Predicting Prognosis in CPEO With mtDNA Deletions: A Case Demonstrating the Advantages of Measuring Heteroplasmy With Novel Droplet Digital Polymerase Chain Reaction Testing

Clinical Correspondence Section Editors: Robert Avery, DO Karl C. Golnik, MD Caroline Froment, MD, PhD An-Guor Wang, MD Predicting Prognosis in CPEO With mtDNA Deletions: A Case Demonstrating the Advantages of Measuring Heteroplasmy With Novel Droplet Digital Polymerase Chain Reaction Testing Nathan A. Lambert-Cheatham, DO, Sophia T. Tessema, MD, MPH, Obada Subei, MD, Ragha C. Sakuru, MD, Matthew D. Fullmer, DO, Elizabeth M. Selner, MS, LCGC, Noemi Vidal-Folch, MS, Howard T. Chang, MD, PhD, Linda Hasadsri, MD, PhD, David I. Kaufman, DO M Address correspondence to Nathan Lambert-Cheatham, DO, Department of Neurology and Ophthalmology, Michigan State University, 804 Service Drive, Suite A217, East Lansing, MI 48824; E-mail: [e-mail] Some of these challenges in correlating genotype with phenotype may be explained by the limited ability of certain testing methodologies to quantify heteroplasmy levels of mtDNA deletions. Older forms of testing, such as multiplex ligation-dependent probe amplification and Southern blot, cannot detect deletions at low heteroplasmy levels. As a result, these methodologies have been increasingly replaced by nextgeneration sequencing (NGS) coupled with long-range polymerase chain reaction (LR-PCR) or real-time quantitative PCR (qPCR). However, these approaches involve a PCR step that is subject to “PCR bias.” Namely, DNA molecules with large deletions are shorter in size and therefore are preferentially amplified, “outcompeting” full-length DNA strands in traditional PCR reactions. This leads to exponential differences in amplification and can cause significant overestimation of heteroplasmy levels (4,5). This, in turn, can lead to misdiagnoses that affect prognostication and patient care. Recently, a new technique called droplet digital polymerase chain reaction (ddPCR) has begun to be used in mitochondrial genetic testing. ddPCR does not show much PCR bias because each molecule of DNA is amplified in its own separate droplet, independent from other DNA molecules. This overcomes errors of amplification efficiency and bias. Studies have shown ddPCR to give more reliable and consistent results (4,5). Below, we present a case of a patient with CPEO-plus with a relatively benign phenotype, despite the initial test result showing a large homoplastic (100% heteroplasmy level) mtDNA deletion. Repeat testing using ddPCR revealed errors in the patient’s previous testing and thus explained her clinical course and prognosis. A 76-year-old woman presented with progressive intermittent binocular diplopia and ptosis. The ptosis was first noted by her husband 8 years prior. Her first occurrence of diplopia was reported 5 years prior. Examination revealed bilateral ptosis and ophthalmoplegia with decreased supraduction, abduction, and adduction (Fig. 1). Oculocephalic reflex did not improve extraocular movements. Motor e260 Lambert-Cheatham et al: J Neuro-Ophthalmol 2023; 43: e260-e263 itochondrial DNA (mtDNA) mutations can present clinically as various syndromes, one of the most common of which is chronic progressive external ophthalmoplegia (CPEO), characterized by ptosis and limitations in extraocular movements. The symptoms and severity of CPEO varies markedly between patients. The age of onset also varies from pediatric to as late as the eighth decade of life. The term “CPEO-plus” has been coined to describe a large range of additional symptoms that can present in patients with CPEO, including retinopathies, heart conduction block, endocrinopathies, sensorineural hearing loss, and generalized muscle dysfunction (1,2). The phenotype of CPEO and many other mitochondrial syndromes is believed to result from mutations affecting the oxidative phosphorylation pathway. Thus, tissues with high metabolic rates, such as the extraocular muscles, are preferentially affected. Many studies have shown the phenotype/ prognosis of patients to correlate with different mtDNA mutations. However, owing to different heteroplasmy levels (percent of cells/mitochondria in a tissue carrying the mutation), patients with an identical mtDNA mutation often present with very different clinical phenotypes, posing significant challenges to predicting a patient’s prognosis/phenotype based on genetic testing (1–3). Large deletions in mtDNA, for instance, can cause CPEO, Kearns–Sayre syndrome, Pearson syndrome, or even Leigh syndrome depending on the tissue type(s) or cells affected and the mutation load within those cells. Department of Neurology and Ophthalmology (NL-C, ST, RCS, HTC, DK), Michigan State University, East Lansing, Michigan; Department of Neurology, Banner Neuroscience Institute (OS), University of Arizona, Phoenix, Arizona; Department of Ophthalmology (MF), Beaumont Hospital, Taylor, Michigan; Department of Pathology (HTC), Sparrow Health System, Lansing, Michigan; and Department of Laboratory Medicine and Pathology (ES, NV-F, and LH), Genomics Laboratory, Mayo Clinic, Rochester, Minnesota. The authors report no conflicts of interest. Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence FIG. 1. Images of patient’s extraocular motility and eyelid ptosis on follow-up examination. The upper panel shows extraocular motility in primary gaze, up gaze, left gaze, right gaze, and down gaze. The bottom panel shows ptosis when relaxed and during maximum attempted eyelid elevation. examination revealed mild gait ataxia and mild bilateral extremity weakness. The visual acuity was 20/30 in both eyes. Fundus examination was significant for dry macular degeneration in both eyes. Her remaining neuroophthalmic examination was unremarkable. Testing for serum inflammatory markers, creatine phosphokinase, and lactic acid levels were within normal limits. MRI orbits revealed mildly atrophic extraocular muscles. A single-fiber electromyogram of her frontalis showed increased jitter and blocking. A quadriceps muscle biopsy showed only a few ragged red fibers that were within normal limits, given her age. Her muscle biopsy and a blood specimen were sent for a whole mitochondrial genome analysis by NGS (involving LR-PCR amplification before NGS). Sequencing of blood yielded normal results. The muscle specimen testing, however, reported a homoplasmic 7.2 kb deletion from mt. 5,790 to 13,000 (Fig. 2A, B). This result was reproducible on repeat LR-PCR testing. Based on the severity of her mitochondrial mutation in this initial NGS study, there was concern for more systemic involvement (e.g., Kearns–Sayre syndrome). The patient was evaluated by cardiology which was significant for atrial fibrillation without signs of conduction block. She was evaluated by endocrinology which was significant for a history of hyperthyroidism. Fundus examination revealed no obvious pigmentary retinopathy, although there were signs of geographic atrophy. Interestingly, full-field electroretinography (ERG) demonstrated bilateral reduced cone function and borderline rod function. Over 7 years of follow-up, the patient developed further limitations in extraocular motility. Her visual acuity declined significantly to CF in the right eye and 20/400 in the left eye, likely from her geographic atrophy. Her diplopia initially worsened before resolving secondarily to her reduced visual acuity. Her gait ataxia and extremity weakness remained only mildly reduced for her age. She developed sensorineural hearing loss potentially related to her mitochondrial deletion. Otherwise, her symptoms were stable on follow-up examinations. The patient’s relatively benign clinical course did not seem to correlate with the reported severity of her mitochondrial mutation. Furthermore, a homoplasmic mutation implied that all muscle fibers should carry identical mutations, and thus, presumably all or at least a higher Lambert-Cheatham et al: J Neuro-Ophthalmol 2023; 43: e260-e263 e261 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence FIG. 2. A. Representative diagram of the patient’s mitochondrial mutation from position 5,790 to 13,000. LR-PCR amplicons (B) and genes probed in ddPCR (C and D) are labeled. The diagram also shows approximate positions of relevant genes and mutations associated with other conditions, including Kearns–Sayre syndrome (KSS); Leber hereditary optic neuropathy (LHON); and mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS). B. Gel electrophoresis of 2 LR-PCR amplicons (DNA segments being amplified) performed on our patient’s muscle and blood samples. Amplicon 1 (LR1) does not contain the mutation of interest, thus would be expected to have only 1 band corresponding to the full-length product. Amplicon 2 (LR2) contains the mutation of interest, thus 2 bands are expected which represent the full-length DNA product and the shorter mutated product. Notably, there was no full-length mtDNA band for the patient’s muscle sample, suggesting a homoplastic mutation. This result was reproducible on repeat LR-PCR testing. No band corresponding to the mutation was found on blood sample testing. This highlights the importance of tissue-specific heteroplasmy and why blood samples are not as sensitive as muscle samples for CPEO genetic testing. Laboratory testing was performed in the CLIAcertified, CAP-accredited genomics laboratory at Mayo Clinic. C. ddPCR result demonstrating a large mtDNA deletion present at 15.2%–21.4% heteroplasmy. A laboratory-developed quantitative ddPCR assay was performed on the Bio-Rad QX200 Droplet Digital PCR system using mtDNA extracted from the frozen muscle tissue. The sample is partitioned into 20,000 nanoliter-sized droplets, and targets are amplified by end point PCR in each droplet, followed by automated measurement of the fluorescence of each droplet in 2 channels. The x axis represents the fluorescence amplitude of the probe for the MTRNR1 gene (green cluster), which lies outside of the deletion breakpoints. The y axis shows the fluorescence amplitude of the probe for MT-ND4 (blue cluster), which is within the deleted segment of mitochondrial DNA. Bio-Rad QuantaSoft Analysis Pro software was used to measure the number of positive and negative droplets for each fluorophore and to quantify the number of copies of MT-RNR1 and MT-ND4 per 20 mL as well. Heteroplasmy is calculated by determining the ratio between MT-ND4 and MT-RNR1 copies/mL. ddPCR testing was performed in the CLIA-certified, CAP-accredited genomics laboratory at Mayo Clinic. ddPCR results were confirmed with whole genome sequencing. D. ddPCR results of comparison patient with an 85% heteroplasmy level mutation. Compared with our patient’s results (C), it can be seen that the blue droplets (MT-ND4+) are significantly fewer than the green droplets (MT-RNR1+), indicating an MT-ND4 deletion with a greater heteroplasmy level. proportion of fibers should have appeared histologically as “ragged red fibers,” an inference obviously incongruent with only rare ragged red fibers seen in her biopsy. These correlation discrepancies raised concerns for the bias error in the original genetic testing. We decided to test her muscle specimen using a novel ddPCR-based assay that revealed a reduced heteroplasmy level of 15.2%–21.4% (Fig. 2C). This new result was confirmed by whole genome sequencing (WGS), which does not involve a PCR amplification step before NGS, and was also independently verified by an external reference laboratory using qPCR. This case highlights the advantages of ddPCR testing and the potential drawbacks of other techniques reliant on a traditional PCR amplification step or techniques using a more qualitative approach to the determination of heteroplasmy. As shown in our case, novel ddPCR may provide more accurate genetic testing results that better predict the prognosis of patients. Further studies are e262 Lambert-Cheatham et al: J Neuro-Ophthalmol 2023; 43: e260-e263 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Clinical Correspondence necessary to elucidate the full prognostic potential of ddPCR technology. STATEMENT OF AUTHORSHIP Conception and design: N. Lambert-Cheatham, S. Tessema, O. Subei, H. Chang, D. Kaufman, L. Hasadsri; Acquisition of data: N. Lambert-Cheatham, S. Tessema, O. Subei, M. Fullmer, H. Chang, R. C. Sakuru, E. Selner, N. Vidal-Folch; Analysis and interpretation of data: N. Lambert-Cheatham, E. Selner, N. Vidal-Folch, H. Chang, L. Hasadsri, D. Kaufman. Drafting the manuscript: N. Lambert-Cheatham, O. Subei; Revising it for intellectual content: E. Selner, N. Vidal-Folch, H. Chang, L. Hasadsri. Final approval of the completed manuscript: N. Lambert-Cheatham, S. Tessema, O. Subei, R. C. Sakuru, M. Fullmer, H. Chang, E. Selner, N. VidalFolch, L. Hasadsri, D. Kaufman. Lambert-Cheatham et al: J Neuro-Ophthalmol 2023; 43: e260-e263 REFERENCES 1. Yamashita S, Nishino I, Nonaka I, Goto YI. Genotype and phenotype analyses in 136 patients with single large-scale mitochondrial DNA deletions. J Hum Genet. 2008;53:598–606. 2. Heighton JN, Brady LI, Sadikovic B, Bulman DE, Tarnopolsky MA. Genotypes of chronic progressive external ophthalmoplegia in a large adult-onset cohort. Mitochondrion. 2019;49:227–231. 3. Auré K, Ogier de Baulny H, Laforêt P, Jardel C, Eymard B, Lombès A. Chronic progressive ophthalmoplegia with large-scale mtDNA rearrangement: can we predict progression? Brain. 2007;130:1516–1524. 4. Taylor SC, Laperriere G, Germain H. Droplet Digital PCR versus qPCR for gene expression analysis with low abundant targets: from variable nonsense to publication quality data. Scientific Rep. 2017;7:1–8. 5. Belmonte FR, Martin JL, Frescura K, Damas J, Pereira F, Tarnopolsky MA, Kaufman BA. Digital PCR methods improve detection sensitivity and measurement precision of low abundance mtDNA deletions. Scientific Rep. 2016;6:1–11. e263 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.