Prevention Signal for Statins in Liver Cancer Supports Prophylaxis
San Antonio—Statins appear to provide significant protection against liver cancer and should be considered in patients at high risk for the disease, according to the investigator of the largest meta-analysis conducted to date on the question.
In 20 studies with approximately 2.6 million patients worldwide, use of the cholesterol agents was associated with a 43% reduction in the risk for hepatocellular carcinoma (HCC), said Muhammad T. Sarmini, MD, a gastroenterologist affiliated with the Cleveland Clinic in Cleveland. This new study is not the first meta-analysis, but it is the largest, Dr. Sarmini added.
Three of the studies he assessed were randomized controlled trials, six employed a cohort design, and 11 were case–control studies.
When the data were combined, the odds ratio (OR) of developing HCC was 0.57 for statin users compared with statin nonusers. The confidence intervals were relatively tight and statistically significant (95% CI, 0.49-0.66; P<0.05), he said.
image
The data were consistent when stratified by geographic location: 45% in Asia and 46% in North America, but 29% among studies conducted in Europe (P<0.05 for all). In people infected with the hepatitis B virus, statins reduced the risk for HCC by 45%, which was about the same for those without the virus.
When the data were stratified by study design, the risk reduction for statins was 47% for case–control studies and 37% for cohort studies. The data from the three randomized controlled trials constituted the exception. In these, the risk reduction did not reach significance (OR, 0.7; 95% CI, 0.63-1.5).
Randomized trials represent the strongest test of clinical effectiveness, so the lack of statistical significance raises questions about the potential benefit of the therapy. But Dr. Sarmini maintained that long-term data are critical for showing a treatment effect.
“The follow-up in the randomized trials was relatively short, and this limited the ability to demonstrate a protective effect,” Dr. Sarmini said. In this data set, there were only 81 cases of HCC. Despite the greater weight typically given to randomized trials, Dr. Sarmini doesn’t believe this finding negated the overall conclusion, even if he acknowledged that definitive evidence of benefit from statins requires “prospective randomized research.”
However, he believes, based on this large meta-analysis, “statins should be considered in high-risk patients for HCC, such as those with a viral hepatitis infection.”
One of the limitations of the new work was a lack of information about types, doses or duration of statin exposure. The relevance of these characteristics was underlined by a recently published study of Swedish registry data that explored this same question (Ann Intern Med 2019;171[5]:318-327). This study, which Dr. Sarmini did not include in his meta-analysis, looked at 8,334 statin users matched to the same number of nonusers. The researchers found that lipophilic (hazard ratio [HR], 0.57) but not hydrophilic (HR, 0.95) statins were associated with a significant HCC risk reduction. The risk reduction with lipophilic statins appeared to be dose dependent.
The new meta-analysis “shows a remarkably strong signal that statins might have salutary effects in patients with cirrhosis and reduce carcinogenesis,” said David E. Kaplan, MD, an associate professor at the University of Pennsylvania, in Philadelphia. But Dr. Kaplan cautioned that meta-analyses “cannot remove the fundamental inherent bias, confounding by indication.”
Such bias is not only present in general but might be particularly relevant in this specific situation, he added. “Only cirrhotic patients whose livers are still compensated enough to make enough cholesterol to have hypercholesterolemia receive statins, so prescriptions are given predominantly to patients with lower intrinsic risk of hepatocellular carcinoma,” Dr. Kaplan explained. “Only an adequately powered large randomized controlled trial can truly answer this critical question.”