Allison’s results astounded cancer specialists. Nature published a review in December 2011 and noted that the antibody to CTLA-4 “provides realistic hope for melanoma patients, particularly those with late stage disease who otherwise had little chance of survival. More broadly, it provides clear clinical validation for cancer immunotherapy in general.” I asked Harold Varmus why Allison had had success where other researchers in immunotherapy had failed. “We need to understand what we do,” he said. “Jim made things understandable.”

  “You’ve got to be careful about using the word ‘cured,’ because some patients have residual tumors,” Allison said. “But it doesn’t matter, because their cancers are not growing. And, in others, tumors just pop up and then go away. So it’s become something of a chronic condition,” rather than a death sentence. Allison moved to Sloan-Kettering to be closer to the clinical trials conducted by Wolchok and others. “I just wanted to be the advocate who is keeping it in everybody’s face,” he said.

  In the fall of 2003, Sharon Belvin was a twenty-two-year-old student teacher with plans to marry the following June. She ran between four and five miles a day and began to notice that her chest hurt after her morning workout. The student health service thought that she might have viral bronchitis, picked up from the children in her class. But her symptoms did not improve, and she was given other diagnoses, including asthma and pneumonia. Before long, she found it uncomfortable even to walk. On a visit to her mother, Belvin saw the family physician, who found a lump on her clavicle. A biopsy showed that she had metastatic melanoma. “It shocked me,” Belvin told me. “I was never a sunbather. And I never had any lesions on my skin.” A week before her wedding, she completed her evaluation. A body scan “lit up like a Christmas tree,” she recalled. “I ended up having chemotherapy on Monday, Tuesday, and Wednesday, and got married on Saturday.” During four months of therapy, the tumors shrank a bit. Then they began to grow again. An MRI showed that the melanoma had spread to her brain. Belvin went to Sloan-Kettering, where the brain tumor was treated with radiation. After recovering from the procedure, she received interleukin-2, to stimulate her T cells. The therapy caused such a severe reaction that “my skin peeled off all over my body,” Belvin said. “I was so violently ill, I don’t remember half of what happened.” Worse yet, the treatment failed to stop the cancer’s growth. “The doctor told me, ‘If you are going to take a vacation, you’d better do it now.’” Belvin and her husband went on a Caribbean cruise.

  When she got back, Belvin returned to the hospital and had twelve liters of fluid drained from her chest. Then Wolchok offered Belvin treatment with the antibody to CTLA-4, which was still an experimental therapy. “By that point, I had told my husband, ‘If this doesn’t work, I don’t know how much more I can take,’” she recalled. Wolchok gave her an informed-consent release that listed all the possible side effects. “It was pages and pages of this could happen to you and that could happen to you. I didn’t read one page. I just signed at the bottom and said, ‘Give it to me.’”

  The antibody was infused through one of Belvin’s veins, and she had a drastic reaction: her body shook and she experienced drenching sweats, as well as an immune attack on her thyroid gland. “I thought I was dying, the rigors were so bad,” she recalled. After four treatments given every three weeks, Belvin went for a set of scans. “I remember how Dr. Wolchok came in with this huge smile on his face, and he was like, ‘This is great!’ He was just floored.” The massive tumors in her lungs had shrunk significantly.

  Wolchok did not want to raise Belvin’s hopes too much. But “every single scan that I had after that time, the tumors kept shrinking,” she said. Eight years after her diagnosis, she still has no signs of the cancer.

  Belvin’s case is remarkable, but it contradicts the popular notion that boosting the immune system is a “natural” way to treat cancer, free of the harsh side effects associated with chemotherapy or radiation. The results of immunotherapy can include an attack on the skin, intestines, lungs, liver, thyroid, pituitary gland, kidneys, and pancreas. When T cells are stimulated to an intensity that destroys cancer cells, they can also cause collateral damage to normal tissue. Wolchok told me, “You may need to cross the line to toxicity for the immune system to be effective against a cancer. It’s not a free ride.” Because Belvin’s thyroid gland was destroyed by the therapy, she now requires replacement hormones.

  Steven Rosenberg, the chief of surgery at the National Cancer Institute, who played a key role in developing interleukin-2, also conducted some of the early studies with the antibody to CTLA-4. He noted that the bowel often became severely inflamed with the treatment: “You have, like, eight liters of diarrhea a day. The colitis is atrocious and would be lethal in almost everybody. If you don’t put those patients on corticosteroids immediately, they’ll die.”

  “In the field of oncology, the bar is set so low,” Rosenberg told me. He welcomes the outcomes for patients like Belvin but is cautious about the long-term benefits of similar treatments. “I believe that the antibody to CTLA-4 will cure some patients with melanoma, although the follow-up is short.” But unless all detectable cancer disappears, he said, “the tumors are going to grow back eventually.”

  Rosenberg has pioneered a different strategy, called “adoptive cell transfer,” in which T cells are taken from a patient’s tumor and given immune stimulants such as interleukin-2, which cause them to replicate. Then they are put back into the body. In the latest of three trials of patients with melanoma who underwent adoptive cell transfer at the National Cancer Institute, nine of twenty-five patients have been in complete remission for more than five years. Across all three trials, five patients who had received earlier, unsuccessful treatment with the antibody to CTLA-4 are in remission.

  Sam Breidenbach, who runs a construction company in Wisconsin, was one of those five. In September 1999, his wife noticed a small mole on his back. He went to the hospital at the University of Wisconsin in Madison and was told that he had melanoma. It was caught early, and the doctors, after removing it, said that the cancer did not appear to have spread. But three years later, while playing volleyball, he lunged to spike the ball, and felt a pull at his left flank. “It was this roly-poly little nodule on my left hip, at the top of the bone”—a metastasis from the original melanoma. “A local oncologist just basically said, ‘You’ll be lucky to live five years,’” Breidenbach recalled. He returned to the hospital in Madison, where he was given high doses of interferon. “For the first month, I was just totally dead. I couldn’t do anything.” The treatment was ineffective. Within months the melanoma had appeared in the lymph nodes of his left groin.

  Breidenbach found out about Rosenberg through his daughter, who was in a violin class with a girl whose father had been treated for melanoma at the National Cancer Institute. Breidenbach contacted Rosenberg, who treated him with an experimental melanoma vaccine. Breidenbach did not respond to the treatment, and the melanoma spread to his liver and lungs. In the summer of 2003, after being treated with the antibody to CTLA-4, he developed excruciating pain in his abdomen—pancreatitis, caused by the toxicity of the immune response. “It was so brutal that they had to stop the treatment,” Breidenbach said. “They were basically out of any other ammunition to throw at me.” His doctor at the University of Wisconsin told him that he couldn’t expect to live more than four to six months. One oncologist suggested chemotherapy, but “I knew the numbers, and my wife and I said, ‘If this is really the remaining time I have on the planet, why make it miserable?’”

  Over the week of Thanksgiving, Rosenberg called and told him that his research team had studied his T cells in the laboratory. “Your cells are jumping out of the petri dish,” Rosenberg said. He explained that Breidenbach’s T cells could be stimulated to recognize and attack melanoma. “Dr. Rosenberg basically told me to get on the plane on Monday and expect to be here for three weeks.” Breidenbach’s T cells had been removed and manipulated in Rosenberg’s lab. Upo
n his arrival at the National Institutes of Health (NIH), they were returned to his body through a catheter entering the vein to his heart. “All the doctors were grinning in the operating room,” he told me. “I felt like it was Dr. Strangelove.” Breidenbach developed a fever of 104 degrees, and his skin erupted in a rash. He went home on Christmas Eve barely able to walk, but within a month the numerous metastases had started to shrink. Today none of the melanoma remains. “My T cells, they were fiery,” Breidenbach concluded. But there was one permanent side effect of the treatment. Along with the cancer, the manipulated T cells attacked the normal cells with melanin, causing vitiligo, in which skin loses its pigment and hair turns white.

  Rosenberg believes that melanoma has a unique relationship with the immune system: there are so many mutations in the tumors that T cells have an easier time recognizing them as foreign. This characteristic makes developing immune therapies easier. “An intense natural immune response just doesn’t exist for other kinds of cancers,” he said.

  But Rosenberg thinks that he has the key to a more wide-ranging approach. “With six hundred thousand Americans dying every year with cancer, we need something for the common cancers,” he said. He acknowledges that targeted drugs, such as Gleevec, can be effective, but he points out that most targeted therapies quickly wane in their efficacy. A recently developed therapy for melanoma dramatically shrank more than half of tumors, but nearly all patients relapsed within a year. A study published in March suggested that as a cancer spreads in the body—from the kidney to the liver and the lungs—the mutations occur in nonuniform ways, so that DNA in liver deposits may differ from DNA in tumors in the lung. This protean progression means that a drug targeted to one mutation may not work against cancer cells throughout the body.

  In Rosenberg’s view, with adoptive cell transfer, these malignancies would all appear equally foreign to the immune system. He is refining the treatment for other cancers by skimming patients’ blood and then inserting a gene into their T cells that targets a different protein, called NY-ESO. The protein, which was identified at Memorial Sloan-Kettering, is normally absent in tissues after fetal development, except in the testis, but it reappears in about a third of all common cancers. “I think adoptive cell transfer is going to be the secret to applying immune therapy to the treatment of many human cancers,” Rosenberg said. “When T cells are genetically engineered to target NY-ESO, there is no difference between melanoma and breast cancer or prostate cancer, or colon cancer, ovarian cancer, sarcoma, and so on.”

  Varmus agrees that this approach might make a wider array of tumors susceptible to therapy, and in early trials Rosenberg’s strategy has been promising. In 2008 Anita Robertson, a sixty-three-year-old accountant from Long Beach, California, had a large sarcoma growing in her hip, a type of tumor similar to the one that killed Elizabeth Dashiell. In July 2010, after treatment with genetically altered T cells, Robertson was discharged from the NIH hospital. A CAT scan in September showed that the sarcoma had begun to shrink; it is now more than 50 percent smaller. Once immobile and in pain from the cancer, she now can drive, shop, and attend church.

  Using a similar approach, researchers at the University of Pennsylvania have eradicated chronic lymphocytic leukemia in three patients who were no longer responding to other therapies. This month Rosenberg reported remissions in eight of nine patients with advanced lymphoma, and in three of those patients the cancer disappeared completely.

  “We’ve got much, much better now with adoptive cell transfer,” Rosenberg told me, “but it’s not widely available.” The treatment has to be individually designed for each patient, which makes it enormously expensive and so less valuable to pharmaceutical companies. “They want a drug, and they don’t care if you spend five hundred million dollars developing the first vial, as long as they can produce the second vial for a dollar,” Rosenberg said. Because his work is experimental, it has been supported by federal funds. Eventually, however, these therapies will be priced by calculating how much they offset the costs of conventional treatments. Although the new procedures could run to hundreds of thousands of dollars, they might still prove less costly than the money spent on chemotherapy, hospitalization, and hospice care for the many patients who currently cannot be cured.

  Jedd Wolchok, however, argues that common cancers may not require adoptive cell therapy. He talks about the “three E’s” in immune therapy: elimination, equilibrium, and escape. Therapy should aim for total elimination of the cancer, but “we need to think about immune-system equilibrium,” in which the cancer, though present, does not grow or spread. After decades of frustration and failure in the clinic, most scientists are wary of predicting whether immune therapy will be able to completely cure the majority of cancer patients. Tumors have mutated to escape the effects of radiation, chemotherapy, and targeted agents; the body’s immune responses may not be unique.

  Though CTLA-4 is still the focus of much research, scientists have now identified at least five other inhibitors on T cells. Initial studies show that treatments directed at these inhibitors can shrink some of the most deadly tumors, including those of the lung and the colon. Mario Sznol, an oncologist at Yale, has conducted clinical trials with an antibody directed against one of the inhibitors, a protein called PD-1. “I believe that in the future we can customize immune therapy to the individual patient,” he said. Doctors will examine the specific characteristics of a tumor and then treat patients with the appropriate antibody.

  Allison’s laboratory is an open space that occupies a large part of the fifteenth floor of the Zuckerman Research Building at Sloan-Kettering. The day I visited, postdoctoral fellows and graduate students were analyzing data on their computers from recent experiments. In a corner was an intravital microscope, which can show cells and tissues in a living animal. Allison demonstrated how an anesthetized mouse is injected with the antibody to CTLA-4. Previously, the T cells of the mouse had been labeled with a fluorescein dye and sensitized to a protein from a tumor. Using the intravital microscope, “you can actually watch the T cells move into the lymph node,” Allison said. They appeared as bright green circles coursing through thin gray vessels. “And then the T cells jump—they leave the lymph node and attack the tumor.”

  In another part of the lab, a postdoctoral fellow had arranged a series of mice that had been inoculated with melanoma. Some served as controls, and black masses an inch or more grew on their flanks. Others had received the antibody to CTLA-4 or to PD-1 or a combination. “The most dramatic regression is seen with the combination,” Allison said, pointing to the flanks of mice where the tumors had shrunk to small black dots. Clinical trials in patients have begun with combining one antibody against CTLA-4 and another against PD-1 in order to remove two distinct brakes on the T cell.

  Last year the antibody to CTLA-4, marketed under the name Yervoy, was approved by the Food and Drug Administration to treat melanoma. It was a vindication for immune therapy and an important step in the treatment of cancer. Yet this branch of research has also uncovered how far we have to go to understand the mutations that make cancer the most protean of diseases. “The future is about thoughtful combinations, different antibodies, perhaps with targeted therapies,” Wolchok told me. “There won’t be a single silver bullet for everyone.”

  DAVID OWEN

  The Artificial Leaf

  FROM The New Yorker

  DANIEL NOCERA WAS a science-minded high-school junior in New Jersey at the beginning of the Arab oil embargo in 1973. American fuel prices soared, the stock market crashed, Congress prohibited speed limits higher than 55 miles an hour, and President Nixon banned the sale of gasoline on Sundays. At the end of the decade, the Iranian revolution, followed closely by the outbreak of war between Iran and Iraq, precipitated a second oil crisis. By then Nocera was a graduate student in chemistry at the California Institute of Technology. Within a short time, he had decided to devote his science career to energy.

  Most of the energy we use comes from
photosynthesis. Green plants store energy from the sun in chemical bonds, and we exploit that energy when we eat plants, or when we eat animals that have eaten plants, or when we burn either plants or substances ultimately derived from plants: firewood, peat, coal, oil, natural gas, ethanol. Photosynthesis has been understood in a general way for a long time and is familiar even to grade-school students—water and carbon dioxide in; oxygen and carbohydrates out—but the process is complex, and until fairly recently important parts of it remained mysterious. Nevertheless, Nocera decided in the early eighties that the chemistry of green plants was the likeliest place to seek an answer to civilization’s long-term energy difficulties. “For the past two hundred years, we’ve run this other experiment, with fossil fuels, and it’s not working out so well,” he told me last August in his office in the chemistry department at the Massachusetts Institute of Technology. (Next January he will move to Harvard.) “I wanted to go back to what worked for two billion years before that.”