The "Jellyfish" Vaccines
To the 13 prostate cancer patients, their experimental treatment at Duke seemed like a dream-scenario from the most optimistic future of cancer care. They had already endured surgical removal of their prostates and hormone therapy--with its side effects including impotence, hot flashes, diarrhea, and nausea. This new treatment, however, consisted of just three simple injections, each of which produced at most the kind of mild inflammation that might occur after a flu shot. Better yet, after each injection, the patients could immediately go home and get back to their daily business.
Perhaps most remarkably, these men were receiving not some exotic new cancer-killing chemical, but a vaccine made from their own dendritic cells. These cells--biological Paul Reveres that alert the immune system against invaders--had been ingeniously treated to enable them to sensitize the immune system to the presence of the men's prostate cancers, then reinjected into the bloodstream. There, the dendritic cells educated "naïve" killer T cells--the immune system's shock troops--to attack cancer cells wherever they might encounter them.
This initial prostate cancer vaccine trial aimed only to determine the vaccine's safety, emphasizes Johannes Vieweg, MD, an assistant professor of urology and immunology, who reported the results in the February 2002 Journal of Clinical Investigation. But the researchers got more than they hoped for. Besides showing that the vaccines produced no significant toxicity, the study results show a significant, consistent immune reaction against the cancer cells, even though the treatments had been done with an early version of the vaccines and had been limited in scope.
These results hint at early success in what the Duke researchers hope will be an important new front in the war against cancer. "This is the first study that has data on the safety and immunological efficacy of this type of cancer vaccine," says Vieweg. "And while this work was done in prostate cancer patients, we believe our method may prove to work in most other cancers as well."
Such vaccinations will not cure patients at this point, cautions cancer biologist Eli Gilboa, PhD, Joseph and Dorothy Beard Professor of Experimental Surgery and professor of immunology--but they may prove powerful enough to reduce the rate of cancer recurrence and prolong remission times. Indeed, Gilboa asserts that the non-toxic nature of immunotherapy means it could prove far superior to chemotherapy: "I believe we are in the midst of a revolution," he said. "I think immunotherapy will eventually replace much of chemotherapy as we know it."
Cancers are notoriously adept at evading detection by the immune system. They are, after all, not foreigners invading the body, but the patient's own cells run amok. For that reason, attempts to develop immune-alerting vaccines have been exceedingly difficult and frustrating, yet promising enough to pursue.
Perhaps the earliest such experiments, conducted by New York surgeon William Coley in the 1890s, were inspired by his observation that cancer patients who developed infections near their tumors showed a reduction in tumor size. Coley began directly injecting live bacteria into the tumors--and indeed, some of them shrank. Since then, the scientific quest for cancer vaccines has grown more sophisticated and multifaceted, with researchers using techniques from gene therapy to killed tumor cells themselves to try to spur the immune system to attack cancer.
At the moment, among the more promising strategies seems to be one that, at first, few people believed would even work. Developed at Duke by Gilboa and his colleagues, it involves engineering dendritic cells to fight cancer.
Discovered just three decades ago, dendritic cells are both scarce in the body and exotic-looking--like microscopic jellyfish extending tangles of delicate tendrils that entwine themselves about neighboring cells. In action, dendritic cells engulf foreign proteins, chemically shred them, and display the shreds on their surface like molecular signal flags. These flags, called "antigens," tell the T cells what to attack. Not until Gilboa and his colleagues achieved startling results with their admittedly unorthodox experiments, however, has the medical community recognized the cells' ability to serve as the basis for cancer vaccines.
The Duke researchers initially tried loading the cells with antigenic proteins that were prevalent in cancer cells but rare in normal cells. The aim was to induce the dendritic cells to display the antigens, triggering an immunological attack on the cancer.
The researchers had some success, but isolating and mass-producing the proteins proved frustratingly difficult.
Then Gilboa and his colleagues Smita Nair and David Boczkowski hit upon an idea for one of those slightly off-the-wall experiments that yields a Eureka moment. What if it would be possible to get the dendritic cells to produce the antigen proteins themselves? If researchers simply soaked the cells in the messenger RNA that constitutes the genetic blueprint for those proteins, the cells might use the RNA as a template to create the proteins and then their antigens. The long-shot experiment was tempting, because it is far easier to use standard genetic engineering techniques to mass-produce messenger RNA than to make proteins.
"There was originally some very healthy skepticism, because the myth was that RNA is very unstable," recalls Gilboa. "People asked how it could be possible to just dump RNA on dendritic cells and expect them to pick it up and translate it to proteins." But to everyone's surprise and delight, the scientists found that the dendritic cells readily took the bait, using the RNA to make antigens that they presented on their surface.
From Mice to Men
In 1996, the scientists reported proof-of-concept experiments in mice, in which an RNA-based dendritic-cell vaccine dramatically reduced the spread of lung cancer and prevented its return. In subsequent test tube studies, they found that dendritic cells loaded with RNA for a human tumor antigen called carcinoembryonic antigen (CEA)--produced in breast, lung, and colorectal cancers--stimulated an immune response. Based on those studies, Professor of Surgery Kim Lyerly, MD, and his colleagues have begun a trial to test the CEA-based vaccine in cancer patients (see accompanying article).
The researchers also launched studies of vaccines made using total RNA extracted from patients' tumors, instead of just the RNA for specific known antigens inside the tumor. Such a total-RNA approach could yield vaccines that did not depend on any prior knowledge of the antigens in a patient's tumor.
The scientists first tested a total-RNA vaccine in mice, and the results surprised even the careful, reserved Gilboa. "It was remarkable," he said. "Again, it was quite an unorthodox approach. We pursued it because it offered the practical advantage of broad applicability, but we did not expect that it would be particularly powerful in producing immunity. When we found an exceptionally powerful immune response, we tended not to believe our data because they were so counterintuitive." Not until other laboratories successfully tested the Duke approach was he fully convinced, Gilboa said.
Building on Gilboa's basic studies, Vieweg immediately began human trials of both vaccine approaches. In the prostate cancer vaccine trial, patients with metastatic prostate cancer are receiving vaccines made by treating dendritic cells with RNA for prostate-specific antigen (PSA), a hallmark of prostate cancer. The results of the phase I trial described in the Journal of Clinical Investigation article showed several indications of success: an absence of toxicity, activation of the patients' T cells to fight the cancer, and the encouraging fact that PSA levels rose more slowly in the treated patients. Also ongoing are trials of a vaccine for metastatic renal cell carcinoma, in which patients receive vaccines made from their own total tumor RNA and dendritic cells--eliminating any possibility of an unwanted immune response to foreign proteins.
While the initial clinical success has confirmed their approach, Vieweg cautions that the place of vaccines in cancer therapy can only be established by further trials.
"With these initial patients, we have not cured them, but we may have at least in part shrunk their tumors or significantly retarded tumor growth. However, the protocol in this safety study called for application of only three vaccinations. From these initial results, we believe there will be no real downside to continuing vaccine therapy to keep patients' cancer in check."
What's more, Vieweg pointed out, the initial trials were in patients whose cancer had already spread outside the prostate. The ultimate clinical use of such vaccines might be much different, he says. "About 50 percent of prostate cancer surgeries do not eliminate the tumor entirely," said Vieweg. "Currently, there is simply no cure for residual tumor cells after surgery or radiation. We believe that a vaccine could help eliminate the ticking time bomb of the remaining cancer cells in these patients."
A Universal Vaccine?
Meanwhile, said Gilboa, the Duke researchers will continue basic studies both on the defined-antigen and whole-tumor vaccines.
One highly promising branch of their defined-antigen effort involves working to create a universal cancer vaccine using telomerase, a protein enzyme that prolongs the life of cells. While cancer cells show highly activated telomerase production, in normal cells the enzyme is repressed to low levels.
In experiments in which the scientists loaded dendritic cells with RNA representing the telomerase protein, the resulting vaccine stimulated an immune response that killed a wide variety of cancer cells. And, the vaccine slowed the growth of melanoma, bladder and breast cancers implanted into mice. No other vaccine has been able to induce such a broad immune response, said Gilboa. "Telomerase is a particularly promising antigen, because the more aggressive the cancer gets the higher the telomerase expression," added Vieweg. "So, when we treat cancer with a telomerase-based vaccine, the more aggressive a tumor is, the more visible it is to the immune system."
Telomerase does have drawbacks--by itself, it is not a strong antigen--but both Vieweg and Gilboa see antigen "cocktail" vaccines including telomerase as well as other antigens, such as PSA, as a highly promising approach.
Another ambition is to learn to activate dendritic cells against cancers without even removing them from the patient--reducing both cost and complexity of immunotherapy. The researchers are also working to supercharge the immune system itself, using substances called immunomodulators.
And, they are seeking to deal cancers a one-two punch by exploring how immunotherapy will work with other cancer treatments. For example, they are combining vaccines with "antiangiogenesis" treatments--which prevent tumors from growing blood vessels, starving them of a blood supply.
"Cancers are not cured by a magic silver bullet," emphasized Vieweg, who, like Gilboa, sees cancer vaccines as only one weapon in a steadily growing arsenal. "They will require multimodal therapy applied in the right strategic context."
At this point, said Gilboa, the only real frustration is that the researchers' imaginations have outstripped their resources in terms of people, facilities, and support for research and clinical trials. "We have so many ideas, and these therapies promise to be so benign yet so powerful and specific, that we never feel we are moving fast enough," he said.