Nemaline bodies are not always congenital: A case of sporadic late‐onset nemaline myopathy (SLONM) with monoclonal gammopathy of undetermined significance (MGUS)

Sporadic late-onset nemaline myopathy (SLONM) is an acquired adult-onset myopathy characterized by the presence of nemaline bodies/rods in myofibres.1, 2 Due to its rarity and non-specific clinical features, its diagnosis is often missed or delayed. A 66-year-old male presented with insidious onset, progressive weakness of proximal upper and lower limbs and neck of 3 years duration. He did not have a history of distal or cranial muscle weakness, muscle twitching, fatigability, sensory, autonomic, or systemic symptoms. He had diabetes mellitus of 30 years duration and a history of atorvastatin intake for 10 years prior to the present symptom onset, which he had discontinued thereafter. Neurological examination showed mild to moderate wasting of periscapular and gluteal muscles and symmetric scapular winging. There were no fasciculations, and the tone was normal. He had weakness of proximal upper and lower limb muscles without selectivity and mild facial muscle weakness. Neck flexors were weaker than extensors, with weak trunk muscles and negative Beevor's sign. He had a hyperlordotic posture with waddling gait. Ankle deep tendon reflexes were diminished. There were no features of respiratory dysfunction, myotonia, fatigability, sensory abnormalities, or any other neuraxis involvement. Serum creatine kinase (CK) was mildly elevated on a few occasions, maximum value being 356 IU/L (normal: 38–174 IU/L). Electromyoneurography revealed a myopathic process and mild symmetric sensory axonal polyneuropathy. Repetitive nerve stimulation did not show a decremental response. Muscle magnetic resonance imaging showed fatty infiltration of glutei, lower limb adductors, posterior spinal, paraspinal, deltoid, pectoralis, and periscapular muscles associated with myoedema. He was evaluated extensively for late onset myopathy. Myositis antibody immunoblots, anti-HMGCoA reductase antibody, and antibody to voltage-gated PQ calcium channels were negative. Antinuclear antibody panel (immunoblot) and paraneoplastic antibody panel (tissue immunofluorescence assay) were normal. Endocrine work up for thyroid, parathyroid and adrenal cortical hormones were negative. Viral markers were negative. An initial muscle biopsy from vastus lateralis was performed elsewhere 1 year into symptom-onset, which had shown mild myopathic features. However, the sample was formalin-fixed, and enzyme histochemistry was not performed. Next-generation sequencing with an 83-gene panel for muscular dystrophy and congenital myopathy did not reveal any pathogenic variants. Genetic testing for Becker muscular dystrophy, facioscapulohumeral muscular dystrophy, and spinal muscular atrophy were negative. The dried blood test for Pompe disease was normal. Serum IgG kappa light chain was elevated nearly 1.5 years after symptom onset and increased from 3.8 g/dL to 9.06 g/dL over the next 1.5 years. Bone marrow biopsy did not show atypical cells, and there was no renal dysfunction or hypercalcemia. A repeat muscle biopsy from the right biceps brachii (3 years after symptom onset) showed myopathic features: endomysial adipose tissue infiltration, mild increase in endomysial collagen, rounding of fibers, and variation in fiber size. There were many atrophic round fibers, a few atrophic fibers with clumped nuclei, some hypertrophied fibers, regenerating fibers and fibers with internalized nuclei (Figure 1a). Occasional fibers showed few vacuoles. There is no evidence of myonecrosis or inflammatory infiltrate. Modified Gomori Trichrome (MGT) showed numerous intracytoplasmic nemaline bodies, predominantly in the atrophic fibers and mild increase in red reaction product at the periphery of a few (Figure 1b). The nemaline bodies were short rod-shaped and were noted in about 25%–30% of myofibres. Nicotinamide adenine dinucleotide tetrazolium reductase (NADH-TR) and succinate dehydrogenase (SDH) showed increased staining intensity at the periphery in some fibers and some moth-eaten fibers (Figure 1c). There were no cytochrome c oxidase (COX)-deficient fibers on COX-SDH (Figure 1d). ATPase at pH 9.5 showed both type 1 and type 2 fiber atrophy. Immunohistochemistry for alpha, beta, gamma, delta-sarcoglycan, merosin, dysferlin, and emerin showed preserved labeling in all myofibres. Major histocompatibility complex (MHC) I, MHC II, myxovirus resistance protein A (MxA), and p62 immunohistochemistry were negative. Transmission electron microscopy (EM) showed numerous sarcoplasmic nemaline bodies (Figure 1e,f). Correlating with the clinical profile, features were consistent with SLONM. He had initially taken intravenous immunoglobulin (IVIg) pulse 2 years after symptom onset, had noted a transient benefit and completed nine courses. Following the latest muscle biopsy, he was initiated on chemotherapy: lenalidomide-bortezomib-dexamethasone combination for four cycles and was then continued on bortezomib-dexamethasone for another 10 months. He had mild improvement in power during this period but continues to have significant proximal lower limb weakness. Nemaline bodies are derived from Z-line proteins and best visualized on light microscopy by MGT as cytoplasmic or subsarcolemmal rod-shaped reddish structures.1 They are hallmark of “nemaline myopathy,” which is a congenital myopathy. However, they can also be seen physiologically in myotendinous junctions, extra-ocular eye muscles and ageing muscles, and pathologically in adult-onset SLONM and HIV-associated myopathy.1 Nemaline bodies in the latter cases are considered acquired and hence, these conditions are better called “myopathies with rods” and not “nemaline myopathy.” With onset in the fourth to fifth decade, SLONM presents predominantly with progressive proximal muscle weakness and upper and/or lower limb atrophy, often with neck, axial, and bulbar muscle involvement.2, 3 Cardiac involvement such as dilated cardiomyopathy, ventricular dysfunction/failure and conduction abnormalities have been described.2, 3 Nearly half of SLONM cases are associated with monoclonal gammopathy of undetermined significance (MGUS)2 and have more rapid progression than those without.2, 4 Due to its putative autoimmune etiology, patients benefit from immunosuppressive therapy with or without IVIg therapy, and cases associated with MGUS, respond well to autologous stem cell transplantation.2, 4 Given the therapeutic implications, it is important to diagnose SLONM and to screen for paraproteinemia. In the case presented here, there was a delay of 3 years between disease onset and diagnosis, highlighting the practical difficulties. Perhaps the rarity of the disease and nonspecific clinical features lead to considering more common myopathies such as inflammatory myopathies and adult-onset muscular dystrophies. In a study on 18 cases of late-onset myopathies with slowly progressive limb-girdle muscle weakness and remaining unsolved after whole-exome sequencing, 10 were found to have rods in the muscle biopsy, fitting into the SLONM category.5 Whenever muscle biopsies are done, it is important to analyse snap frozen samples with MGT and enzyme histochemical stains, which help detect cytoplasmic inclusions/structures and myofibrillar abnormalities which are otherwise undetectable. In SLONM, the proportion of myofibres that show nemaline bodies can vary widely between cases (1%–94%)2, 3, 6 with mean ranging from 11% to 28%.2, 6 Intranuclear rods can also be present. Unlike the inherited myopathy, nemaline bodies in SLONM can have different morphologies: typical rod-like feature (about 55%), a finer sand-like appearance (about 31%) or mixed features.3 Other additional changes in muscle include myofiber atrophy, myofibrillar disintegration, core-like areas, sarcoplasmic vacuolation, and intermyofibrillary changes. Mitochondrial changes such as ragged-red fibers and COX-deficient fibers are common. Less frequently, myonecrosis and inflammatory infiltrate can be encountered.2 Inflammation in SLONM is generally mild with predominance of macrophages. 66.7% of biopsies can show variable sarcolemmal HLA-ABC expression, and 78.3% can exhibit fine granular p62 expression, posing potential histopathological overlap with idiopathic inflammatory myopathies.3 However, nemaline bodies are not a feature of inflammatory myopathies. Since the histopathological diagnosis hinges on the presence of nemaline bodies, pathologists should be aware that not all biopsies with these structures are diagnosed as congenital “nemaline myopathy.” The histological features that are distinct in SLONM and different from inherited nemaline myopathy include: nemaline bodies in < 50% of myofibres, highly atrophic fibers filled with rods, necrotic fibers, regenerating fibers, and increased endomysial connective tissue.6 In clinically suspected cases, when nemaline bodies are not detectable in MGT, it is worthwhile to perform EM.2 Yet another point to be noted is that they may be absent at symptom onset and a repeat biopsy maybe necessary later in the disease course.2 To conclude, SLONM diagnosis requires a high degree of clinical suspicion and extensive work-up to exclude other muscle diseases. Muscle biopsy plays a key role in its diagnosis, and samples should be processed fresh for MGT staining and enzyme histochemistry. Silvie Thomas: Data acquisition; drafting the manuscript; editing. Deepti Narasimhaiah: Manuscript review and editing; final approval. Sruthi S. Nair: Data acquisition; clinical data collection; manuscript review and editing. Bejoy Thomas: Data acquisition; radiology data collection; drafting the manuscript and review. Rashmi Santhoshkumar: Data acquisition; electron microscopy data collection; drafting the manuscript. Anita Mahadevan: Data acquisition; electron microscopy data collection; manuscript reviewing. Rajalakshmi Poyuran: Conception; manuscript review and editing; final approval. The authors have no funding to report. The authors declare no conflict of interest. Ethics approval was not obtained. As per the Institutional Ethics Committee (IEC) of the Institute, manuscripts/report based on since case are exempt from obtaining IEC approval. Informed consent from the patient was obtained. Patient identifiers are not included in the manuscript and the data is anonymised. Data sharing is not applicable to this article as no new data were created or analyzed in this study.