In mid-July, the nonpartisan research arm of the federal government, the Government Accountability Office (GAO), released its long-awaited verdict on how the Food and Drug Administration (FDA) is handling compassionate use. As anyone who saw Dallas Buyers Club knows, compassionate use (also known as pre-approval access and expanded access) is when a drug company allows a patient who has no other treatment options to try a drug that is still in development and not FDA approved for use or sale. These patients are generally too sick to participate in a clinical trial—which has inclusion and exclusion criteria to try to keep the trial data from getting too “messy”—but if their physician, the drug company, the FDA, and their physician’s institution (in that order) all permit the effort, they can try these drugs in development through compassionate use.

On and off over the past decades, the FDA has come under fire for being too slow or too bureaucratic in handling compassionate use requests. One of the major recent critics is the libertarian Goldwater Institute, which published a policy report stating that: “For patients suffering from terminal illnesses, the federal Food and Drug Administration (FDA) is the arbiter of life and death. … The FDA, however, often stands between the patients and the treatments that may alleviate their symptoms or provide a cure. To access these treatments, patients must either go through a lengthy FDA exemption process or wait for the treatments to receive FDA approval, which can take a decade or more and cost hundreds of millions of dollars. Sadly, over half a million cancer patients and thousands of patients with other terminal illnesses die each year as the bureaucratic wheels at the FDA slowly turn.”

According to the FDA’s data, the statement made here about the FDA’s compassionate use program is not accurate. Yes, FDA approval may take a long time, but patients are able to access drugs (if the drug company permits) via compassionate use as soon as—or even before—the drug is first tried in humans. The FDA does a case-by-case review of all requests and approves more than 99 percent of them. For emergency requests, the review is completed in less than a day. For non-emergency requests, the review takes a median of four days. So while it is true that terminally ill patients may die from their disease before a cure is available, it is not true—as the Goldwater Institute claims—that the FDA is a gatekeeper standing between these patients and available cures.

All this has been said and debated back and forth so many times that some patients are confused. Is the FDA friend or foe? Is the Goldwater Institute, which has led the finger-pointing at the FDA, making valid criticisms or using untruths to gin up support for its anti-regulatory activities? Are “right-to-try” laws, such as the ones on the books in 37 states and just passed by the US Senate, that promise to expedite patient access to investigational medical products by cutting the FDA out of the picture a way forward or just a mirage?

What The GAO Report Tells Us About The FDA’s Speed

Accordingly, members of Congress asked the GAO to weigh in on whether the FDA was a help or a hindrance with regard to compassionate use. Using data from the FDA and other sources, the GAO determined that “From fiscal year 2012 through 2015, FDA reviewed nearly 5,800 expanded access requests and allowed nearly all of them to proceed. Specifically, of the 5,753 expanded access requests the agency reviewed, it allowed 5,697 (99 percent) to proceed.” As for the few requests the FDA did not allow, “the most common reasons were incomplete applications, unsafe dosing, the requested drug’s demonstrated lack of efficacy for its intended use, the availability of adequate alternative therapies and inadequate information provided in the application on which to base a decision” (GAO report, p. 22).

All of these make sense as grounds for denial, and they also help us understand what individual physicians could face if we cut the FDA out of the picture, as right-to-try laws permit. Individual physicians deal with significant burdens including 15-minutes-per-patient appointments and an ever-growing medical literature. They rarely know everything about drugs in the pipeline, even just those relevant to their particular specialty. Thus, every so often we see a request to the FDA in which the physician proposes using an unsafe dose of the investigational drug, or to try an investigational drug that has already been shown not to work on this condition, or to use an investigational drug for a disease that already has an approved, tested treatment. These are all important things to screen for before giving a patient an investigational drug, and the FDA does this screen rapidly, for the benefit of the patients who will be getting the drug.

The GAO reported that there had been some complaints that the “process for physicians trying to request expanded access to drugs for single patients was complex and cumbersome” (GAO report, p. 25). To this the GAO responded that the “FDA has undertaken efforts to improve the expanded access program. These efforts include publishing a simplified application form, finalizing existing guidance and making changes to its website for single-patient expanded access IND requests, and working with the Reagan-Udall Foundation—a non-profit established by Congress to assist FDA—to develop a website to help patients and physicians navigate the expanded access process” (GAO report, p. 25). The GAO found that “most stakeholders, including physicians, spoke positively about the new application” (GAO report, p. 26). However, the GAO investigators found that “even with these changes, a few of the stakeholders that we spoke with said the revised website and forms may still be hard to understand, particularly for physicians and patients who have no prior experience with expanded access requests” (GAO report, p. 26).

Educating The Health Care Workforce

Nobody finds a process easy the first time they do it, whether that is registering for a course, getting a marriage license, or navigating a new grocery store. Exacerbating the fact that physicians and patients who seek compassionate use may not have done it before and thus need guidance is the fact that compassionate use is not something taught to health care workers—even though they are who a patient would expect to assist him or her with the process. Drug development—and, in fact, clinical research itself—is not something every physician or health care worker knows about: The physicians and health care workers have to seek out opportunities to learn about them. And with regard to compassionate use, the most common way for health care workers to learn the process is by doing it.

The GAO found that compassionate use requests were disproportionately distributed across various clinical specialties, with the most occurring in the areas of anti-infectives, antivirals, hematology, and oncology (GAO report, p. 18). For this reason, I propose that physicians in training, at least in these areas, should be taught about compassionate use: What it is, why it exists, and how to lodge a request. Established physicians can get this information via continuing education courses, and expert groups within these areas, such as the American Society of Clinical Oncology, should make user-friendly guidance available to their members. This doesn’t mean education should be restricted to health care workers: Patient advocacy organizations would be important resources for patients seeking guidance. Indeed, Kids V. Cancer offers a navigator service that helps physicians and patients understand how to file a request for compassionate use access to an investigational drug. But such recommendations for health care workers and patient advocacy organizations are outside of the purview of the GAO report, which focused on the FDA.

Where There Is Room For Greater Clarity

The GAO found that there was one thing the FDA could do better. Namely, some of those interviewed by the GAO “raised concerns that FDA is not clear about how it uses expanded access adverse events data …. Some of the manufacturers noted that use of such data may influence FDA in making final approval decisions and that this possibility can contribute to a manufacturer deciding not to grant patients access to their drug through the expanded access program” (GAO report, p. 30). This has certainly been a concern of drug companies. If patients are too ill to be enrolled in a clinical trial, aren’t they more likely to have a problem while getting the investigational drug simply because they are so sick? And if a patient dies or has a serious complication while on the company’s investigational drug, how might anyone be able to figure out if the drug caused the problem? Will the FDA err on the side of caution and tell the drug company that its drug is unsafe? Will the agency demand additional clinical trials or animal research to make sure that the drug was not to blame for that patient’s experience? Will the already long and expensive process of developing a drug become longer and more expensive because the drug company allowed a sick patient to try its drug via compassionate use?

With all these questions looming, it is clear why a company may decide, as a business decision, that it cannot allow its investigational medical products to be used in compassionate use. Thus, it is important, as the GAO recommends, for the FDA to be crystal clear in conveying to drug manufacturers and other stakeholders what the ramifications of a bad outcome in a patient getting compassionate use access to an investigational drug would be.

Fortunately, the FDA is already doing so. The GAO report does not note that the FDA has been engaged in conducting a series of audits on these very questions and has been publishing its findings. Specifically, FDA investigators reported looking at data from January 2005 through December 2014. At issue was how often a bad outcome in a patient receiving a drug via compassionate use led to a “clinical hold,” in which the FDA paused further research on a product until vital safety questions were figured out. They found that 0.2 percent of clinical holds were “related to adverse events observed in patients receiving drug treatments under expanded access.” This low number ought to provide manufacturers reassurance that a patient’s bad experience while receiving a product via compassionate use is not a kiss of doom for the drug being developed.

The FDA researchers didn’t stop there. Next, they audited the agency’s regulatory decision making from 2010 to 2016, looking to see if adverse events in a compassionate use setting had led to the FDA either not approving a drug or changing an approved product’s label from what it would have been had the adverse event not happened. They reported, “none of the negative regulatory marketing decisions were based on the adverse experiences reported under expanded access. The vast majority of deaths and serious adverse events that occurred under expanded access were not interpreted by FDA to be due to the investigational drug and did not affect product labeling.”

Contrary to widespread industry concerns that serious adverse events in a compassionate use program will lead to problems with the FDA, analysis of years of data does not show this to be the case. This finding needs to be spread far and wide, and the FDA should use all its capabilities to do so. But, in my opinion, what we are left with is a situation where education is key: education of all stakeholders about how to navigate the process of compassionate use and education of the pharmaceutical industry that providing access to investigational drugs is not likely to cause companies serious problems down the road. Education is where we need to concentrate our efforts—not deregulation, not speeding up a system that is already quite quick, and not legislating new laws that don’t address the real issues. Research that dives into how to create and promote education about compassionate use will be a critical next step.