br Table br Description of oral tumors with cyclin D
Description of oral tumors with cyclin D1 expression (n = 68a).
a Corresponding to 54 patients.
2.2. Statistical methods
SPSS-Windows 15.0 (SPSS Inc., Chicago, IL) was used for descriptive statistics (means, standard deviations and percentages or studied vari-ables). For comparing tumor cyclin D1 expression between categories of studied variables, we used SUDAAN 7.0 (Research Triangle Institute, Durham, NC, USA) to account for clustering (multiple OSCC's per pa-tient). We used a t-test adjusted for clustering (DESCRIPT procedure in SUDAAN) for variables with two categories (for example, tumor size after collapsing categories, etc), or ANOVA adjusted for clustering (REGRESS procedure in SUDAAN) for variables with more than 2 ca-tegories (ie, Stage, etc).
Association of clinical-pathological variables and tumor Ki-67 expression with tumor cyclin D1 expression (% cell count) in oral tumors (n = 68).
Variable na Nucleus
0.341 Degree of diﬀerentiation
a Diﬀerences with n = 68 due to missing values.
b DESCRIPT SUDAAN procedure 7.0 to account for clustering (multiple cancers within patients).
c REGRESS SUDAAN procedure 7.0 to account for clustering (multiple can-cers within patients). d r = Pearson correlation.
Table 3 displays the statistical associations found between the ex-pression of nuclear cyclin D1 and the clinicopathological variables under study. Nuclear cyclin D1 expression was significantly associated with high histological grade (p = 0.030). In tumors with cyclin D1 expression in both nucleus and cytoplasm (n = 23) (Fig. 1), the per-centage of immunopositive (±)-Baclofen was significantly associated with an
Fig. 1. The image depicts numerous tumor cells with nuclear expression (red arrow) or with both nuclear and cytoplasmic expression (yellow arrow). Some of the cells with nuclear and cytoplasmic expression show cytoplasmic pro-longations that characterize an invasive morphology (Ramos-García et al., 2018) (immunohistochemical technique; 40x) (For interpretation of the refer-ences to colour in this figure legend, the reader is referred to the web version of this article).
invasive morphology (p = 0.045). Nuclear expression of cyclin D1 was significantly associated with tumor cell ki-67 expression (p = 0.018).
Our study demonstrates that cyclin D1 overexpression in OSCC is significantly associated with clinicopathological parameters that are classically considered as indicators of poor prognosis, including a low degree of tumor diﬀerentiation, an elevated proliferation rate, and an invasive morphology of tumor cells.
Cyclin D1 is a key protein in oral oncogenesis, increasing cell pro-liferation (Ramos-García, Gil-Montoya et al., 2017; Ramos-García, Ruiz-Ávila et al., 2017; Sherr & Roberts, 2004) and exerting numerous emerging functions (Pestell, 2013), including the regulation of cell migration and mitochondrial metabolism and the inhibition of cell diﬀerentiation and DNA repair, considered hallmarks of cancer (Hanahan & Weinberg, 2011). Amplification of the CCND1 gene is the main oncogenic mechanism of cyclin D1 overexpression in OSCC (Ramos-García, Gil-Montoya et al., 2017; Ramos-García, Ruiz-Ávila et al., 2017), although its transcriptional activation can also be in-creased by the alteration of pathways involved in human carcinogenesis such as MAPK, Wnt, and NF-κβ, among others (González-Moles, Plaza-Campillo et al., 2014; Ramos-García, Gil-Montoya et al., 2017; Ramos-García, Ruiz-Ávila et al., 2017). These mechanisms lead to uncontrolled cell proliferation, which would hypothetically lead to the development of larger OSCCs and increase the risk of lymph node involvement (Hanahan & Weinberg, 2011; Leemans, Braakhuis, & Brakenhoﬀ, 2011; Ramos-García, Gil-Montoya et al., 2017; Ramos-García, Ruiz-Ávila et al., 2017). However, in the present and previous studies (Bova et al., 1999; Guimarães, de Carli, Sperandio, Hanemann, & Pereira, 2015; Huang et al., 2012; Jia, Wei, Gong, Gan, & Yu, 2013; Kuo et al., 1999; Maahs, Machado, Jeckel-Neto, & Michaelses, 2007; Mishra & Das, 2009; Miyamoto, Uzawa, Nagaoka, Hirata, & Amagasa, 2003; Shah et al., 2009; Shiraki et al., 2005; Soni et al., 2005; Vora, Shah, Patel, Trivedi, & Chikhlikar, 2003; Yun, Wang, Cao, Okada, & Miki, 2010), nuclear expression of cyclin D1 was not significantly associated with larger tumor size (Bova et al., 1999; Huang et al., 2012; Kuo et al., 1999; Mishra & Das, 2009; Shah et al., 2009; Shiraki et al., 2005; Soni et al., 2005; Vora et al., 2003; Yun et al., 2010) or with the presence of lymph node metastases (Guimarães et al., 2015; Jia et al., 2013; Kuo et al., 1999; Maahs et al., 2007; Miyamoto et al., 2003; Shah et al., 2009; Shiraki et al., 2005; Soni et al., 2005; Vora et al., 2003). Our recent meta-analysis (Ramos-Garcia et al., 2018; Ramos-García et al., 2018) on the prognostic value of cyclin D1 reported the influence of its overexpression on T and N status, especially in tongue cancer. Tongue carcinoma, which behaves diﬀerently from other intraoral carcinomas (higher T and N status and worse survival) may also be diﬀerent from a molecular perspective, especially with respect to cyclin D1 over-expression. Hence, discrepancies between the present findings and previous studies may be due to their inclusion of tumors from diﬀerent sites in which the importance of cyclin D1 may vary. We also observed a statistically significant association between the nuclear expression of cyclin D1 and low tumor diﬀerentiation degree, as also reported in our meta-analysis (Ramos-Garcia et al., 2018; Ramos-García et al., 2018). Molecular mechanisms regulating increased proliferation via cyclin D1 would alter the homeostasis of cells towards a more proliferative phe-notype (Ramos-García, Gil-Montoya et al., 2017; Ramos-García, Ruiz-Ávila et al., 2017), typically associated in cancer with an immature and oncogenically more aggressive cell phenotype (Hanahan & Weinberg, 2011). Some studies have suggested a putative role for cyclin D1 in the inhibition of cell diﬀerentiation in various cell lineages, either through the formation of complexes with their binding partners CDKs 4 and 6 (Nakajima et al., 2011; Shen et al., 2006) or independently of CDK activity (Nishi, Inoue, Schnier, & Rice, 2009; Ratineau, Petry, Mutoh, & Leiter, 2002). A recent study by our group (Ramos-García et al., 2018) observed a less diﬀerentiated morphology in OSCC cells with cyclin D1