Wednesday, November 30, 2011

Paying to Volunteer

This past weekend, while many were getting together with family and friends, sharing in feasts, watching TV and spending too much money on holiday shopping, a story erupted among skeptic circles. Like many such matters, it began with the actions of an apologist for what is, at best, questionable science and at worst rank quackery. An individual, Marc Stephens, put himself forth as representing one Dr. Stanislaw Burzynski. Mr. Stephens, you see, took it upon himself to defend the reputation of the good doctor, going so far as to rattle his saber and issue legal threats to Andy Lewis (aka Le Canard Noir) of Quackometer and teenage skeptic Rhys Morgan. They were to take down their less than complimentary writings about Dr. Burzynski or face a libel suit. We've seen this kind of thing before.

As most of you know, the routine generally goes something like this: person Q comes up with some rather extraordinary claim; a critical thinker closely examines the claims and offers some legitimate criticisms; Q, rather than addressing the criticisms, threatens to sue the critical thinker for libel; then, since most people don't want to deal with the headaches and expense of defending such a case, they'll probably back down. Basically, that's what happened here, except that instead of tucking their tails and slinking into quiet obscurity, Andy and Rhys politely requested that Mr. Stephens explain which specific statements he felt were libelous and the reasons why. Instead of taking the polite, ethical and standard approach of pointing out the particular offensive passages, Stephens continued to try to bully them into silence. This is not what one does if one wishes to avoid making a fool of oneself. His actions only served to fuel ridicule for him and the Burzynski Clinic he purported to represent in the matter. I'm not going to go into all of that any further. If interested, Liz Ditz has, as usual, a plethora of links to other blogs covering the story, as do Josephine Jones and Anarchic Teapot. For a discussion on the science of Burzynski's claims and a review of a infomercial documentary film about Burzynski, head over to Respectful Insolence for Orac's usual close scrutiny.

Instead, I want to talk about a question that arose regarding Burzynski's research. Namely, that he charges lots and lots of money to participate in his clinical trials. An individual seeking help from him will quickly find themselves racking up a bill in the tens or hundreds of thousands of dollars. Often, these will be people who have run out of options: they have been diagnosed with inoperable cancer or may have pursued conventional treatments to their limits. They are desperate and may be quite willing to spend the money required to grab at any hope they can. And I understand that. I don't blame them. If I were in a similar situation, I'm not certain what I would do, but that need for something, for some hope, would definitely be very, very strong.

But isn't that illegal?

How can he charge people who are volunteering to participate in a clinical trial? Isn't there some law against that?

Technically, it isn't necessarily against the law. There is a regulation (21 CFR 312.8) that allows researchers to charge study participants for the study drug in certain situations. For example, if I wanted to study some new use for Zantac and needed to buy the drug from the manufacturer, I could pass that cost along to the study participants. In the case of unapproved drugs, like Burzynski's "antineoplastons", then there are certain conditions which must be met.

First, the sponsor must:
(i) Provide evidence that the drug has a potential clinical benefit that, if demonstrated in the clinical investigations, would provide a significant advantage over available products in the diagnosis, treatment, mitigation, or prevention of a disease or condition;
Burzynski would need to show, through in vitro or animal studies, for example, or early phase I trials, that his antineoplastons have might be more effective or have fewer side effects (or both) compared to any alternative treatments that subjects would normally be able to receive. He would need to compare his drug to the "gold standard" used for a specific disease (e.g., glioblastoma).

But that's not enough to be allowed to charge for his drug. The sponsor must also:
(ii) Demonstrate that the data to be obtained from the clinical trial would be essential to establishing that the drug is effective or safe for the purpose of obtaining initial approval of a drug, or would support a significant change in the labeling of an approved drug (e.g., new indication, inclusion of comparative safety information);
The antineoplastons developed by Burzynski would need to be backed by some convincing explanation that the results of the clinical trial would support approval by the FDA. Assuming that point one, above, is met, this shouldn't be difficult to accomplish. In fact, the stronger the initial evidence suggesting a significant improvement over approved products, the easier this argument would be to make.

Assuming that the sponsor has some evidence suggesting their product will be better and that the clinical trial would lead to approval, there is still a third condition that must be met. They still must:
(iii) Demonstrate that the clinical trial could not be conducted without charging because the cost of the drug is extraordinary to the sponsor. The cost may be extraordinary due to manufacturing complexity, scarcity of a natural resource, the large quantity of drug needed (e.g., due to the size or duration of the trial), or some combination of these or other extraordinary circumstances (e.g., resources available to a sponsor).
Usually, sponsors bear the costs of conducting a study. They seek grants from nonprofit organizations, government agencies like the National Institutes of Health or even for-profit companies. In order to charge for the drug, the sponsor would have to face some seriously large expenses. Obtaining the drug would have to be so expensive that the normal avenues to fund a study would be insufficient to cover costs. In Burzynski's case, the antineoplastons, now synthetically developed, were originally obtained from urine. Not exactly a scarce resource, but the isolation of the amino acids might be a difficult and costly procedure. Then again, from what has been reported so far, subjects get a huge amount, pharmocologically speaking, of the study drug, up to around 7g (yes, grams) per kilogram.

So let us assume that Burzynski has the evidence (i), would be able to obtain data from the clinical trial leading to FDA approval (ii) and, after all other funding sources had been exhausted, would be unable to conduct the trial without charging for the drug (iii). What could he actually charge for?

I'll quote this section in its entirety (underlined emphasis added):
(d)Costs recoverable when charging for an investigational drug. (1) A sponsor may recover only the direct costs of making its investigational drug available.

(i) Direct costs are costs incurred by a sponsor that can be specifically and exclusively attributed to providing the drug for the investigational use for which FDA has authorized cost recovery. Direct costs include costs per unit to manufacture the drug (e.g., raw materials, labor, and nonreusable supplies and equipment used to manufacture the quantity of drug needed for the use for which charging is authorized) or costs to acquire the drug from another manufacturing source, and direct costs to ship and handle (e.g., store) the drug.

(ii) Indirect costs include costs incurred primarily to produce the drug for commercial sale (e.g., costs for facilities and equipment used to manufacture the supply of investigational drug, but that are primarily intended to produce large quantities of drug for eventual commercial sale) and research and development, administrative, labor, or other costs that would be incurred even if the clinical trial or treatment use for which charging is authorized did not occur.
So only the cost to make the drug or to acquire the drug from another manufacturer can be charged to subjects. Research and development costs, administrative costs of running the study, wages or salaries for study staff are all off limits; the subject cannot be charged for those. Likewise, any procedures which are not part of normal care for the disease and are used only for the purposes of the study generally cannot be charged to subjects (i.e., biopsies to confirm cancer diagnosis would be billable, blood draws for study-related measurements would not). The sponsor then needs to provide documentation justifying the costs to be recovered by charging subjects.

Once all of these requirements are satisfied and submitted to FDA, the sponsor cannot charge subjects until they receive written authorization from the agency. If the FDA feels that charging subjects interferes with clinical trial enrollment or smacks of commercial sale of the drug, they can revoke authorization.

Not knowing all of the details, I can't really offer a strong opinion on whether or not Burzynski is meeting these criteria. It seems, based on some of the things I've read, that he's charging for more than just the investigational drug and standard treatment-related costs. I'd also be very curious to see the approval from FDA allowing Burzynski to charge subjects, if such an approval actually even exists. I wouldn't be all that surprised if he hadn't actually received authorization or if he went a bit beyond the specific authorization he did receive. I'm fully willing, though, to be proven wrong.

It may be legal, but isn't it unethical?

I have to apologize. I've been a bit long-winded on the legalities. If you're still with me, here's the bit that people really want to know. Is the system broken if it allows subjects to be charged, when they are already taking on the risks of a drug which, by definition, has not yet been shown to be safe and effective for the condition under study? Isn't that wrong?

The case of Burzynski's trials, spanning several decades, gives us a good excuse to discuss this ethical question. The exorbitant prices he charges his subjects make it easy to cry foul. When we think about clinical trials and human subjects research, we think about how the whole point of the research is to figure out whether the drug is a) safe and b) actually does something that has a meaningful benefit for patients. Anyone volunteering for a drug study is necessarily taking a risk beyond what they would normally face. There may be as yet unknown side effects of the drug, some of which may be quite serious. The drug might be less effective than available alternatives (or even completely ineffective), leading to possible worsening of the patient's condition. When they are already facing such risks, it seems wrong and unfair to ask any more of them, like making them pay financially.

But, just like with the ethical dilemmas around paying study subjects, where payment may serve as an inducement to participate against their own best interests, charging subjects isn't necessarily unethical in and of itself. There may be situations where there really is justification to do so, where the study would not be able to be conducted without such measures. There may be a product which really could help future patients, not to mention the study subjects themselves. Although early studies may show some promise, we all know (or should know) that research dollars don't come easy. There are a lot of project out there, some worthy, others not quite so worthy, all vying for limited resources. Despite a researcher's best efforts and intentions, the cost of an investigational drug may simply be too great to move forward. In those situations, completely banning the practice of charging subjects for the study drug may delay development of newer, better drugs or even prevent their development altogether.

The ethical questions here would be a lot nicer if everything were black and white, but life is seldom so simple as that. As with everything, there is nuance; a multitude of factors rise up for consideration. There needs to be room to accommodate those subtleties. While no regulation is folly, an iron fisted prohibition is likewise foolhardy.

Getting back to Burzynski specifically, even if he were to justify to the FDA the need to charge subjects, he would still need IRB approval to do so. If the IRB thinks that billing subjects for study-related procedures or the drug itself is unethical, then they may require that subjects will not be billed. FDA authorization, however, would act as a strong influence to favorably sway the IRB. That said, any IRB worth its salt (about which there is some question regarding the Burzynski Research Institute's IRB) would definitely require that the consent forms for the trials clearly state which specific charges would be billed to the subject, including the amount of the charges, as well as a statement that insurance companies might not pay for those costs. If you are reading this and are considering enrolling in one of Burzynski's trials, make sure to very carefully read the consent forms and keep a copy for yourself. In fact, I would very much like to see a copy of the consent form from any past or current subjects (feel free to redact signatures). You can e-mail a copy to me by clicking this link.

The brouhaha around Burzynski and Stephens (why do I see an ogre in sweater and skirt with a club in one hand and a pom-pom in the other?) has brought the doctor quite a bit of attention and raised ethical questions to consider. It will be very interesting to watch how this all plays out. And if anyone happens to find a copy of the FDA authorizations (separate ones for each clinical trial, of which there are quite a few, it seems) for Burzynski to charge subjects, I'd consider it a kindness to get a glimpse.

Addendum: Almost forgot to mention something. After all the stuff I discussed above, I want to make one thing clear: charging subjects to participate in a study is not a common thing. Most of the time, subjects remain financially neutral or even get paid to participate in a study. If there are any charges, they are typically only in treatment studies and involve standard-of-care procedures, not the investigational aspects of the study.

Update: Karl Withakay had a very insightful comment over at NeuroLogica. If Burzynski is charging subjects to be in his study, then that means his studies are not randomized, controlled trials, which brings into question the validity of any results. You can't blind people to the treatment they're getting if they're paying for the study drug. If he is using randomization and controls, and thus charging all subjects for the investigational drug whether they're getting the study drug or not, then he's violating FDA regulations, employing illegal practices, since he cannot charge a subject for something they are not receiving.


  1. All the trials Dr B is running are phase 2 - these are not necessarily randomised.
    CancerHelp UK says;
    Not all treatments tested in a phase 1 trial make it to a phase 2 trial. Phase 2 is sometimes written as phase II. These trials may be for people who all have the same type of cancer, or who have several different types of cancer. Phase 2 trials aim to find out

    * If the new treatment works well enough to test in a larger phase 3 trial
    * Which types of cancer the treatment works for
    * More about side effects and how to manage them
    * More about the best dose to use

    Although these treatments have been tested in phase 1 trials, you may still have side effects that the doctors don't know about. Drugs can affect people in different ways.

    Phase 2 trials are often larger than phase 1. There may be up to 100 or so people taking part. Sometimes in a phase 2 trial, a new treatment is compared with another treatment already in use, or with a dummy drug (placebo). If the results of phase 2 trials show that a new treatment may be as good as existing treatment, or better, it then moves into phase 3.

    So you're looking for dosage, toxicity and any inklings of efficacy. The trials aren't actually designed to show true efficacy - you need phase 3 for that. After 2,000 patients treated in the past 10 years (according to the clinic website) you have to ask why he doesn't have enough data to show that he is ready for phase 3.

    There is one phase 3 trial meant to be starting this year. But the paucity of convincing data after so many years of trials is rather suspicious. If there was significant evidence of efficacy (assuming his methodology is capable of showing it, which is highly doubtful based on previous criticisms of his trials then he could licence it to a pharma company (or even set up his own) and make $$$$$$$$$.

    As he hasn't, one can only ask questions as to why. And at the prices that his clinic is charging to take part in the trials (although he doesn't charge for the actual drug, he charges for pretty much everything else around it), there is a fairly obvious answer.

    Incidentally, I wonder how many of his success stories are actually down to chemotherapy?

    1. Phase 2 trials are normally divided into 2a for proof of concept (does the drug work as predicted by non-clinical studies?), and 2b (what dose works best?). So they do test efficacy, but not necessarily in a population that reflects normal clinical practice. What baffles me is why the FDA has approved protocols that don't meet the design requirements of phase 2 - Burzynski's studies can't possibly test efficacy. See

  2. This also raises the interesting question of the size of Burzynski's trials, and the consequent total gross income therefrom.

    Clearly, a trial with one subject cannot be in any way randomised, or capable of any form of statistical analysis. So presumably any legitimate trial would have to assemble a cohort of appropriate subjects before commencing, rather than, as seems to be the case, recruiting as and when subjects appear and/or come up with the cash. And if there is a defined size for a particular trial, what happens to early recruits while waiting for the study group to come up to size, particularly if they have terminal conditions?

    Which raises the question of the income generated. With charges of up to £200,000 per subject, trials involving single figures of subject numbers may be raising 7-figure sums. Are there any actual figures for this?

  3. @Andwhynot - that's not really how clinical trials work, certainly based on what I know of cancer trials in the UK. Patients are recruited onto the trial as and when they turn up, treated according to the protocol and followed up. So early recruits will still be treated according to the protocol, exactly the same as later ones. Each person is a unique dataset.

    If a trial fails to recruit enough people over time, the funding body may close the trial early, and the researchers just have to make do with he data they have up until that point. The reasons trials fail to recruit can be many and varied - not enough suitable patients diagnosed with that type of cancer for example.

    Trials need to be big enough to be statistically powerful. For larger trials, quite a lot of time will go into calculating how many people are needed to get meaningful results for that particular disease.

    The main issuem, as I mention above, is that Burzynski claims to have treated 2,000 people over the past 10 years. Where the f*ck is this data? And is it enough to justify charging $$$$ for the current trial

  4. @Andwhynot

    @K77 is correct. You don't need to wait until all of the subjects you need are enrolled. Once the study is approved by an IRB, you can begin enrolling subjects. As K77 mentioned, one of the problems that may occur to stop a trial early are slower than expected enrollment. The protocol should have some plan to deal with poor enrollment, drop-outs and early terminations. In something like Burzynski's trials, I would expect something like an intent to treat (ITT) analysis, where all subjects enrolled are counted in the final analysis. But, from a couple of the earlier published studies I've seen, that doesn't seem to be the case, since early terminations were dropped from analysis completely. One wonders the reasons.

    Looking at, Burzynski has 61 studies registered, only one of which was completed. There are a couple terminations due to slow enrollment, some withdrawn before enrollment began, and a whole lot of studies that have unknown status (no updates in the last two years). That's kinda fishy as well.


You need a Google, LiveJournal, WordPress, TypePad, AIM or OpenID account to comment. If you need one of those, click on the appropriate link.