MSACL 2025 Abstract

Spatial Post-Translational Modifications as Precision Indicators of Breast Cancer Risk and Progression: Are these the NextGen Biomarkers?

Taylor Hulahan (1), Jaclyn Dunne (1), Denys Rujchanarong1, Danielle Scott (1), Laura Spruill (1), Mengjun Wang (1), Heather Jensen-Smith (2), George E. Sandusky (3), Harrison Taylor (1), Jade Macdonald (1), Ashlyn Ivey (1), Michael Anthony Hollingsworth (2), Lauren E. Ball (1), Anand S. Mehta (1), Graham Colditz (4), Jeffrey Marks (5), E. Shelley Hwang (5), Robert West (6), Richard R. Drake (1), Harikrishna Nakshatri (3), Marvella Ford (1), Peggi M. Angel (1)
(1) Medical University of South Carolina, Charleston, SC (2) University of Nebraska Medical Center, Omaha, NE (3) Indiana University School of Medicine, Indianapolis, IN (4) Washington University at St. Louis, St. Louis, MO, United States (5) Duke University, Durham, NC, USA (6) Stanford University School of Medicine, Stanford, CA 94305, USA

Peggi Angel, PhD (Presenter) MUSC Proteomics Center

Presenter Bio: Peggi Angel is tenured Professor at Medical University of South Carolina Department of Pharmacology & Immunology and Co-Director of Mass Spectrometry Imaging. Dr. Angel’s work focuses on the contribution of spatial chemical biology to the external, endogenous environmental in disparities of disease risk, progression, and therapeutic resistance. She has developed multiple mass spectrometry imaging approaches to spatial biology all of which are designed for use on clinically archived human specimens of tissues, cells and fluids, and are developed working with clinicians. Notably, she is the inventor of a spatial method targeting the collagen proteome in formalin-fixed, paraffin-embedded tissues that integrates collagen proteomic maps with spatial transcriptomics and microscopy studies. Dr. Angel has over 14 years cumulative experience in 5 biotech startups including Glycopath, Inc., a company that leveraged glycosylation patterns as a prognostic or diagnostic tool; she currently serves on the board of N-Zyme Scientifics, a company that produces enzymes for targeted mass spectrometry imaging. Dr. Angel is committed to creating a collaborative mass spectrometry imaging community and serves as Past President for the Americas Region of the International Mass Spectrometry Imaging Society, as a Trustee for the International Mass Spectrometry Imaging Society, and as Secretary on the USHUPO Board of Directors. Dr. Angel is devoted to coaching and mentoring, particularly for females and minorities, serving on multiple committees to advise and mentor young scientists in entrepreneurship within multidisciplinary teams

Relevant Financial Disclosures (within past 24 months, reported on Jul 17, 2026) No relevant financial relationship(s) to disclose. Conflict mitigation NOT REQUIRED.

INTRODUCTION:
It is a dichotomy that collagen breast stroma is clinically imaged for breast cancer risk and progression, yet molecular regulation of the breast collagen proteome remains poorly defined. Collagen proteins represent the majority of the extracellular microenvironment in any tissue; emergence of collagen stroma is significantly associated with risk of breast cancer (BC). Post-translational modifications (PTMs), particularly those on collagen proteins, play a significant role in cell function within the tissue microenvironment leading to disease. In current studies, we test if the breast microenvironment shows specific PTMs linked to histopathological and clinical characteristics over the continuum of breast health.

METHODS:
Formalin-fixed, paraffin-embedded sections of breast tissue were prepared for multiplexed glycomic and proteomic imaging experiments with appropriate institutional review board permissions. Tissues were dewaxed, sprayed with enzymes (HTX Imaging), digested, and imaged using published protocols. Multiplexed, sequential glycomics used EndoF3 (N-zyme Scientifics) to target core fucosylated N-glycans and PNGaseF (N-zyme Scientifics) to target all other N-glycans. After glycomics, collagenase was used to target collagens and other extracellular proteins. Serial sections were used to define select cell markers (PhotoCleavable Mass-Tags (PCMT); AmberGen). Glycomic, proteomic, and PCMT imaging were done by MALDI FT-ICR or MALDI QTOF (timsTOF fleX, Bruker). Recently reported proteomic workflows were used to derive peptide identifications. Images were visualized by commercial software (SCiLS; Bruker or Weave; Aspect Analytics). Exported data were statistically evaluated (Prism v10).

RESULTS:
This report summarizes data from several independent mass spectrometry imaging studies across the continuum of breast health. Collagen proteins increase with breast density and contain PTMs that facilitate localized cell recruitment, and signaling. Very few studies investigate spatial distribution and regulation of the collagen proteome across the breast continuum. Tissue sections and tissue microarrays were used to investigate post-translational modifications in healthy breast from the Susan G. Komen tissue bank (n=86 samples), ductal carcinoma in situ (DCIS) (n>210) normal adjacent and matched invasive breast carcinomas (IBC; n=79) and triple-negative breast cancer (TNBC; n=80). A main comparator was differences in populations at high risk versus low risk for breast cancer mortality. Area under the receiver operating curve (AUROC) ≥70%, p-value≤0.05 was used to distinguish between groups; nonparametric testing p-value <0.05, FDR controlled, was used to evaluate staging.

ECM-targeted proteomic imaging reported collagen peptides post translationally modified by hydroxylation of proline (HYP) are spatially regulated within the breast microenvironment. In normal breast, single collagen peptides from stroma could differentiate by clinical imaging category (47 peptides, AUROC≥70%, p-value 0.01). In DCIS, specific collagen peptide intensities decreased with progression to invasive breast cancer and associated with an 19.5% decrease in disease free survival probability (hazard ratio 2.8, p-value 0.041). In a cohort of 27 patients with matched primary tissue and later tissue collected with invasive breast cancer, collagen peptides with hydroxylated prolines showed a significant decrease (Mann-Whitney test with multiple comparison correction performed q-value 0.002). Exploratory machine learning suggested a 6-peptide signature could predict between primary DCIS and all later recurrence (AUROC 99.6%, Accuracy 98.5%, positive predictive value 97.1%, negative predictive value 100%. TNBC showed differences in staging with increases in specific peptides from stage I-III, followed by decreases at metastasis. A total of 9 peptides associated with survival based on HYP status, including discoidin domains binding sites, known for immune exclusion. HYP variants tested against normal and TNBC breast epithelial cells demonstrated changes in transcriptional signaling dependent on the site localization of proline hydroxylation.

CONCLUSION:
The collective studies suggest that post-translational modification on collagen peptides contribute to profound pathological changes within the breast microenvironment. Larger studies are warranted to leverage PTM variation as breast cancer targets for improved clinical management, new therapeutic avenues, and as predictors of outcome.