Liquid Biopsy Detection of TP53, mGSTP1, and mRASS1Fa from Circulating Tumor DNA Improves the Non-invasive Diagnosis of Hepatocellular Carcinoma

Monday, April 13, 2026

3:09 PM - 3:18 PM ET

Location: 718B

Presenting Author(s)

Wali Badar, MD

Resident physician
University of Illinois at Chicago, United States

Author/Co-author(s)

TW

Taryi Wint, MBBS (she/her/hers)

Research Associate
University of Illinois at Chicago, United States
NK

Nusrat Khan, PhD

Post Doctoral Student
UIC, United States
OF

Omar Farooqui, None

Undergraduate Student
UIC, United States
SK

Shaheer Khan, BS

Undergraduate Student
Loyola, United States
Kyle M. Schachtschneider, PhD (he/him/his)

Vice President of R&D, Services
Sus Clinicals Inc., United States
LE

Lobna Elkhadragy, PhD

Postdoctoral Research Associate
University of Illinois at Chicago, United States
RG

Ron C. Gaba, MD, MS

Professor
University of Illinois at Chicago, United States

Purpose: Alpha fetoprotein (AFP) is used to screen high risk patients for hepatocellular carcinoma (HCC) but has limited sensitivity, demanding the identification of new biomarkers. This study assessed the feasibility and performance of detecting mutated TP53, methylated GSTP1 (mGSTP1), and RASS1Fa (mRASFF1a)-the most common point mutation and epigenetic variants in HCC-from circulating tumor DNA (ctDNA) to enhance HCC screening.

Materials and Methods: This single center prospective study enrolled both HCC and control patients with chronic liver disease. Plasma was collected, at the time of diagnosis for HCC patients. PCR assay was used to measure TP53, mGSTP1, and mRASSF1a concentration in ctDNA. AFP was quantified as part of routine clinical evaluation. Univariable analysis was performed for demographic data and biomarkers. Area under receiver operator curve analysis (AUROC) was applied to assess the diagnostic performance of AFP alone and in combination with TP53, mGSTP1, and mRASSF1a. Combination model was created by fitting a logistic regression model for the probability of HCC diagnosis.

Results:

85 patients (56 HCC, 29 control; M/F 55/30, mean age 62 years) were enrolled. Liver disease etiology liver disease included: viral (27/85, 32%), metabolic (26/85, 31%), alcohol (25/85, 29%), and other (7/85, 8%). Tumor stage included BCLC 0 (7/56, 13%), A (29/56, 52%), B (17/56, 30%), and C (3/56, 5%). AUROC for AFP was 0.889 and for the combined model was 0.955. Accuracy for AFP alone was 0.790 vs 0.935 for combined model. Negative predictive value for AFP was 0.409 vs 0.692 for the combined model. Among HCC subjects with low AFP (< 20 ng/mL), 85% (33/39) had increased TP53 mutation, mGSTP1, or mRASS1Fa levels.

Conclusion: ctDNA detection of TP53 mutation and mGSTP1/mRASS1Fa complements AFP as a screening biomarker for HCC and may improve HCC diagnosis by detecting HCC tumors when AFP is low. Future studies will focus on further validation and prognostic assessment.