Progressive Proximal Muscle Weakness Due to a 51 CAG Repeat Expansion in Exon 1 of the Androgen Receptor Gene: A Case Report of Kennedy Disease

Introduction

Spinal and bulbar muscular atrophy (SBMA, Kennedy disease) is a rare X-linked neuromuscular disorder caused by an androgen receptor (AR) CAG repeat expansion on Xq11–12 [1]. Kennedy disease affects approximately 1 in 100 000 individuals, with higher prevalence in Finland’s Wazard region, and typically presents in the third to fifth decades of life [1,2]. SBMA is a slowly progressive lower motor neuron disorder characterized by limb and bulbar weakness, muscle atrophy with fasciculations, and androgen insensitivity features such as gynecomastia. The additional presence of tongue atrophy with tremor, together with the absence of dermatomyositis-specific cutaneous findings, favors the diagnosis of Kennedy disease [3].

Due to overlapping clinical features, Kennedy disease is frequently initially diagnosed as dermatomyositis, polymyositis, myasthenia gravis, or amyotrophic lateral sclerosis (ALS), with reports describing a case with up to 10 years of being initially misdiagnosed as polymyositis [1,4]. However, polymyositis can be excluded by measuring muscle enzymes, performing muscle imaging, and electromyography [5]. Kennedy disease is diagnosed by neurological examination and confirmed by detecting a mutation with more than 35 CAG repeat expansion in exon 1 of the AR gene on the X chromosome [6]. Recently, Cai et al found that the genotype was correlated with the clinical phenotype and that rehabilitation can help slow the progression of muscle atrophy [7]. Moreover, there has been a lack of standardized treatment for this condition.

This report describes the case of a 52-year-old man with a 10-year history of progressive limb weakness due to Kennedy disease, confirmed by identifying a 51 CAG repeat expansion in exon 1 of the AR gene on the X chromosome.

Case Report

A 52-year-old man had been experiencing progressive limb weakness for 10 years. Ten years ago, he initially experienced mild bilateral hand weakness, accompanied by small-amplitude action tremors, and did not seek any medical care. Four years later, he developed difficulty climbing stairs, with gradual worsening of hand weakness and tremor as the disease progressed. During the first 9 years of disease progression, he reported no sensory disturbances or sphincter dysfunction, and his weakness remained non-fluctuating. These changes did not affect his ability to perform activities of daily living; therefore, he did not seek medical attention.

One year ago, worsening leg weakness reduced his walking tolerance to less than 100 meters, and he was admitted to a local hospital. Laboratory investigations revealed persistently elevated creatine kinase levels (CK), ranging from 808 to 2300 U/L (normal 39–308 U/L), and mildly elevated transaminase AST at 47.7 U/L (normal <40 U/L) and ALT at 49.5 U/L (normal <41 U/L). Magnetic resonance imaging (MRI) ruled out lumbar or spinal cord pathology. A brain MRI scan was also normal. A nerve conduction study (NCS) performed at that time showed demyelination and involvement of sensory and motor nerves. A quadriceps femoris biopsy revealed no significant findings. Because the underlying cause remained unclear, he was diagnosed with polymyositis and chronic inflammatory demyelinating polyneuropathy (CIDP). At a local hospital, he began immunosuppressive therapy comprising methylprednisolone 16 mg/day, hydroxychloroquine (HCQ) 200 mg/day, and mycophenolate mofetil at an undocumented dose, maintained for several months. Although he experienced transient improvement, his symptoms continued to progress despite ongoing treatment.

Due to persistent limb weakness and sustained CK elevation despite immunosuppression, he was referred to our hospital. Upon admission, a neurological examination revealed proximal weakness in the upper (Medical Research Council [MRC] grade 4/5) and lower limbs (MRC grade 3/5), with preserved distal strength (MRC grade 5/5). Notably, he exhibited proximal muscle atrophy, gynecomastia (Figure 1), and bilateral tongue atrophy with tremors (Figure 2). Deep tendon reflexes were diminished (1+ in all limbs), while sensory function remained intact. Babinski and Hoffmann signs were absent. Given the clinical phenotype and prior diagnoses, inflammatory myopathy remained in the differential; additional considerations included a lower motor neuron disorder, ALS, a primary myopathy, and myasthenia gravis. Accordingly, further immunologic testing was performed at the local hospital and was negative for antinuclear antibodies and polymyositis-specific autoantibodies.

A nerve conduction studies and electromyography (NCS/EMG) was performed to further characterize the neuromuscular pathology. It demonstrated normal motor conduction but absent sensory nerve action potentials (SNAPs) in the lower limbs and reduced SNAP amplitudes in the upper limbs. Needle EMG revealed chronic neurogenic changes with increased insertional activity and large-amplitude, prolonged-duration, polyphasic motor unit action potentials in limb muscles, without repetitive discharges. These findings suggested a motor neuron disorder rather than an inflammatory myopathy. Given the presence of gynecomastia, hormonal evaluation was conducted, showing normal levels of free testosterone at 21.8 pg/mL (normal 15–50 pg/mL), estradiol at 35.7 pg/mL (normal 11.3–43.2 pg/mL), total testosterone at 15.7 nmol/L (normal 6.680–25.700 nmol/L), and DHEA-SO4 at 1.74 μg/mL (normal 0.443–3.31 μg/mL), ruling out an endocrine etiology.

His atypical phenotype and electrophysiological profile (a lower motor neuron-predominant syndrome with proximal/bulbar weakness, gynecomastia, and lingual atrophy with tremor) suggested a diagnosis of spinal and bulbar muscular atrophy (Kennedy disease) over amyotrophic lateral sclerosis, polymyositis, and myasthenia gravis, and a genetic evaluation was performed. The initial hospital assay was non-informative because it interrogated only single-nucleotide variants and did not assess repeat expansions. In consultation with medical genetics, whole-exome sequencing was performed at a CLIA-certified commercial clinical laboratory (3billion Inc., Seoul, Republic of Korea). Genomic DNA extracted from a buccal swab sample underwent exome capture and next-generation sequencing. Repeat expansion analysis was conducted using ExpansionHunter (v5.0.0), a validated bioinformatics tool designed to detect short tandem repeat expansions from short-read sequencing data, which identified a hemizygous pathogenic CAG repeat expansion in exon 1 of the androgen receptor (AR) gene, with a repeat number of 51 (range 50–53) (Figure 3). These findings were consistent with a molecular diagnosis of Kennedy disease.

During the period of diagnostic uncertainty, and after comprehensive counseling regarding the potential benefits and risks of available treatment options, the patient consented to immunosuppressive therapy with rituximab, methotrexate, and plasmapheresis. However, no clinical improvement was observed. Following confirmation of Kennedy disease, immunosuppressive therapy was discontinued. After 6 months of follow-up, the patient remains hemodynamically stable, with a mild progressive weakness of the lower limbs, and no disease-specific treatment has been initiated.

Discussion

This case highlights the diagnostic challenges in distinguishing Kennedy disease (spinal and bulbar muscular atrophy), a rare neuromuscular disorder that remains underrecognized, from inflammatory myopathies such as polymyositis. Clinicians should maintain a high index of suspicion for Kennedy disease in patients presenting with chronic progressive proximal weakness, tongue atrophy with tremor, gynecomastia, reduced sensory nerve action potentials, and negative inflammatory panels. It also emphasizes the need to use appropriate genetic testing with CAG repeat expansion analysis rather than single-point mutation detection to establish an accurate diagnosis. Early recognition leads to the avoidance of unnecessary immunosuppressive therapy and cost-effectiveness, particularly in limited-resource countries.

Kennedy disease is a polyglutamine expansion disorder caused by a CAG repeat expansion in exon 1 of the androgen receptor gene on Xq11–12. Normal alleles contain 11 to 36 repeats; the European Federation of Neurological Societies sets a diagnostic threshold at ≥35 repeats, with most symptomatic cases harboring 38 to 62 repeats [6,8,9]. The mutant AR protein causes progressive degeneration of anterior horn cells and dorsal root ganglia, resulting in motor neuron dysfunction, muscle weakness, and distal sensory impairment [3]. Additionally, the toxic gain of function in skeletal muscle and partial loss of AR activity contribute to androgen insensitivity features, including gynecomastia and reduced fertility [3,8]. Elevated creatine kinase in SBMA may reflect combined neurogenic and myopathic mechanisms. The accumulation of polyglutamine-expanded androgen receptor protein within muscle tissue has been linked to impaired myogenic differentiation and defective fusion of satellite cells, suggesting that SBMA directly affects myofibers in addition to motor neurons and sensory fibers [10]. In contrast, in polymyositis (PM), CK elevation arises from an immune-mediated mechanism predominantly driven by cytotoxic CD8+ T lymphocytes. These T cells recognize muscle fibers as “foreign” and initiate a direct cytotoxic attack, leading to muscle fiber necrosis and sarcolemmal disruption. The resultant leakage of CK into the circulation manifests as proximal symmetric muscle weakness [11]. Although the underlying mechanisms differ, both conditions present proximal weakness and elevated CK, which can contribute to diagnostic confusion.

The age at disease onset varies from 18 to 64 years, with most cases manifesting in the fourth or fifth decade of life [9]. The diagnosis of SBMA is confirmed by genetic testing, identifying an expanded CAG repeat in the androgen receptor gene [3]. Although not required for diagnosis, muscle biopsy can reveal neurogenic as well as myogenic atrophy [4]. In our case, CK levels were markedly elevated, liver enzymes mildly increased, and the muscle biopsy was non-contributory. NCS showed normal motor conduction with absent or reduced SNAPs, consistent with SBMA.

Kennedy disease is commonly initially diagnosed as dermatomyositis, polymyositis, myasthenia gravis, or ALS due to its overlapping clinical characteristics. Myasthenia gravis typically shows diurnal fluctuation, preserved reflexes, positive repetitive nerve stimulation, and thymic abnormalities on imaging [4]. In contrast, Kennedy disease lacks symptom fluctuation, shows diminished reflexes, normal thymus, and no specific findings on nerve stimulation. In addition, motor neuron disease, particularly ALS, poses the greatest diagnostic challenge due to its overlap with Kennedy disease in age at onset, clinical features, NCS/EMG, and muscle biopsy findings. Unlike ALS, which involves both upper and lower motor neurons, progresses rapidly, and lacks sensory or endocrine abnormalities, KD presents with lower motor neuron signs, possible sensory involvement, endocrine features, elevated CK, and a slower course [1,4]. Genetic testing confirms the diagnosis.

In our case, the initial diagnosis of polymyositis underscores the difficulty in distinguishing neurogenic from inflammatory myopathies, as both present with progressive proximal muscle weakness and elevated CK levels. Polymyositis, an immune-mediated inflammatory myopathy, typically features subacute proximal muscle weakness, muscle enzyme elevation, myopathic EMG findings, and inflammatory infiltrate with muscle fiber necrosis on biopsy, and a variable response to immunosuppressive therapy [5]. Moreover, while electrodiagnostic studies in polymyositis typically show normal motor and sensory conduction with a myogenic EMG pattern and increased spontaneous activity [3], our patient’s NCS/EMG revealed reduced or absent SNAPs, consistent with chronic denervation and sensory nerve involvement, findings more suggestive of a neurogenic rather than myopathic process. Therefore, polymyositis can be excluded by checking for elevated levels of muscle enzymes, detection of inflammation by MRI scans and biopsy, and electromyography. Additionally, key features of Kennedy disease, such as tongue atrophy, tongue tremors (absent in polymyositis), and gynecomastia, were initially overlooked, further delaying diagnosis. This case highlights the need to consider Kennedy disease in patients with progressive proximal weakness, bulbar involvement, and fluctuating CK levels when immune-mediated myopathies do not follow the expected course. Compared to the case reported by Harutunian et al, which described a patient with Kennedy disease initially diagnosed as polymyositis, our patient had a similar disease duration of approximately 10 years but only began comprehensive evaluation within the past year [1]. They also received multiple immunosuppressants for over a decade, without improvement. The correct diagnosis of Kennedy disease in that case was only considered after the appearance of gynecomastia, lower facial weakness, chin myokymia, and tongue atrophy, which led to confirmatory genetic testing. This is quite similar to our clinical case, highlighting the importance of recognizing the early signs of Kennedy disease. In contrast, in our case, weakness of the lower limbs was the initial presenting symptom, and the patient had 2 children, whereas Cai et al [7] reported gynecomastia as the first symptom, accompanied by progressive infertility and sexual dysfunction. This difference may be explained by the late onset of the disease, and gynecomastia is often overlooked.

Genetic testing remains the definitive diagnostic tool for Kennedy disease. In our case, the initial genetic assay was non-informative because it assessed only single-nucleotide variants and did not evaluate repeat expansions. SBMA is caused by expansion of the CAG repeat in exon 1 of the androgen receptor gene; accordingly, quantifying AR CAG repeat length is the diagnostic standard, with ≥35 CAG repeats required to confirm the diagnosis [12]. Although PCR-based fragment length analysis and repeat-primed PCR are considered the standard for confirming CAG repeat expansion, next-generation sequencing combined with dedicated repeat expansion detection tools can reliably identify pathogenic expansions [13,14], as demonstrated in our case. Moreover, the estimated repeat size was clearly within the pathogenic range and was strongly correlated with the patient’s characteristic clinical phenotype. We acknowledge that the absence of fragment length analysis is a limitation of this report.

Currently, there is no disease-modifying treatment for SBMA. Randomized, placebo-controlled clinical trials investigating androgen-reducing agents, including leuprorelin and dutasteride, have failed to demonstrate significant efficacy in improving key functional outcomes such as swallowing ability and muscle strength [9,15]. Given the progressive nature of SBMA, a multidisciplinary approach involving neurology, rehabilitation, and respiratory support is essential. Future research should focus on identifying targeted therapies that address the underlying pathophysiology of the disease.

Through this case, we recommend a streamlined diagnostic algorithm: teach weakness-pattern recognition; and SBMA characteristics (bulbar symptoms, lingual atrophy/tremor, gynecomastia); use NCS/EMG cues (reduced or absent sensory nerve action potentials with relatively preserved motor conduction); and order early AR CAG repeat testing to avoid unwarranted immunosuppression. Prospective validation should quantify improvements in diagnostic yield, time to diagnosis, and cost savings.

This report has several limitations. First, it describes a single patient, limiting generalizability. Second, PCR-based fragment length analysis was not performed to orthogonally confirm the repeat size. Nevertheless, the clinical–genetic concordance strongly supports the diagnosis.

Conclusions

Kennedy disease is a rare neuromuscular disorder that can mimic other neurodegenerative or inflammatory conditions. This case underscores the importance of considering Kennedy disease in patients with progressive proximal muscle weakness, fluctuating creatine kinase levels, bulbar involvement, signs of androgen insensitivity, negative results on polymyositis workup, and no response to immunosuppressive therapy. Accurate diagnosis requires genetic testing with CAG repeat expansion analysis, as single-point mutation testing can yield false-negative results. Increased clinical awareness is important to help prevent unnecessary immunosuppressive therapy and improve cost-effectiveness, particularly in limited-resource countries. A structured diagnostic approach to patients with weakness – including targeted history and examination for bulbar signs and key NCS/EMG features – may facilitate early recognition of Kennedy disease. Given the lack of disease-modifying treatments, further research is needed to better understand the disease course and identify potential therapeutic strategies.