August 13, 2018
Comments on: Limited Population Pathway for Antibacterial and Antifungal Drugs; Draft Guidance for Industry; Availability
[FDA-2018-D-2032]
As organizations representing patients, consumers, and public health experts, we thank you for the opportunity to respond to the FDA’s proposed new “limited population pathway.”
Everyone agrees that there is a need to develop new antimicrobials to treat serious or life-threatening infections. Unfortunately, these are often difficult to test in clinical trials. When the FDA approves new drugs that were tested on patients who did not have resistant bacteria and for whom safe and effective options already exist, it raises ethical and scientific questions regarding testing new antimicrobials.
A key issue is that just having more drug options on the market often does not help patients. An analysis of the antibiotics approved between 1980 and 2009 found that 43% (26 drugs out of 61) were taken off the market due to poor sales or safety and efficacy problems.[1] The only way to make sure that new drugs are safe and effective is by requiring well-designed and valid clinical trials.
Congressional language for this limited approval pathway (LPAD) is quite vague. However, it is important to note that the FDA’s proposed guidance includes a description of the pathway that had previously been inserted in an early version of the 21st Century Cures legislation and then intentionally deleted by the HELP Committee before the Senate passed the legislation. This indicates that Congress did not support the wording that the FDA now proposes.
The law that defines all FDA approval decisions, including the LPAD pathway, requires substantial evidence from “adequate and well-controlled studies demonstrating efficacy.” The guidance itself states that the pathway does not allow for drugs to be approved without meeting the normal standards. However, the guidance in part defines “limited population” as the subset of “patients for whom the drug could potentially be effective.” {italics added}
Very few patients have an “unmet need” – a situation when none of the drugs currently available work for their infection. That makes it very difficult to study new drugs on the most relevant patients. For that reason, the guidance suggests that an experimental drug should be tested in a broader population of patients with the intention that, if approved, the drug would be indicated for a narrower “limited population” of patients who do not have good options. This is further implied by the use of noninferiority trials, which requires comparison to a drug known to work for the condition treated in the trial. However, this would mean that the drug is not tested in the population (the one with the unmet need) for which the drug is intended.
If the drug is not tested on the population for which it is intended, it would be impossible to determine the efficacy and safety for that intended population. For example, patients with kidney problems are more likely to have resistant infections, but antibiotics are less effective in those patients.[2]
Drugs approved by testing in a more general population would not provide patients or their physicians with the evidence needed to determine appropriate treatment for patients in the “limited population.”
The questionable ethics of the proposed clinical trials are clear. Patients in the clinical trial would be exposed to the risks of the trial, including a new and potentially less safe or less effective treatment, without the likelihood that they actually need the drug even if it works for the limited population. If, in contrast, the new drug was expected to be safe and effective for the broader population – the types of patients tested in the clinical trial – then it would not need to go through the LPAD.
This presents an unethical exposure to risk for clinical trial participants. It also raises many concerns about informed consent: How will the doctor explain to the clinical trial participants that, if they are randomly assigned to receive the experimental drug, it might be less effective than a drug that is known to work?
This ethical issue is compounded because the new drug might be approved as more effective than other drugs for the “limited population” when there is actually no evidence to support that claim. To make that claim of superiority, the drug must demonstrate statistical and clinical superiority in clinical trials for the limited population. That is not possible in the types of clinical trials being proposed, and to make that claim is scientifically erroneous and ethically inappropriate.
The proposed guidance describes the use of small clinical trials that are not powered to test meaningful efficacy and safety endpoints or that use wide noninferiority margins. Intentionally designing a study that cannot be used to evaluate efficacy and safety is unethical for patients in the trial, since they are exposed to the risks without the potential for benefit for themselves or others. It is a wasteful use of research resources and can be potentially harmful for patients.
It is also important to note that FDA’s continued use of noninferiority trials is leading to less effective drugs. When a drug is compared to another in a noninferiority trial, the intent is generally stated to be that the drugs are equally effective. But to deal with natural variation between trials, the noninferiority margin allows for the new drug to be potentially a little less effective than the old drug.
The problem is that the new drug will sometimes be significantly worse than a drug that is already on the market, especially as the percentage difference allowed by the FDA has gotten larger. This inferiority grows during subsequent comparisons of each new drug to a slightly older drug, with each new drug just a little worse than the drug it was compared to. This could result in new drugs on the market that are substantially inferior to old drugs. Even a noninferiority margin of 10% means that 1 in 10 patients may receive a new, less effective drug than the older therapy.[3]
Using less effective drugs can increase the risk of antimicrobial resistance, if newer, poorer performing drugs work by the same mechanisms as older drugs. Resistance to new drugs has been found soon after approval and use.[4] Thus new drugs may not actually help the problem of resistance and may make the situation worse if the new drugs are inferior to the older ones and there is no effective stewardship.
The guidance states that patients with serious disease and unmet need are willing to accept greater uncertainty or higher risk. As organizations that work with patients and consumers, we have found that is often not the case. It is our understanding that patients and consumers who are not faced with chronic fatal diseases have also expressed the need for FDA to focus more on safety.
The FDA recognizes the need to warn patients about the different standards for drugs approved through the limited pathway. For that reason, the guidance states that the labeling should include the words “Limited Population” adjacent to the drug’s name and include a statement about the indication for “limited and specific population of patients.” This is completely inadequate because it does not clearly describe the limited scientific evidence used to support approval.
Randomized and double-blinded superiority trials can be small and provide best available treatment by comparing the standard of care and placebo to the standard of care plus the new drug as an “add-on” treatment. This is common for cancer trials.[5] [6] FDA should adapt those strategies for antimicrobials rather than considering evidence from small trials of patients that are substantially different than the indication FDA approves.
Sincerely,
National Center for Health Research
Jacobs Institute of Women’s Health
MRSA Survivors Network
National Center for Health Research
National Physician Alliance
National Women’s Health Network
Our Bodies Ourselves
TMJ Association
USA Patient Network
Washington Advocates for Patient Safety
Woodymatters
- Outterson K, Powers JH, Seoane-Vazquez E, Rodriguez-Monguio R, Kesselheim AS. Approval and withdrawal of new antibiotics and other antiinfectives in the U.S., 1980-2009. J Law Med Ethics. 2013: 41:688-96. 10.1111/jlme.12079.
- Deak D, Outterson K, Powers JH, Kesselheim AS. Progress in the fight against multidrug-resistant bacteria? A review of US Food and Drug Administration-approved antibiotics, 2010-2015. Ann Intern Med 2016;165:363-72. 10.7326/M16-0291. 27239977
- Doshi P, Hur P, Jones M et al. Informed consent to study purpose in randomized clinical trials of antibiotics, 1991 through 2011. JAMA Intern Med 2017;10:1-8. 10.1001/jamainternmed.2017.3820
- Humphries RM, Yang S, Hemarajata P, et al. First report of ceftazidime-avibactam resistance in a kpc-3-expressing Klebsiella pneumoniae isolate. Antimicrob Agents Chemother 2015;59:6605-7. 10.1128/AAC.01165-15. 26195508
- Fojo T, Bates S. Mechanisms of resistance to PARP inhibitors—three and counting. Cancer Discov 2013;3:20-3. 10.1158/2159-8290.CD-12-0514. 23319766
- Gottesman MM. Mechanisms of cancer drug resistance. Annu Rev Med 2002;53:615-27. 10.1146/annurev.med.53.082901.103929. 11818492