A number of genetic aberrations have been reported in end-stage squamous cell carcinoma of the head and neck, including p16INK4a and p14ARF (INK4a/ARF) inactivation rates of 70–85%. Still, the cell cycle-regulatory genes p16INK4a and p14ARF remain poorly understood in oral cavity premalignant lesions. This study evaluated INK4a/ARF locus alterations in 26 patients (28 samples) deemed to be at increased risk for malignant transformation to squamous cell carcinoma due to the diagnosis of severe oral epithelial dysplasia. Microscopically confirmed dysplastic oral epithelium and matching normal tissue were laser capture-microdissected from paraffin sections, DNA was isolated, and molecular techniques were used to evaluate p16INK4a and p14ARF gene deletion, mutation, loss of heterozygosity (LOH), and hypermethylation events. Deletion of exon 1β, 1α, or 2 was detected in 3.8%, 11.5%, and 7.7% of patients, respectively. INK4a and ARF mutations were detected in 15.4% and 11.5% of patients with severe dysplasia of the oral epithelium. All identified mutations occurred in the INK4a/ARF conserved exon 2. Allelic imbalance was assessed using three markers previously reported to show high LOH rates in head and neck tumors. LOH was found in 42.1%, 35.0%, and 82.4% of patients for the markers IFNα, D9S1748, and D9S171, respectively. Hypermethylation of p16INK4a and p14ARF was detected in 57.7% and 3.8% of patients, respectively, using nested, two-stage methylation-specific PCR. The highest rates of p16INK4a hypermethylation occurred in lesions of the tongue and floor of the mouth. In addition, p16INK4a hypermethylation was significantly linked to LOH in two or more markers. These data support that INK4a/ARF locus alterations are frequent events preceding the development of oral cancer and that p16INK4a inactivation occurs to a greater extent in oral dysplasia than does p14ARF inactivation.

Molecular alterations in a number of cell cycle-regulatory genes have been identified in end-stage SCC3 of the head and neck, including INK4a/ARF inactivation rates of 70–85% (1, 2, 3, 4). The INK4a/ARF locus is located on chromosome 9p21 and has the unique distinction of encoding two cell cycle-regulatory genes, p16INK4a and p14ARF (reviewed in Refs. 5, 6, 7, 8). Briefly, alternative splicing of the first exon and common downstream exons permits one gene to encode two different products, which function via two distinct pathways to inhibit cell cycle progression. The tumor-suppressive activity of p16INK4a is ascribed to its ability to bind both cdk4 and cdk6. This in turn inhibits the catalytic activity of the cdk4/6-cyclin D complex, blocks retinoblastoma phosphorylation, and ultimately prevents cell cycle progression (5). In contrast, p14ARF interacts with the oncogenic protein MDM2, inducing stabilization of p53 and enhancing p53-related functions (9). Codeletion of p16INK4a and p14ARF has been described in a number of tumor types, including those of the head and neck (6). Thus, a single alteration in the INK4a/ARF locus can potentially disrupt the p16-Rb and p14-p53 tumor suppressor pathways and facilitate cancer development. Moreover, a recent study by Bates et al. (10) reported the existence of cross-talk between the pathways, illustrating the complicated nature of these two gene products.

Numerous studies recognize the prominent tumor suppressor function of p16INK4a; however, INK4a/ARF locus alterations in premalignant oral disease remain incompletely investigated and poorly understood. In this study, we sought to comprehensively determine the mode and incidence of p16INK4a and p14ARF alterations in patients diagnosed with severe oral epithelial dysplasia and to assess relationships between INK4a/ARF alterations and various patient characteristics. The risk of malignant transformation among patients with severe oral epithelial dysplasia has been reported to be 36% over a period 8 years (11). Genetic alterations in the 9p21 chromosomal region have been linked to malignant progression, and there is limited evidence suggesting a role for p16INK4a in cancer recurrence (4, 12). The main modes of p16INK4a inactivation in SCC of the head and neck are known to include homozygous deletions, mutations, and gene hypermethylation events (5). A limited number of studies have assessed p16INK4a inactivation in premalignant lesions of the head and neck and support a role for it in the development of precancerous lesions (13, 14, 15). However, in the populations previously assessed for p16INK4a inactivation, the pathologic diagnoses often varied, and not all modes of inactivation were evaluated within a single population. Thus, the current study was undertaken to characterize inactivation of p16INK4a and the lesser understood p14ARF in patients diagnosed with severely dysplastic oral lesions. Dysplastic areas were laser capture-microdissected in an attempt to obtain a pure population of cells for assessment of deletion, mutation, LOH, and hypermethylation events in a high-risk population with morphologically similar lesions. The fact that the lesions are all histologically severe dysplasia controls for pathological grade as a potential confounder and may foster improved understanding of a molecular profile predictive of risk for progression or cancer recurrence in years to come.

Patient Specimens.

Twenty-eight biopsies from 26 consecutive patients with histologically confirmed severe oral epithelial dysplasia were obtained from the archives of the College of Dentistry, Department of Oral Surgery and Pathology at The Ohio State University from 1997 to 2000. One patient provided three sequential biopsies over the course of 3 years. Each case was reviewed by two board-certified oral pathologists (one of whom was S. R. M.). All biopsies were fixed in neutral buffered formalin, routinely processed, and embedded in paraffin. Severe epithelial dysplasia was defined as full thickness epithelial changes without invasion of the basement membrane.

Microdissection and DNA Isolation.

Up to six (8-μm) sections of each case were cut, mounted on noncharged glass slides, dewaxed in xylene, rehydrated in graded alcohols, and briefly stained with H&E (15 s each). Severely dysplastic oral epithelium was identified via bright-field microscopy and microdissected using the PixCell-II Laser Capture Microdissection System (Arcturus Engineering, Mountain View, CA). On average, 2200 laser hits were fired using a 15-μm spot size to obtain sufficient material. Histologically normal connective tissue was microdissected from each case to serve as a control for molecular analysis. Captured cells were digested overnight at 55°C in proteinase K lysis buffer [0.5 mg/ml proteinase K, 50 mm Tris, 1 mm EDTA, and 0.5% Tween 20 (pH 8.5)]. Samples were heated for 10 min at 95°C to inactivate proteinase K and centrifuged, and the supernatant was used for PCR.

PCR Amplification and Analysis.

Exons 1α, 1β, and 2 of the INK4a/ARF locus were amplified as described previously (16). Briefly, intron-based primers were used for the amplification of exons 1α and 1β. Exon 2 was amplified in two fragments to meet the size limitation associated with SSCP analysis (300 bp). In addition, one primer in each set for exon 2 amplification was intron-based to exclude potential coamplification of a similar gene family member or pseudogenes. Exon 3 was not analyzed because it contains only a small portion of the coding sequence for p16INK4a and none of the coding sequence for p14ARF. The 20-μl PCR mixture contained 1× PCR buffer, 1.5 mm MgCl2, 0.5 mm deoxynucleotide triphosphate, 0.5 μm of each primer, 1× Invitrogen enhancer, 1 unit of Taq polymerase, and 3 μl of genomic DNA. Amplification was performed in a MJ Research Peltier Thermal Cycler (Waterton, ME) with PCR conditions of 95°C for 2 min followed by 40 cycles of 95°C for 50 s, 60°C for 20 s, 72°C for 50 s, and a final elongation step of 72°C for 7 min.

Mutation Analysis of p16INK4a and p14ARF.

CDKN2A alterations were detected by a nonradioactive SSCP technique and DNA sequencing as described previously (16). Positive SSCP results were confirmed by analysis of the corresponding replicate PCR sample, isolation of the mutant band, reamplification of the shifted band, and reexamination by SSCP and DNA sequencing. Experimental times and primer information were as reported previously by our laboratory (16).

Detection of Homozygous Deletions.

The p16INK4a and p14ARF genes were processed as described under PCR amplification and analysis, except that multiplex PCR was performed using HPRT as an internal control gene. HPRT was coamplified with the specific exons of interest and electrophoresed through a 20% polyacrylamide Tris-borate EDTA minigel (Invitrogen, San Diego, CA) for deletion assessment.

MSP.

Aberrant methylation of the INK4a/ARF locus was determined by a nested, two-stage MSP method developed by Palmisano et al. (17). Procedural modifications included direct utilization of 1 μl of stage 1 PCR products before stage 2 PCR and changes in PCR conditions permitting optimal amplification with newly designed primers. Briefly, DNAs were bisulfate-modified using the CpGenome DNA modification kit (Intergen, Purchase, NY), and PCR was conducted to amplify a 208- and 259-bp fragment of the p16INK4a and p14ARF genes, respectively. Stage 2 MSP using unmethylated and methylation-specific primers required 1 μl of stage 1 product for the second PCR. The primer sequences used to amplify p16INK4a were 5′-GGAGAGGGGGAGAGTAGGT-3′ (outside forward), 5′-CTACAAACCCTCTACCCACCT-3′ (outside reverse), 5′-TGGGGAGTAGTATGGAGTTGGTGGT-3′ (unmethylated forward), 5′-CAACCCCAAACCACAACCATAA-3′ (unmethylated reverse), 5′-CGGGGAGTAGTATGGAGTCGGCGGC-3′ (methylated forward), and 5′-GACCCCGAACCGCGACCGTAA-3′ (methylated reverse). Four previously described sequences were used to evaluate p14 methylation status (9), as were two newly designed methylation-specific forward primers. The outside p14ARF primers were 5′-TGGGTTTTAGTTTGTAGTTAA-3′ (forward) and 5′-CTCCTCAATAACATCAACAC-3′ (reverse). Methylation-specific p14ARF primers were 5′-TTTTTGGTGTTAAAGGGTGGTGTAGT-3′ (unmethylated forward), 5′-ACCACACACACACCAAATCCA-3′ (unmethylated reverse), 5′-GTGTTAAAGGGCGGCGTAGC-3′ (methylated forward), and 5′-CGCGACGAACCGCACGCGCG-3′ (methylated reverse). Amplification conditions for the first PCR included 95°C for 2 min, denaturing at 95°C for 50 s, annealing at 60°C (p16) and 54°C (p14) for 50 s, extension at 72°C for 50 s for 40 cycles, and a final 10-min extension step. Conditions for the second PCR using methylation-specific primers were the same, except that the annealing temperatures were 64°C and 62°C, respectively, for p16INK4a and p14ARF, and the number of cycles was lowered to 25. Placental DNA and methylation-positive cell lines (HS4, p16; HeLa, p14) served as negative and positive controls.

Allelic Imbalance.

Genomic DNA from premalignant and patient-matched normal control tissues was amplified using fluorescent PCR methods and the microsatellite markers D9S1748, D9S171, and the IFNα cluster (Research Genetics, Huntsville, AL). The 5′ forward amplification primer of each of the PCR primer pairs was synthesized with fluorescent labels. Amplification was conducted as described above, except that the final PCR volume was 25 μl. Individual amplification reactions were pooled per patient for both normal and premalignant tissues. Next, 3 μl of the respective pooled panel were added to 3 μl of loading buffer and the TAMRA-500 molecular weight marker. Reactions were denatured and placed on ice, and 1.5 μl of each panel were loaded onto a 4.25% denaturing polyacrylamide Tris-borate EDTA gel. The data were collected on an ABI 377 automated sequencer, and GeneScan and Genotyper software was used to quantify normal versus premalignant amplicon patterns for each fluorescent marker. LOH markers were chosen based on their location on 9p and the fact that they previously showed high levels of LOH in head and neck tumors. Allele loss was calculated by the method of Canzian et al. (18), which determines allelic imbalance by calculating the ratio of the two allele pairs in normal tissue to that of the premalignant tissue, with a shift of 40% or more considered LOH positive.

Statistical Analysis.

Associations between patient characteristics (nominal and ordinal scale) and gene inactivation events were analyzed using χ2 contingency tables or Fisher’s exact test. Relationships between patient age and gene inactivation were assessed using the unpaired t test. All statistical tests were two-sided, with P < 0.05 considered statistically significant.

Patient Characteristics.

A total of 28 biopsies showing severe epithelial dysplasia, from 26 patients, were evaluated for alterations in the INK4a/ARF locus. Age at biopsy ranged from 26 to 87 years, with a mean patient age of 65 years. The study population was comprised of 15 male and 11 female patients. The most common sites of dysplasia in this population were the tongue (32%), floor of the mouth (25%), and buccal mucosa (25%). In this study, no significant differences emerged with regard to gender and site of dysplasia. However, among female participants, dysplasia most commonly occurred on the tongue (54%), followed by the floor of the mouth (18.2%). Similarly, recent data from the National Cancer Institute Surveillance, Epidemiology, and End Results database reported the most common site for head and neck cancer among females to be the tongue, followed by the floor of the mouth (19, 20). In contrast, male patients most commonly presented with dysplasia on the floor of the mouth (33%), followed by the buccal mucosa (26.7%) and tongue (20%). According to Surveillance, Epidemiology, and End Results data, the greatest number of head and neck cancers among males occurred on the tongue, followed by the lip and floor of the mouth (20).

Twenty-five percent of patients exhibited dysplasia with oral leukoplakia, and one patient exhibited dysplastic erythroplakia. Leukoplakia exhibiting dysplastic changes has previously been reported to impart increased risk for malignant transformation to SCC (12, 21). A study by Silverman et al. (11) found 36% of dysplastic leukoplakias transformed to carcinoma over a period of 8 years. In our study, the clinical diagnosis of leukoplakia was not significantly linked to gender, lesional site, or inactivation of p16INK4a or p14ARF; however, we urge caution because the number of patients with leukoplakia is small (n = 7), reducing the statistical power to detect meaningful differences. Pathology reports indicated that CIS or superficially invasive SCC could not be ruled out in >50% of patients under study. This ruling did not correlate with any patient characteristics or p16INK4a or p14ARF gene inactivation events. Thus, the fact that CIS or superficially invasive SCC could not be ruled out appeared to be more a function of biopsy sampling and biopsy size than linkage to individual tissue morphology.

Allelic Imbalance on Chromosome Arm 9p.

LOH on the short arm of chromosome 9p is the most common defect reported in SCC of the head and neck, ranging from 45% to 95% (22, 23, 24). A number of studies have also found high frequencies of LOH in oral premalignant lesions and found a positive correlation between LOH and histological progression, cancer development, and cancer recurrence (11, 14, 24, 25). In the current study, LOH was the most frequent molecular alteration detected. Allelic loss at chromosome 9p21 was assessed using three markers reported to be highly polymorphic in head and neck tumors or oral premalignant lesions (14, 15, 26, 27). The markers were also chosen based on their relative location on chromosome 9p. Details of LOH are presented in Fig. 1. Allelic loss at 9p21 was detected in at least one microsatellite marker in 76.9% of patients. As summarized in Fig. 1 C, LOH occurred in 42.1%, 35.0%, and 82.4% of cases that were informative for the markers IFNα, D9S1748, and D9S171, respectively. Approximately, 58% of cases presented with LOH in a single marker, 15.4% of cases presented with LOH in two markers, and 7.7% of cases presented with LOH in all three markers.

In our analysis of LOH and patient characteristics, the mean age of patients with LOH in one or fewer markers was statistically higher (69.2 years) than the age of patients (53.1 years) with LOH in two or three markers (P = 0.020). Furthermore, LOH in two or more markers was significantly associated with p16INK4a hypermethylation (P = 0.007). All cases with LOH in two or three markers were hypermethylated for p16INK4a. An earlier study assessing prevalence of p16INK4a alterations in head and neck tumors reported that 65% of the cases with biallelic gene inactivation showed LOH and hypermethylation (25). In this study population, we found no microsatellite instability in any of the three markers evaluated. This contrasts with a recent study by Shahnavaz et al. (15) that found higher rates of microsatellite instability than LOH in oral precancerous lesions and subsequent SCC that developed in 11 patients. Our study findings may differ with respect to microsatellite instability for a number of reasons, including the fact that different microsatellite markers were assessed. The patient populations were also from different geographic areas and may have had other disparate characteristics impacting chromosomal defects. In addition, there were differences in the histological grade of the premalignant lesions in the two patient populations. All 26 patients in our study were diagnosed with severe dysplasia, whereas in the study by Shanavaz et al. (15), only 2 patients had severe dysplasia, and the others presented with mild or moderate dysplasia. Other groups have reported similar LOH rates in oral dysplasias or CIS in this chromosomal region (28).

In this study cohort, LOH in two or more markers was significantly linked to lesions on the floor of the mouth (P = 0.023). Although tobacco exposure information was not available for this patient cohort, recent studies have emerged linking LOH to tobacco exposure among patients with upper aerodigestive tract cancers (29, 30, 31, 32). Furthermore, studies support site specificity of head and neck SCC in both smokers and nonsmokers (27, 29, 30). Smoking-associated head and neck SCCs develop on the floor of the mouth; whereas nonsmokers present with oral cancers of the tongue, followed by the buccal and alveolar mucosa (30). These relationships merit further study in a larger cohort in which tobacco and alcohol exposure information is available.

p16INK4a and p14ARF Methylation Status.

The frequency of aberrant methylation for the genes p16INK4a and p14ARF was determined for all 26 patients and 28 dysplastic biopsy specimens. The results are summarized in Table 1, and a representative gel is pictured in Fig. 2. The detection of both methylated and unmethylated sequences in same lesion may be due to the fact that the population of cells analyzed, although histologically similar, remains genetically diverse and/or may indicate that methylation is not an all or nothing event in oral precancer. A total of 57.7% of patients (15 of 26) had methylation for the p16INK4a gene, whereas only a single patient (3.8%) was positive for p14ARF methylation. Similar rates of p16INK4a hypermethylation have been detected in SCCs of the head and neck (33, 34), and even higher levels of p16INK4a methylation (82%) were recently detected in severely dysplastic oral lesions (24). We attribute high levels of p16INK4a methylation detection to the fact that a sensitive two-stage nested MSP approach was used coupled with laser capture microdissection, permitting assessment of a rather pure population of severely dysplastic cells.

In addition to the association between p16INK4a methylation and increased LOH, methylation was significantly associated with the site of dysplasia (P = 0.001). In terms of the p16INK4a-methylated samples, 53% were biopsies from the tongue, 40% were biopsies from the floor of the mouth, and 6.7% were biopsies from the buccal mucosa; there were no p16INK4a-methylated samples from the palate or lip. Alternatively, 88% (8 of 9) of tongue lesions and 85% (6 of 7) of biopsies from the floor of the mouth were positive for p16INK4a methylation. In contrast, only 15% of buccal mucosa biopsies were p16INK4a methylation positive, and none of the biopsies from the palate or lip were hypermethylated. The floor of the mouth and ventrolateral tongue are considered sites with increased risk for malignant progression (31). A recent report by Koch et al. (30) found that floor of the mouth cancers rarely develop in nonsmokers. In addition, Soria et al. (34) recently evaluated aberrant methylation in bronchial brush samples from former smokers and found significantly higher p16INK4a methylation in former smokers with a history of previous cancer compared with former smokers without a history of cancer. Unfortunately, as mentioned previously, tobacco and alcohol exposure information was not available in this patient population. Thus, it is possible that the linkage between the site of dysplasia and methylation status is due in part to exposures for which we cannot account. Methylation did not appear to be statistically linked to any other clinicopathological features evaluated. Aging has been one feature previously reported to be linked to methylation events; however, in the population under study, we found no significant difference in mean age between p16INK4a methylation-positive and -negative patients. To the contrary, methylation-positive patients were on average 8.8 years younger than methylation-negative patients (61.1 years versus 69.9 years, respectively). To our knowledge, no other study has evaluated p14ARF methylation in oral premalignant lesions. Only one patient was positive for p14ARF methylation, supporting that p16INK4a methylation is more prevalent in patients with high-grade oral dysplasia and p16INK4a hypermethylation may be of greater importance in oral cancer development. Gene hypermethylation reportedly leads to chromosomal instability, which in turn supports further gene-inactivating events via LOH, mutation, or deletion and ultimately contributes to the transcriptional silencing of tumor suppressor genes. We did not detect any significant correlation between p16 or p14 hypermethylation and INK4a/ARF mutation or deletion events; however, greater frequencies of mutation, deletion, and p14 hypermethylation events would likely be required to accurately access such relationships, given the cohort size.

p16INK4a and p14ARF Homozygous Deletion.

A summary of INK4a/ARF inactivation events in patients with high-grade oral cavity dysplasia is presented in Table 1. Homozygous deletion of exon 1α was the most prevalent and was found in 11.5% (3 of 26) of patients. Our findings are similar in terms of deletion prevalence to those of another study reporting selective loss of exon 1α in 12% of patients with mild dysplasia of the oral epithelium (10). Exon 2 deletions were detected in two patients (7.7%), both of whom had a deletion of exon 1α. Concomitant loss of exons 1α and 2 has previously been reported in patients with oral SCC (35). Perhaps there are unique subpopulations that differ in their deletion profiles, or exon 2 deletions may be relatively late events, following other modes of inactivation. Only one patient (patient 11) was found to have an exon 1β deletion, and this same patient had LOH in the centromeric marker D9S171 as well as in the telomeric marker IFNα.

p16INK4a and p14ARF Mutation Analysis.

Fig. 3,A summarizes the INK4a/ARF mutations detected in severely dysplastic lesions of the oral cavity and predicts resultant alterations of p16INK4a and p14ARF transcripts and proteins. SSCP analysis of the INK4a/ARF locus revealed abnormal band shifts in approximately 15% (4 of 26) of patients with high-grade dysplasia. This appears to be the first assessment of INK4a/ARF mutations in histologically severe oral epithelial dysplasias. A representative shift and subsequent sequence analysis are pictured in Fig. 3, B and C. These novel findings indicate that mutational events occur at about the same incidence as deletional events in oral premalignant lesions but that the exon targeted for mutational events may differ. All mutations were single-base changes located in conserved exon 2. As discussed above, exon 1α seemed to be the preferred target for deletion events, followed by exon 2. Earlier work in our laboratory evaluated 100 SCCs of the head and neck for INK4a/ARF alterations and found 27% of them to exhibit sequence alterations in this locus. In the head and neck tumors evaluated, 91% of mutations involved exon 2, and 9% of mutations involved exon 1α, whereas no mutations were found in exon 1β (16). In our current analysis of severe dysplasias of the oral cavity, we observed no mutations in exon 1β or 1α, implying that mutations in exon 2 are more prevalent in precancerous lesions as well as in SCCs.

Sequence alterations were identified in four patients, including changes in codons 58, 74, 82, 89, and 148 of p16INK4a and codons 72, 89, 96, and 103 of p14ARF. In terms of sequence changes affecting p16INK4a, three were missense mutations, one was a silent mutation, and one was a polymorphism in codon 148, which has been previously reported in SCCs of the head and neck (8, 16, 36). In addition to earlier reports of a G→A transition at codon 148, a C→T transition in codon 58 has been identified as a frequent mutation resulting in a p16 truncation (reviewed in Ref. 8). However, we found a G→A transition in codon 58, which results in a p14ARF truncation. One patient (patient 1) exhibited changes in two different codons (codons 82 and 89), whereas all other mutations involved a single codon. We are unaware of other reports of mutations in codon 82. Four of the detected mutations are expected to result in altered p16 and p14 proteins.

To summarize, these data support that p16INK4a is frequently inactivated in dysplastic oral epithelial lesions via LOH, hypermethylation, deletion, and mutation events, in descending order. Thus, oral biopsies that morphologically represent premalignant lesions already possess genetic aberrations commonly found in end-stage oral cancer. The temporality of INK4a/ARF alterations cannot be assessed in this cohort, given that only a single biopsy was available from all but one of the study participants. Interestingly, in the one patient contributing three biopsies over 3 years, divergent p16 aberrations were found, supporting the observation that even within a specified genetic region, the clonal evolution of premalignant cells is not necessarily linear.

Although the sample size of 26 patients or 28 samples appears relatively small, this is the largest cohort of patients with premalignant lesions of the same histological category to be assessed for p16INK4a and p14ARF alterations. Characterization of INK4a/ARF inactivation events in populations at increased risk for malignant transformation may provide insight into which alterations or combinations of inactivation events are most critical for cancer development. To date, only one of the study patients has developed a SCC, but we will continue to follow these patients with the intent of improving our understanding of the role that p16INK4a and p14ARF plays in cancer progression.

Fig. 1.

LOH analysis in severe oral epithelial dysplasias. A, allelogram depicting the three exons of p16INK4a (exons 3, 2, and 1α) and p14ARF (shared exons 3 and 2, with alternatively spliced exon 1β) and the relative positions of the microsatellite markers IFNα, D9S1748, and DS171. B, traces of allelic loss at chromosome 9p21–22 in patients 25, 20, and 9, respectively. The normal tissue results are shown in the top trace, and patient-matched dysplastic tissue is depicted in the bottom trace for each patient and marker. The top traces (normal tissue) show amplification of two alleles, whereas the bottom traces of patient-matched dysplastic tissue show allelic loss (as indicated by the arrow). C, summary of LOH results by patient. Patient age at the time of biopsy is given in years. The mean LOH levels reported reflect the percentage of LOH-positive patients out of the total number of patients informative for each marker. NORM, normal or retention of both alleles; NI, noninformative for the marker. Over 75% percent of patients with severe dysplasia showed LOH in at least one marker, with the highest LOH reported for the centromeric marker D9S171.

Fig. 1.

LOH analysis in severe oral epithelial dysplasias. A, allelogram depicting the three exons of p16INK4a (exons 3, 2, and 1α) and p14ARF (shared exons 3 and 2, with alternatively spliced exon 1β) and the relative positions of the microsatellite markers IFNα, D9S1748, and DS171. B, traces of allelic loss at chromosome 9p21–22 in patients 25, 20, and 9, respectively. The normal tissue results are shown in the top trace, and patient-matched dysplastic tissue is depicted in the bottom trace for each patient and marker. The top traces (normal tissue) show amplification of two alleles, whereas the bottom traces of patient-matched dysplastic tissue show allelic loss (as indicated by the arrow). C, summary of LOH results by patient. Patient age at the time of biopsy is given in years. The mean LOH levels reported reflect the percentage of LOH-positive patients out of the total number of patients informative for each marker. NORM, normal or retention of both alleles; NI, noninformative for the marker. Over 75% percent of patients with severe dysplasia showed LOH in at least one marker, with the highest LOH reported for the centromeric marker D9S171.

Close modal
Fig. 2.

Representative examples of MSP for p16INK4a and p14ARF. The presence of a visible PCR product in Lanes U indicates the presence of unmethylated genes, and the presence of a visible PCR product in Lanes M indicates the presence of methylated genes. The results from four cases are shown for each gene. Placenta (Plac) served as an unmethylated control, and the HS4 and HeLa cell lines served as methylation-positive (Pos) controls for p16INK4a and p14ARF, respectively. H2O served as a negative control for the overall reaction.

Fig. 2.

Representative examples of MSP for p16INK4a and p14ARF. The presence of a visible PCR product in Lanes U indicates the presence of unmethylated genes, and the presence of a visible PCR product in Lanes M indicates the presence of methylated genes. The results from four cases are shown for each gene. Placenta (Plac) served as an unmethylated control, and the HS4 and HeLa cell lines served as methylation-positive (Pos) controls for p16INK4a and p14ARF, respectively. H2O served as a negative control for the overall reaction.

Close modal
Fig. 3.

Mutations of the INK4a/ARF locus in dysplastic oral cavity lesions and their effect on p16INK4a and p14ARF transcripts. A, details of INK4a/ARF locus alterations by patient and site. M, missense; T, truncation; P, polymorphism; S, silent; NA, not applicable. B, representative example of p16INK4a mutation analysis as detected by cold SSCP. Dysplastic epithelium from patient 3 resulted in p16INK4a exon 2 shifts, as compared with normal and placental control tissues. C shows the sequencing trace indicating a G→A transition in codon 148.

Fig. 3.

Mutations of the INK4a/ARF locus in dysplastic oral cavity lesions and their effect on p16INK4a and p14ARF transcripts. A, details of INK4a/ARF locus alterations by patient and site. M, missense; T, truncation; P, polymorphism; S, silent; NA, not applicable. B, representative example of p16INK4a mutation analysis as detected by cold SSCP. Dysplastic epithelium from patient 3 resulted in p16INK4a exon 2 shifts, as compared with normal and placental control tissues. C shows the sequencing trace indicating a G→A transition in codon 148.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by The National Institutes for Dental and Craniofacial Research Grant RO1-DE11943 and National Cancer Institute Grants T32 CA09338 and P30 CA16058.

3

The abbreviations used are: SCC, squamous cell carcinoma; LOH, loss of heterozygosity; CIS, carcinoma in situ; MSP, methylation-specific PCR; INK4a/ARF, p16INK4a and p14ARF; SSCP, single-stranded conformational polymorphism; cdk, cyclin-dependent kinase.

Table 1

Summary of INK4α/ARF locus alterations in patients with severe dysplasia of the oral cavity

Patient no.GenderLesion siteaLOHbMutationcDeletionMethylationd
p16p142p16p14
1e Male Buccal mucosa − − − − − 
1be Male Buccal mucosa − − − − − − − 
1ce Male Buccal mucosa − − − − − − 
Male Floor of mouth − − − − 
Female Palate − − − − − 
Male Palatea − − − − − 
Male Floor of moutha − − − − 
Female Tongue − − − − − − 
Male Buccal mucosa − − − − − − − 
Male Lip − − − − − − − 
Male Palatea − − − − − − − 
10 Male Buccal mucosa − − − − − − 
11 Male Floor of mouth − − − − − 
12 Male Floor of mouth − − − − − − 
13 Female Lip − − − − − − − 
14 Male Tongue − − − − − − 
15 Female Floor of mouth − − − − − − 
16 Female Tongue − − − − − − 
17 Female Tonguea − − − − − − 
18 Female Tongue − − − − 
19 Male Buccal mucosa − − − − − − − 
20 Female Tonguea − − − − − − − 
21 Male Tonguea − − − − − − 
22 Male Tongue − − − − − − 
23 Female Tongue − − − − − − 
24 Female Buccal mucosa − − − − − − − 
25 Female Floor of mouth − − − − − − 
26 Male Floor of moutha − − − − − − − 
           
Percentagef   76.9 15.4 11.5 7.7 11.5 3.8 57.7 3.8 
Patient no.GenderLesion siteaLOHbMutationcDeletionMethylationd
p16p142p16p14
1e Male Buccal mucosa − − − − − 
1be Male Buccal mucosa − − − − − − − 
1ce Male Buccal mucosa − − − − − − 
Male Floor of mouth − − − − 
Female Palate − − − − − 
Male Palatea − − − − − 
Male Floor of moutha − − − − 
Female Tongue − − − − − − 
Male Buccal mucosa − − − − − − − 
Male Lip − − − − − − − 
Male Palatea − − − − − − − 
10 Male Buccal mucosa − − − − − − 
11 Male Floor of mouth − − − − − 
12 Male Floor of mouth − − − − − − 
13 Female Lip − − − − − − − 
14 Male Tongue − − − − − − 
15 Female Floor of mouth − − − − − − 
16 Female Tongue − − − − − − 
17 Female Tonguea − − − − − − 
18 Female Tongue − − − − 
19 Male Buccal mucosa − − − − − − − 
20 Female Tonguea − − − − − − − 
21 Male Tonguea − − − − − − 
22 Male Tongue − − − − − − 
23 Female Tongue − − − − − − 
24 Female Buccal mucosa − − − − − − − 
25 Female Floor of mouth − − − − − − 
26 Male Floor of moutha − − − − − − − 
           
Percentagef   76.9 15.4 11.5 7.7 11.5 3.8 57.7 3.8 
a

Dysplastic lesions with leukoplakia.

b

LOH indicates the number of positive LOH markers detected per patient sample analyzed.

c

All mutations detected were in exon 2.

d

Hypermethylation of p16INK4a correlates with lesions present on the tongue and floor of the mouth and is significantly linked to the presence of LOH in two or more markers (P ≤ 0.01), χ2 test.

e

Tissue was obtained from the same patient over 3 consecutive years.

f

Percentage refers to the percentage of patients whose tissue contained the genetic aberration.

1
Cairns P., Polascik T. J., Eby Y., Tokino K., Califano J., Merlo A., Mao L., Herath J., Jenkins R., Westra W., Rutter J. L., Buckler A., Gabrielson E., Tockman M., Cho K. R., Hedrick L., Bova G. S., Isaacs W., Koch W., Schwab D., Sidransky D. Frequency of homozygous deletion at p16/CDKN2 in primary human tumours.
Nat. Genet.
,
11
:
210
-212,  
1995
.
2
Reed A. L., Califano J., Cairns P., Westra W. H., Jones R. M., Koch W., Ahrendt S., Eby Y., Sewell D., Nawroz H., Bartek J., Sidransky D. High frequency of p16 (CDKN2/MTS-1/INK4A) inactivation in head and neck squamous cell carcinoma.
Cancer Res.
,
56
:
3630
-3633,  
1996
.
3
Riese U., Dahse R., Fiedler W., Theuer C., Koscielny S., Ernst G., Beleites E., Claussen U., von Eggeling F. Tumor suppressor gene p16 (CDKN2A) mutation status and promoter inactivation in head and neck cancer.
Int. J. Mol. Med.
,
4
:
61
-65,  
1999
.
4
Danahey D. G., Tobin E. J., Schuller D. E., Bier-Laning C. M., Weghorst C. M., Lang J. C. p16 mutation frequency and clinical correlation in head and neck cancer.
Acta Otolaryngol.
,
119
:
285
-288,  
1999
.
5
Rocco J. W., Sidransky D. p16 (MTS-1/CDKN2/INK4a) in cancer progression.
Exp. Cell Res.
,
264
:
42
-55,  
2001
.
6
Sharpless N. E., DePinho R. A. The INK4A/ARF locus and its two gene products.
Curr. Opin. Genet. Dev.
,
9
:
22
-30,  
1999
.
7
Chin L., Pomerantz J., DePinho R. A. The INK4a/ARF tumor suppressor: one gene−two products−two pathways.
Trends Biochem. Sci.
,
23
:
291
-296,  
1998
.
8
Ruas M., Peters G. The p16INK4a/CDKN2A tumor suppressor and its relatives.
Biochim. Biophys. Acta
,
1378
:
F115
-F177,  
1998
.
9
Esteller M., Cordon-Cardo C., Corn P. G., Meltzer S. J., Pohar K. S., Watkins D. N., Capella G., Peinado M. A., Matias-Guiu X., Prat J., Baylin S. B., Herman J. G. p14ARF silencing by promoter hypermethylation mediates abnormal intracellular localization of MDM2.
Cancer Res.
,
61
:
2816
-2821,  
2001
.
10
Bates S., Phillips A. C., Clark P. A., Scott F., Peters G., Ludwig R. L., Vousden K. H. p14ARF links the tumour suppressors RB and p53.
Nature (Lond.)
,
395
:
124
-125,  
1998
.
11
Silverman S., Jr., Gorsky M., Lozada F. Oral leukoplakia and malignant transformation. A follow-up study of 257 patients.
Cancer (Phila.)
,
53
:
563
-568,  
1984
.
12
Lydiatt W. M., Davidson B. J., Schantz S. P., Caruana S. C., Chaganti R. S. K. 9p21 deletion correlates with recurrence in head and neck cancer.
Head Neck
,
20
:
113
-118,  
1998
.
13
Papadimitrakopoulou V. A., Izzo J., Lippman S. M., Lee J. S., Fan Y. H., Clayman G., Ro J. Y., Hittelman W. N., Lotan R., Hong W. K., Mao L. Frequent inactivation of p16INK4a in oral premalignant lesions.
Oncogene
,
14
:
1799
-1803,  
1997
.
14
Mao L., Lee J. S., Fan Y. H., Ro J. Y., Batsakis J. G., Lippman S., Hittleman W., Hong W. K. Frequent microsatellite alterations at chromosome 9p21 and 3p14 in oral premalignant lesions and their value in cancer risk assessment.
Nat. Med.
,
2
:
682
-685,  
1996
.
15
Shahnavaz S. A., Bradley G., Regezi J. A., Thakker N., Gao L., Hogg D., Jordan R. C. K. Patterns of CDKN2A gene loss in sequential oral epithelial dysplasias and carcinomas.
Cancer Res.
,
61
:
2371
-2375,  
2001
.
16
Poi M. J., Yen T., Li J., Song H., Lang J. C., Schuller D. E., Pearl D. K., Casto B. C., Tsai M-D., Weghorst C. M. Somatic INK4a-ARF locus mutations: a significant mechanism of gene inactivation in squamous cell carcinomas of the head and neck.
Mol. Carcinog.
,
30
:
26
-36,  
2001
.
17
Palmisano W. A., Divine K. K., Saccomanno G., Gilliland F. D., Baylin S. B., Herman J. G., Belinsky S. A. Predicting lung cancer by detecting aberrant promoter methylation in sputum.
Cancer Res.
,
60
:
5954
-5958,  
2000
.
18
Canzian F., Salovaara R., Hemminki A., Kristo P., Chadwick R. B., Aaltonen L. A., da la Chapelle A. Semiautomated assessment of loss of heterozygosity and replication error in tumors.
Cancer Res.
,
56
:
3331
-3337,  
1996
.
19
Forastiere A., Koch W., Trotti A., Sidransky D. Head and neck cancer.
N. Engl. J. Med.
,
345
:
1890
-1900,  
2001
.
20
Canto, M. T., and Devesa, S. S. Oral cavity and pharynx cancer incidence rates in the United States 1975–1998. Oral Oncol., in press, 2002.
21
Lumerman H., Freedman P., Kerpel S. Oral epithelial dysplasia and the development of invasive squamous carcinoma.
Oral Surg. Oral Med. Oral Pathol. Oral Radiol. Endod.
,
79
:
321
-329,  
1995
.
22
Tabor M. P., Brakenhoff R. H., van Houten V. M. M., Kummer J. A., Snel M. H. J., Snijders P. J. F., Snow G. B., Leemans C. R., Braakhuis B. J. M. Persistence of genetically altered fields in head and neck cancer patients: biological and clinical implications.
Clin. Cancer Res.
,
7
:
1523
-1532,  
2001
.
23
Nawroz H., van der Riet P., Hruban R. H., Koch W., Ruppert J. M., Sidransky D. Allelotype of head and neck squamous cell carcinoma.
Cancer Res.
,
54
:
1152
-1155,  
1994
.
24
Papadimitrakopoulou V. A., Izzo J., Mao L., Keck J., Hamilton D., Shin D. M., El-Naggar A., Hollander P., Liu D., Hittelman W. N., Hong W. K. Cyclin D1 and p16 alterations in advanced premalignant lesions of the upper aerodigestive tract: role in response to chemoprevention and cancer development.
Clin. Cancer Res.
,
7
:
3127
-3134,  
2001
.
25
Miracca E. C., Kowalski L. P., Nagai M. A. High prevalence of p16 genetic alterations in head and neck tumours.
Br. J. Cancer
,
81
:
677
-683,  
1999
.
26
Califano J., van der Riet P., Westra W., Nawoz H., Clayman G., Piantadosi S., Corio R., Lee D., Greenberg B., Koch W., Sidransky D. Genetic progression model for head and neck cancer: implications for field cancerization.
Cancer Res.
,
56
:
2488
-2492,  
1996
.
27
Rosin M. P., Cheng X., Poh C., Lam W. L., Huang Y., Lovas J., Berean K., Epstein J. B., Priddy R., Le N. D., Zhang L. Use of allelic loss to predict malignant risk for low-grade oral epithelial dysplasia.
Clin. Cancer Res.
,
6
:
357
-362,  
2000
.
28
Guo Z., Yamaguchi K., Sanchez-Cespedes M., Westra W. H., Koch W. M., Sidransky D. Allelic losses in OraTest-directed biopsies of patients with prior upper aerodigestive tract malignancy.
Clin. Cancer Res.
,
7
:
1963
-1968,  
2001
.
29
Wynder E. L., Bross I. J., Feldman R. M. A study of the etiological factors in cancer of the mouth.
Cancer (Phila.)
,
10
:
1300
-1323,  
1957
.
30
Koch W. M., Lango M., Sewell D., Zahurak M, Sidransky D. Head and neck cancer in nonsmokers: a distinct clinical and molecular entity.
Laryngoscope
,
109
:
1544
-1551,  
1999
.
31
Boffetta P., Mashberg A., Winkelmann R., Garfinkel L. Carcinogenic effect of tobacco smoking and alcohol drinking on anatomic sites of the oral cavity and oropharynx.
Int. J. Cancer
,
52
:
530
-533,  
1992
.
32
Sanchez-Cespedes M., Decker P. A., Doffek K. M., Esteller M., Westra W. H., Alawi E. A., Herman J. G., Demeure M. J., Sidransky D., Ahrendt S. A. Increased loss of chromosome 9p21 but not p16 inactivation in primary non-small cell lung cancer from smokers.
Cancer Res.
,
61
:
2092
-2096,  
2001
.
33
El-Naggar A. K., Lai S., Clayman G., Lee J-K. J., Luna M. A., Goepfert H., Batsakis J. G. Methylation, a major mechanism of p16/CDKN2 gene inactivation in head and neck squamous carcinoma.
Am. J. Pathol.
,
151
:
1767
-1774,  
1997
.
34
Soria J-C., Rodriguez M., Liu D. D., Lee J. J., Hong W. K., Mao L. Aberrant promoter methylation of multiple genes in bronchial brush samples from former cigarette smokers.
Cancer Res.
,
62
:
351
-355,  
2002
.
35
Tsai C-H., Yang C-C., Chou L. S-S., Chou M-Y. The correlation between alteration of p16 gene and clinical status in oral squamous cell carcinoma.
J. Oral Pathol. Med.
,
30
:
527
-531,  
2001
.
36
Wu C. L., Roz L., Mckown S., Sloan P., Read A. P., Holland S., Porter S., Scully C., Paterson I., Tavassoli M., Thakker N. DNA studies underestimate the major role of CDKN2A inactivation in oral and oropharyngeal squamous cell carcinomas.
Genes Chromosomes Cancer
,
25
:
16
-25,  
1999
.