Cyclin D1 and p16 expression in recurrent nasopharyngeal carcinoma

Nasopharyngeal carcinoma (NPC) is a prevalent malignancy in Southeast Asia with reported incidence rate ranging from 10 to 53 cases per 100,000 persons per year [1]. Although it is a relatively uncommon cancer in western countries with an incidence rate of less than 1 case per 100,000, it poses a significant health problem in regions of the US where there is large population of Asians. External beam radiation therapy is considered the primary mode of treatment for previously untreated NPC. This is mainly due to the high degree of sensitivity of this tumor to radiation as well as the anatomic constraints for surgical access in this highly complex region. Despite advances in diagnostic and treatment modalities, loco-regional failure is still significant with reported rates of 15.6% to 58% [1, 2]. The addition of chemotherapy concurrent with radiation has been shown to improve survival in patients with NPC [3].

Although the molecular events responsible for the pathogenesis of NPC remain to be elucidated, the final common pathway appears to be a disruption of the mechanisms involved in the regulation of cell cycle progression, leading to uncontrolled cell proliferation. The most critical point in cell cycle regulation is the G₁ checkpoint. It is here that complex interactions take place to determine whether the cell will exit the cell cycle and go into a quiescent state (G₀) or enter into the S phase and proceed with cell division [4]. These complex interactions involve a large number of regulatory proteins such as cyclins, cyclin-dependent kinases (CDKs), and CDK-inhibitors. Among the cyclins involved in G1 phase, cyclin D1 appears to be most strongly implicated in human carcinogenesis. Cyclin D1, through its interaction with cyclin-dependent kinase-4 or 6 (CDK-4/6), forms a complex that inactivates the tumor suppressor protein retinoblastoma (pRb) through phosphorylation [4]. Phosphorylation of pRb releases transcription factors such as E2Fs which then activate a series of events that allow entry into S-phase and cell division [4]. An increased level of cyclin D1 expression has been reported in a number of malignancies [5] including esophageal, ovarian, breast, uterine, colon, lung, prostate, lymphoma, as well as head and neck cancers. Between 30% to 83% of head and neck squamous cell carcinoma (HNSCC) were reported to be associated with cyclin D1 over-expression [6, 7]. The progression of cells from G1 to S phase, on the other hand, is blocked by a potent tumor suppressor protein, p16, which acts to disrupt the cyclin D1/CDK-4/6 complex. Deletions, mutations, or methylation of the p16 gene has been implicated in the development of a variety of human malignancies, including head and neck cancer [6, 8].

The disruption of the "Retinoblastoma/cyclin D1/p16 pathway" involved in the regulation of the G1 checkpoint also appears to play an important role in the tumorigenesis of NPC. Over-expression of cyclin D1 and/or loss of p16 has been reported in several studies on NPC. These studies found an increased level of cyclin D1 expression in 30% [7] to 66% [9] and the lack of expression of p16 in 40% to 70% of previously untreated NPC specimens [9‐12].

Although a large body of data exists on cyclin D1 and p16 expressions in HNSCC including NPC, to the best our knowledge, no previous studies have been reported on the expression level of these proteins in recurrent HNSCC. We set out in this study to examine the expression level of cyclin D1 and p16 in recurrent NPC that have failed previous treatment with radiation +/- chemotherapy.

Twenty-seven recurrent NPC specimens were examined in this study. The average age at the time of salvage nasopharyngectomy was 51 years (range, 27 to 72 years). There were 18 men and 9 women. Thirteen patients were Asian and 14 were non-Asians. Patients were restaged using the 1998 AJC Cancer Staging Manual and 16 were staged as recurrent (r)T1, 2 as rT2, 5 as rT3, and 4 as rT4. Other relevant clinicopathological characteristics of these patients were shown in additional file 1.

The results of the immunohistochemical staining of these 27 recurrent NPC specimens were presented on additional file 1. Cyclin D1 over-expression was limited to tumor cells and no immunostaining was noted on the surrounding normal epithelium of the nasopharynx. Thus, the normal epithelium served as internal negative control. On the other hand, normal tissue stained positive for p16 and was used as internal positive control. Representative samples of cyclin D1 and p16 immunostaining were shown in Figure 1. Absence of staining for cyclin D1 and p16 were illustrated in panel C and D respectively. In Panel E and F, 3+ immunostaining was demonstrated for cyclin D1 and p16 respectively. Of the 27 recurrent NPC specimens, 7 (26%) stained positive for cyclin D1 proteins while only 1 (4%) demonstrated immunostaining for p16. Of the 7 specimens that stained positive for cyclin D1 protein, 1 stained strongly (3+), 3 stained moderately (2+), and 3 stained weakly (1+).

Length of follow-up ranged from 5 months to 164 months, with mean of 45 months. Four patients were lost to follow-up and were not included in the survival analysis. At the time of last contact, patients were classified as no evidence of disease (NED), alive with disease (AWD), or died of disease (DOD). The 5-year disease-free survival following salvage nasopharyngectomy for the 23 patients (excluding the 4 patients that were lost to follow-up) was 50% (Figure 2). Using the Log-rank test, no statistical significance in disease-free survival was found to correlate with variables such as gender, staining with cyclin D1 (Figure 3), staining with p16, WHO classification (Figure 4), or recurrent stage (Figure 5). However, there was a statistically significant difference (p = 0.003) in 5-year disease-free survival between Asian (88.9%) and non-Asian patients (22.2%) (Figure 6).

In this study, we examined the expression level of cyclin D1 and p16 in recurrent NPC that have failed previous treatment with radiation +/- chemotherapy. These specimens were procured at the time of salvage nasopharyngectomy. We found over-expression of cyclin D1 in 26% and the lack of expression of p16 in 96% of recurrent NPC specimens. While the level of over-expression of cyclin D1 (26%) in recurrent NPC specimens was similar to the levels of over-expression (30%[7] to 66%[9]) found in primary untreated NPC, the lack of expression of p16 in recurrent NPC specimens (96%) was significantly higher than those typically reported for primary untreated NPC (40% to 70%) [9‐12].

There are a number of studies that attempted to correlate the level of expression of cyclin D1 and p16 in cancer with prognosis. Several investigators reported association between over-expression of the proto-oncogene cyclin D1 with worse prognosis in terms of shorter survival and increased incidence of recurrence [14‐16]. This poor prognosis in tumors over-expressing cyclin D1 was found to be independent of other clinicopathological parameters such as tumor stage, nodal or distant metastasis, or degree of differentiation [14‐16]. Other studies however either failed to show this association between over-expression of cyclin D1 and poor clinical outcome [17] or found a reverse association between over-expression of cyclin D1 and poor prognosis. In a series of 65 patients, Hwang et al [9] reported that loss of cyclin D1 was closely related to local recurrence after radiation therapy for NPC. Our data also failed to support the association between over-expression of cyclin D1 with increased risk of recurrence. If over-expression of cyclin D1 was associated with increased risk of recurrence, we would expect to find a higher proportion of recurrent NPC specimens to over-express cyclin D1 compared to previously untreated NPC. However, the level of over-expression of cyclin D1 in recurrent NPC specimens found in this study (26%) was not higher, but slightly lower than those reported for previously untreated NPC (30% to 66%) [7, 9, 18].

The other interesting finding from this study was the lack of p16 expression in almost all (96%) of the recurrent NPC specimens. This was in contrast to the 40% to 70% reported rate of p16 loss in previously untreated NPC (table 1). This high proportion of recurrent cancer with loss of p16 activity suggested that p16 loss may confer survival advantage by making these cancer cells more resistant than their p16 positive counterparts to conventional treatment with radiation +/- chemotherapy. Other investigators have also reported on the association between loss of p16 expression and worse prognosis. Hwang et al [9] reported on a statistically significant correlation (p = 0.047) between absence of p16 expression with higher rate of local recurrence in NPC. In another study involving 84 NPC samples, Makitie et al [12] found a statistically significant association (p = 0.022) between absence of p16 expression and inferior overall survival rate. The absence of p16 expression found in nearly all of our recurrent NPC samples in this study seemed to corroborate the finding of these investigators. Taken together, these findings would suggest that the loss of p16 tumor suppressor proteins in NPC may confer resistance to radiotherapy and may be associated with higher rate of failure after therapy with radiation +/- chemotherapy.

Finally, it is interesting to note that Asian patients, following treatment with salvage nasopharyngectomy, have a statistically significant better 5-year disease-free survival compared to Caucasians and Hispanics (p = 0.003) (Figure 5). The reason for this difference is unclear. The Asian and non-Asian group did not differ significantly in their level of cyclin D1 expression. Four out of 13 Asians and 3 out of 14 non-Asians stained positive for cyclin D1.

Although this is the first study looking at the expression levels of cyclin D1 and p16 in recurrent NPC, this study has several limitations. First, pre-irradiated NPC specimen were not available for use as control due to the fact that most of these patients were referred from outside institutions (in many cases, from institutions outside of US) after treatment failure with radiation therapy. The unavailability of pre-irradiated NPC specimen raised the issue of whether the lack of expression of p16 in these recurrent NPC samples was pre-existing (pre-treatment) or the result of selection pressure from treatment. The data presented in this study will not be able to answer this question. Our assumption here was that these recurrent tumors would have the same staining characteristics showing absence of p16 expression even before treatment with radiation. Irrespective of whether the p16 expression was lost during tumor transformation (before treatment) or during treatment with radiation, this marked reduction of p16 in recurrent NPC specimen underscored the importance of p16 alteration in NPC and provided support for therapeutic strategies that aim to restore p16 expression. There have been pre-clinical studies which demonstrated the ability of exogenously introduced p16 to inhibit proliferation and induce apoptosis in NPC cell culture [19]. Another limitation of this study was the small number of cases (n = 27) examined. However, both the low prevalence of NPC in the US and the reluctance of most head and neck surgeons to surgically salvage recurrent NPC made access to a large number of surgically removed recurrent NPC specimen available for immunohistochemical study difficult.

In this study, we have determined the level of expression of cyclin D1 and p16 in recurrent NPC that have failed primary treatment with radiation +/- chemotherapy. While the level of expression of cyclin D1 in recurrent NPC was similar to that of previously untreated HNSCC, almost all of the recurrent NPC samples lacked expression of p16 protein. This finding suggested that the loss of p16 tumor suppressor protein in NPC may be associated with higher rate of failure after treatment with radiation +/- chemotherapy. Thus, p16 expression level appeared to be a clinically useful prognostic marker in NPC and warranted further investigations. Further, therapeutic strategies targeting the p16 pathway may be a promising approach to improve the treatment outcome of NPC. Several promising studies have demonstrated the ability of exogenously introduced p16 to inhibit proliferation and induce apoptosis in NPC cell culture [19].