Betty Lane didn't have a whole lot of options left. Diagnosed with stage 3b lung cancer in the summer of 2011, she underwent surgery and chemotherapy, which drove away her disease for about six months. But then it returned, this time in her liver. The Fayetteville, Tenn., psychotherapist enrolled in a drug trial, but her cancer didn't respond.
Then Lane's doctor at Vanderbilt-Ingram Cancer Center in Nashville suggested that they try a drug used to treat breast cancer. Lane's lung tumor had revealed a glitch in a gene more commonly abnormal in breast and gastric tumors, resulting in a mutant form of a protein targeted by the drug Herceptin. "I feel very fortunate," says Lane, 55, whose cancer is now shrinking, "to live in an age and a time when this is happening."
Lane is a prime example of the promise of personalized medicine, which matches treatments to patients based on their genetic and other biological information. "There are now over 80 personalized treatments available," says Edward Abrahams, president of the Personalized Medicine Coalition, an education and advocacy organization. Herceptin and Gleevec, approved in 2001 to treat a certain type of leukemia, are two of the oldest. Just this past May, the Food and Drug Administration approved two drugs for advanced melanoma driven by certain mutations that join a crop of new therapies approved in the last few years. Also, in May, the Mayo Clinic launched an Individualized Medicine Clinic at its three locations in Minnesota, Arizona and Florida aimed at getting cutting-edge guidance to cancer patients who have failed standard treatments and to people with mysterious ailments that may have a genetic cause.
There are obstacles along this road, to be sure. While the National Institutes of Health says the cost of sequencing a genome is now about $5,700, compared to $95,000 a dozen years ago, all that information has to be analyzed and interpreted, and somehow made accessible to physicians and patients. And identifying the precise cause of a disease is no guarantee that there's a treatment. If there is, it's apt to be very expensive, since the cost of drug development must be spread out over a smaller number of patients.
Still, the beachheads are being established, most notably in cancer. Vanderbilt-Ingram, Massachusetts General Hospital, Memorial Sloan-Kettering Cancer Center in New York and MD Anderson in Houston all now regularly test patients for certain mutations. At Vanderbilt-Ingram, all lung-cancer patients are tested for more than 40 glitches in 10 different genes, and melanoma patients are screened to see if they have one of 43 mutations in six genes, says William Pao, director of the hospital's division of hematology and oncology. Vanderbilt also tests colon and breast tumors to see if patients are eligible for trials of experimental drugs targeting particular mutations.
Researchers at Washington University School of Medicine in St. Louis took more drastic measures in 2011 to help their colleague Lukas Wartman. The Wash U oncologist and leukemia researcher was fighting leukemia himself, and his prognosis after a second relapse was grim. Rather than just screen his cancer for the small number of known genetic changes that occur with the disease, the researchers sequenced the entire genome of the cancer's DNA as well as that of Wartman's own normal cells, looking for any changes that could be driving the cancer. They also sequenced the RNA from his cancerous cells.
What they found, says Wartman, was that a certain gene called FLT3 was unexpectedly overproduced by his cancer cells. It so happens that a drug approved for kidney cancer inhibits that gene, and within days, Wartman started taking it. The drug sent his leukemia into remission, though he eventually had to call a halt because it also lowered his blood cell counts, a side effect. Now "I'm in uncharted waters," he says, adding that he remains in remission and is weighing the risks and benefits of taking the drug again.
Wartman points out that this is very much an emerging technology. "We've gone from a kind of Model-T Ford to a Maserati in just a few years," he says. But his case is a "very clear example" of personalized medicine's potential power, he says. He predicts that the next decade will usher in a whole new era of tailored cancer treatments.
Most will probably not be miracle cures. For many patients taking the new targeted therapies, the added benefit is measured in months; cancer cells can eventually find their way around drugs that target a specific glitch. But for others, a reprieve lasts for years. The 10-year survival rate for chronic myeloid leukemia is now above 80 percent, compared to 20 percent in the pre-Gleevec days. The hope is that different drugs could be prescribed in combination or succession and turn cancer into a chronic disease.
On other fronts, doctors are using genetic information to predict how a patient will respond to a drug. The HIV drug abacavir, for example, can cause a potentially deadly reaction in 5 to 8 percent of patients, so they get a genetic test to see if they should take the drug. Tests to gauge whether a person will get the full benefit of the anti-clotting drug Plavix, to figure out the best dose of the blood thinner warfarin, and to see who's likely to experience muscle pain or weakness on the anti-cholesterol drug simvastatin are also available.
Figuring out who will experience drug toxicity is key, because rare side effects have the potential to sharply curb the use of effective drugs. Clozapine, for example, is an antipsychotic that works but is a last-resort drug due to a dangerous side effect affecting the white blood cells, says Jeffrey Lieberman, chairman of psychiatry at the Columbia University College of Physicians and Surgeons. Researchers are now searching for genetic signatures that can predict who is at risk.
Increasingly, too, the roots of baffling diseases are being uncovered by peering deep into the genome. At the Medical College of Wisconsin and the Children's Hospital of Wisconsin, for example, researchers most famously identified a unique mutation in the DNA of young Nic Santiago Volker, who developed a crippling disease as a toddler that ravaged his digestive tract and required the removal of his colon. From 2007 until 2010, pretty much all of Nic's nutrition was delivered intravenously, recalls his mother, Amylynne Santiago Volker.
In 2009, when Nic was 4, sequencing of the 1 percent of his genome containing instructions for making proteins found a mutation linked to an immune-system disorder that doctors suspected was also responsible for his GI illness. In 2010, Nic was treated with a cord-blood transplant, a treatment for the immune disorder. A few months afterward, he was able to eat. "He got a steak first. That's what he wanted," his mother says.
The transplant has by no means solved all of Nic's problems. He developed encephalitis and has a seizure disorder, his mother says. He continues to grow more slowly than his peers. He will need more reconstructive surgeries. But for now, at 8, he is active in a way his mom couldn't have imagined just a few years ago.
"We're completely amazed," she says. "Before, he couldn't walk, climb or jump. Now he has more energy than the entire household." Howard Jacob, director of MCW's Human and Molecular Genetics Center, says 19 children have now had their genomes sequenced, leading to five different diagnoses.
Unlike Nic's mysterious programming, the gene controlling cystic fibrosis, CFTR, has been known since 1989. The disease is caused by a variety of mutations in CFTR, which affects salt and water flow in and out of cells, and impacts various organs including the lungs and pancreas. Vertex Pharmaceuticals has developed a drug targeting one of those mutations and is studying whether, used in combination with other drugs, it could also provide a boost to patients with different genetic mistakes.
"I noticed a difference the next day," says Caitlin O'Hara, 29, who was diagnosed with cystic fibrosis at age 2 and was able to access the twice-daily pill through Vertex's compassionate use program when her lung function was too poor to get her into a clinical trial. O'Hara had had a relatively normal childhood, except that she needed IV antibiotics every couple of years, and a surgery to treat infection, a common problem for CF sufferers. But in her mid 20s, her symptoms worsened. "I had to cancel trips all the time, because I was too sick. I was avoiding any situations where I'd unexpectedly be asked to walk," she says.
Two weeks after she started the drug, a pulmonary function test confirmed she was improving. "I haven't been in a hospital in two years," she says. "And a year after starting the drug I went to Paris for two weeks by myself, which is something I never thought I'd be able to do unless I got a lung transplant." That still may be necessary eventually. But meantime, "I can live my day-to-day life," she says.
While CF is triggered by changes in just one gene, challenges like coronary artery disease, type 2 diabetes and depression are far more complex, the result of an as-yet-undetermined combination of genetic predisposition and environmental influences. Individualized treatment regimens may be options someday, when more is known about the diseases. And researchers hope that some of these ailments, including cancer, can be avoided. Could we ever get an individualized prescription for prevention? "That's the trillion-dollar question," says Peter Tonellato, director of Harvard's Laboratory for Personalized Medicine.
At this point, researchers are hunting for genetic variants that might be associated with an elevated risk of disease. A team led by Sekar Kathiresan, a genetics expert and director of preventive cardiology at Massachusetts General Hospital, has discovered 45 different genetic variants that can identify a subset of people who have a greatly increased risk of a heart attack. If borne out by future research, those at high risk might someday be told to take aspirin and statins earlier in life, while those not high risk could be advised to wait, he says.
Also ripe for personalized prevention is Type 2 diabetes. About 25 different genetic variants raise the risk of the disease, but even knowing that information isn't a particularly good predictor of whether someone will come down with it, says Allen Spiegel, endocrinologist and dean of the Albert Einstein College of Medicine of Yeshiva University. More finely honed predictors could identify the people who need intensive help to change their diet and exercise habits or who should take medications to prevent the onset of diabetes.
One researcher provides a futuristic glimpse of what this analysis might look like if extended beyond genomic sequencing to other "omes" like the proteome (the proteins produced by the human body) and metabolome (chemicals). Michael Snyder, director of the Stanford Center for Genomics and Personalized Medicine, had his genome sequenced and underwent frequent blood draws to measure thousands of molecules in his blood. His genome predicted an increased risk of Type 2 diabetes, which didn't really worry him because he was at a healthy weight and didn't have other risk factors. But his glucose levels soon spiked, and he was diagnosed with the disease.
"I changed my eating habits a lot" and exercised more, says Snyder. Six months later, his glucose levels had returned to normal.
Snyder foresees a time in preventive medicine when gene sequencing is routine and regular blood tests can capture many more pieces of information than the number of white or red blood cells or your blood glucose level. To be sure, his is a research project with a single subject – himself – so a lot more work needs to be done. Snyder is also now tracking his microbiome (the microbes that live on and in the body) and epigenome (chemical changes to the DNA that can be triggered by the environment). Whether this kind of personalized information, collected and used on a larger scale, could improve care or lower costs is still a question for the future.
Meanwhile, for Betty Lane, Nic Santiago Volker, Lucas Wartman and Caitlin O'Hara, the era of personalized medicine is now. And the concept has already proved its worth.