Stanislaw Burzynski publishes 42.5% of one clinical trial

Damn it.

After hearing the horrible news that Laura Hymas had died on the same day that I found out that the pro-quackery “health freedom” Alliance for Natural Health (ANH) was supporting a federal “right to try” bill, I thought it would just be a one-off post about Burzynski, his allies, and the human toll exacted by his cancer quackery. No such luck. The reason is that yesterday morning I awoke to e-mails and Tweets from multiple colleagues and bloggers making me aware of an article published in a journal that (once again) I had never heard of, Child’s Nervous System. The article is by—surprise! surprise!—Stanislaw Burzynski and the usual suspects at the Burzynski Clinic and Burzynski Research Institute and entitled The response and survival of children with recurrent diffuse intrinsic pontine glioma based on phase II study of antineoplastons A10 and AS2-1 in patients with brainstem glioma.

My first reaction was: Hmmmph. It looks as though Burzynski finally found a journal editor naive or ignorant enough to publish one of his clinical trials. Then I looked at the abstract and realized that I had jumped to a conclusion. Burzynski hasn’t published one of his phase II trials at all. He’s published only less than half of a phase II trial, as you will soon see. In fact, he’s published exactly 42.5% of one of his phase II clinical trials.

Looking over the methods, I noted that this is trial BT-11. That means that it’s one of the 72 clinical trials submitted by Burzynski in the mid-1990s in order to take advantage of FDA willingness to consider approving his clinical trials. This was a willingness forged in the furnace of Rep. Joe Barton’s constant flogging and investigating of the FDA and his frequent dragging of then FDA director David Kessler in front of his committee for some tender loving rhetoric about his “persecution” of Burzynski, interspersed with crying families of cancer patients who were afraid that the FDA might actually manage to convict Burzynski. Again, remember, these families honestly believed that Burzynski was responsible for their loved ones still being alive and that Burzynski was the only doctor who could save them from death due to their various advanced tumors of the brain and other organs. It’s not the patients that I castigate, rather, it’s Burzynski’s cynical use of these patients as “human shieldsagainst the FDA. In any case, let me just refresh your memory about these 72 clinical trials, quoting the immortal words of Burzynski’s lawyer, Richard Jaffe:

CAN-1 [a “wastebasket” clinical trial by Burzynski approved by the FDA and earlier described by Jaffe as an “artifice, a vehicle we and the FDA created to legally give the patients Burzynski’s treatment”] allowed Burzynski to treat all his existing patients. That solved the patients’ problems, but not the clinic’s. A cancer clinic cannot survive on existing patients. It needs a constant flow of new patients. So in addition to getting the CAN-1 trial approved, we had to make sure Burzynski could treat new patients. Mindful that he would likely only get one chance to get them approved, Burzynski personally put together seventy-two protocols to treat every type of cancer the clinic had treated and everything Burzynski wanted to treat in the future…Miracle of miracles, all of Burzynski’s patients were now on FDA-approved clinical trials, and he would be able to treat almost any patient he would want to treat!

Now, more than 15 years later, we finally see one of these 72 trials published—sort of. There is far less to Burzynski’s clinical trial than meets the eye. After reading the study, my reaction to it was: Huh? That’s all you could come up with in over 15 years? Really?

I’ll take care of the formalities first. From the methods:

The study was designed as a single-arm, two-stage, phase II trial of ANP as monotherapy. The study was listed by the National Cancer Institute (NCT00003459). It was supervised by an independent Institutional Review Board (BRI-IRB, BCBT- 11). The study was performed

This “independent” IRB is, of course, the same IRB that is chaired by an old crony of Burzynski’s from Baylor who also happens to be the chair of the board of the Burzynski Research Institute. It’s also the same IRB that was recently roundly condemned in an FDA warning letter for abusing the single patient IND process, failing to provide adequate informed consent, and failing to protect patients interests and make sure that the risks of BRI clinical trials are minimized.

Of course, the line in the paper that leapt off the page at me was this one:

The protocol permitted admission of patients with brainstem glioma for whom standard curative treatments are not available. A total of 40 patients were enrolled, but the scope of this paper has been limited to the 17 RPDIPG patients.

So why isn’t Burzynski reporting on the other 23 patients? This is what I mean when I say that Burzynski is reporting less than half of a phase II trial: 17 patients out of 40, which equals 42.5% of the patients on that trial. RDIPG, by the way, stands for “recurrent pediatric diffuse intrinsic pontine glioma.” I also note that the inclusion criteria for the original BT-11 trial include “histologically confirmed (except if medically contraindicated) brain stem glioma that is unlikely to respond to existing therapy and for which no curative therapy exists.” So why limit it to just 17 patients? It’s not explained in the manuscript other than the sentence above. I’m also puzzled by the choice of dosage, in which the infusion rates of antineoplaston therapy were scaled according to the patient’s age, rather than the more standard method of calculating overall daily dose based on either body weight or body surface area and then simply administering that dose. It’s not necessarily wrong, but it’s nonstandard.

It’s also important to note that this is not a randomized trial. There is no control group. It’s more like a combined phase I/II trial than a strictly phase II trial given its lack of randomization and control group plus its dose escalation design. Again, this is not wrong; it’s how many early phase II trials are done. It’s just that, even if the trial were promising, it would not in any way constitute definitive evidence that antineoplastons are efficacious against brain tumors. In any case, I think I’ll just cut to the chase here and show you the results. Here’s Figure 4, a Kaplan-Meier survival curve showing overall survival (OS) and progression-free survival (PFS).

Bzfig4

Not particularly impressive, is it? It sure doesn’t look like any sort of miracle cure. However, it should be noted that DIPG is a horrible tumor, and the median survival tends to be on the order of several months, with survival beyond two years rare—rare, but, as I note, not “never,” as Burzynski claims. Moreover, as dismal as this survival curve is, in fairness, one has to note that it’s better than typically reported for DIPG. (Yes, DIPG is just that bad.)

Still, there’s some rather deceptive reporting going on there. No, it’s not outright lies. However, it’s reporting that fails to put results in context. For example, Burzynski reports a complete response in 6% of patients. That would be one patient out of seventeen. That one patient, it would appear, is a patient with a grade I pilocytic astrocytoma. This is a tumor that is not considered to be malignant. No, seriously. It’s generally not considered a malignant tumor:

Juvenile pilocytic astrocytoma (JPA) is a rare childhood brain tumor. In most cases, the tumor is a benign, slow growing tumor that usually does not spread to surrounding brain tissue. Symptoms of a JPA will vary depending upon the size and location of the tumor. Most symptoms result from increased pressure on the brain and include headaches, nausea, vomiting, balance problems and vision abnormalities.

So, in reality, excluding this one patient Burzynski’s long term survival result, even among these seventeen patients, is, as one would expect for DIPG, zero. Not surprisingly, that same patient is the one who had what was described as a “complete response” (CR). Excluding that patient, as would be appropriate, makes Burzynski’s CR rate zero as well. In fact, there were two cases of pilocytic astrocytoma included with this group, and the other patient with this tumor type was reported as a partial response (more the 50% tumor shrinkage). So, even in this apparently cherry-picked group of patients, Burzynski’s results are not what they seem and not as impressive as he presents them. That’s leaving out the toxicity:

Safety assessments were analyzed based upon the total number of enrolled patients (n=17). Grade 4 toxicities including hypokalemia and fatigue occurred in 6 % and hypernatremia in 18% of the group. Grade 3 fatigue and urinary incontinence occurred in 6 % and somnolence in 12 %. In patient 2 (Table 2), grade 4 hypernatremia was detected when he was on mechanical ventilation in the intensive care unit. This patient received terminal care and was disconnected from the ventilator upon request of the members of his family. His attending physician determined brain tumor as the cause of death.

Remember the definition of grade 4 toxicity. According to the Common Terminology Criteria for Adverse Events (CTCAE), it’s toxicity that is potentially life-threatening, and grade 3 is severe, as in requiring treatment or even hospitalization. I’m also not convinced about this hypernatremia in that it’s listed as grade 4, but then Burzynski dismisses the patient’s death as being due to the brain tumor. Yet, clearly even from this description it seems as though the development of grade 4 hypernatremia is what prompted the family to decide to stop treating him and switch into the palliative care mode, disconnecting him from the ventilator. In any case, if you add up the toxicities, grade 3 or 4 toxicities occurred in 42% of patients. Contrary to how Burzynski’s treatment is represented, both by him and his supporters, this is hardly “nontoxic” therapy.

Finally, leave it to Burzynski’s supporters to sabotage him in their enthusiasm to “prove” what a genius he is. The ANP Coalitionremember these guys?—published a blog post the other day announcing with pride the publication of 17/40 (or 42.5%) of Burzynski’s clinical trial. In it, they exulted:

The results of this study depict a high level of efficacy along with a more than acceptable tolerance of ANP.

One of the article’s sterling successes is Jessica Ressel. You can meet Jessica under our website’s “Meet the Miracles” button.

Who remembers Jessica Ressel? I do. In fact, she played a prominent role in the very first movie made by everyone’s favorite Leni Riefenstahl to Dr. Burzynski’s—oh, never mind. I’m referring, of course, to Eric Merola’s first Burzynski “epic” that portrayed Burzynski as a misunderstood genius and “brave maverick doctor.” In fact, her diagnosis of pilocytic astrocytoma is very consistent with what her tumor looked like on imaging. Indeed, even two and a half years ago I thought that it looked like a small, odd tumor. Indeed, Burzynski’s description of the tumor in his paper does not jibe with what I saw when I watched Merola’s first movie about him. There, the tumor was small. In the paper, the the tumor was described as “tumor involving 72 % of pons at the baseline.” Go back and look at the images I took from the film and showed in my original post on Merola’s first movie. Does the lesion look as though it takes up 72% of any structure? Not to me, it doesn’t. It’s easily covered by a yellow “radiologist’s mark.” I will admit that I’m not a radiologist, in particular a neuroradiologist. Even so, the lesion shown on the scan that Eric Merola included in his medical records for his movie is small. Even a “dumb surgeon” such as myself can recognize that, and the largest measurement ever recorded for the tumor was 1.7 cm, with many times the measurement being recorded as well under 1 cm. This same “dumb surgeon” also recognized that there was something very strange about the tumor size measurements recorded during Ressel’s treatment. Again, go back and read my review if you want the details. (I love being able to say that, given that I wrote it two and a half years ago.) Suffice to say, Ressel’s case is not convincing evidence that antineoplastons are effective against DIPG.

The bottom line is that this “study” is nothing more than a typical ploy by Burzynski and his allies to convince the rubes that, alone of all the scientists studying such seemingly untreatable tumors, Burzynski can offer a chance of curing them. The problem, of course, is that there’s no evidence that he can. He’s only presenting 43% of his patients in this study. My guess is that the rest of the patients did poorly, and if that’s the case, then his real survival rates are probably less than 50% of what he is reporting because he’s excluded them from the denominator. That would bring his results right into line with everyone else’s—or even make them worse than average.

Then there are the scans shown in Figures 2 and 3. Look at them. They show different sections of the brain. (Use the eyes to compare.) Given that, they don’t actually demonstrate whether or not there is a response. There are a number of other oddities, too, such as how long it took to accrue so few patients, why the study is only now being published seven or eight years after the most recent data reported, and the like. Come to think of it, this isn’t even really 42.5% of a clinical trial, and even this undoubtedly carefully cherry picked segment of one of Burzynski’s 72 phase II trials is completely underwhelming.

I’ll conclude by pointing out that, upon seeing this study, I wrote an e-mail to Concezio Di Rocco, the editor of Child’s Nervous System, expressing many of these concerns in a less “insolent” fashion. Here was his response:

Thank you for your letter. I have already asked some eminent neuroncologists to write a comment in the direction you suggest.

Bob Blaskiewicz also wrote an e-mail to the editor, and this was the response he got:

I have invited eminent neurooncologists to write a comment to be added to the articles. Unfortunately the reviewers of the journal failed to provide their promised evaluation so that after several months there were “no negative comments” to justify rejection. Now I look for receiving comments from Canada and UK to be published in the journal. Should you be willing to contribute you are welcome.

I’m not entirely sure how to interpret this, but this seems to be saying that Burzynski’s article was not peer-reviewed. Basically, it appears that Burzynski submitted his article, and the editor, not getting any comments back from his peer reviewers, just decided to publish it. This suggests that the policy of his journal is that the default action for a manuscript is to publish unless there are negative comments. I’ve never heard of a journal with such a policy before, which makes me wonder if I’m missing or misinterpreting something here. Either way, I’m not surprised. The only way Burzynski can publish his tripe is if he slips his study into a journal with—shall we say?—less than rigorous editorial standards.

Same as it ever was, which is why my final reaction remains: Hmmmph. It looks as though Burzynski finally found a journal editor naive or ignorant enough to publish one of his crappy clinical trials.