Expanded access programs are conducted as research protocols in clinical settings and must be approved by the FDA and by ethical review committees or institutional review boards (IRBs). An expanded access program site may be located in a local doctor's office or a large university HIV research clinic (which is where I work), yet each must adhere to the regulatory agency's requirements for record keeping and safety reporting. While small clinics and independent physician researchers may use a single national IRB for ethical oversight, universities typically host a local IRB that must approve all human research conducted at that institution.
Clinical research sites typically get little if any funding from drug companies to run expanded access programs for investigational HIV drugs, yet these projects probably consume more site staff time than most conventional research studies. A few busy research sites, particularly those associated with large universities, have recently refused to open expanded access programs due to the uncompensated burdens of administration. If this trend continues, then in the future there will be fewer opportunities for patients to obtain access to desperately needed new treatment options.
The drug companies usually want to rush the expanded access protocol through the IRB submission process. This is admirable, since it means the drug will get dispensed sooner to people who need it, but it usually results in sloppier work being done at all levels, with multiple protocol revisions being necessary. Often the protocol and/or template consent form will need to be re-written one or more times because the protocol wasn't thought through carefully before being released to the sites, or because late breaking data needs to be incorporated, or because errors were made that weren't caught in the proofreading process. Each rewrite results in the sites having to make protocol revision submissions to the IRB. Revision submissions can be almost as time consuming and paperwork-generating as initial submissions.
When rushed and/or overworked, contract research organizations (CROs) -- third party clinical trial specialists often hired by drug companies to actually manage expanded access programs -- are more likely to misplace key documents such as signature pages, original financial disclosure forms, etc. This means duplicated effort for the sites if they have to print new forms, obtain new signatures, and send the signed forms to the sponsor again. Depending on the form, this may also result in a new IRB submission as well.
My experience at the local level is that the more I am rushed, the more likely I am to make a typo on a consent revision or some other form, or to miss one sheet out of the mountains of papers the investigator is required to sign. In the world of regulatory compliance, you can't just put a line through a typo and write the correction above it with your initials and date. You have to generate a document highlighting the correction, and generate a clean document incorporating the correction, and then get this copied and approved by the sponsor, CRO, and IRB.
The grants and contracts specialist at the research site has to do the same work for an expanded access study as for any other study. The host institution will take the same "overhead" fees from the company's payment. IRB processing and review fees are the same. Nurses, doctors and data managers have to do the same preparation in terms of becoming familiar with the protocol, the study forms, and the data recording and reporting processes, and must set up files and notebooks just like they do for any other study. All this takes time, which costs money.
Any time a patient with limited treatment options is enrolled in a study with a new agent, a resistance test should be done to make sure the patient's virus is sensitive to the drugs in rest of the ART regimen. If the only active drug is the investigational agent, then there is a risk of developing resistance since the patient will be on "virtual monotherapy." For a salvage study being used for drug development, sponsors usually pay for this resistance testing, and occasionally help pay for the optimized background regimen. For expanded access studies however, the resistance test must be paid for by the patient. If the patient has inadequate insurance coverage for this, the site usually ends up absorbing the cost. If the patient is unable to pay for the drugs in the new optimized background regimen, then site staff often ends up spending additional time to help the person enroll in compassionate use or the state's AIDS Drug Assistance Program (ADAP).
Another major reason for the increased costs to research sites offering expanded access programs is the health status of the patients often enrolling in these studies. In a typical Phase II or III drug development study, there are tight inclusion and exclusion criteria designed to ensure the integrity of the data and protect patient safety by filtering out patients with very advanced disease. This also means that the likelihood of a study patient experiencing an adverse event is fairly low, and usually not expected.
This is not the case for expanded access studies. The patients who enroll in these studies are often very sick, to the point that the new drug is their last (but very slim) hope. The inclusion/exclusion criteria for expanded access studies are minimal. Patients with high viral loads, low to non-existent CD4 counts, and pre-existing conditions are allowed to enroll. Unfortunately, these patients are easily prone to severe illness. Furthermore, a sick patient will come to the clinic more often than a healthier patient, and will require more time from site personnel at every level.
Once a patient becomes a study patient, every adverse event, every illness, and every lab abnormality, even if it is obviously related to a pre-existing condition and not related to the study drug, must be reported to the sponsor and to the IRB. Moreover, the event usually must be reported several times, because there are usually follow up reports generated. It may seem irrelevant to a study drug that a patient with a pre-existing end-stage brain infection has a seizure after enrolling in an expanded access study. However, because there is no way to completely rule out that the drug did not exacerbate the condition to some degree, the event must be reported. Every follow up visit, every new lab report, and every recurrence of the event must be reported again, until the time when and if the condition is resolved.
In addition, each one of these adverse event reports then becomes a safety letter that is sent to every other site using the drug in research. These safety reports must be filed in each site's regulatory binder and reported to their IRB, and then documentation of IRB receipt must be filed and sent to the sponsor. This can result in multiple shelves of binders full of paperwork. Every on-study serious adverse event, from every clinic in the world doing research on the drug, must be filed at every site and reported to the IRB at every site that is using the drug in any research project. For example, during the tenofovir (Viread) expanded access study in 2001, I accumulated 10 full binders of safety report filings by the time the study closed. For a few months, I was processing up to 10 or more tenofovir safety reports a day, spending half my time on it. Most of these were obviously related to late-stage HIV complications, but since an association to the drug could not be ruled out, all of the events had to be reported. For every report and follow-up report on an adverse event related to pre-existing C. difficile or seizure disorder, there might be another report that could lead to a real association of the drug to a condition like renal disease or bone density problems.
We want to do everything we can to get potential life saving drugs to patients, but clinics also have to pay the doctors who diagnose and evaluate the events and the study staff who have to process the paperwork. In addition, the lab work for safety labs, if the patient's insurance doesn't cover it, often must come out of the site's budget. Expanded access studies often consider safety labs to be part of "standard of care." They expect reporting of all labs, including routine viral load, CD4 and CBC counts, but seldom pay for the cost of these labs.
The bottom line is until the FDA approves a drug, it can only be accessed through a research study. Even expanded access studies are by definition research studies, and can be audited by the FDA just like any other study. Therefore all the same work must be done to stay in compliance with Good Clinical Practice, multiplied by the fact that these studies, by nature of enrolling very sick people, can eat up a major percentage of the staff's work hours. Since the expanded access studies are often conducted at the same time as the FDA submission process, the drug companies often generate even more regulatory paperwork than normal.
Companies should be willing to pay the sites for the work they do on expanded access studies. The sad fact is that some research sites are now declining to participate in these programs because they are unwilling to take on the extra burden. It's not that the sites don't want to help people obtain early access to promising new HIV drugs; it's that they can't afford to pay the people who have to perform the significant amount of work involved.
Harry C. S. Wingfield, MFA, is a site regulatory specialist at a major U.S. clinical research institution.