NEW YORK – University of California, Los Angeles researchers have modified a circulating tumor cell (CTC) purification system to help detect early-stage hepatocellular carcinoma (HCC) in patients by identifying biomarkers in tumor-derived extracellular vesicles (EVs).
The two-step process, coined "EV Click Chip," involves collecting and purifying extracellular vesicles from a patient's blood sample before performing reverse-transcription Droplet Digital PCR (RT-ddPCR) on the EVs to identify early-stage HCC biomarkers.
The Click Chip platform comprises a custom microfluidic chip that integrates tetrazine antibody (Tz)-grafted silicon nanowire substrates with a network of microchannels altered to induce chaotic mixing. To perform biorthogonal ligation-mediated capture, the team grafts trans-cyclooctene (TCO)-modified capture antibodies to EVs in a liquid sample, such as blood or urine. When the sample runs through the chip, the Tz and TCO react and bind to the EVs.
Hsian-Rong Tseng, a pharmacology professor at UCLA and inventor of Click Chip, explained that his team explored the use of a disulfide cleaving agent that snips the embedded disulfide bonds in the antibody linkers. Released by breaking the disulfide bonds, the HCC-derived EVs are eluted out of the chip while minimizing impurities.
While Click Chip was originally developed and is still used by the UCLA team to identify gene rearrangements in CTCs, Tseng said that his team also wanted to focus on monitoring the transition from cirrhosis to early-stage HCC cancer.
The group therefore adapted the method to purify HCC-specific EVs rather than CTCs. The new method also incorporates a customized RT-ddPCR panel to quantify 10 well-defined HCC-specific mRNA transcripts.
Yazhen Zhu, study coauthor and assistant researcher in the pharmacology department at UCLA, explained that the team pursued EVs as a source for potential biomarkers because they are shed into a patient's bloodstream from the tumor at a much earlier stage than CTCs. While CTCs have a high surface area with several potential biomarkers, most EVs instead only contain one biomarker, which requires the combined use of EV Click Chip with multiple capture antibodies.
"We wanted to directly target tumor-derived EVs and therefore use Click Chip chemistry to purify the EVs, rather than antigen-antibody or immunoaffinity-based EV isolation methods," Tseng added.
In a validation study published in Nature Communications on Monday, Tseng and his team developed an antibody cocktail to recognize and capture HCC EVs from clinical plasma samples. The researchers included surface antigen markers that are highly expressed in HCC EVs, including anti-EpCAM, anti-ASGPR1, anti-CD147, and anti-GPC-3.
The group then injected 150-μl plasma samples into EV Click Chips from a cohort of 153 patients, including 46 treatment-naïve HCC patients and a control group that included patients with liver cirrhosis, chronic hepatitis B/C without liver cirrhosis, other cancers with or without metastasis, and healthy donors. After immobilizing HCC-specific EVs on the Click Chip platform, the researchers then eluted the targeted EVs out of the chips.
To avoid signals from non-HCC-specific EVs, Tseng's team detected and quantified HCC EV-specific gene expression after extracting mRNA by lysing collected EVs from the plasma samples. After running a customized panel using Bio-Rad's QX200 ddPCR instrument, the team analyzed the data to quantify copy numbers of gene transcripts detected for each gene and computed a digital score that distinguished HCC patients from at-risk cirrhotic patients.
"Instead of using a conventional immunohistochemistry way to [collect EVs], we use an antibody cocktail to address the heterogeneity of HCC cells," Zhu explained. "EVs might be even more important, as it may only have one of EpCAM, ASPGR1, or CD147, and thus you're [establishing] a tumor-specific EV."
The researchers found that EV Click Chip had a clinical sensitivity of 94 percent and specificity of 89 percent, with an area under the operating curve of 0.93, for early-stage HCC cancer detection.
In addition, Zhu highlighted that EV Click Chip distinguished HCC from noncancer (liver cirrhosis, chronic hepatitis, and health donors) and other cancer cohorts, with a clinical sensitivity of about 96 percent and specificity of 89 percent.
"This is important, since in liver cancer, you need to differentiate between primary HCC and metastatic liver cancer," Zhu said. "Even if you isolate EVs from other liver cancers, we only read HCC-specific cancers, which allows us to distinguish HCC from over cancers very well."
Because the study's HCC cohort was relatively small, however, the authors acknowledged that they will need to screen a larger number of HCC patients across all the disease's etiologies as well as perform longitudinal follow-up testing.
Tseng and his colleagues have also begun developing a proprietary HCC panel that they believe will boost the platform's AUC and clinical sensitivity and specificity. The researchers will conduct additional validation studies on 10 to 60 potential biomarkers by gathering blood samples containing EVs from institutional biorepositories.
"We will start to collect blood samples in the US first, [as] we have collaborated with UCLA's [Ronald Reagan] Medical Center, [the] Cedars-Sinai Medical Center, as well as Washington University," Tseng said in an email. "Meanwhile, we will be exploring potential collaborations with medical institutions in Asia."
Tseng noted that EV Click Chip's surface markers can be replaced with different tumor-specific markers to capture other types of tumor-derived EVs. His team has found that EV Click Chip can be used to purify tumor-derived EVs in both Ewing sarcoma and liver cancer by applying different markers.
Zhu said that the group has proposed a retrospective study to examine HCC cancer patient etiologies, where it will determine the platform's diagnostic performance for early-stage HCC, as well as for patients with cirrhosis, HPV, and nonalcoholic fatty liver disease.
An important advantage of focusing on EVs as a biomarker for HCC, Tseng highlighted, is that they can be easily preserved in serum and plasma, while CTCs are often damaged or not preserved properly.
"Many institutions have retrospective collections of EVs … and over the past four decades, [they are] one of the most well-annotated [collections in] databases," Tseng said. "We believe it'll be a goldmine to easily perform retrospective studies on the platform."
While acknowledging that next-generation sequencing-based early cancer detection approaches may offer high clinical sensitivity and specificity, Tseng pointed out that they often need between one to two weeks to produce diagnostic results.
Meanwhile, he highlighted that EV Click Chip only requires about eight hours to produce diagnostic results when combined with downstream RT-ddPCR. The team now plans to reduce the time to four hours using Qiagen's Qiacuity digital PCR system.
Tseng noted that his team has filed for a provisional patent for the EV Click Chip platform with the US Patent and Trademark Office.
If future validation study results indicate clinical potential for early stage HCC detection, Tseng said that his team may potentially license the patent from UCLA and commercialize the technology through a startup. Tseng believes the envisioned assay would "be in line with" a laboratory-developed test and offered out of a CLIA-certified, CAP-accredited laboratory.
Because the group will observe a relatively small number of biomarkers, Tseng noted that, combined with Click Chip, the cost per assay may be significantly less than sequencing-based methods.
"This will be technology that patients will be able to use to test periodically, not daily, but easier than WGS testing in order to monitor their likelihood to progress from cirrhosis to HCC," Tseng said. "No one can afford an expensive assay and wait two weeks for a report."