Marty Tenenbaum shouldn't be here today. Almost 20 years ago, the computer scientist and e-commerce pioneer was diagnosed with metastatic melanoma, an aggressive cancer that, at the time, had no effective treatments. Told he should measure the rest of his life in months, not years, Tenenbaum scoured the Web for a clinical trial that might buy him more time.
He decided to bet his life on a Phase III trial of Canvaxin, an investigational vaccine designed to stimulate the immune system to combat tumors. In an earlier phase of clinical testing, the vaccine had been shown to extend the lives of people with the deadliest form of skin cancer. But in the trial Tenenbaum got into, Canvaxin failed to demonstrate that kind of survival benefit in enough people, so it was abruptly halted. The vaccine, deemed a flop, was destroyed.
A few outlier patients, however, responded remarkably to Canvaxin before the trial was closed. Tenenbaum, now 72 and cancer-free, was one of them.
The history of oncology is peppered with similarly curious (or miraculous) anecdotes about one or two patients, like Tenenbaum, who had recoveries so spectacular they defied explanation. Often these patients failed to respond to multiple courses of therapy and eventually sought treatment in a clinical trial as a last resort.
Until recently, such dramatic outcomes have left patients thanking some higher power for their Lazarus-like recovery, and physicians and researchers scratching their heads. Not enough was known about cancer's basic biology, and the technology did not exist to understand why someone fared so well on a drug that provided little to no help for most patients.
But today, thanks to powerful new genome-sequencing technologies, which are getting faster and cheaper every day, it's increasingly possible to pinpoint genetic mutations and other molecular abnormalities that play a role in some patients' astounding recoveries. By studying these patients—known as "super responders" or "exceptional responders"—a growing number of researchers hope to not only learn how and why a patient responded to a specific treatment but to identify other patients who could benefit from the same regimen.
Dr. David Carbone, a lung cancer specialist and genetics expert at Ohio State University's (OSU) Comprehensive Cancer Center, says he's seen a "fair share" of super responders come through his clinic, but one in particular stands out: a 66-year-old woman with advanced lung cancer. Neither surgery nor chemotherapy could help her, and within six months of her diagnosis, she was admitted to a hospice. However, her health remained stable, and she sought a second opinion.
"For her, it was a shot in the dark," says Carbone. "She had no expectations." He enrolled her in a clinical trial for sorafenib, a drug that blocks the function of certain enzymes that play a role in the development of tumor formation. It is approved in the U.S. for advanced forms of liver, kidney and thyroid cancers, but not for lung cancer.
The woman's tumors began to shrink almost immediately. Within two months, they had disappeared entirely, and her disease was kept at bay for another five years. Only nine others among the 306 patients in the trial responded to the drug, says Carbone, but she "by far had the best and longest-lasting response of them all."
Although she eventually relapsed and succumbed to her disease years later, her off-the-charts response to sorafenib prompted Carbone to take a deeper, more intensive look at the genetics of her tumor. He and his team performed whole-genome sequencing to look for genetic mutations in the DNA of the patient's cancer cells before her use of sorafenib. They also sequenced RNA—molecules that carry genetic messages within the cell—from the woman's tumor and healthy cells.
Their analyses revealed more than 100 genetic abnormalities in her cancer cells, compared with her healthy cells, but one stood out: a mutation in a gene called ARAF that had never been linked to cancer. Further research demonstrated that the abnormal ARAF gene formed tumors and that these tumors were inhibited by sorafenib.
OSU has since added ARAF to the panel of cancer-causing genes it routinely screens in patients with all cancers, in the hopes of identifying others with the rare mutation who may respond to targeted therapy. "If we can show that a particular gene mutation is making one person's tumor vulnerable to a drug, there's a chance that other patients with the mutation—including those with different kinds of cancers—may benefit from the same treatment," Carbone says.
Muting the Mutations
Decades of cancer research have shown that cancer is a remarkably diverse disease. Even cancers that begin in the same part of the body are radically different at the DNA level. Lung cancer, for example, is now understood not as one disease but as a collection of subtypes—each characterized by a spectrum of mutated genes and other abnormalities—that require different treatment approaches.
Because several known cancer-causing mutations occur in multiple types of tumors, cancer is increasingly defined not just by the organ in which it originated but by the mutations that drive its growth. "These mutations can sometimes be targeted with the same drug, but it's unfortunately not a given," Carbone says. Melanoma patients who have a mutation in a gene called BRAF respond well to drugs that block the activity of the BRAF protein. Lung cancer patients with the BRAF mutation also respond well to the drug, but colorectal cancer patients with that same mutation do not. Still, Carbone says, "knowing which mutations are present in an individual patient is the first step in helping to precisely tailor a patient's treatment to the genetic features present in his or her cancer cells."
Dr. Glen Weiss, director of clinical research and Phase I and Phase II clinical trials at Cancer Treatment Centers of America and a clinical associate professor at the Translational Genomics Research Institute in Phoenix, has also treated patients whose cancer took an unexpected trajectory. One of them was a 54-year-old woman who was dying of ovarian cancer. "She had exhausted all other treatment options. She came to me already having started to get her affairs in order," he says.
On a hunch, Weiss treated her with an experimental drug called a PARP inhibitor as part of a clinical trial. In earlier studies, PARP inhibitors had been shown to be effective in ovarian cancer patients who had a mutation in the BRCA gene, which the woman had.
Weiss was stunned to find that she was cancer-free six weeks after beginning treatment. "Rather than just having some tumor shrinkage or disease control for a period of time—as is usually the best case for that particular class of drugs—she recently celebrated four years with no sign of disease," he says.
Late last year, the Food and Drug Administration approved the first medication of this type, olaparib, for the treatment of women with ovarian cancer who no longer respond to other treatments and who are likely or suspected to have BRCA mutations. PARP inhibitors are also under study for patients with other cancers who harbor BRCA mutations, such as breast, pancreatic and prostate cancers.
Cancer Commons
But what about those patients who've had an exceptional response outside a clinical trial? After all, only about 3 percent of the 1.7 million people diagnosed with cancer each year in the U.S. take part in one. "Surely there are other super responders, but unless these cases are published in medical journals or shared at medical meetings, we just are not hearing about them," Carbone says. It is not uncommon for research data to be published years after being generated.
This is where Tenenbaum re-enters the picture. He drew on his experience as a super responder to start Cancer Commons. The nonprofit organization, based in Palo Alto, California, aims to place data relating to exceptional responders in a free, searchable online database. "If there was another patient who had similar mutations as me and who had a miraculous response to a drug, I'd want to know before I made any decisions about my treatment. Wouldn't you?" he says.
Both health care providers and patients can contribute via smartphone data that is immediately rendered anonymous. Powerful analytics then sift through that data, along with information from other sources, like physicians' notes, clinical guidelines and journal articles, to provide treatment recommendations. "As more data are added, patterns emerge that you could not have seen in just one patient or even in bigger clinical trials where positive responses from one or two patients get lost in the rest of the data," Tenenbaum explains.
In theory, doctors would tap the database for insights into how to treat a patient based on the experiences of super responders and other patients who share the same genetic mutations and other genomic characteristics. Patients can also query the database. Together, doctors and patients will form a knowledge network around certain genetic mutations that Tenenbaum envisions as "similar to LinkedIn."
Others groups are also seeking to tap into patient experiences—the good, the bad, the spectacular—that are not captured by clinical trials. The American Society of Clinical Oncology's CancerLinQ and health care technology company Flatiron Health, both of which aim to cull data from millions of electronic health records, will allow physicians to base their treatment choices on the experiences of similar patients. And cancer institutes such as the Dana-Farber Cancer Institute in Boston and Memorial Sloan Kettering Cancer Center and the Icahn School of Medicine at Mount Sinai, both in New York City, have created their own databases in the hopes of helping doctors match the right treatment to the right patient.
Says Tenenbaum, "You could say we're hoping to make the exception routine."
This article is one in a series from Newsweek 's 2015 Cancer issue, exploring challenges and innovations in cancer treatment and research. The complete issue is available online and at newsstands.
