Michelle Christina Johansen, MD

Tada H, Shiffman D, Smith JG, Sjögren M, Lubitz SA, Ellinor PT, et al. Twelve–Single Nucleotide Polymorphism Genetic Risk Score Identifies Individuals at Increased Risk for Future Atrial Fibrillation and Stroke. Stroke. 2014

The role of genetics in determining fate is no longer a topic reserved for science fiction. For decades the practicing neurologist has recognized the importance of understanding genetically based disease processes and strived for potential treatment targets. Modern science is ushering in improved ways to probe the human genome allowing for a new frontier where genes are used to predict disease and assess risk. 



Melander et al utilized a community based cohort in Malmo Sweden of over 30,000 patients in search of a biomarker that could identify individuals at increased risk for atrial fibrillation (AF).  The team designed a population based prospective study to test two hypotheses:

1) a AF genetic risk score (AF-GRS) combining 12 single nucleotide polymorphisms (SNPs) would associate with risk for AF beyond the established risk factors (hypertension, smoking, obesity, diabetes, age, male sex and heart disease)
2) AF-GRS composed of the same 12 SNPs would also be associated with risk of ischemic stroke.

The study was composed of 27,471 patients from the Malmo Diet and Cancer observational cohort excluding only those who did not provide enough DNA for genotyping, those with prevalent AF and one lost to follow up. When assessing for stroke risk, those with TIAs, prevalent ischemic stroke or hemorrhagic strokes were excluded from analysis. Patient characteristics such as hypertension, diabetes, body mass index, cigarette use, prevalence of AF/ischemic stroke/coronary heart disease at time of enrollment were determined by medical record and ICD 9 codes.

The primary end point was time to first occurrence of AF and time to first occurrence of stroke.  These outcomes were determined by diagnosis code and stroke was validated within the Malmo stroke registry. 

During the follow up period (median 14yrs), 2,160 participants suffered a first AF event and 1,495 had a first ischemic stroke. The AF-GRS was significantlyassociated with incident AF after adjusting for established risk factors. The genetic risk score was separated into quintiles based on the contribution of differently weighted SNPs. The 12 SNPs were chosen based on prior literature suggesting a linkage to AF. Those patients in the top quintile of the AF-GRS had a two-fold increased risk of AF compared to the bottom quintile. 

The authors interestingly compared this risk of AF to that caused by hypertension. The magnitude was equivalent or having a high AF-GRS appeared to confer the same risk as having hypertension. 

Similarly, those in the top quintile of the AF-GRS had about 23% greater risk of ischemic stroke compared to the bottom quintile. This risk was not as strong as that conferred by a diagnosis of hypertension. The authors took the data one step further and investigated the association between AF-GRS and ischemic stroke in patients WITH a diagnosis of AF that preceded or coincided with the stroke. There was felt to be an association between the AF-GRS and these stroke events (HR 1.81).

The authors unfortunately did not perform a head to head comparison of CHADS2 to the AF-GRS but did find that adding AF-GRS to CHADS2 improved reclassification of patients into a higher degree of risk. 

So where does this leave the practicing Vascular Neurologist? Are these results generalizable? 

The authors acknowledge that the study was limited to a population in Malmo Sweden. They discuss another recently performed study where the same SNPs were associated with risk in Europeans as well as a case-control study that demonstrated association in a Japanese population. 

Is this enough? Can these results be applied in other countries including the United States? A review of the demographics of the Malmo participants demonstrates that the mean age was 58 and the BMI was 25. In America, we face an increasingly aging and obese population. We know that genetics as well as environment play a role in the pathophysiology of stroke. While the results of this study are exciting and promising, a similar study would need to be performed to ensure its clinical utility.

Assuming that the results do apply to all demographics, how should this impact our practice? If results indicate a patient falls into the top quintile of the AF-GRS, how does one proceed? The authors suggest that this cohort might be placed on more intensive monitoring for AF but should we wait to confirm to treat knowing that the incidence of paroxysmal AF increases with age? Should we subject these patients to implantable loop recorders? How aggressive do we need to be balancing investigation with the risks of presumptive anticoagulation?

Melander et al has provided the Stroke community with a thought provoking study, the nature of which will continue to increase in the upcoming years as the cost of genome sequencing declines. While the prospect of learning more about the genetics behind a disease process is thrilling, the neurologist of today needs to become acutely attuned to the ethical and practical implications of these results.