They don’t call it “cheat-lation” for nothing (updated)

Last year, a seeming victory for the protection of human subjects from being subjected to pseudoscience. It began when Kimball C. Atwood IV, MD; Elizabeth Woeckner, AB, MA; Robert S. Baratz, MD, DDS, PhD; and Wallace I. Sampson, MD published a lengthy criticism of the NIH Trial to Assess Chelation Therapy (TACT) in Medscape, pointing out that it was a boondoggle that was not only not based on sound science but was in fact risky to patients and riddled with conflicts of interest and administered by highly dubious practitioners. If you want to know just how bad the National Center for Complementary and Alternative Medicine (NCCAM) could go in terms of studying quackery and doing a very poor job of it, it’s really worth reading the entire article. Of course, I will concede that TACT wasn’t entirely NCCAM’s fault. Its scientific advisory board voted against funding it because, it concluded, there was an insufficient scientific basis. Unfortunately, quackery fan and antivaccine zealot Representative Dan Burton applied pressure, and suddenly NCCAM did have $30 million after all to fund this study. The article by Atwood et al is really, really long, but worth reading, as is, I would humbly posit, my discussion of the issue. As a result, in September the NIH decided to suspend TACT until the allegations in the article could be more carefully investigated.

In the meantime, woo-meisters everywhere mounted a counterattack against the Medscape article. Perhaps most hilarious of these was an article published in, so very appropriately, the Journal of American Physicians and Surgeons by Beth Clay entitled Study of Chelation Therapy Should Not Be Abandoned. Harriet Hall did a comprehensive takedown of the misinformation contained in this article; so I won’t go into detail. Suffice it to say that it was a perfect article to publish in JPANDS. Indeed, it went beyond the normally high level of crankery that JPANDS promotes. I can boil down Ms. Clay’s points very easily:

  1. I know chelation works because practitioners and patients tell me it does. Never mind all that nasty science that says it’s incredibly implausible and all those already existing clinical studies that show no effect greater than a placebo.
  2. Atwood et al are a bunch of poopyheads.
  3. The TACT investigators are a bunch of great guys, and never mind all those conflicts of interest and dubious activities.

Oh, and pay no mind to those deaths on study.

At the time TACT was shut down, given the revelations in the article by Atwood et al, I was quite certain that it would never be opened to accrual again. The whole study was rotten to the core. Not only was it based on a highly implausible (at best) physiological mechanism and flying in the face of more than enough evidence from clinical trials to conclude that chelation therapy for heart disease doesn’t work, but its informed consent left out important risks, and several investigators were of–shall we say?–dubious character. If there’s one thing that’s paramount in any human subjects research, it’s that full informed consent is an absolute bedrock principle that cannot be waived or watered down. The Helsinki Declaration demands it, as does the Common Rule, which governs human subjects research and the composition and activity of the Institutional Review Boards (IRBs) that approve and oversee all human subjects research in order to protect the subjects. I wondered at the time just how the hell any IRB could have approved this pseudoscientific atrocity of a study. It turns out that the various local IRBs that had to approve the TACT study at the various locations at which it was being done. Moreover, many of the locations were “alternative medicine” clinics that dispensed a veritable cornucopia of woo, along with having a financial interest in chelation therapy. If you want an idea of the quality of some of these sites, Dr. R. W. Donnell’s magical mystery tour of NCCAM chelation study sites (for example, Tequesta Family Practice) is worth a read, and he even points out that the blinding protocol was inadequate to the task of keeping the drugs unknown to the investigators, to boot.

I was wrong. TACT is back up and running. True, the NIH’s Office for Human Research Protections (OHRP) issued a preliminary determination letter, but somehow, despite finding some serious problems with TACT, all it did was to slap its investigators on the wrist and tell them to “fix the problems.” For example, although the NIH told them, in essence, to “fix the consent form,” it didn’t penalize them for leaving out critical information on the original form, nor did it consider the consent issue enough to scuttle the study, as it should have been.

The second most telling part of the letter is this passage:

(3) The complainant alleged that there was a failure to ensure that risks to subjects are minimized and that risks to subjects are reasonable in relation to anticipated benefits, if any, to subjects, and the importance of the knowledge that may reasonably be expected to result as required by HHS regulations at 45 CPR 46.l11(a)(1) and (2). In specific, the complainant alleged that:

(a) The basis in the protocol for the claim that chelation may be a reasonable treatment for coronary artery disease is that the supposition that removing toxic heavy metals from the body will treat coronary artery disease, the “heavy metals” hypothesis. Calcium disodium EDTA, the form of EDTA used for treatment of lead poisoning would be consistent with this hypothesis and less dangerous than the disodium EDTA used as the agent in the TACT study. The use of disodium EDTA in this trial is more consistent with the “decalcification hypothesis” which has been demonstrated to be invalid.

(b) Biochemical literature has demonstrated that the “heavy metals hypothesis” is implausible and demonstrates that the chelation mixture used in the TACT actually has pro-oxidant effects in vitro.

(c) The trial was begun in the absence of prior supporting laboratory, animal, or human phase 1 or 2 studies, contrary to the usual requirements for a phase 3 trial.

(d) Since the mid-1970’s court documents and newspapers have reported at least 30 deaths associated with intravenous disodium EDTA.

In regards to allegations (a) through (d), we acknowledge your response that the “purpose of a clinical trial is to integrate all of the evidence and make a finding based on a public health need which will lead to clinical recommendations and a putative change in clinical practice,” as stated in your response. We also acknowledge your statement that you adhered to all regulatory obligations requiring that “…all serious adverse events, including deaths, were reported to the medical monitor, to the DSMB, the FDA and the IRB panels.” The test article used in the TACT study is under an IND issued by the FDA, which may consider the kind of information stated by the complainant in support of this allegation in the determination whether to issue an IND. Given this, our office is deferring the allegations noted above to FDA for appropriate investigation and/or action.

Any single one of these should have been enough to shut the study down, particularly (d), given that no mention of the risk of death was apparently in the “informed” consent. (c) in particular is so outside the norm for a human clinical trial that I still shake my head that any trial without supporting laboratory, animal, or human phase I or II was ever approved by the NIH, much less a $30 million phase III trial. Whenever woo-meisters complain that there’s a double standard and that they don’t get a fair shake, I’ll rub their noses in this study, which was approved based on, in essence, no data. I can tell you with absolute certainty that if I were to propose a clinical trial with no cell culture or animal data to back it up, no granting agency would fund it, and no IRB would approve it. Yet the quackery that is chelation therapy for coronary artery disease was funded to the tune of $30 million with, in essence, no evidence to support its efficacy other than anecdotes. No, strike that. TACT was approved in the face of more than enough clinical trial evidence to show that chelation is no better than a placebo. So what does the NIH do? It punts to the FDA.

However, I’m with Kim Atwood. If you want to see the absolute ethical and moral bankruptcy in this letter, this next passage sums it up:

Several site co-investigators have been disciplined for substandard practices by state medical boards, several have been involved in insurance fraud, and at least three are convicted felons.

Wow. Those are some serious allegations. What did the NIH say in response? Take a guess:

We note that your investigation revealed that in fact several of the TACT study investigators have been accused of substandard practices by state medical boards, involved in insurance fraud, and at least three are convicted felons. While concerning, these things do not automatically preclude an investigator from participating in research and do not automatically indicate a failure of risks to subjects to be appropriately minimized. The details and circumstances surrounding these incidents must be considered by the IRB when ascertaining the “acceptability of proposed research in terms of institutional commitments and regulations, applicable law, and standards of professional conduct and practice” as required by HHS regulations at 45 CFR 46.107(b). Based on the information available to us, we determine that, while true, the alleged facts in themselves do not give rise to a violation of 45 CFR 46.111. Please note the related recommendation at (C), below.

Yes, it’s true that having felons as investigators probably doesn’t necessarily preclude a safe study, but it sure as hell would make me wonder about the dedication of these felons to high ethical standards. More importantly, the insurance fraud reveals a serious conflict of interest, because many of these “investigators” made a lot of money doing chelation therapy and thus had a major stake in not having TACT shut down.

So what did (C) say? Not much more:

HHS regulations at 45 CFR 46.107(b) state that “the IRB shall be able to ascertain the acceptability of proposed research in terms of institutional commitments and regulations, applicable law, and standards of professional conduct and practice.” We note that, as stated in (A)(3)(e), above, your investigations revealed multiple instances of substandard practices, insurance fraud, and felony activity on the part of investigators. We recommend that the IRBs that reviewed this research re-examine the processes for evaluating study investigators to determine they are obtaining sufficient site and investigator information that is adequate to comply with HHS regulations found at 45 CFR 46.107(b)

In other words, yes you guys had felons for investigators, substandard practices, and insurance fraud, but if you just have your IRBs review their processes for reviewing investigators, and you’ll be just fine. Never mind the pseudoscience. Never mind that you didn’t provide the IRBs approving the project with accurate scientific information and risk estimates. None of that matters. It can all be made right if you just play nice now.

If you want to know why NCCAM should be defunded, dismantled, and whatever bits of it that have any value distributed to other Institutes in the NIH, look no further than the offense to science and medical ethics that is TACT and the tepid reaction of the OHR{ to its endangerment of patients. In no other area of human subjects research other than CAM research would a study with so little basis in science and so much evidence against its hypothesis be approved, much less to the tune of $30 million. Too bad powerful supporters of quackery in the legislature (Dan Burton is especially guilty in this case) care far more about proving that their woo works, no matter what, than they care about science or, more importantly, protecting the human subjects who are recruited for CAM studies like TACT from undisclosed risks, harm, or even accurate information about the science (or, more commonly, the lack thereof) behind the treatments being tested.