Validation of a Multivariate Serum Profile for Epithelial Ovarian Cancer Using a Prospective Multi-Site Collection

Abstract

In previous studies we described the use of a retrospective collection of ovarian cancer and benign disease samples, in combination with a large set of multiplexed immunoassays and a multivariate pattern recognition algorithm, to develop an 11-biomarker classification profile that is predictive for the presence of epithelial ovarian cancer. In this study, customized, Luminex-based multiplexed immunoassay kits were GMP-manufactured and the classification profile was refined from 11 to 8 biomarkers (CA-125, epidermal growth factor receptor, CA 19-9, C-reactive protein, tenascin C, apolipoprotein AI, apolipoprotein CIII, and myoglobin). The customized kits and the 8-biomarker profile were then validated in a double-blinded manner using prospective samples collected from women scheduled for surgery, with a gynecologic oncologist, for suspicion of having ovarian cancer. The performance observed in model development held in validation, demonstrating 81.1% sensitivity (95% CI 72.6 – 87.9%) for invasive epithelial ovarian cancer and 85.4% specificity (95% CI 81.1 – 88.9%) for benign ovarian conditions. The specificity for normal healthy women was 95.6% (95% CI 83.6 – 99.2%). These results have encouraged us to undertake a second validation study arm, currently in progress, to examine the performance of the 8-biomarker profile on the population of women not under the surgical care of a gynecologic oncologist.

Comprehensive Serum Profiling for the Discovery of Epithelial Ovarian Cancer Biomarkers

“…This assured sample quality and removed the possibility of any collection, processing or biological biases in the sample set, a concern for many other studies [30] . No normal healthy samples were used in this study, as they are typically easier to classify than benign conditions [21] and introduce confounding factors such as lower stress levels compared to patients facing surgery [38] . As expected, the median patient age was higher in individuals with ovarian cancer (61 years) than those with benign conditions (51 years) and increased with the stage of disease present ( Table 1 ; [28] , [36] ).…”

“…Sera were from a prospective collection undertaken by Correlogic Systems, Inc. specifically to develop and validate the performance of an ovarian cancer test [21] . All samples were collected under a uniform protocol from 11 different sites, which were monitored for adherence.…”

“…Many other strategies have sought to combine CA-125 with additional markers [16] , [17] , [18] , [19] , [20] . OvaCheck® combines CA-125 with seven other markers and has 81.1% sensitivity and 85.4% specificity as determined in a double-blinded clinical validation study [21] . However, the test performance needs to be validated on a non-specialist population (e.g.…”

“…In a previous report, we profiled 104 common serum biomarkers, 44 autoimmune markers and 56 infectious-disease markers [36] and reported on their individual ability to discriminate ovarian cancer from normal and benign conditions. In related studies, we built panels of biomarkers to improve the performance of the individual biomarkers ( [21] , [37] ). The final panel, OvaCheck, was intended to be used to refer women with symptoms of ovarian cancer to a gynecologic oncologist.…”

Microarrays as Platform for Multiplex Assays in Biomarker and Drug Discovery

“…In a similar approach, Wittekind and colleagues reported a set of proteins as biomarker candidates associated with hepatocellular carcinoma [20]. Nowadays, several studies have been performed and focused on biomarker identification in several oncological pathologies; for example: In ovarian cancer, 11 proteins have been identified by Amonkar et al [21]; Sreekumar et al identified a panel of proteins as biomarker candidates in colon carcinoma cells [22]; Díez et al has identified differentially expressed proteins in B-cell chronic lymphocytic leukemia which are related with target therapeutics [23]. Previously, Below and coworkers have developed an antibody microarray to immunophenotype 1100 leukemia and lymphomas according to the abundance of a panel of 82 antigens or cluster of differentiation (CD) characterized at the surfaces of lymphocytes [24].…”