TRIM72 Immunohistochemical Expression Can Predict Relapse in Colorectal Carcinoma

Large bowel adenocarcinoma is one of the most frequent human neoplasms and despite recent insights into the pathophysiology and molecular basis of this disease, mortality remains high in advanced and metastatic cases. Most guidelines recommend adjuvant chemotherapy for tumours involving lymph nodes, but not for patients with localized stage I or II disease. However, it is well known that approximately 20% of stage II colorectal carcinoma patients eventually recur, mainly with distant or peritoneal involvement and show bad prognosis. It would be important to predict which patients are at increased risk of recurrence to guide potential adjuvant therapy use in this controversial setting. In this sense, only microsatellite stability has been proposed as a predictive tool in some guidelines. The tripartite motif family protein 72 (TRIM72) is a ubiquitin ligase, involved in the cell membrane repair machinery and known to be associated to insulin resistance. Its potential role in colon cancer has recently been proposed. The aim of this study is to determine the potential predictive value of TRIM72 immunohistochemical expression in stage II colon carcinoma. We have retrospectively reviewed a series of 95 patients with stage II colon microsatellite stable carcinomas operated with a curative intent at a single large tertiary hospital in Madrid (Spain) between 2006 and 2012. None of the patients received adjuvant chemotherapy. We reviewed the histopathological slides and constructed a tissue microarray (TMA) of three representative areas to perform immunohistochemical staining for TRIM72. In our series 30 patients (31.7%) recurred after a median follow-up of 17.5 months. Lack of immunohistochemical expression of TRIM72 in the tumor was significantly and independently associated to recurrence. A recent report by Chen et al. has shown that TRIM72 can be measured in plasma for colon carcinoma detection as an alternative to CEA or CA19.9, with lower levels in patients with carcinoma. Our report is the first one to show that lower immunohistochemical expression of TRIM72 predicts recurrence in colon stage II carcinoma. We feel this predictive influence can be related to its crucial role as a central regulator in many signaling pathways (PI3K-AKT, ERK). As an ubiquitin ligase, the lack of TRIM72 could increase the levels of several potential oncogenic molecules and therefore lead to a more aggressive phenotype. It remains to be shown whether chemotherapy could change the clinical behaviour of this bad prognosis group. We propose TRIM72 immunohistochemical analysis as a potential tool to predict recurrence risk in stage II colon carcinoma patients. Our results should be confirmed in larger series, but could open the way to management strategies refinement in this early stage group of patients.