Turning the immune system against cancer

Aug 4 2006

Starting with leukemia antigen
Testing personalized vaccines
Expanding melanoma's immune army
Glioblastoma
Learning to Turn the Human Immune System Against Cancer

Medicine can now prevent a host of diseases with a mere shot of vaccine. Polio and smallpox are almost non-existent, and mumps and chicken pox are rarely seen nowadays.

And for the first time, the prospect of eradicating a specific cancer through vaccination is possible. The newly approved human papillomavirus (HPV) vaccine is designed to curb the 230,000 worldwide deaths due to cervical cancer, which is caused solely by HPV. And the hepatitis B virus, responsible for 70 percent of all liver cancer deaths, is also preventable with a vaccine.

Cancer researchers are working on the next era of vaccines designed to treat cancer that has already developed. These vaccines don't rev up the human immune system to attack an invading microbe, but prime the system to go after a unique biological tag found only on tumor cells.

For example, brain cancer researchers at The University of Texas M. D. Anderson Cancer Center are testing an experimental vaccine that homes to a protein studding the surface of glioblastoma cancer cells. It tricks the body into thinking this protein is foreign and infectious, which alerts killer immune cells. The same kind of strategy is producing very promising results in clinical trials at M. D. Anderson of vaccines for advanced myeloid leukemia as well as other forms of leukemia, aggressive lymphoma and melanoma.

Because of the preliminary nature of therapeutic cancer vaccines - none has yet been approved for use anywhere in the world - researchers can only describe their findings as "promising." But their hope in the therapy is clear. In fact, several M. D. Anderson vaccine clinical trials have shown strong anti-tumor activity and one produced the first clinical demonstration that a vaccine could produce complete molecular remission - meaning, no biological evidence of cancer remained in some treated patients.

This wealth of cancer vaccine research at M. D. Anderson - possibly the most varied and advanced in the nation - has come about because of the strong interest from M. D. Anderson physicians and researchers in basic immunological science. They believe that the human immune system can be used against cancer, and that vaccines may represent the cutting edge of immunologic cancer treatment.

"Hypothetically, once the immune system has been sufficiently stimulated, it would be able to find and destroy every single tumor cell throughout the body," says Yong-Jun Liu, M.D., Ph.D., chair of the Department of Immunology and director of the Center for Cancer Immunology Research (CCIR).

"It could do this without destroying healthy tissue," he says. "That's the goal we strive every day for." Liu and other CCIR researchers believe that, ultimately, the best use of such vaccines will be to eliminate the minimal disease that remains after initial cancer therapy.

"When there is too much disease, the immune system is overwhelmed and a cancer vaccine may not be helpful," says Jorge Cortes, M.D., a professor in the Department of Leukemia. "But after patients have been treated, there is often a low level of disease, so our idea is that we can add a vaccine at that point to eliminate the cancer before it has a chance to grow back again."

"This is an exciting time in cancer research, given our increased understanding of the molecular nature of cancer and the immune response," says Patrick Hwu, M.D., chair of the Department of Melanoma. "Our ultimate success will likely depend on the rational combination of appropriate chemotherapies, targeted therapies, and immunotherapy, such as therapeutic cancer vaccines."

Starting with leukemia antigen

All of the cancer vaccine tactics being developed and tested at M. D. Anderson aim to mislead the cancer patient's immune system into thinking it is attacking bacteria or a virus (see notes). They are designed to strengthen the body's natural defenses against a cancer that has already developed, and this could stop an existing tumor from growing further, prevent cancer from coming back after it has been treated, or eliminate cancer cells not killed by previous treatments.

These efforts are concentrated on a subset of cancers that M. D. Anderson researchers believe are most amenable to cancer vaccines - the leukemia and lymphoma blood-based cancers (because red and white blood and lymph system cells are easier to reach and manipulate) and a few solid tumors, including melanoma, which seem to elicit a natural, if weak, immune response.

In 2003, recognizing this focus, M. D. Anderson opened its Center for Cancer Immunology Research, which is believed to be the first comprehensive program in the United States in which both basic and clinical immunologists work together in open laboratory environments to develop immunological treatments for cancer.

The four M. D. Anderson departments that collaborate within the Center for Cancer Immunology Research - Immunology, Lymphoma and Myeloma, Melanoma, and Blood and Marrow Transplantation - have built up strong basic and translational research programs to develop novel vaccines, manufacture them in on-site cell processing facilities, and test them in patients.

Among the scientists who are leading this effort are Larry Kwak, M.D., Ph.D., chair of the Department of Lymphoma, and Patrick Hwu, the Department of Melanoma chair, both of whom were recently recruited from the National Institutes of Health (NIH). They join Liu, a leading immunologist who came to M. D. Anderson in 2002 from the biotech and pharmaceutical industry, and Jeffrey Molldrem, M.D., professor in the Department of Blood & Marrow Transplantation, who helped pioneer cancer vaccine research at M. D. Anderson.

"We have a number of homegrown vaccines, pioneered in research laboratories at M. D. Anderson, that are true examples of translational research," Kwak says. "The Center for Cancer Immunology Research exists in part to take promising therapeutic agents from our own institutional pipeline into patient care, and this strategy is thriving."

Molldrem's vaccine, developed about 10 years ago to treat advanced myeloid leukemia, demonstrated success from the beginning, and now is being tested across M. D. Anderson in different types of leukemia that originate from myeloid blood cells.

Molldrem and his team were able to find that a special tumor antigen, which they called PR1, is over-expressed in myeloid leukemia cells. The vaccine combines PR1 with a substance that stimulates the immune system and directs T cells to kill the leukemia and leave normal cells alone. The first test of the peptide vaccine demonstrated that it could produce complete responses in some patients with advanced myeloid leukemia for whom no other therapy had been successful.

Such early success "startled" Molldrem, he says. "Initially, we were just trying to see if we could boost immunity to the antigen we had identified - we didn't expect molecular remissions, especially in a phase I trial and in such a refractory group. That's never been described before for any vaccine," he says.

Further testing has shown that the vaccine can induce complete responses in 20 percent of patients with advanced acute leukemia for whom all forms of chemotherapy had failed. Furthermore, the vaccine also appears to offer an immune memory after only three doses, says Muzaffar Qazilbash, M.D., an assistant professor in the Department of Blood and Marrow Transplantation who is leading an ongoing clinical trial.

The PR1 peptide vaccine is slated to be tested at M. D. Anderson in other kinds of leukemia of myeloid origin, such as acute myelogenous leukemia (AML), chronic myelogenous leukemia (CML), and pre-leukemic myelodysplastic syndromes.

A vaccine can even be useful in patients who respond well to treatment, says Cortes, because even the best therapies may not eliminate all traces of disease. For example, Cortes is testing the PR1 peptide vaccine in CML patients who have done well with Gleevec therapy, which keeps the cancer at bay when continuously used, but does not eliminate it. "Even with the great drugs that we now have to treat CML, we get to the point where the last bit of disease stubbornly remains, and we may need something else to keep it under control or eliminate it," Cortes says. "There is some evidence that an immune response can play a role in eliminating these cancers, so using a vaccine is a great idea."

Cortes has just opened a clinical trial testing the concept in CML patients who will continue to be treated with Gleevec. Twenty patients will receive the vaccine and another 20 will receive a combination of the vaccine and interferon, which stimulates an immune response. "Gleevec was the first major breakthrough in CML treatment, and, in my opinion, the next one will be immune modulation," Cortes says.

Testing personalized vaccines

Early vaccine research in lymphoma (a cancer of lymphocytic white blood cells) dates back to studies in mice in the 1970s when a lymphoma tumor antigen was found. That discovery led Kwak, then at the National Cancer Institute (NCI), to test a highly promising lymphoma therapeutic vaccine. First reports of its success, in 1999, demonstrated complete molecular remissions and long-term disease-free survival in 75 percent of patients who used the vaccine after chemotherapy for lymphoma.

The vaccine, now known as Biovaxid after being licensed to a biotech firm by NCI, is now being tested in a national phase III clinical trial. This study, led at M. D. Anderson by Sattva Neelapu, M.D., assistant professor of lymphoma, is testing the therapy in 460 patients with follicular lymphoma, a form of non-Hodgkin lymphoma.

The difference between this vaccine and the PR1 peptide vaccine is that it is an "idiotype" vaccine, tailored to the patient's unique tumor antigens, says Kwak, who has continued his vaccine studies at M. D. Anderson. In this individualized therapy, cells are harvested from a patient's lymph node, and the unique cancer markers on the outside of their cancer cells are identified. To create the idiotype vaccine, researchers fuse the antigen-bearing tumor cells to antibody-producing mouse cells that act as mini-factories, churning out large quantities of the protein antigens, which are then given back to patients with an immune system booster.

In a bid to see if a more "robust" cancer vaccine can be produced, Neelapu, who was part of Kwak's team at the NCI, has opened two trials at M. D. Anderson in which healthy volunteers are vaccinated. These trials, in which Sergio Giralt, M.D., a professor in the Department of Blood and Marrow Transplantation, is collaborating, are for treatment of patients with multiple myeloma who are slated to receive a stem cell transplant from a matched donor.

In this strategy, which will also be tested in patients at the NIH, stem cell donors will be vaccinated with the patient's tumor antigen. "The donor has a healthy immune system and can mount a reaction against the antigen," says Neelapu. Those primed stem cells will then be given back to the patient, "improving the chance of a graf-versus-myeloma effect," he says.

Also being tested at M. D. Anderson are two different idiotype vaccines to treat chronic lymphocytic leukemia (CLL), a cancer in which the bone marrow makes too many lymphocytes.

One of the vaccines, co-developed by William Wierda, M.D., Ph.D., assistant professor in the Department of Leukemia, with collaborators at the University of California, San Diego, uses gene therapy techniques. Tumor blood cells are extracted from patients, and then sent to a cell processing facility at M. D. Anderson, where the cells are infected with a virus that carries a gene that activates the immune system. When the cells are given back to the patients, the transformed leukemia cells manufacture the activation protein and thereby function as a vaccine, says Wierda. "Leukemia cells efficiently stimulate T cells to react against them as well as against nearby leukemia cells that haven't been infected by the virus," he says. The clinical trial is enrolling patients and will test whether a single dose of the vaccine can produce an immune reaction against the leukemia.

The second vaccine, known as MyVax and developed by Genitope Corporation, is now being tested in a phase I/II trial at nine institutions. It combines an idiotype antigen from individual patient's tumor cells with an immune stimulant derived from shellfish. "Because this trial is in the early stages, the intent is to see if the vaccine stimulates an immune response to the point that chemotherapy treatment can be delayed in these patients," says Wierda, who is leading the trial at M. D. Anderson.

"The chemotherapies that we have now for CLL are very successful in putting patients into remission, but the majority will relapse," says Wierda. "If we are able to stimulate an immune reaction against leukemia cells while there is minimal disease, it could offer us a curative strategy."

Expanding melanoma's immune army

Melanoma is one of the few solid tumors that the human immune system seems to "see," says Hwu, the Department of Melanoma chair. But that natural T cell response to an antigen on the melanoma cell is too weak to conquer a growing tumor, so researchers led by Hwu have found ways to isolate T cells from patients' tumors, grow them in large quantities in the lab and give them back to patients. This strategy, known as "adoptive T-cell transfer," differs from that taken by Kwak, because Hwu is providing more immune system "ammunition" to attack the already existing tumor antigens.

Before coming to M. D. Anderson from the NCI in 2003, Hwu found this method could shrink tumors in about half of patients with metastatic melanoma - a higher response rate than any other therapy for this advanced cancer.

His challenge, however, has been to come up with ways to grow the T cells efficiently, Hwu says. "We are trying to develop a way to make this process easier and more effective."

Hwu has begun testing what he hopes is an improved version of this approach. It involves isolating a patient's dendritic cells (immune cells that first detect microbes and alert T cell killers) and infusing them with billions of T cells that have been isolated from patients and expanded in the laboratory. Before patients receive chemotherapy to treat their melanoma, investigators take a blood sample to isolate the dendritic cells, which are then grown in the laboratory and exposed to the patient's melanoma antigen, says Willem Overwijk, Ph.D., an assistant professor in the Department of Melanoma Medical Oncology. "The dendritic cells take up the antigen and are ready to present it to T cells, which will cause the T cells to become activated," says Overwijk, who helped develop this approach.

A clinical trial testing this protocol opened in early 2006, after Hwu and his research team worked for two years with the NIH and the U.S. Food and Drug Administration on the protocol. In this phase I/II study, 50 patients will all receive a short course of chemotherapy while their cells are being grown in the lab. Half then will receive an expanded mixture of their own dendritic and T cells, while the other half will receive only T cells.

"This novel trial is so incredibly complicated and involved that few places can do it," says Overwijk. "M. D. Anderson has the infrastructure and the expertise to make it possible to do all this in house. It may take this kind of effort to make an impact in treatment of melanoma."

While adoptive T cell transfer may be required for advanced melanoma, earlier stage diseases may be treated with vaccines alone. In another trial for patients with early stage melanoma, Hwu and his team are combining a vaccine to stimulate T cells with the prostate cancer drug Lupron, which obstructs production of testosterone. It turns out, says Hwu, that when sex hormones are blocked, the thymus gland begins to produce new T cells. This is a "wonderful surprise," says Hwu, because the thymus gland largely shuts down T cell production after puberty. "The drug will produce a working thymus, which will be primed by the T cell vaccine," he says.

Glioblastoma

Another solid tumor - one that is even more treacherous than melanoma - appears to respond to a novel vaccine being developed and tested at M. D. Anderson.

This vaccine significantly increased life expectancy in patients with glioblastoma multiforme (GBM), the most dangerous type of brain tumor. The results were so surprising that the trial was stopped before full accrual, and a pharmaceutical company acquired the rights to the drug. A larger, multi-institutional, randomized study is being planned, says Amy Heimberger, M.D., an assistant professor of Neurosurgery who led the trial, conducted at both M. D. Anderson and at Duke University Medical Center.

She describes the vaccine as an easy to use "off-the-shelf" treatment that can potentially help up to 50 percent of all GBM patients keep their cancer at bay for a period of time. Interim results of the Phase II clinical trial show that the vaccine significantly delays progression of tumors until the cancer finds a new growth pathway.

According to results Heimberger presented in April at the annual meeting of the American Association of Neurological Surgeons (AANS), median survival for the 23 patients tested is at least 19 months, and only four patients have died from the cancer, That figure surpasses the median survival of 14 months for patients with GBM who are treated with the most current chemotherapy and radiation, and the 4-month median survival for untreated patients, she says.

"We can't say this vaccine is better than chemotherapy because we haven't tested the two treatments head-to-head yet," she says. "However, so far, results have exceeded the expectations we had for this vaccine."

Heimberger and a team of researchers designed the vaccine to alert the brain's immune system to the presence of just one type of protein that studs the outside of a glioma tumor. This protein, epidermal growth factor variant III (EGFRvIII), is found on 30 percent to 50 percent of brain tumors, as well as on some breast and non-small-cell lung cancers, but not on normal tissue. Heimberger believes EGFRvIII drives gliomas to spread, which could explain why these brain tumors are unusually dangerous and invasive.

The vaccine contains a synthesized piece of the EGFRvIII antigen, as well as compounds that stimulate immune system dendritic cells, which then activate the immune system in general and killer T cells in particular. "It tricks the body into thinking that EGFRvIII is foreign, and the T cells are sent in to kill the tumor," Heimberger says.

"This is a proof of concept, and optimal use of the vaccine may be with chemotherapy to further retard progression," says Heimberger. "Still, this is exciting to us because people have been trying to use immunotherapy against gliomas for a long time."

NOTES

Learning to Turn the Human Immune System Against Cancer

It is significantly more challenging to develop vaccines against cancer than against bacteria and viruses, says M. D. Anderson's Patrick Hwu, M.D., chair of the Department of Melanoma.

Preventing infection with a vaccine is relatively straightforward, he says, because it capitalizes on the body's elegant immune system, which is primarily set up to recognize and attack foreign proteins ("antigens") found on invading microbes. "We live in a sea of bacteria and viruses, and our immune system has evolved to recognize the biology of these pathogens in order to mount potent responses against them," he says.

So a traditional vaccine introduces a non-infectious "bit" of the pathogen antigen to a body that has not been infected by it before, and this brief and harmless exposure activates the immune system against the microbe and sets up a long term memory that confers lasting protection.

But cancers are "derived from our own tissues, and therefore our patrolling immune system is not programmed to recognize cancer cells as foreign," says Hwu.

"The immune system is not very willing to attack things that look like itself or the host, and there are all kinds of regulatory circuits that keep the immune system down if everything appears to be normal, even though it might not be," says Willem Overwijk, Ph.D., a translational scientist who works on melanoma vaccines.

That may be because "cancer does not really threaten the existence of the human species, as do infectious diseases, because most cancers occur after age 65," speculates Yong-Jun Liu, M.D., Ph.D., head of M. D. Anderson's Center for Cancer Immunology Research and chair of the Department of Immunology.

However, Liu adds, the existence of autoimmune disease indicates that the immune system is capable of attacking "self" tissue. "The trick is learning how to get it to attack malignant tissue instead," he says.

Another major problem is that viruses and bacteria cause a lot of damage to tissue, which produces inflammation - another red flag that stokes the immune system to act. An effective vaccine might have to do the same, researchers say.

The challenge, then, is to "create" a lasting immune response through a vaccine. While past decades of effort in this direction have only produced tepid results, experts now believe they know enough about human immunity to design rational approaches.

"Cancer vaccine strategies have a long history in medicine and that history hasn't been very favorable," says M. D. Anderson leukemia physician and researcher William Wierda, M.D. "We've gone through periods of excitement and disappointment, and now we are in a period of excitement again."

One reason, he says, is that the research community has been able to find pieces of protein antigens either on the outside or the inside of tumors that are relatively unique to cancer cells. The job now is to manipulate the immune system's dendritic sentinel cells and the destroying T cells to hunt them down and destroy them in the same way that those cells attack microbes.

"A fundamental understanding of the immune system has only been developed over the last four decades or so, and that knowledge is critical for developing cancer vaccines," says Jeffrey Molldrem, M.D., professor in the Department of Blood & Marrow Transplantation.

"For instance, one of the trickiest parts is to identify which antigens to direct the immune response against any given tumor type," he says. "There can be tens of thousands of different proteins and protein variants getting turned over at different times in a cell, so trying to identify which ones the T cell actually sees is kind of like finding a needle in a haystack," Molldrem says. "But now we have a molecular scale for understanding how it works, which is an important tool for directing immune reactions against a tumor."

Investigators at M. D. Anderson now have a toolbox of protein antigens that they can use to create vaccines. The idea is to take specific antigens that are over-expressed on certain tumor cells and manipulate them to stimulate the body's T cells to destroy all cancer cells expressing that antigen, says Hwu. Another strategy is to grow T cells which recognize the tumor, then administer them back to the patient to destroy the cancer cells.

Scientists are also working to imprint each newly primed cell with a long-lasting memory that will enable it to fight cancers as they develop, again in the same way that polio and measles vaccines work over time, Liu says. Patients can then be immunized against cancer recurrence or even against cancer development.

"We are applying principles learned from the natural immune response to pathogens in the generation of new anti-tumor vaccines," Hwu says. "As cancer researchers, our goal is to harness the power of the immune system against cancer."