Is it right for scientists to put the links between cholesterol reducing medication and cancer down to ‘chance’?

Back in July I wrote about a study (the so-called SEAS trial) in which two cholesterol agents (simvastatin and ezetimibe) taken in combination appeared to increase risk of cancer. The full results of this study were published this week in the New England Journal of Medicine [1] and can be read here.

Alongside this research, we have an analysis by some scientists from Oxford University who look at other data from two large trials (the so-called SHARP and IMPROVE-IT trials) which are ongoing, and were designed to assess the potential benefits of the simvastatin/ezetimibe combo [2]. You can read this here.

The analysis found:

Neither the SHARP or IMPROVE-IT trials showed a statistically significant association between the taking of simvastatin/ezetimibe and cancer incidence or mortality from cancer.

When all 3 trials were lumped in together (SHARP, IMPROVE-IT and SEAS), there was no increased incidence of cancer, but the risk of dying from cancer was increased by 45 per cent in those taking the medication (and this was statistically significant).

The feeling of the authors of this analysis is that the excess of cancer and cancer death thrown up by the data is just down to chance. A fluke. Part of their reasoning rests on the fact that when all 3 trials are taken together, there was no statistically significant increase in cancer incidence (just mortality). They also say there is no plausible mechanism for the link.

The fact that there is no plausible mechanism is a not a good defense of ezetimibe, I think. Just because can’t explain how something is happening does not mean it’s not happening. If scientists notice a new adverse effect from a drug, the challenge is then to explore it further in an attempt elucidate its mechanism (as well as its extent). I’m not sure it’s a good idea to use our ignorance as a reason to dismiss our findings. Also, it should be borne in mind that there is quite a lot of evidence linking low cholesterol levels with enhanced risk of cancer.

The results of the 3 trials are inconsistent. But it should be borne in mind that if one or both of these cholesterol-reducing drugs do cause cancer then the likelihood is that it’s going to take some time (several years, probably) for that effect to be seen. Notably, the SEAS trial lasted for 4 years, while SHARP and IMPROVE-IT trials lasted an average of 2.7 and one year respectively. These lengths of time are almost certainly not long enough for any real effect of cancer to be seen.

But I suppose what is noteworthy in all of this is just how readily it seems some scientists are prepared to abandon science when (apparently) it suits them. Results have come in which, when analysed scientifically and statistically, show a link between the taking of simvastatin and ezetimibe and cancer that are very unlikely to be due to chance. And then some scientists have turned round and claimed that the results are, actually, likely to be due to chance.

Not all doctors and scientists take this view, fortunately. An accompanying editorial [3] states: The fact that the combined data from all three trials showed an increase in cancer mortality with ezetimibe should not be assumed to be a chance finding until further data are in. That seems eminently sensible to me, as I hope it will to others who are resistant to the idea of dismissing scientific findings, however unpalatable.

References:

1. Rossebø AB, et al. Intensive Lipid Lowering with Simvastatin and Ezetimibe in Aortic Stenosis. NEJM Published on-line 2nd September 2008.

2. Peto R, et al. Analyses of cancer data from three ezetimibe trials. NEJM Published on-line 2nd September 2008.

3. Drazen JM, et al. Ezetimibe and cancer ” an uncertain association. NEJM Published on-line 2nd September 2008.

9 Responses to Is it right for scientists to put the links between cholesterol reducing medication and cancer down to ‘chance’?

  1. Cybertiger 3 September 2008 at 9:01 pm #

    John Briffa said,

    “Just because can’t explain how something is happening does not mean it’s not happening. If scientists notice a new adverse effect from a drug, the challenge is then to explore it further in an attempt elucidate its mechanism.”

    All this is spookily reminiscent of the MMR vaccine and the link with autism: the clever scientists ‘know of no plausible mechanism for the link’. And after the disgraceful treatment meted out to Andrew Wakefield and his colleagues by the utterly appalling GMC, who will now dare attempt to elucidate that mechanism?

    The sun has set on honest, independent science and we are now witness to ‘the dawn of McScience’ and a new era of scientific treachery.

  2. Matty Maccaro 5 September 2008 at 6:00 pm #

    Thanks again , Dr. Briffa for pointing out that these “scientists” abandon science when it serves their interests. Could their interests be pharmaceutical companies and the financial benefits they reap by promoting best interests of these giants? The health of people who are ill and depending on them to behave in an ethical manner is at stake. We all owe to ourselves to assume more responsibility for keeping ourselves informed.

  3. Doris Smithb 5 September 2008 at 6:52 pm #

    I am very concerned as I take atorvistatin and ezetimibe – I had breast cancer last year. I raised the question of taking the two medications for cholesterol with my doctor recently and he dismissed my fears. What should I do?

  4. Brian 5 September 2008 at 7:15 pm #

    This is silly and smacks of the pattern with Keys et al espousing (and ram rodding) the benefits of a low-fat diet and high cholesterol heart disease relationships. The data never supported their notions but somehow, what they believed was right. The results were spurious, the studies not long enough or the right people not studied.

    And when evidence of insulin and carbohydrate began to appear, what did they do? They went on the attack. One can only hope the objectivity and good sense of Drazen et al doesn’t get swept under the rug or buried in dogma!

    Keep up the good fight!

    Brian

  5. Cathy 7 September 2008 at 7:35 am #

    Oh trolleos, trolleos, wherefore art thou trolleos?
    To Doris – I suggest you get, read and absorb Dr Uffe Ravnskov’s “The Cholesterol Myths” or even Dr Malcolm Kendrick’s “Cholesterol Con” (even though he ignored an email of mine via redflags- ahem). And if your doctor dismissed your fears I would try and get a new doctor (granted not so easy in the UK – if that’s where you are).
    Also check out the material on http://www.thincs.org for some spirited, if not always recent thoughts from the medical/scientific establishment.

  6. Cybertiger 7 September 2008 at 5:20 pm #

    Cathy suggested that “you get, read and absorb Dr Uffe Ravnskov’s “The Cholesterol Myths” …

    As it happened, the redoubtable – and independent – Uffe Ravnskov had a BMJ rapid response posted today,

    http://www.bmj.com/cgi/eletters/337/aug21_1/a993#201600

    And well worth the read it was too.

  7. Neil 8 September 2008 at 12:07 pm #

    Well spotted Cybertiger.

    I wonder what are the chances of a well argued rebuttal to Uffe Ravnskov’s Rapid Response?

Trackbacks/Pingbacks

  1. Can low cholesterol cause cancer? | Dr Briffa's Blog - A Good Look at Good Health - 6 April 2012

    [...] Oddly, this result was put down by prominent scientists as likely to be due to ‘chance’, when the statistics showed that it was statistically significant and therefore highly unlikely to be due to chance. You can read more about this here. [...]

  2. Another New Cholesterol-Lowering Drug is Licensed Despite No Evidence of Benefits for Health | Health Impact News - 6 November 2012

    [...] the effect was statistically significant and therefore highly unlikely to be due to chance). See here for more about [...]

Leave a Reply to Matty Maccaro Click here to cancel reply.