Improving diagnosis of mitochondrial fatty-acid oxidation disorders

INTRODUCTION Mitochondrial Fatty Acid Oxidation Disorders (FAOD, Table 1, Fig. 1) include 12 genetically distinct metabolic disorders, inherited as an autosomal recessive trait, with an estimated cumulative incidence from 1:6,500 to 1:110,000 [1]. Their clinical presentation ranges from fatal acute hypoglycaemic crises in neonates to less severe later onset conditions characterised by myalgia and exercise intolerance. Symptoms differ for each, and phenotypic diversity extends even to patients bearing identical genetic variants [2]. In the most severe cases, neonatal presentation includes recurrent episodes of hypoketotic hypoglycaemic encephalopathy, liver dysfunction, often cardiac dysfunction, and sometimes congenital malformations. Therefore, it is important to rapidly implement emergency protocols for the acute management of metabolic crises [3]. Later onset FAOD are often associated with chronic symptoms such as hypotonia, exercise intolerance, or hepatic dysfunctions. Retinopathy and peripheral neuropathy are specific to long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) and mitochondrial trifunctional protein (MTP) deficiency [2, 4, 5]. Fasting or other physiologic stresses can lead to crises of muscle breakdown (rhabdomyolysis), and cardiac muscle damage [2]. These symptoms greatly impact quality of life. Moreover, left undiagnosed, patients of all ages are exposed to the risk of fatal metabolic decompensations [2]. Access to simple, effective management strategies and dietary therapies can greatly improve quality of life [3, 6]. New-born screening (NBS) programs improve early diagnosis of FAOD [1], but ethical and economic factors still play an important role in its implementation, limiting access in some locations [7]. As such, differential diagnosis of FAOD remains important, to initially recognise symptoms and later to account for possible normal biochemical testing that may occur when blood samples are collected from patients not in catabolism [8]. The objective of this manuscript is to indicate appropriate specimens to be collected and analysed, and to propose diagnostic algorithms according to symptoms. These algorithms are divided into three categories by age and are designed as stand-alone tools to facilitate the prompt diagnosis of mitochondrial FAOD by primary caregivers and thus accelerate the accurate referral of patients to specialists. Where necessary, differential diagnoses are indicated so that physicians may seek further resources. LABORATORY TESTING Table 2 outlines the main laboratory tests involved in FAOD diagnosis. Initially, routine blood analyses that can direct further testing should be conducted, e.g., anaemia in carnitine OCTN2 transporter deficiency (CTD). Routine testing, however, may appear normal unless samples were taken during an acute metabolic decompensation. At such times, hypoketotic hypoglycaemia may be observed, with increased levels of lactic acid, ammonia, and/or creatine kinase. Next, biochemical analysis of acylcarnitine profile by tandem mass spectrometry in blood specimens is central for the diagnosis of all FAOD and serves as the basis for NBS. Results from these assays allow to identify the predominant fatty acid derived metabolites and to characterise carnitine levels. It is again crucial to note that blood samples must be collected during an acute crisis or after fasting, as acylcarnitine profiles can normalise in anabolic state, leading to negative results [8]. High levels of free carnitine, especially in dried blood spots (DBS), should trigger consideration of carnitine palmitoyl transferase type IA deficiency (CPT IA). Very low levels of plasma carnitine (<5 μmol/L), contrasting with a urine carnitine level >5 mmol/mol of creatinine suggests CTD. For patients with low carnitine levels and non-specific acylcarnitine profiles, two options exist. L-carnitine supplementation can be given with subsequent testing of a new fasting blood sample, or blood cells (or eventually cultured skin fibroblasts) may be used for subsequent in vitro flux studies to measure the rate of fatty acid oxidation [9]. If the blood specimens have been collected post-mortem, acylcarnitine profiles can be non-informative, and could induce an incorrect diagnosis [8]. Thus, leftover samples collected during the acute decompensation, before death, should be sought to achieve highest confidence in the results of biochemical genetic tests. Acylcarnitine profiles can be diagnostic (Table 2) [8, 10] and indicate which gene(s) to analyse to further confirm the diagnosis. If next-generation sequencing (NGS) analyses are available, physicians may choose to advance rapidly to this step, and test several candidate genes simultaneously. Elsewhere, targeted Sanger sequencing of both alleles of suspected gene(s) remains a robust technique. When variants of uncertain significance are identified, in vitro flux studies or measurement of enzyme activities are necessary to provide definitive diagnosis.


INTRODUCTION
Mitochondrial Fatty Acid Oxidation Disorders (FAOD, Table 1, Fig. 1) include 12 genetically distinct metabolic disorders, inherited as an autosomal recessive trait, with an estimated cumulative incidence from 1:6,500 to 1:110,000 [1]. Their clinical presentation ranges from fatal acute hypoglycaemic crises in neonates to less severe later onset conditions characterised by myalgia and exercise intolerance. Symptoms differ for each, and phenotypic diversity extends even to patients bearing identical genetic variants [2]. In the most severe cases, neonatal presentation includes recurrent episodes of hypoketotic hypoglycaemic encephalopathy, liver dysfunction, often cardiac dysfunction, and sometimes congenital malformations. Therefore, it is important to rapidly implement emergency protocols for the acute management of metabolic crises [3].
Later onset FAOD are often associated with chronic symptoms such as hypotonia, exercise intolerance, or hepatic dysfunctions. Retinopathy and peripheral neuropathy are specific to long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) and mitochondrial trifunctional protein (MTP) deficiency [2,4,5]. Fasting or other physiologic stresses can lead to crises of muscle breakdown (rhabdomyolysis), and cardiac muscle damage [2]. These symptoms greatly impact quality of life. Moreover, left undiagnosed, patients of all ages are exposed to the risk of fatal metabolic decompensations [2]. Access to simple, effective management strategies and dietary therapies can greatly improve quality of life [3,6].
New-born screening (NBS) programs improve early diagnosis of FAOD [1], but ethical and economic factors still play an important role in its implementation, limiting access in some locations [7]. As such, differential diagnosis of FAOD remains important, to initially recognise symptoms and later to account for possible normal biochemical testing that may occur when blood samples are collected from patients not in catabolism [8].
The objective of this manuscript is to indicate appropriate specimens to be collected and analysed, and to propose diagnostic algorithms according to symptoms. These algorithms are divided into three categories by age and are designed as stand-alone tools to facilitate the prompt diagnosis of mitochondrial FAOD by primary caregivers and thus accelerate the accurate referral of patients to specialists. Where necessary, differential diagnoses are indicated so that physicians may seek further resources. Table 2 outlines the main laboratory tests involved in FAOD diagnosis. Initially, routine blood analyses that can direct further testing should be conducted, e.g., anaemia in carnitine OCTN2 transporter deficiency (CTD). Routine testing, however, may appear normal unless samples were taken during an acute metabolic decompensation. At such times, hypoketotic hypoglycaemia may be observed, with increased levels of lactic acid, ammonia, and/or creatine kinase.

LABORATORY TESTING
Next, biochemical analysis of acylcarnitine profile by tandem mass spectrometry in blood specimens is central for the diagnosis of all FAOD and serves as the basis for NBS. Results from these assays allow to identify the predominant fatty acid derived metabolites and to characterise carnitine levels. It is again crucial to note that blood samples must be collected during an acute crisis or after fasting, as acylcarnitine profiles can normalise in anabolic state, leading to negative results [8].
High levels of free carnitine, especially in dried blood spots (DBS), should trigger consideration of carnitine palmitoyl transferase type IA deficiency (CPT IA). Very low levels of plasma carnitine (<5 µmol/L), contrasting with a urine carnitine level >5 mmol/mol of creatinine suggests CTD. For patients with low carnitine levels and non-specific acylcarnitine profiles, two options exist. L-carnitine supplementation can be given with subsequent testing of a new fasting blood sample, or blood cells (or eventually cultured skin fibroblasts) may be used for subsequent in vitro flux studies to measure the rate of fatty acid oxidation [9].
If the blood specimens have been collected post-mortem, acylcarnitine profiles can be non-informative, and could induce an incorrect diagnosis [8]. Thus, leftover samples collected during the acute decompensation, before death, should be sought to achieve highest confidence in the results of biochemical genetic tests.
Acylcarnitine profiles can be diagnostic (  Severe cases often present as acute and sometimes fatal crises of hypoketotic hypoglycaemia, associated with hepatic failure, hyperammonaemia (Reye-like syndrome) and/or cardiac symptoms (arrhythmia, cardiomyopathy), and possibly congenital malformations (mainly renal cysts and neuronal migration defects). Other cases present with milder, non-specific symptoms such as gross motor or language delay, and, over time, hypotonia and failure to thrive. Fasting and protein-induced hypoglycaemia with hyperinsulinism points to short-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (SCHAD). Specimens should be collected during an acute crisis or after fasting for analysis of the blood acylcarnitine profile and urine organic acids to differentiate organic acidaemias. For patients with normal initial blood acylcarnitine profiles, medical history can provide additional insight. Maternal haemolysis, elevated liver enzymes and low platelets (HELLP) syndrome or acute fatty liver of pregnancy (AFLP) are seen especially in mothers carrying a foetus affected with MTP/LCHAD, and this history should trigger additional studies for these disorders including in vitro flux studies, or directly molecular genetic testing, depending on local facilities.

CHILDHOOD, FROM AGE 2-11 YEARS (FIG. 3)
Affected children <5 years of age can present severe hypoketotic hypoglycaemia and/or cardiac symptoms. Differential diagnosis in this setting is the same as for younger children.
More typically, patients >4-6 years old present muscular symptoms. Exercise intolerance, with episodic rhabdomyolysis is common. Liver dysfunction may be present, and in some cases hypoglycaemic crises are triggered by prolonged fasting, cold exposure, strenuous physical exercise, or intoxication (e.g., accidental ingestion of alcohol). Finally, chorioretinopathy and peripheral neuropathy may be present in MTP/LCHAD, and can be the presenting and only symptom [4,6]. Patients with at least one c.1528G > C allele in the HADHA gene, commonly develop severe retinopathy [5].
Long-chain FAOD (LC-FAOD) should be considered in any patient with cardiomyopathy, especially a new and acute onset. Here, biochemical testing of in vitro FAO flux or directly molecular genetic testing will allow a diagnosis to be made. Note that lactic acid is often mildly elevated in LC-FAOD during episodes of acute metabolic decompensation. For patients presenting with myopathic or neurological symptoms, plasma creatine kinase levels should be measured. (FIG. 4) Cases of late-onset FAOD often present with myopathic symptoms. The most frequent symptoms include exercise-triggered myalgia, rhabdomyolysis, cardiomyopathy, after prolonged fasting (>14 hours) or physiologic stress. Excessive consumption of alcohol during short period of time can also induce symptoms. Hypoglycaemia is less common than in younger patients. Retinopathy or sensory-motor axonal neuropathy are suggestive of MTP/LCHAD.    Whole exome or genome sequencing (ultra-high throughput sequencing) C 0 free carnitine, CACT carnitine acylcarnitine translocase deficiency, CPT IA carnitine palmitoyl transferase IA deficiency, CPT II carnitine palmitoyl transferase II deficiency, CTD carnitine OCTN2 transporter deficiency, FAD flavin adenine dinucleotide, FAOD fatty acid oxidation disorders, LCHAD long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency, MAD multiple acyl-CoA dehydrogenase deficiency (ETF or ETF-QO deficiency), MCAD medium-chain acyl-CoA dehydrogenase deficiency, MFT mitochondrial FAD transporter, MTP mitochondrial trifunctional protein deficiency, SCAD short-chain acyl-CoA dehydrogenase deficiency, SCHAD short-chain 3-hydroxyacyl-CoA dehydrogenase deficiency, ↑ indicating abnormally elevated results from these blood tests, ↓ indicating abnormally decreased results from these blood tests. The acylcarnitines in bold characters are the most clinically relevant species for each disorder.  Table 1.

CHILDREN > 11 YEARS OLD AND ADULTS
If the acylcarnitine profile is normal with elevated plasma creatine kinase levels in vitro flux studies or molecular genetic testing are helpful to reach a diagnosis. Muscle biopsies are nondiagnostic in FAOD.

POST-DIAGNOSIS
Prompt diagnosis ensures immediate referral of patients to specialists and provides access to appropriate care and dietary management, thus reducing the risk of acute metabolic crises, while alleviating the chronic symptoms associated with FAOD. For example, intake of more frequent meals emphasising complex carbohydrates during illness, stress or increased activity, prevents acute decompensations in most FAOD [2]. Beyond dietary modifications, LC-FAOD (VLCAD [very long-chain acyl-CoA dehydrogenase], MTP/LCHAD, CACT [carnitine acylcarnitine translocase], CPT IA, CPT II) can be effectively and safely treated with an anaplerotic drug, triheptanoin. Triheptanoin has been shown to reduce episodes of myalgia, rhabdomyolysis, cardiomyopathy, and hypoglycaemia, along with emergency hospitalisations, leading to marked improvement in quality of life, but is not effective on retinopathy and neuropathy of LCHAD/MTP [6]. L-carnitine supplementation in CTD provides complete relief from symptoms. It may also be necessary in patients with severe secondary carnitine deficiency. SCHAD deficiency responds well to diazoxide [2].
Finally, genetic counselling can be proposed to families of an affected patient. Prenatal diagnosis is possible for all FAOD in chorionic villi or amniocytes, using molecular genetic testing as the preferred technique [2].

CONCLUSION
The implementation of NBS for FAOD in more and more countries will most probably improve their diagnosis, but many patients still do not benefit from NBS. Moreover, it has been documented that NBS programs can miss diagnoses and that some patients can be symptomatic before the results of NBS are available [1,4].
Considering this challenge, we hope that the presented tools will empower general practitioners working in a variety of settings to accelerate differential diagnosis for this group of metabolic disorders. More rapid FAOD diagnoses will improve the health and quality of life of patients, all of whom stand to benefit from the care available at specialised centres where their treatment may be adapted over the course of their lives. Even late onset, cryptic presentations of FAOD leave patients susceptible to potentially fatal metabolic crises. Through the reduction of hospitalisation rates due to acute metabolic decompensations, prompt and accurate referrals will also reduce burdens on local hospitals and associated costs to patients who do not receive appropriate treatment.  Table 1. Note that molecular testing should be preferred to muscle biopsy when available.
conference presentations and honoraria for participating in their Scientific Working Group. CVS is also Scientific Advisor (an unpaid role) for the ERNDIM (European Research Network for evaluation and improvement of screening, Diagnosis and treatment of Inherited disorders of Metabolism). AF declares that his institution has received funding from Amicus and Sanofi-Genzyme. AF has also received honoraria for participation on advisory boards from Amicus. JV declares that he has in the past received research funding from Ultragenyx and Reneo Pharmaceuticals. JV has also received payment or honoraria from CheckRare CME, Excel CEM, PTCE-PER CME, Med-IQ CME and Horizon Pharmaceuticals CME. JV plays are leadership from in SIMD NAMA and holds stocks in American Gene Therapies. CAB declares that she has received honoraria in the past for participating in scientific working groups for Utragenyx and Alnylam, and is unpaid board member of ANPGM, the French Association of Genetic Molecular Practitioners. NG declares that her institution has been the recipient of grants for clinical trials from Sanofi-Genzyme, Takeda and Chiesi. NG has received support for attending scientific meetings from Ultragenyx, Sanofi-Genzyme and Chiesi, as well as honoraria for participation on data safety monitoring and advisory boards from Ultragenyx, Sanofi-Genzyme, Chiesi and Biomarin.