Similarities between familial and sporadic autopsy-proven progressive supranuclear palsy

Abstract

Progressive supranuclear palsy (PSP) is a relatively common neurodegenerative tauopathy clinically characterized by parkinsonism, axial rigidity, and supranuclear gaze palsy. Pathologic findings of PSP are neuronal loss, gliosis, and neurofibrillary tangles in basal ganglia, diencephalon, and brainstem; there is increasing recognition of clinicopathologic variants of PSP.¹

Progressive supranuclear palsy (PSP) is a relatively common neurodegenerative tauopathy clinically characterized by parkinsonism, axial rigidity, and supranuclear gaze palsy. Pathologic findings of PSP are neuronal loss, gliosis, and neurofibrillary tangles in basal ganglia, diencephalon, and brainstem; there is increasing recognition of clinicopathologic variants of PSP.¹

PSP is usually a sporadic condition; however, there are reports of familial aggregation in PSP.² Recently, a genome-wide association study (GWAS) identified several genes associated with increased risk of PSP.³ This report describes clinical, pathologic, and genetic differences in familial and sporadic autopsy-proven PSP.

Methods.

We identified cases of autopsy-proven PSP with family history of neurodegenerative disease, including PSP, parkinsonism, or dementia from available medical records in the brain bank at Mayo Clinic in Jacksonville, FL. Cases considered to have a positive family history were those who had at least one first- or second-degree relative with parkinsonism or dementia. We excluded cases with pathogenic mutations in MAPT. All brains were acquired with appropriate ethical approval, and all autopsies were approved by legal next of kin. Available medical records were abstracted for clinical information in a consecutive series of cases from the period 1998 to 2008, including cardinal clinical features of PSP, such as parkinsonism and unexplained falls, as well as other notable neurologic findings. All cases had standardized neuropathologic evaluation as part of CurePSP brain bank, including assessment of neuronal and glial tau pathology with phospho-tau immunohistochemistry in 21 brain regions.⁴ Atypical PSP included cases with extensive cortical involvement or paucity of tau pathology in cardinal regions.¹ For each case, DNA was extracted from frozen brain tissue and genotyped for APOE and MAPT. A subset of PSP cases (n = 375) was included in the CurePSP GWAS³ (tables e-1 and e-2 on the Neurology® Web site at www.neurology.org). MAPT H1/H2 was determined from a single nucleotide polymorphism, rs8070723. For comparison, genotypes were available on healthy, living control subjects collected from neurologically normal family members or caregivers of patients with Parkinson disease followed at Mayo Clinic Florida.

Student t test was used to compare continuous variables. Fisher exact test or χ² test was used to compare group differences for categorical variables. Pathologic tau lesion scores were compared with the Mann-Whitney rank sum test. Statistical significance was considered for p values <0.05. Statistical analysis was performed with SigmaPlot version 11.0 (Systat Software Inc., San Jose, CA).

Results.

Of the 375 PSP cases, 58 (15%) had a family member with clinical history of PSP, parkinsonism, or dementia. Eleven cases (3%) had a family member with clinical history of PSP. Table 1 summarizes select demographic, clinical, pathologic, and genetic features of familial and sporadic PSP. Pathologically, familial PSP had more frequent atypical pathology compared with sporadic PSP (p = 0.004). In most regions, the severity of tau lesion scores was lower in familial than sporadic PSP, although there were no significant differences after adjustments for multiple comparisons (tables e-1 and e-2). Table e-3 provides genotype and allele frequencies for familial and sporadic PSP for genome-wide significant single nucleotide polymorphisms from the CurePSP GWAS.

Table 1.

Clinical, pathologic, and genetic information for familial and sporadic PSP

Discussion.

We found very few differences in pathologic and genetic characteristics between familial and sporadic PSP. Moreover, on a range of clinical features (table 1), familial and sporadic PSP did not differ regardless of whether one used stringent or liberal criteria for defining familial PSP. We defined familial PSP when probands had either a first- or second-degree relative with either parkinsonism or dementia; however, we acknowledge the difficulty of defining familial PSP based solely on clinical criteria, given the increasing recognition of clinical heterogeneity of pathologically confirmed PSP.¹ In this study, we used different levels of stringency to overcome this weakness, by including dementia, parkinsonism, and the most stringent diagnosis, PSP, to define familial PSP.

Pathologically, we found that familial PSP tended to have less tau pathology compared with sporadic PSP. These results are contrary to those found in other familial neurodegenerative disorders, such as Alzheimer disease (AD); familial AD often has more severe pathology than sporadic AD.⁵ Moreover, familial tauopathies due to mutations in MAPT invariably have severe tau pathology.⁶ This unexpected finding may indicate that genetic factors other than MAPT may have an important role in familial PSP.

APOE ε4 allele frequency was significantly lower in PSP compared with controls, a finding that was noted in the CurePSP GWAS.³ In previous studies, presence of the ε4 allele has been associated with increased AD pathology in PSP.⁷ Familial PSP had a stronger association with the MAPT H1 haplotype than both controls and sporadic PSP. This finding suggests that MAPT H1 affects risk of familial PSP, but not necessarily the severity of tau pathology.

Although this report is based on a large series of autopsy-confirmed PSP, it is still underpowered to find small differences between familial and sporadic PSP, especially for genetic factors with small effect sizes, such as non-MAPT genes implicated in the CurePSP GWAS.³