Until a decade ago, amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) were considered to be distinct neurodegenerative conditions: one was clinically characterized by paralysis and the other by cognitive dysfunction. Epidemiologic, clinical, and neuropathologic data now make it clear that the 2 conditions form a spectrum of disease.1,2 Genetic studies have played a central role in determining the cellular mechanisms underlying the overlap of ALS and FTD. In particular, the pathogenic repeat expansion in the C9ORF72 (Entrez GeneID: 203228) gene was found to account for a large percentage of cases of ALS-FTD.3,4 Despite theseadvances, the relationshipbetweenALSand dementia is only partially resolved. A particularly troubling problem is thatwedonotunderstandwhy somepatientswith ALSdevelopdementiawhile others donot, evenwhen thepatients are close family relatives carrying the same genemutation. Environment and lifestyle factors almost certainly play a role indetermining this outcome.Nevertheless, efforts have focused largely on identifying the genetic factors that lead to dementia in ALS, in part because modern genomic technologies allow these hypotheses to be tested relatively easily. Genetic variants that determine whether an individual carrying the C9ORF72 repeat expansion will present with ALS or FTD have already been reported. For example, homozygosity for the minor allele of rs3173615 in TMEM106B (Entrez GeneID: 54664) protects carriers of the repeat expansion from developing FTD but not from developing ALS.5 These studies focus largely on predicting whether the initial presenting features would be muscle paralysis or frontal lobe dysfunction. In contrast, Chio and colleagues6 followed a cohort of patients with ALS over time to identify those who developed dementia during the course of their illness and those who did not. Using this information, they then tested whether APOE (Entrez GeneID: 348) genotype influenced this outcome and found that the presence of an APOE e2 allele increased the risk of FTD by nearly 3-fold. This is not the first time that APOE has been studied in patients with ALS. Previous studies, however, focused on whether this allele increased the risk of developing ALS or altered the age at onset; results were either negative or unconvincing.7-9 The innovative feature of the study by Chio et al6 is that they evaluated the effect of APOE on longitudinal outcome, not just presentation of disease. This is a subtle but crucially important distinction, as their results show. The most obvious strength of the study by Chio et al6 lies in their use of a population-based, longitudinal registry of patients with ALS, without which they would not have been able to evaluate long-term cognitive status. This Northern Italian registry has been in existence for 20 years,10 and Chio et al6 illustrate how such large-scale data collection efforts can be leveraged to answer important questions in the field. The recognition of APOE as a risk factor for dementia in ALS has immediate implications for clinical care. Knowing that an individual is likely to develop dementia will help prioritize the establishment of a living will at a time when the patient is still able to participate and will help caregivers cope with the patient’s behavioral changes arising from the illness. It will also facilitate genetic counseling of other family members. In addition, the identification of the role of APOE has broader implications for how we think about and perceive neurodegeneration. As neurologists, we are trained to diagnose patients according to their presenting symptoms. This paradigm retains obvious clinical utility, but it is increasingly recognized that neurodegeneration does not respect these silos and is more dynamic in nature. Pathologists frequently observe multiple types of inclusions in brains undergoing autopsy,11 and geneticists are increasingly reporting that mutations in a gene that causes one form of neurodegeneration increase risk in other forms of neurodegeneration.12,13 The study by Chio et al6 is certainly a good step forward in our understanding of the genetic architecture underlying ALS and dementia. Like all good science, however, it raises more questions than it answers. For example, do patients with ALS who are carrying the APOE risk allele have more amyloid deposition or are other protein species involved? Why is the APOE e2 allele a risk factor for FTD, when it is thought to be protective in other neurodegenerative diseases? Why do some patients carrying the APOE risk factor not develop dementia?14 Does this finding suggest the existence of additional risk factors or even protective factors elsewhere in the genome? Genome-wide studies would be the ideal way to search for such loci, but such studies will be challenging owing to the multiple testing involved, the requirement for large sample sizes, and the difficulty in interpreting how these multiple risk factors are operating within the central nervous system. This is a complex problem indeed but an important one to address if we are to arrive at a better understanding of neurodegeneration. Therein lies the rub.