Towards personalised treatment of patients with colorectal liver metastases Hof, Joost
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Hof, J. (2019). Towards personalised treatment of patients with colorectal liver metastases. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the
number of authors shown on this cover page is limited to 10 maximum.
Low tumour expression of miR-19b and miR-196b
predicts a good survival after surgery for colorectal liver metastases
J. Hof
1,2, K.P. de Jong
1, R.H. Sijmons
2, K. Kok
21
Department of Hepato-pancreato-biliary surgery and Liver transplantation, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands.
2
Department of Genetics, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands.
Submitted manuscript
Abstract
Background: Colorectal liver metastases (CRLM) can be cured by surgical resection, but we still lack reliable patient-specific prognostic markers. As microRNAs have been shown to play a role in tumourigenesis and cancer progression, we aimed to identify microRNAs that could serve as potential biomarkers for patient survival and provide support for personalised treatment and follow up after liver surgery.
Methods: MicroRNA expression profiles of resected CRLM of two patient groups were analysed: poor survivors (death from recurrent disease <30 months after surgery) and good survivors (no recurrent disease >60 months after surgery). The expression data was then validated by qPCR assays.
Results: MiR-17-3p, miR-19b and miR-196b were more highly expressed in the poor survivors, and miR-19b and miR-196b were validated by qPCR. In multivariable analysis correcting for clinical characteristics, high expression of miR-19b was independently associated with a poor survival after surgery for CRLM (p=0.037). Although miR-196b was not independently associated with survival (p=0.064), it is complementary to miR-19b:
patients with both low miR-196b and low miR-19b expression always had a good survival.
Discussion: MiR-19b and miR-196b are potential new tissue biomarkers to predict patient
survival after surgery for CRLM.
Introduction
Colorectal carcinoma is the cancer with the second highest incidence and mortality rate in Europe [1]. In the full course of the disease, around 50% of all patients develop liver metastases (CRLM). Recent advances in the treatment of CRLM have extended treatment options to increase curability [2], and many patients now undergo potentially curative resection with five year overall survival rates that range from 15-60%. This large range in survival rates can be partially explained by clinicopathological tumour characteristics [3]. A more accurate survival prediction can be useful for personalised treatment, for example in the administration of adjuvant chemotherapy. Currently, in technically resectable CRLM, the recommendation is to administer adjuvant chemotherapy to patients with unfavourable clinical prognostic features, albeit with little clinical benefit [2]. This indicates that the group of patients with a good survival after resection alone will not benefit from adjuvant chemotherapy and thus can be spared the associated morbidity.
Improving survival prediction might help in better selecting patients for studies on experimental adjuvant therapies and, vice versa, might guide decisions not to offer current chemotherapy. To improve the identification of these patients, we aimed to predict patient survival by studying microRNA expression in surgically resected CRLM.
MicroRNAs are small non-coding 18-22nt RNAs involved in post-transcriptional regulation of gene expression, primarily by silencing mRNA [4]. It is believed that microRNAs have a prominent role in initiation, progression and dissemination of CRLM and are therefore potential biomarkers for prognostication [5]. In the present study we analyse the association between microRNA expression and patient survival. To do this, we generated microRNA expression profiles of tumour specimens from two patient groups at the extremes of survival outcome.
Methods
Patient samples
Patients were selected from a prospectively maintained database. Inclusion criteria were (1) R0 partial liver resection for CRLM, (2) no neoadjuvant chemotherapy, (3) a Fong clinical risk score [6] of 3 or lower, (4) no detectable extrahepatic disease at time of surgery, (5) no other known malignant disease, and (6) availability of fresh frozen (-80°C) resected CRLM material. Follow-up consisted of imaging and measurement of CEA serum levels every 3-4 months during the first 2 years after liver resection and at 6 monthly intervals
4
afterwards up to 5 years. We included samples of two groups of patients with extreme survival rates: poor survivors, who died of recurrent disease within 30 months after partial liver resection, and good survivors, who showed no evidence of recurrent disease up to 60 months after liver resection. The tumour samples were reviewed by an experienced pathologist to make sure vital tumour was present.
RNA/DNA isolation
Both genomic DNA and total RNA were isolated from the same 10 μm-tissue sections (RNA/DNA purification kit, Norgen Biotek Corporation (Thorold, Ontario, Canada)). DNA/
RNA isolation was performed according to the manufacturer’s protocol. Quality check and RNA quantification of samples was carried out by capillary electrophoresis using the LabChip GX (Perkin Elmer (Waltham, Massachusetts, USA)).
MicroRNA expression
MicroRNA expression profiles were generated with the SurePrint G3 unrestricted miRNA 60k microarray (ID-070156, Agilent, Santa Clara, California, USA). This array contains 2549 human microRNAs based on the miRBase database (release 21.0). We restricted our analysis to the human microRNAs indicated as confident in miRBase (ftp://mirbase.org/
pub/mirbase/22). Out of the 1198 confident human microRNAs, 1144 were present on the SurePrint G3 unrestricted miRNA microarray. Experimental procedures were performed according to the manufacturer’s instructions, omitting optional steps (spike-in, purification of labelled RNA). After hybridization and washing, slides were scanned on the Agilent DNA Microarray Scanner. Agilent Feature extraction software was used to digitalize the fluorescent signals. The corresponding grid file was 070156_D_F_20141006. Data analysis was performed using Genespring 14.8 (Agilent), and values were log-transformed and normalized by the percentile shift (90%) method. MicroRNAs were included in further statistical testing if at least 80% of the normalized values were higher than -3 in one of the two survival groups (supplementary figure 1).
MicroRNA quantitative RT-PCR
MicroRNA expression levels were measured using the Taqman MicroRNA Reverse
Transcription Kit (Applied Biosystems, Foster City, California, USA) with an input of 10
ng total RNA. Supplementary table 1 lists the TaqMan primer sets used in the reverse
transcription (RT) reaction, namely RNU44, hsa-miR-196b, hsa-miR-17 and hsa-miR-
19b (Thermo Fisher, Waltham, Massachusetts, USA). Before the actual analysis four
reference genes (RNU24, RNU44, RNU48 and RNU49) were tested by qPCR on CRLM
samples and RNU44 was selected as the most stable (data not shown). qPCR was done in triplo on an Applied Biosystems™ Quantstudio™ 7, and data-analysis performed using the corresponding software. The cycle threshold (Ct) was recorded and a good technical replicate was considered if the triplo Ct values were within 0.5 from each other. If 2 out of 3 Ct values were within 0.5 Ct of each other, the mean of those Ct values was calculated and applied in further calculations. ∆Ct values were calculated using RNU44 as the reference gene.
Microsatellite instability and mutational hotspots
All DNA samples were screened for microsatellite instability and for the most important somatic mutations in CRLM, namely KRAS (codon 12 and 13) and BRAF (V600E).
Microsatellite instability was tested by amplifying 20 ng genomic DNA using primers for 5 mononucleotide microsatellite loci (NR21, NR24, BAT25, BAT26, MONO27). The resulting PCR products were analysed on the ABI 3730xl DNA Analyzer (Thermo Fisher). The MSI status was assessed as MSI-instable when two out of five markers showed instability. For assessment of mutations in KRAS (codon 12 and 13) and BRAF (V600E), genomic DNA was amplified by PCR and the resulting amplicons were analysed by Sanger sequencing. Primer specifics are shown in supplementary table 1.
Statistical analysis
Summary statistics were obtained using established methods and presented as percentages, median (interquartile range, IQR) or mean (±standard deviation, SD).
Variables associated with survival were tested by univariable and multivariable binary logistic regression analyses. Variables with a p-value <0.1 in univariable analysis were entered into the multivariable model. Odds ratio and 95% confidence intervals (CI) were estimated, and a p-value <0.05 was considered significant. Statistical analyses were carried out with IBM SPSS Statistics V22 (IBM, Armonk, New York, USA).
Results
Clinicopathological characteristics
The clinical characteristics of the 44 patients enrolled in this study are displayed in table 1, stratified by survival. Poor survivors had a median follow-up of 15.6 months (IQR 13.7-21.8) and good survivors a median follow-up of 106.1 months (IQR 85.5-123.2). There was no single clinicopathological characteristic that was significantly different between the two
4
survival groups (table 1). However, the multiparameter clinical risk score (CRS) defined by Fong et al. [6] showed a higher risk score in the poor survivors (p=0.016). The molecular characteristics MSI-status, KRAS (codon 12 and 13) and BRAF (V600E) mutations were not different between the poor and good survivors (table 1).
Table 1 Clinical characteristics of the total study population
Poor survival (n=20) Good survival (n=24) P value Patient characteristics
Mean age at liver surgery 62.2 ± 11.0 63.1 ± 7.5 0.637
Male sex 13 (65%) 9 (37.5%) 0.069
Tumour characteristics
Major liver surgery (≥ 3 segments) 17 (85%) 16 (66.7%) 0.162
Size largest CRLM (in cm) 7.7 ± 5.1 4.6 ± 2.6 0.075
Rectal primary tumour 9 (45%) 6 (25%) 0.163
Neoadjuvant chemotherapy 0 0 -
Clinical risk score (CRS)
CRS = 3 (high score) 9 (45%) 3 (12.5%) 0.016
Interval CRLM >12 months 11 (55%) 14 (58.3%) 0.824
CEA >200 mg/μl 5/18 (27.8%) 2/23 (8.7%) 0.107
More than 1 CRLM 6 (30%) 6 (25%) 0.711
CRLM larger than 5cm 10 (50%) 9 (37.5%) 0.405
N+ primary tumour 13 (65%) 11 (45.8%) 0.204
Molecular characteristics
Microsatellite instable (MSI-high) 2 (10%) 1/23 (4.2%) 0.501
KRAS codon 12/13 mutation 8 (40%) 9 (37.5%) 0.865
BRAF V600E mutation 0 0 -
CRLM = colorectal liver metastases, CEA = carcinoembryonic antigen, N+ = lymph node positive.
MicroRNA expression profiles
Twenty-three samples were subjected to microRNA expression profiling: 12 from the good survival group and 11 from the poor survival group. Supplementary figure 1 shows the normalized expression values in all 1144 microRNAs stratified by survival. After excluding miRNA’s with a low expression, 247 out of 1144 microRNA’s entered statistical testing.
Three microRNAs were significantly different expressed between the survival groups.
MiR-196b-5p had the highest fold change (fc 8.3, higher expression in poor survivors) with
a p-value= 0.012. Similarly, miR-122-5p (fc 4.0, p=0.026) and miR-17-3p (fc 4.2, p=0.019)
were more highly expressed in the poor survivors. After multiple testing correction, none
of these microRNAs remained significant. We excluded miR-122-5p from subsequent validation since this microRNA was previously reported as being liver-specific [7]. Indeed, in our series, miR-122-5p showed a high correlation (r=0.864) with the percentage of Albumin mRNA reads in our RNA samples (as determined by RNAseq, unpublished data).
Of note, we excluded samples with an Albumin mRNA percentage higher than 1.6%. We performed logistic regression modelling to identify microRNAs that could predict survival and logically included miR-196b-5p and miR-17-3p. When we included miR-19b-3p in this model survival was correctly predicted in 19 out of 23 patients (supplementary table 2).
Of note, expression levels of miR-17-3p, mir-19b-3p and miR-196b-5p were not correlated with Albumin expression (-0.201 < R < -0.030).
Quantitative RT-PCR
When validating the microarray data, we included an additional 21 patients using the same inclusion criteria: 12 good survivors and 9 poor survivors. The array data for miR-196b- 5p, miR-19b-3p, and miR-17-3p were validated by qPCR for all 44 patients. We obtained reliable qPCR results for 19 patients previously subjected to microarray analysis and for the additional 21 patients. Supplementary figure 2A-C shows the correlation between the qPCR data and microarray data. Of note, the qPCR data of miR-196b and miR-19b were correlated (r=0.653; supplementary figure 2D). The total poor and good survivor groups consisted of 17 and 23 patients, respectively. Poor survivors tended to have a higher miR-196b expression compared to the good survivors (ΔCt values -0.206 (±1.133) vs. 1.008 (±2.184); p=0.071), in agreement with the microarray data. Similarly, miR-19b expression also tended to be higher in the poor survivors vs. good survivors (ΔCt values -2.317 (±0.744) vs. -1.549 (±1.522); p=0.058). In contrast, the expression of miR-17 did not show any differences between the patient survival groups (p=0.170).
Multivariable analysis
For statistical modelling, we dichotomized the microRNA ΔCt values based on the median value. By using a threshold of p <0.10 in univariable testing, three clinical variables and microRNAs 19b and 196b were carried over to the multivariable analysis (table 2). Because the variable ‘high clinical risk score’ was significantly different between the survival groups, we did not select individual variables that are part of the CRS for the regression model to prevent overfitting (table 2). Multivariable analysis showed that a high expression of miR-19b (p=0.037) is independently associated with a poor patient survival (table 2).
Although miR-196b had a p-value >0.05 in multivariable testing, it does add value in predicting patient survival. Patients with both low miR-196b and low miR-19b expression
4
always had a good survival (p=0.001; table 3). The regression model displayed in table 2 has a Nagelkerke R² of 0.557. Strikingly, when we exclude the microRNAs as variables, the Nagelkerke R² drops to 0.350. The multivariable analysis shows that in addition to a high miR-19b expression, male sex and a high CRS also were independent predictors of a poor patient survival (table 2). Out of the 15 patients with a low CRS and a low miR-19b expression, 12 patients had a good survival. Similarly, out of the five patients with a high CRS and a high miR-19b expression, all five patients had a poor survival.
Table 2 Multivariable analysis
Univariable Multivariable
Factors P value OR (95% CI) P value OR (95%CI)
Patient characteristics
Mean age at liver surgery 0.985 0.999 (0.934-1.069)
Male sex 0.023 5.056 (1.248-20.480) 0.032 8.016 (1.198-53.642)
Tumour characteristics
Major liver surgery (≥ 3 segments) 0.238 2.489 (0.548-11.308) Size largest CRLM (in cm) 0.030 1.278 (1.024-1.596)
Rectal primary tumour 0.088 3.187 (0.842-12.072) 0.121 4.180 (0.687-25.428) Clinical risk score (CRS)
CRS = 3 (high score) 0.052 4.667 (0.990-22.008) 0.036 15.143 (1.199-191.203) Interval CRLM <12 months 0.896 0.918 (0.256-3.298)
CEA >200 mg/μl 0.351 2.500 (0.364-17.173)
More than 1 CRLM 0.531 1.545 (0.396-6.035)
CRLM larger than 5cm 0.435 1.667(0.463-6.006) N+ primary tumour 0.292 2.000 (0.552-7.251) Molecular characteristics
Microsatellite instable (MSI-high) 0.397 2.933 (0.244-35.329) KRAS codon 12/13 mutation 0.804 0.848 (0.231-3.114) MicroRNA qPCR
High miR-196b expression (> median) 0.009 6.500 (0.594-26.511) 0.064 5.686 (0.906-35.665)
High miR-19b expression (> median) 0.029 4.800 (1.251 – 18.421) 0.037 7.548 (1.130-50.414)
CRLM = colorectal liver metastases, CEA = carcinoembryonic antigen, N+ = lymph node positive, OR = Odds
ratio, CI = confidence interval
Table 3 Expression of miR-19b and miR-196b.
Both miR-19b and miR-196b
expression is low miR-19b or miR-196b
expression is low Both miR-19b and miR-196b expression is high Total
Good survivors 11 7 5 23
Poor survivors 0 11 6 17
A high expression is defined as a Ct-value below the median value of the particular microRNA marker.
Discussion
In this study we have shown that a low expression of both miR-196b and miR-19b is associated with a good survival after surgery for CRLM. This suggests that measuring the expression levels of these microRNAs might improve the accuracy of the currently available clinical models in predicting patient survival after surgery for CRLM [6]. To our knowledge, there are eight studies in CRLM that correlate high-throughput microRNA expression data to patient survival, of which seven validated their observations by qPCR [8–15]. None of these studies except one showed an association between miR-196b and miR-19b expression and patient survival [8–14]. The one exception, Li et al. [15], reported an association between high tumour expression of miR-196b-5p and favourable patient survival. This is an opposing survival outcome to what we find in our study, even though both studies had comparable number of patients (48 vs. 40) and used similar TaqMan miRNA expression assays (Applied Biosystems, Foster City, California, USA). The difference might be explained by differences in the study populations. We selected patients at the extremes of survival outcome while Li et al. did not. In addition, Li et al. did not report on neoadjuvant chemotherapy or showed a multivariable analysis to correct for clinicopathological factors [15]. Kahlert et al. showed that high expression of miR-19b in the liver invasion front (extending 10 cell rows into the liver), but not in the tumour, was associated with an unfavourable patient survival [8].
Evidence from a number of earlier studies appears to support a biological relation between miR-196b and miR-19b and poor prognosis. Proposed mRNA targets of miR- 196b in colorectal cancer are FAS, which might promote cancer by repressing apoptosis [16], and GATA6, which is a regulator of genes relevant to tumourigenesis and cancer progression [17]. Furthermore, miR-196b-5p is thought to interact with HOXB7 and GALNT5 to stimulate the development of metastases in colorectal cancer [18]. Ren et al. showed the influence of miR-196b in the JAK2/STAT3 pathway and linked miR-196b-5p to 5-FU
4
chemoresistance in vivo [19]. In their mouse study, tumours overexpressing miR-196b-5p showed continual tumour growth while receiving 5-FU chemotherapy. In contrast, the tumour volume in mice treated with silenced miR-196b-5p and 5-FU chemotherapy did not increase over time [19].
With respect to the biological role of miR-19b in CLRM prognosis, it has been identified as one of the major oncomiRs in the 17-92 cluster, a group of microRNAs that share functions in tumourigenesis, tumour invasion and metastasis. The 17-92 cluster is known to be involved in activation of the JAK-STAT pathway and the PI3K/AKT signaling pathway, and has been reported to promote epithelial-to-mesenchymal transition [20]. Two groups studying colorectal cancer cell lines identified miR-19b as an important player in chemoresistance [21,22]. Jiang et al. proposed oxaliplatin-based chemoresistance through SMAD4 targeting, suggesting that miR-19b induces a reduction in tumour cell apoptosis through TGF-beta signalling [22].
In conclusion, we have shown that a low expression of miR-196b and miR-19b in chemo- naïve CRLM tumour tissue is associated with a favourable survival after surgery for CRLM.
Previous studies had identified both these microRNAs as oncogenes and our data support these findings [16–18,20]. Intriguingly, both high expression of miR-196b as well as miR-19 is associated with chemoresistance in primary colorectal cancer [19,21,22]. Future research should aim to prove whether high miR-19b/196b expression, poor patient survival and chemoresistance are all measurements of the same aggressive tumour biology.
Acknowledgements
We thank Kate McIntyre for assistance in the editorial process.
References
[1] Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JWW, Comber H, et al. Cancer incidence and mortality patterns in Europe: Estimates for 40 countries in 2012. Eur J Cancer 2013;49:1374–403. doi:10.1016/j.ejca.2012.12.027.
[2] Van Cutsem E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, Aderka D, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol 2016;27:1386–
422. doi:10.1093/annonc/mdw235.
[3] Simmonds PC, Primrose JN, Colquitt JL, Garden OJ, Poston GJ, Rees M. Surgical resection of hepatic metastases from colorectal cancer: A systematic review of published studies. Br J Cancer 2006;94:982–
99. doi:10.1038/sj.bjc.6603033.
[4] Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004;116:281–97.
[5] Wang D, Liu J, Huo T, Tian Y, Zhao L. The role of microRNAs in colorectal liver metastasis:
Important participants and potential clinical significances. Tumor Biol 2017;39:101042831770964.
doi:10.1177/1010428317709640.
[6] Fong Y, Fortner J, Sun RL, Brennan MF, Blumgart LH. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999;230:309- 318; discussion 318-321. doi:10.1097/00000658-199909000-00004.
[7] Coulouarn C, Factor VM, Andersen JB, Durkin ME, Thorgeirsson SS. Loss of miR-122 expression in liver cancer correlates with suppression of the hepatic phenotype and gain of metastatic properties.
Oncogene 2009;28:3526–36. doi:10.1038/onc.2009.211.
[8] Kahlert C, Klupp F, Brand K, Lasitschka F, Diederichs S, Kirchberg J, et al. Invasion front-specific expression and prognostic significance of microRNA in colorectal liver metastases. Cancer Sci 2011;102:1799–807.
doi:10.1111/j.1349-7006.2011.02023.x.
[9] Pizzini S, Bisognin A, Mandruzzato S, Biasiolo M, Facciolli A, Perilli L, et al. Impact of microRNAs on regulatory networks and pathways in human colorectal carcinogenesis and development of metastasis.
BMC Genomics 2013;14:589. doi:10.1186/1471-2164-14-589.
[10] Manceau G, Imbeaud S, Thiébaut R, Liébaert F, Fontaine K, Rousseau F, et al. Hsa-miR-31-3p expression is linked to progression-free survival in patients with KRAS wild-type metastatic colorectal cancer treated with anti-EGFR therapy. Clin Cancer Res 2014;20:3338–47. doi:10.1158/1078-0432.CCR-13- 2750.
[11] Ellermeier C, Vang S, Cleveland K, Durand W, Resnick MB, Brodsky AS. Prognostic microRNA expression signature from examination of colorectal primary and metastatic tumors. Anticancer Res 2014;34:3957–
67.
[12] Li W, Chang J, Wang S, Liu X, Peng J, Huang D, et al. miRNA-99b-5p suppresses liver metastasis of colorectal cancer by down-regulating mTOR. Oncotarget 2015;6:24448–62. doi:10.18632/
oncotarget.4423.
[13] Pecqueux M, Liebetrau I, Werft W, Dienemann H, Muley T, Pfannschmidt J, et al. A Comprehensive MicroRNA Expression Profile of Liver and Lung Metastases of Colorectal Cancer with Their Corresponding Host Tissue and Its Prognostic Impact on Survival. Int J Mol Sci 2016;17. doi:10.3390/
ijms17101755.
4
[14] Kingham TP, Nguyen HCB, Zheng J, Konstantinidis IT, Sadot E, Shia J, et al. MicroRNA-203 predicts human survival after resection of colorectal liver metastasis. Oncotarget 2017;8:18821–31. doi:10.18632/
oncotarget.13816.
[15] Li W, Chang J, Tong D, Peng J, Huang D, Guo W, et al. Differential microRNA expression profiling in primary tumors and matched liver metastasis of patients with colorectal cancer. Oncotarget 2017;8:35783–91.
doi:10.18632/oncotarget.16206.
[16] Mo J-S, Alam KJ, Kang I-H, Park WC, Seo G-S, Choi S-C, et al. MicroRNA 196B regulates FAS-mediated apoptosis in colorectal cancer cells. Oncotarget 2015;6:2843–55. doi:10.18632/oncotarget.3066.
[17] Fantini S, Salsi V, Reggiani L, Maiorana A, Zappavigna V. The miR-196b miRNA inhibits the GATA6 intestinal transcription factor and is upregulated in colon cancer patients. Oncotarget 2017;8:4747–59.
doi:10.18632/oncotarget.13580.
[18] Stiegelbauer V, Vychytilova-Faltejskova P, Karbiener M, Pehserl A-M, Reicher A, Resel M, et al. miR- 196b-5p Regulates Colorectal Cancer Cell Migration and Metastases through Interaction with HOXB7 and GALNT5. Clin Cancer Res 2017;23:5255–66. doi:10.1158/1078-0432.CCR-17-0023.
[19] Ren D, Lin B, Zhang X, Peng Y, Ye Z, Ma Y, et al. Maintenance of cancer stemness by miR-196b-5p contributes to chemoresistance of colorectal cancer cells via activating STAT3 signaling pathway.
Oncotarget 2017;8:49807–23. doi:10.18632/oncotarget.17971.
[20] Fang L-L, Wang X-H, Sun B-F, Zhang X-D, Zhu X-H, Yu Z-J, et al. Expression, regulation and mechanism of action of the miR-17-92 cluster in tumor cells (Review). Int J Mol Med 2017;40:1624–30. doi:10.3892/
ijmm.2017.3164.
[21] Kurokawa K, Tanahashi T, Iima T, Yamamoto Y, Akaike Y, Nishida K, et al. Role of miR-19b and its target mRNAs in 5-fluorouracil resistance in colon cancer cells. J Gastroenterol 2012;47:883–95. doi:10.1007/
s00535-012-0547-6.
[22] Jiang T, Ye L, Han Z, Liu Y, Yang Y, Peng Z, et al. miR-19b-3p promotes colon cancer proliferation and
oxaliplatin-based chemoresistance by targeting SMAD4: validation by bioinformatics and experimental
analyses. J Exp Clin Cancer Res 2017;36:131. doi:10.1186/s13046-017-0602-5.
Su pp le m en ta ry t ab le 1 . P rim er s pe ci fic s. M ic ro RN A m ar ke rs Ta qm an a ss ay s eq ue nc es RN U 44 ( as sa y 0 01 09 4) CC TG G AT G AT G ATA G CA A AT G CT G AC TG A AC AT G A AG G TC TTA AT TA G CT CTA AC TG AC T hs a- m iR -1 96 b ( as sa y 0 02 21 5) U AG GU AGUU U CC U GU U GU U G G G hs a- m iR -1 7 ( as sa y 0 02 42 1) AC U G CAG U G A AG G CA CU U G U AG hs a- m iR -1 9b ( as sa y 0 00 39 6) U G U G CAAA U CC AU G CAAAA CU G A Fo rw ar d p rim er Re ve rs e p rim er Mi cr os at elli te N R21 TA A AT G TAT G TC TC CC CT G G AT TC CTA CT CC G CAT TC AC A N R24 CC AT TG CT G A AT TT TA CC TC AT TG TG CC AT TG CAT TC CA A M ONO 27 CA CT CC AG CG TGGG AG AC AG GG TGG AT CA A AT TT CA CT TGG BAT 25 TC G CC TC CAA G AA TG TAA G T TC TG CAT TT TA AC TAT G G CT C BAT 26 TG AC TA CT TT TG AC TT CA G CC TA AC CA TT CA AC AT TT TT A AC CC M ut ati on h ot sp ot KR AS c od on 1 2 a nd 1 3 CG AT AC ACG TC TG CA G TC A A G A AT G G TC CT G CA CC AG TA A BR AF V 60 0E ATA AT G CT TG CT CT G ATA G G TG TG A ATA CT G G G A AC TAT G
4
Supplementary table 2. Microarray expression data and logistic regression modelling.
Patient miR-196b-5p miR-17-3p miR-19b-3p Survival Logistic regressiona
21 -6.78 -6.78 3.1 Good -64.27
7 0.92 -7.03 3.35 Good -55.07
23 2.1 -7.22 2.21 Good -45.33
3 -8.65 -0.12 4.88 Good -12.52
1 -3.1 -2.78 2.59 Good -11.19
16 -0.05 -2.39 3.51 Good -9.31
22 1.04 -1.42 4.3 Good -3.78
14 -3.7 -0.09 4.8 Good -2.57
13 1.59 -1.22 4.32 Good -0.85
8 0.92 -1.95 3.24 Good -0.65
10 1.09 -1.6 3.68 Poor -0.36
17 2.57 -1.39 4.22 Poor -0.02
5 0.1 -0.58 4.68 Poor 0.16
6 0.48 -0.31 5.01 Good 0.91
11 0.54 -1.31 3.76 Poor 1.02
4 2.12 -1.01 4.46 Poor 1.15
9 2.64 -1.75 3.63 Good 1.29
15 1.97 -0.64 4.8 Poor 1.91
18 -0.25 -1.71 2.79 Poor 3.55
19 2.07 -1.05 4.07 Poor 3.92
12 4.28 -0.61 4.87 Poor 5.81
2 3.36 0.84 5.68 Poor 12.59
20 4.12 1.33 5.96 Poor 16.77
The normalized expression data, survival, and logistic regression modelling results
aare shown for all 23 patients with microarray expression data.
a
