OL Oncology Letters 1792-1074 1792-1082 D.A. Spandidos 10.3892/ol.2017.5840 OL-0-0-5840 Articles Association of FGFR2 rs2981582, SIRT1 rs12778366, STAT3 rs744166 gene polymorphisms with pituitary adenoma GlebauskieneBrigita 1 VilkeviciuteAlvita 2 LiutkevicieneRasa 1 2 JakstieneSilvija 3 KriauciunieneLoresa 1 2 ZemaitieneReda 1 ZaliunieneDalia 1 Department of Ophthalmology, Medical Academy, Lithuanian University of Health Sciences, LT-50161 Kaunas, Lithuania Neuroscience Institute, Medical Academy, Lithuanian University of Health Sciences, LT-50161 Kaunas, Lithuania Department of Radiology, Medical Academy, Lithuanian University of Health Sciences, LT-50161 Kaunas, Lithuania Correspondence to: Dr Brigita Glebauskiene, Department of Ophthalmology, Medical Academy, Lithuanian University of Health Sciences, Eiveniu 2, LT-50161 Kaunas, Lithuania, E-mail: bglebauskiene@gmail.com 05 2017 10 03 2017 13 5 3087 3099 22082016 13012017 Copyright: © Glebauskiene et al. 2017 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

The aim of the present study was to determine the association between sirtuin 1 (SIRT1), fibroblast growth factor receptor 2 (FGFR2) and signal transducer and activator of transcription 3 (STAT3) polymorphisms, and pituitary adenoma (PA) development, invasiveness, hormonal activity and recurrence. The present study included 143 patients with a diagnosis of PA. The reference group involved 808 healthy subjects. The genotyping of SIRT1 rs12778366, FGFR2 rs2981582 and STAT3 rs744166 was performed using the quantitative polymerase chain reaction method. The SIRT1 rs12778366 polymorphism analysis in the overall group revealed differences in the genotype distribution between patients with PA and control group subjects. The rs12778366 T/C genotype was observed to be different in non-invasive, non-recurrent and inactive PA subgroups compared with the control group, while the C/C genotype was observed to be different in invasive, recurrent and active PA subgroups compared with the control group. STAT3 rs744166 polymorphism analysis in the overall group revealed differences in the genotype distribution between patients with PA and the control groups. The rs744166 G/G genotype was observed to be different in invasive, non-recurrent and active PA subgroups compared with the control group, while the rs744166 A/A genotype was observed to be different in the active PA subgroup compared with the control group, and was also different in terms of invasiveness and recurrence in PA subgroups. The present study demonstrated that SIRT1 rs12778366 is associated with pituitary adenoma development while STAT3 rs744166 is associated with PA invasiveness, hormonal activity and recurrence.

 pituitary adenoma gene polymorphism fibroblast growth factor receptor 2 rs2981582 sirtuin 1 rs12778366 signal transducer and activator of transcription 3 rs744166 Introduction

Pituitary adenomas (PAs), located in a bone cavity termed the sella turcica, are one of the most common types of intracranial neoplasms, with reported estimated prevalence rates ranging between 14.4 and 22.5% in pooled autopsy and radiological series, respectively (1). Although the majority of PAs are benign, it is not uncommon for them to grow large and extend locally into the surrounding structures, invading the sphenoid bone inferiorly, the cavernous sinus laterally (2–7,8) and/or compressing the optic chiasm, if the direction of expansion is suprasellar, thus resulting in neurological complications, including headache and visual impairment (9–17). Certain types of PA are extremely invasive and may cause extensive destruction of the skull base (18). Investigation of tumour invasiveness is required, as this affects the management and prognosis of PA (19). The aim of the present study was to identify possible genes involved in PA tumourigenesis, which may serve as potential diagnostic and prognostic molecular markers. The present study selected 3 genes, sirtuin 1 (SIRT1), fibroblast growth factor receptor 2 (FGFR2) and signal transducer and activator of transcription 3 (STAT3), which are associated with different types of cancer, but are connected in pathogenic processes (20–23).

SIRT1 is a nicotinamide adenine dinucleotide-dependent histone deacetylase (HDAC) (24), which serves an important role in maintaining the balance between cell death and survival through targeting the Ku70-B-cell lymphoma-like protein 4 pathway (25), p53 (26,27) and forkhead box O3 (28), among others. A significant increase in the level of SIRT1 in hepatocellular carcinoma (29), breast cancer (30), prostate cancer (31), ovarian cancer (32), gastric cancer (33), colon cancer (34), glioblastoma (35) and lymphoma (36) was previously suggested to be associated with the development and invasion of these tumours. Furthermore, the rs12778366 polymorphism of the SIRT1 gene was found to be associated with breast cancer (37).

FGFR2 is a member of the FGFR family of tyrosine kinase receptors and participates in the process of tumourigenesis by inducing mitogenic and survival signals, and promoting invasiveness and angiogenesis (38). If cancer cells overexpress an FGFR with altered ligand-binding specificity, FGFs, secreted from neighbouring cells, stimulate the cancer cells, creating a paracrine loop (38). FGFR2 was previously revealed to be overexpressed in bladder (39) and lung cancer (40). Additionally, the importance of the FGFR2 rs2981582 gene polymorphism was investigated in breast (41–50) and prostate cancer (51).

STAT3 is activated in tumour cells and numerous immune cells of the tumour microenvironment, and is associated with tumour cell proliferation, invasion and angiogenesis (52–54). The effect of STAT3 was previously studied in the tumour development of colorectal adenocarcinoma (55), hepatocellular carcinoma (56), multiple myeloma (57), glioblastoma (58), prostate cancer (59), and head and neck cancer (60). The STAT3 rs744166 polymorphism was also evaluated in gastric (61,62), colon (63) and lung cancer (64).

These findings support the hypothesised role of SIRT1, FGFR2 and STAT3 as tumour promoters. However, an association between SIRT1, FGFR2 and STAT3 polymorphisms, and PA development, invasiveness, PA activity and recurrence has not yet been reported. The aim of the present study was to determine these associations.

 Materials and methods <sec> <title>Patients and selection

Permission to undertake the present study was obtained from the Biomedical Research Ethics Committee of Lithuanian Health Sciences University (Kaunas, Lithuania). The study was conducted in the Departments of Ophthalmology and Neurosurgery, Lithuanian Health Sciences University Hospital (Kaunas, Lithuania).

The participants comprised of 143 patients with a diagnosis of PA. The reference group involved 808 healthy subjects. The reference group was created by taking into consideration the distribution of age and gender in the PA group. Therefore, the median patient age of the control group and the PA group did not differ significantly (P<0.05). Demographic data of the study subjects are presented in Table I.

The inclusion criteria were as follows: Determined and confirmed PA via magnetic resonance imaging (MRI); general good condition of the patient; consent of the patient to take part in the study; age ≥18 years; and no other brain tumours or tumours with other localizations.

All PAs were analysed based on MRI findings. The pre-operative MRI investigations were performed with 1.5T MRI scanners (Siemens MAGNETOM Avanto: Siemens AG, Munich, Germany; 1.5 T Philips ACHIEVA: Philips Healthcare, DA Best, The Netherlands) using a head coil and a standard pituitary scanning protocol, obtaining T1-weighted (T1W) sagittal and coronal and T2W/turbo spin echo coronal pre-contrast images, and T1W coronal and sagittal gadolinium-enhanced MR images with the intravenous agent gadodiamide (Omniscan; GE Healthcare Life Sciences, Chalfont, UK). The retrospective analysis of MRI data was conducted by an experienced radiologist. The suprasellar extension and sphenoid sinus invasion by PAs were classified according to Wilson-Hardy classification (Hardy classification, modified by Wilson) (19). The degree of suprasellar and parasellar extension was graded as stages A-E. The degree of sellar floor erosion was graded between I and IV. Grade III, localized sellar destruction, and grade IV, diffuse destruction, were considered to be invasive PAs. The Knosp classification system (4) was used to quantify invasion of the cavernous sinus, in which only grades 3 and 4 define true invasion of the tumour into the cavernous sinus: Grade 0, no cavernous sinus involvement; grades 1 and 2, the tumour pushes into the medial wall of the cavernous sinus, but does not go beyond a hypothetical line extending between the centres of the two segments of the internal carotid artery (grade 1) or it goes beyond such a line, but without passing a line tangential to the lateral margins of the artery itself (grade 2); grade 3, the tumour extends laterally to the internal carotid artery within the cavernous sinus; and grade 4, total encasement of the intracavernous carotid artery.

 DNA extraction and genotyping

The DNA extraction and analysis of the gene polymorphisms of SIRT1 rs12778366, FGFR2 rs2981582 and STAT3 rs744166 were performed at the Laboratory of Ophthalmology at the Institute of Neuroscience of the Lithuanian University of Health Sciences (Kaunas, Lithuania). DNA was extracted from 200 µl venous blood (white blood cells) using a DNA purification kit based on the magnetic beads method (MagJET Genomic DNA kit; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to manufacturer's instructions.

The genotyping of SIRT1 rs12778366, FGFR2 rs2981582 and STAT3 rs744166 was performed using the quantitative polymerase chain reaction (qPCR) method with a Rotor-Gene Q Real-Time PCR Quantification system (Qiagen, Inc., Valencia, CA, USA). All 3 single-nucleotide polymorphisms were determined using TaqMan® Genotyping assays (Applied Biosystems; Thermo Fisher Scientific, Inc.), C_1340370_10 (rs12778366), C_2917302_10 (rs2981582) and C_3140282_10 (rs744166), according to the manufacturer's protocols.

The Allelic Discrimination program (Applied Biosystems; Thermo Fisher Scientific, Inc.) was used during the qPCR. The assay was then continued following the manufacturer protocols. The Allelic Discrimination program was completed, and the genotyping results were received. The program determined the individual genotypes according to the fluorescence intensity rate from different detectors: Molecular marker labeled with VIC fluorescent dye was chosen for the X axis and a molecular marker labeled with FAM fluorescent dye was selected for the Y axis. These dy-labeled probes were included in the TaqMan® Genotyping assays.

 Statistical analysis

Statistical analysis was performed using SPSS 20.0 software (IBM SPSS, Armonk, NY, USA). The data are presented as absolute numbers with percentages in brackets, and as median with minimum/maximum values. The frequencies of genotypes are presented as percentages.

Hardy-Weinberg analysis was performed to compare the observed and expected frequencies of rs12778366, rs2981582 and rs744166 using the χ2 test in all groups. The distribution of rs12778366, rs2981582 and rs744166 single-nucleotide polymorphisms (SNPs) in the PA and control groups was compared using the χ2 test or Fisher's exact test. Binomial logistic regression analysis was performed to estimate the impact of genotypes on PA development. Odds ratios (ORs) and 95% confidence intervals (CIs) are presented. Only statistically significant variables are presented in the tables. The selection of the most suitable genetic model was based on the Akaike Information Criterion (AIC), whereby the best genetic models were those with the lowest AIC values (65). P<0.05 was considered to indicate a statistically significant difference.

 Results <sec> <title>Genotype distribution in the PA patients and the control group

The genotyping of SIRT1 rs12778366, FGFR2 rs2981582 and STAT3 rs744166 was performed in the PA group and the control group subjects (Table II).

The distribution of analysed SIRT1 genotypes and allele frequencies in the control and PA groups did not match the Hardy-Weinberg equilibrium. The SIRT1 rs12778366 polymorphism analysis in the overall group revealed differences in the genotype distribution between patients with PA and control group subjects (P<0.001). The genotype T/C was significantly less frequent in the PA group compared with the healthy controls (0 vs. 17.5%; P<0.001) and the genotype C/C was significantly more frequent in the PA group compared with the healthy control group (18.9 vs. 2.5%, respectively; P<0.001) (Table II).

The distribution of analysed FGFR2 genotypes and allele frequencies in the control and PA groups did not match the Hardy-Weinberg equilibrium. Statistical analysis did not reveal significant genotype (G/G, G/A and A/A) distribution differences between the control and PA groups: 41.6 vs. 39.2%, 53.1 vs. 58.7%, and 5.3 vs. 2.1%, respectively (P=0.174) (Table II).

The distribution of the analysed STAT3 rs744166 genotypes and allele frequencies did not match the Hardy-Weinberg equilibrium in the control group, but it did in the group of patients with PA. STAT3 rs744166 polymorphism analysis in the overall group revealed differences in the genotype distribution between the patients with PA and the control group (P=0.012). The genotype G/G was less frequent in the PA group compared with the healthy controls (9.1 vs. 19.1%, respectively; P=0.003) (Table II).

 Genotype distribution in the PA patients and the control group by gender

All 3 SNPs were analysed in the PA and control groups according to gender (Table III). SIRT1 rs12778366 polymorphism analysis did not revealed any statistically significant differences between females and males with PA in genotype (T/T, T/C and C/C) distribution (80.7, 0 and 19.3% vs. 81.8, 0 and 18.2%, respectively; Table III). Comparing SIRT1 rs12778366 genotype distribution in healthy females and females with PA, significant differences were revealed. The T/C genotype was less frequently present in females with PA compared with the healthy control females (0 vs. 17.5%, respectively; P<0.001) and C/C was more frequent in PA females compared with healthy females (19.3 vs. 2.7%, respectively; P<0.001). The T/T genotype did not exhibit any significant differences when healthy females and females with PA were compared. When analysing genotype distribution in males, T/C genotype distribution showed statistically significant difference between males with PA and healthy males (0 vs. 17.4%, respectively; P<0.001) and the C/C genotype was more frequent in males with PA compared with the control group (18.2 vs. 2.0%, respectively; P<0.001) (Table III).

FGFR2 rs2981582 polymorphism analysis by gender was performed, but it did not reveal any genotype distribution differences between females and males.

STAT3 rs744166 polymorphism analysis did not reveal any significant differences between females and males with PA in the genotype (G/G, G/A and A/A) distribution (8.0, 48.9 and 43.2% vs. 10.9, 45.5 and 43.6%, respectively; Table III) either. When comparing STAT3 genotype distribution between healthy females and females with PA, there were significant differences. The STAT3 rs744166 G/G genotype was less frequently present in PA females compared with healthy control females (8.0 vs. 20.4%, respectively; P=0.004). The STAT3 rs744166 G/A and A/A genotype distribution did not exhibit any significant differences when healthy females and females with PA were compared. STAT3 rs744166 analysis between male groups did not reveal any statistically significant differences (Table III).

 Binomial logistic regression analysis of the patients with PA and the control group

Binomial logistic regression analysis of the patients with PA and the control group was performed (Table IV). SIRT1 rs12778366 analysis revealed that there were significant variables in the co-dominant (OR=7.530; 95% CI: 4.087–13.873; P<0.001), recessive (OR=9.171; 95% CI: 4.982–16.881; P<0.001) and additive (OR=1.584: 95 % CI: 1.187–2.115; P=0.002) models of the patients with PA and the control group (Table IV).

FGFR2 rs2981582 analysis did not reveal any significant variables.

STAT3 rs744166 analysis revealed that there were significant variables in the co-dominant (OR=0.396; 95% CI: 0.211–0.743; P=0.004), recessive (OR=0.425; 95% CI: 0.234–0.771; P=0.005) and additive (OR=0.702; 95% CI: 0.541–0.911; P=0.008) models of the patients with PA and the control group (Table IV).

Binomial logistic regression analysis in the patients with PA and the control group according to gender was performed (Table V). In the SIRT1 rs12778366 analysis there were statistically significant variables in the co-dominant (P<0.001) and recessive (P<0.001) models of males. The co-dominant (P<0.001), recessive (P<0.001) and additive (P=0.013) variables were also significant in females.

Binomial logistic regression analysis of FGFR2 rs2981582 in the patients with PA and in the control group according to gender was performed, but no significant variables were observed.

However, binomial logistic regression analysis of STAT3 rs744166 in the patients with PA and in the control group according to gender showed statistically significant variables only in the co-dominant (P=0.007), recessive (P=0.008) and additive (P=0.012) models of females (Table V).

 Genotype distribution in the control group and the PA patients by different PA subgroups

Analysis of SIRT1 rs1277836, FGFR2 rs2981582 and STAT3 rs744166 polymorphisms was performed by different PA subgroups (Tables VI–VIII).

The SIRT1 rs12778366 T/C genotype was less frequently observed in non-invasive, non-recurrent and inactive PA subgroups compared with healthy controls (0 vs. 17.5%, P=0.021; 0 vs. 17.5%, P<0.001; 0 vs. 17.5%, P<0.001, respectively). However, no differences were observed between non-invasive and invasive, non-recurrent and recurrent, and inactive and active PA subgroups (Tables VI–VIII).

Additional analysis revealed that the C/C genotype was more frequent in invasive, recurrent and active PA subgroups compared with the healthy controls (18.8 vs. 2.5%, P=0.041; 19.4 vs. 2.5%, P=0.047; 15.0 vs. 2.5%, P<0.001, respectively;Tables VI, VII and VIII).

The FGFR2 rs2981582 G/G genotype was less frequently observed in the non-invasive PA subgroup compared with the healthy controls (27.6 vs. 41.6%, respectively; P=0.038), but the G/A genotype was more frequently observed in the non-invasive PA subgroup compared with the control group (72.4 vs. 53.1%, respectively; P=0.004) and the invasive PA subgroup (72.4 vs. 49.4% respectively; P=0.009) (Table VI).

Statistical analysis was performed to evaluate the FGFR2 rs2981582 association with PA activity and recurrence (Tables VII and VIII). This analysis did not reveal any association between SNP and active or non-active PA, and PA without recurrence or with recurrence.

The STAT3 rs744166 G/G genotype was less frequently observed in invasive, non-recurrent and active PA subgroups compared with healthy controls (4.7 vs. 19.1%, P<0.001; 6.2 vs. 19.1%, P<0.001; 8.8 vs. 19.1%, P=0.022, respectively).

The STAT3 rs744166 A/A genotype was more frequent in the active PA subgroup compared with the control group (48.8 vs. 36.0%, respectively; P=0.029). There were differences between non-invasive and invasive, non-recurrent and recurrent PA subgroups as well, with the exception of comparing inactive and active PA. The STAT3 rs744166 G/G genotype was more frequent in non-invasive PA compared with invasive PA (15.5 vs. 4.7%, respectively; P=0.038) and in recurrent PA group comparing to non-recurrent PA (19.4 vs. 6.2%, respectively; P=0.036) (Tables VI–VIII).

 Binomial logistic regression analysis of the control group and the PA patients by different PA subgroups

Binomial logistic regression analysis in the non-invasive PA, invasive PA and control groups was performed (Table IX). Analysing the SIRT1 polymorphism in non-invasive PA group and control group this analysis showed that the co-dominant (P<0.001), recessive (P<0.001) and additive (P=0.025) variables were significant. Binomial logistic regression analysis in the patients with invasive PA and the control group revealed significance of the same co-dominant (P<0.001), recessive (P<0.001) and additive (P=0.010) variables (Table IX).

Binomial logistic regression analysis of FGFR2 rs2981582 in the non-invasive PA and control groups showed that the co-dominant (P=0.017), dominant (P=0.039) and over-dominant (P=0.005) variables were significant, but this analysis in the patients with invasive PA and the control group did not reveal any significance of these models.

Binomial logistic regression analysis of STAT3 rs744166 was also performed (Table IX). The analysis showed that the co-dominant (P=0.004), recessive (P=0.003) and additive (P=0.011) variables were statistically significant only in the invasive PA and control groups (Table IX).

Binomial logistic regression analysis in the inactive PA and control groups, and in the active PA and control groups, was performed for all 3 SNPs (Table X).

Inactive PA group analysis of SIRT1 rs12778366 showed that the co-dominant (P<0.001), recessive (P<0.001) and additive (P<0.001) variables were significant. The analysis of the active PA group revealed significance in the co-dominant (P<0.001) and recessive (P<0.001) models (Table X).

Binomial logistic regression analysis of FGFR2 rs2981582 was performed in the inactive PA, active PA and control groups, but this analysis did not reveal significance in these models.

STAT3 rs744166 analysis in the inactive PA group showed that there were no significant variables. Analysing the active PA group, the present study revealed significance in the co-dominant (P=0.010), dominant (P=0.026), recessive (P=0.027) and additive (P=0.007) models (Table X).

Additional binomial logistic regression analysis of SNPs was performed in non-recurrence, recurrence and control groups. The analysis in the non-recurrent PA and control groups showed that the co-dominant (P<0.001), recessive (P<0.001) and additive (P=0.005) variables were statistically significant (Table XI). In the analysis of PA with recurrence, the co-dominant (P<0.001) and recessive (P<0.001) variables were also significant.

FGFR2 rs2981582 polymorphism analysis did not show any statistical significance.

Binomial logistic regression analysis of STAT3 rs744166 in the non-recurrent and recurrent PA groups, and the control group was performed. This revealed that in the non-recurrent PA and control groups, the co-dominant (P=0.002), recessive (P=0.002) and additive (P=0.005) variables were significant (Table XI). In the analysis of PA with recurrence there were no statistically significant variables.

 Discussion

The impact of SIRT1, FGFR2 and STAT3 gene polymorphisms on the development of various tumours has been analysed in numerous studies (38,42–52,60,63–65), but no studies have investigated the associations with PA development, invasiveness, activity and recurrence.

A study conducted by Rizk et al (37) investigated SIRT1 gene single nucleotide polymorphism rs12778366 in patients with breast cancer, revealing that the SIRT1 rs12778366 T/T genotypes were more frequent, exhibited higher SIRT1 levels than the C/C and C/T genotypes, and were associated with histological grade and lymph node status. The T allele frequency was higher in patients with breast cancer compared with that in normal subjects.

The present study was the first to assess the association between SIRT1 rs12778366 and PA. It was found that the T/C genotype was less frequent in the PA group compared with the healthy controls (0 vs. 17.5%, respectively; P<0.001) and that the C/C genotype was more frequent in the PA group compared with the healthy control group (18.9 vs. 2.5%, P<0.001).

Numerous studies have investigated the FGFR2 rs2981582 polymorphism in breast cancer patients, and have provided controversial data on the impact of this polymorphism on tumour development. Chen et al (43) revealed that the G/A and A/A genotypes of FGFR2 rs2981582 were associated with lower mammographic density and a reduced risk of breast cancer, and Butt et al (42) revealed a statistically significant association between the FGFR2 rs2981582 A/A genotype and breast cancer risk. Shan et al (66) also revealed that patients with the A/A genotype of FGFR2 rs2981582 exhibited an increased risk of breast cancer, while Ledwoń et al (47) revealed that the rs2981582 SNP showed significant association with the familial and sporadic types of breast cancer. On the basis of these findings, the present study aimed to examine whether the polymorphism in the FGFR2 promoter may affect the risk of PA development, activity, recurrence or invasiveness. No differences in genotype (G/G, G/A and A/A) distribution were observed between the control and PA groups (41.6 vs. 39.2%, 53.1 vs. 58.7%, and 5.3 vs. 2.1%, respectively; P=0.174). No significant differences were observed between genotype distribution according to gender, PA activity, invasiveness or recurrence.

Several studies have analysed the STAT3 rs744166 polymorphism in association with various types of tumour, but none have investigated the association between STAT3 rs744166 and PA. Rocha et al (61) reported that the rs744166 polymorphic G allele was associated with gastric cancer, and a significantly decreased risk of non-small cell lung cancer was observed for carriers of STAT3 rs744166 in a study by Jiang et al (64). The present study demonstrated the differences in the distribution of the STAT3 rs744166 polymorphism between patients with PA and control group subjects (P=0.012). The G/G genotype was less frequent in the PA group compared with the healthy controls (9.1 vs. 19.1%, respectively; P=0.003). Analysis in different PA subgroups showed that the STAT3 rs744166 G/G genotype was more frequent in non-invasive PA compared with invasive PA (15.5 vs. 4.7%; P=0.038) and in recurrent PA compared with the non-recurrent PA (19.4 vs. 6.2%, respectively; P=0.036).

Overall, the present study demonstrated that the SNPs SIRT1 rs12778366 and STAT3 require replication in future larger studies, particularly with increased sample sizes to confirm the association of SIRT1 and STAT3 in patients with PA.

 Acknowledgements

The present study received funding from the Research Council of Lithuania (grant no. MIP-008/2014).

 References EzzatSAsaSLCouldwellWTBarrCEDodgeWEVanceMLMcCutcheonIEThe prevalence of pituitary adenomas: A systematic reviewCancer101613619200410.1002/cncr.2041215274075 ColaoADi SommaCPivonelloRFaggianoALombardiGSavastanoSMedical therapy for clinically non-functioning pituitary adenomasEndocr Relat Cancer15905915200810.1677/ERC-08-018118780796 KasputytėRSlatkevičienėGLiutkevičienėRGlebauskienėBBernotasGTamašauskasAChanges of visual functions in patients with pituitary adenomaMedicina (Kaunas)49132137201323893057 KnospESteinerEKitzKMatulaCPituitary adenomas with invasion of the cavernous sinus space: A magnetic resonance imaging classification compared with surgical findingsNeurosurgery33610618199310.1097/00006123-199310000-000088232800 FerranteEFerraroniMCastrignanòTMenicattiLAnagniMReimondoGDel MontePBernasconiDLoliPFaustini-FustiniMNon-functioning pituitary adenoma database: A useful resource to improve the clinical management of pituitary tumoursEur J Endocrinol155823829200610.1530/eje.1.0229817132751 AhmadiJNorthCMSegallHDZeeCSWeissMHCavernous sinus invasion by pituitary adenomasAm J Neuroradiol68938981985 FalbuschRBuchfejderMTranssphenoidal surgery of parasellar pituitary adenomasActa Neurochir929399198810.1007/BF014019783407479 ScheithauerBWKovacsKTLawsERJrRandallRVPathology of invasive pituitary tumors with special reference to functional classificationJ Neurosurg65733744198610.3171/jns.1986.65.6.07333095506 MoonCHHwangSCKimBTOhnYHParkTKVisual prognostic value of optical coherence tomography and photopic negative response in chiasmal compressionInvest Ophthalmol Vis Sci5285278533201110.1167/iovs.11-803421960556 ThomasRShenoyKSeshadriMSMuliyilJRaoAPaulPVisual field defects in non-functioning pituitary adenomasIndian J Ophthalmol50127130200212194569 TrautmannJCLawsERJrVisual status after transphenoidal surgery at the Mayo Clinic 1971–1982Am J Ophthalmol96200208198310.1016/S0002-9394(14)77788-86881243 KaurABanerjiDKumarDSharmaKVisual status in suprasellar pituitary tumoursIndian J Ophthalmol4313113419958822488 MortiniPLosaMBarzaghiRBoariNGiovanelliMResults of transsphenoidal surgery in a large series of patients with pituitary adenomaNeurosurgery5612221233200510.1227/01.NEU.0000159647.64275.9D15918938 el-AzouziMBlackPMCandiaGZervasNTPanagopoulosKPTranssphenoidal surgery for visual loss in patients with pituitary adenomasNeurol Res122325199010.1080/01616412.1990.117399071970620 MarazuelaMAstigarragaBVicenteAEstradaJCuerdaCGarcia-UriaJLucasTRecovery of visual and endocrine function following transsphenoidal surgery of large nonfunctioning pituitary adenomasJ Endocrinol Invest17703707199410.1007/BF033477637868814 GondimJAde AlmeidaJPde AlbuquerqueLASchopsMGomesEFerrazTHeadache associated with pituitary tumorsJ Headache Pain101520200910.1007/s10194-008-0084-019067118 LevyMJJägerHPowellMMatharuMSMeeranKGoadsbyPJPituitary volume and headache: Size is not everythingArch Neurol61721725200410.1001/archneur.61.5.72115148150 StevensJMValentineARKendallBEComputed cranial and spinal imagingA practical introductionLippincott Williams & WilkinsPhiladelphia, PA1541561988 WilsonCBNeurosurgical management of large and invasive pituitary tumorsClinical management of pituitary disordersTindallGTCollinsWFRavenNew York, NY3353421979 LuoJCPengYLChenTSHuoTIHouMCHuangHCLinHCLeeFYClopidogrel inhibits angiogenesis of gastric ulcer healing via down regulation of vascular endothelial growth factor receptor 2J Formos Med Assoc115764772201610.1016/j.jfma.2015.07.02226315480 LuJZhangLChenXLuQYangYLiuJMaXSIRT1 counteracted the activation of STAT3 and NF-κB to repress the gastric cancer growthInt J Clin Exp Med750505058201425664004 NieYErionDMYuanZDietrichMShulmanGIHorvathTLGaoQSTAT3 inhibition of gluconeogenesis is downregulated by SirT1Nat Cell Biol11492500200910.1038/ncb185719295512 BernierMPaulRKMartin-MontalvoAScheibye-KnudsenMSongSHeHJArmourSMHubbardBPBohrVAWangLNegative regulation of STAT3 protein-mediated cellular respiration by SIRT1 proteinJ Biol Chem2861927019279201110.1074/jbc.M110.20031121467030 ImaiSArmstrongCMKaeberleinMGuarenteLTranscriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylaseNature403795800200010.1038/3500162210693811 CohenHYMillerCBittermanKJWallNRHekkingBKesslerBHowitzKTGorospeMde CaboRSinclairDACalorie restriction promotes mammalian cell survival by inducing the SIRT1 deacetylaseScience305390392200410.1126/science.109919615205477 VaziriHDessainSKEatonNg EImaiSIFryeRAPanditaTKGuarenteLWeinbergRAhSIR2(SIRT1) functions as an NAD-dependent p53 deacetylaseCell107149159200110.1016/S0092-8674(01)00527-X11672523 ChenWYWangDHYenRCLuoJGuWBaylinSBTumor suppressor HIC1 directly regulates SIRT1 to modulate p53-dependent DNA-damage responsesCell123437448200510.1016/j.cell.2005.08.01116269335 MottaMCDivechaNLemieuxMKamelCChenDGuWBultsmaYMcBurneyMGuarenteLMammalian SIRT1 represses forkhead transcription factorsCell116551563200410.1016/S0092-8674(04)00126-614980222 ChenHCJengYMYuanRHHsuHCChenYLSIRT1 promotes tumorigenesis and resistance to chemotherapy in hepatocellular carcinoma and its expression predicts poor prognosisAnn Surg Oncol1920112019201210.1245/s10434-011-2159-422146883 SungJYKimRKimJELeeJBalance between SIRT1 and DBC1 expression is lost in breast cancerCancer Sci10117381744201010.1111/j.1349-7006.2010.01573.x20412117 HuffmanDMGrizzleWEBammanMMKimJSEltoumIAElgavishANagyTRSIRT1 is significantly elevated in mouse and human prostate cancerCancer Res6766126618200710.1158/0008-5472.CAN-07-008517638871 JangKYKimKSHwangSHKwonKSKimKRParkHSParkBHChungMJKangMJLeeDGMoonWSExpression and prognostic significance of SIRT1 in ovarian epithelial tumoursPathology41366371200910.1080/0031302090288445119404850 ChaEJNohSJKwonKSKimCYParkBHParkHSLeeHChungMJKangMJLeeDGExpression of DBC1 and SIRT1 is associated with poor prognosis of gastric carcinomaClin Cancer Res1544534459200910.1158/1078-0432.CCR-08-332919509139 StunkelWPehBKTanYCNayagamVMWangXSalto-TellezMNiBEntzerothMWoodJFunction of the SIRT1 protein deacetylase in cancerBiotechnol J213601368200710.1002/biot.20070008717806102 LiuGYuanXZengZTuniciPNgHAbdulkadirIRLuLIrvinDBlackKLYuJSAnalysis of gene expression and chemoresistance of CD133+ cancer stem cells in glioblastomaMol Cancer567200610.1186/1476-4598-5-1817140455 JangKYHwangSHKwonKSKimKRChoiHNLeeNRKwakJYParkBHParkHSChungMJSIRT1 expression is associated with poor prognosis of diffuse large B-cell lymphomaAm J Surg Pathol3215231531200810.1097/PAS.0b013e31816b647818724249 RizkSMShahinNNShakerOGAssociation between SIRT1 gene polymorphisms and breast cancer in EgyptiansPLoS One11e0151901201610.1371/journal.pone.015190126999517 WescheJHaglundKHaugstenEMFibroblast growth factors and their receptors in cancerBiochem J437199213201110.1042/BJ2010160321711248 MarzioniDLorenziTMazzucchelliRCapparucciaLMorroniMFioriniRBracalentiCCatalanoADavidGCastellucciMExpression of basic fibroblast growth factor, its receptors and syndecans in bladder cancerInt J Immunopathol Pharmacol22627638200910.1177/03946320090220030819822079 DaviesHHunterCSmithRStephensPGreenmanCBignellGTeagueJButlerAEdkinsSStevensCSomatic mutations of the protein kinase gene family in human lung cancerCancer Res6575917595200516140923 BayraktarSThompsonPAYooSYDoKASahinAAArunBKBondyMLBrewsterAMThe relationship between eight GWAS-identified single-nucleotide polymorphisms and primary breast cancer outcomesOncologist18493500201310.1634/theoncologist.2012-041923635555 ButtSHarlidSBorgquistSIvarssonMLandbergGDillnerJCarlsonJManjerJGenetic predisposition, parity, age at first childbirth and risk for breast cancerBMC Res Notes5414201210.1186/1756-0500-5-41422867275 ChenYShiCGuoQTNRC9 rs12443621 and FGFR2 rs2981582 polymorphisms and breast cancer riskWorld J Surg Oncol1450201610.1186/s12957-016-0795-726911390 CuiFWuDWangWHeXWangMVariants of FGFR2 and their associations with breast cancer risk: a HUGE systematic review and meta-analysisBreast Cancer Res Treat155313335201610.1007/s10549-015-3670-226728143 CampaDBarrdahlMGaudetMMBlackAChanockSJDiverWRGapsturSMHaimanCHankinsonSHazraAGenetic risk variants associated with in situ breast cancerBreast Cancer Res1782201510.1186/s13058-015-0596-x26070784 SiddiquiSChattopadhyaySAkhtarMSNajmMZDeoSVShuklaNKHusainSAA study on genetic variants of fibroblast growth factor receptor 2 (FGFR2) and the risk of breast cancer from North IndiaPLoS One9e110426201410.1371/journal.pone.011042625333473 LedwońJKHennigEEMaryanNGorycaKNowakowskaDNiwińskaAOstrowskiJCommon low-penetrance risk variants associated with breast cancer in Polish womenBMC Cancer13510201310.1186/1471-2407-13-51024171766 Murillo-ZamoraEMoreno-MacíasHZivERomieuILazcano-PonceEÁngeles-LlerenasAPérez-RodríguezEVidal-MillánSFejermanLTorres-MejíaGAssociation between rs2981582 polymorphism in the FGFR2 gene and the risk of breast cancer in Mexican womenArch Med Res44459466201310.1016/j.arcmed.2013.08.00624054997 OttiniLSilvestriVSaievaCRizzoloPZannaIFalchettiMMasalaGNavazioASGrazianoVBianchiSAssociation of low-penetrance alleles with male breast cancer risk and clinicopathological characteristics: Results from a multicenter study in ItalyBreast Cancer Res138861868201310.1007/s10549-013-2459-4 JaraLGonzalez-HormazabalPCerceñoKDi CapuaGAReyesJMBlancoRBravoTPeraltaOGomezFWaughEGenetic variants in FGFR2 and MAP3K1 are associated with the risk of familial and early-onset breast cancer in a South-American populationBreast Cancer Res Treat137559569201310.1007/s10549-012-2359-z23225170 MilesFLRaoJYEckhertCChangSCPantuckAZhangZFAssociations of immunity-related single nucleotide polymorphisms with overall survival among prostate cancer patientsInt J Clin Exp Med81147011476201526379965 YuHKortylewskiMPardollDCrosstalk between cancer and immune cells: Role of STAT3 in the tumour microenvironmentNat Rev Immunol74151200710.1038/nri199517186030 YuHPardollDJoveRSTATs in cancer inflammation and immunity: a leading role for STAT3Nat Rev Cancer9798809200910.1038/nrc273419851315 NiuGWrightKLHuangMSongLHauraETurksonJZhangSWangTSinibaldiDCoppolaDConstitutive Stat3 activity up-regulates VEGF expression and tumor angiogenesisOncogene2120002008200210.1038/sj.onc.120526011960372 KusabaTNakayamaTYamazumiKYakataYYoshizakiANagayasuTSekineIExpression of p-STAT3 in human colorectal adenocarcinoma and adenoma; correlation with clinicopathological factorsJ Clin Pathol58833838200510.1136/jcp.2004.02341616049285 HeGKarinMNF-κB and STAT3-key players in liver inflammation and cancerCell Res21159168201110.1038/cr.2010.18321187858 PandeyMKSungBAggarwalBBBetulinic acid suppresses STAT3 activation pathway through induction of protein tyrosine phosphatase SHP-1 in human multiple myeloma cellsInt J Cancer1272821392201019937797 GariboldiMBRavizzaRMontiEThe IGFR1 inhibitor NVP-AEW541 disrupts a pro-survival and pro-angiogenic IGF-STAT3-HIF1 pathway in human glioblastoma cellsBiochem Pharmacol80455462201010.1016/j.bcp.2010.05.01120488164 ShinJLeeHJJungDBJungJHLeeHJLeeEOLeeSGShimBSChoiSHKoSGSuppression of STAT3 and HIF-1 alpha mediates anti-angiogenic activity of betulinic acid in hypoxic PC-3 prostate cancer cellsPLoS One6e21492201110.1371/journal.pone.002149221731766 Leeman-NeillRJWheelerSESinghSVThomasSMSeethalaRRNeillDBPanahandehMCHahmERJoyceSCSenMGuggulsterone enhances head and neck cancer therapies via inhibition of signal transducer and activator of transcription-3Carcinogenesis3018481856200910.1093/carcin/bgp21119762335 RochaGARochaAMGomesADFariaCLJrMeloFFBatistaSAFernandesVCAlmeidaNBTeixeiraKNBritoKSQueirozDMSTAT3 polymorphism and helicobacter pylori CagA strains with higher number of EPIYA-C segments independently increase the risk of gastric cancerBMC Cancer15528201510.1186/s12885-015-1533-126186918 YuanKLiuHHuangLRenXLiuJDongXTianWJiaYrs744166 polymorphism of the STAT3 gene is associated with risk of gastric cancer in a Chinese populationBiomed Res Int2014527918201410.1155/2014/52791824864251 RyanBMWolffRKValeriNKhanMRobinsonDPaoneABowmanEDLundgreenACaanBPotterJAn analysis of genetic factors related to risk of inflammatory bowel disease and colon cancerCancer Epidemiol38583590201410.1016/j.canep.2014.07.00325132422 JiangBZhuZZLiuFYangLJZhangWYYuanHHWangJGHuXHHuangGSTAT3 gene polymorphisms and susceptibility to non-small cell lung cancerGenet Mol Res1018561865201110.4238/vol10-3gmr107121948749 AkaikeHInformation theory as an extension of the maximum likelihood principlePetrovBNCsakiF2nd International Symposium on Information TheoryAkademiai KiadoBudapest2672811973 ShanJMahfoudhWDsouzaSPHassenEBouaouinaNAbdelhakSBenhadjayedAMemmiHMathewRAAighaIIGenome-wide association studies (GWAS) breast cancer susceptibility loci in Arabs: Susceptibility and prognostic implications in TunisiansBreast Cancer Res Treat135715724201210.1007/s10549-012-2202-622910930

Demographic characteristics of patients with PA and reference group subjects.

Group n Min/max/median age, years Females, n (%)
PA 143 19/87/52.5 88 (65.67)
Control 808 20/90/58 510 (63.12)
P-value – 0.793a 0.882

P-value for comparison of the median age between the PA and control groups. PA, pituitary adenoma; min, minimum; max, maximum.

Frequency of single nucleotide polymorphisms in patients with PA and the control group.

Gene marker Control group, n (%) P-value HWE PA group, n (%) P-value HWE χ2 P-value
SIRT1 rs12778366
  Genotype
  T/T 647 (80.1) <0.01 116 (81.1) <0.001 91.139 <0.001
  T/C 141 (17.5)a 0 (0.0)a
  C/C 20 (2.5)b 27 (18.9)b
  Total 808 (100.0) 143 (100)
  Allele
  T 1,435 (88.8) 232 (81.1)
  C 181 (11.2) 54 (18.9)
FGFR2 rs2981582
  Genotype
  G/G 336 (41.6) <0.001 56 (39.2) <0.001 3.502 0.174
  G/A 429 (53.1) 84 (58.7)
  A/A 43 (5.3) 3 (2.1)
  Total 808 (100.0) 143 (100.0)
  Allele
  G 1,101 (68.1) 196 (68.53)
  A 515 (31.9) 90 (31.47)
STAT3 rs744166
  Genotype
  G/G 154 (19.1)c <0.001 13 (9.1)c 0.354 8.825 0.012
  G/A 363 (44.9) 68 (47.6)
  A/A 291 (36.0) 62 (43.4)
  Total 808 (100.0) 143 (100.0)
  Allele
  G 671 (41.5) 94 (32.9)
  A 945 (58.5) 192 (67.1)

SIRT1 rs1277836 T/C genotype was significantly less frequent (P<0.001) in the PA group compared with the control group.

SIRT1 rs1277836 C/C genotype was significantly more frequent (P<0.001) in the PA group compared with the control group.

STAT3 rs744166 G/G genotype was significantly less frequent (P=0.003) in the PA group compared with the control group. PA, pituitary adenoma; SIRT1, sirtuin 1, FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; HWE, Hardy-Weinberg equilibrium.

Frequency of single nucleotide polymorphisms in patients with PA and control group according to gender.

Control group, n (%) PA group, n (%)
Gene marker Females Males P-value HWE P-value Females Males P-value HWE P-value
SIRT1 rs12778366
  Genotype
  T/T 407 (79.8) 240 (80.5) 0.811 0.801 71 (80.7) 45 (81.8) 0.866 0.866
  T/C 89 (17.5)a 52 (17.4)b 1.000 0 (0.0)a 0 (0.0)b 1.00
  C/C 14 (2.7)c 6 (2.0)d 0.518 17 (19.3)c 10 (18.2)d 0.866
  Total 510 (100.0) 298 (100.0) 88 (100.0) 55 (100.0)
  Allele
  T 903 (88.5) 532 (89.3) 142 (80.7) 90 (81.8)
  C 117 (11.5) 64 (10.7) 34 (19.3) 20 (18.2)
FGFR2 rs2981582
  Genotype
  G/G 218 (42.7) 118 (39.6) 0.675 0.381 39 (44.3) 17 (30.9) 0.197 0.117
  G/A 265 (52.0) 164 (55.0) 0.398 48 (54.5) 36 (65.5) 0.224
  A/A 27 (5.3) 16 (5.4) 1.0 1 (1.1) 2 (3.6) 0.559
  Total 510 (100) 298 (100) 88 (100) 55 (100)
  Allele
  G 701 (68.7) 400 (67.1) 126 (71.6) 70 (63.6)
  A 319 (31.3) 196 (32.9) 50 (28.4) 40 (36.4)
STAT3 rs744166
  Genotype
  G/G 104 (20.4)e 50 (16.8) 0.378 0.207 7 (8.0)e 6 (10.9) 0.815 0.563
  G/A 229 (44.9) 134 (45.0) 0.986 43 (48.9) 25 (45.5) 0.733
  A/A 177 (34.7) 114 (38.3) 0.312 38 (43.2) 24 (43.6) 1.00
  Total 510 (100) 298 (100) 88 (100) 55 (100)
  Allele
  G 437 (42.84) 234 (39.3) 57 (32.4) 37 (33.6)
  A 583 (57.2) 362 (60.7) 119 (67.6) 73 (66.4)

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in PA females compared with control females.

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in PA males compared with control males.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P<0.001) in PA females compared with control females.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P<0.001) in PA males compared with control males.

STAT3 rs744166 G/G genotype is significantly less frequent (P=0.004) in PA females compared with control females. PA, pituitary adenoma; SIRT1, sirtuin 1; FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; HWE, Hardy-Weinberg equilibrium.

Binomial logistic regression analysis in patients with pituitary adenoma and the control group.

Gene Model Genotype OR (95% CI) P-value AIC
SIRT1 rs12778366 Co-dominant T/T 1.000 720.516
T/C 0 (0.000) 0.995
C/C 7.530 (4.087–13.087) <0.001
Recessive T/T+T/C 1.000 780.895
C/C 9.171 (4.982–16.982) <0.001
Additive – 1.584 (1.187–2.187) 0.002 780.214
STAT3 rs744166 Co-dominant A/A 1.000 801.223
G/G 0.879 (0.603–1.603) 0.504
G/G 0.396 (0.211–0.211) 0.004
Recessive A/A+G/A 1.000 799.670
G/G 0.425 (0.234–0.234) 0.005
Additive – 0.702 (0.541–0.541) 0.008 801.881

OR, odds ratio; CI, confidence interval; AIC, Akaike Information Criterion; SIRT1, sirtuin 1; STAT3, signal transducer and activator of transcription 3.

Binomial logistic regression analysis in patients with pituitary adenoma and control subjects according to gender.

Gene Gender Model Genotype OR (95% CI) P-value AIC
SIRT1 rs12778366 Male Co-dominant T/T 1.000 275.783
T/C 0.000 (0.000) 0.997
C/C 8.889 (3.076–25.076) <0.001
Recessive T/T+T/C 1.000 290.082
C/C 10.815 (3.748–31.748) <0.001
Female Co-dominant T/T 1.000 450.358
T/C 0.000 (0.000) 0.996
C/C 6.961 (3.285–14.285) <0.001
Recessive T/T+T/C 1.000 474.428
C/C 8.483 (4.008–17.008) <0.001
Additive – 1.580 (1.100–2.100) 0.013 497.979
STAT3 rs744166 Female Co-dominant A/A 1.000 496.249
G/A 0.875 (0.542–1.542) 0.583
G/G 0.314 (0.135–0.135) 0.007
Recessive A/A+G/A 1.000 494.549
G/G 0.337 (0.151–0.151) 0.008
Additive – 0.654 (0.469–0.469) 0.012 497.077

OR, odds ratio; CI, confidence interval; AIC, Akaike Information Criterion; SIRT1, sirtuin 1; STAT3, signal transducer and activator of transcription 3.

Frequency of SNPs in patients with PA and in control group according to PA invasiveness.

Gene marker Control group, n (%) P-value HWE Non invasive PA group, n (%) P-value HWE Invasive PA group, n (%) P-value HWE
SIRT1 rs12778366
  Genotype
  T/T 647 (80.1) <0.001 47 (81.0) <0.001 69 (81.2) <0.001
  T/C 141 (17.5)a,b 0 (0.0)a 0 (0.0)b
  C/C 20 (2.5)c,d 11 (19.0)c 16 (18.8)d
  Total 808 (100.0) 58 (100.0) 85 (100.0)
  Allele
  T 1,435 (88.8) 94 (81.0) 138 (81.2)
  C 181 (11.2) 22 (19.0) 32 (18.8)
FGFR2 rs2981582
  Genotype
  G/G 336 (41.6)e <0.001 16 (27.6)e,f <0.001 40 (47.1)f 0.043
  G/A 429 (53.1)g 42 (72.4)g,h 42 (49.4)h
  A/A 43 (5.3) 0 (0.0) 3 (3.5)
  Total 808 (100.0) 58 (100.0) 85 (100.0)
  Allele
  G 1,101 (68.1) 74 (63.8) 122 (71.8)
  A 515 (31.9) 42 (36.2) 48 (28.2)
STAT3 rs744166
  Genotype
  G/G 154 (19.1)i <0.001 9 (15.5)j 0.313 4 (4.7)i,j 0.031
  G/A 363 (44.9) 23 (39.7) 45 (52.9)
  A/A 291 (36.0) 26 (44.8) 36 (42.4)
  Total 808 (100.0) 58 (100.0) 85 (100.0)
  Allele
  G 671 (41.5) 41 (35.3) 53 (31.2)
  A 945 (58.5) 75 (64.7) 117 (68.8)

SIRT1 rs12778366 T/C genotype is significantly less frequent (P=0.021) in non-invasive PA compared with the control group.

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in invasive PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P=0.041) in non-invasive PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly less frequent (P<0.001) in invasive PA compared with the control group.

FGFR2 rs2981582 G/G genotype is significantly less frequent (P=0.038) in non-invasive PA compared with the control group.

FGFR2 rs2981582 G/G genotype is significantly more frequent (P=0.024) in invasive PA compared with the non-invasive PA group.

FGFR2 rs2981582 G/A genotype is significantly more frequent (P=0.004) in non-invasive PA compared with the control group.

FGFR2 rs2981582 G/A genotype is significantly less frequent (P=0.009) in invasive PA compared with the non-invasive PA group.

STAT3 rs744166 G/G genotype is significantly less frequent (P<0.001) in invasive PA compared with the control group.

STAT3 rs744166 G/G genotype is significantly less frequent (P=0.038) in invasive PA compared with the non-invasive PA group. PA, pituitary adenoma; SIRT1, sirtuin 1, FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; HWE, Hardy-Weinberg equilibrium.

Frequency of single nucleotide polymorphisms in patients with PA and in the control group according to PA recurrences.

Gene marker Control group, n (%) P-value HWE Non-recurrent PA group, n (%) P-value HWE Recurrent PA group, n (%) P-value HWE
SIRT1 rs12778366
  Genotype
  T/T 647 (80.1) <0.001 91 (81.3) <0.001 25 (80.6) <0.001
  T/C 141 (17.5)a, b 0 (0.0)a 0 (0.0)b
  C/C 20 (2.5)c,d 21 (18.8)c 6 (19.4)d
  Total 808 (100) 112 (100.0) 31 (100.0)
  Allele
  T 1,435 (88.8) 182 (81.3) 50 (80.6)
  C 181 (11.2) 42 (18.8) 12 (19.4)
FGFR2 rs2981582
  Genotype
  G/G 336 (41.6) <0.001 44 (39.3) <0.001 12 (38.7) 0.067
  G/A 429 (53.1) 66 (58.9) 18 (58.1)
  A/A 43 (5.3) 2 (1.8) 1 (3.2)
  Total 808 (100.0) 112 (100.0) 31 (100.0)
  Allele
  G 1,101 (68.1) 154 (68.8) 42 (67.7)
  A 515 (31.9) 70 (31.3) 20 (32.3)
STAT3 rs744166
  Genotype
  G/G 154 (19.1)e <0.001 7 (6.3)e,f 0.083 6 (19.4)f 0.305
  G/A 363 (44.9) 56 (50.0) 12 (38.7)
  A/A 291 (36.0) 49 (43.8) 13 (41.9)
  Total 808 (100.0) 112 (100.0) 31 (100.0)
  Allele
  G 671 (41.5) 70 (31.3) 24 (38.7)
  A 945 (58.5) 154 (68.8) 38 (61.3)

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in non-recurrent PA compared with the control group.

SIRT1 rs12778366 T/C genotype is significantly less frequent (bP=0.005) in recurrent PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P<0.001) in non-recurrent PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P=0.047) in recurrent PA compared with the control group.

STAT3 rs744166 G/G genotype is significantly less frequent (P<0.001) in non-recurrent PA compared with the control group.

FGFR2 rs2981582 G/G genotype is statistically more frequent (P=0.036) in recurrent PA compared with non-recurrent PA. PA, pituitary adenoma; SIRT1, sirtuin 1, FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; HWE, Hardy-Weinberg equilibrium.

Frequency of single nucleotide polymorphisms in patients with PA and in the control group according to PA activity.

Gene marker Control group, n (%) P-value HWE Inactive PA group, n (%) P-value HWE Active PA group, n (%) P-value HWE
SIRT1 rs12778366
  Genotype
  T/T 647 (80.1) <0.001 48 (76.2)   0.007 68 (85.0) 0.044
  T/C 141 (17.5)a,b 0 (0.0)a 0 (0.0)b
  C/C 20 (2.5)c,d 15 (23.8)d 12 (15.0)c
  Total 808 (100.0) 63 (100.0) 80 (100.0)
  Allele
  T 1,435 (88.8) 96 (76.2) 136 (85.0)
  C 181 (11.2) 30 (23.8) 24 (15.0)
FGFR2 rs2981582
  Genotype
  G/G 336 (41.6) <0.001 25 (39.7) <0.001 31 (38.8) 0.005
  G/A 429 (53.1) 38 (60.3) 46 (57.5)
  A/A 43 (5.3) 0 (0.0) 3 (3.8)
  Total 808 (100.0) 63 (100.0) 80 (100.0)
  Allele
  G 1,101 (68.1) 88 (69.8) 108 (67.5)
  A 515 (31.9) 38 (30.2) 52 (32.5)
STAT3 rs744166
  Genotype
  G/G 154 (19.1)e <0.001 6 (9.5) 0.193 7 (8.8)e 0.915
  G/A 363 (44.9) 34 (54.0) 34 (42.5)
  A/A 291 (36.0)f 23 (36.5) 39 (48.8)f
  Total 808 (100.0) 63 (100.0) 80 (100.0)
  Allele
  G 671 (41.5) 46 (36.5) 48 (30.0)
  A 945 (58.5) 80 (63.5) 112 (70.0)

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in inactive PA compared with the control group

SIRT1 rs12778366 T/C genotype is significantly less frequent (P<0.001) in active PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly less frequent (P<0.001) in active PA compared with the control group.

SIRT1 rs12778366 C/C genotype is significantly more frequent (P<0.001) in inactive PA compared with the control group.

FGFR2 rs2981582 G/G genotype is significantly less frequent (P=0.022) in active PA compared with the control group.

FGFR2 rs2981582 A/A genotype is significantly more frequent (P=0.029) in active PA compared with the control group. PA, pituitary adenoma; SIRT1, sirtuin 1, FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; HWE, Hardy-Weinberg equilibrium.

Binomial logistic regression analysis in non-invasive and invasive PA, and the control group.

Gene PA subgroup Model Genotype OR (95% CI) P-value AIC
SIRT1 rs12778366 Non-invasive Co-dominant T/T 1.000 390.145
T/C 0.000 (0.000) 0.996
C/C 7.571 (3.426–16.426) <0.001
Recessive T/T+T/C 1.000 406.092
C/C 9.221 (4.175–20.175) <0.001
Additive – 1.649 (1.065–2.065) 0.025 425.148
Invasive Co-dominant T/T 1.000 509.448
T/C 0.000 (0.000) 0.996
C/C 7.501 (3.715–15.715) <0.001
Recessive T/T+T/C 1.000 533.418
C/C 9.136 (4.528–18.528) <0.001
Additive – 1.616 (1.120–2.120)   0.010 459.515
FGFR2 rs2981582 Non-invasive Co-dominant G/G 1.000 419.170
G/A 2.056 (1.136–3.136) 0.017
A/A 0.000 (0.000) 0.998
Dominant G/G 1.000 425.028
G/A+A/A 1.869 (1.033–3.033) 0.039
Over-dominant G/G+A/A 1.000 420.961
G/A 2.319 (1.283–4.283) 0.005
STAT3 rs744166 Invasive Co-dominant A/A 1.000 553.313
G/A 1.002 (0.630–1.630) 0.993
G/G 0.210 (0.073–0.073) 0.004
Recessive A/A+G/A 1.000 551.313
G/G 0.210 (0.076–0.076) 0.003
Additive – 0.651 (0.467–0.467) 0.011 558.771

PA, pituitary adenoma; SIRT1, sirtuin 1; FGFR2, fibroblast growth factor receptor 2; STAT3, signal transducer and activator of transcription 3; OR, odds ratio; CI, confidence interval; AIC, Akaike Information Criterion.

Binomial logistic regression analysis in inactive and active PA, and control groups.

Gene PA subgroup Model Genotype OR (95% CI) P-value AIC
SIRT1 rs12778366 Inactive Co-dominant T/T 1.000 402.990
T/C 0.000 (0.000) 0.996
C/C 10.109 (4.868–20.868) <0.001
Recessive T/T+T/C 1.000 419.306
C/C 12.312 (5.933–25.933) <0.001
Additive – 2.045 (1.388–3.388) <0.001 444.842
Active Co-dominant T/T 1.000 497.635
T/C 0.000 (0.000) 0.996
C/C 5.709 (2.675–12.675) <0.001
Recessive T/T+T/C 1.000 521.245
C/C 6.953 (3.260–14.260) <0.001
STAT3 rs744166 Active Co-dominant A/A 1.000 535.474
G/A 0.669 (0.430–1.430) 0.148
G/G 0.339 (0.148-.0776) 0.010
Dominant A/A 1.000 536.765
G/A+G/G 0.592 (0.373–0.373) 0.026
Recessive A/A+G/A 1.000 535.576
G/G 0.407 (0.184–0.184) 0.027
Additive – 0.622 (0.442–0.442) 0.007 533.914

PA, pituitary adenoma; SIRT1, sirtuin 1; STAT3, signal transducer and activator of transcription 3; OR, odds ratio; CI, confidence interval; AIC, Akaike Information Criterion.

Binomial logistic regression analysis in non-recurrent and recurrent PA and control groups.

Gene PA subgroup Model Genotype OR (95% CI) P-value AIC
SIRT1 rs12778366 Non-recurrent Co-dominant T/T 1.000 614.042
T/C 0.000 (0.000) 0.995
C/C 7.465 (3.896–14.896) <0.001
Recessive T/T+T/C 1.000 645.812
C/C 9.092 (4.748–17.748) <0.001
Additive – 1.592 (1.153–2.153) 0.005 678.223
Recurrent Co-dominant T/T 1.000 247.718
T/C 0.000 (0.000) 0.996
C/C 7.764 (2.868–21.868) <0.001
Recessive T/T+T/C 1.000 255.407
C/C 9.456 (3.495–25.495) <0.001
STAT3 rs744166 Non-recurrent Co-dominant A/A 1.000 673.559
G/A 0.916 (0.606–1.606) 0.678
G/G 0.270 (0.119–0.119) 0.002
Recessive A/A+G/A 1.000 671.731
G/G 0.283 (0.129–0.129) 0.002
Additive – 0.653 (0.487–0.487) 0.005 677.047

PA, pituitary adenoma; SIRT1, sirtuin 1; STAT3, signal transducer and activator of transcription 3; OR, odds ratio; CI, confidence interval; Akaike Information Criterion.