乌鸦传媒

A genetic marker used to find patients at risk for heart disease that can be significantly mitigated with statin therapy may not be predictive, researchers say.

In a large replication study involving several case-control analyses, mutations in the KIF6 gene were not associated with coronary artery disease, Themistocles L. Assimes, MD, PhD, of Stanford University, and colleagues reported in the Journal of the American College of Cardiology.

"This is such a big study [that] if there was a significant association between this variant and coronary disease, we would have found it," Tom Quertermous, MD, of Stanford, senior author of the study, said in a statement. The final publication included 69 authors.

Earlier studies had shown that the Trp719Arg polymorphism in the KIF6 gene was associated with an increased risk of coronary artery disease, particularly analyses from the WOSCOPS and CARE trials.

An analysis of data from PROVE IT-TIMI 22 showed statins potentially mitigated some of this risk, as did an analysis of data from elderly patients in the PROSPER study.

Those analyses were largely supported by and performed by researchers from genetic testing company Celera, of Human Genome Project fame, which markets its KIF6 test, Statincheck, to healthcare professionals through its subsidiary Berkeley Heart Labs.

The company has performed 230,000 tests since the KIF6 analysis became available in 2008.

release on its website pointing out several criticisms of the present analysis, including its reliance on case-control studies, not prospective data that the WOSCOPS, CARE, and PROVE IT provided.

"In case-control studies, you don't see an increased risk, but in prospective studies you do see an increased risk," James J. Devlin, PhD, senior director of cardiovascular research at Celera, told 乌鸦传媒. "The question is how do you square those observations?"

Eric Topol, MD, director of the Scripps Translational Science Institute in La Jolla, Calif., told 乌鸦传媒 that the company's analyses were retrospective candidate gene studies, the type of study which "often reveal[s] a spurious positive result."

"The KIF6 variant never showed up in any genome-wide study," he said.

Assimes and colleagues looked at the association in 19 case-control studies of non-fatal coronary artery disease that were either genome-wide association studies or an attempt to replicate the initial positive reports.

These studies included ADVANCE, deCODE CAD, FINRISK, and INTERHEART, among others.

There were a total of 17,000 cases and 39,369 controls, the vast majority of European descent but also including a "modest number" of South Asians, African Americans, and Hispanics.

The researchers found that none of the 19 studies showed an increased risk of coronary artery disease in carriers of the 719Arg allele compared with non-carriers (pooled OR 0.98, 95% CI 0.95 to 1.02).

In fact, one study -- deCODE -- showed a 7% decreased risk of disease.

There was no association for ethnic groups (South Asians in INTERHEART; African Americans in ADVANCE, CATHGEN, and GeneSTAR; or Hispanics and East Asians in ADVANCE) or for those with early-onset disease.

Regression analyses ruled out an increase of 2% or more in heart disease among European 719Arg carriers, the researchers said.

They acknowledged that their study was potentially limited by selection bias since they looked at only non-fatal cases of coronary artery disease. If the mutation increases risk of fatal heart disease, the data could be skewed, although that's not likely, they said.

They also could not control for all traditional risk factors of coronary artery disease, and didn't explore whether statin use modified the relationship because information on drug use wasn't available.

In an accompanying editorial, Topol and Samir B. Damani, MD, PharmD, also of Scripps, wrote that the present study was "well-conducted and elegant," pointing out that several other markers of heart disease have been validated, including apolipoprotein E, lipoprotein(a), and chromosome 9p21.

They also pointed out a lack of a plausible biological mechanism for the relationship between KIF6, heart disease, and statins.

Devlin said mechanistic research was presented at a sub-meeting of the American Heart Association last spring that linked KIF6 to diseased atherosclerotic plaques in mice and humans, and wasn't expressed in normal arteries.

He added that researchers are currently performing other mechanistic studies, and are looking to assess more randomized controlled data from trials of statin therapy, which the present study doesn't assess.

Topol said the findings imply that "we desperately need better regulation for genetic testing." The test currently does not need to be FDA approved, but Celera said it is moving forward with plans to register a pre-market approval application by the end of the year.

The test has already received a CE mark in Europe, and the company has entered into a marketing agreement there with Abbott.

But Topol and Damani wrote that "the KIF6 story should serve as a valuable reminder of the potential pitfalls present in prematurely adopting a genomic test without sufficient evidence."

In a statement, Assimes said, "we know from previous experience that a positive association between a genetic variant and a common disease, such as coronary disease, needs to be consistently observed in many human population studies before it can be believed."