By Bernadette Toner
Mitochondrial disorders are among the most difficult diseases to diagnose accurately due to their complex genetic underpinnings and wide range of clinical phenotypes, but a number of labs have recently turned to next-gen sequencing to assist physicians in reaching a molecular diagnosis.
In the past few months, both Arup Laboratories and Baylor College of Medicine's Medical Genetics Laboratories launched new NGS-based tests for mitochondrial disorders. The two labs join a number of academic and commercial labs, including GeneDx and Transgenomic, that offer targeted panels or exome-based tests to detect mutations in the 37 genes present in mitochondria, nuclear genes that encode mitochondrial proteins, or both.
Baylor and Arup's new offerings differ slightly. Arup has launched a broad mitochondrial disorder panel that sequences the entire 16.5-kilobase mitochondrial genome and 108 nuclear genes. It has also developed a comparative genomic hybridization array to detect large duplications and deletions in the 37 mitochondrial and 108 nuclear genes.
The Mitochondrial Diagnostic Laboratory within Baylor's MGL, meantime, has expanded upon a mitochondrial genome sequencing service that it launched last year with a handful of targeted panels that focus on specific pathways involved in mitochondrial disease.
A Focused Approach
Earlier this month, Baylor launched a line of "MitomeNGS" panels related to mitochondrial disease and metabolic disorders: a nine-gene panel for muscle-related glycogen storage disorders; a 10-gene panel for liver-related glycogen storage disorders; a 14-gene mitochondrial depletion and integrity panel; a 25-gene panel for mitochondrial or metabolic myopathy/rhabdomyosis; a six-gene panel for progressive external ophthalmoplegia; and a four-gene panel for cholestasis.
Six additional MitomeNGS panels are in development and are slated for launch in June or July.
The lab uses custom NimbleGen sequence capture for target enrichment and performs all sequencing on the Illumina HiSeq. In validation studies, all panels achieved 100 percent sensitivity and specificity compared to Sanger sequencing.
Victor Zhang, assistant director of Baylor's Mitochondrial Diagnostic Laboratory, told Clinical Sequencing News that the panels are intended to complement the lab's existing tests. "We're trying to offer a wide spectrum of testing — from single genes for common mutations to panel testing to exome sequencing."
The new panels are designed for physicians who have been able to narrow down their diagnosis to a small subset of potential mitochondrial diseases based on phenotype, but still need help with a molecular diagnosis.
In such cases, single-gene testing would not provide enough information, but broader approaches such as whole-exome sequencing would be overkill.
"There are distinct phenotypes related to the particular genes in these panels," Zhang said. For example, if a patient presents with exercise intolerance, a physician might want to run the metabolic myopathy panel, which contains genes related to that phenotype. "Otherwise, for liver disease, they wouldn't consider those genes in the metabolic panel."
Cost is another factor. "Most clinicians are conscious about cost and try their best to order the most relevant test possible for their patients," Zhang said. "While large gene panels are useful, and exomes are useful as well, if a physician is able to narrow it down to what kind of subgroup, or what pathway it's going to be, it's more cost effective at this point to do that rather than going to broad exome sequencing."
Pricing for the tests will vary based on the client's needs so Zhang could not provide specifics, but he noted that the lab determined that a small panel such as the cholestasis test will save at least $2,000 when compared to ordering the four Sanger sequencing genes individually, while a larger panel like the metabolic myopathy test will save at least $25,000 when compared to ordering the 25 genes individually by Sanger sequencing.
As the lab expands its testing menu for mitochondrial disorders, it recognizes that the range of available options might pose a challenge to some physicians. With this in mind, it has developed a clinical testing algorithm for mitochondrial disorders that serves as a "cheat sheet" for physicians to determine the most appropriate course of testing.
"We understand the complexity of all this, so we're trying to make it easy," Zhang said. "We acknowledge that not all physicians will be the foremost expert in mitochondria, so we want to make sure we get all the information out to the people who need it."
The algorithm serves as a decision tree for physicians that starts with their suspected diagnosis based on phenotype and then recommends specific single-gene tests or panels that would likely lead to a molecular diagnosis. In cases where those options lead to a dead end, the lab recommends exome sequencing.
Casting a Wide Net
Arup, meantime, has taken a more comprehensive approach to sequencing-based mitochondrial disease testing.
"Because of the complexity of mitochondrial disease and the fact that there are many genes involved in these conditions, we are using next-generation sequencing technology to detect mutations in the mitochondrial genome and the nuclear genes at the same time," Rong Mao, medical director of molecular genetics at Arup Labs, explained in a presentation at Cambridge Healthtech Institute's Molecular Medicine Tri-Conference last month.
Mao noted that it's difficult to make a differential diagnosis for mitochondrial disease because patients have very variable phenotypes, even when the same mutations are involved. As a result, the lab developed a broad-based test that assesses both the mitochondrial genome and nuclear genes — combined with a CGH array to detect large duplications and deletions in those genes — in order to cast a wide net to detect causative mutations. It launched the test in early February.
The Arup team developed different protocols for sequencing the mitochondrial genome and nuclear genes.
For the mitochondrial genome, it begins with long-range PCR for DNA enrichment, followed by library prep on Beckman Coulter's SPRI-TE, sample barcoding and pooling, and single-end sequencing on the Illumina HiSeq. Minimum coverage is 200-fold in order to detect low levels of heteroplasmy.
Mao explained that the group amplifies the mitochondrial genome twice, using different sets of primers, because mtDNA is highly polymorphic. "If you're only using one set of primers, very often SNPs can cause allele dropout, so if you have two sets of primers in a different location you can avoid that," she said.
In a validation study using eight Coriell samples with known mutations, the mtDNA sequencing protocol was able to detect all variants and demonstrated the ability to detect heteroplasmy at levels below 10 percent, Mao said.
For the nuclear genes, the team uses RainDance for target enrichment and library prep, followed by single-read sequencing on the HiSeq. Minimum coverage is 50-fold.
The lab uses CLC Bio software to analyze both the mtDNA and nuclear sequence data.
Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at btoner [at] genomeweb [.] com.