The ε2 allele might play a pathogenic role in other neurodegenerative conditions, as it might increase the risk of Parkinson disease. Patients with FTD carrying the ε4 allele show β-amyloid protein deposits more frequently than ε4 noncarriers, although β-amyloid deposition in FTD could be unrelated to primary pathological processes.5 Even if we cannot exclude the possibility that the association between ε2/ε4 genotype and PPA may be due to the insufficient size of our samples, such an association may be explained by molecular positive heterosis. When a certain chromosomal locus has 2 different alleles (for example, alleles A and B), heterozygosity occurs when a subject has both of the 2 alleles, A and B, and can be therefore defined as an AB heterozygote, while subjects having only the A or B allele can be defined as AA homozygotes and BB homozygotes, respectively. Usually, in a 2-allele polymorphism (where a single chromosomal locus has the 2 alleles, A and B), if the allele A is strongly associated with a certain phenotypic effect, subjects carrying the AA genotype (AA homozygotes) should show the greatest phenotypic effect, subjects carrying the BB genotype (BB homozygotes) should show the least phenotypic effect, and AB heterozygotes should show an intermediate phenotypic effect. Molecular positive heterosis is a counterintuitive phenomenon that occurs when subjects heterozygous for a specific genetic polymorphism (for example, AB heterozygotes) show a significantly greater phenotypic effect (positive heterosis) for a quantitative or dichotomous trait than homozygotes for either allele (AA homozygotes and BB homozygotes).6 Although the possible mechanisms involved in molecular positive heterosis are only partially known, this is a relatively common phenomenon that has been shown in several disorders and may be sex-specific, namely, it may be observed only in either men or women.6 Thus, in our series, women with the APOE ε2/ε4 genotype showed an increased risk for PPA compared with women with either ε2/ε2 or ε4/ε4 homozygosity. Further studies are needed to confirm the role of ε2/ε4 genotype as a risk factor in women with PPA and to clarify the possible pathogenic mechanisms related to APOE in this syndrome.