Read our COVID-19 research and news.

Outlier. This bladder cancer patient's metastatic tumors (red arrows) vanished when she took an experimental drug (CT images taken before drug and at 3 months, 6 months, and 15 months).

G. Iyer et al., Science

Genome Sequencing Clears Up a Cancer Medical Mystery

Most experimental cancer drugs never make it to market because they don't help enough people in early clinical trials. But even in "failed" drug trials, researchers may find that a few patients see their tumors shrink dramatically. Since it’s not clear why some respond but most don’t, researchers typically shake their heads and move on. But researchers today report that by sequencing the entire genome of one outlier patient’s tumor, they learned why her cancer disappeared when she took an experimental drug that didn’t help others. That drug now has a new lease on life for this cancer, and such testing may help revive other cancer drugs that showed promise in lab studies but initially failed in clinical testing.

Researchers at Memorial Sloan-Kettering Cancer Center (MSKCC) in New York City were puzzled by the case of a woman with metastatic bladder cancer whose tumors vanished after she was given a drug, called everolimus, which targets a protein involved in cell growth known as mTORC1. Most patients in the trial weren't helped by the drug and it was abandoned as a single agent for bladder cancer. But this one patient has been cancer-free for 2.5 years—a "pretty unprecedented result" for this cancer, which resists chemotherapy, says physician-scientist David Solit of MSKCC and Weill Cornell Medical College in New York City.

Solit's group tested the woman's tumor for mutations in a few genes in the mTORC1 pathway that might explain the tumor's sensitivity, but didn't find anything. So in collaboration with bioinformatician Barry Taylor's lab at the University of California, San Francisco, Solit's group sent a sample of the woman's tumor to a commercial lab for whole genome sequencing.

By comparing the tumor genome to the woman's normal DNA, the researchers found mutations in two genes, NF2 and TSC1, which lab studies have suggested also lie in the mTORC1 pathway. Mutations in TSC1, but not NF2, also turned up in several other bladder cancer samples. Although TSC1 (tuberous sclerosis complex 1) hadn't been on the scientists’ radar, it made sense that this gene might be involved: People born with a TSC1 mutation later develop benign tumors. The researchers then searched for TSC1 mutations in 13 other patients in the same failed drug trial. Four whose tumors shrank had mutations in TSC1, but only one of nine who failed to respond had the change, the researchers report online today in Science.

Solit and others plan to screen for bladder cancer patients that have a TSC1 mutation in their tumor so they can further test everolimus and other mTORC1-targeted drugs in such patients. "Now we have a clear path forward," Solit says. And he thinks the same approach can be used in other clinical trials. "This is going to change the way we view these outlier cases," Solit says. Although researchers already knew that certain cancer drugs work on only patients with specific mutations in their tumors, whole genome sequencing could help identify more such genetic changes, he says.

"This is a great story," says cancer researcher José Baselga of Massachusetts General Hospital in Boston. Baselga, who has led a clinical trial of everolimus for breast cancer, now expects to test patients for TSC1 mutations to see if they explain why some responded better than others. The study also shows that finding drug sensitivity genes will require more than testing a tumor for just a few candidates, Baselga says. "We probably have to go deeper" and sequence entire genomes, he says.