Introduction

Since the 1970s, the U.S. Food and Drug Administration (FDA) has allowed patients with serious diseases who have exhausted other options to access unapproved investigational drugs outside of the clinical trial setting [1]. This process has historically been facilitated by the FDA’s Expanded Access Program, which was first formalized in the late 1980s. Expanded access (EA) can be used for single patients (also known as “compassionate use”), intermediate size populations, or treatment group populations all of which require FDA and ethics review by an institutional review board (IRB). Single patient compassionate use requests can be made for emergency or non-emergency situations and response times average less than 1 day for emergency requests [1]. More than 99% of all requests are approved, and the FDA has taken recent steps to reduce the application to two pages and to provide assistance for the process [1–3].

Despite these efforts by the FDA, 41 states have passed new “Right to Try” (RTT) laws since 2014 [4]. This right to try movement has been led by the Goldwater Institute, a libertarian public policy organization that has advocated for the adoption of “right to try” laws that would reduce FDA powers and oversight from the process of approving non-trial use of experimental medication [5]. RTT laws however do not mandate companies to provide experimental medication to patients and thus fall short of their name and intent. Despite this issue, RTT laws are favored by the public and members of congress on both sides of the aisle. In May 2018, congress passed the federal RTT Act, which created another federal pathway for terminally ill patients to access treatments that had passed Phase I clinical testing without requiring FDA and IRB oversight.

The growing right to try movement, and the passage of the federal RTT law, have brought new scrutiny to the complex ethics of providing experimental medications outside of clinical trials [6]. Access is typically justified ethically by compassion for individual patients with terminal illnesses and respect for their autonomy [7, 8]. However, these goods must always be weighed against patient safety, the long term value of regulatory structures, and the societal benefit of clinical trials to public health [9–11]. The difficulty of balancing safety and public health on the one hand with timely and equitable access on the other [12, 13] makes clear the need to develop robust guidance for patients and clinicians navigating this landscape [14, 15].

The COVID-19 pandemic has made the tension between the desire for rapid access and the need for strong empirical evidence of safety and efficacy through clinical trials more apparent and urgent [16–18]. Questions have been raised about the complex ethics of providing drugs with unknown safety and efficacy data during a pandemic [19, 20], the appropriateness of different regulatory mechanisms in overseeing unapproved treatments for the public [16, 21, 22], and the professional obligations of physicians in utilizing preapproval mechanisms [23].

The 2018 RTT law’s removal of required IRB and FDA oversight places more responsibility on requesting physicians, but little empirical work has explored how physicians perceive their role and responsibilities in non-trial access to investigational drugs [1, 24–26]. Oncology and hematology physicians are among the most frequent prescribers of compassionate use drugs [27, 28]. Understanding how oncologists with compassionate use experience interpret the approval process and navigate its challenges may clarify the practice implications of new policy changes and lend insights to the kinds of guidance clinicians need for existing pathways to be beneficial to their patients. Given that oncologists, even those with experience providing off-trial medication, can conflate definitions, policies and procedures between compassionate use, RTT and off-label [26], capturing their understanding of and experiences with different preapproval pathways is important if we are to design education to inform and aid physicians with varying patient requests for experimental medications. In this study, we interviewed oncologists from an academic medical center with three major geographic sites across the U.S. who had compassionate use experience to understand their experiences, knowledge, and opinions of the approval processes and pathways.

Methods

We conducted semi-structured interviews to capture the views and experiences of academic oncologists at a major cancer center with access to federal RTT and EA programs, the methods of which have been previously reported [26]. This study was designated minimal risk and a waiver of signed consent was approved by the Mayo Clinic Institutional Review Board #19–005556. Prior to obtaining informed consent, all subjects were provided information regarding the study goals, risks and benefits, procedures used including the interview process, recording and transcription, methods to protect privacy including deidentification, deletion of recordings and maintaining confidentiality in reporting data, and procedures for withdrawal. All subjects provided verbal consent to participate which was documented by the interviewer.

Results

Twenty-one medical oncologists from Mayo Clinic in Minnesota, Arizona or Florida participated in our study, representing several oncology sub-specialties. Participants had practiced medicine for 17 years on average (7 to 34 years range) at the time of interviews, and most oncologists had research and administrative duties in addition to their clinical responsibilities. We report five major themes that emerged from our analysis examining oncologists’ attitudes towards EA and reflection on physician obligations to provide off-trial access (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Summary of themes.

https://doi.org/10.1371/journal.pone.0261478.t001

Experiences and considerations for navigating preapproval access to investigational drugs

All oncologists we interviewed had experience navigating compassionate use pathways. Most participants had pursued the pathway for single patient use and some also had experiences enrolling patients in intermediate size EA programs. While several physicians had considered RTT requests, none had requested access to investigational drugs from pharmaceutical companies through this regulatory pathway (see Smith et al. [26] for more details on oncologists’ experiences with RTT).

Nearly all of our participants reported they had raised the possibility of pursuing non-trial access for patients that they thought would benefit. Many oncologists also reported that in one third to one half of cases, the patient was the one who had asked about pursuing an investigational drug. Some participants reported rare cases in which their patients had requested a particular, non-trial pathway such as compassionate use or RTT. In most cases, however, their patients who requested experimental therapies did not specify whether they wanted to enroll in a trial or use non-trial pathways to gain access to those drugs.

Outcomes of the process were mixed for our participants. Even though all of our participants had navigated the process, some had not administered the drugs because their requests were denied by companies or the IRB, their patients died before they could get access to the drug, the drug was approved before their request was completed, or their patient did not yet need the drug they had requested. Among those who had administered the drug, there were a handful of participants who saw dramatic responses from patients. The majority, however, saw moderate, little, or no benefit for their patients. Regardless of the outcomes, most physicians were willing to pursue EA again, although that willingness was often contingent on the presence of administrative support.

Ease with experience: Mixed reviews on the ease of accessing experimental drugs through EA.

Our participants gave mixed reports on the ease of accessing investigational drugs falling in two main groups. One group reported that non-trial access was difficult to obtain due to the complexity of the process and unwillingness of sponsors to make the drug available. The other group reported that it was not difficult for them to gain access and that companies were usually receptive. Some who had success in the past acknowledged that the process is much easier once a physician has been through it, noting they could imagine it being much more difficult now than it was in their particular experience.

It’s not easy at all in my opinion for a number of reasons. One, [companies] actually have to have a willingness and an access program for you to be able to get access to it. Two, you’ve got to make sure that it’s the right thing for your patient. (Participant 15)

I haven’t noticed any significant barriers because I’ve tried only three or four times with two companies and both of them, like I told you, the people on the other side have been really responsive. They just want to make sure we have a good rationale and [that]. . .the patient fits most of the clinical trial-ish criteria that they have set up before. Other than that, I haven’t found much difficulty. (Participant 16)

Regardless of oncologists’ ease with past experiences, there was widespread agreement that the likelihood of gaining access to investigational agents was significantly increased by requesting a drug with enough data to suggest a high likelihood of benefit and applying the drug appropriately to a patient’s specific tumor. Participants also reported that the ease of access to drugs is ultimately determined by the receptiveness and policies of each drug company, and that larger companies were typically easier to work with. Participants also mentioned that other variables, such as insurance coverage and travel to medical centers, may also affect a patient’s ability to use a drug even if it has been made available by the manufacturer.

Oncologists named multiple factors that had played a role in their previous requests (Table 2). First and foremost, participants specified that for a request to be reasonable it must have a strong scientific rationale—typically a genetic mutation or other biomarker offering plausibility of benefit. They also required that the known risks of the drug not outweigh the benefits, that the patient is healthy enough to tolerate possible side effects, and that there are not any other options available to the patient. Participants reported they would be unwilling to consider non-trial access to investigational drugs if there was insufficient safety data and low likelihood of benefit, or if the patient had a poor functional status and was simply taking a “shot in the dark.”

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Oncologists’ reasons for considering non-trial preapproval access to investigational drugs.

https://doi.org/10.1371/journal.pone.0261478.t002

Discussion

The passage of the federal RTT and the recent COVID-19 pandemic have reignited questions about the ethics of using medications without known safety and efficacy, but little is known about physicians’ experiences with and attitudes toward existing pathways that could help guide the development of future policy [25, 26, 32]. The diversity of views and experiences in discussing and requesting experimental drugs through EA in our study provide additional clinical perspectives on patient access to unapproved investigational drugs outside of clinical trials.

While institutional support was available for oncologists in our study to access experimental drugs via EA, several oncologists reported challenges in navigating the multi-stakeholder and multi-administrative processes similar to the experiences of physicians captured in other studies [1, 32]. The biggest barriers our participants identified were the navigation of complex websites and forms, and the receptiveness of drug companies to requests for EA. Institutions can do their part to help streamline administrative processes by providing a centralized process and identifiable experts to help physicians navigate FDA, IRB, and potentially other institutional processes. Additionally, a database like the one used in this study to recruit oncologists could improve access by identifying different drugs, patient conditions, and the indications for use among EA requests to minimize duplicative efforts. Despite administrative hurdles, oncologists in our study were willing to use EA again. Similar attitudes were found in an FDA commissioned study where 94% of 139 physicians reported a willingness to recommend EA programs to a colleague despite difficulties they encountered [1]. This is further echoed by the finding that our cohort of oncologists reported that accessing investigational products was part of their professional obligation to patient care but remained cognizant of the reasons why other clinicians might choose not to offer EA to patients, especially physicians at institutions that do not provide the support they perceived to be necessary to navigate EA pathways.

Oncologists in our study were concerned with many factors beyond the minimum regulatory requirements of EA programs. They required that requests for investigational drugs through EA have a clear scientific rationale, a strong ratio of risks to benefits, and a solid safety profile. They also expected patients to have good functional status and high motivation to pursue EA. This rationale meets current regulatory standards that require a patient have a qualifying condition and not be eligible for clinical trials, but also goes beyond them. Together with our participants’ prioritization of functional status over the terminal nature of their patients’ conditions, this rationale provides important nuance around the factors physicians may use to qualify patients to receive unapproved drugs and suggests substantial differences between regulatory and clinical standards for making unapproved therapies available to patients. Raus has analyzed common justifications for compassionate use programs, arguing that most are grounded in justice, autonomy, or beneficence [33]. While law and policy tend to prioritize self-determination and fairness, oncologists in our study appear to more heavily favor beneficence, ensuring that the decision to pursue these pathways is the option they believe to be best for their patient. Further research into how oncologists and other specialists weigh the factors for offering EA to patients would help in understanding and improving access to experimental drugs for patients outside of clinical trials.

The inclusion of patient motivation as part of the rationale oncologists used to offer EA is interesting given their discussion of the low potential of these drugs for benefit and the importance of managing patient expectations. Many scholars differ in their views as to whether physicians should offer experimental options in addition to all reasonable standard of care for every patient. Recognizing the threat of promoting unrealistic optimism of drugs with no potential benefit, Bunnik and Aarts have argued that patient reported factors including affordability, health literacy, or motivation should not guide a physician’s decision when considering whether to discuss experimental interventions [34]. They separately describe regulatory and ideological shifts toward investigational drugs becoming an expectation in clinical settings, raising concerns about false hope, side effects, and funding challenges [35]. Our study shows evidence of high hopes among patients and the powerful role that patient motivation can play, even in contexts where physicians do not believe a particular drug to be helpful for a patient. While affirming the importance of managing expectations, our findings suggest that patient motivations play a complicated role in justifying compassionate use requests, and further research should explore possible differences between patients’ and physicians’ mindsets in considering compassionate use and clinical trial medications while also continuing to engage with broad ethical analysis.

In our study, oncologists reported that patients’ expectations ranged from realistic outlooks to high hopes about experimental interventions, while oncologists themselves considered experimental drugs obtained through EA as mostly palliative. Patients self-reports have also been shown to vary in terms of their motivation to undertake experimental treatment and their expectations of benefit [34]. The extent of a realistic expectation of benefit from experimental products is likely to be based on the patient’s severity of prognosis, despair, frame, risk behavior, and their social interactions with families and caregivers. Unrealistic expectations may add significant burdens on the physicians correcting these beliefs, and this calls into question whether the requirements of informed consent can be upheld. It remains unclear whether physicians ought to intervene, and if so to what extent, especially when considering denying experimental interventions to out-of-option patients who may have high hopes or are unrealistically optimistic [36]. Bunnik et al. call for more explicit investigational drug guidelines, especially in systems that may be shifting toward expanded access as a default, in order to proactively engage with these thorny ethical questions [35].

Conversations in cancer care are inherently difficult given the psychosocial distress on patients and the difficulty of having end-of-life conversations for physicians [37–41]. Our data suggest that oncologists’ views toward patient motivations and expectations of benefit from experimental oncology drugs may differ from what other studies have shown examining the attitudes of patients. While 54% of U.S. adults favor greater access to experimental treatments prior to trials that demonstrate safety and efficacy [42], only 3–5% of patients are enrolled in cancer clinical trials [43], with significantly fewer patients receiving experimental interventions through EA. The source of these disparate numbers suggests that patients may desire access to experimental drugs but perhaps outside of a clinical trial. While therapeutic misconception (the concept that patients may conflate the goals of a clinical trial with the goals of care) exists among oncology patients, adequate informed consent processes shows that patients become less interested in accessing experimental medication if it means they will be randomized and could be placed in a placebo or even a standard of care arm. These numbers also suggest a potentially large disconnect between what patients theoretically may want versus the realities of experimental medicine where toxicities are overwhelming and the idea of taking novel medicine that could help is no longer theoretical.

Our study has some limitations. As a qualitative interview study, our results are not intended to be generalizable. This study was performed at a single institution with robust support for clinical trials, administrative assistance for completing EA requests, and extensive physician experience with EA programs. Our results are further limited as the Mayo Clinic Cancer Center may not necessarily reflect other oncology settings. Finally, our findings are only reflective of physicians’ views and only include interpretations of patients’ perspectives. Further research is needed with patients themselves to share their own interpretations of similar encounters.

Conclusion

Our study highlights that oncologists at a single medical center with three major locations across the U.S., all of whom have at least some experience with EA and most with research experience, are supportive of EA programs and are willing to support patients who seek experimental treatments outside of clinical trials. Despite our conservative selection of participants with EA experience, several oncologists still experienced hardships and some were unclear about terminology and how best to navigate the multi-stakeholder processes. This suggests that oncologists considering non-trial experimental treatments for patients could benefit from training about preapproval access and need institutional infrastructure to help navigate the application processes. The results from this study also demonstrate that oncologists’ rationales for considering experimental drugs for their patients differ significantly from regulatory standards.

Tools designed to enhance shared decision-making and facilitate conversations about both on- and off-trial medications with patients are needed, and may bridge the gap among discordant views between patient and oncologist [44]. Both physicians and patients are confused with terminology and processes surrounding non-trial preapproval pathways suggesting a need for physician education [1, 26, 34]. While internet databases, social media support groups, and patient advocacy organizations have been highlighted as potential avenues of information about accessing experimental drugs for patients [45], there remains little information among these sources about accessing non-trial experimental treatments compared to information about clinical trials [46]. Tools to foster dialogue between patient and oncologist and support systems for oncologists to navigate non-trial EA are needed so that patients can receive the best care at a crucial juncture late in their illness.

Supporting information

Acknowledgments

We would like to thank the oncologists who took the time to participate in our study.