Will Personalized Medicine Help or Harm Us?

Sian Abrahams’ mother was only 46 when she died of breast cancer. A few years later, her father died of bowel cancer at the age of 76. Her half-brother developed skin cancer. When Abrahams’ older sister was diagnosed with cancer of the peritoneum — a rare cancer closely linked to cancer of the ovaries — Abrahams sought out her family practitioner, who referred her to a geneticist.

He carried out a test on Abrahams and her sister to determine whether they carried mutations on the BRCA1 or BRCA2 genes, both of which carry a substantially increased risk of breast cancer and other cancers in women. The results didn’t arrive for months. And then, at that first appointment, based on her family history alone, Abrahams was told she had a one in two chance of developing cancer.

“That’s when I made the decision to have the hysterectomy,” Abrahams says. She found a surgeon willing to perform an operation to remove her ovaries, cervix, and uterus, sending her straight into menopause. “Because I was only 38, people would say, ‘Do you not want anymore children? You’re still quite young.’ But for me, it was about staying alive for the child that I had.”

Abrahams’ surgery was 11 years ago. But the decision she faced — of whether to act on a set of risk factors based on her genetic profile — is one that may, in time, be faced by all of us. Personalized medicine, also known as precision medicine, is improving scientists’ understanding of how to tailor treatment to an individual patient and how an individual’s genetic makeup might predispose them to particular diseases. A small blood sample is now all it takes to sequence a person’s entire genome. The sequencing can be carried out in a day, though analysis is a more lengthy process.

Research to understand the role genes play in disease is well underway. The U.K. has a project to map the genomes of 100,000 individuals, comparing them to a reference genome to identify potentially harmful genetic variants. Across the world, biobanks, which include human tissue samples from thousands of people, along with anonymized (or de-identified) health care records, are making it possible to investigate the link between particular genes and conditions such as cancer, Type 2 diabetes, and heart disease. The numbers involved are huge: the National Institutes of Health’s All of Us biobank, for example, plans on collecting samples and health records from 1 million volunteers.

“Yes, you can find out about breast cancer, but for about the same amount of money, you can find out all sorts of other things that might be useful to your health.”

Studies on this scale enable researchers to spot genetic variations that might have a small impact, says Janet Olson, an associate consultant at the Mayo Clinic in Minnesota and a specialist in breast cancer. “We’re looking for things that are not even doubling risk — perhaps just 10 percent of an increased risk,” she says. “So we need a lot larger population to be able to tease out those differences between those with that small genetic variant and those that do not have it.”

There have already been some notable discoveries, particularly in the area of pharmacogenomics, which looks at how patients with particular genetic profiles respond to different drugs. It used to be the case that patients with chronic myeloid leukemia, which affects more than 8,000 patients a year in the United States, were treated with the long, complicated, and painful procedure of a bone marrow transplant.

That all changed, says Mark Lawler, chair in translational cancer genomics at Queen’s University Belfast in Northern Ireland, when an important discovery was made: “All patients with chronic myeloid leukemia have the same acquired genetic abnormality in their tumor cells.” A drug called Gleevec, which Lawler describes as the “poster child of personalized medicine,” was developed to target that abnormality. It has been hugely successful, and now 98 percent of patients are still in remission five years after treatment.

Olson cites the work of her colleague Bob Diasio, director of the Mayo Clinic Cancer Center. Using the samples from the Mayo biobank, Diasio found that a particular variant of the DPYD gene affected the way a certain commonly prescribed chemotherapy drug was metabolized. It means, Olson says, that Mayo clinicians can advise patients with this variant, “If you ever get cancer and are thinking of being prescribed this very common chemotherapy drug, don’t get it, because it could kill you — your body can’t handle it.”

In these cases, the decisions are usually straightforward: If a sick patient has this genetic variant, then they can be treated with this drug. But what happens when we are talking about genetic predisposition?

It could be a long way off, but Ruth Chadwick, emerita professor at the School of Social Sciences at Cardiff University and chair of the U.K.’s Human Genome Organisation Committee on Ethics, Law, and Society, believes it is possible that in the future, everyone will have their genome sequenced at birth.

It’s a tantalizing prospect. Many of us love to monitor our own health through devices such as Fitbits and Apple Watches, so what’s not to like about a personalized genetic profile showing what conditions we may or not be predisposed to in later life?

Quite a lot, it turns out. First, there is the obvious risk that it could allow insurance companies to discriminate on the basis of genetic profiles.(Currently, U.S. legislation prohibits discrimination by health insurers or employers on the basis of genetic profiling, but the legislation doesn’t apply to life insurers.)

There are also more nuanced risks. Imagine knowing that you have a predisposition to Alzheimer’s, for example, but being unable to do anything about it. It could also affect your relationship with others, Chadwick points out: “You might want to know what your genetic predispositions are, but your sister may not, and you may share the same genes, so that can lead to family tensions.”

On the other hand, surely being aware of your genetic predisposition could help reduce your risk by taking preemptive action? Unfortunately, the hoped-for clarity about the precise relationship between particular genetic variants and the risk of developing a disease has not been forthcoming.

Jason Torres, who researches Type 2 diabetes at the Wellcome Centre for Human Genetics, says that for many diseases, including Type 2 diabetes, rheumatoid arthritis, and schizophrenia, the relationship is complicated. “There are hundreds, if not thousands of individual genetic variants that increase risk for complex diseases and traits. Each of them has a very small effect and by itself doesn’t do much, but when you take it all together, that’s where the risk comes in.”

It seems that the hope of a better understanding of genetics solving many of our health problems is receding. Even the BRCA1 mutation isn’t a clear-cut indicator that a woman will develop breast cancer — Olson points out that some women with the mutation may never develop the disease, yet we don’t know why. “The more that I know, the more that I know that I don’t know.”

Kyle Brothers, an associate professor at the University of Louisville, has written about the ethics of personalized medicine and agrees. He cautions that we should be aware of the limitations of genetic profiling. “Yes, you can find out about breast cancer, but for about the same amount of money, you can find out all sorts of other things that might be useful to your health.”

So, what of Abrahams? Sadly, her sister died, like their mother, at the age of 46. Yet there is a twist to their story: When the results of the genetic test came back, it showed that neither sister had the BRCA1 or BRCA2 mutation.

Abrahams is nonetheless grateful that the risk of some major cancers in women has now been removed completely. As she approaches her 50th birthday, Abrahams has no regrets: “For me, it was quite black-and-white. I wanted to stay alive to see my son grow up. Nothing else mattered.”

This article was first published on Medium.