Stimulating the immune system to destroy tumor cells has long been a hope—but judging from past studies, perhaps a dashed one. Clinical trials testing various cancer vaccines have failed miserably; in one, a melanoma vaccine called Canvaxin did not improve the survival of patients, an outcome that ultimately forced the drugmaker to sell itself to another firm. But rather than writing off cancer immunotherapy, some researchers argue that the agents have been examined in the wrong way, resulting in erroneous conclusions. With the correct study design, proponents say, cancer vaccines should prove to be promising.

Such optimism arises from data that have surfaced in the wake of failed tests. After phase III trials, reported in 2006, ended in disappointment for the prostate cancer vaccine Provenge (made by Dendreon in Seattle), subsequent analyses revealed that men whose prostate cancer had spread survived a median of 4.5 months longer than those given a placebo. Patients who took the vaccine and went on to receive chemotherapy survived even longer: a median of 34.5 months, versus 25.4 months for patients who received the placebo followed by chemotherapy.

With standard clinical trials, the criteria used to confirm benefit do not apply to vaccines, points out Jeffrey Schlom, chief of the Laboratory of Tumor Immunology and Biology at the National Cancer Institute. He explains that, unlike traditional cancer drugs, vaccines do not shrink tumors. Therefore, measuring response in those terms, as most trials do, will always show them to be ineffective. Instead, Schlom says, “what we’re seeing is increases in patient survival, as opposed to tumor shrinkage.”

Moreover, experimental cancer agents are usually tested in patients who have already received several prior therapies, which blunt the immune system, Schlom remarks. This dulling is irrelevant when it comes to testing new drugs, but it impairs the ability of a vaccine to elicit an immune response. Additionally, because vaccines work better with repeated dosing (booster shots), their benefit may not appear until much later than anticipated.

Armed with the knowledge of how to study vaccines appropriately, scientists ideally should have no trouble conducting proper trials. But in reality, problems persist. One of the major hurdles is the lack of available adjuvants, drugs that augment the activity of the vaccines. Given without an adjuvant, a vaccine is unlikely to provide any real benefit. But finding appropriate adjuvants is difficult, because the Food and Drug Administration does not approve new adjuvants alone, explains Martin “Mac” Cheever, director of solid-tumor research at the Fred Hutchinson Cancer Research Center in Seattle. As Cheever notes, the FDA only approves adjuvants for use with the vaccine with which it was tested.

Consequently, researchers cannot access approved adjuvants with experimental vaccines, because each is already locked up with a specific vaccine. Some adjuvants have been tested as therapy in and of themselves, but they have rarely proved effective and have usually been shelved. “Companies in general are not willing to develop compounds that only work as components of other people’s products,” Cheever comments. This situation leaves vaccine investigators stuck with inadequate study designs.

To be sure, scientists have zeroed in on some promising vaccine-adjuvant combinations, and a few well-designed trials are now under way. GlaxoSmithKline (GSK) is testing its MAGE-3 vaccine in a phase III trial with 2,300 lung cancer patients who have had their tumors surgically removed but received little or no other therapy. The study size requires screening about 10,000 patients for the presence of the MAGE-A3 antigen that the vaccine targets. “A lot of studies are underpowered and not controlled appropriately,” says Vincent Brichard, who heads GSK’s antigen-specific cancer immunotherapeutics program. “With this number of patients, there will be no ambiguity.” Patients will receive the vaccine plus three adjuvants. The study, which will not be completed for at least five years, will evaluate whether the vaccine prevents tumor recurrence rather than causing shrinkage. Other promising vaccine trials include those for WT1 for leukemia (GSK) and Onyvax-P for prostate cancer (Onyvax in London), on top of ongoing studies of Provenge.

Some researchers remain skeptical. One concern is that some vaccines target naturally occurring proteins that the immune system would not normally attack. “We are asking the body to override these existing control mechanisms,” says Mark Kelley of Vanderbilt-Ingram Cancer Center in Nashville, Tenn., who was involved with the failed Canvaxin studies. This “self” immune response could cause autoimmune disease as a side effect. Kelley also worries about the fundamental complexity of vaccines, which makes it harder to achieve full medical potential than with other cancer therapies, he feels.

Schlom, who lived through a decade of negativity over monoclonal antibodies, sees much of the nay-saying as par for the course. When it comes to new medicines, “there is always a period of skepticism,” he recounts. “I have absolutely no doubt that several years from now there will be several vaccines approved for several cancer indications.”

A Cancer Vaccine’s Mystery Deaths
Proponents of cancer vaccines note that past clinical failures nonetheless have shown that the vaccines lengthened patients’ survival times. But that was not the case this past August, when Cell Genesys in South San Francisco halted a clinical trial of the prostate cancer vaccine GVAX when it discovered that more patients receiving the vaccine plus chemotherapy had died (67 deaths) as compared with those receiving chemotherapy alone (47 deaths). The reasons are under investigation, but the 408 patients in the study had the most advanced prostate cancer possible, “exactly the patient population for which [one] would not use a vaccine,” remarks Jeffrey Schlom, chief of the Laboratory of Tumor Immunology and Biology at the National Cancer Institute. An ongoing companion study of GVAX in patients with less advanced prostate cancer should help clarify this issue.

Note: This article was originally published with the title, "Confidence Booster".