Biomarker Developmental Laboratory

Project Summary/Abstract
Cancer screening can detect cancer early and reduce mortality. However, high false positive rates and low
sensitivities of lung (LC) and ovarian cancer (OC) screening result in unnecessary surgeries and missed
cancers. Our overarching goal is to develop circulating biochemical biomarkers that reduce the false positive
rate of the CT scan for LC and increase the positive predictive value of OC screening in combination with
CA125 and trans-vaginal ultrasound (TVUS). We hypothesize that LC and OC patients have different antibody
immune responses from subjects with benign diseases that can be explored for detecting cancer and excluding
non-cancer. We will take a systems approach working with 3 types of antibodies: autoantibodies, anti-microbial
antibodies and anti-aberrant glycoprotein antibodies. Our proposal builds on our extensive experience with
cancer biomarker discovery and immunoproteomics technology. Our previous results on autoantibody
biomarkers have been confirmed in blinded PRoBE phase 2 multicenter validation studies. Our results showed
that autoantibody biomarkers are highly specific for cancer; although, multiple autoantibodies are needed for
adequate predictive value. There is increased attention to the role microbes play in cancer vs. benign disease
development, which can be observed by different overall infection rates and / or immune responses to different
individual antigens from various microbes. Aberrant protein glycosylation is a hallmark of LC and OC, and we
will exploit our unique capability to display glycosylated proteins in high throughput. The successful
implementation of our study entails high throughput methods for assessing antibodies at the proteome level.
With EDRN support, we have developed a set of innovative immunoproteomics technologies, namely high-
density nucleic acid programmable protein array (HD-NAPPA), contra-capture protein array (CCPA) and
multiple in solution protein array (MISPA), that, combined with the largest full-length human and microbial gene
collection (DNASU plasmid repository), enable us to study antibodies against the full human proteome,
microbial proteomes and the human O-glycoproteome for antibody biomarker signatures in cancer. Our
previous success, Videssa® Breast, which is a CLIA-certified blood test for breast cancer used clinically with
mammography, includes a panel of both autoantibody markers and serum protein markers. As with Videssa®,
we will also use the Meso Scale Diagnostics (MSD) MultiArray platform to assess reported serum protein
markers for their utility to complement our antibody markers for LC and OC. We will collaborate with experts on
LC and OC screening at Vanderbilt University Medical Center, Boston University, MD Anderson Cancer
Center, and German Cancer Research Center, who will also provide access to high quality well-characterized
samples from our target populations. We will perform Phase I (PRoBE) discovery by screening protein arrays
with cancer and control sera for cancer- or control-specific antibodies. Candidate biomarkers for both LC and
OC will be Phase 2-validated using ELISA or MISPA assays.

Grant Number: 5U2CCA271903-04
NIH Institute/Center: NIH
Principal Investigator: Karen Anderson