At the 9th International Conference on Frontotemporal Dementia (ICFTD), held October 23 to 25 in Vancouver, British Columbia, Canada, the biggest news splash happened when the first international cohort study of presymptomatic FTD mutation carriers presented its initial results, on brain volume and neuropsychological tests. In a nutshell: It appears as if each causative FTLD gene generates its own particular sequence of events as it ravages the brain, but overall atrophy eats away the insula first. Folded beneath the fronto-lateral surface of the brain, this area of the cerebral cortex is a connectivity hub of the salient network, involved in the regulation of consciousness and emotion, and the body’s homeostasis. The news came as validation but not surprise, as the insula has lit up previously in small studies of presymptomatic FTD gene carriers, as well as in studies of early symptomatic FTD. (Thirty presentations at ICFTD mentioned the insula.) In the paper-and-pencil department, deficits in behavior and cognition began to show five to 10 years after brain atrophy, a few years prior to the estimated age of onset.
Jonathan Rohrer of University College, London, presented results of a first data cut in the Genetic FTD Initiative (GENFI), a 13-center study of carriers of pathogenic mutations in tau, progranulin, and C9ORF72. The work generated respect throughout the conference. “This study is super-important,” said Markus Otto of the University of Ulm, who coordinates a large cohort of symptomatic FTD in Germany (see Part 3 of this series). “GENFI did spectacular work,” said Adam Boxer of the University of California, San Francisco, whose center just nabbed NIH funding to do much the same thing in the United States and other Canadian sites (see Part 4). “Impressive data,” ICFTD co-host Howard Feldman said in his concluding remarks.
Set up in 2011, GENFI is a consortium of centers in Canada, France, Germany, Italy, Portugal, Spain, Sweden, the Netherlands, and the United Kingdom. Martin Rossor and Rohrer at UCL jointly coordinate the network. GENFI is similar to DIAN, the Dominantly Inherited Alzheimer's Network (Nov 2008 news story), in that it aims to chart the evolution of FTD in the decade or more before a person shows symptoms. The idea is to use the autosomal-dominant forms, in which researchers know that a mutation carrier will develop the disease, to learn whether there is a measurable, stereotypical sequence of brain changes that precede the symptomatic phase. If there is an ordered pattern of biomarker change, then these stages can be characterized by comparing mutation carriers to their non-carrying siblings in longitudinal natural history studies. In so doing, GENFI can—as DIAN did—establish a so-called “readiness” cohort of people whose presymptomatic biomarker changes are known and who want to enter therapeutic intervention trials before large swaths of their brain are damaged beyond repair. “Our goal is to create a global initiative for therapeutic trials. Our ultimate goal is to provide a disease-modifying treatment for genetic FTD,” Rohrer said.
For the first two years, the investigators’ goal was to see if they could pull off GENFI. Would they be able to recruit enough participants to build a sizable cohort? After all, FTD is a rare, heterogeneous disease that inflicts both daily dysfunction and fear of inheritance on affected families. Would they agree on standardized assessments? Could they harmonize sample collection and data handling in a uniform way across centers, countries, and languages? Researchers like to do things their own way, but in consortia they have to give up local control on many issues in service of a greater good. They have to adhere exactly to centralized methods for cerebrospinal fluid processing, and perhaps forgo their favorite imaging techniques for ones that hold up more robustly in multicenter settings. They may have to map their data to new standards, fork over results to a common database, and, on manuscripts, drop their name into a large collective of authors. It worked for GENFI. To date, GENFI has enrolled 336 people and 99 have returned for their first follow-up visit. Setting up a standardized protocol was key, Rohrer said.
GENFI welcomes asymptomatic carriers of mutations in tau, progranulin, and the C9ORF72 gene, as well as their mildly symptomatic and mutation-free siblings. Most people prefer not to know their mutation status, Rohrer said. GENFI protects their privacy by having “genetic guardians” enter the data into a database so that clinicians who interact with the participants do not see it and cannot accidentally deduce or disclose whether a given person has the mutation.
Like DIAN, GENFI asks a lot of its participants. They sit for a long battery of tests measuring their social cognition, executive function, memory and other domains of mental function. They submit to lumbar punctures for CSF sampling, blood draws for plasma biomarker research, and lie in a scanner for volumetric, functional, and structural connectivity MRI. Their least favorite assessment? “The neuropsych tests. They do not mind the scanner or the spinal tap,” Rohrer said. The researchers try to help by explaining that the tests are hard, designed to trip up healthy people. Still, these test sessions can be daunting to a person who has grown up watching FTD in a parent and now confronts his or her own risk head-on during each visit.
In September 2013, GENFI researchers froze the data to generate a first batch for analysis. By that time, 220 people had finished their first visit. Of those, 118 are from progranulin, 44 from tau, and 58 from C9ORF72 families. GENFI started just before the C9ORF72 repeat expansion was published in September 2011, hence this first cut reflects a preponderance of people with progranulin mutations. More C9ORF72 families have since joined, Rohrer said. Forty people are symptomatic; 45 are presymptomatic carriers of progranulin, 15 of tau, and 18 of C9ORF72 mutations; 102 are non-carriers. “We would expect in autosomal-dominant disease a 50-50 split between carriers and non-carriers, but we have over-representation of the latter,” Rohrer said. He does not know why, but Boxer, who studies presymptomatic FTD at the University of California, San Francisco, said the same is true at his site.
Comparing carriers and non-carriers, the scientists looked for links between a given measure and the expected time of symptom onset in each carrier. In GENFI this is calculated as the difference between the participant’s age and the average age of onset in their family.
When did the first cognitive signals appear? Carriers and non-carriers started to diverge around five years before estimated onset, though the specific tests were not the same from gene to gene. For tau and C9ORF72, the Cambridge Behavioral Inventory CBI-R showed the first differences; for progranulin, it was executive function and working memory.
Before even these subtle signs appear, the brain appears to have been shrinking for a decade. GENFI measures the volume of frontal, temporal, parietal, occipital, insular, and cingulate cortex, as well as the subcortical areas of the hippocampus, amygdala, thalamus, and striatum. The first difference between carriers and non-carriers was in the insula, at 15 to 10 years prior to estimated onset. In roughly five-year steps, the temporal lobe began to atrophy, then the frontal lobe, then the parietal lobe. The cingulate cortex changed around onset of symptoms and the occipital lobe soon after, Rohrer reported.
Unlike in Alzheimer’s, where three autosomal-dominant genes converge on APP processing, the three main FTD genes are not known to act on a common pathway, and no single pathogenic hypothesis has been formulated. GENFI analyzes all results separately by gene, and is finding a unique order of events for each. At ICFTD, Rohrer showed that atrophy in tau carriers appears to start in the hippocampus 20 years before onset, spreading to the amygdala and temporal lobe. Atrophy in progranulin carriers is first seen in the insula at minus 15 years, then in the temporal and later the parietal lobe; the first subcortical target is the striatum. In C9ORF72 carriers, early atrophy is most widespread, cropping up 20 years before estimated onset in the insular and parietal cortex, hippocampus, and thalamus. In this latter region, the loss of volume is most pronounced at that early stage, Rohrer said. These numbers and sequences will be refined as data from the second cut, taken in October 2014 on 336 participants, are analyzed.
The first cut of follow-up data, on 99 participants, came in for analysis just before ICFTD. A quick peek suggests differences in how fast the brain as a whole shrank in the year between non-carriers (0.1 percent), presymptomatic carriers (0.6), and symptomatic carriers (1.4), Rohrer said. Most of that analysis remains to be done.
Several scientists noted that the estimated of date of onset is just that, particularly for the C9ORF72 repeat expansion whose age at onset ranges widely. At ICFTD, Christine van Broeckhoven, University of Antwerp, Belgium, reported “anticipation” in some families with C9ORF72 hexanucleotide repeat mutations, confirming previous reports (e.g., Chiò et al., 2012). In this strange phenomenon known from the trinucleotide-repeat disease Huntington’s, subsequent generations become symptomatic at younger ages than their parents. Still, in the DIAN cohort, careful analysis of age of onset has shown that it is more predictable, within a narrow range of a few years, than previously thought for most mutations in a given family (Ryman et al., 2014).
GENFI’s suggested sequence of overall changes jibes with previous studies. The insula is a hub of the salience network that is known to be important in FTD, and subtle presymptomatic changes in neuropsychometry and brain imaging are known from isolated cases and smaller series (e.g., Jannssen et al., 2005).
In fact, at ICFTD Lize Jiskoot from Erasmus Medical Center in Rotterdam, Netherlands, presented follow-up data from such a study (Dopper et al., 2013). Liskoot said that two years after baseline, the 39 asymptomatic mutation carriers in her group had not improved on the neuropsych tests from one sitting to the next; unlike the non-carriers, the carriers had no practice effect and did not learn. On functional MRI, the carriers had a decline in the insula; across different imaging techniques used, they declined in frontal areas of the salience network and white-matter tracts connecting those areas, such as the uncinate fasciculus. Jiskoot’s site participates in GENFI.
Also in Vancouver, Emma Dowds of the University of British Columbia presented her group’s finding that eight presymptomatic carriers of a C9ORF72 repeat expansion had a thinner cortex than their unaffected relatives in the right insula and other frontotemporal regions; this was before their cognition declined. UBC is a site in the upcoming North American LEFFTDS cohort study (see Part 4 of this series). For their part, Suzee Lee and colleagues from UCSF showed reduced gray matter in a range of cortical and subcortical regions in C9ORF72 expansion carriers on their poster at ICFTD, but they noted that these differences might be developmental rather than degenerative. For a review of previous work in presymptomatic FTD, see Rohrer et al., 2013.
The next phase of GENFI will start in early 2015 as a five-year study of 600 participants. It will use largely the same protocol plus tau PET. Additional sites in Spain, France, and Germany have joined. To date, financial support for GENFI2 remains piecemeal, with sites scraping together local funding but no overarching grant approved yet.
GENFI2 aims to flesh out the emerging picture with much more data. The scientists want to address the twin questions of when during the presymptomatic decade might be the right time to intervene therapeutically, and how to measure whether the treatment works. To help plan future presymptomatic trials in FTD, Rohrer obtained EU funding for a working group called PreNI. It convenes experts from other neurodegenerative diseases who are conducting similar observational studies and have based therapy trials on them, such as TrackHD. PreNI includes U.S. advisers from DIAN. “We can learn lots from studies that have gone before us,” said Rohrer. The group will meet for the first time this month in London.
Perhaps the most difficult part of presymptomatic cohort studies is to find participants. Once a person has decided to enroll, they tend to be keen to enter a treatment trial, and many participants find purpose in contributing to research. In DIAN, which began in 2008, about 40 percent of the more than 400 participants have since decided to find out their genetic status, too, according to Randall Bateman of Washington University, St. Louis. But not everyone jumps at the chance. Just days before ICFTD, Rohrer said, he had met with a C9ORF72 family and explained GENFI to them, only to hear the day before his talk that they had politely declined to join. “They have grown up with FTD and know it is in the family. Still, they do not want to face it every year when they come to the visit,” said Rohrer. Working with carriers is new territory for neurologists. Rather than seeing symptomatic patients who come to their clinic, they have to cultivate an atmosphere of trust with younger, still-healthy people who harbor fears of an uncertain future.
In Germany, Felix Mueller-Sarnowski, a neurologist at Ludwig-Maximilians-University Munich, crisscrosses the country by train paying home visits to far-flung families with suspected autosomal-dominant Alzheimer’s disease in an effort to build that trust for the German DIAN study. “In the process, I also meet families with FTD. We cannot expect those families to open up easily. We have to support them,” Mueller-Sarnowski told Alzforum. In the United Kingdom, GENFI has set up a local “arms-length” support group, where family members can meet and talk about their experiences with FTD and with GENFI without a GENFI physician listening in. Rohrer jointly edits the website FTD talk www.ftdtalk.org to inform patients and families about FTD research.
In many instances, presymptomatic carriers come to the attention of scientists when their symptomatic relative sees a specialist who notices an autosomal-dominant inheritance pattern in the family history and invites the relatives to learn more about research opportunities. While done locally, this is beginning to coalesce in a more organized way as cohort studies of symptomatic FTD are coordinating with presymptomatic studies nationally and even internationally. In the United States, a push for such twin studies got an infusion of federal funding (see Part 4 of this series). In Germany, clinicians studying a large symptomatic cohort are now reaching out to its autosomal-dominant member families about GENFI2 (see Part 3).—Gabrielle Strobel
- German Network of 700 FTLD Patients Presents Baseline Data
- Meet the Artful Leftie: NIH Jump-Starts U.S.-Canadian FTLD Cohorts
- DIAN: Registry for eFAD to Chart Alzheimer’s Preclinical Decade
- Chiò A, Borghero G, Restagno G, Mora G, Drepper C, Traynor BJ, Sendtner M, Brunetti M, Ossola I, Calvo A, Pugliatti M, Sotgiu MA, Murru MR, Marrosu MG, Marrosu F, Marinou K, Mandrioli J, Sola P, Caponnetto C, Mancardi G, Mandich P, La Bella V, Spataro R, Conte A, Monsurrò MR, Tedeschi G, Pisano F, Bartolomei I, Salvi F, Lauria Pinter G, Simone I, Logroscino G, Gambardella A, Quattrone A, Lunetta C, Volanti P, Zollino M, Penco S, Battistini S, , Renton AE, Majounie E, Abramzon Y, Conforti FL, Giannini F, Corbo M, Sabatelli M. Clinical characteristics of patients with familial amyotrophic lateral sclerosis carrying the pathogenic GGGGCC hexanucleotide repeat expansion of C9ORF72. Brain. 2012 Mar;135(Pt 3):784-93. PubMed.
- Ryman DC, Acosta-Baena N, Aisen PS, Bird T, Danek A, Fox NC, Goate A, Frommelt P, Ghetti B, Langbaum JB, Lopera F, Martins R, Masters CL, Mayeux RP, McDade E, Moreno S, Reiman EM, Ringman JM, Salloway S, Schofield PR, Sperling R, Tariot PN, Xiong C, Morris JC, Bateman RJ, Dominantly Inherited Alzheimer Network. Symptom onset in autosomal dominant Alzheimer disease: a systematic review and meta-analysis. Neurology. 2014 Jul 15;83(3):253-60. Epub 2014 Jun 13 PubMed.
- Janssen JC, Schott JM, Cipolotti L, Fox NC, Scahill RI, Josephs KA, Stevens JM, Rossor MN. Mapping the onset and progression of atrophy in familial frontotemporal lobar degeneration. J Neurol Neurosurg Psychiatry. 2005 Feb;76(2):162-8. PubMed.
- Dopper EG, Rombouts SA, Jiskoot LC, Heijer Td, de Graaf JR, Koning Id, Hammerschlag AR, Seelaar H, Seeley WW, Veer IM, van Buchem MA, Rizzu P, van Swieten JC. Structural and functional brain connectivity in presymptomatic familial frontotemporal dementia. Neurology. 2013 Feb 26;80(9):814-23. Epub 2013 Feb 6 PubMed.
- Rohrer JD, Warren JD, Fox NC, Rossor MN. Presymptomatic studies in genetic frontotemporal dementia. Rev Neurol (Paris). 2013 Oct;169(10):820-4. Epub 2013 Sep 4 PubMed.
- Rohrer JD, Nicholas JM, Cash DM, van Swieten J, Dopper E, Jiskoot L, van Minkelen R, Rombouts SA, Cardoso MJ, Clegg S, Espak M, Mead S, Thomas DL, De Vita E, Masellis M, Black SE, Freedman M, Keren R, MacIntosh BJ, Rogaeva E, Tang-Wai D, Tartaglia MC, Laforce R Jr, Tagliavini F, Tiraboschi P, Redaelli V, Prioni S, Grisoli M, Borroni B, Padovani A, Galimberti D, Scarpini E, Arighi A, Fumagalli G, Rowe JB, Coyle-Gilchrist I, Graff C, Fallström M, Jelic V, Ståhlbom AK, Andersson C, Thonberg H, Lilius L, Frisoni GB, Pievani M, Bocchetta M, Benussi L, Ghidoni R, Finger E, Sorbi S, Nacmias B, Lombardi G, Polito C, Warren JD, Ourselin S, Fox NC, Rossor MN. Presymptomatic cognitive and neuroanatomical changes in genetic frontotemporal dementia in the Genetic Frontotemporal dementia Initiative (GENFI) study: a cross-sectional analysis. Lancet Neurol. 2015 Mar;14(3):253-62. Epub 2015 Feb 4 PubMed.