High Expression of HULC Is Associated with Poor Prognosis in Osteosarcoma Patients

Abstract

Osteosarcoma (OS) is the most common primary bone cancer in childhood. OS is an aggressive disease, and metastatic patients evolve with very poor clinical outcomes. Genetically, OSs are extremely complex tumors, and the related metastatic process is not well understood in terms of the biology of the disease. In this context, long non-coding RNAs (lncRNAs) have emerged as an important class of gene expression regulators that play key roles in the invasion and metastasis of several human tumors. Here, we evaluated the expression of HULC, which is an lncRNA that is associated with the tumor metastatic process, and assessed its potential role as a prognostic marker in OS. HULC expression was evaluated in primary OS samples using real-time RT-PCR. HULC expression status was determined by receiver operating characteristic (ROC) analysis, and its association with survival was assessed using the Kaplan-Meier method. The HULC expression level was not significantly associated with the clinicopathological characteristics of the OS patients. However, our data demonstrated that higher levels of expression of HULC were associated with lower survival rates in OS patients, both in terms of overall and event-free survival. Elevated HULC expression was associated with poor clinical outcomes among the OS patients, which suggests that HULC could be a potential prognostic biomarker in OS.

About the Authors:

Vanessa Regina Maciel Uzan

Affiliation: Pediatric Oncology Laboratory, Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos, SP, Brazil

André van Helvoort Lengert

Affiliation: Pediatric Oncology Laboratory, Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos, SP, Brazil

Érica Boldrini

Affiliation: Barretos Children’s Cancer Hospital, Barretos, SP, Brazil

Valter Penna

Affiliation: Department of Orthopedic Surgery, Barretos Cancer Hospital, Barretos, SP, Brazil

Cristovam Scapulatempo-Neto

Affiliations Pediatric Oncology Laboratory, Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos, SP, Brazil, Department of Pathology, Barretos Cancer Hospital, Barretos, SP, Brazil

Carlos Alberto Scrideli

Affiliation: Department of Pediatrics, Medicine School of Ribeirao Preto, University of Sao Paulo, Ribeirao Preto, SP, Brazil

Alberto Paiva de Moraes Filho

Affiliation: Department of Radiology, Barretos Cancer Hospital, Barretos, SP, Brazil

Carlos Eduardo Bezerra Cavalcante

Affiliation: Department of Radiology, Barretos Children’s Cancer Hospital, Barretos, SP, Brazil

Cleyton Zanardo de Oliveira

Affiliation: Center for Research Support (NAP), Barretos Cancer Hospital, Barretos, SP, Brazil

Luiz Fernando Lopes

* E-mail: [email protected] (DOV); [email protected] (LFL)

Affiliation: Barretos Children’s Cancer Hospital, Barretos, SP, Brazil

Daniel Onofre Vidal

* E-mail: [email protected] (DOV); [email protected] (LFL)

Affiliation: Pediatric Oncology Laboratory, Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos, SP, Brazil

Introduction

Primary bone tumors represent 5–6% of all tumors of childhood and adolescence (0–19 years), and osteosarcomas (OSs) are the most common primary bone tumor in this age group [1]. Approximately 15–20% of patients with OS present with metastasis at diagnosis as detected by radiological examinations such as computerized tomography (CT). Additionally approximately 40% will progress to metastasis in later stages during treatment [2,3]. Metastases are frequent in the lungs (~ 80%) but can also occur in other sites. Metastatic disease sites commonly exhibit resistance to treatment, which results in a low patient survival rate of approximately 20% over five years [3,4]. In Brazil, patients enrolled in clinical studies III and IV of the Brazilian Osteosarcoma Treatment Group (BOTG) exhibited 50.1% and 39% 5-year overall and event-free survival rates, respectively. Additionally, 20.8% of the patients presented with metastases at diagnosis, which reflects a pattern of advanced disease in Brazilian patients [5].

OS treatment involves combined high-dose chemotherapy and resection of the primary tumor and all metastatic disease sites usually via exploratory thoracotomy [6]. The pulmonary recurrence of the disease is frequent and associated with a poor prognosis and a post-relapse 5-year survival of 20% [7]. Thus, the detection of metastasis at diagnosis and the monitoring of metastatic lesions are essential in the treatment of OS. New strategies of treatment have been proposed by the BOTG to improve the survival of OS patients. However, no improvements in survival have been achieved even after the intensification of OS treatment [5].

Genetically, OSs are extremely complex tumors, and despite the knowledge about the molecular changes involved in the pathogenesis of OS [8], metastatic disease is not well understood in terms of the evolution of the disease. Recently, high levels of the expressions of the MAPK7 and MAP2K4 genes have been reported to be significantly associated with metastasis in OS [9].

In this context, long non-coding RNAs (lncRNA) are emerging as important molecular markers of metastatic disease in several human cancers [10–13]. LncRNAs are defined as endogenous RNAs larger than 200 nucleotides that do not contain significant open reading frames (i.e., no open reading frames greater than 100 amino acids) and thus do not result in proteins [14]. LncRNAs comprise a heterogeneous group of RNA molecules that is involved in the regulation of gene expression [15], protein localization [16] and the formation of essential protein complex substructures [17].

Hepatocellular carcinoma up-regulated long non-coding RNA (HULC) is an lncRNA that was first reported to be highly expressed in primary tumors of the liver [18]. Since this report, important roles of HULC have been described in the cellular invasion and metastatic processes of several human tumors [13,19,20].

Evaluations of lncRNA expression in OSs remain rare. Because metastatic disease is a challenge in OS, there is a critical need to identify markers that may indicate or accurately predict the existence of metastatic disease in OS patients. Therefore, we evaluated the expression of HULC in primary osteosarcoma samples at diagnosis and its association with the clinicopathological characteristics of OS Brazilian patients. We also evaluated the role of HULC expression as a potential prognostic marker in OS.

Material and Methods

The study was conducted following national and institutional ethical policies, and it was previously approved by the Barretos Cancer Hospital Ethical Committee and the Medicine School of Ribeirão Preto/USP Ethical Committee (protocol CAAE 13671113.0.0000.5437). The ethical committees classified this study has having minimal risk, ensuring confidentiality, and not resulting in any clinical influences due to changes in treatment or genetic counseling for the participants and their families. For these reasons, both ethical committees waived the need for consent.

Patient samples

Fresh-frozen osteosarcoma tissue samples from 33 patients were retrieved from the Biobank of the Barretos Cancer Hospital and the Department of Pediatrics at the Ribeirão Preto School of Medicine, University of São Paulo. All patients were diagnosed between 2006 and 2013, and the samples were taken from primary tumor biopsies prior the use of any possible systemic treatment and without metastasis at diagnosis. All patients were treated following the new clinical protocol of the Brazilian Osteosarcoma Treatment Group, which is currently referred to as the Latin American Group (GLATO) 2006. In the GLATO 2006 study, initially non-metastatic patients at diagnosis undergo the standard regimen of Cisplatin and Doxorubicin (60 mg/m²/day and 37.5 mg/m²/day, respectively; D1-D2; weeks 1 and 6) followed by Methotrexate (12 g/m²/day; D1; weeks 4–5 and 9–10). Surgery of primary tumor is proposed in week 11–12. After surgery, patients were randomized into a first arm treatment comprising four more cycles of chemotherapy, as described above, for a total of 31 weeks of treatment; and, in a second arm, in which after 31 weeks (same as first arm) Cyclophosphamide (25 mg/m²/day; every day) and Methotrexate (1.5 mg/m²/2x day; D1 and D4, every week) were administered for a total of 104 weeks of treatment.

RNA extraction and cDNA synthesis

The OS samples were re-evaluated by a pathologist to confirm the diagnoses and to guarantee at least 80% tumor cellularity prior to RNA extraction. Total RNA was extracted from the OS samples using a standard protocol with TRIZOL reagent (Invitrogen, Grand Island, NY, USA). The homogenization of the tumor cells in TRIZOL was performed using Precellys 24 (Bertin Technologies, Rockville, MD, USA). RNA was quantified using Qubit 2.0 (Life Technologies, Grand Island, NY, USA), and quality was assessed using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). We considered samples with a RIN ≥ 4 for our analysis.

The RNA was submitted to treatment with deoxyribonuclease I (Sigma-Aldrich, St. Louis, MO, USA) following the manufacturer’s recommendations. cDNA synthesis was performed with 1 μg of DNA-free total RNA using the High-Capacity cDNA Reverse Transcription Kit (Life Technologies, Grand Island, NY, USA) according to the manufacturer's recommendations.

Real-time quantitative PCR analysis of gene expression

The expression analysis of HULC was performed by RT-qPCR using PrimeTime qPCR assays (Integrated DNA Technologies, Coralville, Iowa, USA). PrimeTime assays contain probes and primers that allow for the specific amplification and evaluation of the expression of the gene of interest. For HULC, we used assay Hs.PT.58.3962267.

RT-qPCR was performed with 1 μL of the cDNA, 1x Master Mix Fast Probe KAPA (Kapa Biosystems, Wilmington, MA, USA), and 1x PrimeTime qPCR assay (50X) in a final volume of 20 μL. The reaction was performed as follows: 95°C for 10 minutes; 40 cycles at 95°C for 15 seconds; and 60°C for 1 minute. Additionally, we evaluated the expressions of the endogenous genes GAPDH and ACTB to normalize the expression data.

The normalized expressions of the gene of interest were calculated with the 2^-ΔCt mathematical model wherein ΔCt corresponds to the Ct of the gene of interest subtracted from the Ct of the endogenous gene [21]. The expression analysis was performed with the 7900 HT Fast Real-Time PCR System (Applied Biosystems, Grand Island, NY, USA), and all reactions were performed in duplicate.

Clinical characteristics and statistical analysis

The epidemiological variables considered in the present study were the following: sex and age at diagnosis, and the clinicopathological variables of OS histological type, Huvos grading, tumor localization, tumor size and the occurrence of metastasis during treatment. The categorical variables are expressed as frequencies and percentages.

To initially categorize HULC expression, a cut-off value was determined using receiver operating characteristic (ROC) analysis. The cut-off value was determined as the value that maximized the sensitivity and specificity of the discrimination between non-metastatic patients and patients who experienced metastasis during treatment. Tumors with HULC expression levels below the cut-off were classified as low expression tumors, and those with HULC expression levels above the cut-off were classified as high expression. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy and the respective 95% confidence intervals were estimated. Furthermore, the gene expression levels (i.e., low or high expression) were correlated with the clinicopathological characteristics using the chi-square test or Fisher's exact test.

Overall survival and event-free survival were determined with the Kaplan-Meier method. For overall survival, death from any cause was considered an event, and the follow-up was determined from the date of diagnosis to the date of death. For event-free survival, the occurrence of metastasis during treatment and death were considered events, and the follow-up was determined from the date of diagnosis to the date of the event that occurred first. The differences between the survival curves were analyzed using the log-rank test.

For all analyses, statistical significance was considered at p<0.05. The statistical analyses were performed with SPSS v.21 software.

Results

Associations between HULC expression and the clinicopathological characteristics of the OS patients

The clinicopathological characteristics of the OS population are summarized in Table 1. We determined a HULC expression cut-off value to discriminate the patients with good outcomes (no event occurrences) from the patients with poor outcomes (occurrence of any event during the follow-up) using ROC analysis (Table 2). Based on this cut-off value, the HULC expressions were classified as low or high for the subsequent analysis.

[Figure omitted. See PDF.]

Table 1. Clinicopathological characteristics of the osteosarcoma population.

https://doi.org/10.1371/journal.pone.0156774.t001

[Figure omitted. See PDF.]

Table 2. HULC expression cut-off based on predictive probability of clinical outcome among the OS patients.

https://doi.org/10.1371/journal.pone.0156774.t002

We observed that HULC expression was not significantly associated with the evaluated clinicopathological characteristics of the osteosarcoma patients (Table 3).

[Figure omitted. See PDF.]

Table 3. Associations between HULC expression and clinicopathological characteristics of the OS patients.

https://doi.org/10.1371/journal.pone.0156774.t003

Association of HULC expression with the prognoses of the osteosarcoma patients

The patients with higher HULC expression presented 5-year overall survival rate of 26.0% compared with 75.4% for those with low expression (p = 0.016). Similar results were observed for the EFS; i.e., the patients with high HULC expression presented a 5-year EFS rate of 37.5% compared with 64.5% for those with low expression (p = 0.047; Fig 1).

[Figure omitted. See PDF.]

Fig 1. Kaplan-Meier survival curves of the OS patients according to HULC expression.

A: overall survival, B: event-free survival.

https://doi.org/10.1371/journal.pone.0156774.g001

Regarding the evaluated clinical characteristics, our data revealed that the histological type of OS significantly influenced the EFS rate. The patients with the osteoblastic subtype presented a 5-year EFS of 69.8% compared with 21.4 for the other OS histologies (p = 0.047).

The COX regression analysis of the EFS revealed that HULC expression and histological type were independent prognostic factors for the OS patients (Table 4). These findings suggest that HULC expression is a biomarker of poor prognoses in OS patients and highlight the role of HULC in the biology of the disease.

[Figure omitted. See PDF.]

Table 4. COX regression model of the event-free survival for the OS patients.

https://doi.org/10.1371/journal.pone.0156774.t004

Discussion

Recent estimates suggest that there are more than 20,000 lncRNA transcripts in the human genome, which implies that the lncRNA class represents a large but unknown component of normal cellular networks that may be altered in tumor cells. Therefore, it is reasonable to assume that lncRNA deregulation is involved in the development of many diseases, including cancer [22]. Indeed, lncRNAs are often expressed in tissue-specific patterns; thus, these molecules are attractive new therapeutic targets [23,24].

OS is a complex genetic disease and despite all of the progress achieved by basic and clinical research in recent years, the molecular genetic mechanisms involved in the pathogenesis of osteosarcoma are not yet clear. LncRNA expression has not been well explored in OS, and our knowledge of how lncRNAs act in OS cells, particularly in terms of driving metastatic processes, remains very limited. Because the presence of metastatic disease at diagnosis drives OS patients to more aggressive treatment approaches [5], the identification of patients with non-metastatic statuses at diagnosis who will develop metastatic lesions during treatment is of significant interest. The identification of potential metastatic OSs at diagnosis even when clinical metastasis is not present based on results from standard imaging techniques could be a promising tool in the management of OS patients. In our work, we evaluated the expression of HULC, which is a lncRNA that is associated with the metastatic processes of tumors [13,18–20] in OS pediatric primary samples from patients without metastatic disease at diagnosis.

Initially, HULC was described as a liver-specific non-coding RNA that is highly expressed in primary tumors of the liver [18]. Subsequently, high levels of HULC were detected in the plasma of hepatocellular carcinoma patients and found to be associated with tumor aggressiveness and progression. Based on these data, some authors have suggested that HULC is a potential prognostic biomarker in liver cancer [19]. Moreover, high levels of HULC expression have been described in liver metastases in colorectal carcinoma patients [25], and its expression has been associated with lymph node metastasis and vascular infiltration in pancreatic cancer [13] and advanced metastatic disease in gastric cancer [20].

HULC has been demonstrated to act as a microRNA (miRNA) sponge that binds to and reduces the expression of a number of miRNAs, including miR-372. The reduction of miR-372 expression leads to increased expression of its target PRKACB, which in turn induces CREB phosphorylation. This process results in alterations in the patterns of the deacetylation and methylation of histones, which leads to chromatin remodeling and influences the expression of a series of genes [26].

We found no significant associations between HULC expression and the clinicopathological characteristics of the OS patients, which was probably due to our limited sample. Due to the rarity of this disease, the majority of the studies that have aimed to examine the molecular changes that occur in osteosarcomas have encountered restricted sample sizes [9,27,28]. Although, higher expression of HULC is not significantly associated with development of metastasis during treatment in our work, probably due to the number of patients (10) included in this group, our data demonstrated that elevated HULC expression is associated with poor prognoses among OS patients and influences both overall and event-free survival. Additionally, our observations regarding EFS, in which we considered the occurrence of metastasis during treatment and death as events, allow us to suggest that the high expression of HULC could be associated with primary tumors with a more aggressive behavior and/or metastatic potential.

Recently, elevated expression of HULC was associated with clinical stage and distant metastasis in OS. Additionally, elevated HULC expression has been associated with poorer overall survival in these patients [29], which corroborates our findings. Moreover, these authors demonstrated that the inhibition of HULC using a siRNA technique results in decreases in the proliferation, migration and invasion capacities of OS cell lines [29]. The results obtained in our work, considering a Brazilian OS population with a different genetic background from previous study [29] supports that in general HULC is a potential molecular marker of prognosis in OS. Also, our findings strengthen the observation that HULC seems to have a pivotal role in OS tumorigenesis.

In conclusion, our data suggest that HULC expression is associated with poor clinical outcomes in OS patients and that HULC could be a potential biomarker of the prognosis and occurrence of metastasis in OS. Further evaluations of larger samples are needed to allow for an extensive exploration of the role of HULC in osteosarcoma biology.

Acknowledgments

The authors acknowledge the Center for Research Support (NAP), the Barretos Children’s Cancer Hospital, and the Pio XII Foundation.

Author Contributions

Conceived and designed the experiments: DOV. Performed the experiments: VRMU AHL. Analyzed the data: VRMU AHL EB CZO LFL DOV. Contributed reagents/materials/analysis tools: VP CAS. Wrote the paper: VRMU AHL EB CSN LFL DOV. Evaluated the imaging exams: APMF CEBC.

Citation: Uzan VRM, Lengert AvH, Boldrini É, Penna V, Scapulatempo-Neto C, Scrideli CA, et al. (2016) High Expression of HULC Is Associated with Poor Prognosis in Osteosarcoma Patients. PLoS ONE 11(6): e0156774. https://doi.org/10.1371/journal.pone.0156774

References

1. Mirabello L, Troisi RJ, Savage SA. Osteosarcoma incidence and survival rates from 1973 to 2004: data from the Surveillance, Epidemiology, and End Results Program. Cancer. 2009;115(7):1531–43. pmid:19197972; PubMed Central PMCID: PMCPMC2813207.

2. Bielack SS, Kempf-Bielack B, Delling G, Exner GU, Flege S, Helmke K, et al. Prognostic factors in high-grade osteosarcoma of the extremities or trunk: an analysis of 1,702 patients treated on neoadjuvant cooperative osteosarcoma study group protocols. J Clin Oncol. 2002;20(3):776–90. pmid:11821461.

3. Harting MT, Blakely ML. Management of osteosarcoma pulmonary metastases. Semin Pediatr Surg. 2006;15(1):25–9. pmid:16458843.

4. Bacci G, Rocca M, Salone M, Balladelli A, Ferrari S, Palmerini E, et al. High grade osteosarcoma of the extremities with lung metastases at presentation: treatment with neoadjuvant chemotherapy and simultaneous resection of primary and metastatic lesions. J Surg Oncol. 2008;98(6):415–20. pmid:18792969.

5. Petrilli AS, de Camargo B, Filho VO, Bruniera P, Brunetto AL, Jesus-Garcia R, et al. Results of the Brazilian Osteosarcoma Treatment Group Studies III and IV: prognostic factors and impact on survival. J Clin Oncol. 2006;24(7):1161–8. pmid:16505436.

6. Kager L, Zoubek A, Pötschger U, Kastner U, Flege S, Kempf-Bielack B, et al. Primary metastatic osteosarcoma: presentation and outcome of patients treated on neoadjuvant Cooperative Osteosarcoma Study Group protocols. J Clin Oncol. 2003;21(10):2011–8. pmid:12743156.

7. Briccoli A, Rocca M, Salone M, Guzzardella GA, Balladelli A, Bacci G. High grade osteosarcoma of the extremities metastatic to the lung: long-term results in 323 patients treated combining surgery and chemotherapy, 1985–2005. Surg Oncol. 2010;19(4):193–9. pmid:19515554.

8. Martin JW, Squire JA, Zielenska M. The genetics of osteosarcoma. Sarcoma. 2012;2012:627254. pmid:22685381; PubMed Central PMCID: PMCPMC3364016.

9. Tesser-Gamba F, Petrilli AS, de Seixas Alves MT, Filho RJ, Juliano Y, Toledo SR. MAPK7 and MAP2K4 as prognostic markers in osteosarcoma. Hum Pathol. 2012;43(7):994–1002. pmid:22154052.

10. Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long non-coding RNA HOTAIR reprograms chromatin state to promote cancer metastasis. Nature. 2010;464(7291):1071–6. pmid:20393566; PubMed Central PMCID: PMCPMC3049919.

11. Guo F, Li Y, Liu Y, Wang J, Li G. Inhibition of metastasis-associated lung adenocarcinoma transcript 1 in CaSki human cervical cancer cells suppresses cell proliferation and invasion. Acta Biochim Biophys Sin (Shanghai). 2010;42(3):224–9. pmid:20213048.

12. Iacoangeli A, Lin Y, Morley EJ, Muslimov IA, Bianchi R, Reilly J, et al. BC200 RNA in invasive and preinvasive breast cancer. Carcinogenesis. 2004;25(11):2125–33. pmid:15240511.

13. Peng W, Gao W, Feng J. Long noncoding RNA HULC is a novel biomarker of poor prognosis in patients with pancreatic cancer. Med Oncol. 2014;31(12):346. pmid:25412939.

14. Chen LL, Carmichael GG. Long noncoding RNAs in mammalian cells: what, where, and why? Wiley Interdiscip Rev RNA. 2010;1(1):2–21. pmid:21956903.

15. Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, et al. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell. 2007;129(7):1311–23. pmid:17604720; PubMed Central PMCID: PMCPMC2084369.

16. Willingham AT, Orth AP, Batalov S, Peters EC, Wen BG, Aza-Blanc P, et al. A strategy for probing the function of noncoding RNAs finds a repressor of NFAT. Science. 2005;309(5740):1570–3. pmid:16141075.

17. Clemson CM, Hutchinson JN, Sara SA, Ensminger AW, Fox AH, Chess A, et al. An architectural role for a nuclear noncoding RNA: NEAT1 RNA is essential for the structure of paraspeckles. Mol Cell. 2009;33(6):717–26. pmid:19217333; PubMed Central PMCID: PMCPMC2696186.

18. Panzitt K, Tschernatsch MM, Guelly C, Moustafa T, Stradner M, Strohmaier HM, et al. Characterization of HULC, a novel gene with striking up-regulation in hepatocellular carcinoma, as noncoding RNA. Gastroenterology. 2007;132(1):330–42. pmid:17241883.

19. Xie H, Ma H, Zhou D. Plasma HULC as a promising novel biomarker for the detection of hepatocellular carcinoma. Biomed Res Int. 2013;2013:136106. pmid:23762823; PubMed Central PMCID: PMCPMC3674644.

20. Zhao Y, Guo Q, Chen J, Hu J, Wang S, Sun Y. Role of long non-coding RNA HULC in cell proliferation, apoptosis and tumor metastasis of gastric cancer: a clinical and in vitro investigation. Oncol Rep. 2014;31(1):358–64. pmid:24247585.

21. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. pmid:11846609.

22. Lipovich L, Johnson R, Lin CY. MacroRNA underdogs in a microRNA world: evolutionary, regulatory, and biomedical significance of mammalian long non-protein-coding RNA. Biochim Biophys Acta. 2010;1799(9):597–615. pmid:20951849.

23. Fu X, Ravindranath L, Tran N, Petrovics G, Srivastava S. Regulation of apoptosis by a prostate-specific and prostate cancer-associated noncoding gene, PCGEM1. DNA Cell Biol. 2006;25(3):135–41. pmid:16569192.

24. Amaral PP, Neyt C, Wilkins SJ, Askarian-Amiri ME, Sunkin SM, Perkins AC, et al. Complex architecture and regulated expression of the Sox2ot locus during vertebrate development. RNA. 2009;15(11):2013–27. pmid:19767420; PubMed Central PMCID: PMCPMC2764477.

25. Matouk IJ, Abbasi I, Hochberg A, Galun E, Dweik H, Akkawi M. Highly upregulated in liver cancer noncoding RNA is overexpressed in hepatic colorectal metastasis. Eur J Gastroenterol Hepatol. 2009;21(6):688–92. pmid:19445043.

26. Wang J, Liu X, Wu H, Ni P, Gu Z, Qiao Y, et al. CREB up-regulates long non-coding RNA, HULC expression through interaction with microRNA-372 in liver cancer. Nucleic Acids Res. 2010;38(16):5366–83. pmid:20423907; PubMed Central PMCID: PMCPMC2938198.

27. Zhang Q, Geng PL, Yin P, Wang XL, Jia JP, Yao J. Down-regulation of long non-coding RNA TUG1 inhibits osteosarcoma cell proliferation and promotes apoptosis. Asian Pac J Cancer Prev. 2013;14(4):2311–5. pmid:23725133.

28. Fellenberg J, Bernd L, Delling G, Witte D, Zahlten-Hinguranage A. Prognostic significance of drug-regulated genes in high-grade osteosarcoma. Mod Pathol. 2007;20(10):1085–94. pmid:17660802.

29. Sun XH, Yang LB, Geng XL, Wang R, Zhang ZC. Increased expression of lncRNA HULC indicates a poor prognosis and promotes cell metastasis in osteosarcoma. Int J Clin Exp Pathol. 2015;8(3):2994–3000. pmid:26045809; PubMed Central PMCID: PMCPMC4440118.