Abbreviated Name

Liver CA Ref Set App (Rapid): Chilkoti (2019)

Lead Investigator

Chilkoti, Ashutosh — Duke University

Coordinating Investigator

Zheng, Yingye — Fred Hutchinson Cancer Center

Involved Investigators

• Chilkoti, Ashutosh — Duke University
• Zheng, Yingye — Fred Hutchinson Cancer Center

Abstract

Our objective is to assess the diagnostic performance of a novel multiplexed point-of-care test (POCT) against circulating biomarkers of liver cancer from patients with hepatocellular carcinoma (HCC) and differentiate them from healthy controls and at-risk patients with liver disease (i.e. with cirrhosis).

Aims

Our objective is to assess the diagnostic performance of a novel multiplexed point-of-care test (POCT) against circulating biomarkers of liver cancer from patients with hepatocellular carcinoma (HCC) and differentiate them from healthy controls and at-risk patients with liver disease (i.e. with cirrhosis).

Analytic Method

We will analyze 100 samples from the reference set (50 early stage HCC cases and 50 at risk of HCC with liver disease (cirrhotic). Samples will be analyzed in parallel on 6-plex D4 POCT chips and standard ELISA kits to directly compare the analytical performance between the two. All test points will be run in triplicate. The assay will be evaluated in terms of the following “figures-of-merit” (FOM): limit-of-blank (LOB), limit-of-detection (LOD; synonymous with ‘analytical sensitivity’), linear calibration range (LCR), upper and lower limits of quantification (LOQ) and coefficient-of-variance (COV).

Comments

No comments available.

Outcome

Among the 100 participants in the rapid testing set delivered to Chilkoti, the distribution of AFP as determined by Chilkoti and determined for the main database is exceedingly different. If a classifier employing AFP is constructed from data in the main database and applied to data returned by Chilkoti, the classifier would likely fail. However, there is a strong rank correlation for AFP between observations from the main database and observations returned by Chilkoti. Because of the strong rank correlation, a classifier constructed from Chilkoti data results in an ROC curve with AUC=0.776. AFP from Chilkoti predicts liver cancer status. The distribution of DCP is similar for specimen from Chilkoti and specimen from the same 100 participants observed in the DCP main database. DCP from Chilkoti also predicts liver cancer status. However, DCP does not do as well as AFP for predicting liver cancer status. Also, there is no benefit when DCP is added to classifier model constructed with AFP. None of the other four markers, Axl, DKK-1, GPC-3, or OPN, has any utility for classifying liver cancer, either alone or in combination with AFP.

Secure Outcome

No secure outcome available.

Publications

• No publications available at this time for this protocol.

Biomarkers

• AFP
• DKK1
• Osteopontin

Data Collections

• No data collections available at this time for this protocol.

 Team Project

Start Date

May 15 2019

Estimated Finish Date

May 15 2020

Finish Date

May 27 2020

Protocol ID

457

Protocol Type

Reference Set

Fields of Research

• Proteomics

Collaborative Group

G.I. and Other Associated Cancers Research Group

Cancer Types

• Liver cell carcinoma

Phased Status

1