July 8, 2016

expert reaction to conference abstract about high cholesterol and survival in people with cancer

A conference abstract, presented at the Frontiers in CardioVascular Biology, looked at high cholesterol and survival in people with cancer.

Prof Jack Cuzick, Director of the Wolfson Institute of Preventive Medicine and Head of Centre for Cancer Prevention, Queen Mary University of London (QMUL), said:

“This is almost impossible to interpret accurately from the data provided. How was hyperlipidaemia diagnosed and were tests done on all 929 552 patients or a selected subset? Was hyperlipidaemia linked in any way with cancer incidence or just case fatality? If hyperlipidaemia leads to more of these cancers, but the excess cases have a tendency to be the less aggressive ones, the lower case fatality rate may not be associated with a lower overall mortality rate. This has been seen for HRT and breast cancer, where HRT increases risk, but survival is better, yet overall breast cancer mortality in HRT users appears to also be higher.  For many of these cancers (especially breast and prostate) cancer specific mortality is low so it is important to separate death from the cancer of interest from total (crude) mortality which would be heavily influenced by other causes of death. Weight is a potential confounding factor, as it is related to both survival following cancer diagnosis and hyperlipidaemia. No adjustment for this appears to have been done.”

Prof Sir Richard Peto, Professor of Medical Statistics & Epidemiology and Co-Director of the Clinical Trial Service Unit, University of Oxford, said:

“This suggestion of a protective effect of statins against death from cancer should not be taken seriously, because:

“1. We know that advanced cancer can reduce blood cholesterol substantially (which could well entirely account for their findings); and

“2. We know from large-scale randomised trials that statins have little or no effect on cancer incidence or cancer mortality (so their claim of a protective effect in advanced cancer is implausible).”

Dr Mangesh Thorat, Research Fellow, Centre for Cancer Prevention, Queen Mary University of London, said:

“Although statin use has resulted in a considerable decline in cardiovascular mortality and emerging observational and laboratory data suggest a potential role in cancer, this study has not investigated the actual statin use, so it is not possible to draw any firm conclusions regarding statins in this study. Individuals with high cholesterol could be on several medications and drawing conclusions without carefully analysing actual medication use in all participants is fraught with a serious risk of misinterpretation of data.

Can you tell from the information available whether this is a good quality study?

“I would rate it as incomplete for not having investigated medication use.

Has the study been through peer review yet?

“Abstracts are usually peer reviewed, but the timebound nature of this process and limited data presented often means that it is often not as rigorous as a published full-paper.

Are the conclusions backed up by solid data?

“No, because statin use and other medication use was not measured.

How do the authors know the effect is due to statins rather than to the high cholesterol itself?

“There is no way to know as they have not measured or studied statin use. They only hypothesise. And they also acknowledge in the press release that use of other drugs could potentially confound these results.

How does this work fit with the existing evidence?

“Some very early observational and laboratory data suggests statins may have a potential role in reducing cancer incidence and mortality but more research is needed.

Have the authors accounted for confounders?  Are there important limitations to be aware of?

“They have not accounted for potential confounders, which is a major issue.

What might be a mechanism for the effect observed?

“The evidence here is too weak to speculate.”

Dr Tim Chico, Reader in Cardiovascular Medicine / consultant cardiologist, University of Sheffield, said:

“Conference papers such as this are always preliminary data for discussion, and do not contain enough details to properly evaluate the findings. As such, these findings should be seen as extremely preliminary and certainly should not influence current practice or recommendations.

“There is no evidence presented to say that cancer patients who lived longer were on statins; they simply had a diagnosis of high cholesterol. The authors of this study appear to presume such patients will be taking statins. This would need to be clarified even before we could begin to talk about whether statin treatment might be associated with a better outlook in people with cancer.

“Finding that cancer patients who also have a diagnosis of high cholesterol have a better outcome does not prove any causal link, and although the study attempted to adjust for other factors such as age, gender, and ethnicity, many other factors could explain such a finding.”

* Conference abstract title: ‘Hyperlipidaemia reduces mortality in breast, prostate, lung and bowel cancer’ by PR. Carter et al. will be presented at the Frontiers in CardioVascular Biology conference at 16:00 UK time on Friday 8 July 2016. There is no journal paper.

Declared interests

Prof Jack Cuzick: “No conflicts of interest.”

Prof Sir Richard Peto: “CTSU has a policy of not accepting honoraria, consultancies or other payments directly or indirectly from pharma, other than the reimbursement of travel and accommodation for particular scientific meetings.  The CTSU conducts, analyses and interprets its clinical trials (for which it serves as the regulatory sponsor) and other research independently of the pharmaceutical industry funders and other funders, with the datasets held by the CTSU rather than by the funders. In accordance with our long-term policy, honoraria, consultancy or other payments have not been received directly or indirectly from industry, either personally by me or by the University (except for reimbursement of travel and accommodation for taking part in relevant scientific meetings).”

Dr Mangesh Thorat: “No relevant conflicts.”

Dr Tim Chico: “I am a committee member and Treasurer of the British Atherosclerosis Society, a charity established in 1999 to promote UK atherosclerosis research.”