| Sign In to gain access to subscriptions and/or personal tools. |
Survivin, a Potential Early Predictor of Tumor Progression in the Oral Mucosa
1 Institute of Dental Sciences, Faculty of Medicine, University of Ancona, Ancona, Italy; Correspondence: * corresponding author, Via Carelli 28, 71100 Foggia, Italy, lomuziol{at}tin.it or llomuzio{at}tin.it
Survivin is a recently described apoptosis inhibitor selectively over-expressed in most tumors. Immunohistochemistry was used to investigate a potential role of survivin as an early predictor of malignant transformation in precancerous and cancerous lesions of the oral cavity. Survivin was present in 10/30 cases (33%) of oral precancerous lesions without malignant progression, and in 15/16 cases (94%) of oral precancerous lesions evolved into full-blown squamous cell carcinoma. Tumors that progressed from these precancerous lesions retained widespread survivin positivity (100%). Variations among group means were highly statistically significant (p < 0.001). No significant correlation was found between survivin expression and the degree of dysplasia. High expression of cytoplasmic/nuclear survivin is an early event during oral carcinogenesis and may provide a useful tool for the identification of precancerous lesions at higher risk of progression into invasive carcinoma.
Key Words: survivin precancerous lesion IAP oral squamous cell carcinoma apoptosis
Oral squamous cell carcinoma (OSCC) is the most frequent malignant tumor of the oral cavity (90% of all tumors), and the seventh most frequent cancer in humans. It is now accepted that most OSCC arise in the presence of clinical premalignant lesions (Scully, 1995). Erythroplasias and dysplastic leukoplakias are the most frequent potentially malignant lesions, and about half of OSCCs exhibit associated leukoplakia (Schepman et al., 1999). The presence of severe epithelial dysplasia, morphologically characterized by enlarged nuclei and eosinophilic nucleoli, hyperchromatism, dyskeratosis, and aberrant mitoses, is suggestive of malignant transformation. Although up to a third (3-33%) of oral precancerous lesions will eventually evolve into invasive OSCC over a 10-year interval, no reliable histopathological parameters have been identified that predict their potential for subsequent transformation. A prophylactic surgical management is often impractical, especially in patients with multiple and extensive precancerous lesions. Therefore, novel molecular predictors of malignant progression are needed to identify oral precancerous lesions at greater risk of invasive transformation as candidates for surgical intervention. Considerable interest has focused on the identification of regulators of programmed cell death, or apoptosis, which may influence the cell death/cell viability balance in cancer. In particular, deregulation of apoptosis resulting in aberrantly reduced cell death is thought to participate in cancer by facilitating the insurgence of additional transforming mutations. In addition to pro- and anti-apoptotic bcl-2 molecules, a second gene family of inhibitors of apoptosis (IAP) has recently been identified (Salvesen and Duckett, 2002). Highly evolutionarily conserved from viruses to mammalian cells (Deveraux and Reed, 1999; Salvesen and Duckett, 2002), certain IAP proteins target a downstream step in apoptosis by inhibiting the terminal effectors caspase-3 and -7 and by interfering with processing/activation of the pinnacle caspase, caspase-9 (Deveraux and Reed, 1999; Salvesen and Duckett, 2002). Survivin is a recently characterized IAP protein (Ambrosini et al., 1997), which is found abundantly expressed in solid and hematological malignancies, but which is undetectable in most normal adult differentiated tissues (Altieri, 2001). Despite the redundancy of cell death pathways, survivin appears to be required for cancer cell viability, and interference with survivin expression/function has been associated with catastrophic defects of mitotic transition and induction of mitochondrial-induced cell death (Altieri, 2001). Survivin may also provide a reliable indicator of disease progression, and retrospective analysis of various solid tumors has linked survivin expression to decreased overall survival, negative predictive indicators of aggressive disease, resistance to therapy, and accelerated rates of recurrence (Altieri, 2001). In this study, we sought to investigate the potential expression and impact of survivin in oral precancerous lesions, and its potential predictive value for the identification of lesions at high risk of progression to invasive carcinoma.
Selection of Cases To study longitudinal alterations of survivin expression in the process of oral carcinogenesis, we retrieved 16 cases of oral epithelial dysplasia with subsequent invasive OSCCs from the surgical pathology archives (from 1987–1999) of Hiroshima University Dental Hospital. The histological grading of epithelial dysplasia was classified according to the Histological Typing of Cancer and Precancer of the Oral Mucosa (Pindborg et al., 1997). The lesions were independently scored by two pathologists. If there was disagreement, the H&E slides were evaluated again by the two pathologists until a consensus was reached. Five specimens were obtained from healthy oral mucosa, and 30 from cases of oral epithelial dysplasia that did not evolve into squamous cell carcinoma (after follow-up of 13-292 months). All cases selected for this investigation consisted of single oral epithelial dysplasias without multifocal lesions at the time of diagnosis. All cases of dysplasia included in this study were subjected only to biopsy and standard follow-up. Specimens of human oral epithelia were obtained from patients who had undergone routine oral surgical procedures. We reviewed clinical data to record sex and age of patients. The use of archived human tissues conformed to an informed-consent protocol that had been reviewed and approved by the institutional review board of the Hiroshima University Dental Hospital and accepted by all Departments involved in the study.
Immunohistochemistry An anti-human p27Kip1 mouse monoclonal antibody (diluted 1:100, K25020, Transduction Laboratories, Lexington, KY, USA) and an anti-human p53 mouse monoclonal antibody (diluted 1:100, DO-7, DAKO, Glostrup, Denmark) were used. Moreover, to determine the proliferative cell compartment on consecutive tissue sections and correlate them with survivin expression, we examined the Ki-67 expression by immunohistochemistry using an anti-human Ki-67 mouse monoclonal antibody (diluted 1:50, MIB-1, Immunotech, Marseille, France). The sections were treated with the primary antibodies at 4°C overnight. Negative control slides in the absence of primary antibody were included for each staining. The results of the immunohistochemical staining were evaluated separately by two observers, blind to the histological diagnosis of the single cases and especially to the follow-up data for the respective patients. To evaluate survivin expression in the various samples examined, we applied a scoring method (Lu et al., 1998). A mean percentage of positive tumor cells was determined by the examination of 300 cells in at least five areas at 400x magnification. Cells were assigned to one of the five following categories: (a) 0, < 5%; (b) 1, 5-25%; (c) 2, 26-50%; (d) 3, 51-75%; or (e) 4, > 75%. Cases with weighted scores of less than 1 were weighted as negative. To evaluate p53 expression, we applied a scoring method and assigned a mean percentage of positive tumor cells to one of the two following categories: (a) 0, < 15%; or (b) 1, > 15%. To evaluate Ki-67 expression, we used the following scoring categories: (a) 1, < 5%; (b) 2, > 5-10%; or (c) 3, > 10%. To evaluate p27 expression, we used the following scoring categories: (a) 0, < 5%; (b) 1, > 5-10%; or (c) 2, > 10%.
Statistical Analysis
Survivin Expression in Normal Oral Mucosa In all the normal oral epithelium samples examined, in agreement with previous observations (Lo Muzio et al., 2001), only sporadic cells in the basal and parabasal layers occasionally exhibited weak staining for survivin (Fig. panel A; Table 1
Survivin Expression in Oral Precancerous Lesions Survivin expression was detectable in 10/30 cases (33%) of precancerous lesions (Fig. panels B, C) that did not progress into malignancy, and in 15/16 cases (94%) of precancerous lesions that evolved into squamous cell carcinoma, with weighted overall survivin scores ranging from 1 to 4 (Table 1
Survivin Expression in OSCC Sixteen of 16 cases (100%) of OSCC were defined as survivin-positive, with weighted survivin scores ranging from 1 to 4 (Table 1
Statistical Analysis
In this study, we have shown that survivin is up-regulated early during malignant transformation of the oral cavity, and that its up-regulation is overwhelmingly associated with precancerous lesions that evolved into full-blown invasive squamous cell carcinomas. Consistent with current efforts of genetic fingerprinting of human tumors, novel molecular markers that may predict disease progression, early recurrences, or resistance to therapy are being intensely investigated. Aberrations in apoptotic programs are a hallmark of perhaps all cancers that potentially affect various stages of malignant transformation (Evan and Vousden, 2001). In this context, compelling experimental evidence demonstrated that over-expression of survivin in most human cancers signals more aggressive and disseminated disease, prone to recurrence (Altieri, 2001). Here, we have taken oral carcinogenesis as a model of multistep malignant transformation and demonstrated, at least in the present series, that survivin up-regulation may also provide a novel molecular predictor of disease progression. A potential role of apoptosis regulators of the bcl-2 family in this pathway has remained controversial. While lack of bcl-2 expression in oral premalignant lesions has been reported (McAlinden et al., 2000), deregulated bcl-2 expression in severely dysplastic oral epithelial lesions was linked to progression of OSCC in another study (Singh et al., 1998). However, in contrast to the pattern observed here with survivin, which was largely restricted to the transformed cell population and unambiguously associated with oral premalignant lesions that progressed to OSCC, bcl-2 expression was observed in non-dysplastic basal cells contiguous to neoplastic lesions, and progressively reduced in differentiating carcinomas (Singh et al., 1998). Consistent with the data presented here, survivin expression has been shown in other preneoplastic lesions, including polyps of the colon, and nearly all cases of Bowens disease (SCC in situ) and hypertrophic actinic keratosis (Chiodino et al., 1999; Grossman et al., 1999). The molecular basis for the selective expression of survivin in malignant and premalignant lesions, as compared with normal tissues, has not been completely elucidated (Altieri, 2001). Although the survivin gene is expressed in mitosis in a cell-cycle-dependent manner (Li et al., 1998), expression of survivin in cancer is seen in virtually all tumor cells, not restricted to the mitotic fraction, suggesting that malignant transformation is associated with global deregulation of survivin gene expression (Altieri, 2001). This is consistent with recent data identifying survivin as one of the genes repressed by wild-type p53 (Hoffman et al., 2002; Mirza et al., 2002; Zhou et al., 2002), suggesting that loss of p53, which is one of the most common genetic abnormalities in cancer), may result in de-repression of survivin gene transcription.
Our cases showed only low expression of p53 in cases of precancerous lesions that did not evolve into OSCC, and in particular in 1/8 cases with mild (12.5%), 1/9 cases with moderate (11,11%), and 3/6 cases with severe dysplasia (50%) (Table 1 In our cases, p27 expression was not correlated with differentiation and/or biological behavior of the lesions. Instead, p53 showed a progressively increased expression from mild to severe oral dysplasias and in infiltrating OSCC with worse clinical behavior. Similarly, Ki67 expression levels were increased in severe dysplasias and OSCC. In summary, deregulation of cell-cycle regulatory proteins cooperatively or independently contributes to SCC development and progression. These results further support a potential utility of these biomarkers in assessing the progression of individual head and neck squamous tumors. Analysis of the data presented here suggests that this process of survivin re-expression is an early event during stepwise malignant transformation, which may confer selective growth advantage and resistance to environmental or checkpoint-initiated pro-apoptotic stimuli. In our opinion, the overexpression of survivin represents, in all cases, a hallmark of malignant conversion. Naturally, higher scores of survivin expression are likely to be more indicative of this risk. Considering that, in hyperplastic epithelia without associated dysplasia, survivin positivity may reach 5-10% of cells (never exceeding this value), we consider this a possible cut-off value for positivity able to predict malignant conversion. OSCC is a frequent tumor in humans and carries elevated rates of recurrence (Al-Rajhi et al., 2000) that may involve up to 80% of patients within 2 yrs (Mattijssen et al., 1993). In addition, the five-year survival rates of OSCC are not encouraging (Friedlander et al., 1998). This stresses the need for new molecular markers of disease progression that could reliably identify patients at high risk of developing invasive disease. Consistent with its predictive/prognostic value in other tumors (Altieri, 2001), histological determination of survivin expression in oral premalignant lesions may provide a quick and potentially useful indicator for identifying patients requiring more aggressive therapeutic intervention.
This investigation was supported in part by NIH grants CA78810 and CA91917, in part by Italian grant MIUR 2002068474/2002, and by the University of Ancona (2001). Received for publication January 8, 2003. Revision received August 17, 2003. Accepted for publication August 25, 2003.
Journal of Dental Research, Vol. 82, No. 11,
923-928 (2003) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||



