By Michael McCathy, MD

The Brotman Baty Institute for Precision Medicine (BBI) launched its cell-free DNA (cfDNA) working group in 2018, awarding grants for more than a dozen pilot studies to apply a cfDNA assay developed and validated by BBI to address a wide variety of  clinical challenges, ranging from prenatal screening, cancer diagnosis to detection of transplant rejection. In a recent symposium, the projects’ investigators provided updates on their progress to date.

Dr. Christina Lockwood. Photo courtesy of UW Medicine.

The working group is led by Christina Lockwood, Ph.D. associate professor, UW Department of Laboratory Medicine and director, Genetics and Solid Tumor Diagnostics Laboratory; Steve Henikoff, Ph.D., member of the Basic Sciences Division at the Fred Hutchinson Cancer Research Center and an affiliate professor of genome sciences at the University of Washington; and Sarah Leary, MD associate professor in the UW Division of Pediatric Hematology/Oncology.

cfDNA is made up of free-floating fragments of DNA that can be found in blood, cerebral spinal fluid, urine and other bodily fluids. The DNA, which are usually, double-stranded fragments of about 150 to 300 base pairs in length, is primarily released into the circulation by dying cells as part of natural cell turnover. Generally, the amount found in plasma is very low, though it often rises when there is an increase in cell death, such as occurs as the result of tissue injury and inflammation. Tumor cells, which have a high rate of cell death, can also contribute to the amount of cfDNA found in plasma. This form of cfDNA is called circulating-tumor DNA or ctDNA. ctDNA is one component of so-called “liquid biopsies,” which allow for the diagnosis of cancers and cancer recurrence from blood samples.

cfDNA from patients can be sequenced and analyzed to identify things as point mutations, gene fusions, copy number variation, methylation patterns and chromosomal abnormalities. Discovered in 1948, cfDNA is now routinely used in prenatal screening. The goal of the BBI projects is to refine how cfDNA is collected, processed and analyzed and explore how it can be used to improve care for a variety of diseases.

1. Dying cells release fragments of their DNA into the circulation 2. Free-floating cell-free DNA is extracted from the blood plasma 3. The cell-free DNA is then sequenced making it possible to identify point mutations, copy number variations, chromosomal rearrangements and other genetic variations.

Prenatal screening

Over the past two years, Lockwood and her BBI colleagues have been developing a cfDNA screening assay for prenatal testing, that can be run independent of any outside lab. Lockwood says one advantage of such an assay is that it offers BBI collaborators access to an assay whose processes are understood, standardized and whose outcomes have been validated.

The project has been a collaborative effort by members of the UW Department of Obstetrics and Gynecology and Laboratory Medicine. The project involves establishing a protocol that encompasses the entire process, covering everything from patient selection, to sample collection, DNA processing and sequencing, to the final computational analysis of results. “The 18-month effort required that we identify and standardize all the variables in the process,” Lockwood said. “But the assay’s greatest strength is the novel, innovative computational approaches it uses,” she said. “These advances are practice changing.” In the coming year, Lockwood says the goal of BBI’s cfDNA working group is to develop an assay that is “better, cheaper, faster — and universal.”

Symposium Highlights

What is the longitudinal trajectory of cfDNA concentration over the three trimesters of pregnancy?

Suchitra Chandrasekaran, M.D., assistant professor in the Department of Obstetrics and Gynecology in the Division of Maternal Fetal Medicine, described a project to establish the normal trajectory of maternal and fetal cfDNA plasma concentrations during pregnancy. It is hoped that such data will help establish a baseline with which changes in cfDNA concentrations seen in pregnancy-related diseases can be compared. Such diseases include pregnancy-related hypertension and diabetes and fetal growth restriction. Of particular interest is establishing the trajectory of cfDNA concentrations in obese women, Body Mass Index >30, whose cfDNA concentrations are diluted due to these women’s larger plasma volume. In the study, Chandrasekaran and her co-investigators are collecting cfDNA samples from 15 nulliparous women experiencing a normal pregnancy, five of normal weight, five who are overweight and five who are obese. The distribution of visceral fat will be documented with MRI. 

Utilizing cfDNA metrics to understand pregnancy disorders associated with placental dysfunction

In another study looking at cfDNA derived from the placenta in pregnancy, Raj Swati Shree, M.D., an assistant professor in the UW Division of Maternal Fetal Medicine and co-investigators are investigating whether plasma cell-free DNA assays can diagnose two conditions associated with placental dysfunction: preeclampsia and intrauterine growth restriction (IUGR), a fetal condition that often accompanies preeclampsia. Preeclampsia complicates roughly one in ten pregnancies worldwide and causes considerable maternal morbidity and mortality. Diagnosis, however, which now depends on clinical factors and basic laboratory tests, is often made after the mother already has end-organ dysfunction, including placental dysfunction that can lead to fetal growth restriction. Currently, management primarily relies on delivery of the fetus, often preterm. A simple blood test that can identify mothers at risk would allow for earlier intervention and perhaps lead to the discovery of targets for drug development to prevent or treat the condition. Although IUGR is usually associated with preeclampsia, about 10 percent of cases are unexplained. The goal of the BBI pilot project is to develop cell-free DNA assays that will help diagnose, explain the biology, and improve the prevention and treatment of these conditions in pregnant women.

Can urine cfDNA identify patients with acute kidney injury due to sepsis vs. other causes?

Shreeram Akilesh, M.D., PhD, assistant professor in the UW Department of Pathology, and co-investigators at the Kidney Research Institute are looking to see whether analyzing cell free DNA (cfDNA) from plasma and urine could serve as a biomarker for different kinds of kidney injury. Kidney injury is a common complication of patients hospitalized in intensive care units (ICUs) and increases patient mortality and medical costs. Unfortunately, current clinical tests for kidney injury do not reveal whether the kidney is being damaged by infection/sepsis or other causes. This uncertainty in turn delays delivery of appropriate treatment. “We need better insight into what’s going on inside the black box of the injured kidney,” Akilesh said. In their study, the research team is seeking to take advantage of the fact that cfDNA encodes a fingerprint of the type of injury that is responsible for kidney damage. “We hope to use these signatures to distinguish between kidney injury due to sepsis or other causes because these have very different treatment requirements” Akilesh said. In the future, the team hopes to identify specific biomarkers for many different kinds of kidney injury, including transplant rejection.

A feasibility study of CSF cfDNA sequencing in pediatric brain tumors.

Cell-free DNA from central nervous system tumors can be detected in the cerebral spinal fluid (CSF) and analyzed to detect genetic markers that can help with diagnosis, assist with establishing prognosis and guide treatment. Bonnie Cole, M.D., clinical associate professor in the UW Department of Pathology, described a project that will compare the sequencing results from cell-free DNA obtained from the CSF with sequences obtained from formalin-fixed and paraffin-embedded tumor tissue. The goal is to see whether it is feasible to obtain cfDNA for sequencing in patients with new or relapsed brain tumors and to see whether this information is clinically useful. “We especially want to see if the results can help make the diagnosis without the need for a biopsy. And if the sequencing identifies any therapeutic targets,” Cole said.

Can low-pass genomic sequencing of plasma cfDNA be used as a platform for non-invasive detection of adult glioblastoma?

In related work in adults, P.J. Cimino, M.D., PhD, assistant professor in the UW Department of Pathology, and colleagues are working to determine whether cell-free DNA in serum rather than the CSF can be used to diagnose glioblastomas and other brain tumors. Currently, when patients present with neurological symptoms suggestive of a brain tumor, they typically have an MRI brain scan. Glioblastomas, however, can look like other conditions, including stroke, lymphomas and metastases from cancers elsewhere in the body. Confirming the diagnosis, then, often requires brain surgery so that a sample of the tumor can be biopsied.

An assay that could detect cell-free DNA with changes typical of glioblastoma in the serum would make it possible to make the diagnosis much more quickly and safely. Glioblastoma lends itself to such an approach, Cimino said, because more than 90 percent of tumors have large regions of copy number gains or loss that are easy to detect with low-pass sequencing. “Sometimes the radiology is clear, but sometimes it’s not,” Cimino said. “A blood test might help make the diagnosis.”

Circulating cfDNA from cerebrospinal fluid as a molecular and prognostic marker of epilepsy caused by focal brain malformation

Malformations of cortical development (MCD) can cause a wide spectrum of neurological disorders in children, including intellectual disability, autism and early onset epilepsy. Genes that have been linked to MCD involve cell proliferation, neuronal migration and organization of the brain cortex. These lesions can affect the entire brain, as in megalencephaly, or are focal in distribution. Focal cortical dysplasia (FCD) is an example of these disorders and is the most common cause of focal intractable seizures in children. Ghayda Mirzaa, M.D. assistant professor of pediatrics, and co-investigators have linked FCD to mutations in genes within the mTOR pathway, causing the pathway to be overactive. In her presentation, Mirzaa described a project designed to determine whether cfDNA from the cerebral spinal fluid can be used to molecularly diagnose FCD and other related brain disorders. Such tests could eliminate the need for biopsy of brain tissue that is often difficult to reach and perhaps make it possible to use mTOR inhibitors, now used to treat some cancers, to treat these patients early and either prevent or reduce the severity of epilepsy and intellectual disability from these disorders.

cfDNA from cystic lesions in human-papillomavirus (HPV)-related oropharyngeal squamous cell carcinoma guide treatment

The incidence of oropharyngeal squamous cell carcinoma, the most common cancer of the throat and mouth, has increased more than 200 percent over the past three decades. These cancers typically appear as cystic masses. Currently, the diagnosis is made by drawing fluid from these cystic masses with a fine needle. The fluid is then examined for the presence of cells typical of the cancer. If positive, the findings are then used to stage the cancer, decide on what surgical approach will be used and whether to administer chemotherapy and radiation. This approach, however, yields a high number of false negative results.

In this pilot study, Eric Konnick, M.D., MS, assistant professor of laboratory medicine and associate director of UW Medicine’s Genetics and Solid Tumor Lab, and his co-investigators are looking to see whether analysis of whole-genome copy number variation of cell-free DNA in the cystic fluid could help with diagnosis and determine the aggressiveness of the cancer, allowing the providers to plan the best treatment for these patients. “We want to compare our cell-free DNA findings with the cytologic findings and see if they would make a difference in care,” Konnick said.