That was the situation Angie Watts, 44, faced after she walked into a radiation oncologist’s office last June expecting to discuss the radiation therapy she was about to begin after a lumpectomy for breast cancer. Instead, Dr Timothy M Zagar of the University of North Carolina looked down at a sheet of test results and delivered some shocking news.
A genetic test showed she had inherited an alteration in a gene needed to repair DNA. Radiation breaks DNA, so the treatment might actually spur the growth of her cancer, he said. He urged her not to take the risk and to have a double mastectomy instead. “I’m not a betting man,” he said in a recent interview.
Shaken, Watts called Dr James Evans, a professor of genetics and medicine at North Carolina. He told her the opposite: The mutation she had was not known to be harmful, so he urged her to go ahead with the radiation. A group of doctors met but could not reach a consensus, so, Watts said, “they left it up to me to decide.”
Watts’ experience highlights an unsettling side to the growing use of genetics in medicine, particularly breast cancer care. Doctors have long been tantalised by a future in which powerful methods of genetic testing would allow treatments to be tailored to a patient’s genetic makeup. Today, in breast cancer treatment, testing of tumours and healthy cells to look for mutations has become standard.
But as Watts found out, “our ability to sequence genes has gotten ahead of our ability to know what it means,” said Eric Winer, the director of the breast oncology programme at Harvard’s Dana-Farber Cancer Institute.
The ambiguities and disappointments play out in two areas: There is genetic testing of patients to see if they inherited mutations that predisposed them to cancer, and there is genetic testing of cells from the cancer to look for mutations that drive the tumour’s growth — but if found often cannot be targeted by any drug on the market.
The ability to understand and interpret genetic tests will surely improve. But for now, what sounds like a simple test can leave patients with frightening information but no clear options or guidance for treatment decisions.
“The stakes are very high,” said Evans, the geneticist who counselled Watts. “You have inherently nuanced and confusing tests and widespread ordering and interpretation by doctors who aren’t really equipped to do so,” he said. “The situation is ripe for overinterpretation and misinterpretation.”
Oncologists say it still makes sense to order these tests, which identify mutations in as many as 100 genes. If they find mutations that greatly increase risk of cancer, that is valuable information. But patients need to be prepared for ambiguities. “Typically they are not,” said Dr Elizabeth Campbell, an oncologist and former director of the Duke Women’s Cancer Centre.
In Watts’ case, the mutation was a medical mystery known as a “a variation of unknown significance.” That means it does not destroy the gene’s function but may alter it — leaving the implications entirely uncertain.
With each gene being tested, there is about a 5% chance of finding a variation of unknown significance. So as more and more genes are tested, ambiguous changes can add up fast.
A recent article published in JAMA Oncology involving 897 women 40 and younger with breast cancer found that nearly all had tests for mutations in the BRCA1 and 2 genes, which can increase breast cancer risk. Most did not have known mutations. But the chance of having a mutation of unknown significance — 4.6% — was the same as the 4.5% chance of having a known and risky mutation in the BRCA2 gene.
Watts eventually decided to have radiation therapy and did well. But, she said, “It was scary. There are times I regret ever having genetic testing.” A couple of decades ago, breast cancer seemed to be on the leading edge of personalised treatments. The first precision medicine drug, Herceptin, was developed and approved for a subset of breast cancer patients in the 1990s. Yet now, as powerful new precision medicine drugs elicit striking responses in patients with other cancers — lung, colon, melanoma, blood, gastric — metastatic breast cancer patients have been left out.
“It’s like standing outside a candy store on Sunday when the store is closed, looking in the window,” said Dr Gabriel Hortobagyi, a director of breast cancer research at MD Anderson Cancer Centre in Houston.
There is no obvious reason breast cancer in particular should be so resistant to new therapies. But the situation is one of the starkest examples of the frustrating reality of precision medicine today. While labs can test for hundreds of genes that have been linked to cancer, and while the tests may find likely culprits, there all too often is nothing that can be done.
“As a concept, it is beautiful,” Hortobagyi said. “In practice, we face a number of obstacles. Most breast cancers have not just one but four, six, 10, sometimes 15 or 20 mutations. So which is the driver mutation and which are the passengers? That’s a tall order.”
Even if investigators have a good idea of which mutation to go after, there may be no drug that blocks it. Or there may be a drug being tested in a clinical trial but the woman is not eligible because, for example, she has had two rounds of chemotherapy and the trial’s rules say she can have no more than one. Or there is a drug that was approved for a different cancer, but it costs more than $100,000 a year and her insurer will not pay. Sometimes, a drug that works against a mutation in one type of cancer also works against that same mutation in another cancer, but sometimes it does not.
Dr Norman Sharpless, the director of the Lineberger Comprehensive Cancer Centre at North Carolina, estimates that perhaps one in 1,000 women with advanced breast cancer will benefit from using the approved and experimental drugs available today. “There are a few who benefit tremendously, but when patients come in expecting a cure, most are disappointed,” he said.
Heather Lynn Bowler’s experience was typical. Bowler, a 38-year-old office manager who lives in Raleigh, North Carolina, learned she had breast cancer in November 2013 and had eight rounds of intense chemotherapy and a lumpectomy that seemed to have taken care of it. Then, last April, her cancer came back, having spread outside her breast.
In May, Bowler started the first of 18 rounds of chemotherapy. During her first chemotherapy visit, her oncologist, Campbell, asked if she would be interested in having her tumour’s genes sequenced. A month later, the results were in. Bowler’s tumour had mutated genes, but none of the mutations could be attacked with today’s experimental drugs or with drugs being used in other cancers. “The results added nothing to her care,” Campbell said. “Nothing.”