A case report of riboflavin transporter deficiency: A novel heterozygous pathogenic variant in the SLC52A3 gene

Abstract

Riboflavin transporter deficiency (RTD) is a neurodegenerative disorder that presents from infancy to adulthood with a progressive axonal neuropathy characterized by a variety of neurologic symptoms including hearing loss, weakness, bulbar palsy, and respiratory insufficiency. Pathogenic variants in SLC52A2 and SLC52A3 are implicated in the pathogenesis of RTD type 2 and 3, respectively. Early identification of this disorder is critical, as it is treatable with riboflavin supplementation. We describe a 16-year-old female with a phenotype consistent with RTD3 found to have a novel heterozygous SLC52A3 variant. Though RTD is typically considered an autosomal recessive condition, her heterozygous variant was thought to be disease causing after further genetic analysis and given her improvement in response to riboflavin supplementation. This case highlights the importance of reinterpretation of genetic testing, particularly when there is a high clinical suspicion for disease.

1. Introduction

Brown-Vialetto-Van Laere syndrome (BVVL) was first described in 1894 by Dr. Charles Henry Brown. Fazio-Londe syndrome is considered the same disease entity, but without sensorineural hearing loss, and both disorders have been more recently referred to under the umbrella term Riboflavin Transporter Deficiency (RTD). RTD is categorized into two types, RTD type 2 and 3, corresponding to the recently discovered SLC52A2 and SLC52A3 genes, and has typically been considered an autosomal recessive condition [5,9].

We describe a 16-year-old female with a phenotype consistent with RTD3 and subsequent improvement after riboflavin supplementation. She was found to have a novel heterozygous variant in the SLC52A3 gene. This an example of the phenotypic and genotypic spectrum of RTD that can be inherited in an autosomal dominant fashion with variable penetrance. Given that RTD is an extremely rare metabolic disorder and one of the few with known effective treatment, having a high index of suspicion for the diagnosis is of paramount importance.

2. Case history

A 16-year-old Ukrainian woman with a remote history of sensorineural hearing loss presented for evaluation of one year of progressive bulbar and predominantly distal upper and lower extremity weakness. Symptoms started with right eyelid ptosis and sensory changes on the left face. Over the next three months, she developed bilateral hand tremors and distal weakness followed by progressive bifacial weakness, dysarthria, and dysphagia.

Prior to relocating to the United States, she had evaluations in Ukraine and Romania. Initial work-up showed positive Lyme antibodies and she was treated with a course of antibiotics. Other studies included unremarkable additional infectious studies (Epstein-Barr virus, cytomegalovirus, herpes simplex virus, and varicella-zoster virus), electroencephalogram (EEG) and magnetic resonance imaging (MRI) of the brain. Approximately five months after initial presentation, she had worsened bulbar weakness and distal atrophy. Lumbar puncture (LP) showed elevated protein at 87 mg/dL (ref 15–80 mg/dL) with normal cell count and glucose. Amyotrophic Lateral Sclerosis (ALS) genetic testing and ganglioside antibody panel were negative. She started a variety of vitamin supplements (dosage and duration unknown) and received five doses of intravenous immunoglobulin (IVIG) without improvement.

After one year of persistent symptoms, she emigrated to Chicago for evaluation at the neuromuscular clinic at Ann & Robert H. Lurie Children's Hospital. Examination revealed dysarthria, dysphagia, facial diplegia, tongue atrophy and fasciculations with severe distal weakness and lower extremity areflexia. Her sensation and coordination were normal. She reported a 20-pound weight loss given her bulbar weakness. She did not have any respiratory symptoms. Electromyography/nerve conduction studies (EMG/NCS) showed a primarily motor axonal polyneuropathy. Compound motor action potentials (CMAP) were severely reduced in amplitude with normal conduction velocities (Table 1). Sensory studies were normal. EMG showed evidence of active denervation, with chronic neurogenic changes and a reduced recruitment pattern (Table 2).

Table 1.

Motor nerve conduction studies. APB = abductor pollicis brevis. ADM = abductor digiti minimi. EDB = extensor digitorum brevis. AH = abductor hallucis. DL = distal latency, CMAP = compound muscle action potential, CV = conduction velocity.

Nerve		DL (ms)	CMAP (mV)	CV (m/s)
Right median	Wrist	7.67	0.1	NR
	Elbow	NR	NR
Left median (APB)	Wrist	4.13	1.2	45
	Elbow	8.79	0.9
Right ulnar (ADM)	Wrist	2.67	3.8	51
	Elbow	7.42	3.5
Left ulnar (ADM)	Wrist	2.65	2.3	55
	Elbow	7.35	2.3
Right peroneal (EDB)	Ankle	7.04	0.2	36
	Fib head	16.42	0.2
Left peroneal (EDB)	Ankle	NR	NR	NR
	Fib head	NR	NR
Right tibial (AH)	Ankle	3.88	9.1	41
	Pop fossa	13.98	7.8
Left tibial (AH)	Ankle	3.96	9.1	41
	Pop fossa	14.04	7.8

Table 2.

Electromyography. N = normal, Fib = fibrillations, PSW = polyspike wave, HF = high frequency, MUAP = motor unit action potential, Amp = amplitude, Dur = durations, PP = polyphasia, L = Left, R = Right.

			Spontaneous				MUAP
Muscle	Nerve	Roots	IA	Fib	PSW	Fasc	Amp	Dur	PP	Recruitment
R tibialis anterior	Deep peroneal	L4-L5	N	2+	2+	2+	2+	1+	1+	Reduced
L vastus lateralus	Femoral	L2-L4	N	None	None	None	N	N	N	Reduced
L deltoid (posterior)	Axillary	C5-C6	N	None	None	None	N	N	N	N
L pronator teres	Median	C6-C7	N	None	None	None	1+	1+	1+	Reduced
L thoracic paraspinals	Spinal	T1-T12	N	None	None	None	–	–	–	–
L abductor digit minimi	Ulnar	C8-T1	N	2+	2+	None	1+	1+	1+	Discrete

Upon hospital admission, metabolic, autoimmune, infectious, CSF studies and vitamin levels were unremarkable, including riboflavin levels and acylcarnitine profile. MRI of the brain and cervical spine with and without contrast and computed tomography (CT) of the chest abdomen and pelvis were normal. Nasogastric feeds were initiated due to aspiration of thin and thickened liquids and laryngeal penetration with solids. Audiology evaluation revealed mild to moderate mixed hearing loss and auditory brainstem responses (ABR) were consistent with bilateral auditory neuropathy spectrum disorder. There was bilateral optic nerve pallor on ophthalmologic examination. Given the clinical suspicion for RTD, she was started on 40 mg/kg/day (2800 mg total) of oral riboflavin pending results of whole genome sequencing (WGS) at a reference laboratory (Rady Children's Institute). She was also treated with five rounds of plasmapheresis (five full volume exchanges) while awaiting genetic testing results given the severity of her symptoms and concern for a potential autoimmune neuropathy.

WGS with copy number analysis revealed the heterozygous variant in SLC52A3 NM_033409.4:c.971 A > G (p.Tyr324Cys). No additional sequence or copy number variants involving SLC52A3 were identified by WGS or single nucleotide polymorphism (SNP) array tests. Two heterozygous variants of uncertain significance in the gene HMBS were also reported by WGS: NM_000190.4:c.241C > T (p.Leu81Phe) and NM_000190.4:c.523C > T (p.Arg175Trp). Neither variant was maternally inherited and thus they are unlikely to be in trans; further, acute intermittent porphyria (associated with HMBS gene) was felt to be inconsistent with the patient's presentation and their normal urine porphyrin results. No additional variants thought to be related to her underlying presentation were identified by either assay.

Two weeks after initiation of treatment, she showed mild improvement in weakness and repeat swallow study showed a resolution of aspiration. Two months later, she showed continued slow improvement in her facial and acral weakness and repeat audiology evaluation showed mild improvement bilaterally. At her most recent visit seven months after initiation of treatment, her lower extremity weakness had resolved.

3. Discussion

RTD is a rare neurological disorder which leads to significant morbidity and mortality without early recognition and adequate treatment. It is caused by pathogenic variants in SLC52A2 (RTD2) and SLC52A3 (RTD3), which encode for riboflavin transporters. Although the underlying pathophysiology of neuropathy in RTD is not entirely clear, it is suspected that mitochondrial dysfunction may play a role as riboflavin is a precursor for several metabolites that are involved in the electron transport chain [11]. In mammals, riboflavin cannot be synthesized de novo and is an essential part of the diet to allow for efficient and effective energy metabolism. We report the case of a 16-year old female with a clinical phenotype suggestive of RTD3 who was subsequently found to have the novel heterozygous missense variant in SLC52A3 NM_033409.4:c.971 A > G (p.Tyr324Cys). This variant was classified and reported by the reference laboratory as a variant of unknown significance. However, a subsequent review of the available data suggests that the evidence may indeed reach the threshold for a classification of likely pathogenic [15]. The variant affects a highly conserved amino acid position and is strongly predicted by the computational pathogenicity prediction meta-tool REVEL to be damaging to protein function (REVEL = 0.939; PP3_strong) [7,14]; it is completely absent from the Genome Aggregation Database (gnomAD; PM2_supporting) [6,10]; and the patient's phenotype and response to treatment is consistent with RTD, a highly specific phenotype with limited genetic heterogeneity (PP4). This variant is therefore greatly suspected to be disease-causing.

RTD most commonly presents with sensorineural hearing loss, bulbar palsy, extremity and facial weakness, and respiratory insufficiency [2]. These clinical characteristics can sometimes mimic juvenile-onset amyotrophic lateral sclerosis (ALS) or autoimmune axonal motor neuropathy. A key difference distinguishing RTD from these other disorders is that it is uniquely responsive to high dose riboflavin supplementation with the potential for reversal of symptoms with prolonged treatment. Hearing loss as the initial symptom, as well as facial weakness, early onset bulbar weakness and late onset of symptoms (>10 years old) are more common in RTD3 and were all present in this patient. Optic atrophy, ataxia and sensory abnormalities are more common in RTD2 [8,13]. Her improvement in response to riboflavin supplementation provides additional support that the SLC52A3 variant is clinically relevant. Although she received concurrent treatment with plasmapheresis, previous treatment with IVIG had no benefit.

Inheritance patterns of clinical RTD are complex. To date, most cases of RTD have been inherited in an autosomal recessive fashion, with far fewer case reports of symptomatic individuals with heterozygous variants. Manole et al. [11] reported six RTD patients with five different heterozygous SLC52A3 variants. They found that these patients were similar in phenotype to homozygous or compound heterozygous RTD patients with similar age of onset and symptoms; all six patients had sensorimotor neuropathy with weakness, five had sensorineural hearing loss, three had respiratory failure and three had dysphagia. Khani et al. [12] studied families of RTD patients and found variable presentations in family members possessing the same heterozygous SLC52A3 variant, with some remaining asymptomatic or demonstrating only mild hearing impairment, and others with significant disease burden. In the case of our patient, familial genetic testing revealed that the patient's older brother harbors the same heterozygous variant in SLC52A3. The patient's mother was negative for this variant. Unfortunately, the patient's father was unavailable for testing and clinical evaluation but is presumed to possess the same genetic variant as his children. Currently, the patient's brother remains asymptomatic but awaits formal audiology and ophthalmology evaluations. Of note, it is possible that our patient harbors another SLC52A3 variant that was undetected by WGS and SNP array, leading to biallelic inheritance of RTD, though this is unlikely given prior reports RTD patients with heterozygous pathogenic variants, as well as the lack of additional variants identified by the patient's comprehensive testing by WGS and SNP array.

The mechanisms behind this variable expressivity are unclear. It has been suggested that infectious and autoimmune factors may play a role. Several cases have been reported where onset of RTD symptoms coincided with infection or vaccination, with some responding at least transiently to immunotherapies [1,3,12]. Our patient experienced fevers prior to the onset of symptoms, providing some support that infection may trigger the onset of RTD in the setting of a genetic predisposition, while other family members, such as the patient's brother, remain asymptomatic.

There is limited data on the appropriate treatment regimen for patients with RTD. It has been reported that patients can have significant improvement with riboflavin supplementation with effective doses ranging from 10 to 80 mg/kg/day [13]. Given multiple reports of patients who remained stable but did not improve at lower doses of 10 mg/kg/day [2,4], we elected to start our patient on a higher dose of riboflavin supplementation at 40 mg/kg/day. Riboflavin has few side effects; therefore, it may be reasonable to initiate treatment at a high dose and reduce it only if adverse effects become intolerable.

This case provides additional evidence that RTD can be transmitted in an autosomal dominant fashion with incomplete penetrance and variable expressivity. It highlights the importance of continued re-analysis of genetic testing, particularly in the cases of variants of unknown significance with high clinical suspicion for disease. Additional research is needed to better understand the pathophysiology of RTD, the mechanisms for incomplete penetrance, and the optimal treatment regimen for this unique patient population. To our knowledge, at present there are no studies looking at the correlation between dose of riboflavin supplementation and clinical response. In addition, it remains unknown if asymptomatic individuals with pathogenic variants in SLC52A2 or SLC52A3 should be started on prophylactic riboflavin supplementation.

Author contributions

ET is the first author, wrote the initial draft of the manuscript, and did a literature review on the topic. BM assisted in the care of this patient as an outpatient, and helped edit the final draft of the paper and tables. AD guided further genetic testing for this patient, assisted in interpretation of genetic results, and helped edit the final draft of the paper. LE cared for the patient during her inpatient hospital stay, helped guide her initial workup and management, and helped edit the final draft of the paper. VR helped with the initial differential diagnosis and workup for this patient, and helped edit the final draft of the paper. NK assisted in the interpretation of EMG/NCS results and helped edit the final draft of the paper. AS is the primary outpatient physician for this patient, performed the EMG/NCS for this patient and assisted in the interpretation of the results, and supervised and assisted with the writing of this manuscript.

CRediT authorship contribution statement

Elizabeth S. Tranel: Investigation, Writing – original draft, Writing – review & editing. Bridget McGowan: Investigation, Writing – review & editing. Andy Drackley: Formal analysis, Writing – review & editing. Leon G. Epstein: Writing – review & editing, Investigation. Vamshi K. Rao: Writing – review & editing, Investigation. Nancy L. Kuntz: Writing – review & editing, Investigation. Abigail N. Schwaede: Investigation, Supervision, Writing – original draft, Writing – review & editing.

Declaration of competing interest

VKR has received grants from NS Pharma, RegenxBio, Alexion and Sarepta and nonfinancial support from Ann and Robert H. Lurie Children's Hospital for conduct of clinical trials. VKR has also received personal fees from Biogen, Avexis/Novartis, Capricor, NSPharma, Regenxbio, Genentech-Roche, Scholar Rock, PTC Therapeutics, Sarepta Therapeutics, France Foundation, and MDA outside the submitted work. NLK serves on Medical Advisory Boards and consults with Argenx, Astellas, Biogen, Novartis, Roche and Sarepta Therapeutics. Research funds have been provided directly to Lurie Children's hospital from Argenx, Astellas, Biogen, Novartis, Roche and Sarepta.

Data availability

No data was used for the research described in the article.