Serum MMP-8 and TIMP-1 predict prognosis in colorectal cancer

• Main Authors: Camilla Böckelman, Ines Beilmann-Lehtonen, Tuomas Kaprio, Selja Koskensalo, Taina Tervahartiala, Harri Mustonen, Ulf-Håkan Stenman, Timo Sorsa, Caj Haglund
• Format: Article
• Language: English
• Published: BMC 2018-06-01
• Series: BMC Cancer
• Subjects: Colorectal cancer; Colon cancer; MMP-8; MMP-9; TIMP-1; Prognosis
• Online Access: http://link.springer.com/article/10.1186/s12885-018-4589-x

Abstract Background Almost all of the extracellular matrix (ECM) components can be degraded by the endoproteinases matrix metalloproteinases (MMPs). Important regulators of MMPs, and thereby of the extracellular environment, are tissue inhibitors of metalloproteinases (TIMPs), and especially TIMP-1. Early tumor development, as well as distant metastasis, may be results of an MMP/TIMP ratio imbalance altering the ECM. MMPs are elevated in several inflammatory conditions. Our aim is to investigate the prognostic role of MMP-8, − 9, and TIMP-1 in colorectal cancer (CRC) and their relationship to inflammation. Methods We included 337 colorectal cancer patients and 47 controls undergoing surgery at Helsinki University Hospital in Finland, 1998–2011. Serum levels of MMP-8 and plasma levels of C-reactive protein (CRP) were determined with a time-resolved immunofluorometric assay (IFMA), and MMP-9 and TIMP-1 with commercial enzyme-linked immunosorbent assay (ELISA) kits. Association and correlation analyses were performed with the Mann-Whitney U, Kruskal-Wallis, and Spearman rank correlation tests. Survival curves were constructed according to the Kaplan-Meier method and compared with the log-rank test. Results Among patients with advanced disease, serum levels of MMP-8 and TIMP-1 were elevated. CRC patients with high MMP-8 (HR (hazard ratio) 1.72, 95% confidence interval (CI) 1.17–2.52, P = 0.005) and those with high TIMP-1 (HR 1.80, 95% CI 1.23–2.64, P = 0.002) had worse prognoses. MMP-9 level failed to serve as a prognostic factor. In multivariable survival analysis, Dukes stage, and low MMP-9/TIMP-1 molar ratio (HR 0.46, 95% CI 0.33–0.98, P = 0.042) were independently predicted prognosis. A weak correlation between CRP and MMP-8 (rS = 0.229, P < 0.001), and TIMP-1 (rS = 0.280, P < 0.001) was noted. Among patients showing no systemic inflammatory response, MMP-8 (HR 1.66, 95% CI 1.10–2.53, P = 0.017) and TIMP-1 (HR 1.59, 95% CI 1.05–2.42, P = 0.029) were prognostic factors. Conclusions MMP-8 and TIMP-1 in serum, but not MMP-9, identified CRC patients with bad prognosis. Among patients showing no systemic inflammatory response, MMP-8 and TIMP-1 may associate with poor prognosis.