RESULTS
A total of 184 patients in the database had a clinical diagnosis of PSP (
n=86), CBS (
n=37), or MSA (
n=61). A total of 31 patients (31/184, 16.9%) (males: 22, 71.0%), with a median age at onset of 58 years (range: 20–68 years) and duration of illness of 4 years (range: 1–22 years), had various atypical features and were thus classified into the AAP group (
Table 1). A positive family history was the most frequent atypical feature observed in 9 patients (29.0%). Among these patients, the PSP phenotype was the most common (19/31 patients, 61.3%), followed by MSA (8/31 patients, 25.8%) and CBS (4/31 patients, 12.9%). The subgroup distributions of demographic and atypical features are presented in
Table 1. All patients had at least one MRI of the brain, 27 patients had undergone autoimmune and paraneoplastic antibody panels, and 18 patients had undergone exome sequencing.
- PSP
Among the 86 PSP patients, 19 (22.1%) had atypical features. Among these patients, the most frequent atypical feature was autonomic dysfunction (urinary urge incontinence with or without orthostatic hypotension seen in 6/19 patients, 31.6%), followed by a positive family history (of similar illness in 4 patients and dementia in 1 patient) and prominent visual hallucination seen in 5 patients each (26.3%). Among the 11 patients with an alternative diagnosis, 6 patients had a probable genetic cause, 3 patients had an autoimmune cause (immunoglobulin-like cell adhesion molecule 5 [anti-IgLON5] antibody, anti-N-methyl D-aspartate [NMDA] receptor antibody, and anti-Ma2 antibody in 1 patient each), and the remaining 2 patients had a structural cause (hypothalamic glioma and pineal lesion in 1 patient each).
Among the 6 patients with a genetic cause, 1 patient had compound heterozygous likely pathogenic variants in the NPC1 gene (NPC1: 1408G>A;p.Ala470Pro/c.1816G>C;p. Glu606Gln), confirming the diagnosis of Niemann-Pick disease type C (NPC). Another patient had compound heterozygous pathogenic variants in the CYP27A1 gene (CYP27A1: c.1184+1G>A;p.?/c.1537C>T;p.Arg513Cys), with consistent clinical and radiological findings confirming the diagnosis of cerebrotendinous xanthomatosis (CTX). Another 2 patients had pathogenic variants in the GBA1 gene in a heterozygous state (c.1342G>C;p.Asp448His and c.989G>A;p.Trp330Ter). One additional patient with prominent hallucinations had a variant of uncertain significance in the MAPT gene (MAPT: c.1238G>C;p.Arg413Pro). The remaining patient had mild ptosis, gaze impairment with no significant improvement in the vestibulo-ocular maneuver, and ragged red fibers on muscle biopsy, suggesting the possibility of mitochondrial cytopathy. The patient did not undergo genetic testing.
- CBS
Among the 37 CBS patients, four (10.8%) had atypical features. Among these patients, a positive family history was the most common atypical feature (3/4 patients, 75.0%; similar illness in 2 patients and dementia in 1 patient). All 4 patients had an alternative diagnosis, which included biallelic OPTN mutations in 2 patients (c.1082_1091dup;p.Gln364HisfsTer18/c.1465A>G;p.Lys489Glu and c.1195G>T;p.Glu399Ter), a heterozygous pathogenic mutation in the APP gene in 1 patient (c.2140A>G;p.Thr714Ala) and anti-NMDA receptor antibody encephalitis in the remaining patient.
- MSA
Among the 61 MSA patients, eight (13.1%) had atypical features, seven of whom had the MSA-parkinsonism phenotype (MSA-P), and one of whom had the MSA-cerebellar phenotype (MSA-C). Among these patients, vertical gaze impairment was the most common atypical feature (7/8 patients, 87.5%). Only 2 patients had an alternative diagnosis. One patient (MSA-C) carried a heterozygous pathogenic variant in the PRNP gene (c.305C>T;p.Pro102Leu) with consistent clinical features, resulting in a diagnosis of Gerstmann–Sträussler–Scheinker (GSS) disease. The other patient (MSA-P) carried a novel variant of uncertain significance in the LRRK2 gene (LRRK2: c.7573T>C;p. Ser2525Pro). The remaining 6 patients did not have any alternative diagnoses and continued to have MSA-P with atypical features as a working diagnosis.
DISCUSSION
In this study, close to one-sixth of patients with PSP, CBS, or MSA phenotypes presented one or more atypical features according to the respective diagnostic criteria. More than half of these patients with AAP were found to have an alternative diagnosis. Genetic diagnosis was the most common, followed by an autoimmune cause. A positive family history and inconsistent clinical findings for the respective subtype of APS pointed toward the possibility of an alternative diagnosis.
The clinical diagnosis of APS is often challenging even for those with movement disorders [
6,
7]. Imaging and other relevant investigations help increase the accuracy, but misdiagnosis is not uncommon based on autopsy studies [
8,
9]. The diagnostic criteria available for diagnosing these APSs improve the accuracy and reduce misdiagnosis to a large extent. The red flag signs and exclusion criteria in these diagnostic criteria help identify diseases that mimic APS, thereby improving specificity. However, the strict inclusion criteria may lead to increased specificity while compromising the sensitivity and time to a correct diagnosis [
8,
9]. In the approach of a patient with parkinsonism other than Parkinson’s disease, categorizing them into one of the atypical parkinsonism subtypes helps in planning appropriate investigations. However, the evaluation should proceed to rule out various mimics on the basis of the presence of red flags or exclusion criteria, rather than stopping at just classifying them to one of the phenotypes.
Close to two-thirds of our patients had a PSP phenotype. Among them, more than half had an alternative diagnosis. A positive family history, earlier age at onset, and the presence of clinical findings such as prominent hallucinations and autonomic dysfunction that were inconsistent with a classical PSP diagnosis suggested the possibility of an alternative diagnosis. NPCs with supranuclear gaze palsy and parkinsonism with postural instability and gait impairment often can mimic the PSP phenotype. However, in most patients, the onset is before 40 years of age and may also be associated with a positive family history or consanguinity, as observed in our case. Variants in the
MAPT gene have also been increasingly identified in both sporadic and familial cases with PSP phenotypes. Pathogenic variants in the
MAPT gene can result in inherited phenocopies of PSP, whereas rare variants in the same gene have been shown to result in an earlier age at the onset of sporadic PSP [
3,
10-
12]. The presence of frontal lobe dysfunction, disinhibition, hallucination, and amyotrophy may indicate
MAPT gene abnormalities. Our case with an uncertainly significant variant in the
MAPT gene had frontal lobe dysfunction and hallucinations. Additionally, variants in various other genes, such as
GBA1,
OPTN,
SYNJ1,
SQSTM1, and
LRRK2, have been identified in patients with PSP in Southeast Asia [
10,
13]. One of our patients carried a variant of uncertain significance (VUS) in the
LRRK2 gene, and two of our patients carried pathogenic heterozygous variants in
GBA1. These variants in
GBA1 have been previously observed in patients with Gaucher’s disease. However, their role in PSP patients has yet to be determined. Other genetic causes that can mimic are Perry syndrome due to the
DCTN1 gene variants,
PGRN,
C9orf72,
ATP13A2, spinocerebellar ataxia type-2, CTX, mitochondrial cytopath, and genetic prionopathies [
2].
In addition to genetic causes, various acquired secondary causes can also mimic the PSP phenotype. Autoimmune antibody-related disorders, especially those involving antibodies against Ma2, IgLON5, NMDA, leucine-rich, glioma inactivated 1, and amphiphysin, can also present with a PSP-like phenotype and must be considered in the differential diagnosis. Neurological manifestations secondary to Whipple’s disease, syphilis, and HIV; Mokri syndrome after cardiac bypass or aortic surgery; and structural causes secondary to lesions in and around the midbrain are other causes that need to be considered in PSP mimics [
14].
Among patients with the CBS phenotype, many genetic causes leading to PSP mimics can also result in a CBS-like presentation [
15]. Notably, in 2 patients with
OPTN variants in our cohort, the patients had CBS-like presentations with PSP overlap in the form of gaze impairment. Additionally, genetic variants in the
APP gene have been recently shown to result in a CBS-like phenotype [
16]. Additionally, genetic abnormalities in
PGRN,
C9orf72,
PSEN1,
TBK1,
CSF1R, and
CYP27A1 have been shown to be associated with CBS phenotypes. Other acquired secondary causes that need to be considered include autoimmune disorders (anti-NMDA, anti-glycine receptor, stiff-person syndrome, and antiphospholipid antibody syndrome), sporadic prionopathies, multiple infarct states, osmotic demyelination syndromes, and neuroinfections.
Various genetic and acquired causes can mimic both parkinsonian and ataxic variants of MSA. Among genetic variants, the subtypes of spinocerebellar ataxia, spastic paraplegia type-7, fragile-X tremor ataxia syndrome, various later-onset recessive ataxias, mitochondrial disorders, dentatorubral-pallidoluysian atrophy,
SNCA duplication/triplication, GSS disease presentation of
PRNP genetic abnormalities and Perry syndrome can mimic both the MSA-P and MSA-C subtypes [
17,
18]. Moreover, various types of paraneoplastic cerebellar degeneration and autoimmune cerebellar ataxia can pose diagnostic challenges in the diagnosis of patients with MSA phenocopy. Only one-fourth of our patients with an atypical MSA phenotype received an alternative diagnosis (GSS in one patient and a VUS in
LRRK2).
The current study is a retrospective chart review of cases from our movement disorder database, so there may be significant selection bias toward more atypical cases. Moreover, our center is a referral tertiary care center, and there may be additional referral bias toward more complicated and atypical cases. Another significant limitation is the small sample size. Owing to its retrospective nature, the completeness of the clinical findings may be lacking, and not all patients have undergone investigations uniformly.
- Conclusion
In this retrospective study of “atypical” atypical parkinsonian syndrome from a single center, we report a significant number of various genetic causes presenting as PSP, CBS, or MSA mimics. The presence of red flags and exclusion criteria in the diagnostic criteria of these atypical parkinsonian syndromes points toward alternative causes and helps in planning relevant investigations to identify alternative diagnoses. A large number of patients with atypical features continue to have a working diagnosis of PSP, MSA, or CBS in the absence of a better alternative cause. These patients need to be followed up to determine the evolution of their clinical presentation and whether they develop any additional clinical clues, suggesting an alternative diagnosis.