fb-pixel Skip to main content

There’s a new frontier in detecting cancer: blood tests

Rosemary Jemo posed for a photo at the Health & Wellness Center at Hazleton in Hazleton, Pa.Sean McKeag for The Boston Globe

Rosemary Jemo felt fine when she volunteered for a study in 2018 that required her to give a blood sample. In fact, that was the point of the research — to see if scientists could detect multiple forms of cancer in the blood of people with no symptoms or history of the disease.

So Jemo, a 69-year-old hairdresser and exercise instructor in rural Pennsylvania, was surprised when the experimental test by a Cambridge biotechnology firm indicated she might be sick. She underwent a whole-body PET scan that revealed a malignant tumor the size of a football on an ovary.

“I had no idea,” said Jemo, one of nearly 10,000 women who took the test developed by Thrive Earlier Detection in a study conducted by Johns Hopkins University and Geisinger Health System of Pennsylvania. “I didn’t have any pain. I had nothing.”


Ovarian cancer is often deadly. There is no reliable screening test for it, and more than three-quarters of cases are diagnosed only after the cancer has advanced. Just 47 percent of patients survive five or more years following diagnosis.

Jemo, of Hazle Township, Pa., was fortunate despite the size of the tumor. She underwent surgery to remove it and hasn’t had a recurrence. Now 72 and still teaching exercise classes, she says Thrive’s multi-cancer detection test — which helped find 26 cancers of different types in the study — saved her life.

Thrive is among a handful of startups in the United States pursuing one of the hottest areas of oncology, blood tests intended to find dozens of forms of cancer before patients develop symptoms. Often called liquid biopsies, the gene-based tests home in on tumor DNA and other worrisome biomarkers that can signal cancer. If malignant tumors are found and treated earlier, the thinking goes, it could reduce the country’s 600,000 annual cancer deaths.


“All advanced cancers start out as not advanced, every single one,” said Dr. Bert Vogelstein, a world-renowned cancer researcher at Johns Hopkins who helped found Thrive, which licensed technology from his lab and received funding from the Boston venture capital firm Third Rock Ventures. “Every stage 4 cancer starts out as a stage 1 cancer. [But] once a cancer is advanced and metastatic, it’s extremely difficult to cure the patient.”

None of the blood tests to screen for cancer in asymptomatic people are widely available yet or have won approval of the Food and Drug Administration. Their makers hope that will change in the next few years. A bill introduced in Congress last year with bipartisan support would require Medicare to cover such screening tests, one of which is available now to patients who obtain a doctor’s prescription and pay $949 out of pocket.

Rosemary Jemo of Hazle Township, Pa., rode a stationary bike before leading a fitness class at the Health & Wellness Center at Hazleton.Sean McKeag for The Boston Globe

Some medical experts, however, are skeptical that costly routine screening of healthy patients is wise and say it’s premature to discuss federal health insurance coverage.

Dr. H. Gilbert Welch, a general internist and senior investigator at the Center for Surgery and Public Health at Brigham and Women’s Hospital, said researchers haven’t held large randomized clinical trials to determine if liquid biopsies would save lives. While it may be counterintuitive, he said, earlier cancer detection and treatment don’t necessarily prolong life with aggressive forms of the disease ― they often just lengthens the time between diagnosis and death.

Widespread screening could also backfire, said Welch, who has written extensively on the risks of excessive testing and over-diagnosis. Unintended consequences could include false positives that scare patients as well as detection and invasive treatment of cancers unlikely to progress. The latter concern has dogged other approved screening tests, such as the popular blood test for prostate cancer.


“I don’t want you to hear that liquid biopsy has no role. It may have a great role for people with established cancer” as an alternative to traditional and often painful tissue biopsies, said Welch. “The question is to what extent should we be testing the well to find something wrong? Because the truth is that in almost everyone, you can find something wrong.”

Still, the possibility that blood tests for multiple cancers could become as routine as Pap smears for cervical cancer has led to recent multibillion-dollar business deals.

Exact Sciences, a cancer diagnostics company in Madison, Wis., bought Thrive for $2.1 billion a year ago and is refining CancerSEEK, the test that detected Jemo’s tumor. Exact Sciences operates Thrive as a subsidiary and hopes to start a major clinical trial of the updated test by the end of the year. Results could be ready by 2025.

Grail, a San Francisco-based startup backed by Jeff Bezos and Bill Gates, entered the field before Thrive with a rival blood screen called Galleri, the $949 test that requires a prescription. More than 1,500 doctors prescribed it last year, according to a Grail spokeswoman. Grail recently began recruiting 140,000 volunteers in England for a clinical trial aimed at winning FDA approval. San Diego-based Illumina, the world’s largest gene-sequencing firm, has offered to buy Grail for $8 billion, but European regulators have held up the deal.


Last month Flagship Pioneering, the biotech venture firm that spawned Moderna, unveiled Harbinger Health, which is also developing a cancer detection blood test. Harbinger’s chief executive, Dr. Stephen Hahn, was FDA commissioner during the Trump administration and practiced oncology at MD Anderson Cancer Center in Houston. He said Harbinger’s blood test would use machine learning and artificial intelligence to pick up cancer signals soon after the disease starts.

“We’re pretty good at diagnosing cancer at stage 3 and stage 4,” he said. “We want to do it at stage 1 and stage 2.”

Currently, the only widely used blood test for early detection of cancer is PSA, for prostate-specific antigen. That test was introduced in 1994 and has become routine for men once they reach middle age. Today 90 percent of prostate cancers are detected while the disease is still confined to the prostate gland or nearby, when nearly 100 percent of men with the disease survive five or more years.

In recent years, several biotechs have created blood tests for patients already diagnosed with cancer as an alternative to taking a tissue biopsy to see if tumors have returned or are growing. These tests pick up in the bloodstream small pieces of DNA shed by multiple types of tumors, information that can also guide doctors about what medications would be most effective. The firms marketing them include Cambridge-based Foundation Medicine, a Roche subsidiary with an FDA-approved liquid biopsy covered by Medicare.


Now the focus has shifted to blood draws to screen for multiple cancers in seemingly healthy people. There are only a few FDA-approved tests of any kind to detect cancer in asymptomatic patients, including colonoscopies for colon cancer and mammograms for breast cancer.

A blood test that could flag multiple cancers would be a boon to public health and could create a US market of $30 billion to $100 billion a year, according to Wall Street analyst Vijay Kumar, senior managing director of Evercore ISI’s health care and technology team.

The early version of the Thrive test that flagged Jemo’s ovarian cancer was used on nearly 10,000 women ages 65 to 75 who were encouraged to continue with regular screenings such as colonoscopies and mammograms. If two blood tests detected a possible cancer, patients underwent the PET scan to look for tumors.

After one year, 96 cancers had been diagnosed. Usual screenings found 24; the CancerSEEK blood test helped find 26 others in various organs. The remaining 46 cancers were diagnosed because symptoms appeared or tumors were found other ways, such as an imaging test conducted for another reason.

The blood test generated false positives in about 1 percent of participants.

CancerSEEK missed many more cancers than it flagged, a deficiency acknowledged by a representative of the American Cancer Society in 2020 when study results were reported. Nonetheless, the test increased the number of cases detected through any screening to 50, more than half of the overall total of 96.

“More than 50 percent of the women had their cancer detected by screening methods instead of symptoms,” said Kevin Conroy, chief executive of Exact Sciences. “Today’s sobering reality is that 75 percent of people have their cancer diagnosed only after they present with symptoms.”

Welch, the Brigham cancer researcher, said he welcomes more clinical trials but asserts that aggressive screening doesn’t always mean better health. It can also result in invasive treatment of tumors unlikely to cause serious problems, as sometimes happens with current tests for prostate and breast cancers.

Jemo, the Pennsylvania exercise instructor, doesn’t buy that.

“The sooner you find out and the sooner you are treated, the better it becomes,” she said. “Good or bad, I would rather know.”

Jonathan Saltzman can be reached at jonathan.saltzman@globe.com.