Beware the Cancer Quack – updated

Before getting to the meat of this post, I want to kick things off with some eminently sensible pictorial advice

Believe it or not, this particular poster, by Max Plattner, dates to the period from 1936 to 1938 and yet, as the events of this week have shown, it remains as relevant today as it was in the year if was first published; a year which also saw Jesse Owens win four gold medals at the Berlin Olympics, much to the chagrin of a certain Mr A. Hitler. 1936 also saw the the publication of the very first Billboard magazine charts, from which i discovered, somewhat curiously, that I have copies of three of five recordings* that achieved the highest chart positions during that year.

*For the the record, the three that I have are Billie Holliday’s recording of Gershwin’s ‘Summertime’ and Robert Johnson’s ‘Cross Road Blues’ and ‘Sweet Home Chicago’, lest anyone look up the year and assume that I’m a fan of Bing Crosby and Fred Astaire, the year’s other chart-toppers.

Returning to the 21st century, the story that serves as a reminder of the value of the advice given in that poster is, of course, that of Billie Bainbridge, a four year old girl with a rare and inoperable form of brain cancer.

Actually, if truth be told, the story I’m most interest in isn’t, strictly speaking, about young Billie. It’s actually about the American clinic, the Burzynnski Clinic, that Billie’s family hopes she will be able to attend, if they can manage to raise $200,000 to cover the costs. And its also about the manner in which a PR flack who claims to represent that clinic reacted when a blogger started raising some very pertinent questions about the clinic operating practices and the effectiveness of the treatments it offers.

I really don’t to revisit ground that already been well enough covered elsewhere over the last few days, so if you’re coming to this for the first then you need to start with the following posts by Andy Lewis of The Quackometer:

The False Hope of the Burzynski Clinic

The Burzynski Clinic Threatens My Family

I’d also suggest that you pick up Dorothy Bishop’s article, ‘The weird world of US ethics regulation‘, David Colquhoun’s commentary, which covers some of the ‘science’ behind the activities of this clinic and shows it to be, quite literally, taking the piss, Cancer Research UK’s commentary. ‘Hope or False Hope?‘ and this post, by Josephine James, which includes a fairly comprehensive collection of links to other articles covering this same story.

Last, and by no means least, Zeno’s been looking at some of the business/financial aspects of this story in a post which also includes the following table of the Burzynski clinic’s claimed response rates for common cancers:

Now, I’m no oncologist but I’ve read enough alt-med research papers over the years to know what scientifically meaningless data looks like, and Burzynski’s objective response rate data looks pretty meaningless to me.

For starters, Burzynski does a pretty lousy job of identifying exactly what it he claims to be treating, for example, lung cancer comes in five histological types; Non-small-cell carcinoma (NSCLC), small-cell carcinoma, carcinoid, sarcoma and unspecified and each of these histological types may have their own sub-types – NSCLC sub-types include squamous cell carcinoma, adenocarcinoma, bronchioalveola carcinoma, carcinoid and other. Lung cancers are highly heterogeneous malignancies in which is its not unusual for tumors to consist of more than one subtype and so, if you look at some of the credible published research in this area what you will invariably find is that researchers go to quite some considerable length to spell out just exactly what kind of tumors they’ve been working on as not all types/subtypes respond equally well to particular treatments.

Cancer is a very complicated disease, or rather category of diseases, and its therefore necessary for researchers to be very specific as to the type, and sub-type. of cancer they been working on when presenting their findings, if their results to have any real scientific value.

So, already we’re off to a rather mediocre start in terms of the quality of the evidence on offer and that was enough to prompt me to do a bit more background reading on the subject of cancer research standards which, perhaps, unsurprisingly, threw up another anomaly that merits further investigation.

In clinical trials of non-surgical cancer treatments, how do we assess whether or not the treatment is having any actual effect, given that, in many, if not most cancers, its relatively unusual for a treatment to be so successful that it entirely eradicates the tumors?

The answer is that we look for evidence of shrinkage in the size of the tumor(s) and in order to do that we need some sort of objective standard against which measure this shrinkage, if it occurs, and decide whether its significant enough to provide evidence of a definite response to treatment. That, in very simple terms, what the objective response rate is used for – it measures the percentage of patients who exhibited a clinically significant response to the treatment based on standard assessment criteria, and for cancers thist standard is determined by a set of published rules called RECIST (Research Evaluation Criteria in Solid Tumours), which was initially published in 2000 (version 1.0) and then revised in 2008 (version 1.1). If you’re at all interested, then a full copy of the guidelines can be accessed here, and for our purposes you should take a good close look at section 4.3, ‘Response Criteria’, which starts on page 5 of the pdf, or page 232 if you’re going from the page numbers on the actual pages.

Burzynski’s data table includes the following list of definitions in relation to his table of figures:

OR: Objective Response, includes CR, PR, MR, & IM.

CR: Complete Response. Complete disappearance of all signs of cancer in response to treatment of 4 weeks or longer.

PR: Partial Response. More than 50% decrease in the size of the tumors (the sum of cross-sectional area of the tumors), in response to treatment of 4 weeks or longer.

MR: Mixed Response. Significant decrease (more that 25%) in the size of tumors wifi simultaneous increase in size of some of the other tumors.

IM: Improvement. Decrease in size of the tumors, not confirmed yet by the second follow-up radiological measurement.

SD: Stable Disease. Hb decrease or increase in the size of the tumors, but no progression, in response to treatment of 12 weeks or longer.

PD: Progressive Disease. More then 50% increase in size of the tumors (the sum of cross-sectional area of the tumors), in response to treatment of 4 weeks or longer.

EP: Evaluable Patients. Patients who remained on treatment long enough to enable an objective evaluation of the response.

So, Burzynski’s claimed ‘objective response’ includes any/all patients who had either a complete, partial or mixed response to treatment or who showed an improvement, even if this had not been verified by a second, follow-up, radiological measurement*

*Ideally, measurements should be taken via a CT scan although measurements can also be taken from X-ray photographs for some lung cancers, provided certain conditions are met, i.e. the X-ray photograph must provide an unobsrtructed view of the tumor.

If, however, you consult the RECIST guidance you’ll find no reference whatsoever to any kind of ‘mixed response’ category and, of course, researchers shouldn’t be including results in their data before they’ve been properly verified, so there’s no ‘improvement’ category either.

What you will come across, if you go beyond the guidelines and look at a few journal papers for recent studies conducted using the RECIST standards, are references to a ‘minor response’ category, as here:

PATIENTS AND METHODS: Per protocol, the first three disease assessments were done at 2, 4, and 6 months. For the purpose of the analysis (landmark method), disease response was subclassified in six categories: partial response (PR; > 30% size reduction)*, minor response (MR; 10% to 30% reduction), no change (NC) as either NC- (0% to 10% reduction) or NC+ (0% to 20% size increase), progressive disease (PD; > 20% increase/new lesions), and subjective PD (clinical progression).

*I should point out, before anyone become confused by the numbers here that under the RECIST standard, the size of a tumour is measured in terms of the sum of its diameters in two planes of measurement and not by its cross-sectional area, as used by Burzinski, and the two measures amount to near enough the same thing as makes no difference for our purposes, or those of cancer researchers.

Now all this raises a couple of rather important questions about Burzynki’s response rates.

One is that his data appears to be off protocol in so far as his ‘mixed response’ category is not the same as the ‘minor response’ category that is fairly commonly used by other researchers, even if its not part of the formal RECIST standard, as the latter make no reference whatsoever to any tumours showing an increase in size at the same time as other showing a measurable decrease, albeit one too small to categorised as a partial response. It also, noticeably, includes results that haven’t, at the time of publication, been properly verified, which is also a bit of a no-no.

Burzynski also published his ‘objective response rate’ without any further qualification – there is no data given to show what proportion of the patients who did exhibit a response of some kind fall into each of his four categories, and from the point of view of cancer patients, that’s pretty important information because the figure that they’re naturally going to be most interested in is the ‘complete response’ category as that shows the number of patients for whom the treatment was a complete success.

So, we have no way of knowing exactly how much of Burzynski’s claimed response rate is based on results that have yet to be properly verified at the time of publication and, equally, no way of knowing how much of this same rate is accounted for by his off protocol ‘mixed response’ category, the clinical value of which is, to day the least, distinctly dubious as, ultimately, a treatment which shrinks some tumours, but not others, is only ever going to be of limited value to patients unless it can be used as part of a combination therapy with other treatments that successfully target the tumours that Burzynski’s treatments fail to reach.

To me that all looks just a bit dodgy, especially as Burzynski appears to be touting his wholly unproven antineoplastion therapy as something of a one-size-fits-all miracle cure.

Then, as I was researching this post, the plot thickened even further as I happened across on of Burzynski’s less than happy former customers, Wayne Merritt, who was diagnosed with stage IV pancreatic/liver cancer in 2009. Of everything on the Merritt’s site, one claim, in particular, caught my eye:

Along with the long list of other meds that were supposed to work in conjunction with each other, the Burzynski Clinic gave my husband standard chemotherapy medications. We were never told that two of the medications were conventional chemo medications. We discovered from our local pharmacy that one medication the Burzynski Clinic had charged us over $2300.00 for could have been purchased from the pharmacy for around $170.00.

Leaving aside the alleged 1250% markup, what the Merritt’s are alleging – and this is only an allegation at this point – is that Burzynski has been mixing standard chemotherapy meds in with his antineoplastion therapy without telling his patients.

If this is true then it is an extremely serious breach of medical ethics and it raises further questions about Burzynski’s claimed response rates as, without proper controls, there would be absolutely no way of establishing the extent to which any of his claims may be attributable to to the covert adulteration of his treatments with conventional chemotherapy medications, particularly when you consider that a ‘mixed response’ is pretty much what you’d expect to see in patients who received partial chemotherapy.

Update.

In putting this post together, last night, I was aware of the fact that Burzynski is currently facing legal action by the Texas Medical Board which, if successful, could see his license to practice medicine revoked – the hearing appears to be due to take place next April. What wasn’t clear at the time was exactly why the TMB are now moving to take action against Burzynski but, although details are still rather vague, information posted to sites operated by his supports appear to indicate that the the main charges against him relate to claims that he’s been using off label conventional medications in his ‘trials’.

Again, this information comes with a big ‘if true’ caveat, but if this is indeed true then it would appear to lend some credence to Wayne Merritt’s claim that Burzynski had included conventional chemo meds in the treatment that he was prescribed.

Footnote.

Incidentally, Wayne Merritt is still alive and seemingly doing pretty well even if the very obvious lessons of hsi encounter with the Burzynski clinic haven’t struck home. The ‘what are we doing now’ link on his site leads to a webpage which stands as veritable cornucopia of cancer woo but which whooly omits an mention of the single most salient detail in his story:

Monday 12 September 2011

I have great news! We saw the oncologyst today to get the results of the CT Scan Wayne had last week. The results are as follows: There has been no change. No growth. No new areas of suspect. Total inactivity! What does that mean? Well,… it’s kind of hard to say but we are accepting the fact and believing that the tumors have died! The doctor said she and the doctor who diagnosed would discuss doing a second biopsy to deturmine if there are any live cells within the tumor. If that happenes, and the results are that the tumors are truely dead, then we hope to be able to discuss ceasing the chemotheropy treatments!

So, that’s a win for medical science then?

Sadly, no – not according to the Merritt’s, as their journal entry continues…

We’ve prayed and believed so long and hard for a miracle…. believing that it would come the way our human minds picture it…… instantly. However God has choosen to give us this miracle in bits and pieces. Bit by bit, a little each day. He did not take these tumors away all at once,… but left them there as evidence that they were in fact there,… life threatening,…. and now it appears….. STOPPED IN THEIR TRACKS BY THE POWER OF PRAYER AND FAITH!!!

God has taken us on this journey for a reason. And while we still don’t know where or how the journey will end…… we will declare till the end that God is a miracle working God!

Yes, witness the miraculous power of the cognitive bias and its unmatched ability to blind people to the obvious.