The extracellular matrix (ECM) plays an important role in cancer progression. It can be divided into the basement membrane (BM) that supports epithelial/endothelial cell behavior and the interstitial matrix (IM) that supports the underlying stromal compartment. The major components of the ECM are the collagens. While breaching of the BM and turnover of e.g. type IV collagen, is a well described part of tumorigenesis, less is known regarding the impact on tumorigenesis from the collagens residing in the stroma. Here we give an introduction and overview to the link between tumorigenesis and stromal collagens, with focus on the fibrillar collagens type I, II, III, V, XI, XXIV and XXVII as well as type VI collagen. Moreover, we discuss the impact of the cells responsible for this altered stromal collagen remodeling, the cancer associated fibroblasts (CAFs), and how these cells are key players in orchestrating the tumor microenvironment composition and tissue microarchitecture, hence also driving tumorigenesis and affecting response to treatment. Lastly, we discuss how specific collagen-derived biomarkers reflecting the turnover of stromal collagens and CAF activity may be used as tools to non-invasively interrogate stromal reactivity in the tumor microenvironment and predict response to treatment.

BackgroundExcessive extracellular matrix (ECM) remodeling and a reactive stroma can affect T-cell infiltration and T-cell activity in the tumor and hereby influence response to immune checkpoint inhibitors (ICI). In the pursuit of finding biomarkers that predict treatment response, we evaluated the association between serum biomarkers of collagen and vimentin turnover and outcomes in metastatic melanoma patients treated with the anti-CTLA-4 antibody ipilimumab (IPI).MethodsType III collagen formation (PRO-C3), MMP-degraded type I, type III and type IV collagens (C1M, C3M and C4M), and citrullinated and MMP-degraded vimentin (VICM) were measured with ELISAs in serum from metastatic melanoma patients before (n = 66) and 3 weeks after (n = 52) initiation of IPI treatment. Biomarker levels were associated with Disease Control Rate (DCR) and survival outcomes.ResultsWe found that baseline levels of PRO-C3 (p = 0.011), C1M (p = 0.003), C3M (p = 0.013) and C4M (p = 0.027) were significantly elevated in patients with progressive disease (PD). Univariate Cox regression analysis identified high PRO-C3 (p = 0.021) and C4M (p = 0.008) as predictors of poor overall survival (OS) and the biomarkers remained significant when evaluated with other covariates (PRO-C3 (p = 0.049) and C4M (p = 0.046)). Multivariate analysis identified VICM as a predictor of longer OS (p = 0.026). Similarly, a high C3M/PRO-C3 ratio predicted for increased OS (p = 0.034). Only C3M (p = 0.003) and VICM (p < 0.0001) increased 3 weeks after treatment.ConclusionsECM and tissue remodeling quantified in pre-treatment serum were associated with response and survival outcomes in metastatic melanoma patients treated with IPI. This highlights the importance of addressing the ECM and stromal component non-invasively in future ICI studies.Electronic supplementary materialThe online version of this article (10.1186/s40425-018-0474-z) contains supplementary material, which is available to authorized users.

No abstract

Extracellular matrix (ECM) proteins, such as collagen type I and elastin, and intermediate filament (IMF) proteins, such as vimentin are modified and dysregulated as part of the malignant changes leading to disruption of tissue homeostasis. Noninvasive biomarkers that reflect such changes may have a great potential for cancer. Levels of matrix metalloproteinase (MMP) generated fragments of type I collagen (C1M), of elastin (ELM), and of citrullinated vimentin (VICM) were measured in serum from patients with lung cancer (n = 40), gastrointestinal cancer (n = 25), prostate cancer (n = 14), malignant melanoma (n = 7), chronic obstructive pulmonary disease (COPD) (n = 13), and idiopathic pulmonary fibrosis (IPF) (n = 10), as well as in age-matched controls (n = 33). The area under the receiver operating characteristics (AUROC) was calculated and a diagnostic decision tree generated from specific cutoff values. C1M and VICM were significantly elevated in lung cancer patients as compared with healthy controls (AUROC = 0.98, P < 0.0001) and other cancers (AUROC = 0.83 P < 0.0001). A trend was detected when comparing lung cancer with COPD+IPF. No difference could be seen for ELM. Interestingly, C1M and VICM were able to identify patients with lung cancer with a positive predictive value of 0.9 and an odds ratio of 40 (95% CI = 8.7–186, P < 0.0001). Biomarkers specifically reflecting degradation of collagen type I and citrullinated vimentin are applicable for lung cancer patients. Our data indicate that biomarkers reflecting ECM and IMF protein dysregulation are highly applicable in the lung cancer setting. We speculate that these markers may aid in diagnosing and characterizing patients with lung cancer.

Abstract. BACKGROUND: During cancer the otherwise tightly controlled homeostasis of the extracellular matrix (ECM) is disturbed. The protein composition changes, the ECM stiffens and increased levels of proteases are secreted. The combination of these processes result in release of specific protein fragments (e.g. collagens) to the circulation, which when measured may reflect disease pathogenesis. OBJECTIVE: To investigate if biomarkers of protease-degraded collagen could differentiate ovarian and breast cancer patients from healthy controls when measured in serum. METHODS: The levels of markers reflecting MMP-degradation of type I (C1M), type III (C3M) and type IV (C4M, C4M12) collagen were assessed in serum from ovarian cancer patients (n = 10), breast cancer patients (n = 14) and healthy controls (n = 49) using validated ELISAs. The markers were compared using one way ANOVA and AUC was calculated. RESULTS: All markers were significantly elevated in serum from ovarian cancer patients (p < 0.0001) and breast cancer patients (p < 0.04-0.0001) compared to healthy controls. Furthermore, diagnostically the markers were able to differentiate ovarian (AUROC 90%-93%) and breast cancer patients (AUROC 76%-93%) from healthy controls, with C1M being the strongest differentiator of disease vs. controls. CONCLUSION: Four serum biomarkers reflecting altered MMP-mediated collagen turnover were able to differentiate ovarian and breast cancer patients from healthy controls.Keywords: Cancer, extracellular matrix, collagen, biomarkers, breast cancer, ovarian cancer, matrix metalloproteinase, degradation, remodeling, tumor microenvironment BackgroundWomen's health is significantly influenced by the occurrence of cancers, such as breast and ovarian can- cers. Breast cancer is the leading cause of cancer death among women in the US and EU [1,2], and ovarian cancer, although less frequent, is often discovered in the late clinical stages where the cancer has already spread and no curative interventions are possible [3]. Taken together, the medical needs for breast and ovarian cancer are early diagnosis, prognosis and prediction of a therapeutic response. One of the best ways to achieve this is to use serum biomarkers [4].

The extracellular matrix (ECM) plays a vital role in maintaining normal tissue function. Collagens are major components of the ECM and there is a tight equilibrium between degradation and formation of these proteins ensuring tissue health and homeostasis. As a consequence of tissue turnover, small collagen fragments are released into the circulation, which act as important biomarkers in the study of certain tissue-related remodeling factors in health and disease. The aim of this study was to establish an age-related collagen turnover profile of the main collagens of the interstitial matrix (type I and III collagen) and basement membrane (type IV collagen) in healthy men and women.By using well-characterized competitive ELISA-assays, we assessed specific fragments of degraded (C1M, C3M, C4M) and formed (PINP, Pro-C3, P4NP7S) type I, III and IV collagen in serum from 617 healthy men and women ranging in ages from 22 to 86. Subjects were divided into 5-year age groups according to their sex and age. Groups were compared using Kruskal-Wallis adjusted for Dunn’s multiple comparisons test and Mann-Whitney t-test. Age-specific changes in collagen turnover was most profound for type I collagen. PINP levels decreased in men with advancing age, whereas in women, the level decreased in early adulthood followed by an increase around the age of menopause (age 40–60). Sex-specific changes in type I, III and IV collagen turnover was present at the age around menopause (age 40–60) with women having an increased turnover. In summary, collagen turnover is affected by age and sex with the interstitial matrix and the basement membrane being differently regulated. The observed changes needs to be accounted for when measuring ECM related biomarkers in clinical studies.

Hyaluronan (HA) and collagen are highly expressed in pancreatic cancer (PC) stroma. HA and collagen accumulation increase tumor interstitial fluid pressure, compromising blood flow and drug penetration. The aim of this biomarker study was to determine the clinical utility of serum HA and the propeptide of type III collagen (PRO‐C3) in patients with PC. A cohort from the Danish BIOPAC study (NCT03311776) including patients with histologically confirmed pancreatic ductal adenocarcinoma (n = 809), ampullary carcinoma (n = 44), distal biliary tract cancer (n = 31), chronic pancreatitis (n = 15), intraductal papillary mucinous neoplasm (n = 41), duodenal adenoma (n = 7) and no cancer (n = 25). Healthy controls were available for serum HA (n = 141) and PRO‐C3 (n = 8). The main outcome was overall survival (OS) of patients with PC in relation to pretreatment serum HA and PRO‐C3 levels. Patients with PC had higher baseline serum HA and PRO‐C3 than healthy subjects and patients with benign conditions. Pretreatment serum baseline HA and PRO‐C3 in patients with PC were associated with poorer survival and PRO‐C3 remained prognostic also after adjusting for age, performance status, stage, the presence of liver and peritoneum metastasis, and CA19‐9. Detection of HA and PRO‐C3 may be useful in differentiating between malignant and benign pancreatic conditions. Serum HA and PRO‐C3 were prognostic for OS in patients with PC.

Pancreatic ductal adenocarcinoma (PDAC) patients have poor prognosis and poor response to treatment. This is largely due to PDAC being associated with a dense and active stroma and tumor fibrosis (desmoplasia). Desmoplasia is characterized by excessive degradation and formation of the extracellular matrix (ECM) generating collagen fragments that are released into circulation. We evaluated the association of specific collagen fragments measured in pre-treatment serum with outcome in patients with PDAC. Matrix metalloprotease (MMP)-degraded type I collagen (C1M), type III collagen (C3M), type IV collagen (C4M) and a pro-peptide of type III collagen (PRO-C3) were measured by ELISA in pre-treatment serum from a randomized phase 3 clinical trial of patients with stage III/IV PDAC treated with 5-fluorouracil based therapy (n = 176). The collagen fragments were evaluated for their correlation (r, Spearman) with serum CA19-9 and for their association with overall survival (OS) based on Cox-regression analyses. In this phase 3 PDAC trial, pre-treatment serum collagen fragment levels were above the reference range for 67%-98% of patients, with median values in PDAC approximately two-fold higher than reference levels. Collagen fragment levels did not correlate with CA19-9 (r = 0.049–0.141, p = ns). On a continuous basis, higher levels of all collagen fragments were associated with significantly shorter OS. When evaluating degradation (C3M) and formation (PRO-C3) of type III collagen further, higher PRO-C3 was associated with poor OS (>25th percentile cut-point, HR = 2.01, 95%CI = 1.33–3.05) and higher C3M/PRO-C3 ratio was associated with improved OS (>25th percentile cut-point, HR = 0.53, 95%CI = 0.34–0.80). When adjusting for CA19–9 and clinical covariates, PRO-C3 remained significant (HR = 1.65, 95%CI = 1.09–2.48). In conclusion, collagen remodeling quantified in pre-treatment serum as a surrogate measure of desmoplasia was significantly associated with OS in a phase 3 clinical PDAC trial, supporting the link between desmoplasia, tumorigenesis, and response to treatment. If validated, these biomarkers may have prognostic and/or predictive potential in future PDAC trials.