NCHR Testimony at FDA Meeting on Developing Antibiotics for Bacteria That Rarely Cause Infection

Thank you for the opportunity to speak today. My name is Dr. Stephanie Fox-Rawlings from the National Center for Health Research. Our center analyzes scientific and medical data to provide objective health information to patients, providers, and policy makers. We do not accept funding from drug and medical device companies, so I have no conflicts of interest.

We appreciate FDA and drug sponsors working to determine appropriate methods for testing new drugs for rare bacteria.

Even though good quality superiority trials are challenging, we should not lower the standards for trials and data. Well-designed trials are needed to make sure a new drug actually helps patients.

Fortunately, when a drug is highly effective, the trial doesn’t need to be large to show a significant improvement. For example, a company called Achaogen recently reported a statistically significant 28% reduction in death even though they enrolled only 17 patients in the test group and 20 in the control group.

The goal in developing new antibiotics is to make sure they actually improve the health of patients with the targeted infections compared to drugs that are already available. It is dangerous to approve new drugs that are not as safe and effective as the antibiotics already on the market or drugs that are not studied on patients with the targeted bacteria.

Non-inferiority trials for antibiotics are resulting in the approval of numerous drugs that may be less effective than previously approved antibiotics. It is not ethical to give these drugs to patients that have more effective options. After several rounds of comparing new drugs to somewhat older drugs that were slightly less effective than previously approved drugs, we can end up with new antibiotics that are much less effective than the best available. This is more likely when clinical trials use larger non-inferiority margins or margins equal to the estimated treatment effect.  Wider non-inferiority margins increase the likelihood that the new drug is less effective than the approved drug.

The development of rapid diagnostics would help. In many cases, researchers lack tools to quickly diagnose patients with target bacteria. This means studies are conducted on many patients who do NOT have the targeted microbe. In some cases, most of the patients don’t have the targeted microbe.  This increases the number of patients required in a trial and makes the trial outcomes much more difficult to interpret. Healthcare practitioners run into similar problems when they must decide which antibiotic to prescribe a patient. This trial-and-error exposes patients to increased risk of adverse events from multiple drugs while also delaying appropriate treatment. In contrast, the development of rapid diagnostics would help researchers study the appropriate population and healthcare practitioners prescribe the drug that is most likely to help.

Unfortunately, the Limited Population Pathway of the 21st Century Cures Act was written in a way that could easily increase the number of antibiotics that do not benefit patients. This exacerbates what was already a problem.  Of the 61 new antibiotics approved between 1980 and 2009, 43% were later withdrawn in part due to safety or efficacy reasons (Outterson et al 2013). This rate was about 3 times as often as other drugs from the same period. Unfortunately, smaller clinical trials for limited populations could make this worse because it would increase the risk that results are due to chance rather than proven.

As you all know, once approved, the new drugs are often promoted and prescribed for a much wider patient population than was targeted. This can expose patients to unnecessary risks, lower effectiveness, and generate resistant bacteria. Simply labeling a drug as “Limited Population” is unlikely to be sufficient to limit a drug’s use to appropriate patients.

Developing antibiotics for single bacterial species that infrequently cause infections is difficult. We as a society and as patients want these treatments. However, having ineffective drugs marketed for these specific bacterial species does not help patients and may harm them, in addition to contributing to healthcare costs, which are already higher in the US than other countries where people live years longer. If the FDA wants to be more patient-centered, it needs to ensure that new antibiotics actually work for the intended patients before they are approved. New drugs should be scientifically tested on patients who know that they are participating in a well-designed clinical trial that contributes to knowledge, not by patients who think that they are receiving a proven treatment.