October 29th, 2013
Using CAC Scores to Determine the Best Candidates for a Polypill
CardioExchange’s John Ryan interviewed Khurram Nasir about his research group’s JACC study, which examines whether the coronary artery calcium score (CAC) can be used to determine the target population for a polypill. Results show that patients with a CAC=0 had a very low event rate and high number needed to treat (NNT), whereas the majority of CVD events occurred in patients with a CAC of >100. The authors conclude that avoiding treatment in those with a CAC=0 could allow for a more selective use of the polypill within the population.
Ryan: In your study, you say that the NNT to see a benefit of a polypill in patients with a CAC>100, is ~20. How does this compare with using just clinical and laboratory parameters to decide whether to use the polypill?
Nasir: Across subgroups that met the inclusion and exclusion criteria based on the clinical and laboratory parameters for the four suggested polypill regimens, the NNT in five years to prevent one CVD event ranged from 36-57 and 55-85 for CHD, respectively.
Ryan: You argue that the NNT for a CAC=0 is possibly too high to make it worthwhile. Is there a way of predicting which patients with a CAC=0 will go on to have events that can be applied here?
Nasir: The basic hypothesis for this study stems from observation in a series of reports from our group including a large meta-analysis and retrospective and prospective studies that the absence of CAC confers a favorable prognosis with a very low event rate of ~1 per 1000 patient-years (~1% 10-year event rate). As confirmed in sensitivity analyses in the current study, even if the reduction in the relative risk were as high as 95%, the benefit would be minimal and the resultant NNT for 5 years to prevent a CVD event will be greater than 50 with polypill use. At the same, I do agree that with additional information and testing we may able to identify a subgroup among those with CAC=0 who will be at a relatively higher risk, however, the yield is likely to be minimal for the efforts and resources needed.
Ryan: What are your concerns about introducing CAC as a population-based screening tool?
Nasir: I would like to stress that the idea we are proposing is very different from “screening” with CAC testing, where the goals would be early detection and halting disease progression based on proven interventions, an approach that has its own pros and cons and can be extensively debated.
In this study, we emphasize a unique concept of “de-risking” individuals already deemed high risk enough to be considered for lifelong medications. As extensively demonstrated in the past and now in the current study, CAC testing has the unique ability to identify a sizeable proportion (nearly 50%) who have no signs of early atherosclerotic disease, an extremely low risk of incident CVD, and — as a result — an unacceptably high NNT to prevent one cardiac event.
Our message is very clear: since risk factors and biomarkers cannot really rule out disease, they can only be used to raise risk estimates, and, thus, are inextricably tied to more treatment and unanticipated downstream effects. In this situation, prior to committing to lifelong therapies, the “Power of Zero”, irrespective of traditional baseline risk, provides an excellent opportunity to avoid over-treating and its associated risks. This strategy will allow us to focus on individuals with actual disease, among whom the majority of clinical events will occur, and potentially derive the highest yield from our preventive strategies.
Ryan: The polypill is intended for developing countries. Are you suggesting that they buy a lot of CT scanners?
Nasir: We do not dictate or advocate developing additional assets and infrastructures to cater to this specific need. This will not be an effective use of limited health care resources.
However, at the same time, we need to realize that the concept of a polypill is gaining significant interest in developed countries such as Canada and U.K., as well as in emerging markets such as China and India where we are experiencing an increased penetration of CT technologies in local health care systems.
Depending upon local situations, I think our findings do provide a potential framework to consider whether the added information based on CAC can used for shared decision making on how best to appropriately allocate resources.
While considering the role of CAC testing in these situations, individual stakeholders need to carefully weigh their respective health care systems’ needs, goals, and available resources. Among many important considerations are one’s ability to negotiate a reasonable cost of CAC testing, determine costs for lifelong medication/follow up visits/labs, issues of compliance, side effects of these combination pills, and the potential downstream implications of wider CAC testing.
However, one of the most important things that we seem to ignore and is currently hard to measure is “individual preference”: most of us will agree that we’d like to avoid exposure to unnecessary medications. There is a huge gap in our ability to grasp the importance of how patients (the most important stakeholder) perceive this issue, and can only be better understood by closely engaging them in these decision-making processes.