By Gabrielle Strobel

"I am finding out this month. If I have it, I will never have kids. It's got to stop somewhere in our family." Tom Drury (a pseudonym), 36.

"If I knew I carried the mutation, I would not be able to get up in the morning." Jane Smith (a pseudonym), 26, at risk for eFAD.

Interview with a Genetic Counselor: Jennifer Williamson is a genetic counselor at the Taub Institute for Research on Alzheimer's Disease and the Aging Brain and the Gertrude H. Sergievsky Center at Columbia University, New York 

Genetic Status Protocol: Get a sense of the careful process you may want to ask for should you decide to find out about your genetic status.

Introduction 
Early onset familial Alzheimer disease (eFAD) is an autosomal-dominant genetic disease. This means that in each family it is caused by a mutation in a single gene, and that a single copy of the mutant gene, inherited from one parent, will cause the disease. The discovery in the mid 1990s of eFAD genes aroused the specter of knowing one's genetic future, not only for patients but also for their children and entire families. Indeed, the age of genetic testing is here. DNA testing has been used since the mid-1990s for Huntington disease (HD) and certain types of cancer. The experience with these families provides helpful guidance for families with eFAD, who have turned to genetic testing more recently. Indeed, there is a large literature on genetic testing for HD and a small but growing literature for AD.

Genetic tests can be used to confirm the presence of an eFAD gene in a patient who already has symptoms. Testing can also predict who will develop the disease in the future. Whichever way they are used, genetic tests force us to ponder knotty ethical and life questions. If you are today a perfectly healthy 25-year-old who has seen a parent claimed by this disease, would you want to know if you face the same fate—or have escaped it? How will this knowledge change your life? Will knowing this help you, or harm you? A few brave souls have already taken the step of finding out. Some have learned that they do have an eFAD gene. You will read about them in some of the stories that accompany this series. You will learn how these pioneers have already entered the world of genetic medicine. Some volunteer for trailblazing research studies to understand how their disease will develop, and perhaps to have a chance to try to prevent it. Others have used the knowledge that they carry an AD mutation to have a baby by preimplantation genetic testing, to ensure their child will be free of the eFAD gene.

People who choose DNA testing for early onset familial AD represent the kind of patient envisioned for the future of "genomic medicine"—a new science in which each person's genetic makeup would determine the health care geared toward altering his or her genetic prognosis. Today, people who are at risk of eFAD are stuck in a limbo between knowledge of the genes that cause eFAD and ignorance of how to intervene. The genetic testing for eFAD is available but remains bewildering to lay people and even to many physicians. The preventive care is not in place at all, but academic research efforts in this direction are gearing up (see essay: Where to Turn for Research: Human Studies of eFAD). As this area grows—and experts agree it inexorably will—it is important to develop it with exquisite sensitivity for the best interest of the patients and their families. This includes education of families and physicians, as well as more research on how people who have been tested fare over the long term.

Why Do People Request Genetic Testing for AD? 
Candidates for genetic testing cite a range of reasons, some altruistic, others practical, some psychological. Some of those apply to all categories of DNA testing for AD—diagnostic, predictive, susceptibility—while others are primarily at work in predictive testing.

It's important to realize that while genetic testing can indeed help you address many of these issues, it is not necessary for all of them. You can allay some of your concerns without knowing your genetic status, e.g., prepare finances for your children, draw up advance directives. Genetic counseling can be invaluable to sort those things out. For further reading, see Williamson and LaRusse, 2004.

Checklists for Families and Providers 
Experience with eFAD genetic testing so far has led to some generally accepted recommendations. Every interested family member—not just the family spokesperson—should ask for this information:

• • People are motivated to contribute to genetic research, either for their own benefit or that of their younger relatives and coming generations.
  • People hope that effective treatment will be developed in time for them, and hope to be eligible to participate in clinical trials.
  • People cite the need to know. Particularly as they approach the age of disease onset in their family, their anxiety intensifies and they cannot stand it anymore.
  • Some believe they already have the disease. Fear makes them doubt their mental faculty, and question every instance of forgetting. Many of these cases prove not to carry the mutation.
  • People want to plan their finances: long-term care and disability insurance, retirement, advance directives, and will.
  • People want to plan their families: will they have children or not, get married or not?
  • People want to make changes in their lifestyle: spend more time with family, exercise, eat more healthily, etc.
  • People want to know what to tell their children, family.

  Clinicians can consider these questions before initiating genetic testing with a given family:

  About Genetic Testing 
  Genetic testing for early onset familial Alzheimer disease comes in two major varieties. There is diagnostic DNA testing for people who already have symptoms and predictive testing for relatives who may have inherited a definite disease gene. (A third variety, ApoE susceptibility testing, is allowed for diagnosis but is not recommended for predictive risk assessment of healthy people. ApoE susceptibility testing is more relevant for late-onset AD and for AD that is inherited in a non-Mendelian fashion than for eFAD, and is therefore not discussed in detail here.) Genetic testing is also available for other forms of hereditary dementia, for example Creutzfeldt-Jakob disease and frontotemporal dementia, as well as for some forms of inherited Parkinson disease. The specific details differ for each disease, depending on its underlying genetics, but the general issues are similar.

  Diagnostic Testing 
  In cases where a young or middle-aged person has symptoms of AD and there is a family history of early onset AD, doctors may suggest DNA testing. Clinical practice on this is not uniform, as some physicians believe that genetic testing has value, while others feel that it adds little. Typically, doctors only recommend this if there is a previous known case of AD in the family. That affected person donates blood, which is then tested for a mutation in the genes for presenilin 1, presenilin 2, or APP. (In the U.S. only presenilin 1 testing is available commercially. Testing for the last two genes must be pursued through an academic laboratory.) If these tests come back negative, the question of a possible genetic cause of AD in the family stays unanswered. If they come back positive, the doctor can identify the mutation that causes AD in the family. Once a mutation is found, doctors can test if the same mutation occurs in the DNA of additional family members. In those instances where a mutation is found in a fully diagnosed AD case, DNA testing can then complete the diagnosis in other members of the family. Some clinicians think this can be particularly helpful when a relative has early symptoms and (s)he, or the spouse and siblings, suspect AD but would not otherwise be able to know definitively for a few more years until symptoms become overt. Those are angst-ridden years for the family, and testing can lay an agonizing uncertainty to rest. It can also help people with early AD psychologically, as it confirms that their difficulties are not their fault, and that they were right to cut back work if they had to. For some people, the test result helps them plan their future and focus on creating good times with their families.

  Occasionally, medical geneticists use diagnostic testing in a slightly broader way. They sometimes search the DNA of a person in whom they suspect AD for eFAD mutations even when there is not a clear autosomal-dominant inheritance pattern in the family, or in cases where a parent who had early onset dementia has died and no blood or tissue samples from the parent are available for testing. This can then help diagnose further cases of AD in relatives. This kind of search is best done by specialized research groups, who are adept at distinguishing an inconsequential gene change from a true AD mutation (e.g., Rogaeva et al., 2001).

  Diagnostic DNA testing is done most commonly on the presenilin 1 gene. This happens either through a commercial service offered by Athena Diagnostics, or through a CLIA-certified laboratory at an academic medical center that has a clinical genetics program in AD. CLIA-certified labs can formally confirm mutation findings that cropped up first in the context of a research study. Occasionally, DNA testing is done on the APP or presenilin 2 genes. Mutations in these two genes are rarer than presenilin 1 mutations. No commercial test is available for them, and testing occurs at various academic centers. In the U.S., no company offers a combined test for all three FAD genes, such that a practicing neurologist could simply send a patient's sample and get a test result back. Clinicians disagree about whether that is good or bad. The lack of such a comprehensive commercial test contributes to limiting the use of DNA testing, and probably exacerbates the problem of outright misdiagnosis that families with eFAD frequently encounter (see Brad and Megan interview). At the same time, researchers are mindful that the genetics of AD is more complicated to interpret than that of some other heritable diseases, such as Huntington's. Some AD clinical geneticists worry that if tests results were available too easily, people might end up receiving misleading results. They encourage families to pursue diagnostic testing through qualified centers (see EOAD centers list).

  ApoE testing sometimes becomes part of the diagnostic workup of a person with eFAD. ApoE comes in three types (2, 3, and 4). In the general population, people with ApoE4 face a greater risk of developing AD at a younger age. Some clinicians feel that in a person with an autosomal-dominant eFAD mutation, ApoE genotyping can give some added insight into when AD is likely to begin or how aggressive its course is likely to be. However, in those cases the impact of the primary eFAD mutation far outweighs that of ApoE4, and indeed other clinicians argue that ApoE testing adds little information for these particular families. Some research studies are exploring differences in how ApoE variants affect a person's response to a given therapeutic drug. Once this research solidifies, ApoE testing may add real value to the diagnosis. Until then, it is up to the discretion of the medical team.

  In general, medical centers offer genetic testing as a clinical service. Patients would be required to pay for the diagnostic test out of pocket or through their health insurance. A few centers have research studies that can cover the cost of genetic testing for volunteers who agree to participate in a study. In this setting, the researchers need to know which participants carry an eFAD mutation, but not every participant will want to know. Participants can choose whether or not to be told their test results. Part of the reason why people choose to get tested in a research setting is that this information need not become part of their medical record. Each route to genetic testing—clinical service vs. research study—has advantages and disadvantages, and families should ask up front.

  Letting the Genetic Genie Out of the Bottle 
  In general, diagnostic DNA testing is considered the form of testing that is less controversial. It has a clear-cut benefit—after all, nailing an uncertain, difficult diagnosis in someone who is already ill offers clarity for all. But nothing in genetic testing is absolute, and there are situations where diagnostic testing may do more harm than good. This is a danger in "sporadic" cases. The news of a wholly unexpected AD gene in mom or dad, or a sibling with children, would be a bombshell for the family. It's sure to upset children, siblings, and possibly even relatives further removed. Why test DNA in sporadic cases? Familial Alzheimer disease is genetic and sporadic disease is not—right? This widely held assumption is in fact an arbitrary division and is not always true. Granted, clinicians most readily suspect a faulty gene when early onset AD has shown a pattern of reappearing in the family. But on occasion they are also ordering a genetic test in cases where their patient is the only known sufferer. As AD genetics become better known, some physicians send blood for testing simply because a person's symptoms begin unusually early. With some regularity, back comes news of a presenilin mutation. This can be useful in making the correct diagnosis, but it's dangerous, too. If the physician isn't careful, (s)he might end up springing a terrible surprise on unsuspecting young adult children, namely a 50 percent risk of carrying a gene for AD themselves.

  Where the clinical team suspects that "sporadic" AD is not actually sporadic, it's important that they move cautiously and give the family ample opportunity to come to terms with this new prospect. Depending on the family's circumstances, the delicate balance of a physician's duty to warn versus a family's right not to know may come down on the side of doing less rather than more.

  Consider some examples. Genetic counselor Jill Goldman, then at the University of California, San Francisco, and her colleagues there recently explored this issue with a series of case reports. One tells the story of a 59-year-old Hispanic immigrant who had had early onset AD for 12 years when her husband and daughter brought her in. The family had no health insurance and paid for the visit out of pocket. A genetic counseling session soon made clear that they were neither prepared to learn of the potential risk to their family, nor in a position to benefit from a genetic diagnosis. They worried about the cost of the test and about having to scare their relatives. For them, it seemed as if any further disclosure of genetic risk might cause mostly disruption and anxiety. The father and daughter decided against genetic testing.

  The question that lingered with Goldman was, How can the physician and counselor know beforehand if the caregivers want to be made aware of a possible genetic burden and its implications for the rest of the family? In other words, did this family want to know even as much as they heard in their initial counseling session? Moreover, cultural attitudes to medical knowledge and family structures in different ethnic backgrounds are an important part of this equation. "The availability of genetic tests is creating advances and dilemmas for the standard of care in the diagnosis and prediction of potential hereditary diseases…. In most situations, discussions about genetic testing in families with sporadic cases of dementia should be delayed until the utility and benefits can be established and a long-term relationship with the patient and family can be developed," the authors write (Goldman and Hou, 2004).

  A further facet of this issue emerges from the story of a patient with Creutzfelt-Jakob disease (CJD), a degenerative prion disorder in the group of diseases that includes "mad cow" disease. It poses similar genetic dilemmas. At 53, the man appeared to have CJD. His 23-year-old daughter signed an informed consent for his participation in a research study that included genetic testing, and a known mutation in his prion gene turned up. The daughter was stunned to learn, rather abruptly, that her own odds of getting CJD were 50-50. To her knowledge, her family had never experienced neurologic disease before. The medical team offered predictive testing to the daughter in the context of a full counseling protocol, and after several weeks of consideration, she opted in. The point here is that this woman did not have a chance to make an informed decision about whether she wanted to know her father's status or not. It just came out in the course of reviewing his records.

  The third story tells of a 55-year-old woman with CJD that seemed sporadic. A local physician sent blood off for genetic testing without prior genetic counseling with the patient or her husband. Before the family found out this test result, the patient happened to be referred to UCSF for a research study. In its course, the patient, his wife, and their adult daughter and son-in-law entered genetic counseling, which opened their eyes to the full import of knowing the patient's genetic status. The daughter was expecting a baby, and the family decided they did not want to know the patient's genetic result at that time. The family then had to instruct their local physician to withhold the genetic result. (Sometimes people get tested twice: the local physician had ordered the test clinically, whereas the researchers would have performed it in an academic setting without disclosing its result to the family.)

  The fourth case showcased how genetic testing can fray the family fabric. A 54-year-old patient came in with his wife, adult son, and daughter. He received a diagnosis of probable AD. The family knew of no relatives with AD but chose to have the father tested because his symptoms started at 48. When the family learned he did indeed have a presenilin mutation, disagreements arose. This is a problem with diagnostic testing: it puts on the table the option of predictive testing for relatives, and family dynamics around that can be fraught. In this case, the parents were upset at the risk for their children and did not want them tested. In individual counseling sessions, the daughter, too, said she did not want to know her status, nor her brother's. Yet the son insisted he wanted to find out. He entered the full genetic counseling program but eventually withdrew under pressure from his family (Goldman et al., 2004). Other investigators, too, have reported that many families communicate poorly about their genetic burden and tend to avoid the issue rather than face it openly. The lesson from these four case vignettes is that families who have for years, or generations even, watched AD claim loved ones and cared for them, tend to be aware that something is up genetically. They are somewhat prepared to handle concrete information about it and tend to welcome the genetic validation after years of torment. By contrast, relatives of sporadic cases tend to be unprepared for the idea of an incurable brain disease in their family. For them the realization can bring more shock than relief. To minimize distress and family strife, these families must be led to this awareness gingerly, in a gradual education process that includes the chance to opt out at every step. All things considered, the San Francisco team in their publications encouraged clinicians to consider a genetic reason for isolated cases of EOAD, but to proceed in conjunction with genetic counseling.

  If you are a family member, you can find a genetic counselor by contacting your nearest ADRC. If you live too far from an ADRC, you may want to search the online directory of the National Society of Genetic Counselors, accessed October 2006.

  Predictive Testing 
  When a specific mutation is known in a given family, at-risk siblings, adult children, or other relatives who request this information can find out whether they have inherited the mutation and will most likely get the disease, as well. About 10 percent of eligible relatives choose to do so.

  In practice, what happens more frequently is that a worried relative requests genetic testing, but the clinician cannot offer it because no DNA test has been done on a confirmed case of AD in the family. Or it still cannot be done because all relatives with AD have died, refused testing, or their adult children don't want it. In those cases, the worried relative can benefit from genetic counseling. Delaying predictive testing until a confirmed mutation is known is important, because even if the worried relative had available testing done at considerable cost for all known mutations, a negative result would be inconclusive. Why? The worried relative might indeed have escaped an AD mutation that runs in the family. However, it is just as possible that an as-yet unknown mutation in APP or presenilin, or a different gene altogether, is responsible for AD in the family, in which case the worried person might still carry it. In short, the result of a predictive DNA test in these situations is so vague and incomplete that clinicians discourage it. In other cases, diagnostic DNA testing of the affected family member is being done and comes back negative. Here, too, clinicians cannot offer predictive testing to at-risk relatives because they do not have a mutation in hand. Occasionally, a young adult requests predictive testing who has a 25 percent known risk, that is, someone with an affected grandparent. This situation is difficult when the intervening parent is still alive and does not want to get tested, or is unaware of the child's intention. A positive test on a grandchild would reveal the carrier status of this parent.

  Views about predictive testing range widely. For some, the test's value hinges entirely on a cure. That is true for both at-risk relatives and health care providers. For others, simply knowing their future is reason enough to find out. They find the uncertainty unbearable. In between these two extremes, some at-risk relatives say that the availability of a clinical trial, or a focused research study that gives them a practical opportunity to fight the disease, would be a factor in their decision about whether to get tested. But wherever they are coming from, all involved agree that knowing one's genetic status has implications for the entire family. The decision is intensely personal, and must be made freely, after a period of education and reflection.

  Lessons from Huntington Disease 
  Predictive testing is a fundamentally different use of medical services from what we have been accustomed to through the centuries, because it deals with the future of a currently healthy person, not with the present health status of a sick person. Its use in Alzheimer disease has taken guidance from the model established some years earlier for Huntington disease (HD). A gene for this disease was first localized on chromosome 4 in 1983, and an HD mutation identified in 1993. The discovery of the gene prompted international collaboration for the development of counseling and testing protocols. The experience with these protocols has set the standard for DNA testing in other neurogenetic diseases, such as prion diseases and familial ALS, as well as in heritable cancers of the breast, ovaries, and colon and indeed for serious genetic diseases in general. In the UK, genetics centers first began offering HD presymptomatic testing in 1987 even before the gene itself became known, initially in research settings but subsequently also as a clinical service. Some countries have instituted predictive HD testing in medical genetics centers nationwide. For example, Great Britain offers HD testing at genetic centers of that country's National Health Service, and the universal health systems of other EU nations generally cover the test, too. This record has made it possible for researchers to analyze thousands of such completed tests in various countries around the world (e.g., Harper et al., 2000;Creighton et al., 2003).

  The discovery of the HD gene triggered an intense international debate of the potential benefits versus the ethical and psychological pitfalls of predictive testing. It culminated in guidelines put forward by the International Huntington Association and the World Federation of Neurology. Practitioners across the world have since followed these guidelines broadly and tailored them to the specific resources of their clinical center and to the needs of their respective patient population. The guidelines all emphasize a multidisciplinary protocol that engages a neurologist, clinical geneticist, psychiatrist/psychologist, genetic counselor, and social worker in a series of interactions with the prospective test-taker that begins with several meetings before the test and ends with long-term support afterward (see sample protocol). The guidelines unanimously recommend that parents not test their underage children.

  Some patients, and neurologists, as well, have questioned whether the extensive counseling protocol developed for HD is really necessary for AD. It is expensive, and some families resent having to undergo the intensive process. Leaders of the field disagree about this. Some believe that an emotionally stable person who is in the ongoing care of a neurologist specializing in AD can handle the testing and its aftermath adequately, and examples of that certainly exist. The neurogeneticist Thomas Bird wrote in 1999 that, "some individuals resent what they view as an excessively conservative, paternalistic attitude of academic centers toward testing." Others feel that, if anything, the protocol may be even more important in AD. Why? For one, interpreting genetic test results in AD is more intricate and complicated than in HD. Hundreds of mutations in three different genes cause eFAD, and they must be expertly recognized among the harmless gene variants that also occur in those genes. For another, available genetic testing fails to identify a gene mutation in a substantial fraction of families with autosomal-dominant AD. Furthermore, the age at which a carrier will likely become ill with AD is influenced by what ApoE variant and possibly other genes they have inherited alongside APP or presenilin, but there is no straightforward metric to determine this. By contrast, in HD, the age at which disease strikes can be deduced with some certainty from the nature of the person's mutation and becomes part of the test result. In AD, if relatives prove to be genetically at high risk, the personal consequences for them and their entire family are devastating. Finally, if they have escaped the gene that runs in their family, they are not completely in the clear. They still face the average population risk of developing AD, which rises considerably with age, whereas for HD it is infinitesimal.

  Individual differences in how much family support a person will be able to lean on further reinforce the need for counseling and added support from a social worker for some people. At times, medical teams will simplify the protocol requirements if particular relatives find them too onerous and seem able to cope. All clinicians, however, tend to recommend that people ponder whether they really want to go forward with testing for at least several weeks, or months, after an initial counseling session with either their doctor or a professional counselor. In short, avoid rash decisions. Involve your spouse/partner. You and your loved ones must live with this knowledge for the rest of your lives.

  What Have We Learned from More Than a Decade of Experience with Huntington Genetic Testing? 
  Several research groups have attempted to answer this question. Across the world, about 5 to 20 percent of eligible people decide to take the HD test. This low number surprised medical geneticists; they had expected a great majority of people would demand it. For heritable cancers, the numbers go as high as 50 to even 95 percent. The difference, of course, is that HD has no cure, whereas a positive result for a colon cancer gene means the patient has a real chance to save his/her life with annual screening for polyps and surgery when necessary. For this type of cancer, doctors strike a much more directional tone. Most strongly encourage patients to get tested. By contrast, for HD and AD, geneticists and counselors abide by a strict code of not influencing the individual. Some even have said that the way they present the issues may dissuade people. The high rate of testing for cancer genes portends what is likely to happen to demand for eFAD testing as soon as effective treatments become available.

  Psychological studies of at-risk relatives of families with HD suggest that the personality and coping style of individual people seem important in determining who decides to get tested. Psychologists describe people who choose testing as having high "ego-strength," as being mentally resourceful, as using active problem-solving strategies, and as reaching out for social support more than does the average population (e.g., Evers-Kiebooms et al., 2000). They describe people who decline testing as typically falling into one group who deliberately makes this choice and copes well with uncertainty, and a second group characterized by avoidance, for whom discussing the problem is effectively taboo.

  Research by multiple groups on how people fared after their results found that both carriers and non-carriers felt relief. Carriers were depressed and anxious for some months after learning their status but returned to their prior mental state within a year (they had already come into testing with intense worry). After 3 years, they had adjusted to facing reality while living their lives as normally as possible. They were burdened by worries about the future, but overall, the certainty that they carry the gene had little impact on depression in the long term. In essence, for carriers, over time knowing was no better, or worse, than not knowing and living in fear. Some non-carriers reacted with survivor guilt, and others found it challenging to adjust to the unexpected future of a normal lifespan without HD. On average, however, knowing was better for non-carriers than not knowing. These are summary statements from published group analyses; it's important to remember that a given person's reaction to predictive genetic information is highly individualistic.

  Several studies looked at psychiatric outcomes. Any practitioner's worst nightmare is that carriers become despondent enough to be hospitalized, or even commit suicide. The rate of catastrophic consequences of HD testing was low, but they did occur (see sidebar:The Specter of Suicide). Some investigators believe that extreme outcomes have remained exceptionally rare precisely because the field has worked out a careful approach to predictive testing from the outset.

  One of the main reasons why people over the years have chosen predictive testing for HD, besides relieving uncertainty, is family planning (see sidebar: Taking the Plunge into Genetic Testing). Follow-up studies 1 to 3 years after the test result show that the test has indeed influenced people's reproductive decisions. A minority of affected couples, where one partner had learned (s)he carried an HD mutation, chose prenatal diagnosis, and in cases where the fetus proved to have inherited the mutation, some couples subsequently terminated a desired pregnancy. The emotional suffering in this situation is great. More recently, a small number of affected couples have pursued pre-implantation genetic diagnosis; in this procedure, too, the psychological as well as the financial cost is high, clinicians note. Interestingly, a significant proportion of people who choose DNA testing for HD are already past the mean age of onset for their family (in HD, age of onset can range even more widely than in eFAD.) They report wanting to find out for the benefit of their adult children who approach the age of parenthood and want genetic information to guide their family planning (e.g., Harper et al., 2000).

  Taken together, the experience with HD testing to date has been positive (e.g., Broadstock et al., 2000). Caveats are real and important. "An adequate and systemic multidisciplinary approach, as well as ongoing education of professionals and of the general public are essential to avoid pitfalls," writes Belgian psychosocial researcher Gerry Evers-Kiebooms. Whether predictive testing truly reduces suffering and improves the quality of life of families with HD will remain an open question until data on longer-term outcomes are available. Observations at the 5-year follow-up point are good (Almqvist et al., 2003), but even past that, the particular worry remains of how carriers will fare as they approach the age of symptom onset. Until then, families and their clinical teams should proceed with care (Evers-Kiebooms et al., 2000).

  Alzheimer Disease Predictive Testing: Research Summaries 
  It is difficult to know how many healthy people to date have used this service. In this section, we have collected summaries of cases that have made it into the scientific literature. (For personal stories, see interviews; soon to come.) Each story highlights a different aspect of this complex issue. Together, they reflect the universal human truth that variation marks most everything we do, including genetic testing. No two families are the same. Genetic testing is an individual choice, and how people adjust their relationships and family planning afterward is even more inscrutably personal. Even so, common themes emerge, and there is a cautious consensus that DNA testing for eFAD can be done safely if it's done right.

  The Swedish Experience: The first predictive testing for eFAD was "very discouraging," recalls Lars Lannfelt, a highly regarded Swedish neurologist and researcher. In 1992, Lannfelt's team had discovered a mutation in the APP gene in two Swedish families that go back to the same eighteenth century ancestors. This genetic flaw has since become known as the "Swedish" mutation and remains the basis of numerous AD research studies. By 1995, Lannfelt's group described in the American Journal of Human Genetics what happened when descendants from this family demanded to find out if they would get AD. The researchers told about 60 at-risk descendants of the newly available testing option, 18 chose to participate in a research study without disclosure, and three wanted to know their status. Each of the three nominally had 50 percent risk because one of their parents had been affected, but two were already older than when AD usually struck in their family, so their chances were better. One, however, was younger. The three underwent genetic counseling as developed for HD. They came for extensive discussion at least twice before having their blood drawn, spaced 3 months apart so they would be able to think it through. The two older relatives proved to be negative for the APP mutation and were relieved without suffering long-term psychological consequences. The younger one was a different story. That person was positive. (S)he became extremely sorrowful, even suicidal, for a month, needed close monitoring and support at the clinic, and for many months afterward continued to wrestle with hopelessness and severe depression. One year later, the person was back at work and coping well. The researchers were distressed that despite following a cautious and elaborate protocol, they had in effect plunged a functional person into despair (from Lannfelt et al., 1995). This person has since died from AD, and Lannfelt's group followed him/her to the end, yet to this day Lannfelt feels that his team would have served the person better with counseling but without genetic testing.

  As early as 1995, these authors foresaw that predictive testing for AD might well become widespread, and they urged that it be handled with "exquisite care." This group sets the bar high. They feel that only a preventive treatment can truly counterbalance the uncertain impact of knowing that one will lose one's mind to dementia somewhere in the middle of life. In 2006, the Swedish investigators continue to discourage predictive testing. By e-mail, Lannfelt wrote "We advise against genetic testing…. To be aware of a mutant genotype and know there are no treatment possibilities puts you in a terrible situation…. However, once a promising strategy is available, these patients will be an obvious target group."

  The Indianapolis Experience: Five years after Lannfelt's report, an extraordinary set of papers on predictive testing for eFAD appeared in the Journal of Genetic Counseling. Kimberly Quaid, a research clinician in genetic testing, and colleagues including Martin Farlow and Bernardino Ghetti, the head of the Indiana University Alzheimer Disease Research Center (ADRC), narrated their perspective on testing a family alongside a first-person account by a mutation carrier in that family. The package highlighted the tension that can ensue when a provider's cautious protocol gets pitted against the complexities of family dynamics on the one hand and against the demands of informed, impatient family members on the other.

  The story began when a 42-year-old woman was referred to the Indiana University Alzheimer Disease Research Center with a history of memory loss and disorganized behavior. Her father had shown signs of dementia before dying young from a heart attack, and a younger sibling had cognitive deficits, too. Genotyping of the woman's blood identified a new mutation in the APP gene, and made predictive testing for her five siblings possible.

  The family was invited to meet the ADRC clinical staff; two thirds attended. The researchers laid out the siblings' options: do nothing, participate in clinical research with the separate option of learning their results, and/or learn their genetic risk. Grant funds were available to cover travel and lodging expenses for research participation and for part, but not all, of the predictive testing regime. This became a source of tension later on. At this initial meeting, one sibling stated that all would want to participate to the full extent offered, including predictive testing. The same sibling called back to confirm that all siblings wanted this. In the course of the process, however, cracks among the siblings emerged.

  First, one turned out to already show signs of cognitive impairment in spite of the siblings' denial, and for this person the genetics would confirm a clinical diagnosis. The clinicians found themselves in the position of having to diagnose this person appropriately and privately with only the spouse present, while at the same time keeping all the other siblings within the predictive testing protocol, and the negotiations around this were fraught with tension. Second, another sibling later decided that (s)he did not wish to learn his/her genetic status at all, or even be followed regularly. (S)he needed to fend off family pressure to get tested as a group.

  Communications between the center staff and the sibling who acted as a spokesperson and made joint decisions, as well as the other siblings, were strained at times. One sibling's tendency to view all siblings as a group conflicted with the clinician's approach of treating each as an individual, preferably as a unit with his/her respective spouse. Several siblings saw AD as a family problem and wanted to take the required steps together. Yet each was married, at a different stage of their lives, and they eventually came to see that they needed to include their spouse in the counseling, and receive the test result with only the spouse in the room.

  Moreover, some of the siblings balked at the genetic counseling protocol the center insisted on following. They felt they knew enough to "just find out." They objected to being cajoled into a process that required repeated air travel at their own expense. They felt patronized. For their part, the clinical staff held firm on their professional responsibility to protect the siblings from unforeseen implications of knowing their genetic status, particularly since the siblings were pioneering a new era in genetic medicine. Predictive testing in AD was still so rare. It was the first time this ADRC performed it and they wanted to do it right. Negotiations about cost, individual exceptions, and special arrangements ensued. When the director of the genetic testing program sent a detailed letter to every sibling, laying out core requirements and the reasoning behind it, both sides eventually found a compromise. One sibling underwent counseling at a qualified center near his/her current home, and flew to Indiana to receive the test result.

  One of the three carried the mutation. News of the result met "tearful stoicism" and "stunned silence" on the part of the sibling and the spouse, according to the report. The siblings called and e-mailed with follow-up questions but did not take advantage of the center's offer to schedule further sessions. One year later, they returned for follow-up. The sibling who has AD had progressed mildly, the sibling who had rejected testing still did, and the sibling who turned out positive was pursuing pre-implantation genetic diagnosis in order to have a child who would not carry the mutation. The family remains interested in research participation and came through this crisis intact, indeed closer.

  The path toward this positive overall outcome was draining for all, the authors note, and they offer some lessons for others in the same situation. Direct communication between center staff and each individual family member all along the process is critically important to prevent misunderstandings among siblings and to establish the authority of the clinical director. The underlying challenge of balancing the patient's autonomy with good clinical practice runs through the literature on genetic testing for other diseases, as well. Patients feel entitled to "their" information, and the provider insists on providing it in an appropriate way. Open discussion allows the patient and the provider to reach a mutual understanding. Personal follow-up calls from the provider to each affected family member, just to check in on how each is doing, can further improve relations with the patients and build an ongoing clinical relationship (from Quaid et al., 2000, J. Genetic Counseling, Vol.9[4]).

  The Seattle Experience: By 2001, enough people had chosen predictive eFAD testing and agreed to continue with research that it became possible to advance from anecdotal, single-family reports to studying a group and distilling some general information. The first such study came from the group of Thomas Bird, the neurogeneticist at the University of Washington in Seattle. Bird has studied Huntington disease—considered a model for eFAD—in addition to Alzheimer disease and similar disorders that fall into a group called frontotemporal dementias. The Seattle clinicians followed presymptomatic carriers for up to 3 years to see how knowing had affected them psychologically, how it had shaped their lives, and whether they regretted having taken the test. The clinicians shared their data with the research community in a report in the journal Archives of Neurology. Their overall impression was one of cautious optimism that DNA testing could be worth it, but important caveats turned up as well.

  The study included at-risk descendants from families with eFAD and a related disorder called FTDP-17. Of 251 candidates for testing (meaning each had an affected parent and 50 percent risk), 58 people first expressed interest and 21 eventually opted to go through with it. Sixteen of those were at risk for eFAD, five for FTDP-17, and six of them proved to already be in early stages of disease. These six and their families considered the DNA test helpful in clarifying the situation and helping them plan their lives. Fourteen asymptomatic people were in their thirties and forties. Two had the test done but postponed hearing the results for 2 years. Six were positive for their respective family's mutation. One refused all further contact and the researchers do not know how (s)he is faring. Most others, when asked afterwards if they would recommend DNA testing, said they would leave it up to the individual, but none said they would advise against it. People with positive results were disappointed, but tended to say they were glad to know and to have clearer choices for planning their future. As expected, non-carriers were grateful and relieved. Unexpectedly, however, the follow-up visits and exams showed that some of them continued to wrestle psychologically with AD. Some worried about it all the time even years after the test. Perhaps this should surprise no one as the disease, after all, continued to burden their lives and claim living relatives. The DNA test itself did not cause depression. One person who was depressed had been depressed before the test and turned out to be negative for the disease. The test did not relieve depression, either. Four people scored high on a measure of avoidance, that is, they protected themselves by trying not to let AD enter their minds. Two of them were positive; two others had postponed hearing their results. There were no psychiatric hospitalizations or suicide attempts.

  Curiously, some people have strong feelings about how their test will turn out. One person, upon finding (s)he was positive, said the result "confirmed what I thought." Another had gone through counseling and the blood draw but then felt ambivalent and at the last minute delayed hearing the result to prepare emotionally. Then, when the result proved negative, (s)he said, "I was so convinced I carried the abnormal gene that I have to retrain my mind to think differently…. Now I have a future."

  With regard to concerns about social discrimination, the researchers to date have not heard of a person with a positive result having lost employment or been denied insurance, though this concern is certainly real. The test clearly ripples through personal lives, but not in any particular way. One presymptomatic carrier went through marital separation, one non-carrier went through with a divorce and another with a second marriage; all three said the test influenced their decisions. Likewise, family planning changed but not in any one direction. After finding out, one carrier decided against a second pregnancy but another carrier had a first child. A third person had a first child during the self-imposed hiatus between blood draw and finding out (from Steinbart et al., 2001 and personal interview).

  This is still a small study, and more data are needed before one can generalize more broadly. Moreover, true long-term follow-up data are still not in. What happens 5 years out? Ten years later? Do people who know their genetics react differently as the first disturbing symptoms appear?

  The Barcelona Experience: Recently, Spanish researchers have reported similar results with a smaller group of nine at-risk relatives from families in the Catalan capital of Barcelona. Ranging in age from their twenties to their forties, they came from a family with eFAD, a family with a frontotemporal dementia, and a family stricken with a rare fatal insomnia that is caused by a mutation in the same gene that is also associated with Creutzfeldt-Jakob disease and the cattle disease BSE. (While these diseases are different, they are incurable and raise nearly identical issues of genetic testing.) The relatives chose predictive testing because they wanted to use early treatment in the future. They also cited the anxiety of not knowing their future, the need to plan their families, and the wish to tell their children.

  Five descendants from the family with eFAD chose testing. Two who turned out not to carry the mutation were relieved, though one doubted the test: "I had always been sure that I had the mutation." One decided to have another child. Two relatives had the mutation, and for a fifth the positive test became part of the diagnosis because (s)he was already symptomatic for AD at 38. The two carriers showed sorrow when they heard of their results. One developed intense anxiety in the next few weeks but returned to normal in the post-test follow-up program. (S)he worked less, spent more time with family and hobbies, and decided to tell the children when symptoms would begin. The other carrier had come to the study with high anxiety, which abated when (s)he found out. (S)he stayed emotionally stable, made no major lifestyle changes, and said that relief of a nagging uncertainty was the biggest benefit. Both carriers decided to have no more children and regularly visit a neurologist for follow-up. The relative from the frontotemporal dementia family took a similar course. Upon learning (s)he carries the gene, previously high anxiety eased and (s)he regularly visits a psychiatrist as part of the follow-up. The descendants of the third family all escaped the mutation. Two said they had expected "the worst." None of the non-carriers developed survivor's guilt.

  Unlike U.S. families, the Spanish families did not cite concerns about insurance, largely because Spain has universal national health care, as do other countries in the European Union. The Spanish cited no concerns about employment discrimination, nor did the carriers make significant changes in these areas. There were also no changes in their marriages. The Barcelona study found no depression as a consequence of DNA testing.

  Like most academic investigators in this area, the Spanish scientists used an extensive counseling protocol modeled on HD and emphasize that predictive testing should always be embedded in careful pre-test education and post-test follow-up. Their pre-test program called for the clinical staff (neurologist, psychologist, psychiatrist) to discuss every case and, based on their assessment of the person's ability to cope with a positive result, recommend that the person proceed or not. However, even in cases where the professionals felt the psychological damage might outweigh benefits, they did not deny that person the right to know if (s)he insisted. The patient's autonomy trumped the provider's caution. The Barcelona participants valued the testing program for providing sought-after information, relieving uncertainty, and offering long-term care after the test (Molinuevo et al., 2005).

  In summary, both the Seattle and Barcelona clinicians stated that predictive testing for eFAD can be done safely. They noted, however, that their participants were a self-selected group of people who insisted on having access to genetic testing. This drive to self-determination might be a good indicator for a person's ability to deal with the results in the aftermath. In other words, much as this determination should be honored, others who feel ambivalent about predictive testing must never be pressured into it. On balance, then, the literature available to date tends to agree that DNA testing for eFAD and FTD can be valuable provided it happens in the appropriate context. The risks are real and not fully understood yet.

  • • In which setting will genetic testing occur—clinical or research?
    • Genetic counseling is a benefit: is it offered?
    • What sort of counseling is required?
    • What is optional?
    • Who will notify me of the test results? When? How?
    • Is predictive testing (for relatives without symptoms) available?
    • Will the information be in my medical record?
    • What are my options for participating in research? What procedures would I go through? What knowledge do the researchers hope to gain?
    • What information will I receive as a consequence of my participation?
    • What information will not be given to me?
    • How will I receive this information?
    • Who pays for different parts of the process?
    • Will the test confirm or rule out a diagnosis?
    • Does the patient/family want to know?
    • Will the test result change treatment at all?
    • Is there a reasonable chance of finding a mutation in this family?
    • Who decides in this family? The patient, or relatives?
    • Does the patient understand the implications and limitations of the test?
    • Do first-degree relatives agree about moving forward?
    • What does it cost? Does the family's insurance pay? Can the family afford the test?
    • Does the provider understand the family dynamics?
    • Is there support for potential psychological consequences of the result?
    • What is stronger in the case at hand: the duty to warn or the right not to know?
    • Does the family have an ongoing relationship with the clinical team?

NEXT: The Specter of Suicide