Researchers have learned to coax a patient’s own T cells, like this one shown in yellow, to attack cancer cells, shown in green.
But while there were no proven effective treatments for advanced melanoma, there was an idea floating around. The idea was that the human immune system might be taught to attack and destroy tumors — that the patient’s body might be able to handle cancer all on its own.
Starting in the 19th century, there had been scattered hints that the immune system could sometimes overcome cancer. But nobody had ever been able to turn this observation into a reliable treatment, in part because the biology of the immune system was still mostly a mystery. Even as late as the 1980s, just a handful of people were pursuing this research.
Steven Rosenberg and his patient Linda Taylor in 1984, when Taylor was given a grim diagnosis of metastatic melanoma.
Taylor’s complete response to Rosenberg’s therapy allowed the two to reunite in 2013, almost 30 years later.
One of those people, Steven Rosenberg at the National Cancer Institute (NCI), happened to be running a clinical trial of interleukin-2 (IL-2), a compound that revs up the immune system, and Taylor just happened to hear about it. She’d already been through a lot, but grudgingly, she joined the trial. The IL-2 made her extremely sick with headaches, fever, and chills — at one point, she even stopped breathing and had to be intubated. But within four months, her cancer had melted away.
Not only that, it never came back. Today, 30 years after she was supposed to die, Taylor is alive and cancer-free.
She was very lucky. IL-2 did not help more than 90 percent of the people treated with it in Rosenberg’s trial — in fact, the treatment itself killed a few of them. Even today, high-dose IL-2 eradicates cancer in only about 6 percent of people with advanced melanoma.
But Taylor’s magical recovery had a transformative effect on Rosenberg and the few other researchers who were determined to make cancer immunotherapy work. During the decades of research that followed, her case was a beacon in the fog. It was proof that at least sometimes, the human immune system could overcome cancer.
Now, after years of setbacks, immunotherapy suddenly has become the next big thing in oncology. “Up until maybe two years ago, immunotherapy was seen as something that had promise, but nobody quite ever expected it to work,” says Rosenberg. “The promise is now being realized.” A string of recent successes is leading many cancer doctors to say science may be approaching a turning point in handling at least some kinds of cancer.
The most impressive achievements have been in two variations of immunotherapy: drugs that remove constraints on immune activity, and personalized cell-based therapies that use a patient’s own white blood cells to attack tumors.
These treatments are working against some of the most intractable cancers in some of the sickest patients — people who have run out of options in their fight against malignancies like advanced melanoma and relapsed blood cancers. In June 2014, a team based at the Yale School of Medicine in New Haven, Connecticut, and at Memorial Sloan Kettering Cancer Center in New York announced that a combination of two new immune-activating drugs had kept more than three-quarters of patients with advanced melanoma alive for more than two years. In October, researchers at the University of Pennsylvania and the pharmaceutical company Novartis reported that cell-based immunotherapy had put 27 out of 30 young adults and children with relapsed acute lymphoblastic leukemia (ALL) into complete remission.
“What’s so exciting is that there is more than one concrete example of immune-based therapies being able to eradicate significant malignancies that are otherwise hard to treat,” says oncologist and leukemia specialist Renier Brentjens, director of the cell therapeutic center at Memorial Sloan Kettering.
One reason it has been so slow to develop is that the vision for immunotherapy is a major departure from conventional cancer treatment. Some drugs and forms of radiation are dumb weapons. They carpet bomb the body, doing extensive collateral damage. They often miss cancer cells that have spread to other organs, which sets the stage for metastasis, and tumors often become resistant to even the most toxic treatments.
But they are relatively straightforward tools. To unleash the immune system against cancer, oncologists must learn to control a living weapon, one that has the capacity to zero in on every last malignant cell, transform itself to match even the most resilient cancer — or spiral out of control and harm the patient. Nonetheless, the promise of immunotherapy is startling in its simplicity: With a little help from cancer doctors, the patients will cure themselves.
Way back in the 19th century, doctors noticed that cancer sometimes regressed in people who developed bad infections, presumably because an immune response to the infection also wiped out the tumor. In the 1890s, a doctor named William Coley tried to duplicate this effect by infecting his cancer patients.
“He injected his cancer patients with a mix of bacterial goop, essentially thinking if this doesn’t kill them, maybe it will help them ,” says Margaret Callahan, an oncologist at Memorial Sloan Kettering. Although it did help some, others died. “The concept was great, although clearly he didn’t have the tools to understand it and control it,” she adds.
It wasn’t until the last decades of the 20th century that the science of immunology advanced to the point that researchers understood how they might pull its strings, says Callahan. Now they know about the many varieties of T cells, the white blood cells that coordinate, target, and carry out the immune response. They have identified hundreds of types of cytokines, the chemical messengers that boost and tamp down different aspects of the immune response. Immunologists now understand many of the ways the body keeps a potentially deadly immune response in check: In the 1980s, an immunologist named James Allison discovered the first “checkpoint mechanisms,” which shut down T cells and prevent an immune attack.
Soon after Allison’s discovery, researchers found that cancer cells use these checkpoint mechanisms to shield themselves from T cells, and not long after, drug developers invented a new cancer treatment that can shut down those checkpoints. As with IL-2, this checkpoint blockade drug was for late-stage melanoma. Once again, the drug made people very sick. And once again, the successes were rare but impressive: About 10 percent of the people who took the drug saw their cancers shrink away and stay away for years. It wasn’t a breakthrough, but it was a strong signal that the concept could work. “It was a foot in the door,” Callahan says. The drug, Yervoy (ipilimumab), was eventually approved in 2011.
In 2012, it also was shown to work in some cases of kidney (renal cell) cancer and lung cancer. This success, plus the drug’s high price tag of $120,000 per treatment, inspired drug developers to invent many other checkpoint inhibitor drugs; probably a half-dozen are now in development, and a few others have been extensively tested. Last September, the FDA approved Keytruda (pembrolizumab) for melanoma after it caused cancers to shrink in 24 percent of the people in a clinical trial. “It’s a tremendously exciting time,” Callahan says.
Cancer: The Emperor of All Maladies
Watch the emotional reunion of Steven Rosenberg, chief of surgery at the National Cancer Institute in Bethesda, Maryland, and his former patient Linda Taylor, who received what would turn out to be a lifesaving immunology treatment, in filmmaker Ken Burns’ latest documentary, Cancer: The Emperor of All Maladies. The three-part, six-hour series will explore the past, present, and future of cancer treatment, while weaving in the personal stories of luminaries in the cancer world and the patients they serve. The documentary, which airs March 30, 31, and April 1 on PBS, is based on Siddhartha Mukherjee’s Pulitzer Prize-winning book, The Emperor of All Maladies. To view the trailer, visit cancerfilms.org.
Checkpoint inhibitors basically launch the equivalent of a five-alarm fire in the immune system, and often cause intense side effects. They may work in only some types of cancers; ovarian cancer, bladder cancer, non-small cell lung cancer, and gastric cancer have had positive preliminary results, but just last September, a trial in prostate cancer failed. And so far, checkpoint inhibitors seem to work in only about 20 percent of the people who try them. “It’s a good option in a population that has no other treatment options, but it’s not where we’d like to be,” Callahan says.
For those lucky few, the response often lasts a long time, by the standards of cancer therapy — tens of months or even years, without any additional treatment. In those people, says Callahan, “the chances are exceedingly high that we’ve changed how your immune system treats the tumor. We now have the upper hand. It’s the closest we’ve come to conceding that we have control of this disease, and it’s reasonably close to a cure.”
Yervoy is just the first, the 1.0 version; other checkpoint inhibitors still in tests look promising, and combining several checkpoint inhibitors may raise the success rate another few notches. Callahan is involved in a trial combining Yervoy and another inhibitor, Opdivo (nivolumab), recently approved for advanced melanoma, which remains a difficult disease to treat with an average survival of six months to a year. (Opdivo was also recently approved to treat advanced squamous non-small cell lung cancer.) In the early results reported in June 2014, 42 percent of patients responded to the treatment, and 79 percent had an overall survival rate of two years. If that result holds out through further rounds of testing, it’ll mean a major improvement. Callahan says, “There’s a tremendous amount of optimism.”
Drugs that liberate the immune system may be the simplest way to produce an immune response, but a more daring approach that deploys a patient’s own T cells to destroy cancer has also shown impressive results in the last few years. In a normal immune response, many different types of these white blood cells interact in a complex network, activating and inhibiting each other. Usually they detect and destroy cells that have been infected by viruses or bacteria, and only weakly respond to cancer. But as scientists have learned, they can be tweaked and prodded to go after tumor cells instead.
Rosenberg was tinkering with these methods more than 25 years ago, separating out the tiny number of T cells that often can be found inside the tumors that surgeons remove from a cancer patient, cultivating these tumor-infiltrating lymphocytes (TILs) in a dish until there are billions of them, and injecting them back into the person. The first test in 1988 at the NCI was not too impressive: Only two of 12 patients improved.
Today, refinements to this technique have increased the complete response rate, meaning that at least temporarily, patients showed no sign of the disease, in close to 40 percent of metastatic melanoma patients, with some remissions lasting as long as seven years. But TILs have their limitations: Not every tumor can be accessed by a surgeon, and it’s not always possible to grow TILs in patients. Also, not every patient can withstand the regimen, which currently is available only at a few sophisticated academic medical centers. And so far, TILs have only worked consistently in melanoma — although that may be about to change. In 2012, Rosenberg’s team treated a woman with late-stage bile duct cancer with a modified version of this therapy. She responded well. Today, her liver and lung metastases are continuing to shrink. Rosenberg is also testing the technique on people with gastrointestinal tumors, or ovarian, lung, or breast cancer.
To get around the need for a surgical biopsy, Cassian Yee at MD Anderson Cancer Center in Houston is modifying this method by finding the extremely rare T cells already sensitized to melanoma that are circulating in the blood — as few as 1 in 100,000 cells. In this method, which he calls endogenous T-cell therapy, he grows them in the lab, and then reinjects billions of them back into the patient. In one early study, they stopped cancer growth for 6 out of 11 people with metastatic melanoma, and they caused no serious side effects — but in all but one of those people, the cancer returned within months. Yee has developed a method to get the T cells to convert into a longer-lasting type. There are plans to pair this therapy with the new checkpoint inhibiting drugs.
Another way to solve the problem of finding rare or inaccessible T cells is simply to create them. Normally, T cells are programmed by other immune system cells to detect infectious invaders, but starting in the 1990s, researchers figured out how to redirect that response via genetic engineering. To make these chimeric antigen receptor cells (CAR-T cells), a cancer patient’s T cells are loaded with a new weapon: a man-made targeting system that directs the T cell to identify, attack, and kill tumor cells.
So far, CAR-T cell therapy has mostly been applied to leukemia and other blood cancers of the B cells (another type of white blood cell). To make these cells, a cancer patient’s T cells are harvested and reprogrammed with a detector for a protein called CD-19, which is carried by most B cells. These engineered cells are grown in the lab and then reinjected into the patient.
Three research groups using slightly different versions of this technique have now published more than a dozen studies of CD-19 CAR-T cells. Across all studies, about 80 percent of people treated with CAR-T cells show at least some response, and second- and third-generation designs are expected to yield better results. The most impressive results so far have been in ALL, which often carries a dire prognosis. For adults with relapsed or refractory (meaning resistant to traditional therapies) disease, less than 5 percent survive long-term.
In February 2014, Brentjens’ study at Memorial Sloan Kettering reported that 14 of 16 adults with relapsed ALL responded to CAR-T cell therapy, and seven went on to have potentially curative bone marrow transplants. At the NCI, Daniel “Trey” Lee, assistant clinical investigator in the pediatric oncology branch, is leading the most rigorous trial of CAR-T cell therapy in children and young adults with relapsed ALL. In October 2014, his team announced that 14 of 21 patients had a complete response. Some of those people had never before been able to get their cancer into remission.
Even though these methods are technically complex and may be difficult and expensive to turn into mainstream therapy, drug companies are paying sharp attention. Brentjens and other leaders of CAR-T cell therapy recently launched a biotech company, Juno, to commercialize this technology. Rosenberg and researchers at the NCI are working with another biotech, called Kite. The team at the University of Pennsylvania, which reported the strong results in relapsed ALL last October, has partnered with the drug company Novartis. There is a lot of money at stake: Juno raised $300 million in investor money, said to be the most ever for a start-up biotech, and Novartis is betting big on immunotherapy.