Prediction of polymorphisms in DNA repair genes on the prognosis of bone tumors with methotrexate , intravenous adriamycin and intra-arterial cisplatin chemotherapy

We investigate the relationship of SNPs in XRCC1, ERCC1 and ERCC2 to methotrexate intravenous adriamycin and intra-arterial cisplatin chemotherapy response and survival in bone cancer patients. 281 consecutive patients diagnosed with bone tumor between January 2003 to January 2005, and they were followed up until the end of January 2010. Polymorphisms in XRCC1Arg280His, XRCC1Arg399Gln, ERCC1 Asn118Asn, ERCC1 Gln504Lys, ERCC2 Asp312Asn and ERCC2 Lys751Gln were detected based upon Sequenom MassARRAY platform. All 281 patients were followed up until the end of January 2010. The median follow-up time was about 43.9 months. Our study showed a longer survival rate in XRCC1 399 Arg/Arg genotype than Gln/ Gln, and with a significant increased death risk (HR=0.46, 95%CI=0.310.96). For ERCC1 Asn118Asn, the variant genotype T/T was strongly significantly associated with a higher event free survival when compared with the wide-type C/C, with the adjusted OR (95% CI) of 0.33(0.15-0.97). ERCC2 751 A/A genotype showed increased the event free survival of osteosarcoma (HR=0.45; 95%CI=0.15-0.91), and T allele genotype of ERCC2 Lys751Gln was significantly associated with low risk of death from osteosarcoma (HR=0.53, 95%CI= 0.28-0.95). This study is the first time to report the ERCC1 and ERCC2 gene polymorphisms might well be useful as a surrogate marker of clinical outcome in bone cancer.


INTRODUCTION
Cancer that originates in the bone, termed primary bone cancer, is rare and typically occurs during the children and adolescent growth spurt, with a second smaller peak in the elderly.Osteosarcoma is one of the most frequent bone tumors, with a 3 -5/10 5 population per year worldwide, but this tumor accounted for about 6% of all cancer diagnosed under the age of 20 years (Le et al., 2007).*Corresponding author.E-mail: yubin665544@163.com.
# These authors contributed equally to this work.
The 5 years survival of osteosarcoma is below 20% after surgery, but the chemotherapy after surgery has dramatically improved the survival from 55 to 70% after years (Le et al., 2007).The methotrexate, adriamycin and cisplatin form the backbone of most standard treatment protocols as a triple-drug regimen (MAP).However, patients who have similar clinical characteristics usually showed variable response rate to chemotherapy, with inferior 5 years survival of 40 to 70% (Le et al., 2007).This implies that the therapeutic efficacy has a remarkable inter-individual variability.
Since DNA kinking is the major feature of oxaliplatin-DNA adducts that block DNA replication and lead to cancer cell death (Ishibashi et al., 2011), it is recognized and repaired by the nucleotide excision repair (NER) pathways.It is reported that the inter-individual difference in the NER capacity may influence the efficacy of cisplatin-based chemotherapy and clinical outcomes of the cancer patients.Single nucleotide polymorphisms (SNPs) in the excision repair cross-complementation group 1 (ERCC1) and ERCC2 genes have been found to be associated with the platinum response in different clinical studies.In particular, the polymorphisms of ERCC1 and ERCC2 were reported to be associated with increased survival and better prognosis in several cancer, such as non-small-cell lung cancer, colorectal cancer, gastric cancer, nasopharyngeal cancer, bladder cancer, and breast cancer (Ishibashi et al., 2011;Sun et al., 2011;Cao et al., 2011;Rouissi et al., 2011;Baek et al., 2006;Rajaraman et al., 2008).
The DNA repair gene XRCC1 gene codes for a protein involved in the repair of single-strand breaks (SSB) and in base excision repair (BER) of damaged bases caused by endogenous and exogenous oxidants.Three polymorphisms occurring at conserved sequences in the XRCC1 gene were reported by Shen et al. (1998).These coding polymorphisms, resulting in amino acid substitutions, were detected at codons 194 (Arg-Trp), 280 (Arg-His) and 399 (Arg-Gln).The 399 polymorphism have been associated with a number of cancers, although results have been inconsistent (Mort et al., 2003;Yeh et al., 2005).
Few studies clarified the association between SNPs in ERCC1 and ERCC2 and bone tumor.Only one study conducted in Spain indicated the prediction role of ERCC1 and ERCC2 on the response to cisplain chemotherapy in osteosarcoma patients (Caronia et al., 2009).In the current study, we investigate the relationship of SNPs in ERCC1 and ERCC2 to cisplain response and survival in osteosarcoma patients.

METHODOLOGY Patients
267 consecutive patients diagnosed with osteosarcoma between January, 2003 to January, 2005 from the Second Affiliated Hospital of Inner Mongolia Medical College and Nanfang Hospital of Southern Medical University were enrolled in our study.All the patients were asked to provide blood samples, and informed consent was obtained from the patients or their relatives.Ethnical approval of the study was granted by the Ethics Committee of the Inner Mongolia Medical College.
All the patients received chemotherapy, including intravenous methotrexate (four courses of up to 14 g -2 per day for 1 day), intravenous adriamycin (three courses at 25 to 30 mg -2 per day for 3 days), and intra-arterial cisplatin (three courses at 35 mg -2 per day for 3 days).After surgery, the adjuvant chemotherapy included methotrexate and alternate cylcles of intravenous cisplatin or adriamycin up to 48 weeks of treatment.
On the day of participating, all the patients were asked to provide their next of kins' telephone numbers and mailing addresses to enable our follow-up, and all the patients were followed up every 1 Zhao et al. 2957 month until death.

Genotyping
DNA was extracted from the buffy-coat fractions with TIANamp blood DNA kit (Tiangen Biotech, Beijing, China).Genotyping four SNPs (ERCC1 Asn118Asn, ERCC1 Gln504Lys, ERCC2 Asp312Asn and ERCC2 Lys751Gln) was performed in a 384-well plate format on the Sequenom MassARRAY platform (Sequenom, San Diego, USA).Primers for polymerase chain reaction (PCR) amplification and single base extension (SBE) assays were designed by Sequenom Assay Design 3.1 software (Sequenom, San Diego, CA, USA) according to the manufacturer's instructions.
For quality control, a 5% random sample of cases was genotyped twice by different researchers.The reproducibility was 100%.Amplification was carried out in a total volume of 25 ml containing 0.25 mM of each primers, 0.02 mM dNTPs and 1 mM MgCl2, 1.25U Taq polymerase and 5_PCR buffer.All PCR samples were divided into two tubes, one for the wild probe and another for the mutation probe.The PCR program initiated with a 1 min denaturation at 958°C.The DNA was amplified by one cycle of 958°C for 5 s and 50 cycles of 928°C for 40 s, followed by elongation of 608°C for 40 s.In addition, to ensure the accuracy of this method, 20 randomly selected DNA samples were subjected to PCR-based direct DNA sequencing.

Statistical analysis
Stata 8.0 (StataCorp, College Station, USA) was used to perform statistical analyses.Continuous variables were expressed as mean ± standard deviation (SD), while categorical variables were shown as frequencies and percentages.SNPs in ERCC1 and ERCC2 were assessed in terms of overall survival and event free survival by using regression analysis.The main endpoint of the study was overall survival, which was calculated from the start of treatment to the date of last follow-up or death.Event-free survival was considered from tumor diagnosis to the first of disease recurrence, development of lung or bone metastasis or death.Survival curves were plotted by using the Kaplan and Meier method and compared with the log-rank test.The association between each genotype and survival was estimated by hazard ratios and their 95% confidence intervals (CI) by multivariate Cox's regression models.A P-value <0.05 was considered statistically significant.

RESULTS
All 281 patients were followed up until the end of January 2010.The median follow-up time was 43.9 about months.There was 6 patients loss to follow up due to immigrant to other cities, and a total of 126 patients were died during the 5 years follow-up.The demographic and clinic characteristics of included patients are shown in Table 1.The mean ages of patients were 14.7±5.3(ranged from 5.5 to 38).156 patients showed good response to chemotherapy treatment (55.5%).61(21.7%)patients showed metastasis at diagnosis, and 80 (28.5%) developed metastasis during the follow-up.Our study showed the poor response to treatment and metastasis at diagnosis was associated with higher risk of death, with HR (95% CI) of 1.93 (1.34-5.14)and 2.45(1.20-6.06),respectively (data not shown).A total of 126 patients were died during the follow-up period.The associations between the SNPs and the risk of death from osteosarcoma were studies by using multivariate Cox's regression analysis (Table 2).Individuals with XRCC1 399Gln/Gln genotype were likely to enjoy a longer event free survival of osteosarcoma, with HR (95% CI) of 0.46 (0.31-0.96) (Figure 1).For ERCC1 Asn118Asn, the variant genotype T/T was strongly significantly associated with a higher event free survival when compared with the wide-type C/C, with the adjusted HR (95% CI) of 0.33 (0.15-0.97) (Figure 2).Furthermore, ERCC2 751 A/A genotype showed increased the event free survival of osteosarcoma (HR=0.45;95%CI=0.15-0.91),and T allele genotype of ERCC2 Lys751Gln was significantly associated with low risk of disease recurrence, development of lung or bone metastasis or death (HR=0.53,95%CI= 0.28-0.95)(Figure 3).However, polymorphisms in XRCC1 Arg194Trp, ERCC1 Gln504Lys and ERCC2 Asp312Asn did not have role in the prognosis of osteosarcoma.

DISCUSSION
The drug resistance is a major problem in most solid tumors.Identification of predictive markers for response to chemotherapy is most warranted, since a subgroup of the patients could not benefit from chemotherapy.Pretreatment identification of the predictive gene polymorphism would be very helpful in further treatment.Our study firstly published the associations of ERCC1 and ERCC2 polymorphisms and the prognosis of primary malignant osteosarcoma in Chinese populations, and implied ERCC1 118T/T and ERCC2 A/A are related to prognosis of cancer.The XRCC1 gene encodes the XRCC1 protein, which complexes with three other DNA repair enzymes involved in the BER pathways, including DNA ligase III, DNA polymerase and poly (ADP-ribose) polymerase PARP (Shen et al., 1998).The codon 399 polymorphism resides on the COOH-terminal side of the PARP interacting domain, within the BRCT1 domain, that are thought to mediate several protein-protein interactions (Masson et al., 1998).Amino acid substitutions in the BRCT domain and in the DNA polymerase β interacting domain in hamster is reported to disrupt the functionality of XRCC1 (Shen et al., 1998).The mutations of XRCC1 polymorphisms may increase the risk of cancers by impairing the interaction of XRCC1 with other enzymatic proteins and consequently altering DNA repair activity (Basso et al., 2007;Tudek, 2007), and resulted in carcinogenesis development.Our study found significant an odds ratio of codon 399 Gln allele for osteosarcoma, which strongly implicates that these polymorphisms may alter the normal protein function by encoding for a twisted protein, resulting in altered affinity to its interactive proteins suggesting an association with a deficiency in DNA repair capacity.ERCC1 and ERCC2 are potentially relevant to cancer because of their involvement in the process of NER (Goode et al., 2002).The NER pathway acts predominately on bulky DNA lesions and ultraviolet (UV) damage, and polymorphisms in ERCC1 and ERCC2 may change the function of NER in repairing DNA damage by chemotherapy.It implied that the High ERCC1 and ERCC2 levels are related to increased removal of cisplatin-induced   DNA adducts and cisplatin resistance.Previous experimental and epidemiological studies showed that the ERCC1 118C allele and ERCC2 751G allele are associated with higher mRNA levels and DNA singlestrand break repair than ERCC1 118T allele and ERCC2 751A allele genotypes, and thus, to induce the resistance to cisplatin-based chemotherapy which used to damage the DNA of cancer cells, and the polymorphisms in ERCC1 118T allele and ERCC2 751A allele are associated with decreased risk of death from several cancers, such as non-small-cell lung cancer, gastric cancer, nasopharyngeal cancer, and breast cancer (Sun et al., 2011;Cao et al., 2011;Baek et al., 2006;Rajaraman et al., 2008).Only one previous study reported that the ERCC2 751G allele are associated with response to cisplatin chemotherapy and shorter event-free survival in osteosarcoma patients (Caronia et al., 2009).The results of our study are in line with previous studies, and indicated that ERCC1 118T/T and ERCC2 751A/A are associated with better event free survival in osteosarcoma patients.
There are two limitations in our study.First, because of the rarity of osteosarcoma, we only had a limited number of cases, which would decrease the study power to find the difference.Second, we collected the cases from one hospital visitors, which may be a threat to the validity of the results, and there is a strong need of analysis on the two gene polymorphisms on various populations.Further large sample studies are warranted in various populations.
Overall, as the first study to investigate the association of four SNPs, ERCC1 Asn118Asn, ERCC1 Gln504Lys, ERCC2 Asp312Asn and ERCC2 Lys751Gln, with risk of death from osteosarcoma in Chinese population, and out study found the ERCC1 118T/T and ERCC2 A/A are related to prognosis of osteosarcoma.Further studies in Chinese populations with large sample size are still warranted.

Figure 3 .
Figure 3. Kaplan-Meier analyses of polymorphisms in ERCC2 Lys751Gln on the even free survival of osteosarcoma.

Table 1 .
Characteristics of patients with osteosarcoma.