An integrated proteomics approach identified multiple differentially expressed serum proteins, including Ceruloplasmin and Complement C3, in African American women with breast cancer.
Case-Control (n=34)
Does an integrated proteomics and multivariate pattern analysis identify specific serum biomarkers for breast cancer in treatment-naive African American women?
An integrated proteomics approach successfully identified specific serum protein biomarkers for breast cancer in treatment-naive African American women, demonstrating potential for precision oncology in underrepresented populations.
Abstract Background: Breast cancer is the leading cause of morbidity and mortality among African American women. Identifying population-specific serum biomarkers is essential for early detection and risk stratification. To address limitations of traditional univariate analyses, we developed an integrated platform combining classical proteomics (2D-DIGE, MALDI-TOF/TOF, LC-MS/MS) with Random Forest (RF) and cumulative distribution function (CDF)-based analyses to robustly discover and validate serum biomarkers. Population: Two cohorts of treatment-naïve African American women were examined following depletion of high-abundance proteins. The primary cohort included eleven breast cancer patients median age 50 years, range 29-74, and 11 age-matched healthy controls analyzed by 2D-DIGE and MALDI-TOF/TOF. An independent cohort of 6 patients and 6 controls was assessed by LC-MS/MS (MudPIT). Patients represented both pre- and postmenopausal status and diverse tumor subtypes and receptor profiles. Controls were healthy women without a cancer history. Methods: Proteomic analysis using 2D-DIGE gels combined with MALDI-TOF/TOF allowed the detection and identification of differentially expressed serum proteins. LC-MS/MS data were analyzed with RF (1,000 bootstrap iterations) using Gini importance to rank peptide relevance, and the CDF evaluated distribution differences using an S statistic derived from 1,000 permutations (S threshold greater than or equal to 3). The combination of RF and CDF enabled the detection of relevant signals under conditions of high dimensionality, collinearity, and potentially non-Gaussian distributions. Results: The integrated approach revealed multiple differentially expressed serum proteins in breast cancer. Representative biomarkers included Ceruloplasmin, Complement C3; Alpha-1B-glycoprotein, angiotensinogen precursor, Insulin-like growth factor-binding protein complex acid labile subunit, Hemopexin precursor, and vitamin D binding protein. Complementary 2D-DIGE and MALDI-TOF/TOF analyses allowed direct determination of these differential protein patterns, while LC-MS/MS combined with RF and CDF prioritized peptides with high discriminatory power and independently confirmed their statistical significance. Conclusion: The integration of classical proteomics techniques (2D-DIGE and MALDI-TOF/TOF) with LC-MS/MS combined with RF and CDF analysis enables reliable detection and prioritization of serum biomarkers. This approach combines the sensitivity of multivariate proteomics with non-parametric statistical rigor, making it suitable for small, high-dimensional cohorts, and demonstrates potential for identifying biomarkers specific to underrepresented populations, supporting precision oncology applications in breast cancer. Citation Format: Padma P. Tadi Uppala, Hyun J. Kwon, Elmer C. Rivera, Sharon S. Lum. Serum protein signatures specific to breast cancer in treatment-naive African American women identified using integrated proteomics and multivariate pattern analysis abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6333.
Uppala et al. (Fri,) conducted a case-control in Breast cancer (n=34). Integrated proteomics with Random Forest and CDF analysis vs. Healthy controls was evaluated on Differentially expressed serum proteins. An integrated proteomics approach identified multiple differentially expressed serum proteins, including Ceruloplasmin and Complement C3, in African American women with breast cancer.