Methylation status as a predictor of intravesical Bacillus Calmette-Guérin ( BCG ) immunotherapy response of high grade non-muscle invasive bladder tumor

Background and Aims. Genetic and epigenetic alterations play an important role in urothelial cancer pathogenesis. Deeper understanding of these processes could help us achieve better diagnosis and management of this life-threatening disease. The aim of this research was to evaluate the methylation status of selected tumor suppressor genes for predicting BCG response in patients with high grade non-muscle-invasive bladder tumor (NMIBC). Materials and Methods. We retrospectively evaluated 82 patients with high grade non-muscle-invasive bladder tumor (stage Ta, T1, CIS) who had undergone BCG instillation therapy. We compared epigenetic methylation status in BCGresponsive and BCG-failure groups. We used the MS-MLPA (Methylation-Specific Multiplex Ligation-Dependent Probe Amplification probe sets ME001 and ME004. The control group was 13 specimens of normal urotel (bladder tissue)). Results. Newly identified methylations in high grade NMIBC were found in MUS81a, NTRK1 and PCCA. The methylation status of CDKN2B (P=0.00312**) and MUS81a (P=0.0191*) is associated with clinical outcomes of BCG instillation therapy response. CDKN2B and MUS81a unmethylation was found in BCG failure patients. Conclusion. The results show that the methylation status of selected tumor suppressor genes (TSGs) has the potential for predicting BCG response in patients with NMIBC high grade tumors. Tumor suppressor genes such as CDKN2b, MUS81a, PFM-1, MSH6 and THBS1 are very promising for future research.


INTRODUCTION
Bladder Cancer (BC) is the second most common malignancy of the urinary tract and the seventh most common cancer in men and the 17 th in women 1 .BC is nearly three times more common in men that in women 2 .The median age at the time of diagnosis is around 70 years for each gender.BC has two categories.Approximately 75% of patients with BC present with disease confined to the mucosa or submucosa (non muscle invasive bladder tumor -NMIBC).Carcinoma in situ (CIS) is a very specific subgroup of NMIBC, because it is not a papillary lesion but a flat tumor, which is why CIS can be missed in cystoscopy 3 .CIS is always a high grade tumor.Without any treatment, approximately 54% of patients with CIS progressed to muscle-invasive or metastatic disease 4 .
The second category covers patients with muscle-invasive BC (MIBC).These patients have a higher prevalence of progression rates and higher cancer-specific mortality 5 .Patients with NMIBC are indicated to transurethral tumor resection (TUR) alone or with adjuvant treatment (intravesical chemotherapy or intravesical Bacillus Calmette-Guérin (BCG) immunotherapy).Indication for adjuvant therapy is determined in the European Association of Urology Guidelines (Guidelines on NMIBC).
BCG is an attenuated mycobacterium developed as a vaccine for tuberculosis that has demonstrated antitumor activity in BC intravesical instillation, and significantly reduces the progression risk of high grade papillary lesion and CIS.
The pathologic characteristics and tumor grade are very important.From the standpoint of genetic development, biological behavior, and management, high grade and low grade cancers may be regarded as essentially separate diseases 6 .The therapy of MIBC is radical cystectomy -hard mutilating surgery with urinary diversion.This procedure has significant impact on the quality of the patient's life.
Patients with superficial (NMIBC) high grade tumor and CIS are the focus of our interest.It is a borderline subgroup, where radical and non-radical treatment are both possible.Non-radical therapy means complete tumor resection with adjuvant intravesical immunotherapy (BCG vaccine).
Radical therapy means radical cystectomy.At present we have no markers which would be helpful in deciding optimal therapy.Weighing the risk of failure of non-radical treatment against overtreatment by radical therapy constantly presents a dilemma.
The etiology of BC is multifactorial, driven by the multistep accumulation of environmental, genetic and epigenetic factors.Epigenetic analyses include DNA methylation, microRNAs, and histone modification.Methylation status differences were evaluated in our study subgroups.
DNA methylation is a covalent chemical modification by addition of a methyl group at the carbon-5 position of the cytosine ring.Most cytosine methylation occurs in the sequence context 5'-CG-3'.Methylation is mediated by the DNA cytosine methyltransferases.Changed methylation in the transcribed region has a variable effect on gene expression.A new model for the mechanism of carcinogenesis has been proposed in which hypermethylation of unmethylated cytosine-phosphate-guanine (CpG) islands in the promoter regions of tumor suppressor genes (TSG) in normal cells silence these genes and this leads to the cells becoming cancerous 7,8 .

Patients
We retrospectively evaluated 82 patients with high grade NMIBC (stage Ta, T1, CIS) who had undergone BCG instillation therapy in our urology department.Patients were divided to two groups: BCG response and BCG failure.BCG response we defined as absence of recurrence of, progression of, or death from the disease during follow-up.Our conditions were: complete deep TUR with the presence of detrusor muscle in the specimen; no previous intravesical therapy; complete 6-weekly induction of BCG instillation therapy followed by maintenance BCG therapy just 12 months in patients in this study.BCG therapy was terminated in cases of failure.Schedules of maintenance therapy are unclear in general.Our local schedule covers three weekly doses every three months if there in no reccurence according to cystoscopy.Minimal follow up was 2 years.These conditions were satisfied in 66 patients.Patients' tumor tissues were retrieved from the pathology archive and prepared for DNA isolation and methylation analysis.
The control group comprised 13 specimens of normal urotel (bladder tissue) obtained during cystectomy for benign (fistulas) or malignant (tumor) diagnosis.
The study was approved by the Ethics Committee of the University Hospital Hradec Kralove.

Tumor samples
Formalin-fixed and paraffin-embedded (FFPE) samples of bladder cancer were retrieved from the archive of the department of pathology.All slides were reviewed by a pathologist.

Control samples
Normal urothelium was exfoliated from the urinary bladder wall specimen.The inclusion condition was nor-mal urothelium macroscopically and microscopically evaluated by an experienced pathologist.Only one third specimens could be used for analysis according to these strict criteria.Informed consent for bladder tissue manipulation was obtained before surgery.Control samples were harvested from healthy urothelium according to the strict rules of the Local Ethics Committee which is why the group is limited.

DNA isolation
DNA was extracted from FFPE samples using a Qiagen DNA extraction kit (Hilden, Germany) according to the manufacturer's protocol with minimum modification.The procedure consists of 6 steps: 1. Removing paraffin: paraffin is removed by dissolving out with xylene; 2. Lysis: the sample is lysed under denaturing conditions with proteinase K (56 °C, overnight); 3. Heating: 10 min incubation at 70 °C reverses formalin crosslinking; 4. Binding: DNA bids to the membrane and contaminants flow through; 5.Washing: residual contaminants are washed away; 6. Elution: pure, concentrated DNA is eluted from the membrane.The concentration of isolated DNA was measured by two techniques: fluorimetric (Qubit, Invitrogen) and spectrophotometric (Nanodrop ND -1000, Thermo Fisher Scientific).

Methylation-Specific Multiplex Ligation-Dependent Probe Amplification (MS-MLPA)
The present study used the MS-MLPA probe sets ME001 and ME004, which can simultaneously check for aberrant methylation in several tumor suppressor genes (Tables 1, 2).Probe sequences, gene loci and chromosome locations can be found at http://www.mlpa.com.Individual genes were evaluated by two probes, which recognized different Hha1 restriction sites in their regions.The experimental procedure was carried out according to the manufacturer's instructions, with minor modifications.
In short, DNA (100 ng) was dissolved up to 5 μL in TE-buffer (10 mM Tris•Cl; 0.5 mM EDTA; pH 9.0), denatured and subsequently cooled down to 25 °C.After adding the probe mix, the probes were allowed to hybridize (overnight at 60 °C).Subsequently, the samples were divided into two: in one half, the samples were directly ligated, while for the other half ligation was combined with the HhaI digestion enzyme.This digestion resulted in ligation of the methylated sequences only.PCR was performed on all the samples using a standard thermal cycler (GeneAmp 9700, Applied Biosystems), with 35 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s and extension at 72 °C for 1 min.with a final extension of 20 min at 72 C. Aliquots of 0.6 μL of the PCR reaction were combined with 0.2 μL LIZ-labeled internal size standard (Applied Biosystems, Foster City, CA, USA), and 9.0 μL deionized formamide.After denaturation, fragments were separated and quantified by electrophoresis on an ABI 3130 capillary sequencer and analyzed using GeneMapper4.0(both Applied Biosystems).Peak identification, and values corresponding to peak size in base pairs (bp) and peak areas, were used for further data processing.Methylation dosage ratio was obtained by the following calculation: Dm = (P x /P ctrl )Dig/ (P x /P ctrl )Undig, where Dm is the methylation dosage ratio, P x is the peak area of a given probe, P ctrl is the sum of the peak areas of all control probes, Dig stands for HhaI digested sample and Undig for undigested sample.Dm can vary between 0 and 1.0 (corresponding to 0-100% of methylated DNA).
Based on previous experiments, we considered a promoter to show methylation if the methylation dosage ratio was ≥ 0.10, which corresponds to 20% of methylated DNA (ref. 9,10).CpG universal methylated and unmethylated DNA (Zymoresearch, Irvine, CA, USA) were used in every run as controls.

Statistical analysis
We used the nonparametric Mann-Whitney test for evaluation of differences in age and scores.Fisher's exact two proportions test was used for differences in genes.Associations with P-value <0.05 were considered to be significant.We used Bonferroni modification of significance level.
Both patient groups and the control group were homogeneous without statistically significant variation.We evaluated age, gender, tumor size, smoking, CIS presence, recurrence and progression score in time resection (Table 3).There was no statistically relevant difference in gender, tumor size, CIS presence or smoking between the two BCG groups.The recurrence score was higher in the BCG failure group in time TUR (P=0.0109).The progression score was not statistically different between the BCG groups (P=0.188).

Methylation results
Methylation analysis (complete or partial) was performed on 57 patients with cancer and 13 control patients.DNA isolation was unsuccessful in 9 patients, and hence methylation analysis could not be performed.There was a BCG response in 41 patients and BCG failure in 16 patients (Table 4).
Methylation status in CDKN2B (P=0.00312** ) and MUS81a (P=0.0191*) is associated with the clinical outcomes of BCG instillation therapy response.CDKN2B and MUS81a unmethylation was found in BCG failure patients.

DISCUSSION
Our study results confirm the benefit of BCG instillation in patients with high grade NMIBC and CIS.BCG instillation is standard therapy for these patients.Metaanalyses have confirmed that BCG instillation after TUR is superior to TUR alone or TUR and chemotherapy [11][12][13][14] .BCG therapy reduces relative progression risk by about 27% (ref. 15).Without any treatment, approximately 54% of patients with CIS progressed to muscle-invasive disease 4 .Radical cystectomy is the sole radical procedure.BCG progression score in time TUR correlates with BCG failure in clinical practice.The outcome of the BCG progression score may have been affected by the small sample in our study.Toxicity and an optimal BCG schedule are limitations of BCG therapy.BCG intravesical treatment is associated with numerous side effects (cystitis, hematuria, prostatitis, epididymo-orchitis, arthralgia, fever, sepsis, allergic reactions)(ref. 16).Serious side effects however are described in < 5% of patients 17 .The optimal BCG instillation schedule is unknown.The general consensus is 6-weekly induction of BCG instillation introduced by Morales 18 , but the optimal number of inductions and the optimal frequency and duration of maintenance instillations remain unknown 19 .Although BCG has good clinical outcomes in general, we have many patients with BCG failure.This problematic group is the focus of our interest.Clinical markers for the prediction of BCG response or failure are missing.BCG failure patients spend a lot of time in conservative procedures and radical therapy is delayed.This could be fatal in terms of the patient's oncology prognosis 20 .
2][23][24][25][26][27][28][29][30] ).Methylation status is a potent indicator for distinguishing patients responding to BCG from those who are failing to do so and who need the radical therapy approach 31,32 .We used the MS-MLPA probe sets ME001 and ME004 (MRC-Holland, Amsterdam, The Netherlands) selected by an experienced biochemist, suitable for identification of new typical bladder cancer methylation.These MS-MLPA sets have not been reported for predicting BCG response.
CDKN2B is cyclin-dependent kinase inhibitor 2B.This gene lies adjacent to the tumor suppressor gene CDKN2A in a region that is frequently mutated and deleted in a wide variety of tumors.This gene encodes a cyclin-dependent kinase inhibitor, which forms a complex with CDK4 or CDK6, and prevents the activation of the CDK kinases, and thus the encoded protein functions as a cell growth regulator that controls cell cycle G1 progression.The expression of this gene was found to be dramatically induced by TGF beta, which suggested its role in TGF beta-induced growth inhibition.Two alternatively spliced transcript variants of this gene, which encode distinct proteins, have been reported (provided by RefSeq, Jul 2008).Methylation of CDKN2B has been found in several cancer types including laryngeal squamous cell carcinoma, acute myeloid leukemia, and in ovarian cancer, where the presence of methylation in the CDKN2B gene was present mainly in clear cell carcinoma [34][35][36] .
MUS81 (Structure-Specific Endonuclease Subunit) is a protein coding gene.The MUS81 protein belongs to a conserved family of DNA structure-specific nucleases that play important roles in DNA replication and repair.Xing et al. 2015 demonstrated an essential role for the MUS81 nuclease in the maintenance of replication fork integrity 37 .
Unmethylation of CpG islands in CDKN2B especially and MUS81a TSGs is connected with BCG failure in our study.The mechanism by which unmethylation of CDKN2b and MUS81a favor the BCG failure is unknown.
The literature is sparse on data that predict of BCG response in the context of methylation status.Polyaminemodulated factor-1 (PMF-1) methylation status is described by Alvarez-Múgica as statistically relevant for prediction of BCG response in patients with high grade NMIBC (ref. 32).Agundez published combination methylation statuses for TSGs MSH6 and THBS1 (MS-MLPA ME 002 probe set), with prediction of progression of high risk NMIBC after BCG instillation therapy 31 .

CONCLUSION
Genetic and epigenetic alterations play an important role in urothelial cancer pathogenesis.Deeper understanding of these processes could help us towards better diagnosis and management of this life threatening disease.Of epigenetic changes hypermethylation, hypomethylation and unmethylation are prospective research topics.According to our research results we can conclude that the methylation status of selected TSGs has the potential for predicting BCG response in patients with NMIBC high grade tumors, but cannot yet be safely relied on for common clinical application.TSGs such as PMF-1, MSH6, THBS1, CDKN2b and MUS81a are very promising for future research.

Table 3 .
Demographic characteristics of the subgroups.