Reality Check

APOE e4, the gene variant associated with Alzheimer’s, is now stirring up drama in the world of contact sports.

Apolipoprotein E epsilon 4 (APOE e4) is the genomics equivalent of Kim Kardashian: a gene variant that seems to have developed a taste for the limelight. Already established as the most significant genetic risk factor for late-onset Alzheimer’s disease (LOAD), APOE e4 has more recently been linked to the risk of a poor recovery after traumatic brain injury (TBI), plunging the gene into the ongoing furor over contact sports and the risk of concussion.

What we know of the association between APOE and poor recovery from TBI raises as many questions as it answers, but there is substantial evidence to suggest that it is a risk factor. Multiple studies have demonstrated a link between the APOE e4 variant and worse outcomes after TBI, including prolonged recovery times and worse performance on cognitive tests weeks and months following the injury. In a 1997 case-control study, carriers of a single copy of the APOE e4 allele were twice as likely to have a poor outcome six months post-TBI, and subsequent studies suggested that a poor outcome was 3.5 to 14 times more likely for APOE e4 carriers.

Debates about the use of APOE testing have until now focused solely on its well-documented link to LOAD. Central to the arguments against routine testing is the absence of a medical intervention for LOAD. In fact, in a 2011 commentary in Genetics in Medicine by Jonathan Berg, Muin Khoury, and James Evans detailing a binning system for deciding what to do with incidental findings that result from genome sequencing, APOE was a poster child for Bin 2: “Clinically valid but not directly actionable.”

The case against APOE genotyping hinges on the notion that susceptibility testing has some risk and limited value. Carrying even two copies of APOE e4 is not a definitive test for LOAD — the variant is neither necessary nor sufficient to develop the disease — and in general clinicians have been uneasy with the idea of giving healthy individuals information about their risk for a frightening disease that they cannot prevent. Concern about the possibility of psychological distress and the potential for stigma or discrimination has led a number of professional organizations to advise against the routine use of APOE genotyping, or to caution that it should be done only in the context of expert genetic counseling.

Perhaps people deserve the right to decide for themselves if carrying an APOE e4 allele might be a sort of tipping point in the difficult choice of whether to participate in sports with a high risk of traumatic brain injury.

On the other side of this debate is the widely cited REVEAL study, which looked at disclosure of APOE status to the adult children of parents with LOAD. In this healthy, albeit acutely concerned, cohort, the anxiety provoked by APOE genotyping was minimal and transient. The REVEAL study is not so much an argument for genotyping as it is an argument against the idea that people cannot handle this type of information. However, as the authors themselves have pointed out, “if APOE genotyping had been provided without genetic counseling or to subjects who had no family history of Alzheimer’s disease, the results might have been different.”

The connection to TBI, on the other hand, might be an argument for offering genotyping more routinely. It is true that APOE genotype will not definitively predict who will fall within the 10 to 20 percent of individuals with TBI who experience symptoms for months or even years. But concussions are a major concern for young athletes and their families. Football is the sport most affected by media coverage of concussion risk. For example, participation in the Pop Warner youth football league, the nation’s largest youth football program, fell almost 10 percent from 2010 to 2012; the league’s chief medical officer identified concern over head injuries as the No. 1 reason.

Unlike LOAD, TBI is a risk that can be mitigated. Should we be offering an imperfect test for susceptibility that may turn out to be understated or overstated — or totally wrong? Many are concerned that we will be unable to explain the limitations of the test. But isn’t this, in all its limited and uncertain glory, exactly what personalized medicine is going to be? Perhaps people deserve the right to decide for themselves if carrying an APOE e4 allele might be a sort of tipping point in the difficult choice of whether to participate in sports with a high risk of traumatic brain injury.

This isn’t a simple question. A clinician could make the argument that APOE genotyping should be available but not offered in a medical setting. Offering the test clinically gives it an imprimatur of reliability that it may not deserve, while relegating use to direct-to-consumer (DTC) venues puts the important job of explaining the test into the hands of people who are not incentivized to stress the limitations. A recent article in STAT detailed efforts by several DTC companies to market APOE genotyping to athletes, none with any success. Athletes, the article concluded, are not interested.

Recently, my colleagues and I published a study suggesting that the failure to attract an audience may say more about the messenger than the message. Our study of 845 NCAA athletes showed an enormous openness to genetic testing for an increased risk of poor recovery after TBI, with nearly three-quarters describing themselves as “possibly” or “very” interested in testing. We explained the possibility of learning about a potential susceptibility to LOAD, which we had hypothesized might be a deterrent to testing, and found that it actually made student-athletes more interested in testing. We asked hypothetical questions about their willingness to share the information, and despite acknowledging real concerns about how it might affect them, the vast majority of students indicated that they would share their APOE genotype with parents, coaches, and physicians.

The answers we need are less about saying no on behalf of our patients and their loved ones and more about figuring out ways to explain uncertainty and put testing in context.

Although we explicitly raised the issue that it could affect a coach’s willingness to play them or a doctor’s willingness to clear them for return to play after injury, a majority of student-athletes said that they did not believe this would influence their parents, coaches, and physicians, or that it might, but they would disclose nonetheless. More of them said they would be concerned about telling their doctors (41 percent) than their coaches (31 percent). Almost all of the student-athletes (93 percent) were willing to undergo APOE testing if it were required by the school as a condition of participation.

Is it far-fetched to imagine schools or leagues requiring testing? All NCAA schools already require sickle cell carrier status testing, despite the absence of definitive information on the implications of being a carrier. (Intensive exercise is thought to have caused the sudden deaths of several young athletes who carried a single sickle cell mutation.) TBI, by contrast, poses a far more extensive problem for leagues and schools. A lawsuit against the NCAA settled in 2015 allows athletes to sue individual institutions for damages secondary to concussions sustained as players. A suit filed this August accuses the Pop Warner football league of negligence and fraudulent misrepresentation with regard to the risk associated with youth tackle football. For athletic programs, the liability involved with high-risk contact sports may be an existential issue. But while individuals may choose to use a test like APOE e4 despite its limitations and uncertainties, third parties should not have that option.

It’s been argued that student-athletes are not well positioned to advocate for privacy rights, and our survey does nothing to contradict that assumption. One argument for making APOE genotyping available clinically might be to ensure that the decision remains in the hands of athletes and their families, as well as to provide them with unbiased counseling and some explanation of the link to LOAD. Is this likely to happen in a doctor’s office? Perhaps not, but it is certainly more likely to happen there than anywhere else.

APOE testing is complicated by its connection to multiple conditions and by uncertainty, both of which are going to be the rule and not the exception as we plow ahead with genomic medicine. We can put gene variants into bins, but they keep climbing back out of them.

The answers we need are less about saying no on behalf of our patients and their loved ones and more about figuring out ways to explain uncertainty and put testing in context. If we can do that, then we can offer the public more reality check and less reality television.