p21 and p27 are both G1-checkpoint CDK inhibitors and have an approximately 42% amino acid homology at the amino-terminal domain, which mediates inhibition of CDK and interacts with various CDK complexes . Alterations of expression or function of these two critical regulators in the cell cycle control has been implicated in the etiologies and disease prognosis of human malignancies . In this study, we found that each of three SNPs of p27 and p21 genes was associated with a significantly increased risk of SPM in patients with index SCCHN. We also observed significant associations of the combined variant genotypes of three polymorphisms with increased risk of SPM in patients with index SCCHN, and the trend in risk was statistically significant in a dose-response manner. In addition, the patients with either variant allele of three SNPs were more likely to develop a SPM than the patients with no variant genotypes. Approximately 15% of SCCHN patients develop a SPM, a significant cause of posttreatment morbidity and mortality for the diseases. Although the diagnostic and therapeutic approaches, such as surgery, radiotherapy, and chemotherapy, for SCCHN patients have been improved, the poor prognosis for SCCHN patients has not significantly improved, partly because of the high frequency of SPM. Therefore, p27 and p21 polymorphisms may serve as a marker for genetic susceptibility to SPMs after index SCCHN, and for identifying high-risk subgroups of SCCHN patients who might benefit from management of alternative treatment and predictable patient outcome for an improved survival and a better quality of life. Moreover, identifying markers of risk for SPM among cancer survivors would greatly enhance secondary prevention, which is currently limited to rather simplistic clinical post-treatment screenings.
The roles of both p27 and p21 proteins in modulating cell cycle regulation has been well established. For example, overexpression of p27 was reported to inhibit CDK activation and entry into the S phase of the cell cycle . As p27 may be involved in pathways regulated by both mitogenic and antiproliferative extrinsic signals, the expression level of p27 was found to be positively correlated with cell differentiation, and loss of its function may subsequently contribute to tumorigenesis [22, 23]. p27 is rarely mutated in human malignancies; however, reduced expression of p27 was found to be frequent in various types of cancers, such as gastric, breast, prostate, and non-small cell lung cancers [24, 25]. Furthermore, the reduced expression of p27 was also found to be correlated with poor clinical prognosis in head and neck cancers [26, 27]. On the other hand, p21 protein is a downstream target of p53. In response to DNA damage, increased expression of p21 following p53 activation leads to either cell-cycle arrest at the G1 checkpoint or apoptosis. Through inhibition of proliferating cell nuclear antigen-dependent DNA replication and mismatch repair in vitro, p21 expression can suppress tumor growth . Additionally, the overexpression of p21 may prevent mammalian cell proliferation and inhibit all cyclin-CDK complexes, suggesting that p21 is a universal inhibitor of cyclin-CDK complexes . Moreover, overexpression of p21 and the subsequent overall reduced CDK activity was found to be associated with cell differentiation . While somatic mutations in p21 gene are rare in human malignancies , reduced tumor expression of p21 has been associated with poor clinical prognosis . Because both p27 and p21 play important roles in cell cycle control regulation, it is suggested that alterations of p27 and p21 genes, such as above mentioned genetic polymorphisms, may, at least in part, reflect increased susceptibility of SCCHN patients to SPM. Recent studies have indicated that SNPs of genes in cell cycle control play an important role in carcinogenesis and may lead to altered susceptibility to different cancers, including primary SCCHN and their SPM [13, 15, 17, 18, 33–35]. Therefore, individuals carrying polymorphic CDK inhibitors that may affect its protein function are likely to be more susceptible to cancer.
Among p27 and p21 polymorphisms, the p27 T109G polymorphism is within a region (amino acids 97-151) physically interacting with the Jun activation domain-binding protein 1, which triggers proteolytic degradation of p27 . Therefore, it is speculated that this polymorphism may affect p27 degradation. So far, the p27 T109G polymorphism has been reported to be associated with risk and prognosis of prostate carcinoma and breast cancer [14–16], whereas the results were not consistent. These inconsistent results may be due to differences between studies of cancer types, patient population, and different risk factors for various types of cancers. Other factors in the studies such as small sample size, inclusion of different ethnic groups in a single study, gene-gene or gene-environment interactions, or inadequate adjustment for confounding factors could also cause the inconsistent results. The functional relevance of this p27 polymorphism needs further investigations as p27KIP1 exerts anti- and pro-tumorigenic activities . Likewise, both p21 C98A and C70T polymorphisms are thought to cause p21 functional change, because p21 C98A (at codon 31 in exon 2) may affect the DNA-binding zinc finger motif , while p21 C70T (within the 3' untranslated region in exon 3) lies in a crucial region for cell differentiation, proliferation and tumor suppression [39, 40]. Moreover, several previous studies have reported that these p21 polymorphisms may affect protein expression and activity and may have an effect on carcinogenesis [13, 18, 31]. However, the exact mechanisms for these observations remain unknown, and therefore, future studies on biological functions of these polymorphisms are needed.
Although the current study had a relatively larger cohort of SCCHN patients, there were few inherent limitations as previously described [9, 10, 18]. Briefly, our data on demographics, exposure, and clinical variables for the cohort were collected prospectively, but SPM outcome was recorded retrospectively. Because this cohort included multiple ethnicities, our results might not be generalizable to other ethnic populations. Additionally, because the time for SPM follow-up was limited due to the majority of patients with stage III and IV index cancer, these patients may have less opportunity to develop a SPM, and patients with late stage disease may be lost to follow-up or die relatively soon after recruitment or diagnosis. Furthermore, a screening bias for detecting tobacco-associated SPM or non-tobacco associated SPM might also exist. Finally, lack of tumor human papillomavirus (HPV) data and relatively low SPM rate might bias our estimates of association. Presently, we have not collected enough HPV data in our analysis for adjustment, and the low SPM rate might be due to a high proportion of never-smoker patients and use of strict criteria for SPM detection in this study. All of these confounding factors will be considered in our future studies.
In conclusion, our data suggest that p27 and p21 polymorphisms appear to alter individual susceptibility to SPM in patients with index SCCHN and that the p27 (T109G) polymorphism may individually or in combination with p21 (C98A and C70T) polymorphisms to increase risk of SPM in patients with an index SCCHN. However, future larger and well designed studies with longer follow-up time are needed to verify our findings.