Characterizing expanded access and compassionate use programs for experimental drugs

Objective We sought to determine the characteristics of “expanded access” and “compassionate use” programs registered in ClinicalTrials.gov and to determine the percentage of drugs provided through these programs that ultimately received FDA marketing approval. Results We identified 398 expanded access and compassionate use programs (hereafter referred to as expanded access programs) registered on ClinicalTrials.gov. Industry funded 61% (n = 241) of programs individually or collaboratively, while NIH and the US Federal Government rarely funded programs (3% [n = 11] and 2% [n = 6], respectively). Most programs provided access to drugs (71% [n = 282]), 11% to biologics (n = 43), and 10% to medical devices (n = 40). These programs covered 460 unique conditions, the most common being HIV (n = 26), leukemia (22), and multiple myeloma (n = 14). Only 2% of programs reported results in ClinicalTrials.gov. Most programs (82%) were open to enrolling adults and seniors (n = 326). These programs provided access to 210 unique experimental drugs, of which 76% have received FDA approval.


Introduction
Proposed federal Right to Try legislation, advocated by the White House, [1] allows physicians to prescribe experimental therapies unapproved by the U.S. Food and Drug Administration (FDA) to terminally ill patients. Proponents say these laws, which have already passed in 36 states, are vital to providing potentially lifesaving therapies to terminally ill patients who do not qualify for clinical trials. Critics contend patients can already access experimental drugs through the FDA's Expanded Access Program, which grants 99% of requests but depends on consent by medical product manufactures [2].
Expanded access programs, sometimes also called "compassionate use" programs, provide patients with serious or immediately life threatening diseases or conditions access to investigational products outside of clinical trials. Under the current Federal Food, Drug, and Cosmetic Act (FD&C Act), patients seeking expanded access generally need their physician to determine they have no alternative available therapy and no access to a clinical trial for their disease or condition. Additionally, the patient's treating physician must agree to apply for expanded access on their behalf and the medical product company must agree to provide the investigational drug [3]. Currently, a company can decline a request or it can approve the request, but is not obligated to provide the drug for free [4].
Much of the scholarly literature on expanded access programs focuses on the role of the FDA [5,6] or other regulatory bodies [7] in providing patients access to experimental drugs, or the ethics [8][9][10] and legal aspects of these programs [11,12]. A few papers also explore the role of social media and patient advocacy in fostering pre-approval access [13,14]. Very few papers provide empirical data or data-driven characterizations of expanded access programs.
The main, perhaps only, empirical paper on expanded access programs was published by the FDA. It details how many expanded access requests it received over the last ten years and their rate and handling of approvals (99% of requests were approved) [15]. In general, the literature provides little to no insight on the characteristics of these programs or whether the provided experimental drugs ever receive FDA approval.
Given the contentious debate surrounding the Right to Try laws [16] and lack of empirical information about expanded access programs in the literature, we sought to determine the characteristics of "expanded access" and "compassionate use" programs registered in ClinicalTrials.gov, the trial registry maintained by the U.S. National Institutes of Health (NIH). We also sought to determine the percentage of drugs provided through these programs that ultimately received FDA marketing approval.

Methods
We searched ClinicalTrials.gov to identify all expanded access and compassionate use programs registered by July 1, 2016. After identifying all relevant trials in the ClinicalTrials.gov database using the search terms "compassionate use" and "expanded access" we reviewed trials manually for relevance and removed duplicate trials (trials listed multiple times under the same NCT number). ClinicalTrials.gov is a publicly available registry and database maintained by the National Library of Medicine (NLM) at the National Institutes of Health (NIH), created in 1997 as a results of the Food and Drug Administration Modernization Act of 1997 (FDAMA).
For each identified program, we abstracted the program's registration date, funding source, product type (drug, biologic or device), condition treated, patient age, number patients enrolled, whether the program was for a single or multiple patients, and reporting of trial results.
For drugs only, we determined FDA approval rates by cross-referencing the drug provided in the expanded access program with Drugs@FDA. Drugs@FDA is a public database maintained by the US FDA, listing most drug products approved by the FDA since 1939. We used descriptive statistics, such as means, proportions, and medians, and all analyses were performed using Microsoft Excel v2013 (Redmond, Washington).

Results
We found 527 trials using our search criteria on Clini-calTrials.gov. After removing 54 duplicate trials and 75 irrelevant trials, 398 programs met our final inclusion criteria as compassionate use (CU) or expanded access (EA) programs (hereafter referred to as expanded access programs). Of these 398 programs, 306 were labeled as expanded access only programs, 49 were compassionate use only, and 43 programs used both terms. The earliest recorded program started in June of 1989. It was for "Compassionate use of tetrabenazine in the treatment of abnormal movements, " sponsored by Baylor College of Medicine.
Industry funded 61% (n = 241) of the 398 programs individually or collaboratively. In contrast, the NIH and the US Federal Government rarely funded programs (3% [n = 11] and 2% [n = 6], respectively) ( Table 1). 36% of programs are classified as funded by an institution "Other" than the Industry, NIH, or US Federal Government. The "Other" category generally refers to university or academic sponsors.
The FDA subsequently granted marketing approval for drugs provided in 68% (n = 192) of expanded access programs. These programs provided access to 210 unique experimental drugs, of which 76% (n = 160 of 210) have received FDA approval. We did not analyze FDA approval rates for the biologics or devices.

Discussion
Nearly 400 expanded access and compassionate use programs were registered on ClinicalTrials.gov as of July 2016. Most programs were sponsored by medical product manufacturers. This suggests that even without federal Right to Try legislation, the pharmaceutical industry is establishing programs to make experimental therapies available to terminally ill patients.
Most (76%) provided drugs in expanded access programs eventually received FDA approval. Thus, provided drugs in registered expanded access programs are, more times than not, eventually deemed safe by the FDA. Notwithstanding, the fact that nearly 25% of expanded access drugs have yet to receive FDA approval, shows that we cannot entirely eliminate safety and efficacy questions in expanded access and compassionate use programs.
It is reasonable to allow the FDA to retain its oversight and approval role for these programs, in order to help mitigate safety risks for patients-especially since it approves 99% of expanded access and compassionate use requests. Currently, patients wishing to gain access to experimental drugs through compassionate use and expanded access programs, must obtain FDA approval. Proposed Right to Try Legislation generally removes this step of needing FDA oversight and approval.
Legislative efforts should also aim to expand patient access to clinical trials, which in some cases could alleviate the need for expanded access and compassionate use programs. Currently, clinical trial access is often limited by several factors, including a patient's location, age, and health status [17]. Most clinical trials enroll younger, healthier, and whiter patients than the typical patient population [18][19][20]. The 21st Century Cures Act makes some progress in expanding access to clinical trials.
Expanded access programs raise broader ethical and regulatory questions, including whether (and how much) product manufactures should re-direct investigational products and resources from formal clinical trials to patients requesting expanded access and how to finance these programs. In addition, experts are using [21] or advocating the use [22] of expanded access programs as "real world evidence" for drug safety or efficacy. While the reliability of using these data for this purpose is debatable, it's worth noting that, thus far, we found only 2% of programs reported results in ClinicalTrials.gov.

Limitations
Important information about expanded access and compassionate use programs was frequently missing from ClinicalTrials.gov, limiting precise insight into exactly which patients and how many are accessing experimental agents. This limitation should decrease over time. The FDA Amendments Act (FDAAA) final rule, released in September 2016, newly requires expanded access programs be registered on ClinicalTrials.gov and be kept up-to-date. Thus, ClinicalTrials.gov should improve in robustness over time for understanding U.S. expanded access programs.