Tumor Budding in Colorectal Carcinoma Showing a Paradoxical Mitotic Index (Via PHH3) With Possible Association to the Tumor Stromal Microenvironment.

Abstract

Colorectal carcinomas (CC) are one of the most commonly diagnosed malignancies. Tumor budding (the histologic process of dissociation that occurs at the invasive margin of colorectal cancer), has significant prognostic implications, in that higher tumor budding is associated with adverse histopathologic and clinical outcomes. Because of this prognostic significance, more research is needed to further understand the pathologic and immunohistochemical (IHC) associations pertaining to this important prognostic variable. In this study, we will further evaluate selective clinopathologic and IHC variables with possible association to tumor budding. A total of 234 cases of CC diagnosed in our health system were retrospectively reviewed and routine hematoxylin and eosin-stained slides of these cases were collected. A representative slide for tumor budding was selected per case and selective IHC staining was performed. Clinicopathologic data were collected for each case and analyzed in relation to tumor budding scores. In exploratory analyses, tumor budding scores per individual investigator and consensus tumor budding scores were compared with selected IHC stains (MLH1, PMS2, and PHH3) as well as numerous clinicopathologic variables. We found a paradoxical association between tumor budding and mitosis score using PHH3 immunostaining in univariate and multivariable analysis. Furthermore, patients with intact nuclear expression for MLH1 and/or PMS2 are more likely to have higher tumor budding compared with patients with lost expression. For multivariable analysis, the following covariates were significantly associated with higher tumor budding: the presence of lymphovascular invasion, higher pathologic tumor stage, and finally infiltrating border was more likely to be associated with higher tumor budding compared with cases with a pushing border. Regarding nonmucinous versus mucinous CC, nonmucinous adenocarcinoma (MCA) was more likely to be associated with higher tumor budding compared with MCA. Numerous clinicopathologic variables were found to be associated with tumor budding including lymphovascular invasion, tumor stage, infiltrating tumor border, non-MCA was more likely to be associated with higher tumor budding compared with MCA, possibly related to MUC-2 and MSI. Furthermore, regarding the paradoxical association between tumor budding and mitosis score using a PHH3 immunostaining (high tumor budding having lower mitosis), this is possibly related to the tumoral stomal microenvironment and cancer associated fibroblasts. An idea for a future study would be to look at the maturity of cancer-associated fibroblasts (immature vs. mature) and the tumoral stroma microenvironment, with regards to markers of tumor aggressiveness such as mitosis. In addition, we found that patients with intact nuclear expression for MLH1 and/or PMS2 were more likely to have higher tumor budding compared with patients with lost expression, possibly related to mismatch repair CC's not being as reliant on tumor budding. Future research will hopefully concede further insight into the variables that affect tumor budding, especially regarding the tumoral microenvironment and variations between different patient populations, inclusive of patients lacking activity of the mismatch repair. Ultimately, this will allow for better prognostic information, and more precise treatment modalities.