Combined Effect of Genetic Polymorphisms in P53, P73, and MDM2 on Non-small Cell Lung Cancer Survival

ORIGINAL ARTICLE

Combined Effect of Genetic Polymorphisms in P53, P73, and MDM2 on Non-small Cell Lung Cancer Survival Li Liu, PhD,* Chen Wu, PhD,† Ying Wang, PhD,* Rong Zhong, PhD,* Shengyu Duan, MD,* Sheng Wei, PhD,* Songyi Lin, MD,* Xinyu Zhang, PhD,‡ Wen Tan, PhD,† Dianke Yu, PhD,† Shaofa Nie, MD,* Xiaoping Miao, PhD,* and Dongxin Lin, PhD†

Introduction: Multiple biologically relevant polymorphisms may have more accurate prediction of cancer prognosis compared with single polymorphism because of the modest effect. This study investigated whether the functional polymorphisms in P53 pathway genes, P53 Arg72Pro (rs1042522), P73 G4C14-to-A4T14 (rs2273953 and rs1801173), and MDM2 T309G (rs2279744), alone or in combination, affect survival in advanced non-small cell lung cancer (NSCLC) patients. Methods: A total of 199 stage III–IV NSCLC patients with platinum-based chemotherapy were recruited between 2002 and 2004. Associations between genotypes and survival were assessed using Kaplan-Meier method. Cox proportional hazard models were performed to identify significant variables. Results: During the median 26.5 months of follow-up, the P53 Pro/Pro genotype was strongly associated with shorter overall survival compared with the Arg/Arg genotype (12.0 versus 20.0 months; log-rank p ⫽ 0.002; hazard ratio ⫽ 1.86; 95% confidence interval [CI], 1.15–3.02). Pairwise combination analysis showed that patients carrying the variant P53 Pro/Pro-P73 GC/GC or P53 Pro/Pro-MDM2 GG genotypes had survival time only half of that for those carrying the wild-type genotypes, with hazard ratio being 2.47 (95% CI, 1.20 –5.10) and 2.00 (95% CI, 1.15–3.46), respectively. Furthermore, a combined effect was seen with survival time being *Department of Epidemiology and Biostatistics and MOE Key Lab of Environment and Health, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan; †State Key Laboratory of Molecular Oncology and Department of Etiology and Carcinogenesis, Cancer Institute and Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing; and ‡Tsinghua National Laboratory for Information Science and Technology, Department of Computer Science and Technology, Tsinghua University, Beijing, China. Disclosure: The authors declare no conflicts of interest. Address for correspondence: Xiaoping Miao, PhD, Department of Epidemiology and Biostatistics and MOE Key Lab of Environment and Health, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China. E-mail: [email protected] or Dongxin Lin, PhD, State Key Laboratory of Molecular Oncology and Department of Etiology & Carcinogenesis, Cancer Institute and Hospital, Chinese Academy of Medical Science and Peking Union Medical College, Beijing 100021, China. E-mail: [email protected] The first two authors contributed equally to this study. Copyright © 2011 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/11/0611-1793

gradually shorter with increasing number of unfavorable genotypes in these three genes (ptrend ⫽ 0.039), indicating a gene-dose effect in association with survival. Conclusions: These findings suggest that genetic polymorphisms in the P53 pathway may be promising biomarkers for individualized chemotherapy and prognosis of NSCLC patients. Key Words: Non-small cell lung cancer, Survival, P53, P73, MDM2. (J Thorac Oncol. 2011;6: 1793–1800)

F

or the disseminated non-small cell lung cancer (NSCLC) patients, systemic chemotherapy is the only proven approach of treatment, of which platinum agent as one of the most active anticancer agents has been widely used and yielded improved patient survival.1 However, great interindividual variability in response to platinum drugs correlated to a spectrum of outcomes ranging from limited therapeutic effect to toxic reaction occurred among patients.2 Recently, pharmacogenetic studies evolving from single genetic polymorphism to microarray-based interrogation of thousand genes have tried to identify biomarkers that may predict individual outcome of chemotherapy aiming to select patients who would derive significant benefit from maximized efficacy and minimal toxicity. However, few or none has been established in the clinical setting yet.3 Platinum drugs kill tumor cells by means of cytotoxic mechanism, which is mostly used by the formation of platinum-DNA adducts and ultimately triggering apoptosis.4 Emerging evidence has suggested that functional singlenucleotide polymorphisms (SNPs) involved in cell-cycle control, DNA repair, and apoptosis may modulate response to platinum-based chemotherapy and affect outcome of NSCLC patients treated with the regimen.2,3 P53, a well-recognized transcriptional factor, regulates the expression, cellular location, and activity of key components in cellular processes including cell-cycle arrest, DNA repair, and apoptotic cell death in response to cellular stress including chemotherapy.5 Therefore, normal P53 activity is critical for chemosensitivity of cancer cells, and chemoresistance may be developed when P53 activity is inhibited.4 Multiple functional SNPs occur in the P53 gene and the most frequently studied is the G to C change at codon 72 in exon

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

1793

Liu et al.

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

4 resulting in the amino acid substitution of Arg to Pro (Arg72Pro, rs1042522). This SNP has been shown to alter P53 function because it is located in the proline-rich region that is required for P53 to induce apoptosis.6 Emerging evidence has shown that the P53 Arg72Pro polymorphism is not only associated with lung cancer risk but also affects individual sensitivity to platinum-drug chemotherapy and patients’ survival.7–11 In response to cellular stress, the activation of P53 also requires communication and coordination with other components of the P53 pathway where P53 acts as a central node and a set of genes are involved in cell-cycle arrest and apoptosis. In the P53 pathway, P73 is a structural and functional homologue of P53 that can bind to the P53 responsive elements and transactivate an overlapping set of P53 target genes and thus also implicated in cell-cycle control and apoptotic cell death.7 Accumulating evidence has demonstrated that P73 is essential for apoptosis induced by platinum drugs, and inactivation of P73 by mutations leads to resistance of tumor cells to cytotoxic agents.8,9 Although P73 mutation is rare in all cancers, loss of heterozygosity at P73 locus is relatively common in cancer including lung cancer.10 Two common SNPs at nucleotides 4 (G to A) and 14 (C to T) in the 5⬘ untranslated region of exon 2 in P73 (G4C14-toA4T14; rs2273953, rs1801173) have been identified to be in complete linkage disequilibrium with each other as one variation. This dinucleotide polymorphism lies just upstream of the initiating AUG of exon 2, a region of the transcript that may theoretically form a stem-loop structure and possibly affect the gene expression.11 The association of this dinucleotide polymorphism with risk of NSCLC has been documented in several epidemiological studies.12–14 However, the role of the P73 G4C14-to-A4T14 polymorphism in response to platinum-drug chemotherapy and prognosis of NSCLC has not been investigated. Another major regulator of P53 is human homolog of mouse double minute 2 (MDM2), which inhibits P53 activity by directly binding to the N-terminal transcription of domain, mediating P53 location and degradation through the ubiquitination system.15 Under physiologic conditions, P53 is maintained at low level by MDM2 for normal cell growth. However, overexpression of MDM2 may attenuate P53 function, enabling damaged cells to escape the cell cycle checkpoint control.16 Numerous studies have shown that a subset of tumors overexpressed MDM2, which was associated with cancer progression and lack of response to chemotherapy.17–19 A common T to G polymorphism at nucleotide 309 (T309G, rs2279744) in the promoter region of MDM2 has been widely investigated in cancer, and the results suggested that this polymorphism may accelerate cancer progression and prevent tumor cells from killing by chemotherapy because of its association with increased MDM2 expression and attenuated activity of P53 pathway.20 –22 Because of the key role of the P53 pathway in cellular process in response to DNA damage, we hypothesized that the functional polymorphisms in P53, P73, and MDM2 may have cooperatively impact on outcome of NSCLC treated with platinum-based chemotherapy. To test this hypothesis, we recruited 199 advanced NSCLC patients treated with

1794

platinum-based chemotherapy and investigated whether the functional polymorphisms in these genes, alone or in combination, affect patients’ survival.

PATIENTS AND METHODS Patients Patients with histologically confirmed NSCLC who received platinum-based chemotherapy were enrolled between September 2002 and October 2004 from Cancer Hospital of Chinese Academy of Medical Sciences (Beijing, China). The eligible patients were required to have pathologically documented stage III or IV NSCLC, measurable lesions, Eastern Cooperative Oncology Group (ECOG) performance status of 0, 1, or 2, and normal results of completed blood counts, liver function, and urinalysis. Pretreatment staging involved medical history and physical examination, chest x-ray, chest computed tomography scan, bone scan, and abdominal ultrasonogram. A total of 199 patients were eligible to participate in this study and had adequate blood DNA for genotyping. All the patients had completed follow-up and clinical information. Demographic and clinical data were systemically recorded at entry (including age, gender, smoking status, performance status, disease stage, and tumor histology) and along follow-up (treatment regimen, tumor recurrence and/or metastasis, and death). In this study, smoking status was categorized as never-smoker (having smoked less than 100 cigarettes during their lifetime) or ever-smoker (current smoker and ex-smoker). Written informed consent for the use of DNA sample and clinical information was obtained from each subject. The study was conducted under the approval of the institutional review boards of the Chinese Academy of Medical Sciences Cancer Institute and the Tongji Medical College of Huazhong University of Science and Technology.

SNP Genotyping Genomic DNA was extracted from blood sample collected from each patient at the time of recruitment. P53 Arg72Pro genotypes were determined by polymerase chain reaction-restriction fragment length polymorphism assay,23 whereas P73 G4C14-to-A4T14 and MDM2 T309G genotypes were determined using the tetra-amplification refractory mutation system-polymerase chain reaction method.23,24 A 15% blind, random sample of study subjects was genotyped twice, and the reproducibility was 100%.

Statistical Analysis Association between genotypes and demographic or clinical characteristics of patients were assessed using ␹2 test. For each polymorphism, the Hardy-Weinberg equilibrium was tested by Person’s ␹2 test. Overall survival (OS) was calculated as the time from diagnosis to lung cancer-related death or the date of last follow-up. Progression-free survival (PFS) was taken from the date of diagnosis to the date of disease progression, death without progression, or last follow-up. Median follow-up time was calculated among censored observations only. Association between genotypes and survival were estimated using the Kaplan-Meier method, and

Copyright © 2011 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

statistical significance was assessed by the log-rank test. Cox proportional hazards models were used to evaluate the independent effect of genetic polymorphisms on OS and PFS adjusted for age, gender, performance status, smoking status,

TABLE 1.

tumor stages, histological type, and chemotherapy. The joint effect of P53, P73, and MDM2 genotypes on NSCLC survival was also evaluated. All tests were two-tailed with p less than 0.05 as significant level. All statistical analyses were performed using software SPSS v12.0.

RESULTS

Patient, Treatment, and Follow-Up Characteristics

Characteristics

Patient Characteristics and Clinical Outcomes

n (%)

Median age, yr (range) Sex Female Male Smoking status Smoker Never-smoker ECOG performance status 0 1 2 Histology Adenocarcinoma Squamous carcinoma Adenosquamous Disease stage III (A/B) IV Chemotherapy regimens Cisplatin ⫹ vinorelbine/taxane Carboplatin ⫹ vinorelbine/taxane Oxaliplatin ⫹ vinorelbine Median follow-up time (mo) Events, deaths Median survival time (mo)

56 (29–74) 70 (35.2) 129 (64.8) 118 (59.3) 81 (40.7) 99 (49.7) 73 (36.7) 27 (13.6) 139 (69.8) 36 (18.1) 24 (12.1) 43 (21.6) 156 (78.4) 141 (70.9) 45 (22.6) 13 (6.5) 26.5 159 (79.9) 16.0

ECOG, Eastern Cooperative Oncology Group.

TABLE 2.

SNPs in P53 Pathway and NSCLC Survival

The main demographic and clinical characteristics of patients are shown in Table 1. By the time of the final analysis (May 31, 2008), the median follow-up time was 26.5 months (ranging from 6.0 to 60.0 months), and 159 patients died of cancer. The median OS and PFS time were 16.0 (95% confidence interval [CI], 13.5–18.5) and 6.9 (95% CI, 6.0 – 7.8) months, respectively. Overall, age, sex, smoking, ECOG performance, histological type, disease stages, and chemotherapy regimens were not significantly associated with patients’ OS and PFS in this study.

Association Between Single Genotype and Patients’ OS or PFS The frequency distributions of all genotypes among patients are presented in Table 2. All genotype frequencies for the P53 Arg72Pro, P73 G4C14-to-A4T14, and MDM2 T309G were in Hardy-Weinberg equilibrium (p ⫽ 0.860, 0.233, and 0.803, respectively). No associations were detected between any of these polymorphisms and sex, age, performance status, smoking status, histological type, disease stages, or chemotherapy regimens. The P53 Arg72Pro polymorphism was significantly associated with patients’ survival. Patients with the Pro/Pro genotype had median OS of 12.0 (95% CI, 8.5–15.5) months significantly shorter than that for those with the Arg/Arg genotype (median OS, 20.0 months; 95% CI, 15.3–24.7 months; log-rank p ⫽ 0.002; Figure 1). Moreover, patients

Associations Between P53 Arg72Pro, P73 G4C14-to-A4T14, and MDM2 T309G Genotypes and NSCLC Survival Overall Survival

Genotype P53 Arg72Pro Arg/Arg Arg/Pro Pro/Pro ptrend P73 G4C14-to-A4T14 GC/GC GC/AT AT/AT ptrend MDM2 T309G TT TG GG ptrend

n

Median (mo)

Log-Rank p

HRa

58 100 41

20.0 16.0 12.0

0.095 0.002

Referent 1.46 1.86

114 69 16

16.0 17.0 18.0

0.732 0.709

Referent 0.97 0.83

44 101 54

17.0 17.0 15.0

0.615 0.914

Referent 0.88 1.05

Progression-Free Survival 95% CI

1.00–2.15 1.15–3.02 0.009

0.69–1.39 0.44–1.57 0.606

0.58–1.34 0.66–1.67 0.769

Median (mo)

Log-Rank p

HRa

7.6 6.9 5.3

0.085 0.004

Referent 1.45 1.99

6.5 8.3 5.7

0.119 0.769

0.79 1.15

6.6 6.5 9.0

0.985 0.140

Referent 1.05 0.70

95% CI

0.98–2.15 1.23–3.22 0.005

0.56–1.13 0.61–2.19 0.574

0.69–1.60 0.44–1.12 0.099

a Adjusted for sex, age, smoking status, performance status, stages, histology, and chemotherapy regimens. NSCLC, non-small cell lung cancer; HR, hazard ratio; CI, confidence interval.

Copyright © 2011 by the International Association for the Study of Lung Cancer

1795

Liu et al.

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

FIGURE 1. Kaplan-Meier curves of (A) overall survival (log-rank p ⫽ 0.009) and (B) progression-free survival (log-rank p ⫽ 0.009) for all 199 non-small cell lung cancer patients with different P53 Arg72Pro genotypes.

with the heterozygous Arg/Pro genotype also had shorter OS (median OS, 16.0 months; 95% CI, 12.8 –19.2 months) compared with carriers of the Arg/Arg genotype, although the difference was marginally significant (log-rank p ⫽ 0.095). Combined analysis showed that patients carrying at least one P53 72Pro allele had significant shorter OS (median OS, 14.0 months; 95% CI, 11.0 –17.0 months) than patients carrying no such an allele (log-rank p ⫽ 0.021). Similarly, significant association was also observed between P53 Arg72Pro polymorphism and PFS (Table 2; Figure 1). The median PFS for the Arg/Arg genotype was 7.6 (95% CI, 5.7–9.5) months, which was significantly longer than that for the Pro/Pro genotype (median PFS, 5.3 months; 95% CI, 4.1– 6.5 months; log-rank p ⫽ 0.004) or that for combined variant genotypes (median PFS, 6.5 months; 95% CI, 5.8 –7.2 months; log-rank p ⫽ 0.020). In Cox proportional hazards model including age, sex, ECOG performance status, smoking status, histological type, disease stage, and chemotherapy regimens as covariates, the prognostic significance of P53 Arg72Pro polymorphism for OS still existed. Compared with the Arg/Arg genotype, the Arg/Pro and Pro/Pro genotypes had significantly higher risk of death, with HRs being 1.46 (95% CI, 1.00 –2.15) and 1.86 (95% CI, 1.15–3.02), respectively (ptrend ⫽ 0.009). Similarly, multivariate Cox proportional hazards model also demonstrated an allele-dose effect for the P53 Arg72Pro polymorphism on PFS (ptrend ⫽ 0.005), with the Pro/Pro genotype having the highest hazard ratio (HR) of 1.99 (95% CI, 1.23– 3.22). Nevertheless, no significant associations were observed between patients’ survival and the P73 G4C14-to-A4T14 or MDM2 T309G genotypes (Table 2).

1796

Association Between Combined Genotypes and Patients’ OS and PFS In combined analysis of P53 Arg72Pro and P73 G4C14-to-A4T14 polymorphisms (Table 3), patients with the P53 Pro/Pro-P73 GC/AT ⫹ AT/AT or P53 Pro/Pro-P73 GC/GC genotypes had significantly shorter OS compared with those carrying the P53 Arg/Arg-P73 GC/AT ⫹ AT/AT genotypes (12.0 versus 24.0 months; log-rank p ⫽ 0.031 and 10.0 versus 24.0 months; log-rank p ⫽ 0.001). Similarly, shorter PFS was also presented in patients carrying both P53 72Pro and P73 4G14C alleles (Table 3; Figure 2). Moreover, an allele-dose effect of combined P53 and P73 genotypes on OS was found (Ptrend ⫽ 0.009), with the P53 Pro/Pro-P73 GC/GC genotype having the highest HR of death (2.47; 95% CI, 1.20 –5.10). Similar results were seen for PFS, showing that the combined genotypes P53 Arg/Pro-P73 GC/GC, P53 Pro/Pro-P73 GC/AT ⫹ AT/AT, or P53 Pro/Pro-P73 GC/GC had poor PFS and increased HR compared with the P53 Arg/Arg-P73 GC/AT ⫹ AT/AT genotype (ptrend ⫽ 0.003). When the P53 Arg72Pro and MDM2 T309G polymorphisms were combined for analysis, shorter OS and PFS were seen in patients with the P53 Pro/Pro-MDM2 GG genotype than in patients with the P53 Arg/Arg-MDM2 TT ⫹ TG genotype (median OS, 10 versus 20 months; log-rank p ⫽ 0.003 and median PFS, 5.5 versus 7.0 months, log-rank p ⫽ 0.037; Figure 3). Multivariate Cox proportional hazard model showed an allele-dose effect of the combined P53 Pro and MDM2 G variants on increase in HR for poor OS and PFS (ptrend ⫽ 0.006 and 0.024). Patients carrying two variant alleles of both P53 72Pro and MDM2 309G (P53 Pro/ProMDM2 GG genotypes) had HRs of 1.74 (95% CI, 1.01–3.01)

Copyright © 2011 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

TABLE 3.

SNPs in P53 Pathway and NSCLC Survival

Association Between Combined P53-P73 or P53-MDM2 Genotypes and NSCLC Survival Overall Survival

Genotypes P53 Arg/Arg Arg/Arg Arg/Pro Arg/Pro Pro/Pro Pro/Pro ptrend P53 Arg/Arg Arg/Arg Arg/Pro Arg/Pro Pro/Pro Pro/Pro ptrend MDM2 TT ⫹ TG GG TT ⫹ TG GG ptrend

n

Median (mo)

Log-Rank p

P73 GC/AT ⫹ AT/AT GC/GC GC/AT ⫹ AT/AT GC/GC GC/AT ⫹ AT/AT GC/GC

24 34 39 61 22 19

24.0 17.0 17.0 16.0 12.0 10.0

0.633 0.141 0.192 0.031 0.001

MDM2 TT ⫹ TG GG TT ⫹ TG GG TT ⫹ TG GG

44 14 76 24 9 32

20.0 17.0 16.0 15.0 13.0 10.0

P73 GC/AT ⫹ AT/AT GC/AT ⫹ AT/AT GC/GC GC/GC

56 29 89 25

18.0 17.0 16.0 9.0

HR

Progression-Free Survival a

95% CI

HRa

Median (mo)

Log-Rank p

95% CI

Referent 1.19 0.61–2.30 1.70 0.93–3.13 1.54 0.88–2.70 1.79 0.91–3.51 2.47 1.20–5.10 0.009

11.2 6.8 7.1 6.9 6.1 4.0

0.226 0.089 0.030 0.011 0.002

Referent 1.77 0.92–3.42 2.01 1.09–3.70 1.87 1.07–3.29 2.35 1.18–4.65 3.20 1.53–6.73 0.003

0.671 0.184 0.082 0.287 0.003

Referent 0.98 0.46–2.05 1.37 0.88–2.14 1.72 0.97–3.04 1.47 0.63–3.42 2.00 1.15–3.46 0.006

7.0 10.0 6.5 8.3 4.0 5.5

0.290 0.232 0.816 0.157 0.037

Referent 0.57 0.27–1.23 1.35 0.85–2.15 1.00 0.57–1.77 1.63 0.70–3.81 1.74 1.01–3.01 0.024

0.297 0.583 0.280

Referent 1.18 0.71–1.95 1.09 0.73–1.64 1.26 0.72–2.22 0.495

6.5 9.0 6.5 7.0

0.441 0.258 0.723

Referent 0.72 0.43–1.21 1.17 0.77–1.77 0.77 0.43–1.35 0.967

a Adjusted for sex, age, smoking status, performance status, stages, histology, and chemotherapy regimens. NSCLC, non-small cell lung cancer; HR, hazard ratio; CI, confidence interval.

FIGURE 2. Kaplan-Meier curves of (A) overall survival (log-rank p ⫽ 0.028) and (B) progression-free survival (log-rank p ⫽ 0.031) by combined P53 Arg72Pro and P73 G4C14-to-A4T14 genotypes.

for cancer progression and 2.00 (95% CI, 1.15–3.46) for death. The analysis for combined P73 G4C14-to-A4T14 and MDM2 T309G genotypes was also performed, but no significant association was found (Table 3).

Finally, we evaluated the combined effect of the three polymorphisms in P53, P73, and MDM2 genes, with the genotypes associated with increased HR of death as unfavorable genotypes including the P53 Pro/Pro, Arg/Pro, P73

Copyright © 2011 by the International Association for the Study of Lung Cancer

1797

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

Liu et al.

FIGURE 3. Kaplan-Meier curves of (A) overall survival (log-rank p ⫽ 0.045) and (B) progression-free survival (log-rank p ⫽ 0.046) by combined P53 Arg72Pro and MDM2 T309G genotypes. TABLE 4.

Combined Effect of Polymorphisms in P53, P73, and MDM2 on NSCLC Survival

No. of Unfavorable Genotypesa 0 (Group 1 (Group 2 (Group 3 (Group ptrend

1) 2) 3) 4)

Overall Survival n

Median (mo)

17 73 91 18

20.0 18.0 14.0 10.0

Progression-Free Survival b

Log-Rank p

HR

0.576 0.115 0.243

Referent 1.16 1.62 1.60

95% CI 0.60–2.24 0.86–3.04 0.72–3.60 0.039

Median (mo) 7.7 6.8 6.6 7.0

Log-Rank p

HRb

0.440 0.113 0.750

Referent 1.35 1.66 1.10

95% CI 0.70–2.30 0.88–3.14 0.50–2.34 0.437

a The P53 Arg/Pro, Pro/Pro, P73 GC/GC, and MDM2 GG genotypes were considered as unfavorable genotypes. Group 1 carried no unfavorable genotypes in any gene, group 2 carried one unfavorable genotype in any one gene, group 3 carried unfavorable genotypes in any two genes, and group 4 carried unfavorable genotypes in all three genes. b Adjusted for sex, age, smoking status, performance status, stages, histology, and chemotherapy regimens. HR, hazard ratio; CI, confidence interval.

GC/GC, and MDM2 GG. On the basis of the number of unfavorable genotypes carried, patients were categorized into four groups: group 1, having no unfavorable genotype; group 2, having unfavorable genotypes in any one gene; group 3, having unfavorable genotypes in any two genes; and group 4 having unfavorable genotypes in all three genes. We found that the median OS is gradually shorter with increasing number of unfavorable genotypes, and the HR for death also increased in an allele-dose-dependent manner (ptrend ⫽ 0.039), although the association between the combined genotypes and OS was not statistically significant (Table 4; Figure 4).

DISCCUSION In this study, we evaluated the prognostic effect of three functional polymorphisms in the P53 pathway in advanced NSCLC patients treated with platinum-based chemo-

1798

therapy. We found that the P53 Arg72Pro polymorphism played substantial role in the prognosis of NSCLC not only in itself but also in combination with P73 G4C14-to-A4T14 and MDM2 T309G polymorphisms. To the best of our knowledge, this is the first study investigating the joint effect of P53, P73, and MDM2 polymorphisms on prognosis of advanced NSCLC patients with platinum-based chemotherapy. Accumulating evidence indicates that inherited variations in P53 pathway components may define patient populations in their abilities to produce apoptosis of cancer cells in response to DNA damage induced by chemotherapeutic agents.16,25 As one of the most frequently studied variations in the P53 pathway, P53 Arg72Pro polymorphism has been shown to affect prognosis of various cancers,26 –30 although the reported results are not all consistent for NSCLC patients.31–33 In this study, we found that patients carrying the

Copyright © 2011 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

SNPs in P53 Pathway and NSCLC Survival

FIGURE 4. Kaplan-Meier curves of (A) overall survival (log-rank p ⫽ 0.112) and (B) progression-free survival (log-rank p ⫽ 0.181) by combined unfavorable genotypes in P53, P73, and MDM2 genes.

P53 72Pro allele had both significantly poor OS and PFS. This result is parallel to the laboratory findings showing that apoptotic rate was higher in cells expressing the 72Arg protein than in cells expressing the 72Pro protein in response to platinum treatment.34 Furthermore, our result is consistent with the recent study on NSCLC, reporting that patients with the P53 72Pro/Pro genotype are more likely to be resistant to platinum-based chemotherapy and display worse survival than patients with the P53 72Arg/Arg genotype.31 Interestingly, the variant 4A14T allele of G4C14-to-A4T14 polymorphism in the P73 gene showed a tendency of association with longer survival in this study. This result is consistent with two previous studies showing prognosis significance of the 4A14T allele for better survival in colorectal cancer and breast cancer.35,36 Although to be elucidated, the underlying mechanism for the association between the P73 G4C14-toA4T14 polymorphism and prognosis of cancer could be the involvement of this polymorphism in P73 translational initiating or alternative splicing.35,37 For the MDM2 T309G polymorphism, we did not observe its association with survival in advanced NSCLC patients, which is consistent with the observation from the study by Chien et al.38 that the T309G polymorphism was not associated with survival in late-stage patients, whereas inconsistent with previous studies reporting significant association of this polymorphism with survival in stage I NSCLC patients.38,39 This discrepancy might be due to distinct disease stage between our study and previous studies. In our study, the modest effect of genetic polymorphism could be overwhelmed by complex somatic mutations that occur in advanced cancer. Nevertheless, the reasons for the inconsistent results between previous studies and ours need to be explored in further studies.

Intriguingly, analysis of the collective effect of P53 Arg72Pro, P73 G4C14-to-A4T14, and MDM2 T309G polymorphisms showed an association of combined genotypes with survival in a genotype-dose effect manner. These results are in line with the functional interactions among these three polymorphisms in P53, P73, and MDM2. Previous studies have shown a cooperation effect between P53 and P73 in platinum-induced apoptosis in vitro in cancer cell lines.37,40 In addition, previous study also showed that the influence of P53 72Arg72Pro polymorphism on response to platinumbased cancer chemotherapy in head and neck cancer patients is through modulation of P73-dependent apoptosis, suggesting a P53-P73 interaction.8 Because MDM2 and P53 are in a feedback loop,41 joint effect between P53 Arg72Pro and MDM2 T309G polymorphisms are biologically plausible. It has been shown that the P53 72Pro variant displays more compacted affinity for the TAFII32 and TAFII70 transcriptional factors compared with the 72Arg variant and thus had higher ability to transactivate MDM2 309G allele.42 Moreover, P73 and MDM2, the two regulators of P53 function, also interact with each other in the P53 pathway. MDM2 can bind to the N-terminal domain of P73 because of the structural similarity of P73 and P53 but not degrade P73,43 whereas P73 can elevate the level of MDM2 by increasing transcription and protein stability of MDM2.44 Thus, the activity of P53 could be altered by cooperation between MDM2 and P73. In view of the interactions among P53, P73, and MDM in apoptosis, one may expect that the functional polymorphisms in the P53, P73, and MDM2 genes have joint effect on response to and outcome of chemotherapy. In conclusion, our results showed a substantial cumulative effect of P53 Arg72Pro, P73 G4C14-to-A4T14, and MDM2

Copyright © 2011 by the International Association for the Study of Lung Cancer

1799

Journal of Thoracic Oncology • Volume 6, Number 11, November 2011

Liu et al.

T309G polymorphisms on OS in advanced NSCLC patients treated with platinum-based chemotherapy, suggesting that these polymorphisms may serve as promising biomarkers for individualized chemotherapy and prognosis of NSCLC patients.

22. 23.

ACKNOWLEDGMENTS Supported by Program for New Century Excellent Talents in University NCET-10-0388 (to X.M.).

24.

REFERENCES

25.

1. NSCLC Meta-Analyses Collaborative Group. Chemotherapy in addition to supportive care improves survival in advanced non-small-cell lung cancer: a systematic review and meta-analysis of individual patient data from 16 randomized controlled trials. J Clin Oncol 2008;26:4617– 4625. 2. Molina JR, Yang P, Cassivi SD, et al. Non-small cell lung cancer: epidemiology, risk factors, treatment, and survivorship. Mayo Clin Proc 2008;83:584 –594. 3. Coate LE, John T, Tsao MS, et al. Molecular predictive and prognostic markers in non-small-cell lung cancer. Lancet Oncol 2009;10:1001– 1010. 4. Ahmad S. Platinum-DNA interactions and subsequent cellular processes controlling sensitivity to anticancer platinum complexes. Chem Biodivers 2010;7:543–566. 5. Hrstka R, Coates PJ, Vojtesek B. Polymorphisms in p53 and the p53 pathway: roles in cancer susceptibility and response to treatment. J Cell Mol Med 2009;13:440 – 453. 6. Dumont P, Leu JI, Della Pietra AC III, et al. The codon 72 polymorphic variants of p53 have markedly different apoptotic potential. Nat Genet 2003;33:357–365. 7. Muller M, Schleithoff ES, Stremmel W, et al. One, two, three—p53, p63, p73 and chemosensitivity. Drug Resist Updat 2006;9:288 –306. 8. Bergamaschi D, Gasco M, Hiller L, et al. p53 polymorphism influences response in cancer chemotherapy via modulation of p73-dependent apoptosis. Cancer Cell 2003;3:387– 402. 9. Leong CO, Vidnovic N, DeYoung MP, et al. The p63/p73 network mediates chemosensitivity to cisplatin in a biologically defined subset of primary breast cancers. J Clin Invest 2007;117:1370 –1380. 10. Nomoto S, Haruki N, Kondo M, et al. Search for mutations and examination of allelic expression imbalance of the p73 gene at 1p36.33 in human lung cancers. Cancer Res 1998;58:1380 –1383. 11. Kaghad M, Bonnet H, Yang A, et al. Monoallelically expressed gene related to p53 at 1p36, a region frequently deleted in neuroblastoma and other human cancers. Cell 1997;90:809 – 819. 12. Li G, Wang LE, Chamberlain RM, et al. p73 G4C14-to-A4T14 polymorphism and risk of lung cancer. Cancer Res 2004;64:6863– 6866. 13. Jun HJ, Park SH, Lee WK, et al. Combined effects of p73 and MDM2 polymorphisms on the risk of lung cancer. Mol Carcinog 2007;46:100 – 105. 14. Schabath MB, Wu X, Wei Q, et al. Combined effects of the p53 and p73 polymorphisms on lung cancer risk. Cancer Epidemiol Biomarkers Prev 2006;15:158 –161. 15. Grochola LF, Zeron-Medina J, Meriaux S, et al. Single-nucleotide polymorphisms in the p53 signaling pathway. Cold Spring Harb Perspect Biol 2010;2:a001032. 16. Vazquez A, Bond EE, Levine AJ, et al. The genetics of the p53 pathway, apoptosis and cancer therapy. Nat Rev Drug Discov 2008;7:979 –987. 17. Rayburn E, Zhang R, He J, et al. MDM2 and human malignancies: expression, clinical pathology, prognostic markers, and implications for chemotherapy. Curr Cancer Drug Targets 2005;5:27– 41. 18. Strazisar M, Mlakar V, Glavac D. The expression of COX-2, hTERT, MDM2, LATS2 and S100A2 in different types of non-small cell lung cancer (NSCLC). Cell Mol Biol Lett 2009;14:442– 456. 19. Heist RS, Zhou W, Chirieac LR, et al. MDM2 polymorphism, survival, and histology in early-stage non-small-cell lung cancer. J Clin Oncol 2007;25:2243–2247. 20. Bond GL, Hu W, Bond EE, et al. A single nucleotide polymorphism in the MDM2 promoter attenuates the p53 tumor suppressor pathway and accelerates tumor formation in humans. Cell 2004;119:591– 602. 21. Bond GL, Hu W, Levine A. A single nucleotide polymorphism in the

1800

26.

27. 28. 29.

30. 31. 32. 33. 34. 35. 36. 37. 38.

39. 40. 41. 42. 43. 44.

MDM2 gene: from a molecular and cellular explanation to clinical effect. Cancer Res 2005;65:5481–5484. Kulkarni DA, Vazquez A, Haffty BG, et al. A polymorphic variant in human MDM4 associates with accelerated age of onset of estrogen receptor negative breast cancer. Carcinogenesis 2009;30:1910 –1915. Zhang X, Miao X, Guo Y, et al. Genetic polymorphisms in cell cycle regulatory genes MDM2 and TP53 are associated with susceptibility to lung cancer. Hum Mutat 2006;27:110 –117. Yuan P, Miao XP, Zhang XM, et al. [Association of the responsiveness of advanced non-small cell lung cancer to platinum-based chemotherapy with p53 and p73 polymorphisms]. Zhonghua Zhong Liu Za Zhi 2006; 28:107–110. Viktorsson K, De Petris L, Lewensohn R. The role of p53 in treatment responses of lung cancer. Biochem Biophys Res Commun 2005;331: 868 – 880. Huang ZH, Hua D, Du X. Polymorphisms in p53, GSTP1 and XRCC1 predict relapse and survival of gastric cancer patients treated with oxaliplatin-based adjuvant chemotherapy. Cancer Chemother Pharmacol 2009;64:1001–1007. Toffoli G, Biason P, Russo A, et al. Effect of TP53 Arg72Pro and MDM2 SNP309 polymorphisms on the risk of high-grade osteosarcoma development and survival. Clin Cancer Res 2009;15:3550 –3556. Toyama T, Zhang Z, Nishio M, et al. Association of TP53 codon 72 polymorphism and the outcome of adjuvant therapy in breast cancer patients. Breast Cancer Res 2007;9:R34. Chen J, Li D, Killary AM, et al. Polymorphisms of p16, p27, p73, and MDM2 modulate response and survival of pancreatic cancer patients treated with preoperative chemoradiation. Ann Surg Oncol 2009;16: 431– 439. Yamasaki M, Miyata H, Fujiwara Y, et al. p53 genotype predicts response to chemotherapy in patients with squamous cell carcinoma of the esophagus. Ann Surg Oncol 2010;17:634 – 642. Han JY, Lee GK, Jang DH, et al. Association of p53 codon 72 polymorphism and MDM2 SNP309 with clinical outcome of advanced nonsmall cell lung cancer. Cancer 2008;113:799 – 807. Chua HW, Ng D, Choo S, et al. Effect of MDM2 SNP309 and p53 codon 72 polymorphisms on lung cancer risk and survival among non-smoking Chinese women in Singapore. BMC Cancer 2010;10:88. Matakidou A, El Galta R, Webb EL, et al. Lack of evidence that p53 Arg72Pro influences lung cancer prognosis: an analysis of survival in 619 female patients. Lung Cancer 2007;57:207–212. Sullivan A, Syed N, Gasco M, et al. Polymorphism in wild-type p53 modulates response to chemotherapy in vitro and in vivo. Oncogene 2004;23:3328 –3337. Pfeifer D, Arbman G, Sun XF. Polymorphism of the p73 gene in relation to colorectal cancer risk and survival. Carcinogenesis 2005;26:103–107. Li H, Yao L, Ouyang T, et al. Association of p73 G4C14-to-A4T14 (GC/AT) polymorphism with breast cancer survival. Carcinogenesis 2007;28:372–377. Tokuchi Y, Hashimoto T, Kobayashi Y, et al. The expression of p73 is increased in lung cancer, independent of p53 gene alteration. Br J Cancer 1999;80:1623–1629. Chien WP, Wong RH, Cheng YW, et al. Associations of MDM2 SNP309, transcriptional activity, mRNA expression, and survival in stage I non-small-cell lung cancer patients with wild-type p53 tumors. Ann Surg Oncol 2010;17:1194 –1202. Dong J, Ren B, Hu Z, et al. MDM2 SNP309 contributes to non-small cell lung cancer survival in Chinese. Mol Carcinog 2011;50:433– 438. Righetti SC, Perego P, Carenini N, et al. Cooperation between p53 and p73 in cisplatin-induced apoptosis in ovarian carcinoma cells. Cancer Lett 2008;263:140 –144. Levine AJ, Oren M. The first 30 years of p53: growing ever more complex. Nat Rev Cancer 2009;9:749 –758. Yang M, Guo Y, Zhang X, et al. Interaction of P53 Arg72Pro and MDM2 T309G polymorphisms and their associations with risk of gastric cardia cancer. Carcinogenesis 2007;28:1996 –2001. Lohrum MA, Vousden KH. Regulation and function of the p53-related proteins: same family, different rules. Trends Cell Biol 2000;10:197– 202. Wang XQ, Ongkeko WM, Lau AW, et al. A possible role of p73 on the modulation of p53 level through MDM2. Cancer Res 2001;61:1598 – 1603.

Copyright © 2011 by the International Association for the Study of Lung Cancer