Journal of the National Cancer Institute Advance Access originally published online on February 12, 2008
JNCI Journal of the National Cancer Institute 2008 100(4):237-238; doi:10.1093/jnci/djn030
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© Oxford University Press 2008.
NEWS |
Rethinking Cancer Vaccine Trials: Would New Measures of Success Make a Difference?
Jeffrey Schlom, Ph.D., is wary of what he calls "paradigm paralysis" for judging the value of a cancer vaccine. To date, no therapeutic cancer vaccine has earned U.S. Food and Drug Administration approval. Schlom, chief of the laboratory of tumor immunology and biology at the National Cancer Institute, hopes to change that by honing a different view of success.
|
Writing last year in Clinical Cancer Research, Schlom and his colleagues argued that even though cancer vaccines have yet to show evidence of spurring a patient's immune system to shrink tumors, patients receiving them (mostly in randomized phase II trials) tend to live longer and respond better to chemotherapy or hormone treatment. Such observations have Schlom and others questioning if it's more appropriate to think about vaccine effectiveness as how patients respond rather than by how much a tumor shrinks.
"The classic criteria for measuring the efficacy of a therapeutic may not apply to vaccines," Schlom said. "It appears in [cancer vaccine] trial after trial that very little shrinkage is going on, yet patients are living longer." According to these standard criteria, called RECIST (Response Evaluation Criteria in Solid Tumors), a complete response means the disappearance of tumors, whereas a partial response is tumor shrinkage of at least 30%. Although using RECIST as a surrogate endpoint for survival may work well enough for cell-killing treatments, doing so doesn't always translate well to other types of therapies, the authors said. Instead, they suggest taking into account the immune system response, for example, and other measures such as time to progression or treatment failure.
For example, look at a randomized, placebo-controlled phase III trial of the targeted therapy sorafenib for advanced renal cell carcinoma. Progression-free survival doubled for patients taking the therapy, yet the tumor response rate was less than 10%. Or the opposite: A comparison between high-dose and low-dose interleukin 2 for metastatic renal cell carcinoma showed much more tumor shrinkage with high-dose interleukin 2 but no evidence of better survival.
"There are many descriptive cases, rather than published data, where people look like they are progressing on an agent according to X-rays, but they feel fine and don't go on to anything else," said Elizabeth Jaffee, M.D., an oncology professor at Johns Hopkins University in Baltimore. "Eventually their tumor regresses, suggesting that there was an immune response. More and more of that is being described. We have targeted biological therapies that sometimes keep the tumor from progressing. If the patient is otherwise feeling well, there's no reason why you wouldn't call that a success."
In the article, Schlom and his colleagues reviewed five prostate cancer vaccine trials in which patients who received vaccines may have responded better to chemotherapy or hormone treatment. Schlom lacks conclusive evidence––the trial endpoints were tumor shrinkage, not long-term survival––but he thinks that the vaccines are somehow "priming" the immune system, which either keeps the tumor in check or makes therapies more effective. He's worried that the vaccines could be abandoned as dead ends because they didn't reach their primary endpoints, despite their possible value in keeping patients alive longer.
"Right now, the way studies are done, [vaccines] would probably be thrown out without going on to further testing," said Howard Kaufman, M.D., chief of the division of surgical oncology at Columbia University in New York. "There might be a lot of potentially useful vaccines that aren't getting a fair look."
New Endpoints
Changing the thinking about evaluating vaccines’ effectiveness can be tricky. Kaufman admits that deciding on trial endpoints is hardly an exact science. While "we have to ultimately look at survival," he said, tumor shrinkage isn't a failsafe surrogate. "If a drug shrinks a tumor, we move it forward, though many drugs will shrink tumors and the patients still don't live any longer." For vaccines, time to progression or treatment failure might be better surrogates for overall survival.
"The ideal endpoint would be an immunological response that correlates with clinical response," Kaufman said. "Over the last 2–3 years, either the vaccines have gotten better or the assays have become more sensitive, because we're beginning to see a correlation between immunological endpoints and clinical outcomes. If that's true, then you have an artificial surrogate endpoint that you can study and you use to predict which vaccines are best to take into the adjuvant or prevention setting."
But some experts are skeptical of using endpoints other than tumor shrinkage in cancer vaccine trials. Immunotherapy researcher Steven Rosenberg, M.D., Ph.D., chief of NCI's surgery branch, is one of those skeptics. "In oncology, improved survival and quality of life are the goals," Rosenberg said. "Shrinkage of tumors as a result of therapy is generally a rapid surrogate of improved survival. Stabilization of tumor growth can often occur as a part of the natural history of the tumor in a patient, and if stabilization is to be used [as a response criterion] then it must be demonstrated in a randomized controlled trial."
|
Cancer as an increasingly chronic, stable disease is a relatively new phenomenon, and many advise caution before deciding what this new perception may really mean. "Short of a randomized, prospective trial, we still need some objective way to measure whether something good is happening or not" from a therapy, said Michael Lotze, M.D., vice chair of research in the department of surgery at the University of Pittsburgh. "Softer measures haven't been good in the past."
And even using patient survival––the so-called ultimate endpoint––doesn't satisfy everyone studying these phase II trials. Many drug companies resist using survival because measuring survival versus using surrogate outcomes usually takes much longer and increases costs, said Paul Chapman, M.D., a melanoma vaccine researcher at Memorial Sloan-Kettering Cancer Center in New York.
Combination Therapies
Gauging the effectiveness of a cancer vaccine and its supposed subtle systemic effects—by whatever measure—is no easy task. Endpoints aside, how vaccines will best be used is still open to debate. Kaufman thinks that cancer vaccines can be used as cancer prevention tools in high-risk individuals because the vaccines appear to be safe. "Prevention is where they could be most effective today, and we can identify high-risk individuals," he said.
Many see patients who are at high risk for developing metastatic disease, such as those with advanced prostate cancer, as the perfect population for a vaccine trial because a vaccine might help prevent recurrence. But such trials can be long and costly. "From an immunotherapy point of view, it would be an ideal group," said Walter Stadler, M.D., director of genitourinary oncology at the University of Chicago, "but it would take 2,000 patients in a trial that lasts 8 or 9 years." What's more, in prostate cancer there's no standard treatment if the disease returns again.
Another complication is that vaccines are increasingly being used with other therapies to treat cancer. As scientists' understanding of the immune system grows, determining the interactions among therapies, natural substances, and the immune system will be a key to developing appropriate vaccine trial endpoints. Although vaccines have been less effective in patients who have undergone many prior rounds of chemotherapy, researchers are beginning to believe that there are benefits to combining vaccines with such therapies.
"We’ve shown that combining vaccines with other therapies can be effective," Schlom said. While "some of the current trials are starting to use survival as an endpoint, and there's still resistance about combination therapies because they are so new, the next decade will see a lot of combining immunotherapies with other treatments."
The strategies differ. Some trials give the vaccine first, with a goal of boosting the immune system. "When you give chemotherapy, and radiation following that, there have been studies showing that patients do very well," Schlom noted. Others are looking at chemotherapy first to reduce some of the tumor and possibly enhance the vaccine's effects. Some preclinical models have indicated that chemotherapy and radiation can alter the phenotype of the tumor cell, making it more susceptible to vaccines.
Jaffee is testing a pancreatic cancer vaccine on patients after surgical resection, in the hope that the less cancer is present, the less tumor is available to inhibit the immune response to a vaccine. But even then, the tumor has several ways of inactivating vaccine-induced immune responses. "The only way to get an effective cancer vaccine is to involve other agents to try to turn off some of these inhibitors of immune response and vaccine potency from the cancer," she said.
At the same time, dozens of substances—including cytokines, monoclonal antibodies, and inhibitory factors—may boost vaccine effectiveness. Chapman noted that melanoma cells also tend to express fewer human leukocyte antigen molecules, becoming difficult for the immune T cells to see. "We're learning about new additional resistance mechanisms such as inhibitory elements of the immune system like regulatory T cells that suppress immune response against cancer," he said. "As we learn about the immune system, we see there are substantial hurdles to overcome."
With this knowledge, researchers are in a better position to design studies, Kaufman said. "What's missing is a critical reevaluation of the endpoints of these trials and how we use statistics in these trials. We should find ways to do smaller trials to test the therapies as quickly as possible."
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||