New Trial Approach Needed for Targeted Cancer Therapies

Roxanne Nelson

September 28, 2011

September 28, 2011 (Stockholm, Sweden) — Randomized controlled trials are the gold standard in medical research, but for new targeted therapies, a different strategy might be needed.

Traditional large-scale clinical trials are not the most effective means of testing targeted therapies, which are geared to smaller and more specific patient populations, according to an analysis presented here at the 2011 European Multidisciplinary Cancer Congress (EMCC).

Over the long term, gains in survival can be improved in cancer patients by running smaller and faster clinical trials with less stringent evidence criteria, explained lead author Marie-Cécile Le Deley, MD, associate professor of clinical epidemiology and biostatistics at the Institut Gustave-Roussy, Villejuif, France.

"Common cancers are more and more frequently recognized as consisting of small subsets with particular abnormalities," said Dr. Le Deley in a press briefing that highlighted the study. "These abnormalities may be targeted with specific therapies."

Traditional clinical trials generally compare a potential new treatment with standard therapy in an attempt to establish the superiority of the experimental agent, she explained.

"Traditional trials are designed to avoid wrong decisions," Dr. Le Deley pointed out. They require the recruitment of many patients, and large clinical trials can take many years to reach definitive results, especially if the disease is rare.

New Strategies Needed

The ability to test promising targeted therapies is hindered by the need to invest huge resources to run a single large trial that might continue for a number of years.

New strategies and a more flexible process are needed to test drugs targeted at the many small subsets of cancers. "What happens if we reduce the sample size of the randomized clinical trials and if we relax the evidence criteria used to adopt an experimental agent as the new standard," she asked.

Simulations Show Gain in Survival

To answer that question, Dr. Le Deley and colleagues compared the performance of different trial design strategies. The researchers simulated a series of 2-treatment superiority trials that took place during a 15-year period.

The trial parameters used in the simulations included the number of positive trials to establish superiority (1 vs 2), the α level (.025 to .50), and the number of trials over the 15 years (the trial sample size).

Different disease scenarios, median survivals, accrual rates, and distributions of treatment effect were also studied. The metrics used included gain in survival rate (hazard ratio [HR], year 15 vs year 0) and the risk for an overall effect of harm (HR > 1).

For all of the proposed scenarios, the authors observed that overall gains were greater using the criterion of 1 positive trial than 2, and as the α level increased from .025 to .50. They noted that gains increased substantially as the α level increased from .025 to .20, but tended to reach a plateau for levels beyond .20.

Overall, gains in survival were achieved with sample sizes that were smaller than those under the traditional criteria required for clinical trials, Dr. Le Deley explained. The results were consistent under different assumed distributions for treatment effect, and the greatest gains were achieved in disease scenarios with the smallest expected median survival.

"The downside of this approach is that we also reduce the certainty of the findings," she explained. But this can be offset by the fact that many more clinical trials can be conducted and errors can be quickly corrected.

This novel approach to trials is not meant to replace the traditional design; there will still be a need for classic large-scale clinical studies in oncology, Dr. Le Deley said. But when appropriate, smaller and more targeted trials could increase the number of studies and move more quickly in recruiting patients, reporting results, and getting new therapies to patients much sooner.

Because the findings would be less definitive, the results could lead to a new temporary standard being set that would end up not working better than the existing best therapy. "The fact that we will conduct many more trials will allow such errors to be quickly remedied," she said.

She also pointed out that many of the new targeted agents have fewer safety issues than the older cytotoxic treatments. "We feel that the risk of accepting a few therapies that offer no benefit — but with a very low chance of a truly poorer outcome — for a greater long-term benefit at the end seems reasonable," Dr. Le Deley added.

"Rush Slowly"

This is a strategy of "rush slowly," said Anne-Lise Børresen-Dale, PhD, cochair of the EMCC scientific program and moderator of the press briefing, "and an idea long in coming."

However, working with regulators could be an uphill battle, Dr. Børresen-Dale, who is from the Department of Genetics, Institute for Cancer Research, the Norwegian Radium Hospital, in Oslo, pointed out. "They may be reluctant at first, but we need to work with them now, to get started."

She agrees that the patients who will probably derive the most benefit from these small trials are the ones with rare cancers for which there are no or a very limited number of therapies.

"Other patients may want to participate as well," Dr. Børresen-Dale said. "They may be interested in being part of a trial that will show quick results and a possible benefit for them."

2011 European Multidisciplinary Cancer Congress (EMCC). Abstract 1206. Presented September 26, 2011.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.