PROSPER: Pravastatin reduces cardiovascular events in the elderly

November 18, 2002

Chicago, IL -Pravastatin was associated with a significant 15% reduction in the primary end point of coronary death, MI, or stroke in the Prospective Study of Pravastatin in the Elderly at Risk of vascular disease (PROSPER) trial, reported at the American Heart Association November 2002 meeting today. The results are also published in the Lancet this week[1].

Coronary mortality was reduced by 24%, but no effect was seen on stroke, cognitive function, or disability. There was also a statistically significant increase in cancer, but the investigators dismissed this as the play of chance.

Principal investigator Prof James Shepherd (University of Glasgow, Scotland) said the results extended to elderly individuals the statin treatment strategy currently used in middle-aged individuals.

Most studies previously conducted with statins have excluded elderly patients, as the association between plasma cholesterol and risk of coronary artery disease diminishes with increasing age. However, previous trials have suggested a reduced risk of stroke with statins, and results of observational studies have raised the possibility that statins may reduce the rate of cognitive decline in the elderly. The PROSPER trial was therefore conducted to investigate these issues further.

The trial randomized 5804 men and women aged 70 to 82 with a history of or risk factors for vascular disease to pravastatin 40 mg once daily or to placebo. They were followed for 3.5 years, which Shepherd noted was a shorter time than other statin trials, but as these people were in their last decade of life, the investigators believed a shorter follow-up time was more appropriate. He said that because of this, they did not expect to see a reduction in mortality, which has been shown in some of the previous statin trials with longer follow-up times.

But as it turns out, coronary mortality was the 1 component of the primary composite end point that was significantly reduced with pravastatin.

PROSPER: Major results

End point

Placebo (%)

Pravastatin (%)

Hazard ratio


Coronary heart disease death or nonfatal MI or fatal or nonfatal stroke 16.2 14.1 0.85 0.014
Coronary heart disease death or nonfatal MI 12.2 10.1 0.81 0.006
Fatal or nonfatal stroke 4.5 4.7 1.03 0.81
Nonfatal MI 8.7 7.7 0.86 0.10
Nonfatal stroke 4.1 4.0 0.98 0.85
Transient ischemic attack 3.5 2.7 0.75 0.051
All cardiovascular events 18.0 15.7 0.85 0.012

Deaths from various causes

End point

Placebo (%)

Pravastatin (%)

Hazard ratio


Coronary heart disease 4.2 3.3 0.76 0.043
Stroke 0.5 0.8 1.57 0.19
Vascular 5.4 4.7 0.85 0.16
Nonvascular 5.1 5.6 1.11 0.38
Cancer 3.1 4.0 1.28 0.082
Trauma or suicide 0.2 0.1 NA NA
All causes 10.5 10.3 0.97 0.74

Shepherd noted that 48 people would need to be treated to prevent 1 primary end-point event and that among the highest-risk patients the number needed to treat to prevent 1 event came down to 25.

On the issue of why there was no reduction in stroke, Shepherd said he did not think the trial had long enough follow-up to show this. "In previous trials, prevention of strokes shows up at about 5 years," he commented. He added that because there was no reduction in stroke, it was no surprise that there was no reduction in cognitive decline, especially as all patients had good cognitive function at entry. The cognitive decline results mirror those of the Heart Protection Study with simvastatin, which also showed no effect on this end point.

Subgroups: more benefit in those with low HDLs at entry

Subgroup analysis showed that the treatment effect was the same in most subgroups, such as those with/without previous vascular disease, men/women, smokers/nonsmokers, and those with/without hypertension and regardless of baseline LDL cholesterol. However, there did seem to be a relationship with baseline HDL level, with those individuals with lowest HDL levels gaining the most benefit from treatment, Shepherd noted.

Discussing the trial, Dr Stephen Fortman (Stanford University, Palo Alto) agreed with Shepherd that these results should encourage broader application of the statins in older patients but added that "what we really need to do is to identify these individuals at a younger age." He said it was interesting that the 34% decrease in LDL with pravastatin in this trial was associated with only a 19% reduction in MI. "This suggests that cholesterol reduction is probably more potent at younger ages."

Increase in cancer no concern?

The 1 controversial finding of the study was a significant 25% increase in the incidence of cancer in the pravastatin group, with 199 new cases of cancer found in the placebo group vs 245 in the pravastatin group. Many years ago, when the statins were first available, there were concerns about lowering cholesterol in the elderly after an inverse relationship between plasma cholesterol and cancer rates were reported, especially in older persons. But this concern has been allayed by the large-scale trials of statins, which have not suggested a link.

Shepherd said he was not concerned about the increased cancer risk found in this study for several reasons. These included the fact that there was no pattern that indicated cancer of any 1 particular tissue type was increased and that the general increase appeared in the first year of follow-up, which was "far too soon" to have been caused by pravastatin. In addition, from the incidence range of cancer in this age range of people, it would have been expected to see 290 cancers in each group. "As both groups were lower than this, we can be reassured that individuals recruited into PROSPER brought their cancers with them in a hidden state," Shepherd commented. The investigators also incorporated the cancer finding in a meta-analysis of all statin trials that showed no overall increase in risk.

Fortman agreed that the cancer increase was most likely a chance finding but said that it did warrant further close observation, especially as this was the first trial of the statins specifically in the elderly. "I would like to see a meta-analysis of all elderly patients from all the statins trials," he added.

Shepherd concluded that the important outcome in this trial was the 24% reduction in coronary mortality, and there was thus clearly no justification in withholding statin therapy from the elderly.

Generally well received

Most cardiologists questioned by heartwire after the presentation said they would now treat their elderly patients if they weren't doing so already. Dr Doug Weaver (Henry Ford Hospital, Detroit) commented that: "The relative benefit is not as great as in younger patients, but the absolute benefit is still very good. I will treat my elderly patients." Dr Jennifer Adgey (Belfast, Northern Ireland) said she was "very surprised" that there was no reduction in stroke, but that she already used statins in all her elderly patients with CHD. She added that she would like to see more information on the half of the PROSPER population without CHD.

Dr Harvey White (Green Lane Hospital, Auckland, New Zealand) was also disappointed that there was no reduction in stroke, but said he would definitely treat elderly patients for the CHD benefit. "It's a pity they couldn't continue the study to look for an effect on stroke, as that is what we all wanted to see in this population, but we can't deprive our patients of a reduction in heart disease mortality, so I guess now we'll never know."

On the cancer increase, Weaver said he wouldn't discount this altogether. "This is probably just a chance finding, but it needs further investigation. It is interesting that breast and gastrointestinal were the 2 cancers that showed the largest increase, as these are the 2 that have provoked concern before. I would like to see a meta-analysis of these 2 cancers from all the previous studies in all patients and also in just the elderly patients." Adgey and White were less concerned. Adgey commented: "I found the fact that the expected cancer cases in this age group were much higher than shown in either group very reassuring. I don't think there is a problem." White agreed, saying he didn't think cancer was an issue.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.