Higher Insulin-like Growth Factor Binding Protein 3 Is Associated with Increased Incidence of Colorectal Cancer in Older Men Independently of Conventional Risk Factors and of Insulin-like Growth Factor 1

Presentation Number: OR21-5
Date of Presentation: April 2nd, 2017

Yi Xian Chan*1, Paul Chubb2, Ken Ho3, Graeme J Hankey1, Gerry Fegan4 and Bu Beng Yeap1
1University of Western Australia, Perth, Australia, 2Fiona Stanley Hospital, Perth, Australia, 3Princess Alexandra Hospital/ University of Queensland, Brisbane, Australia, 4Fiona Stanley Hospital, Perth, AUSTRALIA


Abstract: Insulin-like growth factor 1 (IGF1) has anabolic and growth-promoting effects. There are theoretical and clinical concerns regarding its potential to promote tumour growth. Circulating IGF1 is bound to binding proteins (IGFBPs), which modulate bioavailability of IGF1. Furthermore, IGFBPs may also inhibit or potentiate actions of IGF1, and IGFBP3 has been demonstrated to have both proliferative and apoptotic effects on colon cancer cell lines (1). We hypothesised that IGF1 and its binding proteins are differentially associated with the incidence of common cancers. This was a prospective cohort study assessing associations of circulating total IGF1 and its binding proteins (IGFBP1 and IGFBP3) with the incidence of any cancer, prostate, colorectal and lung cancer, and cancer-related deaths in community-dwelling older men. Plasma total IGF1, IGFBP1 and IGFBP3 (DSL, Beckman Coulter, Gladesville, Australia) were measured in 4042 men aged 70 years and above. Cancer outcomes from 2001-2004 until 20th June 2013 were ascertained using data linkage. Analyses were performed using proportional hazards competing risks models, and adjustments were made for conventional risk factors and potential confounders. Results are expressed as subhazard ratios (SHR). There were 907 men who were diagnosed with cancer during a median of 9 years follow up. Of these, there were 359, 139 and 125 men who were diagnosed with prostate, colorectal and lung cancers respectively. The number of men fasted at the time of blood sampling was 3167 (78.4%) men. After adjustments, total IGF1 was not associated with the incidence of any cancer (SHR=0.96, 95% CI 0.89, 1.04; p=0.308 for every 1 standard deviation increase in IGF1), prostate or colorectal cancer. In the univariate analysis, higher total IGF1 was associated with a lower incidence of lung cancer (SHR=0.82, 95% CI 0.67-0.99; p=0.038), however this was not significant after adjusting for potential confounders (SHR=0.84, 95% CI 0.68, 1.03; p=0.097). In the fully-adjusted model, higher IGFBP3 was associated with increased incidence of colorectal cancer (SHR=1.20, 95% CI 1.01, 1.43; p=0.041). This effect was not attenuated by inclusion of total IGF1 into the multivariate model (SHR=1.28, 95% CI 1.03, 1.58; p=0.025), or when further adjustments were made for fasting status (SHR=1.27, 95% CI 1.03, 1.58; p=0.028). There were no other associations of IGFBP1 or IGFBP3 with cancer incidence. Neither total IGF1, IGFBP1 or IGFBP3 were associated with cancer-related deaths. In conclusion, higher IGFBP3 predicted increased incidence of colorectal cancer in older men. This association was independent of conventional risk factors and total IGF1. Therefore, IGFBP3 may be a biomarker for or contributing factor to incidence of colorectal cancer. Further studies are warranted to clarify the direction of causality and explore potential underlying mechanisms.


Nothing to Disclose: YXC, PC, KH, GJH, GF, BBY