Abstract 3840: Novel aspects of tumor fibrosis: Non-invasively assessed type XXII collagen, and not type I collagen, depicts highy elevated levels in patients with various solid tumor types

Abstract

Abstract Background: The major constituents of the tumor microenvironment (TME), are besides the tumor cells themselves, cancer associated fibroblasts (CAFs), stromal cells and the extracellular matrix (ECM). CAFs synthesize excess amounts of collagens leading to a fibrotic TME with poor response to anti-cancer treatment. Traditionally, tumor fibrosis is defined by quantification of fibrillar collagens such as type I collagen (COL1). So-called Fibril-Associated Collagen with Interrupted Triple helices (FACITs) are less studied in this context. Type XXII collagen (COL22) is a minor FACIT located at the basement membrane. COL22 may therefore provide a different value compared to standard COL1 assessment traditionally associated with tumor fibrosis. Methods: Levels of COL1 (PRO-C1) and COL22 (PRO-C22) was measured non-invasively using competitive enzyme-linked immunosorbent assays in serum from patients diagnosed with bladder-, breast-, colorectal-, head and neck-, kidney-, lung-, melanoma-, pancreatic-, prostate-, and stomach cancer (n=220) and compared to healthy controls (n=33) (Cohort 1). Based on results from Cohort 1, PRO-C22 was evaluated in an independent cohort (Cohort 2) comprising patients with pancreatic ductal adenocarcinoma (PDAC) (n=39) and healthy controls (n=20). Receiver operating characteristic (ROC) curve analysis was performed to assess the diagnostic performance of PRO-C22. Kaplan-Meier curves and Cox proportional hazard models were used to assess the association between PRO-C22 and overall survival (OS) after dichotomizing PRO-C22 at the median cutpoint. Results: In Cohort 1, PRO-C22 was significantly higher in patients with all individual cancer types compared to healthy controls (p<0.01 to p<0.0001). In support, the area under the ROC (AUROC) for separation of patients with cancer from healthy controls using PRO-C22 ranged from 0.89 to 0.98 (p<0.0001). As comparison PRO-C1 assesment showed no significant difference. PRO-C22 was also significantly higher in patients with PDAC (Cohort 2) when compared to healthy (p<0.0001) with an AUROC of 0.87 (p<0.0001). Furthermore, high PRO-C22 levels were associated with shorter OS (median OS days 162 and 1363 for ‘high’ and ‘low’ PRO-C22 subgroups, respectively, p=0.0011). Likewise, high PRO-C22 levels at baseline predicted increased risk of mortality (HR=3.53, 95% CI 1.59-7.86, p=0.0020). Conclusion: These proof-of-concept data indicate that PRO-C22 may exhibit potential as a novel diagnostic and prognostic tumor fibrosis biomarker in cancer. PRO-C22 may be superior to type PRO-C1. Larger comprehensive clinical cohorts are needed to fully understand the biology associated with COL22 in cancer as well as the biomarker potential of PRO-C22 in various cancer types. Citation Format: Emilie Albrecht Madsen, Jeppe Thorlacius-Ussing, Christina Jensen, Neel I. Nissen, Astrid Z. Johansen, Inna M. Chen, Julia S. Johansen, Tina Manon-Jensen, Hadi M. Diab, Lars N. Jørgensen, Morten Karsdal, Nicholas Willumsen. Novel aspects of tumor fibrosis: Non-invasively assessed type XXII collagen, and not type I collagen, depicts highy elevated levels in patients with various solid tumor types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3840.