Skip to Main Content

Business

March 27, 2006
Volume 84, Number 13
pp. 23-27

Body, Heal Thyself

After false starts, cancer immunotherapies tiptoe toward commercialization

Lisa M. Jarvis

At a panel discussion at the Biotechnology Industry Organization's annual CEO & Investor Conference last month, the presenters couldn't resist commenting on the title of their session: "Cancer Immunotherapies Coming of Age." Peter Working, senior vice president of R&D at Cell Genesys, went so far as to thank the moderator for ending the sentence in a period, rather than a question mark.

Antigenics Photo

Hope In A Bottle Antigenics will find out within weeks just how effectively its vaccine is helping patients with kidney cancer live longer.

The point was well-taken. Once hyped as the next big thing in the biotechnology business, cancer immunotherapy companies have fallen off the radar of the investment community. Their lofty stock prices of five years ago have come down to earth, and some industry observers wonder whether the technology can ever live up to expectations.

The theory underlying therapeutic cancer vaccines appears sound. Vaccines have traditionally been used as preventive measures; they teach the immune system how to make the antibodies it will need to fight a disease so that it is prepared for future battles. Companies pursuing therapeutic vaccines, also called active immunotherapies, believe the immune system can learn how to ward off a disease even after it has already taken hold.

No therapeutic cancer vaccines have made it to market, and several high-profile, late-stage development failures have reinforced the naysayers' convictions that they don't work. Just last year, Philadelphia-based Aphton said its cancer vaccine Insegnia did not prevent tumor growth or lengthen the life of patients with pancreatic cancer, and Carlsbad, Calif.-based CancerVax dropped its cancer immunotherapy Canvaxin from development after disappointing results in Phase III studies in advanced melanoma.

Despite these setbacks, cancer vaccine companies are optimistic that their technology is poised to come into its own, and they believe that 2006 could be a big year in terms of proof of concept. Companies point to the long roller-coaster ride that almost all new technologies must endure before achieving a true breakthrough, drawing a parallel with the tumultuous decades that preceded the launch of the first therapeutic monoclonal antibodies.

"That cycle of hyperbole and unreasonable rejection and disappointment and more measured success is unfortunately long, 15 to 20 years," says Steven A. Sherwin, chairman and chief executive officer of Cell Genesys.

Sherwin should know. When he was at the National Cancer Institute, he conducted the first trial of a monoclonal antibody in a cancer patient and rode through that cycle before finally seeing success. In the late 1980s, he says, "the press was talking about monoclonal antibodies as magic bullets for cancer." Those early MAbs, however, were mouse antibodies and produced allergic reactions in humans. Thus, they showed only limited efficacy in treating disease.

In the antibody space, the critical breakthrough came when technology evolved to enable the production of humanized antibodies. Ten years after the media frenzy over MAbs, several billion-dollar products are on the market.

The analogy with the development of MAbs is good, says Thomas B. Okarma, president and CEO of Geron, because many cancer immunotherapies that have gone through and failed late-stage trials have been based on old technology.

"Immunotherapy against cancer is not simple," Okarma explains. Cancer immunotherapies, he notes, must overcome two critical hurdles. First, cancer patients by definition are immunocompromised, which means vaccines must work extra hard to generate an adequate immune response. Second, such therapies must also break the body's tolerance of its own antigens, the proteins or other substances that stimulate an immune response.

Another major hurdle has been designing clinical trials that will appropriately measure the success of an immunotherapy. Traditionally, cancer patients have been treated with cytotoxic agents, which upon entering the body, immediately start killing rapidly dividing cells, both the sick and the healthy. This approach produces a measurable result, such as tumor shrinkage, after a well-defined treatment period.

Using that same yardstick in a trial for a cancer immunotherapy will inevitably lead to failure, says Garo H. Armen, president and CEO of cancer vaccine developer Antigenics. "Priming the immune system takes time," he notes, and teaching it to recognize and destroy cancer can take months.

"The traditional end point is not going to work," agrees Fleur Pijpers, oncology analyst at the health care consulting firm Datamonitor. Monitoring the immune response would be a valuable clinical end point for companies, she notes, but good methods of measuring that response across multiple tumor types have yet to be developed.

The natural learning curve that the immune system must navigate means that cancer immunotherapies are unlikely to be successful in patients with late-stage cancers. Yet the traditional approach with new cancer drugs has been to initiate trials in the sickest patients. Companies developing cancer vaccines are now figuring out that they need to choose patient populations with a better survival outlook in order to show efficacy.

The body's learning curve leads to yet another hurdle: determining just how many months it takes for the immune system to start to combat cancer on its own. With few precedents to follow, companies have had to figure out a dosing regimen and whether to give booster shots to keep the immune system on track.

It is also unclear whether a cancer vaccine will ever actually induce tumor shrinkage or, as many believe is more likely, only keep the tumor from growing so that a patient can essentially live with the cancer indefinitely. As they are doing with the newer monoclonal antibody drugs, cancer vaccine companies are generally measuring their product's ability to prolong survival rather than its impact on tumor size.

Despite the complexity of teaching the body to, in essence, heal itself, companies are confident that they have managed to overcome most of the hurdles. At least 12 cancer immunotherapies are in Phase III trials and dozens more are in Phases I and II.

"If you look at where vaccines are today, we have very substantial, good results in Phase II trials with a variety of different therapeutic vaccines," says Alex McPherson, president and CEO of Biomira. He points out that some analysts expect the cancer immunotherapy market to be worth $6 billion by 2010.

Datamonitor has come up with the less heady figure of $600 million in therapeutic cancer vaccine sales in 2010. However, although she is cautious about the near-term prospects, Pijpers believes that as technology evolves, significant commercial breakthroughs could easily raise that forecast.

Regardless of the current lack of commercial success, the developmental pipeline has become crowded with cancer immunotherapies. Datamonitor found at least 64 companies active in the space that offer a huge range of approaches to teaching the immune system to combat cancer.

The products that have made it to late-stage trials can be broken down into three general categories: patient-specific vaccines, which are made from the patient's own tumor; whole-cell vaccines, which address a wide range of antigens; and antigen-specific vaccines, which target one particular antigen that is believed to be responsible for cancer growth. The first method creates a personalized vaccine, whereas the latter two generate off-the-shelf vaccines that can be used on the general population.

Seattle-based Dendreon, with its personalized approach, is widely viewed as the front-runner in the race to get a cancer vaccine to the market. The company is preparing a Biologics License Application (BLA) for Provenge, which has been granted fast-track status by the Food & Drug Administration for use in men with metastatic, androgen-independent prostate cancer.

Provenge is in a sense both personalized and general, says David L. Urdal, Dendreon's chief scientific officer and vice chairman. The personalized component comes from a blood sample taken from a patient in order to isolate dendritic cells, which deliver antigens to T-cells and B-cells, the lymphocytes that control the immune response. Those dendritic cells are then cultured with the general component, prostatic acid phosphatase, a recombinant protein found in 95% of prostate cancers.

The dendritic cells, now loaded with the antigen, are then injected back into the patient to trigger a reaction from the immune system. Patients receive three doses of the vaccine over the course of a month.

The method appears to be offering patients some hope. The combined data from two Phase III trials in androgen-independent patients showed that the vaccine extended survival by four months, a 23% improvement over a placebo. And just over twice as many patients given the vaccine (33%) compared with those given a placebo (15%) were still alive 36 months after treatment.

Dendreon has also completed enrollment in a Phase III trial in patients with the earlier stage and milder androgen-dependent prostate cancer. "The idea is that Provenge would be most effective when the tumor burden is the smallest," Urdal notes. There has also been evidence that administering the vaccine prior to chemotherapy rather than after it might be an effective sequence.

As the first player to ask FDA for approval for a cancer immunotherapy, Dendreon could be facing a tough advisory panel. The company must be armed with the data to address the slew of new issues related to clinical trial design and manufacturing. "When you're in the lead, you're marching, but you also have a lot of arrows in your back," Urdal acknowledges.

On the other hand, there is also the potential for an enormous upside, he points out. Being first to market not only allows a company to capture market share, but it also means setting the standard for other cancer immunotherapy companies to meet or beat. Needham & Co. stock analyst Mark Monane says that in the U.S. alone, roughly 100,000 prostate cancer patients could benefit from Provenge. He expects it to be launched in mid-2007 and to hit peak sales of $1 billion in 2012.

Dendreon has, in a sense, a hybrid of a personalized and general vaccine, whereas New York City-based Antigenics is taking a purely personalized approach. The company's lead candidate, Oncophage, is made entirely from the patient's own tumor in order to tap into cellular regulatory components called heat-shock protein-peptide complexes.

Inside a healthy cell, heat-shock proteins (HSPs) are worker bees, helping to fold proteins and to chaperone abnormal peptides out of the cell. But when HSPs themselves are found outside a cell, it is a flag that the cell is dead and triggers an immune response against the disease or infection that killed the cell.

Antigenics tries to mimic flags unique to each patient by isolating from their tumor an HSP called gp96 and the peptides associated with it, then returning the HSP-peptide complexes to the patient to stimulate the immune system to target those specific cancer cells. Patients receive a 12-dose regimen of the vaccine, getting one shot per week for a month, then every two weeks for two months.

Antigenics will soon find out just how powerful this personalized approach can be. The final results from its Phase III trial of Oncophage in kidney cancer are expected to be unveiled sometime in April. At that stage, assuming positive data, Antigenics will discuss with FDA the best path forward to gain marketing approval.

There are limitations to Antigenics' approach. The patient's tumor must be both operable and sufficiently large to enable the production of the vaccine, conditions that could limit the types of cancer that the treatment can address. Furthermore, significant challenges in maintaining sterility conditions are associated with transporting a tumor sample after surgery to the company's facility in Lexington, Mass., and then back to the patient, not to mention the cost of an individualized product.

"No question that making something that's customized is more onerous than making something off-the-shelf," Armen says. However, he points out that Oncophage's manufacturing process does not involve growing cell culture, which significantly lowers the fixed costs of making the vaccine. Armen expects Oncophage to be priced similarly to the monoclonal antibodies that are currently marketed for cancer.

If Antigenics is offering the customization of a Bentley, technology from Edmonton, Alberta-based Biomira enables the assembly-line approach of the Model T Ford. Biomira's vaccine targets a specific antigen that is overexpressed in specific tumor types but in the package of a generalized vaccine that the company says is less expensive to manufacture.

The company's liposome-based vaccine technology involves embedding in an artificial membrane both a tumor-specific antigen and an immune-system-stimulating molecule that acts as an adjuvant. Given separately, the antigen and adjuvant are not potent enough to provoke the immune system, "but as soon as you put them together into the housing of the liposome, you see a massive response," says Peter Emtage, Biomira's vice president for R&D and technical operations.

The most advanced liposomal vaccine to come out of the company's program is L-BLP25, which targets MUC1, a protein found on the surface of non-small-cell lung cancer (NSCLC) cells. Preliminary data from a Phase IIb trial showed that the vaccine was able to extend average survival in NSCLC patients to 30.6 months, compared with 13.3 months for the unvaccinated group.

In January, Biomira licensed L-BLP25 to Merck KGaA of Germany. Merck took all financial and developmental responsibility for the vaccine, including executing a Phase III trial in lung cancer, expected to begin enrollment later this year.

Meanwhile, rather than using a single agent from within the cell as Biomira does, South San Francisco-based Cell Genesys believes that using whole cells is the key to evoking a strong enough immune response to actually affect patient survival. "Tumors change while they grow," Sherwin, the CEO, explains. "If you pick one antigen and it mutates off the cancer cell or gets physically shed, then your immunotherapy has produced an immune response against a molecule that is no longer on the tumor cell."

Like Dendreon's Provenge, Cell Genesys' main product, GVax, works by activating dendritic cells. But instead of harvesting those cells from the patient's blood as Dendreon does, Cell Genesys has devised an off-the-shelf product by genetically modifying whole tumor cells to secrete granulocyte-macrophage colony-stimulating factor (GM-CSF), a hormone that jump-starts the immune system's ability to recognize and kill cancer cells. The product is then irradiated to make it safe to be given to patients.

Cell Genesys Photo

Proposed Mechanism After Cell Genesys's GVax is injected into a patient with prostate cancer, granulocyte-macrophage colony-stimulating factor secreted by genetically modified tumor cells in the vaccine (light spots) activates the immune system (dotted circles) to respond to the cancer cells (dark spots).

The most advanced studies of GVax are in prostate cancer, with two Phase III trials comparing it with Sanofi-Aventis' Taxotere, the current standard of care. Phase II studies showed that the product extended average patient survival by roughly seven months compared with the published median survival time for Taxotere.

Cell Genesys is also conducting two Phase II trials to test whether GVax can keep cancer from recurring in patients in remission. GVax is given to pancreatic cancer patients after they have undergone surgery to remove their tumors and have completed radiation therapy.

From a manufacturing standpoint, whole-cell vaccines may be the most difficult to produce. "It is new ground, and it is challenging," Sherwin acknowledges. But confident that it can tackle the problem, Cell Genesys has invested in a commercial-launch facility in Hayward, Calif.

With companies taking a range of approaches to coaxing the immune system into attacking cancer cells, the biopharmaceutical industry is still waiting to see which vaccine method—personalized, antigen-specific, or whole-cell—will be most successful. Each category could produce marketed products within the next two to five years, a clear sign that the technology has come a long way.

Still, some industry observers believe that yet another generation of technology could prove much more powerful. Each of the current approaches has one or even multiple limitations—from manufacturing challenges, to the type of tumors that can be addressed, to the ability to evoke a strong enough immune response.

"We think the way forward would be for different companies to get together and combine forces," Datamonitor's Pijpers notes. For example, it would make sense for a company with a great adjuvant to pair with another company with a vaccine that targets multiple antigens.

So far, there is little evidence that companies are ready to team up. The closest example comes from Menlo Park, Calif.-based Geron, which, in what some would view as the ultimate validation of its technology, is collaborating with the big pharma company Merck & Co. to develop a vaccine targeting telomerase, an enzyme that plays an instrumental role in immortalizing cancer cells. The nine-month-old deal will leverage both companies' immunization platforms against Geron's telomerase target.

Merck uses adenoviruses and plasmids to target telomerase, and Geron uses dendritic cells. The collaboration enables the companies to compare their platforms in animal models and could lead to a combined approach that would elicit the most powerful immune response.

On its own, Geron's technology has already shown promise in the clinic. The company uses the patient's own dendritic cells to generate T-cells against the telomerase. Telomerase is a particularly intriguing target because it is a universal antigen—all cancers depend on it—yet it is not found in most normal cells and tissues, Okarma says.

Geron's technology also activates the dendritic cells in such a way that the T-cells attacking telomerase are generated regardless of a patient's tissue type. The end result is a universal antigen that can be used for patients of all tissue types, Okarma says.

Geron has validated the approach in Phase I/II clinical trials in prostate cancer patients. The vaccine induced an extraordinary immune response, Okarma notes, showing the vaccine's ability to break tolerance in cancer patients. Furthermore, the level of prostate-specific antigen—the marker of disease progression in prostate cancer-was reduced or eliminated after vaccination. That result shows that the T-cells being induced are actually killing the cancer cells.

Despite the positive results, Okarma acknowledges the drawbacks of Geron's current technology. Like Dendreon, Geron is using patient-specific dendritic cells. A blood sample is taken from a patient, pulsed with RNA for the telomerase protein component, and then injected back into the patient. Though Geron has made strides to reduce the cost of the vaccine, "it will always be more expensive than a vaccine made in multidose production lots," Okarma says.

To address this limitation, the company is now developing a technology platform that uses embryonic stem cells that are charged with telomerase to generate general vaccines that behave like the patient-specific vaccines.

Stem cell production "is really going to change the landscape in cancer immunotherapy," Okarma says. "You've then used modern immunology infused with modern tumor biology to create a powerful generic platform. I think then you'll really be at a point when the technology has come of age."

For cancer vaccines, the cycle of hyperbole, disappointment, and success now appears closer to the end than to the beginning. Okarma believes his stem cell plan could be less than three years away, and several other vaccines are poised to test the waters before then. "There will be more failure, but I am absolutely confident there will be success," Sherwin says. "It only takes one success to wipe away years of skepticism."

Chemical & Engineering News
ISSN 0009-2347
Copyright © 2010 American Chemical Society