The test also managed to detect cancer reccurence as early as 11 months before traditional methods.

“Lung cancer, when it recurs, is a pretty aggressive disease,” Lin told Business Insider. Even the earliest stage, Stage IA, carries about a 50-50 chance of survival over a five-year period. For the worst stage, metastatic lung cancer, the five-year survival rate drops to 1%.

Lin conceded the data aren’t perfect, even if the test seems to work. His future studies will increase the sample size to hone in on the specificity of the test — or the ability for it to pick up one kind of cancer and not just cancer DNA in general.

As the company pursue that goal, Natera will make the blood test available for research use on cancer patients later this fall. At some point in 2018, armed with better data about which cancer varieties the test screens for best, Natera plans to launch a cancer-specific test for clinical use.

An even bigger jump, Lin said, is moving from testing cancer patients in remission to testing healthy patients. When doctors screen people that have had cancers with a possibility of recurrence, the chances are much greater of putting the test to work. But in healthy people it becomes more of a shot in the dark, and researchers can’t be as sure they’re not getting a false positive.

Decades down the line, Lin hopes patients will be able to learn if they have cancer in the same way they learn if they have high cholesterol. The presence or absence of cancer DNA would be just another piece of data on the long printout of bloodwork.

In the meantime, Lin advises people to rely on the current methods. Methods like colonoscopies and mammograms aren’t perfect, but they still reduce cancer-related risk of death dramatically.

“Those adoption rates could be higher,” Lin said. “So even with existing technologies, people should have this done.”