# Development and Evaluation of a Droplet Digital PCR Assay for the Accurately Detecting the CircHIPK3 in Plasma Samples from Patients with Hepatocellular Carcinoma

## TL;DR

A new droplet digital PCR method was developed to accurately detect a specific RNA in blood samples from liver cancer patients.

## Contribution

A novel ddPCR assay was developed and validated for sensitive detection of circHIPK3 in low-load plasma samples.

## Key findings

- ddPCR showed higher linearity (R² = 0.994) compared to qPCR (R² = 0.9877) for circHIPK3 detection.
- ddPCR had a lower detection limit (0.594 copies/μl) than qPCR (5.753 copies/μl).
- ddPCR detected 11 additional positive samples that qPCR missed in clinical plasma samples.

## Abstract

Hepatocellular carcinoma (HCC) is an increasingly prevalent malignant neoplasm on a global scale. Circrna HIPK3 (circHIPK3) has been identified as playing a key role in HCC tumorigenesis and as a novel biomarker. In this study, we aimed at developing a sensitive and accurate method for the detection of circHIPK3 in low load plasma samples using a droplet digital PCR (ddPCR). We designed circHIPK3 gradient primers and probes and optimized the PCR system to improve performance. Then we assessed and compared the linearity and sensitivity of ddPCR and quantitative PCR (qPCR) using the circHIPK3 plasmid DNA as a template. Using these methods, circHIPK3 concentrations were quantitatively determined in 3 cell lines and 40 plasma samples to assess clinical stability. Within the plasmid concentration range of 31-36 copies/μl, the ddPCR exhibited a linear fitting equation of Y = 1.037X-0.1724 with R2 value of 0.9940, which surpassed the corresponding R2 value of 0.9877 for qPCR. Furthermore, the limit of blank (LOB) and limit of detection (LOD) for ddPCR were determined to be 0.157 copies/μl and 0.594 copies/μl, respectively, which were significantly lower than the LOD of qPCR (5.753 copies/μl). In clinical samples, ddPCR demonstrated a commendable correlation with qPCR, evidenced by a Kappa value of 0.677 (p < 0.05, 95% CI [0.503-0.851]) and an intraclass correlation coefficient (ICC) of 0.903 (95% CI [0.831-0.946]). Notably, ddPCR identified 11 positive samples that qPCR failed to detect. The ddPCR-based circHIPK3 liquid biopsy method emerged as a highly sensitive and accurate approach, lending itself well to clinical applications.