It has been recognized that tumors arise and progress as a result of multiple genetic alterations, such as chromosomal aberrations, DNA changes (e.g., mutations, amplifications, or deletions), and/or mRNA alterations through epigenetic changes . In HNSCCs including oral SCCs, exposure to environmental agents, such as tobacco smoke, alcoholic beverages, and viruses, including human papillomavirus, may affect the process of carcinogenesis [2, 20]. Previous studies have demonstrated that the karyotypes of human oral SCC cells are near-triploid and contain numerical and structural chromosomal abnormalities including balanced and unbalanced translocation, deletions, dicentric chromosomes, gene amplification in the form of extra-chromosomal double minutes or intra-chromosomal regions of homogeneous staining . Although these chromosomal aberrations are clonal genetic changes, many numerical and structural variations were detected in the oral SCCs, suggesting that this kind of cancer exhibits a high rate of CIN, and that CIN may play an important role in their tumorigenesis. In this study, we investigated CIN status using FISH on FNA biopsy samples from primary oral SCCs, and analyzed the association between CIN grade and clinical and histopathological factors. Our findings clearly demonstrated that CIN grade may be a significant predictor of recurrence and poor outcome in patients with this malignancy.
CIN has been observed in almost all solid tumors and studied by various methods, including FISH. The relationship between CIN status determined by FISH and poor prognosis has been studied in a number of human malignancies, including lung cancer , malignant astrocytic tumor , and lymphoma . These studies have demonstrated that CIN status can be effectively detected using FISH analysis and is significantly associated with poor prognosis. For example, Nakamura et al. reported that 28% of non-small cell lung cancers they investigated had heterogeneity of all four chromosomes (chromosomes 3, 10, 11, and 17) examined, and were judged to be carrying CIN . Using univariate and multivariate analysis, they also showed that CIN was strongly associated with a worse prognosis, suggesting that CIN can be considered an independent indicator of poor prognosis in non-small cell lung cancer. With regard to head and neck cancers, including oral SCC, a number of studies have investigated numerical chromosomal alterations by means of FISH. Hardisson et al. characterized numerical aberrations in routine surgical specimens of 50 primary human SCCs of the pharynx and larynx using FISH with DNA probes specific for the centromeric sequences in chromosomes 8, 9, 11, and 17 . They showed that numerical abnormalities of these chromosomes were present in the vast majority (92%) of these cancers. Although they clearly demonstrated an overall positive correlation between DNA ploidy determined by DNA flow cytometry (FCM) and numerical chromosomal aberration determined by FISH, they did not examine the relationship between chromosomal status and clinical and histopathological factors including recurrence and survival. Soder et al. analyzed numerical chromosomal changes during the progression of HNSCC from low-stage non-metastasizing tumors to high-stage metastasizing tumors and lymph node metastasis using the FISH technique with 6 centromeric DNA probes (for chromosomes 1, 7, 9, 11, 17, and 18) . They demonstrated a high correlation (p <.0001) between tumor metastatic potential and aneuploidy. Unfortunately, however, they did not examine whether these chromosomal aberrations have any impact on the clinical outcome of this cancer. Moreover, Bergshoeff et al. have reported that CIN detection by FISH employing two centromeric DNA probes (foe chromosomes 1 and 7) in oral SCC is strongly associated with regional tumor outgrowth (p =.018) . However, they also did not demonstrate a correlation between numerical chromosomal changes and prognosis. Thus, although these studies investigated numerical chromosomal aberrations in oral SCC using FISH with DNA probes specific for several centromeric sequences, they did not examine the correlation between CIN, clinical and histopathological factors, and outcome. In the present study, therefore, to establish a simple and practical method for evaluation of CIN grade in primary oral SCCs, we selected only three chromosomes 7, 9, and 11 and investigated their numerical aberrations using FISH for interphase nuclei obtained by FNA biopsy, together with the possible impact of CIN status on clinical outcome.
We demonstrated that high-grade CIN was significantly correlated with poorer outcome (OS) by univariate (p =.003) and multivariate analysis (p =.041). Why, then, is CIN status in oral SCC associated with a poorer prognosis? There are at least two possible answers. First, in general, CIN may induce accumulation of genetic alterations. CIN gives rise to an increased rate of loss or gain of whole chromosomes or large parts of chromosomes during cell division. Consequently, this may result in an imbalance of chromosome number (aneuploidy) and an increased rate of loss of LOH . These genetic alterations may easily accelerate changes in the expression of many types of cancer-related genes such as oncogenes and tumor suppressor genes. Therefore, genetic alterations of genes that may contribute to acquisition of a malignant phenotype are more frequent in cancer cells with high-grade CIN than in those without. Another possibility is that abnormalities of several significant genes located on chromosomes 7, 9, and 11 may affect the progression of oral SCCs. Epidermal growth factor receptor (EGFR), p16 (CDKN2A), and cyclin D1 (CCND1), which may play an important role in the tumorigenesis and progression of this cancer, are located on chromosomes 7, 9, and 11, respectively. Our previous studies clearly demonstrated that numerical aberrations of these genes are a reliable predictor of recurrence and outcome in oral SCCs [17, 18, 28–30]. We have tried to compare chromosome number aberrations (this study) with copy numbers of these three cancer-related genes localized on the same chromosomes (previous studies), both performed on the same samples. As a result, the concordance rate of chromosome 7 number and EGFR gene copy number were 65-100%(average: 93%), chromosome 9 and CDKN2A gene copy number were 17-100% (average: 81%), and chromosome 11 and CCND1 gene copy number were 18-100% (average: 80%). These findings have indicated that chromosome number aberrations may significantly effect on the gene copy number on the same chromosome. However, only a limited number of samples, chromosomes, and genes were investigated in this study. Moreover, there were very large variations in the concordance rate of each sample. For examples, with regard to chromosome 11 and CCND1 gene, although the average rate is 80%, the minimum rate is only 18%. Therefore, to clarify the association between CIN and copy number aberrations of these cancer-related genes, further investigations are needed.
Several mechanisms have been proposed to explain the induction of CIN in cancer cells, including defects in chromosomal segregation, telomere stability, cell cycle checkpoint regulation, and repair of DNA damage . Although a large number of genes responsible for CIN have been identified in yeast, only a few such genes have been determined in humans [31–33]. These genes include hBUB1, MAD2, BRCA1, BRCA2, and hCDC4 [34–38]. Although mutations of these genes have been reported, such defects are rare during carcinogenesis in many kinds of human malignancies. The present findings suggest that the majority of these genetic defects occur as a result of altered expression of known genes that contribute to the development of CIN or mutation in, as yet, unidentified genes. Recently, several studies have indicated that the spindle protein NuMA has been shown to be critical for spindle assembly, and plays a key role in multipolar spindle formation, a frequent cause of CIN in cancer cells. Moreover, the NUMA1 gene amplification and overexpression has been reported in oral SCC, suggesting abnormalities of this gene may contribute to the development of CIN in this cancer [39, 40]. However, additional examinations are required to clarify the mechanism for CIN in cancer cells.