Identifying local factors that predispose patients to early, pre-cancerous lesions may make it possible to stratify screening based on risk, but risk factors that occur early in CRC development are not well defined. Our group has shown that patients with adenoma have reduced apoptosis in their normal mucosa, demonstrating a field effect that predisposes the individual to a higher risk of developing precancerous adenomas [14
]. We hypothesized that SOCS3 might be a good biomarker of colorectal neoplasia risk, but contrary to expectation the study was negative.
Patients with inflammatory bowel diseases (Crohn's disease or ulcerative colitis) have an increased risk of developing CRC, which is associated with the degree and duration of intestinal inflammation (reviewed in [16
]). In addition, recent data suggest that individuals with chronic, low levels of inflammation (such as increased circulating IL-6 and TNFα) have increased odds of having adenoma and thus increased CRC risk [6
]. SOCS3 is an anti-inflammatory protein that limits IL-6 induction of STAT3, as well as TNFα induction of NFκB. IEC-specific silencing of SOCS3 leads to a dramatic increase in tumor load in a mouse model of inflammation-associated CRC [1
]. Furthermore, recent studies show that STAT3 activation is increased, and SOCS3 is silenced in tumors of patients with ulcerative colitis-associated and sporadic CRC [2
]. The current study tested the hypothesis that patients with adenoma may have lower SOCS3 in the normal mucosa than patients without adenoma, thus contributing to a permissive environment for aberrant growth. However, our study found that low or silenced SOCS3 expression does not occur in the normal mucosa of patients with colorectal adenoma.
One potential limitation of the study is possible effect of standard bowel preparation on SOCS3 expression. While we have not directly tested this for SOCS3, similar studies of other genes have shown that there is no significant change in mucosal gene expression in patients using bisacodyl or polyethylene glycol bowel preparations [18
]. Another possible limitation is that SOCS3 mRNA levels do not reflect changes in SOCS3 phosphorylation, which leads to its targeting for proteasomal degradation [19
]. However, other studies show that SOCS3 is silenced by DNA hyper-methylation in CRC tumors [2
], indicating that evaluating changes in gene expression is an appropriate measure of SOCS3 in this study. Finally, while our results could indicate that SOCS3 is more important in the underlying pathogenesis of inflammation-associated rather than sporadic CRC, recent studies comparing SOCS3 expression in both ulcerative colitis-associated and sporadic tumors found that SOCS3 was decreased and there was no significant difference between the two groups for SOCS3 [2
]. Taken together with these studies, we conclude that SOCS3 silencing occurs later in the progression from adenoma to adenocarcinoma, and is not an independent, early biomarker of CRC risk in the normal mucosa.