Immune cells (orange) may more readily attack tumor cells (brown) with many protein-altering mutations.

STEVE GSCHMEISSNER/SCIENCE SOURCE

Mutation-counting blood test could predict if cutting-edge immunotherapies can beat a cancer

Some cancers generate the seeds of their own destruction. Certain random mutations that accumulate in rapidly dividing tumor cells can spur the immune system to attack the cancer. Researchers are now learning that the extent of such mutations can predict whether a cancer will respond to new, powerful, immune-based therapies. A recently unveiled blood test for this so-called tumor mutational burden (TMB) could help make it a practical tool for guiding cancer treatment.

Cancer researchers can already gauge TMB by sequencing a panel of select genes in biopsied tissue, an approach that recently demonstrated strong predictive power in a large lung cancer trial. Some cancer physicians now use tissue TMB tests in select cases. But the less-invasive blood test, which analyzes tumor DNA shed into a person's circulation, could reveal TMB in the many patients where tissue testing doesn't work. "We'll see [TMB] more and more," says Naiyer Rizvi, an oncologist at Columbia University Medical Center. Still, TMB testing currently takes too long for routine clinical practice, he adds, and some in the cancer field question how useful it'll ultimately prove.

Tests that can predict whether immunotherapy will work in a patient are badly needed, especially for so-called checkpoint inhibitors, which release a brake on immune cells and enable them to attack tumors. Since the Food and Drug Administration (FDA) in 2014 approved the first antibody drug targeting the "checkpoint" protein called PD-1, these drugs have transformed cancer care. University of California (UC), Los Angeles, research oncologist Antoni Ribas notes that in May, half the cancer patients admitted to his hospital had been on checkpoint inhibitors in the previous 6 months. "It's a remarkable thing that we're using these agents so much," he says. In some patients the response is dramatic, but most still don't benefit, and others are never prescribed the drugs. And except for the 4% of patients whose tumors have a specific DNA repair defect, doctors cannot reliably tell who will benefit.

Enter the TMB tests. Most assays estimate the number of protein-altering mutations in a tumor by sequencing a limited number of genes from its DNA; that tally likely reflects the density of mutant protein fragments, known as neoantigens, on the surface of cancer cells. Such fragments aren't helping the tumor grow; they're just a byproduct of error-prone tumor cell division. But they do appear foreign to the immune system—and the more neoantigens, the more likely that immunotherapy will shrink the tumor and keep it at bay.

In the lung cancer trial, which was reported in April at the American Association for Cancer Research (AACR) annual meeting in Chicago, Illinois, researchers found that mutational load in tumor tissue predicts whether a checkpoint inhibitor combination will help lung cancer patients more than standard chemotherapy does. More than 40% of lung cancers showed a high TMB, and the patients with those tumors, on average, did much better on the immunotherapy. Rizvi says the phase III trial of 1739 patients should lead to FDA approval of the tissue-based test, which was developed by Foundation Medicine, a company in Cambridge, Massachusetts, for use in lung cancer. (In mid-June, Swiss pharma giant Roche agreed to acquire the company.)

More evidence for the predictive value of TMB emerged at the annual meeting of the American Society of Clinical Oncology (ASCO) this month in Chicago. UC Davis oncologist David Gandara reported a retrospective analysis of seven different trials of the checkpoint inhibitor Tecentriq in lung and bladder cancer, melanoma, and other tumors. When the TMB was high, as shown by the same tissue test, the tumor response rate to the drug doubled. "The future is now for TMB," Gandara said at the ASCO meeting.

Tissue TMB testing, however, "is very expensive, it requires a lot of tissue, and it's not standardized," says Yale University pathologist David Rimm. In the trial reported at the AACR meeting, doctors only got enough tumor tissue from 58% of patients. Rizvi adds that the whole process can take 3 weeks, too long to wait for newly diagnosed patients.

The blood TMB test, also from Foundation Medicine, may prove just as effective as the tissue test. At the ASCO meeting, Vamsidhar Velcheti of the Cleveland Clinic in Ohio reported early results from a prospective trial of Tecentriq in lung cancer patients who took a blood test for TMB. The drug shrunk more than 36% of tumors that had a high mutational load but only 6% of low-TMB tumors. Patients with high-TMB tumors went three times longer without their cancer growing back that those with low-TMB tumors did.

Velcheti only reported on the first 58 patients, making any conclusion tentative, Hossein Borghaei, an oncologist at the Fox Chase Cancer Center in Philadelphia, Pennsylvania, cautioned at the meeting. A 580-patient trial is underway. Rimm agrees the initial results need validation. "They're just doing pilot studies and saying, ‘Wow, look what we've found.' And it is cool what they've found."

In April, FDA designated the blood TMB test a "breakthrough device" that merits a priority review. But whether from blood or biopsies, it's not clear TMB will give doctors and patients the outcomes or certainty they crave. Rimm points out that trials haven't yet shown that high-TMB patients live longer on immunotherapy than on chemotherapy. And Ribas predicts TMB "will be one component" of a future combination biomarker.