The recent Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) Trial showed that the combined treatment had no impact on the progression of aortic stenosis – but an excess of incident cancers was identified in the simvastatin-ezetimibe group when compared to placebo (105 vs 70, p=0.01). Deaths from cancer were also more frequent in the simvastatin-ezetimibe group (39 deaths vs 23 in the placebo arm, p=0.05). This was unexpected and required further investigation. Ezetimibe prevents the absorption not only of cholesterol but also of phytosterols and other phytonutrients that have been implicated to have protective roles against cancer. It is therefore possible that the drug could have an effect on the growth of cancer cells.
The Clinical Trial Studies Unit and Epidemiological Studies Unit at Oxford University had access to interim cancer data from 2 large ongoing clinical trials, the Study of Heart and Renal Protection (SHARP) and the Improved Reduction of Outcomes:Vytorin Efficacy International Trial (IMPROVE-IT). These are ongoing studies and have been unblinded for cancer outcomes only. The follow-up to date is short – with a mean of 2.7 years and 1.0 years respectively, compared to 4.1 years in the SEAS trial. However the large size of these trials (n=9264 patients in SHARP and n=11353 in IMPROVE-IT) compared to SEAS (n=1873), means that more data with respect to cancer is available. The Oxford team sought to assess whether the excess in incident cancers and cancer deaths was also seen in the ongoing trials. Was the increased risk of cancer in the SEAS trial due to chance or was this a true finding?
The analysis by the Oxford Group failed to confirm the increase in incidence of cancer recorded in SEAS (p=0.61 for incidence of all cancers in the combined IMPROVE –IT and SHARP active treatment arms when compared to placebo). However a non-significant increase in cancer mortality was observed (p=0.07). Whilst none of these trials were specifically designed to look at cancer risk, cancer death is an end-point that should be reliable. On combining the cancer mortality data from SEAS, SHARP and IMPROVE-IT there was an increase in cancer mortality risk in the active treatment arms (134 deaths) compared to controls (92 deaths, risk ratio 1.45, 95% CI 1.02-2.05, p=0.007).
The results are the outcome from several data driven analyses and do not test a single pre-specified hypothesis and therefore must be interpreted with caution. The findings would suggest that the initial findings from SEAS are due to chance as if there is an increase in the risk of cancer death then an increase in the incidence of cancer should also be observed. This was not the case in the combined analysis. There is a serious risk of misinterpreting interim data and also a risk to the integrity of ongoing trials from interim data release. Safety trials need to be specifically designed to rule out a level of increased risk that is clinically unacceptable when compared to the level of benefit achieved by the intervention. For the interim physicians and patients are left with a quandary as to the efficacy and safety of ezetimibe.
- Peto R, Emberson J, Landray M etal. Analyses of Cancer Data from Three Ezetimibe Trials. N Engl J Med 2008;359:1357-66
- Drazen JM, D’Agostino RB, Ware JH etal (ed). Ezetimibe and Cancer – An uncertain Association. N Engl J Med 2008;359:1398-9
- Fleming TR (ed). Identifying and Addressing Safety Signals in Clinical Trials. N Engl J Med 2008;359:1400-2