NEW YORK – Lipid nanoparticles that inhibit cellular uptake of cell-free DNA (cfDNA) and engineered antibodies that protect cfDNA from degradation are showing early success in boosting liquid biopsy cfDNA yields, which could eventually lead to improved cancer diagnostic and prognostic assays.
Investigators from the Broad Institute and the Koch Institute for Integrative Cancer Research at the Massachusetts Institute of Technology collaborated to develop the two methods, which are currently published as preprints on BioRxiv while undergoing peer review.
Viktor Adalsteinsson, associate director of the Gerstner Center for Cancer Diagnostics at the Broad Institute and a lead investigator on both studies, also presented his team's work at the Molecular Medicine Tri-Conference last week in San Diego.
Adalsteinsson explained that limited amounts of cfDNA is an obstacle in procedures such as cancer screening and minimal residual disease testing (MRD) for therapy selection.
"Patients with advanced cancer may still have a false negative liquid biopsy test and it's often because the full tumor genome just wasn't drawn in the tube of blood," Adalsteinsson said. "If you think of a stage I cancer with a billion cells in it, there may only be one tumor cell equivalent of cell-free DNA in a tube of blood, which is a remarkable bottleneck."
"There's a lot of MRD not detected across many different studies right after surgery," he added, "and even in therapy selection, testing sensitivity remains an issue. So if you give a priming agent [beforehand], you could potentially get the full tumor genome in a tube of blood and enable therapy selection testing for more patients."
The first priming agent consists of liposomes coated with 1,2-distearoyl-sn-glycero-3-phosphoethanolamine (DSPE). These lipid nanoparticles slow down cellular uptake of cfDNA by the cells of the mononuclear phagocyte system (MPS) by competing for uptake with cfDNA-laden mononucleosomes.
Although nanoparticle-mediated MPS uptake saturation has been explored therapeutically to decrease nanomedicine accumulation, this appears to be the first time that it has been used in a diagnostic application.
Testing in macrophage cell lines and in mice demonstrated that the DSPE-based liposomes efficiently attenuated mononucleosome uptake in a dose-dependent manner without impairing phagocytosis.
Importantly, the team recovered significantly more cfDNA and therefore also more circulating tumor DNA (ctDNA) in liposome-treated mice than in those treated with PBS as a control.
The team found that liposome treatment resulted in the recovery of more tumor-specific mutations relative to control mice. On the other hand, they noticed a concurrent decrease in tumor fraction with increasing priming agent dose. They hypothesized that this could be due to increased shedding from white blood cells in response to liposome exposure, although this needs to be clarified in further studies.
Finally, they noted that liposome treatment had no effect on tumor progression in mice.
Meanwhile, the team's antibody-based agent consists of a mouse IgG2a antibody with its Fc effector function engineered out in order to limit potential immune reactions and to make it last longer in the blood.
The antibody, which binds both double- and single-stranded DNA, significantly delayed the clearance of double-stranded DNA (dsDNA) injected into mice over the course of one hour of serial blood draws.
Mice primed with the antibody retained 1.9 percent of the dsDNA in their blood at one hour, compared to 0.04 percent in non-primed mice.
Adalsteinsson and his colleagues next showed that the low fraction of persistent dsDNA in antibody-primed mice was due to an initial rapid clearance by the MPS. While the researchers could attenuate this clearance by engineering antibodies whose Fc domains would not interact with the Fc receptors on cells of the MPS, it also points to the possibility of delivering both priming agents together.
"We're doing that experiment now," Adalsteinsson said, "and there does appear to be synergy."
Additionally, the Broad research team compared the amount of circulating tumor DNA (ctDNA) recoverable from different doses of primer agent-treated mice, untreated mice, and dummy agent-treated mice.
Using mice with large lung tumors and a panel of 1,822 tumor-specific sites, they recovered approximately 19 times more ctDNA from primer agent-treated mice than from those treated with the non-primer antibody.
"Certainly the use of antibodies and liposome nanoparticles is a completely original way of improving the cfDNA yields, and the authors did a nice number of controls to demonstrate that they're not significantly perturbing any host (mouse) biology," said Billy Lau, a Stanford University researcher who specializes in single-cell sequencing approaches to cancer.
Although Lau expressed enthusiasm for the eventual clinical applications of these studies, he noted that several items remained to be evaluated. Importantly, these will include follow-up studies to confirm that the cfDNA profiles remain informative while avoiding any safety and toxicity concerns stemming from the priming agents themselves.
"The liquid biopsy field is hot simply because blood draws themselves are generally safe and noninvasive," he said. "It's going to take a lot of hard work to convince the community to use this, especially in the context of routine screening."
Adalsteinsson noted that for both approaches, he and his colleagues chose molecules that were unlikely to provoke immune reactions upon injection.
"For the nanoparticle formulation," he said, "we intentionally chose to include lipids that were part of FDA-approved formulations. [And] for the antibody, we are using [one] that we've engineered out of the Fc effector function, [which] causes complement activation."
Activation of the complement system, he explained, triggers numerous inflammatory responses that, while useful in fighting infection, can have unwanted effects, such as promoting tumor growth.
"So far, we've seen no signs of acute or long-term toxicity with either of these agents," Adalsteinsson said. "But certainly, we have more work to do."
Adalsteinsson also noted that while these cfDNA priming methods could be integrated into numerous workflows, they complement other methods developed in his own lab to improve DNA sequencing accuracy and to improve the sensitivity of MRD testing for genome-wide mutations.
Patents have been filed on each method, which the MIT investigators hope to commercialize after further development and testing.
Although Adalsteinsson said that it is too early to comment on specific plans related to licensing and commercialization, "we think these could have quite a widespread impact as they address a hurdle that's upstream of essentially all liquid biopsy tests."
"I am just really enthusiastic about the potential clinical utility if the follow-up work pans out," said Stanford's Lau. "These proof-of-concept studies are very solid."