October 7, 2002
Attacking tumors from within: Cancer-vaccine research advances at
Dana-Farber
Glenn Dranoff, MD
In Alfred Hitchcock's movie "Sabotage," a mild-mannered cinema owner attempts to terrorize London by shutting down the city's electrical system. Scotland Yard is slow to pick up his trail; he seems too ordinary to be a danger. Londoners themselves are more amused than frightened by the power outage. Slowly, though, a detective becomes suspicious of the shady figures visiting the cinema and decides to move in for an arrest. His action comes too late, however, as the owner has already sent off a bomb that explodes on a crowded bus.
Change the setting from London to the microscopic world of living tissue, replace the actors with human cells, and the film could be an allegory for the drama that takes place between cancer cells and the immune system, the body's defender against disease. If the immune system fails to detect and wipe out cancer cells, as it occasionally does, it is not because it's derelict in its duties, but because cancer cells blend in so easily with their neighbors. They behave, at least in their early stages, much like normal cells, giving few signs of their potential to become malignant.
Cancer cells can evade the immune system's radar because, in a real sense, they aren't alien to the body at all. A cancer cell is actually a normal cell—be it from the liver, breast, bone marrow, or other tissue—that has picked up some glitches in its genetic programming. The immune system, so proficient at finding and destroying truly foreign entities such as bacteria and parasites, often isn't sharp enough to detect the tiny, subtle differences that distinguish cancer cells from normal ones.
To use security parlance, cancer cells sometimes aren't attacked by the immune system because they don't "fit the profile" of a dangerous element. Rather than assault cancer cells and risk harming normal ones as well, the immune system seems to err on the side of caution—a restraint that, by allowing cancer cells to grow unchecked, can have life-threatening consequences.
All these facts have shaped the quest for new immune-based therapies against cancer—for vaccines that use the body's own disease-fighting forces as a weapon against tumors. Instead of trying to rev up or supercharge the immune system's response to tumors, which would risk damaging both healthy and diseased tissue, researchers are seeking ways of making cancer cells more conspicuous to the immune system.
At Dana-Farber, vaccine research has brought a shrewd ingenuity to the fight against cancer. Scientists using the tools of molecular biology are developing techniques that cause cancer cells to betray themselves as a potential enemy and, in effect, invite their own destruction.
The strategies for accomplishing this vary, but they all came about as a result of deepening understanding of how the immune system functions and interacts with cancer cells and other abnormal cells. Advances have been rapid; techniques that just a few years ago were being studied in the laboratory have begun to be tested in patients.
Although cancer-vaccine research was once the work of a few, relatively isolated scientists, it is becoming more of a collaborative venture in which investigators from different laboratories share findings and ideas. "Dana-Farber scientists are contributing to what may one day be seen as a revolution in vaccine research," says the Institute's Senior Vice President for Experimental Medicine Lee Nadler, MD. "There's a dynamism created by bringing together leaders in a particular field. We're poised to have an even greater impact in vaccine research as we move ahead."
Here, Paths of Progress surveys the main currents of cancer-vaccine research at the Institute.
Perfecting the process
As it evolves from a concept to a usable tool, a cancer vaccine follows much the same course as any invention: after initial tests give a glimmer of promise, it is adapted and refined until its full potential is realized.
So it is with the melanoma skin cancer vaccine developed by DFCI's Glenn Dranoff, MD. Created in the laboratory and first tested in animals, this vaccine works as a kind of "truth serum" for melanoma cells, forcing them to reveal their presence to the immune system. To make it, doctors removed a portion of a patient's tumor and mixed it with retroviruses engineered to carry a gene called GM-CSF. (Retroviruses are viruses that carry their genetic material as RNA.) The viruses slipped inside the tumor cells and deposited their gene cargo, causing the cells to produce the GM-CSF protein. The tumor cells were then re-injected into the patient.
GM-CSF, researchers well knew, attracts a powerful immune attack. When investigators took tumor samples from patients and examined them under a microscope, they found that the melanoma cells were engulfed by immune-system cells—evidence of a vigorous immune response.
The results of the first human trial of the vaccine, in 1998, were so impressive that researchers have pressed ahead with efforts to make the vaccine more potent and simpler to produce, and they have extended its use to other types of cancer.
To streamline the process, researchers switched from using retroviruses to adenoviruses (weakened cold viruses) to deliver GM-CSF. The advantage is that, unlike retroviruses, adenoviruses can infect tumor cells that are not actively dividing. They also release their gene payload into cells quicker than retroviruses do and tend to be safer for patients.
"With the old method, it took three months, on average, to prepare and produce a vaccine for a patient," Dranoff notes. "Now, it takes just 10 days to two weeks."
Dranoff and his colleagues have also tested the approach in Phase I clinical trials for patients with lung cancer, ovarian cancer, and acute leukemia. Two of the trials have been completed, and they demonstrated not only that the vaccines are safe (the main aim of Phase I studies), but that they elicit an immune response in the vast majority of patients, Dranoff states.
The results of the lung cancer vaccine trial prompted a pharmaceutical firm to undertake a more advanced, Phase II trial, which confirmed the earlier findings. It's likely, Dranoff says, that the vaccine will soon be tested in a Phase III trial, which will show definitively whether it can prolong the survival of lung cancer patients.
Donald Kufe, MD
Unmasking a danger
Cancer cells are masters of camouflage. The surface molecules that identify them as cancerous are, in effect, too muted to attract much notice from the immune system.
As long as cancer cells maintain their low profile, they're relatively safe from attack. Find a way to make their surface markings blaze forth in all their microscopic menace, however, and the immune system will move against them with all the forces at its disposal.
That is what researchers led by Donald Kufe, MD, of Dana-Farber have sought to do with a group of vaccines now in clinical trials.
Their approach involves the use of antigen-presenting cells, which roam the circulatory system and act as the body's customs inspectors. When they encounter another cell, they strip special molecules, called antigens, from the cell and display them on their surface. As a result, the antigens become readily "visible" to the rest of the immune system. If the antigen pattern indicates an abnormal cell, the immune system's T cells hurry in to destroy it.
Kufe (pronounced "Keefe") and his colleagues were the first to develop vaccines by fusing patients' tumor cells with antigen-presenting cells called dendritic cells, then injecting them into patients. The theory was that by unmasking the tumor cells and targeting them for attack, the dendritic cells would stimulate a particularly fierce immune reaction against tumors.
The first round of clinical trials to use the technique demonstrated that the vaccine is active in the treatment of patients with advanced cancers. The trials, conducted at Dana-Farber and elsewhere, have studied the technique's safety and effectiveness in patients with breast cancer, renal (kidney) cancer, melanoma, and brain tumors. "In all four groups, we have seen anti-tumor activity, including partial and complete remissions of tumors, with no toxicity or side effects," Kufe remarks. "The experience of the first trials has led us to believe that we can improve the vaccine."
Already, there are clues. The clinical trials showed the vaccine works best in patients who have not previously had chemotherapy, perhaps because chemo weakens the immune system, leaving the vaccine less to "work with." The Kufe lab and Japanese scientists have discovered that the substance interleukin-12 primes the vaccine to take action. And Kufe and his colleagues are exploring using more mature dendritic cells in vaccines, reasoning that they will perform better than younger ones.
"This work provides a foundation for us to improve the vaccine—to determine who is most likely to benefit and how it can be made more potent," Kufe says. "It's often in this refinement stage that a therapy's true effectiveness is realized."
A precise calibration
The ideal cancer vaccine would give the immune system a license to hunt only cancer cells. Unleashing a wider attack is risky because it might engulf healthy cells as well as cancerous ones.
One solution would be to identify features that are unique to cancer cells and can be recognized by the immune system.
At Dana-Farber, a search began six years ago for molecules that are located in cancer cells but not, for the most part, in normal cells, and that the immune system has not previously encountered. Introducing these molecules—called universal tumor antigens—to the immune system would be the equivalent of tossing a thief's sweaty handkerchief to a pack of bloodhounds. The immune system would focus its energies on tracking down and destroying tumor cells bearing those antigens.
Dana-Farber's Lee Nadler, MD, and his colleagues Robert Vonderheide, MD, PhD, and Joachim Schultze, MD, identified two such antigens. One is called hTERT, a molecule discovered several years ago by Matthew Meyerson, MD, PhD, then of the Massachusetts Institute of Technology and now of Dana-Farber. The other is called CYP1B1.
Experiments are under way to determine if either of these molecules can be used in cancer vaccines. Vonderheide (who is now at the University of Pennsylvania Cancer Center) pioneered a technique involving hTERT currently being tested by Dana-Farber's Nicholas Haining, MD. In the technique, doctors vaccinate patients with a small piece of hTERT known as a peptide in hopes of sparking an anti-tumor response.
An alternative method, also being studied, is to mix the peptide with patients' dendritic cells, revealing the peptide to the immune system. Clinical trials have shown that both techniques are safe for patients, while laboratory work indicates that the immune system does react to the peptide's presence.
Another set of experiments involves the molecule CYP1B1. The Institute's John Gribben, MD, DSc, in conjunction with Zycos of Lexington, Mass., is conducting a study in which it is given to patients as a "DNA vaccine." The vaccine is injected into patients and causes their cells to produce the CYP1B1 antigen. This is especially true for antigen-presenting cells, whose role is to alert the rest of the immune system to disease. Accord-ing to Gribben, the trial is going well, and the vaccine hasn't produced any side effects.
Nadler, Gribben, and Haining are developing a panel of universal tumor antigens that eventually may serve as targets for treating all human cancers. They plan to test techniques for making these antigens particu-larly visible to the immune system. Says Nadler, "We are uniquely positioned to accomplish this goal. In the next several years, we hope to combine these strategies in patients to 'mop up' residual cancer cells that remain after surgery, radiation therapy, and chemotherapy. The real promise for immune-based therapy is to attack the last remaining cells. This may bridge the gap between treatment and cure."
Cellular switch
The standard advice for people newly diagnosed with chronic lymphocytic leukemia (CLL) is to "watch and wait"—have their condition monitored until it progresses and specific symptoms begin to show. Even then, medication can only control, not cure, the disease.
The CLL vaccine currently undergoing a clinical trial led by Gribben is more ambitious. It seeks to send the disease into permanent remission, offering patients a new alternative to chemotherapy, the mainstay of current treatments.
Instead of trying to clue in the immune system to the presence of molecules that exist only on cancer cells, the vaccine attempts to outfit CLL cells with a molecular "switch" they don't normally possess.
"One way to think of leukemia is as a cancer of the immune system," Gribben states. "Normal white blood cells have a 'ligand'—a surface molecule called CD40—that switches on the immune system's response to disease. When those cells are leukemic, the CD40 switch is missing."
To restore it, Gribben's technique uses a retrovirus that carries the DNA instructions for producing CD40. The vaccine is made by taking a blood sample from a patient and mixing leukemic blood cells with the retrovirus. The retrovirus inserts its DNA into the cells, causing them to manufacture the CD40 ligand. Patients receive 10 injections of the vaccine over several months.
The results of the Phase I trial are promising, Gribben reports. "All of the participants experienced a drop in their white blood cell counts after one vaccination," he says, a key sign that the immune system had responded to the vaccine. Although the levels later rose, Gribben hopes the regimen will prove more effective in a similar trial at DFCI, Brigham and Women's Hospital, and Massachusetts General Hospital.
"Future trials will help us determine if the immune response sparked by CD40 is enough to cure the disease, or if it will find a way of returning."
This story first appeared in the Fall/Winter 2002 issue of Paths of Progress magazine.
