Despite advances in the detection and treatment of lung cancer, the prognosis for patients with lung cancer is poor, partly as a result of recurrences. We retrospectively analyzed the relationship between recurrence and survival in patients with non–small cell lung cancers (NSCLC), and the promoter methylation of p16, GSTP1, FHIT, H-cadherin, and RARβ2 genes to identify a prognostic molecular marker associated with the recurrence of NSCLC. Methylation status from 335 paraffin blocks was determined by methylation-specific PCR. Of the 335 NSCLC samples, promoter methylation was detected in 35% for p16, 39% for RARβ2, 42% for H-cadherin, 7% for GSTP1, and 21% for FHIT. Recurrence was observed in 39% (132 of 335) of the patients. Recurrence was significantly associated with histology (P = 0.001) and pathologic stage (P = 0.009). Hypermethylation of any single gene was not associated with recurrence in patients. However, cohypermethylation of p16 and FHIT genes in stage I NSCLCs was associated with an increased risk of recurrence [odds ratio, 6.43; 95% confidence interval (CI), 1.04-20.19; P = 0.02] and poor recurrence-free survival after surgery (hazard ratio, 2.03; 95% CI, 1.09-6.23; P = 0.02). In addition, their survival after recurrence was also 4.62 times poorer (95% CI, 1.27-16.48; P = 0.005) than for those without cohypermethylation of both genes. In conclusion, the present study suggests that cohypermethylation of p16 and FHIT genes in patients with stage I NSCLC may be a valuable biomarker for predicting the recurrence-associated prognosis of the disease. (Cancer Res 2006; 66(8): 4049-54)

Lung cancer is the most frequent cause of cancer-related death in the world, causing more than one million deaths worldwide each year. Despite advances in the detection and treatment of lung cancer, the overall 5-year survival rate remains <15%. The poor prognosis of patients with lung cancer is partly a result of recurrence, which appears in ∼20% to 50% of patients after curative surgical resection with appropriate lymph node dissection (13). Thus, the development of biomarkers for predicting recurrence and the implementation of efficient preventive methods to prevent recurrence in these patients is clearly imperative.

The aberrant methylation of normally unmethylated CpG islands is an epigenetic change that induces the transcriptional silencing of tumor suppressor genes. The hypermethylation of CpG islands at the promoter region of >40 tumor suppressor genes has been reported in lung cancer. Among those genes, p16, retinoic acid receptor β (RARβ), H-cadherin (CDH13), and fragile histidine triad (FHIT) genes are important in the pathogenesis of lung cancer and are frequently inactivated by aberrant methylation of CpG islands at their promoter regions (4). Accordingly, epigenetic modification may be a candidate marker for predicting the prognosis of non–small cell lung cancers (NSCLC). In addition, patients with epigenetic modifications could benefit from treatment with demethylating agents after surgery.

Although clinical and histopathologic factors that might assist in the prediction of tumor recurrence after curative resection of NSCLCs have been studied by many groups, epigenetic alterations associated with recurrence after curative resection has been reported by a few groups. Kim et al. (3) studied 61 patients with NSCLC and reported that P2 hypermethylation of RARB2 and unmethylation of DAPK could be used as prognostic markers in predicting the early recurrence of NSCLC. To identify a useful prognostic biomarker for disease recurrence after a curative resection of NSCLC, we investigated the relationship between recurrence and the aberrant methylation of p16, RARβ2, H-cadherin (CDH13), GSTP1, and FHIT genes in a large sample.

Study population. A total of 335 NSCLC patients, who underwent definitive surgical resection at the Samsung Medical Center in Seoul, Korea, between May 1994 and December 2001, participated in this study. Written informed consent for the use of paraffin-embedded tissues, as approved by the Institutional Review Board at Samsung Medical Center, was obtained from each patient before surgery. All available paraffin blocks were reviewed by a thoracic pathologist (J. Han). Information on sociodemographic characteristics was obtained using an interviewer-administered questionnaire. Postoperative follow-up was scheduled at 1, 2, and every 3 months during the first 2 years after surgery, and every 6 months thereafter, or more frequently if needed. Chest X-ray, chest computed tomography scan, carcinoembryonic antigen, and other serum chemistries were scheduled at every follow-up visit. Whenever patients did not keep the postoperative follow-up schedule, specialized nurses called patients and checked their health status. The median duration of follow-up after a curative resection was 3.7 years.

Recurrence was evaluated from information obtained as of June 30, 2005 from our hospital records and those from other hospitals. The data of patients who died of any causes, whose cancers did not recur before the end of the study, or who were lost to follow-up during the study period, were treated as censored data when calculating recurrence. Second primary tumors were carefully differentiated from recurrence in this study. Recurrence patterns were classified into two categories, locoregional and distant. We defined a locoregional recurrence as evidence of a tumor in the supraclavicular nodes, mediastinal nodes, pleural effusion or seeding, bronchial stump, and other lobes of the ipsilateral lung, whereas distant recurrence was defined as metastasis to the contralateral lung, brain, bone, liver, adrenal, and other organs. Simultaneous locoregional and distant recurrence was considered with the distant recurrence group. Of the 132 patients who had a recurrence, 79 (60%) had undergone radiotherapy, 38 (29%) had received chemotherapy, and 15 (11%) had received surgery.

The 335 patients consisted of 234 men (70%) and 101 women (30%), ranging in age from 15 to 89 years. The mean age at the time of the surgical resection for the first primary NSCLC was 59.8 years. One hundred and ninety-eight of 335 patients with primary NSCLCs had stage I disease, 84 patients had stage II disease, 49 patients had stage III disease, and 4 patients had stage IV disease. The histologic distribution of the first primary cancers at the time of the surgical resection was 36% adenocarcinomas, 53% squamous cell carcinomas, and 11% other cell types.

DNA extraction from paraffin block. Formalin-fixed, paraffin wax–embedded blocks containing at least 75% neoplastic tissues were cut into 10-μm-thick tissue sections. Serial tissue sections from each paraffin block were placed on slides prior to DNA extraction and stained with H&E to evaluate the admixture of tumorous/nontumorous tissues. Areas that corresponded to tumor were carefully microdissected from the surrounding normal stromal tissues, collected in 15 mL centrifuge tubes, and deparaffinized overnight at 63°C in xylene, and then vortexed vigorously. After centrifugation at full speed for 5 minutes, supernatants were removed, ethanol was added to remove residual xylene, and then removed by centrifugation. After ethanol evaporation, tissue pellets were resuspended in lysis buffer ATL (DNeasy Tissue Kit, Qiagen, Valencia, CA) and the genomic DNA was isolated using a DNeasy Tissue kit according to the manufacturer's instruction.

Methylation-specific PCR. The methylation status of the promoter region of the p16, RARβ2, GSTP1, H-cadherin, and FHIT genes was determined by methylation-specific PCR, as described by Herman et al. (Fig. 1; ref. 5). Two sets of primers were designed for each gene, one specific for DNA methylated at the promoter region and the other specific for unmethylated DNA. Primer sequences and annealing temperatures for the methylation-specific PCR were previously published by our group and others (57). Briefly, 1 μg of genomic DNA was denatured by incubation with 0.2 mol/L NaOH for 10 minutes at 37°C. Aliquots of 3 mol/L sodium bisulfite (pH 5.0) and 10 mmol/L of hydroquinone (both from Sigma Chemical, Co., St. Louis, MO) were then added, and the solution was incubated at 50°C for 16 hours. The modified DNA was purified by use of a Wizard DNA purification system (Promega Corp., Madison, WI), followed by ethanol precipitation. Modified DNA was stored in aliquots at −20°C until required.

Figure 1.

Methylation analysis of the p16, RARβ2, H-cadherin, GSTP1, and FHIT genes in NSCLCs. Methylation-specific PCR for the p16, RARβ2, H-cadherin, GSTP1, and FHIT genes were done using unmethylation-specific (U) and methylation-specific (M) primer sets. Twenty microliters of PCR product was run on 2% metaphore agarose gel, stained with ethidium bromide, and visualized under UV illumination. Patient identification numbers indicated (top). DNA from normal lymphocytes served as a negative controls for methylated alleles. Negative control samples without DNA were included for each PCR.

Figure 1.

Methylation analysis of the p16, RARβ2, H-cadherin, GSTP1, and FHIT genes in NSCLCs. Methylation-specific PCR for the p16, RARβ2, H-cadherin, GSTP1, and FHIT genes were done using unmethylation-specific (U) and methylation-specific (M) primer sets. Twenty microliters of PCR product was run on 2% metaphore agarose gel, stained with ethidium bromide, and visualized under UV illumination. Patient identification numbers indicated (top). DNA from normal lymphocytes served as a negative controls for methylated alleles. Negative control samples without DNA were included for each PCR.

Close modal

The PCR mixture contained 1× PCR buffer [50 mmol/L KCl, 67 mmol/L Tris (pH 8.7), 1.5 mmol/L MgCl2], deoxynucleotide triphosphates (1.25 mmol/L each), primers (300 ng each per reaction), 2.5 units of Taq polymerase, and bisulfite-modified DNA (50 ng). Reactions were hot-started at 94°C before adding 2.5 units of Taq polymerase. Amplification was carried out over 35 cycles (1 minute at 94°C, 1 minute at the annealing temperature, and 1 minute at 72°C), followed by 4 minutes at 72°C. Fifteen microliters of the PCR reaction were loaded on 2% agarose gel, stained with ethidium bromide (Life Technology, Inc., Gaithersburg, MD), and visualized under UV illumination.

DNA from peripheral blood lymphocytes of a healthy individual was treated with SssI methyltransferase (New England Biolabs, Inc., Beverly, MA), subjected to bisulfite modification, and used as a positive control for the methylated alleles. Bisulfite-modified DNA from normal lymphocytes served as a positive control for the unmethylated alleles, and unconverted DNA from normal lymphocytes was used as a negative control for methylated alleles. Negative control samples without DNA were included in each PCR set.

Statistical analysis. The Wilcoxon rank sum test and χ2 test (or the Fisher's exact test) were used to analyze continuous and categorical variables by univariate analysis, respectively. Multivariate logistic regression analysis was conducted to estimate the relationship between the development of recurrence and the covariates found to be statistically significant in univariate analysis after controlling for potential confounding factors, and to calculate odds ratios. The effect of promoter methylation on time to death or recurrence was estimated by the Kaplan-Meier method, and the significance of differences in survival between the two groups was evaluated by the log-rank test. The primary end point was recurrence, and the secondary end point was death. Recurrence-free survival was calculated from the date of surgery to the date of recurrence. Overall survival was measured from the date of surgery to the time of last follow-up or death. Cox proportional hazards regression analysis was used to estimate the hazard ratios of independent factors for survival, after controlling for potential confounding factors such as age, sex, histology, and smoking. All statistical analyses were two-sided, with a 5% type I error rate.

Clinicopathologic characteristics and recurrence. The associations between disease recurrence and clinicopathologic features are listed in Table 1. At a median follow-up of 44 months, 132 patients (39%) had developed recurrence of disease; distant recurrence in 68%, distant and local recurrence in 12%, and local recurrence in 20%. The mean age of patients was similar between those with recurrence and those without (P = 0.86). The recurrence occurred slightly more frequently in males (40%) than in females (39%), but the difference was not statistically significant (P = 0.85). Smoking packyears was not associated with the risk of recurrence (P = 0.36). However, recurrence occurred more frequently in current smokers than those who never smoked or ex-smokers (P = 0.03). The prevalence of recurrence was significantly different according to histologic subtypes (P = 0.001). Recurrence occurred in 65 (54%) of 121 patients with adenocarcinoma and in 56 (31%) of 178 patients with squamous cell carcinoma. A significant relationship was also found between recurrence and pathologic stage (P = 0.009): recurrence occurred in 31% of those with stage I, in 54% of those with stage II, in 45% of those with stage III, and in 75% of those with stage IV. T grade was not associated with recurrence (P = 0.70), but the prevalence of recurrence increased with N grade (P = 0.001): 32% (71 of 223) for N0, 53% (48 of 90) for N1, and 65% (13 of 22) for N2. No relationship was found between recurrence and differentiation (P = 0.15).

Table 1.

Clinicopathologic characteristics (n = 335)

Recurrence
P
NoYes
Age* 59 ± 10 59 ± 12 0.86 
Sex    
    Male 141 93  
    Female 62 39 0.85 
Packyears 52 ± 29 57 ± 25 0.36 
Smoking status    
    Never 24  
    Former 103 60  
    Current 76 65 0.03 
Histology    
    Adenocarcinoma 56 65  
    Squamous 122 56  
    Others 25 11 0.001 
Pathologic stage    
    1 136 62  
    2 39 45  
    3 27 22  
    4 0.009 
Differentiation    
    Well 27 29  
    Moderate 125 72  
    Poor 43 28  
    Undifferentiated 0.15 
Recurrence
P
NoYes
Age* 59 ± 10 59 ± 12 0.86 
Sex    
    Male 141 93  
    Female 62 39 0.85 
Packyears 52 ± 29 57 ± 25 0.36 
Smoking status    
    Never 24  
    Former 103 60  
    Current 76 65 0.03 
Histology    
    Adenocarcinoma 56 65  
    Squamous 122 56  
    Others 25 11 0.001 
Pathologic stage    
    1 136 62  
    2 39 45  
    3 27 22  
    4 0.009 
Differentiation    
    Well 27 29  
    Moderate 125 72  
    Poor 43 28  
    Undifferentiated 0.15 
*

Values are mean ± SE.

Data are missing for six patients.

Recurrence rate and methylation. The relationship between recurrence and the promoter hypermethylation of five genes was investigated to identify biomarkers that would be useful as a risk factor of recurrence. Of the 335 NSCLC samples, promoter methylation was detected in 35% for p16, 39% for RARβ, 42% for H-cadherin, 7% for GSTP1, and 21% for FHIT. In univariate analysis, promoter methylation of any single gene was not associated with recurrence (data not shown). Also, the incidence of tumor recurrence was not found to be associated with cohypermethylation of any genes studied by crude data analysis (data not shown). Because tumor recurrence was found to be significantly associated with pathologic stage at the time of curative resection (P = 0.009; Table 1), we stratified the data according to pathologic stage and reanalyzed the relationship between cohypermethylation of genes and the development of recurrence. Using recursive partitioning analysis, we defined recurrently divergent groups on the basis of combinatiorial gene hypermethylation status. Among the five genes studied, cohypermethylation of p16 and FHIT genes was significantly associated with tumor recurrence in stage I cancers (Fig. 2).

Figure 2.

Prevalence of recurrence according to cohypermethylation of p16 and FHIT genes. Prevalence (56%, 18 of 32) of recurrence in patients with cohypermethylation of p16 and FHIT genes is significantly higher than those without cohypermethylation (27%, 44 of 166) of both genes in only stage I (P = 0.001), but not stage II and stage III.

Figure 2.

Prevalence of recurrence according to cohypermethylation of p16 and FHIT genes. Prevalence (56%, 18 of 32) of recurrence in patients with cohypermethylation of p16 and FHIT genes is significantly higher than those without cohypermethylation (27%, 44 of 166) of both genes in only stage I (P = 0.001), but not stage II and stage III.

Close modal

For stage I cancers, recurrence was found to occur in 56% (18 of 32) of patients with cohypermethylation of p16 and FHIT genes and in 27% (44 of 166) of those without, and this difference was statistically significant (P = 0.001). The prevalence of recurrence for stage II and stage III cancers was not significantly different between patients with and without cohypermethylation of p16 and FHIT genes (Fig. 2).

Multivariate logistic regression analysis of recurrence. The data was stratified according to pathologic stage, and stratified multivariate logistic regression analysis was done to control for the potential confounding effects of variables, such as age, sex, smoking status, differentiation, adjuvant therapy and histology, and to calculate the odds ratios (Table 2). The coefficient for age variable was not determined to be statistically significant in our univariate analyses (P = 0.86), but age was considered to be a biologically important variable, and it was thus included in the multivariate analysis in order to better construct a parsimonious model. Recurrence occurred at 3.22 times higher prevalence [95% confidence interval (CI), 1.49-4.11; P = 0.001] in adenocarcinomas than in nonadenocarcinomas (squamous cell and other cell types). The risk of recurrence for stage I cases with the cohypermethylation of p16 and FHIT genes was determined to be 6.43 times as high as that of the reference group (95% CI, 1.04-20.19; P = 0.02). There was no relationship between cohypermethylation of p16 and FHIT genes and the risk of recurrence in stage II and stage III (data not shown).

Table 2.

Multivariate logistic regression analysis of recurrence in stage I NSCLCs (n = 198)

Odds ratio (95% CI)P
Histology   
    Nonadenocarcinoma 1.00  
    Adenocarcinoma 3.22 (1.49-4.11) 0.001 
p16*FHIT*   
    No 1.00  
    Yes 6.43 (1.04-20.19) 0.02 
Odds ratio (95% CI)P
Histology   
    Nonadenocarcinoma 1.00  
    Adenocarcinoma 3.22 (1.49-4.11) 0.001 
p16*FHIT*   
    No 1.00  
    Yes 6.43 (1.04-20.19) 0.02 

NOTE: Adjusted for age, sex, smoking status, differentiation, and adjuvant therapy.

*

p16*FHIT indicates cohypermethylation of p16 and FHIT genes.

Survival analysis. Overall 5-year survival rates for stage I and stage II were 72% and 42% (P < 0.001), respectively. Five-year recurrence-free survival rates were 57% for stage I and 35% for stage II, and this difference was statistically significant (P < 0.001). The recurrence-free and overall survivals were analyzed between patients with and without cohypermethylation of p16 and FHIT genes. Data was stratified by disease stage because stage is an independent risk factor in NSCLC. Kaplan-Meier survival estimates in stage I are shown in Fig. 3.

Figure 3.

Kaplan-Meier survival curves for p16 and FHIT methylation in stage I NSCLCs. Survival according to cohypermethylation of p16 and FHIT genes in stage I NSCLCs: (A) overall survival, (B) recurrence-free survival, and (C) survival after recurrence were poorer in patients with cohypermethylation of p16 and FHIT genes compared to those without. Black line, group without hypermethylation of p16 and FHIT genes. Green and red lines, unmethylated p16/methylation FHIT and methylated p16/unmethylated FHIT, respectively. Blue line, the group with cohypermethylation of p16 and FHIT genes. P values were calculated for all four groups using the log-rank test.

Figure 3.

Kaplan-Meier survival curves for p16 and FHIT methylation in stage I NSCLCs. Survival according to cohypermethylation of p16 and FHIT genes in stage I NSCLCs: (A) overall survival, (B) recurrence-free survival, and (C) survival after recurrence were poorer in patients with cohypermethylation of p16 and FHIT genes compared to those without. Black line, group without hypermethylation of p16 and FHIT genes. Green and red lines, unmethylated p16/methylation FHIT and methylated p16/unmethylated FHIT, respectively. Blue line, the group with cohypermethylation of p16 and FHIT genes. P values were calculated for all four groups using the log-rank test.

Close modal

For the 198 stage I cases, the overall 5-year survival rates were 52% and 78% in patients with and without cohypermethylation of p16 and FHIT genes (P = 0.03; Fig. 3A). Recurrence-free 5-year survival rates for the 198 stage I cases with and without cohypermethylation of p16 and FHIT genes were 43% and 69%, respectively. This difference was statistically significant (P = 0.01; Fig. 3B). Survival after recurrence in recurrent cases was finally examined with respect to cohypermethylation of p16 and FHIT genes. Of the 62 patients that did experience recurrence in stage I, survival after recurrence was found to be extremely poor in patients with cohypermethylation of p16 and FHIT genes (Fig. 3C). The median survival after recurrence in recurrent stage I cases with and without cohypermethylation of p16 and FHIT genes was 12 and 26 months, respectively; this difference was statistically significant (P = 0.009). For stage II or stage III cases, overall survival, recurrence-free survival, and survival after recurrence was not significantly different in patients with and without cohypermethylation of p16 and FHIT genes (data not shown).

Cox proportional hazards analysis. The stratified Cox proportional hazards model was created according to pathologic stage to determine whether cohypermethylation of p16 and FHIT genes was an independent prognostic factor of survival associated with recurrence, after controlling for potential confounding factors, including age, sex, smoking, differentiation, histology, adjuvant therapy, and the types of treatment received for recurrent cases (Table 3). The present data showed that the cohypermethylation of p16 and FHIT genes was a negative prognostic factor for survival. Patients with cohypermethylation of p16 and FHIT genes in stage I were found to have poorer prognosis than those without cohypermethylation of both genes (hazard ratio, 2.67; 95% CI, 1.21-7.64); this difference was statistically significant (P = 0.02). The recurrence-free survival of stage I cases was also poorer in patients with cohypermethylation of p16 and FHIT genes than those without (hazard ratio, 2.03; 95% CI, 1.09-6.23; P = 0.02). For recurrent stage I cases, the hazard of failure after recurrence was about 4.62 times higher (95% CI, 1.27-16.48; P = 0.005) in patients with cohypermethylation of p16 and FHIT genes than those without. There was no relationship between patient survival and the cohypermethylation of p16 and FHIT genes in stage II and stage III (data not shown).

Table 3.

Stratified Cox proportional hazards analysis in stage I NSCLCs

Hazard ratio (95% CI)P
(A) Overall survival (n = 198)   
    p16*FHIT* 2.67 (1.21-7.64) 0.02 
(B) Recurrence-free survival (n = 198)   
    p16*FHIT* 2.03 (1.09-6.23) 0.02 
(C) Survival after recurrence (n = 62)   
    p16*FHIT* 4.62 (1.27-16.48) 0.005 
Hazard ratio (95% CI)P
(A) Overall survival (n = 198)   
    p16*FHIT* 2.67 (1.21-7.64) 0.02 
(B) Recurrence-free survival (n = 198)   
    p16*FHIT* 2.03 (1.09-6.23) 0.02 
(C) Survival after recurrence (n = 62)   
    p16*FHIT* 4.62 (1.27-16.48) 0.005 

NOTE: Adjusted for age, sex, smoking status, differentiation, histology, adjuvant therapy, and the type of treatment for recurrent cases.

*

p16*FHIT indicates cohypermethylation of p16 and FHIT genes and reference group is patients without cohypermethylation of both genes.

The long-term survival of patients with NSCLC remains poor because of the high incidence of lymph node metastasis and because of early recurrence after curative surgical resection. Therefore, it is critically important to identify groups at high risk of recurrence, to achieve effective treatment after surgery. In the present study, 39% (132 of 335) of the patients experienced recurrence after curative resection, and the prognosis of these cases was poor compared with those with no recurrences (data not shown). We examined any possible association of methylation status of p16, RARB, FHIT, GSTP1, and ECAD genes with recurrence to identify molecular markers to predict the recurrence of lung cancer after surgery. In the present study, the methylation status of no individual gene had a statistically significant effect on the risk of recurrence in the univariate analysis. These findings are consistent with the findings of Kim et al. (3). Pathologic stage in this study was significantly associated with the risk of tumor recurrence (P = 0.009). Therefore, we stratified the data according to the pathologic stage and reanalyzed the relationship between tumor recurrence and the hypermethylation of those genes. We found that the cohypermethylation of both p16 and FHIT promoters in patients with stage I NSCLC was associated with the poor prognosis of patients and an increased risk of recurrence, suggesting that the cohypermethylation of p16 and FHIT genes may contribute to recurrence after curative resection of stage I NSCLC. The hypermethylation of p16 and FHIT genes in this study were significantly associated with the loss of protein expression (data not shown), suggesting a truly functional inactivation of the two genes. The protein expression was assessed by immunohistochemistry. Second primary lung cancers developed in 18 (5%) of the 335 NSCLCs, but the development of second primary lung cancers was not associated with the hypermethylation of p16 and/or FHIT (data not shown).

It remains unclear how cohypermethylation of p16 and FHIT genes contributes to the recurrence of NSCLC. The p16 gene product is an inhibitor of cyclin-dependent protein kinase 4 (cdk4) which phosphorylates the serine/threonine residues of the retinoblastoma protein (8, 9). The p16 protein inhibits cell cycle progression through G1 into S phase, past the G1 checkpoint by maintaining the retinoblastoma protein in an unphosphorylated state (10, 11). The effect of Fhit on the cell cycle remains the subject of debate (1214), but reports about the effects of Fhit on apoptosis have produced consistent results. FHIT reexpression in FHIT−/− negative cells that lack Fhit protein expression has been shown to suppress tumor formation by inducing apoptosis in a variety of human cell lines, including lung cancer cell lines (1420). The induction of apoptosis by FHIT gene transfer in lung cancer cell lines has been associated with the activation of caspase-8 (16). Thus, it is likely that cohypermethylation of p16 and FHIT genes may contribute to the process of multistep carcinogenesis of the lung through different mechanisms, cell cycle and apoptosis control, respectively.

Several reports have shown that the hypermethylation of the p16 gene is detected in the earliest stage of lung cancer and increases with tumor progression. Aberrant methylation of the p16 gene was frequently detected in precursor lesions to lung tumors in rats that were treated with tobacco-specific 4-(methylnitrosamino)-I-(3-pyridyl)-1-butanone and increased with disease progression from basal cell hyperplasia (17%) to squamous metaplasia (24%), and then to carcinoma in situ (50%; ref. 21). Aberrant methylation of the p16 gene was also detected in sputum, bronchial lavage, or bronchial brush samples from subjects without any evidence of lung cancer, which suggested that p16 methylation may occur in preneoplastic stage (2226). The loss of Fhit also occurs very early in lung cancer (2729), which suggests that Fhit is involved in the initiation of lung cancer tumorigenesis rather than in the progression of lung cancer. Given that hypermethylation of both genes is an early event in the carcinogenesis of lung cancer, the cells that remain after a complete tumor resection in stage I cases with cohypermethylation of p16 and FHIT genes might contribute to the growth advantage and expansion of cells by the failure of both cell cycle and apoptosis control, and eventually increase the risk of recurrence.

After surgery of an initial carcinoma, part of a preneoplastic precursor lesion (i.e., the field) with genetically altered cells may remain in the patient and present a continuous risk for developing a new cancer. However, data about genetic alteration of preneoplastic cells in a field surrounding a tumor were not available in this study. Thus, we could not discriminate a recurrent carcinoma that has developed from minimal residual cancer cells that were left in or near the surgical margins and a second field tumor that developed from preneoplastic precursor cells clonally related to the cells of the excised initial tumor (30), which is in line with the “field cancerization” concept (31). Some of the second field tumors in this study might be misclassified as locally recurrent tumors. Additional studies are needed to differentiate a second field tumor and a recurrent tumor, to validate the usefulness of cohypermethylation of p16 and FHIT genes in identifying a group at high risk of recurrence after surgery.

Fhit-deficient normal cells and cancer cells are more resistant to radiation treatment (UVC and ionizing radiation) and drugs (mitomycin C and cisplatin; ref. 32) and have stronger IR-induced S and G2 checkpoint responses than Fhit+/+ cells (33, 34), suggesting an association between FHIT gene inactivation and increased survival after DNA damage. The overactive checkpoints are known to be regulated by the ATR-CHK1 pathway, which contributes to the radioresistance of Fhit−/− cells (35). Accordingly, the effect of treatment on survival after recurrence in recurrent stage I cases was controlled by stratification and multivariate analysis. However, stratification of data according to the types of treatment for recurrent stage I cases did not show a significant difference in the survival after recurrence between patients with and without cohypermethylation of p16 and/or FHIT genes (data not shown). In contrast, the survival was significantly poorer (hazard ratio, 4.62; 95% CI, 1.27-16.48; P = 0.005) in patients with cohypermethylation of p16 and FHIT genes than those without, after adjusting age, sex, smoking status, histology, and the types of treatment in multivariate analysis (Table 3C). The lack of effect of p16/FHIT methylation on survival after recurrence in the stratified data may have resulted from uncontrolled confounding factors or from the small number of cases in each type of treatment. Among 62 recurrent stage I cases, only 38 patients received radiotherapy and 16 cases received chemotherapy. More studies are needed with a large number of samples to show the effect of Fhit deficiency on survival after recurrence, after controlling the types of treatment received for recurrence.

In the present study, lymph node metastasis, male gender, and adenocarcinoma were found to be significantly associated with an increased risk of recurrence after complete curative resection. The relationship between recurrence and histology is controversial. Our data showed similar results to those of the Lung Cancer Study Group (36) and to recent data reported by Okada et al. (37). In the Lung Cancer Study Group, cancer recurrence was more frequent in non–squamous cell types. Okada et al. (37) also showed that advanced stage, high involvement of lymph nodes, male gender, and non–squamous cell cancer were independent, unfavorable prognostic factors in patients with completely resected lung cancer. In contrast, some groups (2, 38) reported that histologic type did not play a statistically significant role in the incidence of recurrence, and Rena et al. (39) reported that adenocarcinoma has better 5-year survival rates than did squamous cell carcinoma.

Although no significant relationship was observed between cohypermethylation of p16 and FHIT genes and the risk of recurrence in stage II and stage III, this may result from the small number of the sample size. Further study in a larger sample is needed to investigate the significance of cohypermethylation of p16 and FHIT genes in identifying groups at high risk of recurrence in advanced cases. Even though independent studies on a larger scale are required to validate its usefulness before clinical application, the cohypermethylation status of p16 and FHIT genes could be valuable as a molecular biomarker in identifying patients at high risk of recurrence during follow-up treatment after complete tumor resection. These patients might benefit from a more aggressive treatment strategy. The ras gene activation is known to be an early event in lung adenocarcinoma and is a good predictor of poor prognosis. Therefore, the combination effect of ras gene activation with p16/FHIT methylation on prognosis for adenocarcinoma of the lung needs to be investigated. In conclusion, cohypermethylation of p16 and FHIT genes might be a strong indicator of a high risk of recurrence as well as an independent prognostic factor in cases of stage I NSCLC.

Grant support: Samsung Biomedical Research Institute (B-A5-102) and the SRC/ERC program of MOST/KOSEF (R11-2005-017).

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.

The authors thank Eun-Kyung Kim for data collection and management, and Hoon Suh for sample collection.

1
Thomas P, Rubinstein L. Lung Cancer Study Group. Cancer recurrence after resection: T1 N0 non-small cell lung cancer.
Ann Thorac Surg
1990
;
49
:
242
–7.
2
Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage 1 lung cancer.
J Thorac Cardiovasc Surg
1995
;
109
:
120
–9.
3
Kim YT, Lee SH, Sung SW, Kim JH. Can aberrant promoter hypermethylation of CpG islands predict the clinical outcome of non-small cell lung cancer after curative resection.
Ann Thorac Surg
2005
;
79
:
1180
–8.
4
Zöchbauer-Müller S, Minna JD, Gazdar AF. Aberrant DNA methylation in lung cancer: biological and clinical implications.
Oncologist
2002
;
7
:
451
–7.
5
Herman JG, Graff JR, Myöhänen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands.
Proc Natl Acad Sci U S A
1996
;
93
:
9821
–6.
6
Kim HJ, Kwon YM, Kim JS, et al. Tumor-specific methylation in bronchial lavage for the early detection of non-small cell lung cancer.
J Clin Oncol
2004
;
22
:
2363
–70.
7
Zöchbauer-Müller S, Fong KM, Virmani AK, Geradts J, Gazdar AF, Minna JD. Aberrant promoter methylation of multiple genes in non-small cell lung cancers.
Cancer Res
2001
;
61
:
249
–55.
8
Sherr CJ. G1 phase progression: cycling on cue.
Cell
1994
;
79
:
551
–5.
9
Weinberg RA. The retinoblastoma protein and cell cycle control.
Cell
1995
;
81
:
323
–30.
10
Serrano M, Hannon GJ, Beach D. A new regulatory motif in cell cycle control causing specific inhibition of cyclin D/CDK4.
Nature
1993
;
366
:
704
–7.
11
Kamb A, Guis NA, Weaver-Feldhaus J, et al. A cell cycle regulator potentially involved in genesis of many tumor types.
Science
1994
;
264
:
436
–40.
12
Mascaux C, Martin B, Verdebout JM, Meert AP, Ninane V, Sculier JP. Fragile histidine triad protein expression in nonsmall cell lung cancer and correlation with Ki-67 and with p53.
Eur Respir J
2003
;
21
:
753
–8.
13
Burke L, Khan MA, Freedman AN, et al. Allelic deletion analysis of the FHIT gene predicts poor survival in non-small cell lung cancer.
Cancer Res
1998
;
58
:
2533
–6.
14
Ji L, Fang B, Yen N, Fong K, Minna JD, Roth JA. Induction of apoptosis and inhibition of tumorigenicity and tumor growth by adenovirus vector-mediated fragile histidine triad (FHIT) gene overexpression.
Cancer Res
1999
;
59
:
3333
–9.
15
Dumon KR, Ishii H, Vecchione A, et al. Fragile histidine triad expression delays tumor development and induces apoptosis in human pancreatic cancer.
Cancer Res
2001
;
61
:
4827
–36.
16
Roz L, Gramegna M, Ishii H, Croce CM, Sozzi G. Restoration of fragile histidine triad (FHIT) expression induces apoptosis and suppresses tumorigenicity in lung and cervical cancer cell lines.
Proc Natl Acad Sci U S A
2002
;
99
:
3615
–20.
17
Sevignani C, Calin GA, Cesari CR, et al. Restoration of fragile histidine triad (FHIT) expression induces apoptosis and suppresses tumorigenicity in breast cancer cell lines.
Cancer Res
2003
;
63
:
1183
–7.
18
Sard L, Accornero P, Tornielli S, et al. The tumor-suppressor gene FHIT is involved in the regulation of apoptosis and in cell cycle control.
Proc Natl Acad Sci U S A
1999
;
96
:
8489
–92.
19
Ishii H, Dumon KR, Vecchione A, et al. Effect of adenoviral transduction of the fragile histidine triad gene into esophageal cancer cells.
Cancer Res
2001
;
61
:
1578
–84.
20
Pavelic K, Krizanac S, Cacev T, et al. Aberration of FHIT gene is associated with increased tumor proliferation and decreased apoptosis-clinical evidence in lung and head and neck carcinomas.
Mol Med
2001
;
7
:
442
–53.
21
Belinsky SA, Nikula KJ, Palmisano WA, et al. Aberrant methylation of p16(INK4a) is an early event in lung cancer and a potential biomarker for early diagnosis.
Proc Natl Acad Sci U S A
1998
;
95
:
11891
–6.
22
Belinsky SA, Palmisano WA, Gilliland FD, et al. Aberrant promoter methylation in bronchial epithelium and sputum from current and former smokers.
Cancer Res
2002
;
62
:
2370
–7.
23
Kersting M, Friedl C, Kraus A, Pankow BW, Schuermann M. Differential frequencies of p16INK4a promoter hypermethylation, p53 mutation, and K-ras mutation in exfoliative material mark the development of lung cancer in symptomatic chronic smokers.
J Clin Oncol
2000
;
18
:
3221
–9.
24
Palmisano WA, Divine KK, Saccomanno G, et al. Predicting lung cancer by detecting aberrant promoter methylation in sputum.
Cancer Res
2000
;
60
:
5954
–8.
25
Soria J-C, Rodriguez M, Liu DD, Lee JJ, Hong WW, Mao L. Aberrant promoter methylation of multiple genes in bronchial brush samples from former cigarette smokers.
Cancer Res
2002
;
62
:
351
–5.
26
Zöchbauer-Müller S, Lam S, Toyooka S, et al. Aberrant methylation of multiple genes in the upper aerodigestive tract epithelium of heavy smokers.
Int J Cancer
2003
;
107
:
612
–6.
27
Sozzi G, Pastorino U, Moiraghi L, et al. Loss of FHIT function in lung cancer and preinvasive bronchial lesions.
Cancer Res
1998
;
58
:
5032
–7.
28
Geradts J, Fong KM, Zimmerman PV, Minna JD. Loss of Fhit expression in non-small-cell lung cancer: correlation with molecular genetic abnormalities and clinicopathological features.
Br J Cancer
2000
;
82
:
1191
–7.
29
Nelson HH, Wiencke JK, Gunn L, Wain JC, Christiani DC, Kelsey KT. Chromosome 3p14 alterations in lung cancer: evidence that FHIT exon deletion is a target of tobacco carcinogens and asbestos.
Cancer Res
1998
;
58
:
1804
–7.
30
Braakhuis BJM, Brakenhoff RH, Leemans CR. Second field tumors: a new opportunity for cancer prevention?
Oncologist
2005
;
10
:
493
–500.
31
Slaughter DP, Southwick HW, Smejkal W. “Field cancerization” in oral stratified squamous epithelium.
Cancer
1953
;
6
:
963
–8.
32
Ottey M, Han S-Y, Druck T, et al. Fhit deficient normal cancer cells are mitomycin C and UVC resistant.
Br J Cancer
2004
;
91
:
1669
–77.
33
Fong LYY, Fidanza V, Zanesi N, et al. Muir-Torre-like syndrome in Fhit-deficient mice.
Proc Natl Acad Sci U S A
2000
;
97
:
4742
–7.
34
Turner B, Ottey M, Otoczek M, et al. The FHIT/FRA3B locus and repair deficient cancers.
Cancer Res
2002
;
62
:
4054
–60.
35
Hu B, Han S-Y, Wang X, et al. Involvement of the Fhit gene in the ionizing radiation-activated ATR/CHK1 pathway.
J Cell Physiol
2005
;
202
:
518
–23.
36
Thomas PA, Jr., Rubinstein L. Malignant disease appearing late after operation for T1N0 non-small cell lung cancer. The Lung Cancer Study Group.
J Thorac Cardiovasc Surg
1993
;
106
:
1053
–8.
37
Okada M, Nishio W, Sakamoto T, Harada H, Uchino K, Tsubota N. Long-term survival and prognostic factors of five-year survivors with complete resection of non-small cell lung carcinoma.
J Thorac Cardiovasc Surg
2003
;
126
:
558
–62.
38
Ramacciato G, Paolini A, Volpino P, et al. Modality of failure following resection of stage I and stage II non-small cell lung cancer.
Int Surg
1995
;
80
:
156
–61.
39
Rena O, Oliaro A, Cavallo A, et al. Stage I non-small cell lung carcinoma: really an early stage?
Eur J Cardiovasc Surg
2002
;
21
:
514
–9.