NEW YORK (GenomeWeb) – Researchers from Baylor College of Medicine and the Children's Hospital of Philadelphia have developed a next-generation sequencing panel for mosaic overgrowth syndromes that can be used both pre- and post-natal.
The team described the eight-gene panel this week in the Journal of Molecular Diagnostics, validating it in 50 consecutive clinical samples, including two prenatal samples. The researchers reported that the NGS panel was more sensitive than Sanger sequencing, since it could detect pathogenic mutations at a frequency of less than 2 percent.
Mosaic overgrowth syndromes can be characterized by a number of features, including variable asymmetry; developmental delay; and vascular, epidermal, and skeletal anomalies. In addition, individuals tend to be more susceptible to tumor development.
Recently, somatic variants in the PI3K/AKT/mTOR pathway have been reported as causative for some overgrowth syndromes. That pathway is also known to be associated with various forms of cancers and there are therapies that target the pathway. Identifying variants using Sanger sequencing, however, has proven to be challenging because many of the disorders are mosaic, with pathogenic variants only in the affected tissue. Sanger sequencing can only detect variants at around a 15 percent to 20 percent frequency or higher, while NGS can often detect variants at a 1 percent frequency.
The researchers designed a 17-amplicon NGS panel that analyzed eight genes using Thermo Fisher Scientific's Ion AmpliSeq technology and ran it on the Ion PGM. For five genes — AKT1, AKT2, AKT3, PIK3R2, MTOR—the panel targeted known pathogenic mutations. The researchers also included all exons of the PIK3CA gene, as well as four known variants in the GNAS gene associated with McCune-Albright syndrome and the entire PTEN gene.
To validate the panel, the team ran it on eight samples, five with known pathogenic variants and three normal controls. The NGS panel detected all known variants, with the lowest frequency variant being present at less than 2 percent frequency. All known normal samples were negative.
The researchers then tested the panel on 50 patient specimens, running it on a variety of sample types, including blood, affected and unaffected tissues, amniotic fluid, and cultured amniotic fluid. In total, they identified a pathogenic variant in 28 out of the 50 samples, including 25 samples with a pathogenic variant in the PIK3CA gene, two with PIK3R2 mutations, and one sample with a pathogenic variant in AKT1. Allele frequencies ranged from 1 percent to 49 percent.
In most of the cases where both affected and unaffected tissue were collected, pathogenic variants were only identified in affected tissue, "demonstrating the somatic nature of most variants," the authors wrote. However, there were four cases in which pathogenic variants were also identified in unaffected tissue, but at lower allele frequencies. The authors concluded that either sample contamination occurred, or "variants might be present in a larger area than what was initially assessed."
In one of the prenatal samples, the researchers ran the panel because an ultrasound had identified multiple congenital anomalies. The team ran the panel on cultured amniotic fluid and identified a pathogenic variant in PIK3CA at an allele frequency of 38 percent. The variant was not identified in either postnatal blood or maternal blood samples by Sanger sequencing.
In the second prenatal sample, a balanced translocation had been identified affecting the X chromosome, and ultrasound had showed a foot abnormality. The overgrowth panel identified a pathogenic mutation in the PIK3CA gene from affected tissue specimens that was confirmed by Sanger. Normal tissue and direct amniotic fluid were both negative.
The authors reported that for these types of disorders, that are caused by somatic mutations and are mosaic, an NGS panel could have benefits over both Sanger sequencing as well as broader exome or whole-genome sequencing. Sanger sequencing often does not have the resolution to detect low-frequency variants and it cannot analyze multiple genes at once. But, while exome and whole-genome sequencing are "ideal for new discoveries and detecting genetic defects for patients with unknown genetic diseases," the authors wrote, their "ability to detect low-level mosaic sequence variants is limited because of the low sequencing depth."
For instance, they described one patient who was suspected of having an overgrowth syndrome or tumor of unknown characteristics and was referred to exome testing. The exome test was negative, but the panel identified pathogenic variant in the PIK3CA gene present at a 4.2 percent frequency, which is below the lab's limit of detection of 5 percent for exome sequencing.
The researchers were also more likely to identify causal variants in patients for whom they analyzed affected tissue, as opposed to a blood sample. Of the 37 cases for which they had affected tissue, they identified a causal variant in 24, or around 65 percent. By contrast, for 13 cases from which they tested a blood sample, they only identified pathogenic mutations in four, or 31 percent.
The mutations identified also suggest a potential therapeutic option, the researchers noted. Currently, most overgrowth syndromes are treated by surgery, which are often only temporary remedies and can lead to large lesions. But, the "causal role of the pathogenic variants in the PI3K/AKT/mTOR signaling pathway bring to light the possibility of using PI3K/AKT/mTOR inhibitors for the treatment of these devastating disorders," they wrote.