Comparison of the prevalence of KRAS-LCS 6 polymorphism ( rs 61764370 ) within different tumour types ( colorectal , breast , non-small cell lung cancer and brain tumours ) . A study of the Czech population

Aims. A germline SNP (rs61764370) is located in a let-7 complementary site (LCS6) in the 3'UTR of KRAS oncogene, and it was found to alter the binding capability of the mature let-7 microRNA to the KRAS mRNA. The aim of the study was to evaluate the frequency of the KRAS-LCS6 variant allele in different cancer types that included patients with colorectal cancer (CRC), breast cancer (BC), non-small cell lung cancer (NSCLC) and brain tumour patient subgroups from the Czech Republic. The occurrence of this genetic variant was correlated with the presence of selected somatic mutations representing predictive biomarkers in the respective tumours. Methods. DNA of tumour tissues was isolated from 428 colorectal cancer samples, 311 non-small cell lung cancer samples, 195 breast cancer samples and 151 samples with brain tumour. Analysis of SNP (rs61764370) was performed by the PCR+RFLP method and direct sequencing. KRAS, BRAF and EGFR mutation status was assessed using real-time PCR. The status of the HER2 gene was assessed using the FISH method. Results. The KRAS-LCS6 TG genotype has been detected in 16.4% (32/195) of breast cancer cases (in HER2 positive breast cancer 3.3%, in HER2 negative breast cancer 20.1%), in 12.4% (53/428) of CRC cases (KRAS/BRAF wild type CRC in 10.6%, KRAS mutant CRC in 10.1%, BRAF V600E mutant CRC in 18.5%), in 13.2% (41/311) of NSCLC samples, (EGFR mutant NSCLC patients in 8%, EGFR wild type NSCLC in 12.9%), and 17.9% (27/151) of brain tumour cases. The KRAS-LCS6 TG genotype was not significantly different across the studied tumours. In our study, the GG genotype has not been found among the cancer samples. Conclusions. Based on the findings, it is concluded that the occurrence of the KRAS-LCS6 TG genotype was statistically significantly different in association with status of the HER2 gene in breast cancer. Furthermore, significant association between the mutation status of analysed somatic variants in genes of the EGFR signalling pathway (KRAS, BRAF, EGFR) and the KRAS-LCS6 genotype in colorectal cancer and NSCLC has not been established.


INTRODUCTION
Germline SNP (rs61764370) has been recently extensively studied in relation to several types of cancer, especially in connection with associated risk, prognosis, and/or treatment response/resistance to therapy [1][2][3][4][5][6][7][8][9][10][11] .The let-7 family of miRNAs has been found to play an important role in tumorigenesis by regulating the expression of multiple oncogenes, including KRAS.The KRAS (Kirsten rat sarcoma viral oncogene homolog) gene is one of the most frequently mutated oncogenes in human cancer 12 , with almost a quarter of different tumour types showing altered functions of the KRAS gene.It encodes a protein, which plays a major role in signal transduction and cell proliferation, is a crucial target of let-7 microRNAs (miR-NAs) and is downregulated by these miRNAs (ref. 5,13,14).Let-7 miRNAs downregulate the KRAS protein expres-sion by binding to a specific site in the 3'-untranslated region (3'-UTR) of the KRASmRNA (ref. 13).Altered expression of the let-7 family of microRNAs is implicated in many human cancers [15][16][17] .The SNP (rs61764370), also known as KRAS-LCS6, is located in the 3'-untranslated region of the KRAS oncogene, consists in a T-to-G base change, and it was found to alter the binding capability of the mature let-7 to KRASmRNA, which resulted in both increased KRAS expression and reduced let-7 levels 1,4 .Another group of authors has reported that cancer associated non-coding SNP (rs61764370) variant exerts biological effect not through transcriptional modulation of KRAS but rather by tuning the expression of microRNA let-7 (ref. 18).
On the basic of the current evidence and the importance of both miRNA let-7 and KRAS oncogenes in human solid tumours, the purpose of our study was to investigate the frequency of the KRAS-LCS6 variant allele in different cancer types that included the following subgroups: colorectal cancer (CRC), breast cancer (BC), non-small cell lung cancer (NSCLC) and brain tumours.Furthermore, the occurrence of this genetic variant was correlated with the presence of selected somatic mutations representing predictive biomarkers in the respective tumour types in Czech population.

MATERIAL AND METHODS
A retrospective analysis was used to test 1085 samples of DNA of tumour tissue from patients in whom KRAS, BRAF, HER2 and EGFR predictive markers (gene status) were analysed in our laboratory between 2008 and 2014 to determine their eligibility for targeted biological treatment.Furthermore, prognostic markers were analysed to determine brain tumour prognosis.The material for the tumour DNA extraction was collected from 428 samples with a diagnosis of metastatic colorectal cancer confirmed histologically, stage III-IV: 277 male subjects (64.7%) aged 35-91 years / 151 female subjects (35.3%) aged 31-94 years; the median age of the CRC cohort was 65 years.The NSCLC group consisted of 311 samples of tumour DNA: 202 male subjects (65.0%) aged 34-99 years / 109 female subjects (35.0 %) aged 32-89 years; the median age of the NSCLC cohort was 67 years.A total of 195 samples of the BC included 121 samples of tumour DNA of invasive ductal carcinoma (62.0%), 15 samples of invasive lobular carcinoma (7.7%), 59 other BC samples (30.3%): 2 male subjects (1.0%) aged 51-77 years / 193 female subjects (99.0%) aged 25-87 years; the median age of the BC cohort was 60 years.A total of 151 samples of tumour DNA in the brain tumour cohort included 117 samples of gliomas, grade I-IV (77.5%), 23 samples of meningiomas (15.2%), 11 samples of other brain tumours (7.3%): 75 male subjects (49.7%) aged 10-77 years / 76 female subjects (50.3%) aged 1-81 years; the median age of the brain tumour cohort was 56.5 years.
The DNA of tumour tissues was isolated from different types of materials: fresh-frozen tissue, formalin-fixed paraffin-embedded tissue and cytology specimens.The commercial kits cobas ® DNA Sample Preparation Kit or MagNA Pure Compact Nucleic Acid Isolation Kit (Roche Diagnostics Corporation, GmbH, Mannheim, Germany) were used to isolate the DNA.The DNA quality and quantity was assessed using the NanoPhotometer™ IMPLEN (Implen, GmbH, München, Germany).
Analysis of SNP (rs61764370) (KRAS-LCS6) was performed by polymerase chain reaction (PCR) and enzymatic digestion.Twenty microliters of PCR reaction mixture contained: PCR water (Thermo Scientific, Vilnius, Lithuania), PCR Master Mix (2x) (Thermo Scientific, Vilnius, Lithuania), specific primers (0.2 pmol of sense and antisense primers) and DNA.The primers used were as follows: sense, 5′-CCTGAGTAGCTGGGATTACA-3′ and antisense, 5′-GGATACCATATACCCAGTGCCTT-3′ for the KRAS-LCS6 polymorphism (amplify a fragment of 232 bp).The PCR conditions were as follows: the initial denaturation was at 95 ˚C for 5 min, followed by 35 cycles of 95 °C for 20 s, 60 °C for 15 s and 72 °C for 10 s; the final amplification was at 72 °C for 5 min, PCR products (232 bp) were checked by electrophoresis on 3% agarose gel.
Some positive results were confirmed by direct sequencing.The sequencing reaction was performed using the BigDye ® Terminator v3.1 Cycle Sequencing Kit (Life Technologies, Foster City, CA, USA). 5 μL of PCR products were cleaned with Exonuclease I and FastAP Thermosensitive Alkaline Phosphatase (Thermo Scientific, Vilnius, Lithuania) by incubating at 37 °C for 15 min, followed by 15 min at 80 °C, and held at 4 °C.The reaction mix consisted of 3 μL BigDye ® Terminator v3.1, 0.5 μL of Big Dye ® Terminator Sequencing buffer (5x), 2.5 μL of primer at a final concentration 1.25 pmol/μL and 4 μL of cleaned template in a total volume of 10 μL.The reactions were performed according to the following protocol: 96 °C for 1 min; 25 cycles of 96 °C for 10 s, 50 °C for 5 s, 60 °C for 3 min.
The products of the sequencing reaction were cleaned using the BigDye ® XTerminator Purification Kit (Life Technologies, Foster City, CA, USA) in accordance with the manufacturer's instructions.After purification, samples were run on Applied Biosystems ® 3130 Genetic Analyzer and analysed using Sequencing Analysis Software version 5.2 (both Life Technologies, Foster City, CA, USA).
The statistical analyses were performed using the chisquared test and Fisher's exact test, the level of significance 5% (when at least one expected value is < 5, Fisher exact tests are recommended rather than chi square).The statistical evaluation was performed using the Stata software, version 13.

RESULTS
We assessed the frequency of the KRAS-LCS6 variant allele in 1085 samples of DNA of different cancer types, including CRC, BC, NSCLC and brain tumour.Using the chi-squared test, there was found a significant difference in the distribution of the G-allele (the TG genotype) across individual groups of BC samples divided into two groups on the basis of the HER2 gene status, (P = 0.026).Table 1.
The 95% confidence interval (CI) for occurrence of the TG genotype was determined (Fig. 1).
Further statistical analysis of the KRAS-LCS6 (rs61764370) polymorphism in relation to the mutation status of the KRAS and BRAF genes was performed in patients with colorectal cancer (Table 3).Using he chisquared test, no significant statistical difference was found in the distribution of the G-allele (the TG/GG genotype) across individual groups of CRC samples divided into three subgroups on the basis of the mutation status of the KRAS and BRAF genes, (P = 0.081).
A statistical analysis of the KRAS-LCS6 (rs61764370) polymorphism in relation to the mutation status of the EGFR gene was performed in patients with non-small cell lung cancer (Table 4).
The Fisher's exact test indicated no significant statistical difference in the distribution of the G-allele (the TG  genotype) across individual groups of NSCLC samples divided into two groups on the basis of the mutation status of the EGFR gene, (P = 0.752).

DISCUSSION
In selected groups of BC depending on the HER2 gene status, our results show a significant difference in the distribution of G allele (TG genotype) of the KRAS-LCS6 polymorphism.The KRAS-LCS6 TG genotype SNP Fisher's exact test, P = 0.752 (rs61764370) was found to be 20.1% in the HER2 negative subgroup compared to 3.3% in the HER2 positive breast subgroup.These results can implicate that KRAS-LCS6 TG genotype SNP (rs61764370) may lead to an alteration in the HER2 gene expression profile.These results are inconsistent with the findings published by Cerne 10 , who reported no significant association of the TG/GG genotype in group of BC selected depending on the HER2 status (the authors analysed 506 cases of BC/ 63 cases were HER2 positive).In addition, they found a positive association of SNP (rs61764370) only in HER2 positive tumours, high histopathological grade and both associations were detected only in the HRT users.
In the non-selected group of breast cancer samples, the frequency of the TG genotype of SNP (rs61764370) was shown to be 16.4%, and the results are in line with Cerne 10 , who found a frequency of 17.2% for the TG genotype in a group of sporadic breast cancers.Furthermore, there is no significant difference in the frequency of the G-allele (TG/GG genotype) KRAS-LCS6 polymorphisms either across cancer groups or between cancer and general population reported in study by Chin LJ 1.Similarly, Paranjape 6 found no difference in the overall frequency distributions of the KRAS-variant genotype in cases of breast cancer and controls.In this study, a significant association of the KRAS-LCS6 was only found in a group with breast cancer in premenopausal patients with oestrogen/ progesterone-negative tumours.
In the group of colorectal cancers, a total of 428 histologically confirmed samples were examined and the frequency of the TG genotype SNP (rs61764370) was found to be 12.4%.No statistically significant difference in the distribution of the TG genotype, 10.6% KRAS/BRAF wildtype, 10.1% KRAS mutant, 18.5% BRAF mutant was found for individual groups of the CRC samples divided into three subgroups depending on the mutation status of the KRAS and BRAF genes.Despite an evident tendency of the given allele to be present nearly twice as often in the BRAF mutant CRC patients compared to KRAS/BRAF wild-type and KRAS mutant patients, the chi-squared test indicated no statistically significant difference.The identified frequency of the KRAS-LCS6 G-allele (TG genotype) in the KRAS/BRAF wild-type group is in line with the data published by Páez 19 .A different result was found in the case of patient stratification to KRAS wild-type group and KRAS mutant group by Graziano 2 .These authors found significantly greater frequency of the KRAS-LCS6 G allele in the KRAS mutant group.Previous studies have shown that SNP is a negative prognostic marker in late-stage co-lon cancer 2 , a positive prognostic indicator in early-stage colon cancer 4 and a predictor of response to treatment in colon cancer 2,5,9 .However, the systematic review and meta-analysis of Langevin and Christensen 20 indicated no clear evidence to support an association between the KRAS-LCS6 genotype and an overall or progression-free survival among colorectal cancer patients; even after analyses of subgroups stratified by stage and anti-EGFR treatment status.In our analysis, no such result was found and no significant association between the KRAS and BRAF mutation status and the KRAS-LCS6 genotype for the colorectal carcinoma group was established.This is in line with the results published in a Norwegian study 21 and in Sha 22 .
In the group of non-small cell lung cancer, a total of 311 histologically confirmed samples were examined.The variant TG genotype was observed in 13.2%.Our results are in line with the results published by Nelson HH 23 , whose group of adenocarcinoma patients exhibited a frequency of variant alleles of 14.7%.No statistically significant difference in the distribution of the G allele (TG genotype) was found between individual groups of the NSCLC samples divided into two subgroups depending of the mutation status of the EGFR gene.
In a group of patients with brain tumours, constituting a heterogeneous group of 151 histologically confirmed samples, the TG genotype of SNP (rs61764370) was detected in 17.9%.Even in this group, no significant difference in comparison with other types of tumour was found.In the literature available to date, no study has been published so far making a comparison with a similar group of brain tumours possible.
The occurrence of this SNP varies geographically across the global population.This follows from the ALFRED (The ALlele FREquency Database) expected heterozygosities (rs61764370) for persons of European descent 24 are 0.05-0.25.Chin LJ 1 reported that in the general population the frequency of the G-allele of KRAS-LCS6 is 5.8% and about 7.6% in populations of European descent, based on the genotyping of 2.433 samples (representing 46 geographic populations).

CONCLUSION
The determination of risk factors for cancer provides improved tools for its diagnosis and treatment.Because of the importance of both miRNA let-7 and KRAS oncogenes in human solid tumours and their potential use as a prognostic/predictive marker, the frequency of the G-allele of the SNP (rs61764370) KRAS-LCS6 in different types of solid tumours was evaluated.Based on the findings, it is concluded that there is a significant difference in the KRAS-LCS6 TG genotype associated with gene HER2 status in breast cancer.This result can implicate that the KRAS-LCS6 G-allele (TG/GG genotype) may lead to alteration in HER2 gene expression profile (could be a potential genetic marker of a developing HER2 negative breast cancer).A further analysis of BC patients is necessary to confirm these results.
However, no significant difference in the occurrence of the KRAS-LCS6 G-allele (TG/GG genotype) across the studied tumour types (CRC, NSCLC, BC and brain tumour) was found.

Table 1 .
Distribution of the KRAS-LCS6 (rs61764370) genotype in BC patients depending on the HER2 gene status.

Table 3 .
Distribution of the KRAS-LCS6 (rs61764370) genotype in CRC patients depending on the mutation status of the KRAS and BRAF genes.

Table 4 .
Distribution of the KRAS-LCS6 (rs61764370) genotype in NSCLC patients depending on the mutation status of the EGFR gene.