Circulating syndecan-1 and glypican-4 predict 12-month survival in metastatic colorectal cancer patients

Abstract

Cell surface syndecans and glypicans play important roles in the development and prognosis of colorectal cancer (CRC). Their soluble forms from proteoglycan shedding can be detected in blood and have been proposed as new prognostic biomarkers in several cancer entities. However, studies on circulating syndecan-1 (SDC1) and glypican-4 (GPC4) in CRC are limited. We, therefore, evaluated the impact of plasma SDC1 and GPC4 on the prognosis of metastatic (m)CRC patients. The present study included 93 patients with mCRC. The endpoints were progression-free survival (PFS) and overall survival (OS) at 12 months. SDC1 and GPC4 levels were measured in plasma using enzyme-linked immunosorbent assays. Plasma levels of SDC1 and GPC4 were significantly correlated. Significant correlations of these two markers were also found with carcinoembryonic antigen (CEA). Kaplan-Meier curve analyses indicated that PFS and OS probabilities significantly decreased with increasing levels of SDC1 and GPC4, respectively. Multivariable Cox regression analyses showed that both markers were significantly associated with PFS and OS independently from clinicopathological characteristics including CEA. Respective adjusted hazard ratios (HR) together with corresponding 95% confidence intervals for one standard deviation change of SDC1 were 1.32 [1.02-1.84] for PFS and 1.48 [1.01-2.15] for OS. Adjusted HRs [95% confidence intervals] of GPC4 were 1.42 [1.07-1.89] for PFS and 2.40 [1.51-3.81] for OS. Results from area under the receiver operating characteristic curve analyses suggest that GPC4 and SDC1 add additional prognostic values to CEA for OS. In conclusion, we showed significant associations of circulating SDC1 and GPC4 with poor survival of mCRC patients.

Keywords: biomarker; colorectal cancer; glypican-4; prognosis; proteoglycans; shedding; syndecan-1 (SDC1).

Figure 1

Plasma levels of SDC1 and GPC4 with respect to disease progression and mortality. Plasma levels of (A) SDC1 and (B) GPC4 with respect to disease progression and mortality are shown as box plots using the Tukey method for plotting the whiskers and outliers. P-values were calculated using the Mann-Whitney U test.

Figure 2

Kaplan Meier estimates of survival according to tertiles of SDC1 and GPC4. Kaplan–Meier curves of (A) PFS stratified by tertiles of SDC1 (B) OS stratified by tertiles of SDC1 (C) PFS stratified by tertiles of GPC4 (D) OS stratified by tertiles of OS. P-values were obtained by Log-Rank-Mantel-Cox-tests. SDC1, syndecan-1; GPC4, glypican-4; PFS, progression-free survival; OS, overall survival.

Figure 3

Association of plasma SDC1 and GPC4 with progression free survival and overall survival. Results were obtained from Cox proportional hazards regression analyses and are presented as hazard ratios (HR) and 95% confidence intervals [CI] for one standard deviation change of (A) SDC1 and (B) GPC4. Model 1 includes age, sex, number of metastatic sites, and scheduled therapy; model 2 additionally includes CEA. PFS, progression-free survival; OS, overall survival; SDC1, syndecan-1; GPC4, glypican-4; CEA, carcinoembryonic antigen.