Long Non-coding Rna As Predictors Of Cisplatin Response In Ovarian Cancer

A retrospective cohort study was conducted on 150 of Formalin fixed paraffin embedded tissue of ovarian cancer. The enrolled ovarian cancer patients have treated by intravenous platinum compound based therapy either in the form of “Cisplatin or Carboplatin” for three to four cycles. As specimens were retrospectively collected; patients were selected based on their response to standard regimen of Cisplatin based chemotherapy into Cisplatin sensitive and resistant. Fifty non-cancerous begin ovarian lesions were analysed for normalization of lnc_RNAs expression. Ovarian cancer specimens were categorized according to the clinical and histopathological features into standard and high risk groups.

The study was approved by the ethical committee of Ain-Shams University Hospitals; the study was conducted according to the World Medical Association Declaration of Helsinki. A written consent was signed from each participant in whom was assured that all information generated in this study remained confidential. A unique identification number was assigned to each patient at the time of enrolment.

Bioinformatics

We searched for the most linked Lnc-RNAs to Cisplatin resistance in cancer; accordingly, target gene modulation, and drug pharmacogenomics or chemoresistance pathways were identified using different databases such as NRDT [46] and Pharmaco-miR [47]. With the aim to select three Lnc_RNAs that are related to Cisplatin resistance while, they interlink to each other; we collect all information about LncRNA-target gene-drugs. There are of large number and therfore data was very difficult to interpret in a database. Therefore we decided to study the RNA-drug interactions using a network-based approach. Basically, we searched in PubMed –NCBI (National Center for Biotechnology Information) for all recent studies (published from 2015 onwards) that investigating the LncRNAs involved in drug chemoresistance.

The result of this screening was manually curated in order to avoid and remove papers with generic statements and not direct links between ncRNAs and drugs. Only the analysis that proved (by in vitro/in vivo) experiments the existence of a direct association between ncRNAs and chemoresistance were then analysed.We thus built a network of three long non-coding RNAs that are linked to Platinum analogues resistance and furthermore interlinked to each other’s by alternating pathways. We obtained a fully connected network of three drug/ncRNA interactions “PVT1, TUG1 and MEG3) (Image 1). A graph theory measures were considered to define the interlink between non-coding RNAs and Cisplatin therapeutics resistance. The selected LncRNAs pathways in the network were based on closeness centrality and shortest path.

Finally, we performed a structure analysis using Cystoscope (48) to identify different clusters of ncRNAs and drugs. The clusters were presented by visual graph (Image 1) for convenient visualization.

Total RNA purification and reverse transcription from FFPE Tissue Sections

Deparafinatization of FFPE tissues was performed using the deparaffinization Solution (Cat no: 19093), then the total RNA was extracted using RNeasy FFPE Kit (cat no: 73504). The procedure was conducted according to the manufacturer’s protocol (Qiagen, Hilden, Germany). Extracted RNA purity was assessed by measuring the optic density at 260 and 280 nm using UV spectrophotometer (Eppendorf, Germany). Total RNA were reverse transcribed in a total volume of 20 μl using the QuantiTect RT kit according to the manufacturer’s instructions (Qiagen, Hilden, Germany). The reaction protocol was adjusted at 37°C for 60 min, then at 95°C for 5 min.

Lnc_RNAs expression analysis by real time PCR

The cDNA was amplified using specific primers sequences for the PVT1, TUG1 and MEG3; the used primer sequences are [Hs_PVT1; ID: LPH17013A, Hs_TUG1 ID: LPH18394A and Hs_MEG3 RT2; ID: LPH02974A primer assays cat no: 330701, respectively]; Qiagen, Hilden, Germany. The Hs_GAPDH_1_SG QuantiTect Primer Assay (GAPDH) cat no: 249900, ID: QT00079247 was used as housekeeper gene. The reaction mix and cycling protocols were adjusted according to the manufacturer instruction using using the the 5 plex Rotor Gene PCR Analyzer (Qiagen, Hilden, Germany). The average expression level of all markers was also used to perform data normalization. The RQ of gene expression was determined using the comparative ΔΔCt method.

Statistical analysis

The results were presented as mean ± standard deviation for normally distributed continuous variables and as median and interquartile range (IQR) for not normally distributed one. Categorical variables were presented as frequencies and percentages. We used Mann Whitney test for non-parametric comparisons between the two groups. We used Kaplan Meier Survival analysis for comparison of survival between different groups. The cut-off values were determined using the median level of long non-coding RNA. Receiver operator curves were used to evaluate the prognostic accuracy. The survival between different groups was compared using log rank test. Statistical analysis was performed by IBM SPSS Statistics V23

Discussion

This study was set out to identify the significance of the long non-coding RNA assay in ovarian cancer. In this study, we revealed three significant markers for ovarian cancer survival and its resistance to Cisplatin. Based on our analysis, the level of the TUG, PVT and MEG3 in the ovarian cancer patients were more than healthy control (P-value = 0.001). Moreover, we found significant difference in the levels of the three long non conding RNA between the Cisplatin resistant and sensitive cases. There was a significant increase in the log level of PVT1 and TUG in resistant cases more than sensitive cases (median = 11.3, range (4.6-36.2), P-value = 0.001) and (median = 77.7, range (22.7 -191), P-value = 0.02). In contrast, MEG3 had significant higher levels in Cisplatin sensitive population more than Cisplatin resistant population (P-value = 0.004). Moreover, the levels of TUG, PVT and MEG3 were affected by the grade of the tumour, pathology and metastasis.

For overall survival, we found that only high levels of Lnc-PVT and Lnc-TUG1 more than the cut-off values had better survival (MST = 25.8, 95% CI [23-29], P-value =0.01) and (MST = 30, 95% CI [30-31], P-value =0.001), respectively. Furthermore, For progression free survival, Lnc-TUG1 and Lnc-MEG3 had significantly higher MST than other cases for Lnc-TUG1 (MST = 27, 95% CI [26-29], P-value =0.01) and (MST = 27, 95% CI [24-30], P-value =0.02).

This is considered the first study to assess the three long noncoding RNA in ovarian cancer and understand their influence on the Cisplatin resistant cases and their survival. In literature, there was no reports regarding the prevalence of Cisplatin resistant ovarian tumours. However, the worse prognosis and recurrence associated with these cases caught the attention to stand on the causes of this problem [26, 28].

Based on our results, there was increased expression of TUG1, MEG3 and PVT in ovarian cancer compared to healthy control. The PVT overexpression in the tumour cells were evident in many tumours not only ovarian tumours and it is considered as oncogene [33, 36]. A study in ovarian cell tumour found that it enhanced the tumour progression and invasion. Moreover, it caused Cisplatin resistance through inhibition of apoptosis [34]. Based on our study, high expression of PVT was associated with better survival despite its effect on chemotherapy resistance. In our study, its level was affected by pathological type and grade. studies found that the level of PVT is correlated with tumour staging and invasions, thus, can be used as a tumour marker [37]. Moreover, it was found to be associated with lower overall survival and progression free survival in epithelial ovarian carcinoma which is inconsistent with our results [38]. Another study found the same results which is inconsistent with our results [39]. Furthermore, it could predict the relapse in ovarian cancer patients into three groups: high, mixed and low risk [38].

For TUG1, it was found that it promotes cell proliferation and invasion [40]. Moreover, it was found that it mediated the resistance to Cisplatin which is consistent with our study. However, we could not find in literature a study that assessed its relation to ovarian cancer survival. Based on its mechanism for inducing cell proliferation and resistance, our results seem inconsistent with the reported mechanisms as the high TUG1 levels were associated with better survival. In other tumours, TUG1 level was associated with shorter survival and Cisplatin resistance through its effect on EZH2 and apoptotic pathways [40-43].

High expression of MEG3 was associated with better survival in our study which is consistent with the reported mechanisms in literature. The MEG3 was used as a target in ovarian cancer for decreasing Cisplatin resistant. This is consistent with our study as we found a significant lower level in Cisplatin resistant cases compared to sensitive ones. Moreover, in another study, it upregulated p53, β-catenin and survivin causing apoptosis of tumour cells thus enhancing survival [44]. In lung cancer patients, it was found to be correlated with chemo sensitive response [45]. However, it is not well established as a tumour marker in ovarian carcinoma.

In conclusion, based on our results and proposed mechanisms in literature, more studies are needed to identify the PVT, TUG1 and MEG3 as a biomarker for ovarian cancer and cisplatin resistance.

Conclusion

TUG, PVT1 and MEG3 can be used as a biomarker for cisplatin resistance in ovarian cancer. Their accuracy as a predictor for resistance and survival are high. However, their use as an indicator for overall survival needs more assessment as it is contradicting to its molecular mechanisms in our study.