Low expression of chloride channel accessory 1 predicts a poor prognosis in colorectal cancer: The question is still open.

Abstract

We read with interest the findings by Yang et al, who aimed to explore the relationship between the level of chloride channel accessory 1 (CLCA1) and the prognosis of patients with colorectal cancer (CRC).1 They concluded that the CLCA1 level is a potential predictor of prognosis in primary human CRC, and low expression of CLCA1 predicts disease recurrence and lower survival, suggesting its role in selecting those patients most likely to benefit from adjuvant chemotherapy.1 Within the context of personalized medicine, the identification of molecular markers that are useful as prognostic biomarkers that can predict response or facilitate the clinical management of patients with CRC is warranted. Despite rapid advances in the development of prognostic biomarkers or marker-guided therapy, to the best of our knowledge only a very small number of markers have been applied for patients with CRC.2 Therefore, although we appreciate the findings by Yang et al, it is our opinion that some aspects of the study need to be discussed in more detail. In their study, Yang et al evaluated the expression of CLCA1 in 36 patients with CRC with different clinical characteristics, including age, stage of disease, and histological grade.1 First, the low number of patients, nonrandomized selection of patients, and lack of validation of emerging data in an independent cohort (preferably multicenter settings) severely limit the conclusions reached and justify the need for the confirmation of their findings to illustrate the reproducibility and sensitivity of CLCA1 expression.3 Second, the statistical methods used for data analysis do not appear to be clearly described; to our knowledge, whether Yang et al1, 2 performed multivariate analyses to demonstrate that the candidate marker was an independent predictor of response or disease recurrence and whether they adjusted for different variables with known prognostic value in CRC, such as grade and lymph node status, remain unknown. In addition, a Bonferroni correction considering all the variables and clinical features of the population as well as the studied biomarker would be required to avoid false-positive associations.4 In particular, using the Bonferroni method, a universal null hypothesis should be postulated in which ≥2 groups are identical compared with all the variables, and could provide a correct answer if the variants were independent, but that was not the case in the study by Yang et al.1 Colon cancer tissues are characterized by a cellular heterogeneity representing a mixture of normal epithelial ducts, stroma, invasive or in situ tumor cells, and infiltrating immune cells. Because of the tumor heterogeneity within tissues, the enrichment of CRC cells from tissue biopsies, even within the same clinical subgroup of patients, using laser microdissection would help to isolate a relatively pure population and decrease false-positive, tumor cell-specific signals.5, 6 Moreover, the evaluation of tumor heterogeneity and possible evolution of cancer cells after disease recurrence should be documented within multiple samples of a single tumor and repeated biopsies performed to decrease the risk of avoidable mistakes.5-7 In addition, immunohistochemistry is a sensitive and empirical method and outcomes depend on a pathologist's visual expertise and the antibody used. Thus, validation of the results of the study by Yang et al1 and reevaluation of the slides by at least 2 independent pathologists may decrease the risk of error,7 although it is beneficial to know whether those who performed the data analysis and interpretation were also blinded to the experimental part. Last, although the authors explored the functional role of CLCA1 in Caco-2 cells,1 it would nevertheless be worthwhile to perform the in vitro experiments in primary cell cultures or a freshly generated xenograft from tumor tissues because these models have been shown to better resemble the genetic characteristics of disease and predict drug activity.8, 9 We thank Yang et al for their worthwhile findings, but believe that additional parameters and analysis with the incorporation of proper methodology, larger populations, and successive powered statistical analysis are essential to validate the candidate biomarker beyond already available clinical factors for the clinical management of patients with CRC. No specific funding was disclosed. The authors made no disclosures. Soodabeh ShahidSales, MD Cancer Research Center School of Medicine Mashhad University of Medical Sciences Mashhad, Iran Majid Ghayour Mobarhan, MD, PhD Faezeh Ghasemi Department of Modern Sciences and Technologies Biochemistry of Nutrition Research Center School of Medicine Mashhad University of Medical Sciences Mashhad, Iran Sharareh Gholamin, MD Institute of Stem Cell Biology and Regenerative Medicine Stanford University School of Medicine Stanford, California Amir Avan, PhD Department of Modern Sciences and Technologies Biochemistry of Nutrition Research Center Cancer Research Center School of Medicine Mashhad University of Medical Sciences Mashhad, Iran