Molecular pathology selected abstracts

Create PDF

Editors: Donna E. Hansel, MD, PhD, chief, Division of Anatomic Pathology, and professor, Department of Pathology, University of California, San Diego; James Solomon, MD, PhD, resident, Department of Pathology, UCSD; Richard Wong, MD, PhD, molecular pathology fellow, Department of Pathology, UCSD; and Sounak Gupta, MBBS, PhD, molecular pathology fellow, Memorial Sloan Kettering Cancer Center, New York.

Using circulating cell-free fetal DNA to test for monogenic disorders

March 2019—Screening for fetal chromosomal abnormalities can be performed by noninvasive methods in which fetal cell-free DNA (cfDNA) circulating in the maternal blood is isolated and analyzed. However, the standard of care for screening for monogenic diseases remains population-based carrier screening—testing the parents for their carrier status of deleterious genes. This carrier screening algorithm is useful for assessing the risk of some recessive genetic disorders that could be transmitted to offspring, but it is not effective for detecting common congenital diseases, such as some skeletal dysplasias that are often caused by a dominant de novo mutation. The authors of this study designed a novel noninvasive prenatal screening method that uses cfDNA to screen for such common monogenic disorders. After extracting and isolating the cfDNA from maternal plasma, the authors used unique molecular indexing to reduce noise created by polymerase chain reaction and sequencing errors. In this method, a unique molecular index (UMI) consisting of a six base-pair molecular barcode was attached to each molecule. After hybridization capture to enrich the library for genes of interest, the library was sequenced to a depth such that each individual UMI-labeled molecule was likely to be sequenced multiple times. These multiple reads could then be deduplicated to create a consensus read, drastically reducing noise and improving the accuracy of variant calls. This method is especially useful when the variant is present at a very low frequency, which is often the case when examining cfDNA. The bioinformatic pipeline also incorporated a statistical method that used the fraction of fetal DNA and sequence read counts to further filter variant calls. The authors examined a clinical cohort of 458 samples from 131 clinics in the United States, Europe, and Asia; of these, 422 samples met the criteria for analysis. At collection, the average gestational age was 16.8 weeks, and the patient population included advanced maternal or paternal age, abnormal ultrasound findings, abnormal serum screening, or family history of genetic disease. Thirty-two cases in this clinical cohort were positive, and many of the pathogenic variants were detected in patients in whom an abnormal fetal phenotype was identified by ultrasound. For example, de novo pathogenic variants were detected in the genes COL1A1, COL1A2, and FGFR3 in patients who exhibited skeletal abnormalities, and alterations in TSC2 were detected in two patients for whom cardiac rhabdomyomas were seen on ultrasound. Clinical follow-up was available for a subset of the cases. All of the positive cases were confirmed with prenatal invasive testing, postnatal genetic testing, or clinical examination after birth. Similarly, no false-negative cases were seen. The authors concluded that this study demonstrates the utility of a noninvasive assay to screen for monogenic disorders. Such an assay would be extremely useful for patients with high-risk pregnancies or abnormal ultrasound findings.

Zhang J, Li J, Saucier JB, et al. Non-invasive prenatal sequencing for multiple Mendelian monogenic disorders using circulating cell-free fetal DNA. Nature Med. 2018. doi: 10.1038/s41591-018-0334-x. Published January 28, 2019.

Correspondence: Dr. Jinglan Zhang at jinglanz@bcm.edu

Liquid biopsy of cerebrospinal fluid to monitor gliomas

Central nervous system gliomas, the most common malignant brain tumor in adults, are classified by histologic features, but the current standard of care also requires molecular classification. Features such as IDH mutation status and co-deletion of chromosome arms 1p and 19q are crucial for prognostic classification and to guide treatment. However, the process of obtaining tissue from the central nervous system (CNS) is extremely invasive and involves significant risk to the patient. Although circulating tumor DNA (ctDNA) from CNS tumors is rarely present in plasma, examining the cell-free DNA in cerebrospinal fluid (CSF) obtained via lumbar puncture could be a minimally invasive way of monitoring tumor evolution and guiding therapy. The authors conducted a study of such a technique in which they examined 85 patients with primary CNS gliomas. The patient population included 46 patients with glioblastoma (WHO grade IV) and 26 and 13 patients with WHO grade III and WHO grade II gliomas, respectively. All patients had previously undergone surgical resection, and the vast majority had also received radiation and systemic chemotherapy. CSF was collected from the patients when a clinical indication for lumbar puncture arose, such as suspicion of infection, increased intracranial pressure, or leptomeningeal tumor spread. The CSF samples were analyzed using a targeted next-generation sequencing assay. CtDNA with at least one tumor-derived genetic variant was detected in 42 of the 85 (49.4 percent) patients. By comparison, ctDNA was seen in plasma in only three of 19 (15 percent) patients tested, often at very low levels, confirming that CSF is a better source of glioma ctDNA. As a control experiment, the authors also examined CSF from patients with non-neoplastic neurological conditions. No oncogenic variants were detected in these samples. Radiographic findings of tumor progression, high tumor burden, and leptomeningeal or subependymal spread were associated with ctDNA in the CSF. The ctDNA in the CSF was associated with worse overall survival in a multivariate analysis that included tumor burden, extent of resection, and IDH status. The alterations seen in the ctDNA included protein-coding mutations within TP53 or IDH1 genes, copy number alterations such as EGFR amplification, promoter mutations affecting the TERT, and structural rearrangements such as the EGFR-variant III deletion. In 36 of the 42 (85.7 percent) patients with ctDNA in CSF, the findings could be compared with somatic variants seen in a prior tissue biopsy. While the findings were similar in most cases, a few cases exhibited differences in the ctDNA suggestive of tumor evolution. For example, in five patients, all of whom received temozolomide, a remarkably higher tumor mutation burden was seen in the ctDNA compared with findings in the tissue biopsy. Overall, analysis of ctDNA from CSF appears to be a viable noninvasive strategy for assessing the molecular landscape of CNS tumors to guide therapy and monitor tumor evolution.