Long non-coding RNAs in colorectal cancer: Novel oncogenic mechanisms and promising clinical applications

Abstract

Colorectal cancer (CRC) is the third most common malignancy and ranks as the second leading cause of cancer-related deaths worldwide. Despite the improvements in CRC diagnosis and treatment approaches, a considerable proportion of CRC patients still suffer from poor prognosis due to late disease detection and lack of personalized disease management. Recent evidences have not only provided important molecular insights into their mechanistic behavior but have also indicated that identification of cancer-specific long non-coding RNAs (LncRNAs) could benefit earlier disease detection and improve treatment outcomes in patients suffering from CRC. LncRNAs have raised extensive attention as they participate in various hallmarks of CRC. The mechanistic evidence gleaned in the recent decade clearly reveals that lncRNAs exert their oncogenic role by regulating autophagy, epigenetic modifications, enhancing stem phenotype and modifying tumor microenvironment. In view of their pleiotropic functional role in malignant progression, and their frequently dysregulated expression in CRC patients, they have great potential to be reliable diagnostic and prognostic biomarkers, as well as therapeutic targets for CRC. In the present review, we will focus on the oncogenic roles of lncRNAs and related mechanisms in CRC as well as discuss their clinical potential in the early diagnosis, prognostic prediction and therapeutic translation in patients with this malignancy.

1. Background

Colorectal cancer (CRC) remains the third most common malignancy, and ranks as the second leading cause of cancer-related deaths worldwide[1]. Surgical resection, chemotherapy and radiotherapy are currently the primary therapeutic strategies for the management of CRC patients, among which surgical resection is preferentially considered for patients with an earlier-stage cancer while chemoradiotherapy is a treatment modality often offered in later disease stages[2]. Despite encouraging advances in the earlier disease diagnosis and improved treatment options, approximately 40–50% of patients with a stage II and III CRC suffer from poor survival outcomes due to a variety of reasons including inadequacy of the currently-available treatments and lack of availability of adequate risk-assessment biomarkers[3]. Classic serological biomarkers, including CEA, CA19–9, CA724 and CA125, are often routinely used for the early diagnosis for CRC, however, their sensitivity and specificity is far from satisfactory[4]. Hence, there is an urgent need for the identification of novel and reliable biomarkers for the individualized management of CRC, which potentially would lead to an improved overall outcome from this disease.

Noncoding RNAs (ncRNAs) represent a class of RNAs that do not directly participate in the transcription of mRNAs, but indirectly influence gene regulation. With the routine application of sophisticated next generation sequencing approaches, the identification of novel classes of ncRNAs continue to evolve each day. Presently, the major categories of such RNA include transfer RNAs (tRNAs), ribosomal RNAs (rRNAs), small nucleolar RNAs (snoRNAs), microRNAs (miRNAs) and long noncoding RNAs (lncRNAs)[5]. The LncRNAs are characterized as a group of ncRNAs that are more than 200 nucleotides long and exert their gene regulatory function through a variety of cellular processes such as chromatin modification, transcriptional regulation and post-transcriptional alterations[6]. Increasing evidence suggests that lncRNAs play an important role in tumor progression[7], metastasis[8], autophagy[9] and chemoresistance[5] via multiple mechanisms. Furthermore, in the recent years, lncRNAs have garnered increasing attention as potential clinical biomarkers for the early cancer screening, determining patient prognosis, and predicting response to various drug treatments[6]. A recent review has provided various strategies to develop lncRNA-based targeted drugs in pancreatic ductal adenocarcinoma (PDAC) such as small interfering RNA, CRISPR-Cas9 and antisense oligonucleotides[10]. In breast cancer, emerging evidence has highlighted the crucial role of exosomal lncRNAs in drug resistance, firmly supporting their therapeutic potential[11]. Taken together, due to their vital role in cancer biology and therapy, a deep and comprehensive understanding of lncRNAs will undoubtedly benefit the biomarker-directed precision medicine in cancer patients.

In the present review, we firstly focus on the biological functions and related oncogenic mechanisms associated with lncRNAs during CRC development. We also comprehensively discuss the roles of lncRNAs in predicting therapeutic efficacy in CRC. Finally, we summarize the potential clinical significance of lncRNAs in the management of patients with CRC. Taken together, our review aims to provide novel insights into the critical role of lncRNAs in CRC, offers a bench to bedside perspective, and highlights their important clinical potential as disease biomarkers for directing precision medicine approaches in cancer patients.

2. LncRNAs and hallmarks of CRC

Data gathered in the last two decades have clearly highlighted the biological roles of various lncRNAs in a variety of diseases, and particularly in cancer. In this context, a few specific examples of lncRNAs that associate with various hallmarks of CRC are illustrated in Figure 1. The subsequently text will delineate these features for their biological and clinical significance in further detail.

Figure 1:

Representative LncRNAs and related oncogenic mechanisms in the proliferation, metastasis, angiogenesis, chemoradiotherapy, autophagy, glycolysis, stemness and immunity of colorectal cancer.

2.1. Proliferative signaling pathways

There is unequivocal evidence that specific lncRNAs play a central role in mediating the oncogenic cellular signaling in CRC, by impacting key signaling pathways including the Wnt/b-catenin, PI3K/AKT, JAK2/STAT3 and p53 pathways shown in more detail in Table 1.

Table 1.

Summary of the oncogenic roles of long non-coding RNAs in colorectal cancer.

LncRNA	Effector	Functions	Targets	Signaling pathway	Cell lines	Reference
CCAL	oncogene	proliferation, invasion and migration	AP-2α	Wnt/β-catenin	LOVO	[18]
CCAT2	oncogene	proliferation and metastasis	TCF7L2	Wnt/β-catenin	HCT116,COLO320	[19]
CCAS11	oncogene	proliferation and metastasis	hnRNP-K	Wnt/β-catenin	SW480, SW620	[20]
CRNDE	oncogene	proliferation and chemoresistance	miR-181a-5p	Wnt/β-catenin	HCT116, SW480	[21]
SNHG1	oncogene	proliferation and metastasis	β-catenin/TCF-4	Wnt/β-catenin	SW480, LOVO	[28]
SNHG1	oncogene	proliferation	miR-154-5p, EZH2	miR-154-5p/CCND2	HCT116, HCT8	[46]
SNHG1	oncogene	proliferation and migration	miR-137	miR-137/RICTOR	LOVO, HT29	[47]
ZEB1-AS1	oncogene	proliferation and apoptosis	miR-181a-5p	Wnt/β-catenin	RKO, HCT116, LOVO	[22]
H19	oncogene	proliferation	miR-200a	Wnt/β-catenin	HCT116, SW480	[23]
H19	oncogene	migration and invasion	RAS	RAS/MAPK	SW480, HCT116	[48]
H19	oncogene	proliferation, invasion and EMT	miR-29b-3p	Wnt/β-catenin	HT29, SW480	[64]
H19	oncogene	proliferation and EMT	miR-138, miR-200a	miR-138/ Vimentin, miR-200a/ZEB1,ZEB2	SW620, HT29	[56]
HCG18	oncogene	proliferation and invasion	miR-1271	Wnt/β-catenin	HCT116, SW480	[24]
PlncRNA-1	oncogene	proliferation, migration and invasion	PI3K	PI3K/AKT	SW480, HCT116	[36]
PlncRNA-1	oncogene	proliferation, migration, invasion and EMT	miR-204	Wnt/β-catenin	SW480, SW620, HCT116, HT29	[25]
HOTAIR	oncogene	proliferation and chemoresistance	miR-203a-3p	Wnt/β-catenin	CoLo205, SW620	[26]
HOTAIR	oncogene	proliferation, metastasis and chemoresistance	miR-214	JAK2/STAT3	SW620, SW480, HCT-8	[41]
XIST	oncogene	proliferation and invasion	miR-34a	Wnt/β-catenin	SW480, HCT116	[27]
XIST	oncogene	proliferation, EMT and stem cell formation	miR-200b-3p	miR-200b-3p/ZEB 1	HCT116, SW620	[55]
HNF1A-AS1	oncogene	proliferation, migration and invasion	β-catenin	Wnt/β-catenin	HCT116, SW620	[29]
HIF1A-AS2	oncogene	proliferation, invasion and EMT	miR-129-5p	miR-129-5p/DNMT3A	SW620, HCT116	[75]
BCAT1	suppressor	proliferation and invasion	β-catenin	Wnt/β-catenin	HCT116, SW480,	[32]
LINC00365	oncogene	proliferation, migration, and invasion	CDK1	Wnt/β-catenin	SW480, HT29	[30]
NEAT1	oncogene	proliferation, migration and invasion	DDX5	Wnt/β-catenin	HCT116, SW1116, HT29	[31]
AB073614	oncogene	proliferation, migration and invasion	740Y-P	PI3K/AKT	SW480	[34]
AB073614	oncogene	migration, invasion and EMT	JAK	JAK/STAT3	SW480, HCT116	[76]
LncRNA-422	suppressor	proliferation, migration and invasion	PI3K, AKT, mTOR	PI3K/AKT	SW480, SW620	[35]
TTN-AS1	oncogene	proliferation, migration, invasion and EMT	miR-497	PI3K/AKT	SW480, SW620	[37]
CASC2	oncogene	proliferation	miR-18a	JAK2/STAT3	CACO2, HT29	[42]
PURPL	oncogene	proliferation	MYBBP1A	p53	HCT116, RKO, SW48, DLD1	[44]
ROR	oncogene	proliferation and viability	p53	p53	M5, HT29	[45]
EPB41L4A-AS1	oncogene	proliferation, migration, invasion and EMT	RhoA, Rac1	Rho/ROCK pathway	HCT116, SW620	[49]
B3GALT5-AS1	suppressor	proliferation and EMT	miR-203	miR-203/ZEB2, SNAI2	HCT116, SW620	[59]
GNAT1-1	suppressor	proliferation and metastasis	RKIP	RKIP/NF-κB/Snail	SW480, LOVO	[50]
UICLM	oncogene	proliferation, EMT and stem cell formation	miR-215	miR-215/ZEB2	SW620, DLD-1	[57]
SNHG6	oncogene	proliferation, migration, invasion	UPF1	TGF-β/Smad	RKO, HCT116	[58]
SNHG6	oncogene	EMT	miR-101-3p	miR-101-3p/ZEB 1	RKO, HCT116	[58]
LINC01413	oncogene	proliferation, migration, invasion and EMT	hnRNP-K	hnRNP-K/ZEB 1	HT29, LOVO	[60]
CHRF	oncogene	migration, invasion and EMT	miR-489	miR-489/ TWIST1	HCT116, SW480	[61]
SNHG15	oncogene	proliferation and migration	Slug	NA	SW1116, HCT116, SW480,SW620	[62]
TUG1	oncogene	migration and invasion and EMT	miR-600	miR-600/KIAA1199	HCT116, LOVO, SW620	[63]
SLCO4A1-AS1	oncogene	proliferation, migration, invasion and EMT	β-catenin	Wnt/β-catenin	HCT116, SW480	[65]
CYTOR	oncogene	migration, invasion and EMT	β-catenin	Wnt/β-catenin	HCT116, SW620, HCT-8	[66]
CTD903	suppressor	invasion, migration and EMT	NA	Wnt/β-catenin	RKO, SW480, SW620, HCT116, DLD-1	[67]
TCF7	oncogene	proliferation, migration and invasion	NA	Wnt/β-catenin	DLD-1, LOVO	[68]
MALAT1	oncogene	EMT and chemoresistance	E-cadhrin	EZH2	HT29, SW480, SW620	[70]
HOXD-AS1	oncogene	proliferation, invasion, EMT and stem cell formation	miR-217	EZH2	HCT116, LOVO	[73]
BDNF-AS	suppressor	proliferation, migration and invasion	EZH2-mediated H3K27me3	GSK-3β	HCT116, LOVO	[74]
BC200	oncogene	proliferation, invasion and EMT	STAT3, β-catenin	JAK/STAT3, Wnt/β-catenin	HCT116, HT29	[77]
UCA1	oncogene	proliferation, invasion, migration and EMT	CAFs	mTOR	SW480	[78]
ZFAS1	oncogene	proliferation, EMT and angiogenesis	miR-150-5p	miR-150-5p/ VEGFA, Akt/mTOR	HCT116, HCT8	[79]

2.1.1. Wnt/β-catenin signaling

The Wnt/β-catenin signaling cascade is one of the most important pathways that controls a variety of biological processes in CRC, including cell proliferation and apoptosis[12][13]. The Wnt/β-catenin signaling cascade is regulated by the β-catenin complex, which consists of APC, Axin2, CK1 and GSK-β[14]. More specifically, the β-catenin interacts with various transcription factors such as the TCF/LEF family members to activate the downstream gene targets[15]. Previous studies have reported Wnt/β-catenin pathway members were frequently activated in CRC[16][17]. Several lncRNAs, such as CCAL[18], CCAT2[19] and CASC11[20] have been validated to affect Wnt/β-catenin signaling pathway through their indirect binding to key proteins. The CCAT2 lncRNA enhances Wnt pathway activity through TCF7L2 and facilitates overactivation of oncogenic MYC[19], while CASC11 activates Wnt/β-catenin signaling to promote tumor growth by targeting hnRNP-K in CRC cells[20]. Some lncRNAs can exert effects on Wnt/β-catenin pathway by sponging or binding to specific miRNAs, and this list of lncRNAs include CRNDE[21], ZEB1-AS1[22], H19[23], HCG18[24], PlncRNA-1[25], HOTAIR[26] and XIST[27].

Furthermore, other lncRNAs can directly regulate the core effectors of this signaling pathway. For instance, lncRNA SNHG1 can upregulate β-catenin, TCF-4, cyclin D1 and MMP-9 expression to influence growth of CRC cells[28]. Likewise, the lncRNA HNF1A-AS1 can directly regulate the expression of β-catenin, cyclinD1, and c-myc – all of which play a vital role in Wnt/β-catenin signaling pathway[29]. Upregulation of the LINC00365 can lead to increased CDK1 protein expression and resultant activation of the Wnt/β-catenin signaling pathway[30]. On similar lines, NEAT1 indirectly activates the Wnt/β-catenin signaling pathway by regulating protein DDX5[31], while BCAT1 acts as a tumor suppressor by downregulating β-catenin expression in CRC cells[32].

2.1.2. PI3K/AKT signaling pathway

It is well-established that PI3K/AKT signaling plays a crucial role in the proliferative ability of CRC cells[33]. Recent studies have demonstrated numerous lncRNAs promote CRC proliferation via PI3K/AKT signaling pathway. The lncRNA AB073614[34], lncRNA-422[35] and PlncRNA-1[36] regulate proliferation in CRC cells through PI3K/AKT signaling pathway. The siRNA mediated knockdown of AB073614 expression significantly inhibited the proliferation, migration, and invasion of SW480 cells by targeting the PI3K/AKT-mediated signaling pathway[34]. LncRNA-422 overexpression inhibited cell proliferation, migration and invasion, and can also be regarded as a tumor suppressor through its ability to regulate PI3K/AKT/mTOR pathway in CRC[35]. In addition, PlncRNA-1 exerts its proliferative effects by targeting the PI3K/Akt signaling pathway in CRC[36] and TTN-AS1 enhances the proliferative ability of CRC cells by activating miR-497-mediated PI3K/Akt/mTOR signaling[37].

2.1.3. JAK2/STAT3 pathway

Similar to the PI3K/AKT pathway, the JAK2/STAT3 signaling pathway exerts a significant growth regulatory influence in driving neoplastic progression in CRC[38][39]. Moreover, this signaling pathway is suggested to serve as a crucial regulatory mechanism for the persistent growth of cancer stem cells following radiotherapy treatment in CRC[40]. Accumulating evidence suggests that certain lncRNAs control CRC cell proliferation via JAK2/STAT3 pathway by influencing the expression of phosphorylated STAT3 or by acting as competing endogenous RNAs (ceRNAs). For instance, ST6GAL1 and HOTAIR have been demonstrated to serve as direct targets of miR-214, wherein the expression of ST6GAL1 is modulated by HOTAIR by acting as a sponge for miR-214. Furthermore, ST6GAL1 regulates the elevated metabolic sialylation of c-Met, which finally impacts the CRC cell proliferation through JAK2/STAT3 pathway. The HOTAIR/miR-214/ST6GAL1 axis controls the malignant behavior of CRC cells by activating JAK2/STAT3 pathway[41]. In addition, the lncRNA CASC2 can directly upregulate PIAS3 expression by functioning as a competing endogenous RNA (ceRNA) for miR-18a and leads to the suppression of expression of genes downstream of STAT3, consequentially inhibiting the CRC tumor growth - both in-vitro and in in-vivo experimental models[42].

2.1.4. p53 pathway

The p53 pathway is a well-established pathway that regulates cellular proliferation and apoptosis in cancer cells[43]. In the recent years, several studies have indicated that specific lncRNAs can modulate the growth of CRC cells through p53 pathway. In this context, PURPL was shown to suppress basal p53 levels by interacting with MYBBP1A, a protein that binds to and stabilizes p53, and promotes tumorigenicity in CRC[44]. The inhibition of lncRNA-ROR significantly suppresses cell proliferation and viability, but enhances cell apoptosis, partially by modulating p53[45].

2.1.5. Other mechanisms

In addition to the specific CRC-associated growth pathways described above, several lncRNAs have been reported to affect other mechanisms in CRC. For instance, SNHG1 was shown to be involved in CRC growth by interacting with EZH2 and miR-153–5p[46] or through miR-137/RICTOR axis[47]. Likewise, H19 has been shown to promote the proliferation of colon cancer cells via MAPK signaling pathway[48]. Furthermore, PB41L4A-AS1 may participate in the growth of CRC by activating the Rho/Rho-associated protein kinase signaling pathway[49], while GNAT1–1 acts as a tumor suppressor through regulation of the RKIP-NF-κB-Snail cascade[50].

2.2. Epithelial-mesenchymal transition (EMT) mediated invasion and migration

The EMT is a biological process where epithelial cells acquire a mesenchymal phenotype with enhanced invasive and migratory potential[51]. It is now well-established that EMT acts as a critical mechanism that accounts for tumor invasion, metastasis and chemotherapeutic resistance[52][53]. Previous studies have demonstrated that lncRNAs regulate EMT-mediated invasion and migration mainly through transcription factors, Wnt/β-catenin signaling and epigenetic modifiers in CRC.

2.2.1. Transcription factors related pathway

The EMT process is regulated by several transcription factors, such as the zinc finger E-box-binding homeobox-(ZEB)1, snail family transcriptional repressor 2 (SLUG), snail family transcriptional repressor 1 (SNAI1), TWIST-related protein 1 (TWIST) and ZEB2[54]. LncRNAs can interact with these transcription factors to induce EMT in CRC, leading to the invasion and migration of tumor cells. Some of the lncRNAs (XIST, H19, UICLM, SNHG6, B3GALT5-AS1 and LINC01413) also act as regulators of ZEB1/2 related pathway and most of them mediate their effects by acting as competing endogenous RNAs (ceRNAs). For instance, XIST regulates CRC metastasis by competing with miR-200b-3p to modulate the expression of ZEB1[55]. Likewise, H19 acts as a ceRNA for miR-138 and miR-200a, and inhibits the expression of Vimentin, ZEB1, and ZEB2[56]. On the other hand, UICLM promotes liver metastasis in CRC by regulating the expression of ZEB2 by serving as a ceRNA for miR-215[57], and SNHG6 plays an oncogenic role in CRC cells by inducing EMT through the regulation of ZEB1[58]. A recent research article revealed that B3GALT5-AS1 directly binds to the promoter region of miR-203, downregulated its expression, targeted ZEB2 and SNAI2 and induced EMT and liver metastasis in CRC[59]. In contrast, LINC01413/hnRNP-K/ZEB1 axis upregulates cell proliferation and EMT in CRC by inducing YAP1/TAZ1 translocation[60]. Furthermore, lncRNA CHRF-induced miR-489 loss facilitates metastasis and EMT process in CRC cells probably via TWIST1 signaling pathway[61]. Finally, SNHG15 maintains Slug stability and promotes invasion and migration of CRC through ubiquitin-proteasome signaling pathway[62].

2.2.2. Wnt/β-catenin signaling pathway

The neoplastic progression of cancer cells through EMT is significantly associated with the activation of intracellular stem-associated pathways including Wnt/β-catenin, Notch, TGF-β, and Hedgehog pathways. As one of the major signaling pathways involved in EMT, the Wnt/β-catenin signaling contributes to the activation of transcription factors such as ZEB and Snail, upregulation of the expression of mesenchymal genes and concurrent downregulation of the E-cadherin expression[12][13]. In this regard, several lncRNAs can regulate EMT by targeting Wnt/β-catenin signaling pathway. For instance, TUG1[63], H19[64] and PlncRNA-1[25] has been shown to act as ceRNAs and accelerate metastasis and EMT in CRC through β-catenin signaling. Similarly, TUG1 promotes KIAA1199 expression leading to the induction of EMT and subsequent metastasis in CRC cells through inhibition of miR-600 expression[63]. H19/miR-29b-3p/PGRN axis promotes EMT in CRC through the Wnt signaling pathway as well[64]. PlncRNA-1/miR-204/Wnt/β-catenin regulatory network promotes cell proliferation, EMT and hepatic metastasis in CRC[25]. Other lncRNAs, such as SLCO4A1-AS1[65], CYTOR[66], CTD903[67] and TCF7[68] can modulate β-catenin expression to exert their oncogenic functions. On similar lines, SLCO4A1-AS1 promotes cellular proliferation, migration, invasion and EMT in CRC cells through activation of Wnt/β-catenin signaling pathway[65]. The positive feed-forward circuit of CYTOR-β-catenin promotes EMT and metastasis in CRC[66]; however, the underlying mechanism of CTD903 and TCF7 still require further exploration.

2.2.3. Epigenetic regulatory pathway

Increasing studies have demonstrated that epigenetic regulation plays a pivotal role during EMT, which is also manifested through methylation and acetylation changes[69]. The EZH2 protein has been proved as a critical factor in histone methylation[70], and DNMT3A methyl transferase is responsible for transferring methyl groups to CpG islands, leading to gene repression[71]. Specific lncRNAs, such as MALAT1[70], XIST[72], HOXD-AS1[73], BDNF-AS[74] and BCAT1[75] can exert regulatory function of EMT by targeting EZH2 and DNMT3A. In addition, lncRNA XIST promotes tumor metastasis and EMT in CRC possibly through the regulation of miR-137-EZH2 axis[72]. The HOXD-AS1 lncRNA promotes cancer progression and EMT by modulating the protein expression levels of AEG-1 and EZH2 in CRC cell by competing for miR-217[73]. LncRNA BDNF-AS functions as a tumor suppressor in CRC progression by suppressing GSK-3β expression through its binding to the EZH2 and H3K27me3 within the GSK-3β promoter, which leads to the inhibition of EMT-induced migration[74]. Finally, HIF1A-AS2 can directly bind to miR-129–5p, and positively affect EMT-mediated cell invasion by regulating the expression of miR-129–5p and DNMT3A[75].

2.2.4. Other mechanisms

There are various lncRNAs that regulate invasion, migration and EMT through other mechanisms. For instance, LncRNA AB073614 and BC200 induce invasion and EMT in CRC cells by regulating the JAK/STAT3 pathway[76][77]. Likewise, UCA1 and ZFAS1 can lead to migration and invasion of CRC through EMT through mTOR signaling pathway[78][79], while EPB41L4A-AS1 inhibits the migration, invasion, and EMT in CRC cells through Rho/ROCK pathway[49].

2.3. Glycolysis metabolism

Tumor growth depends on a low-efficiency ATP production process which utilizes glucose uptake for aerobic glycolysis[80]. In this context, the lncRNA PKM2 catalyzes the last step during glycolysis by forming pyruvate and ATP from phosphoenolpyruvate (PEP) and ADP[81]. The glycolytic process can also be influenced by other lncRNAs in CRC. For instance, FEZF1-AS1 and MAFG-AS1 promote pyruvate kinase activity and aerobic glycolysis by regulating PKM2 signaling[82][83]. Furthermore, lncRNAs are able to influence glycolysis in CRC through MYC-mediated pathway, such as LINRIS[84] and GLCC1[83]. The lncRNA KCNQ1OT1 promotes CRC by enhancing aerobic glycolysis via hexokinase-2[85], while HNF1A-AS1 regulates glycolysis by modulating miR-124/MYO6 expression in CRC cells[86].

2.4. Tumor angiogenesis

Angiogenesis is a complex and multistep process driving tumor proliferation and metastasis. There are various regulators involved in tumor angiogenesis, such as vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), angiopoietin 1 and 2 (Ang-1 and Ang-2), and transforming growth factor-α and -β (TGF-α and -β)[87]. Recent studies have found several lncRNAs promoted or inhibited CRC angiogenesis by various mechanisms. ZFAS1 and TPT1-AS1 induce CRC angiogenesis by upregulating VEGFA related pathway[79][88]. MALAT1 mediates angiogenesis by sponging miR-126–5p in CRC[89]. LncRNA AK001058 regulates tumor growth and angiogenesis in CRC via methylation of ADAMTS12[90]. LncRNA GAS5 inhibits angiogenesis and metastasis of CRC through the Wnt/β-catenin signaling pathway[91].

2.5. Chemoradiotherapy resistance

Chemotherapy and targeted drugs, including 5-fluorouracil (5-FU), platinum, hydroxycamptothecin, vincristine, methotrexate, irinotecan, paclitaxel and cetuximab have been widely used for treating CRC[92]. However, a considerable proportion of patients eventually acquire therapeutic resistance to these treatment modalities. The underlying mechanisms for chemoradioresistance are complex and may be related to DNA repair capacity, DNA damage response, apoptotic factors, loss of cell cycle checkpoint control and drug metabolism[5]. Recent studies have highlighted the crucial role of lncRNAs in the chemoradiotherapy resistance in CRC (Table 2).

Table 2.

Summary of the role of long non-coding RNAs in determining therapeutic efficacy in colorectal cancer.

LncRNA	Effector	Therapy	Signaling pathway	Cell lines	Reference
HOTAIR	Resistance	5-FU	NF-κB/TS	HT29, SW480	[93]
HOTAIR	Resistance	Radiotherapy	miR-93/ATG12	HCT116, SW480	[127]
UCA1	Resistance	5-FU	miR-204-5p/CREB1/BCL2/RAB22A	HCT116, HT29, LOVO, SW480	[94]
UCA1	Resistance	Radiotherapy	NA	CCL244	[126]
SNHG6	Resistance	5-FU	miR-26a-5p/ULK1	RKO, HT29, HCT116	[95]
TUG1	Resistance	5-FU	miR-197-3p/TYMS	HCT116, SW1116, HCT8	[96]
TUG1	Resistance	MTX	miR-186/CPEB2	HT29, HCT8	[117]
HOTAIRM1	Sensitivity	5-FU	miR-17-5p/BTG3	HCT116, SW480	[97]
HAND2-AS1	Sensitivity	5-FU	miR-20a/PDCD4	HCT116, SW480	[98]
LINC00152	Resistance	5-FU	miR-139-5p/NOTCH1	HT29, LOVO, HCT116, SW480	[99]
NEAT1	Resistance	5-FU	miR-150-5p/CPSF4	SW480, HCT116	[101]
H19	Resistance	5-FU	miR-194-5p/SIRT1	HCT8, HCT116	[100]
H19	Resistance	MTX	Wnt/β-catenin	HT29	[118]
LINC00957	Resistance	5-FU	NA	HCT8, LOVO	[102]
PVT1	Resistance	5-FU	NA	HCT8, HCT116	[103]
PVT-1	Resistance	Cisplatin	NA	RKO, LOVO	[108]
SnaR	Resistance	5-FU	NA	SNU-C4, SNU-C, 5HCT116	[104]
GIHCG	Resistance	5-FU	NA	LOVO, SW480	[106]
GIHCG	Resistance	Oxaliplatin	NA	LOVO, SW480	[106]
LINC00261	Sensitivity	Cisplatin	Wnt/β-catenin	SW480	[107]
SNHG14	Resistance	Cisplatin	miR-186/ATG14	SW620, SW480	[109]
CACS15	Resistance	Oxaliplatin	miR-145/ABCC1	HT29, HCT116	[110]
KCNQ1OT1	Resistance	Oxaliplatin	miR-34a/Atg4B	HCT116, SW480	[111]
KCNQ1OT1	Resistance	MTX	miR760/PPP1R1B	HT29, Caco2	[116]
MEG3	Resistance	Oxaliplatin	miR-141/PDCD4	HT29, HCT116	[112]
CRNDE	Resistance	Oxaliplatin	miR-136/E2F1	SW480, HCT116	[113]
LINC00525	Resistance	Oxaliplatin	miR-507/ELK3	HCT116, HT29	[114]
MALAT1	Resistance	Oxaliplatin	EZH2	HT29, SW480, SW620	[70]
MALAT1	Resistance	Radiotherapy	miR-101-3p	CCL244, LOVO	[124]
CCAL	Resistance	Oxaliplatin	Wnt/β-catenin	SW480, HCT116	[115]
LUCAT1	Resistance	Oxaliplatin	RPL40/MDM2/p53	SW480,RKO, HCT116	[105]
LUCAT1	Resistance	5-FU	RPL40/MDM2/p53	SW480,RKO, HCT116	[105]
BANCR	Resistance	ADR	miR-203/CSE1L	HCT116, LOVO	[119]
XIST	Resistance	DOX	miR-124/SGK1	HCT116, LOVO	[120]
MIR100HG	Resistance	Cetuximab	Wnt/β-catenin	CRC cells	[121]
CRART16	Resistance	Cetuximab	miR-371a-5p/ERBB3/MAPK	Caco-2	[122]
OIP5-AS1	Sensitivity	Radiotherapy	miR-369-3p/ DYRK1A	LOVO, SW480	[123]
p21	Sensitivity	Radiotherapy	Wnt/β-catenin	SW1116, SW620, LS174T, HT29, LOVO	[125]

2.5.1. LncRNAs and resistance to 5-FU

Some lncRNAs modulate resistance to 5-FU by interacting with miRNAs; and the list of these lncRNA include HOTAIR[93], UCA1[94], SNHG6[95], TUG1[96], HOTAIRM1[97], HAND2-AS1[98], LINC00152[99], H19 [100]and NEAT1[101]. For instance, HOTAIR contributes to 5-FU resistance through suppression of miR-218 and activation of NF-κB signaling in CRC[93]. On the other hand, UCA1 enhances cell proliferation and 5-FU resistance in CRC by inhibiting miR-204–5p[94], while SNHG6 acts through ULK1-induced autophagy by sponging miR-26a-5p[95]. In addition, the suppression of LINC00957 can reverse 5-FU resistance in CRC by down-regulation of P-gP[102]. Other reports have indicated that inhibition of PVT-1 can reverse 5-FU resistance through MDR related proteins, including MRP1, P-gp, mTOR and Bcl-2[103]. The lncRNAs snaR[104], LUCAT1[105] and GIHCG[106] also have impact on 5-FU resistance in CRC. For instance, knockdown of snaR decreases cell death following 5-FU treatment, which indicates that snaR loss decreases in-vitro sensitivity to this chemotherapeutic drug[104]. Similarly, GIHCG has been confirmed to facilitate cell survival following 5-FU treatment[106], while LUCAT1 induces cell cycle arrest and resistance to 5-FU by binding UBA52 and activating the RPL40-MDM2-p53 pathway in CRC cells[105].

2.5.2. LncRNAs and resistance to platinum drugs

Like a double-edged sword, lncRNAs can not only increase sensitivity to platinum drugs, but also induce resistance to these compounds in CRC cells. As for cisplatin, LINC00261 sensitizes HCT116, HCT8, HT29, SW480 human colon cancer cells to cisplatin therapy via β-catenin pathway[107]. Another report showed that PVT-1 inhibits cisplatin sensitivity in HT29, SW480, HCT116, RKO, LOVO cells[108]. SNHG14 stimulates cell autophagy to facilitate cisplatin resistance of SW620, SW480 cells by regulating miR-186/ATG14 axis[109]. With regards to treatment with oxaliplatin, a number of lncRNAs promote resistance to this drug by sponging specific miRNAs, including CACS15[110], KCNQ1OT1[111], MEG3[112], CRNDE[113] and LINC00525[114]. Additionally, MALAT1 can promote oxaliplatin-based chemotherapeutic resistance in CRC patients[70]. Likewise, CCAL promotes oxaliplatin resistance in CRC cells by increasing β-catenin expression levels[115], while LUCAT1 can influence sensitivity to this drug treatment in CRC via p53 signaling pathway[105]. Finally, lncRNA GIHCG has been reported to induce chemoresistance to oxaliplatin and indicates poor prognosis in CRC[106].

2.5.3. LncRNAs and resistance to other chemotherapeutic and targeted drugs

In addition to specific lncRNAs that have been reported to associate with 5-FU and platinum-based drugs, others have been related to various other chemotherapies and targeted drugs. In this regard, lncRNA KCNQ1OT1[116] and TUG1[117] have been reported to mediate methotrexate resistance in CRC through the sponging of miR-760 and miR-186. In another report, H19 was shown to mediate methotrexate resistance in CRC through Wnt/β-catenin pathway[118]. LncRNA BANCR enhances adriamycin resistance in CRC by acting as a ceRNA[119], while suppression of XIST can inhibit doxorubicin resistance by upregulation of miR-124 and downregulation of SGK1[120]. Other lncRNAs such as MIR100HG and CRART16 have been reported to mediate cetuximab resistance in CRC by interacting with specific miRNAs[121][122]. In this context, MIR100HG-derived miR-100 and miR-125 have been reported to modulate cetuximab resistance via Wnt/β-catenin signaling[121], while CRART16 overexpression contributes towards cetuximab resistance through miR-371a-5p/ERBB3/MAPK axis[122].

2.5.4. LncRNAs and resistance to radiotherapy

Radiotherapy is a localized treatment for CRC, which is primarily used in combination with chemotherapy. LncRNAs are important determinants for resistance to radiotherapy. For example, lncRNA OIP5-AS1 regulates radiotherapeutic resistance by targeting DYRK1A through miR-369–3p in CRCcells[123]. Similarly, MALAT1 modulates radioresistance to CRC via miR-101–3p sponging[124]. Other reports have shown that lncRNA-p21 enhances the sensitivity of radiotherapy in CRC by targeting the Wnt/β-catenin signaling pathway[125], while downregulation of UCA1 and HOTAIR enhances the radiosensitivity in this malignancy[126][127].

3. LncRNA-associated novel oncogenic mechanisms

3.1. LncRNAs and cell autophagy

Autophagy is a common biological process during environmental changes such as starvation where cells degrade macromolecules to provide energy for fundamental biological processes[128]. The formation of the autophagosome comprises of three main steps: initiation, nucleation, and elongation. The molecular mechanisms underlying each step consist of several conserved autophagy-related genes (ATGs). Previous studies have suggested that autophagy plays a dual role in oncogenesis, including regulation of oncogenes and tumor suppressor gene expression and participating in response to chemoradiotherapy[129]. The expression of lncRNAs can regulate autophagy in CRC and play a vital role in tumor proliferation, EMT and chemoradiotherapy. The lncRNA SLCO4A1-AS1[130], HAGLROS[131], UCA1[132] and MALAT1[133] promote CRC cell proliferation by enhancing autophagy via miRNA related axis. In addition, lncRNA CPS1-IT1 suppresses EMT and metastasis in CRC by inhibiting hypoxia-induced autophagy through inactivation of HIF-1α[134]. Likewise, GAS5 promotes autophagy by targeting miR-222–3p through the GAS5/PTEN-signaling pathway in CRC[135]. Other reports have shown that H19 confers resistance to 5-FU in CRC by promoting SIRT1-mediated autophagy[100]. In addition, NEAT1 knockdown attenuates autophagy to elevate 5-FU sensitivity by targeting miR-34a[136], while SNHG14 stimulates cellular autophagy to facilitate cisplatin resistance in CRC by interacting with miR-186/ATG14 axis[109].

3.2. LncRNAs and cancer stem cells

The cancer stem cells (CSCs) are defined as a small subpopulation of cancer cells with embryonic stem cell (ESC) characteristics. CSCs have been confirmed to initiate tumor growth, slow-cycling cellular turnover, as well as the resistance to therapy[137]. In CRC, CSCs can be modulated by several lncRNAs through different pathways. For instance, suppression of HOTAIR inhibits the malignant characteristics of CD133(+) CSCs in CRC[138]. The lncRNAs BCAR4[139], cCSC1[140], LINC00525[114], ROR[141], LINC00657[142] and MALAT1[143] modulate CSC phenotype of CRC cells by targeting miRNAs related signaling. Similarly, GAS5[144] and GATA6[145] are critical for maintaining stemness, as evidenced by high expression of GATA6 in the cancer stem cells in CRC, leading to recruitment of the NURF complex onto the Ehf promoter and mediate its transcription, promote tumor initiation and progression[145]. Additionally, lncRNA B4GALT1-AS1 promotes colon cancer cell stemness and migration by recruiting YAP to the nucleus and enhancing YAP transcriptional activity in colorectal neoplasia[146].

3.3. LncRNAs and epigenetic mechanisms

Epigenetic alterations include reversible events within the human genome, and the primary processes include DNA methylation, histone modifications and chromatin remodeling[147]. In CRC, several epigenetic changes can be regulated by lncRNAs. The lncRNA HOTAIR contributes to histone H3 lysine 27 methylation and lysine 4 demethylation by interacting with both LSD1/CoREST/REST complex and PRC2[148]. This lncRNA is also reported to reprogram chromatin organization in CRC[149]. It has been reported that NEAT1[150] and MALAT1[151] contribute to chromatin remodeling by regulating chromatin modifiers. DACOR1 interacts with both chromatin and targets DNMT1, a DNA methyltransferase, to influence DNA methylation levels in colon cancer cells[152]. Furthermore, GAS5 inhibits progression of CRC by interacting with and triggering YAP phosphorylation and degradation, and is negatively regulated by the m6A reader YTHDF3[153]. Similarly, m6A-induced lncRNA RP11 triggers the dissemination of CRC cells via upregulation of ZEB1[154]. CCAT2 overexpression has been found to promote malignant transformation in colon cells through inducing chromosomal instability via BOP1/AURKB cascade[155].

3.4. LncRNAs and tumor microenvironment

Recent studies have suggested that lncRNAs derived from carcinoma-associated fibroblasts (CAF) promote CRC progression and therapeutic resistance. H19 enriched in CAFs activates the β-catenin pathway by serving as a competing endogenous RNA sponge for miR-141; thus enhancing the stemness and chemoresistance of CRC cells[156]. Likewise, CAF-derived CCAL interacts with human antigen R to upregulate β-catenin expression and thereby induce the resistance to oxaliplatin in CRC cells[115]. LncRNAs can also affect the cancer-related immune cells in tumor microenvironment. NKILA promotes tumor immune evasion by sensitizing T cells to activation-induced cell death[157]. On the same line, lncRNA GM16343 overexpression in CD8 + T cells promotes the secretion of interferon γ and therefore enhances the antitumor immune response[158], while SATB2-AS1functions as a tumor suppressor by regulating TH1-type chemokines expression and immune cell density[159].

4. Potential of LncRNAs in clinical applications

In addition to their biological role in mediating cancer cell growth in various human malignancies, lncRNAs are also emerging as important substrates for the development of cancer biomarkers for the early detection, prognosis prediction, and predicting therapy response to various chemotherapies and developingtreatments (Figure 2). Subsequent sections will describe the current state of knowledge with regards to lncRNAs and their clinical significance in CRC.

Figure 2:

Schematic summary for the potential of lncRNA to serve as diagnostic, prognostic, personalized therapeutic and disease monitoring biomarkers in cancer patients.

4.1. LncRNAs as noninvasive biomarkers for early cancer diagnosis

Though invasive colonoscopy has significantly contributed to the early diagnosis of CRC, non-invasive biomarkers are imperative for improving the diagnostic accuracy for this disease, reducing the medical expense and increasing participation of average-risk population in cancer screening efforts. In this regard, accumulating evidence suggests that lncRNAs could act as novel noninvasive diagnostic biomarkers for CRC patients (Table 3). It has been reported that CCAT1 is overexpressed during the multistep normal-adenoma-carcinoma sequence, and can serve as a potential early diagnostic biomarker[160][161]. Similarly other lncRNAs, such as BLACAT1[162], BCAR4[163] and CRNDE[164] are frequently overexpressed in CRC with corresponding AUC values of 0.85, 0.936 and 0.921 respectively. Additionally, the combination of two mRNAs, KRTAP5–4 and MAGEA3, and lncRNA BCAR4, can discriminate CRC patients from healthy controls with an AUC value of 0.857[163].

Table 3.

Summary of the diagnostic performances of long non-coding RNAs in colorectal cancer.

LncRNAs	Expression	CRC cases vs. controls	AUC	Sensitivity	Specificity	Sample	Reference
BLACAT1	Up	30 vs.30	0.858	83.3%	76.7%	Serum	[162]
BCAR4	Up	76 vs.76	0.936	NA	NA	Serum	[163]
KRTAP5-4, MAGEA3 and BCAR4	-	76 vs.76	0.857	NA	NA	Serum	[163]
CRNDE-b	Up	161 vs. 222	0.921	85%	96%	Plasma	[164]
CRNDE-g	Up	161 vs. 222	0.892	80%	96%	Plasma	[164]
CRNDE-h	Up	161 vs. 222	0.888	80%	96%	Plasma	[164]
NEAT1	Up	100 vs. 100	0.787/0.871	69%/70%;	79%/96%	Blood	[165]
91H	Up	76 vs.76	0.870	84.48%	80.36%	Plasma	[166]
PVT-1	Up	76 vs.76	0.786	68.97%	82.14%	Plasma	[166]
MEG3	Down	76 vs.76	0.819	75.86%	82.14%	Plasma	[166]
GAS5	Up	76 vs.76	0.642	63.79%	62.5%	Plasma	[166]
CCAT1-L	Up	76 vs.76	0.748	75.86%	73.21%	Plasma	[166]
91H, PVT-1 and MEG3	-	76 vs.76	0.877	82.76%	78.57%	Plasma	[166]
LINC00654	Up	114 vs.58	0.734	NA	NA	Plasma	[167]
LINC00909	Up	114 vs.58	0.886	NA	NA	Plasma	[167]
SNGH11	Up	114 vs.58	0.868	NA	NA	Plasma	[167]
ZFAS1	Up	114 vs.58	0.850	NA	NA	Plasma	[167]
ZFAS1, SNHG11, LINC00909 and	-		0.937	NA	NA	Plasma	[167]
LINC00654
LINC02418	Up	125 vs. 125	0.8978	95.2%	66.4%	Serum	[169]
RP11-296E3.2	Up	60 CRC	0.663	69.0%	62.1%	Plasma	[171]
RP11-296E3.2 and CEA	-	60 CRC	0.722	72.4%	71.4%	Plasma	[171]
BANCR, NR_026817, NR_029373, and NR_034119	-	120 vs. 120	0.881	89.17%	75.83%	Serum and tissue	[168]

Detecting NEAT1 expression in the whole blood sample has a diagnostic AUC of 0.787 (NEAT1_v1) and 0.871 (NEAT1_v2), with a corresponding sensitivity of 69% (NEAT1_v1) and 70% (NEAT1_v2) and a specificity of 79% (NEAT1_v1) and 96% (NEAT1_v2), respectively[165]. On similar lines, lncRNAs such as 91H, PVT-1 and MEG3 have been shown to be expressed at significantly higher levels in plasma from CRC patients and can be regarded as promising diagnostic biomarkers for early-stage CRC; especially considering that they have superior diagnostic accuracy compared to CEA and CA19–9[166]. Similarly, SNGH11, ZFAS1, LINC00909 and LINC00654 exhibit high diagnostic performance for CRC, especially in the early-stage disease[167]. Additionally, four lncRNAs (BANCR, NR_026817, NR_029373, and NR_034119) are significantly dysregulated in both tissue and serum samples of CRC patients and this distinctive 4-lncRNA panel has a better diagnostic performance than CEA[168]. Finally, LINC02418 is highly expressed in the serum of CRC patients and is considered as a promising, novel biomarker that can be used for the clinical diagnosis of CRC[169].

4.2. LncRNAs as prognostic and predictive biomarkers for clinical outcomes in cancer

LncRNAs can be considered as prognostic biomarkers of colorectal cancer for predicting patient survival and therapeutic response, as depicted in Table 4. Retrospective clinical studies have revealed that various lncRNAs including LUADT1[170], LEF1-AS1[171], H19, MIR31HG, HOTAIR, WT1-AS, LINC00488[172], XIST[173], CCAT1,2[174], DANCR[175], PVT-1[176], SBDSP1[177], BANCR[178], UCA1[179] and CRNDE[180] are frequently over-expressed in CRC, and their expression significantly correlates with shorter survival time in these patients. Additionally, high expression LUADT1[170] and XIST[173] correlates with tumor size, metastasis, and TNM staging in CRC. Similarly, the expression of CRNDE associates with high tumor stage and lymph node metastasis[180]. Other reports have shown that SBDSP1 expression is significantly associated with tumor differentiation, depth of invasion and TNM stage. Patients with higher SBDSP1 expression have significantly shorter disease-free survival and overall survival, compared to those with lower expression of this lncRNA[177]. Upregulation of BANCR may be associated with the lymph node metastasis and poor survival in CRC patients[178]. Similarly, the overexpression of UCA1 is associated with poor prognosis and more advanced clinicopathological features, including tumor differentiation, lymph node metastasis, distant metastasis, tumor node metastasis, stage, tumor invasion depth and tumor size[179]. In contrast, low expression HOTTIP is an independent prognostic marker for overall survival in CRC patients[181]. Dysregulated expression of AFAP1-AS1 has been confirmed to associate with oncogenesis and tumor progression, and can serve as a novel potential molecular biomarker in tumor prognosis[182][183]. Low expression of NR_029373 and NR_034119 have both been identified as independent unfavorable prognostic factors affecting the disease-specific survival rate of CRC patients[168]. Finally, the expression of NEAT1 has been associated with tumor recurrence and unfavorable prognosis in CRC [184].

Table 4.

Summary of the prognostic significance of long non-coding RNAs in colorectal cancer.

LncRNAs	Expression	Patient No.	Univariate analysis HR (95%CI)		Multivariate analysis HR (95%CI)		Reference
			OS	PFS/DFS/RFS	OS	PFS/DFS/RFS
NEAT1_v1	Up	191	0.415(0.227–0.758)	NA	NA	NA	[165]
NEAT1_v2	Up	191	0.426(0.231–0.784)	NA	0.519 (0.280–0.965)	NA	[165]
NEAT1	Up	239	1.88(1.32–2.69)	1.93(1.38–2.71)	1.70(1.18–2.45)	1.80(1.27–2.55)	[184]
PVT-1	Up	210	2.07(1.42–3.06)	2.01(1.26–2.64)	1.76(1.16–2.37)	2.15(1.39–2.76)	[176]
HOTTIP	Down	52	NA	NA	4.949(1.685–11.984)	NA	[181]
XIST	Up	196	1.284(1.143–1.442)	1.242(1.115–1.384)	1.197(1.064–1.346)	1.165(1.044–1.300)	[173]
CCAT1	Up	135	4.06(1.47–16.8)	3.88(1.67–8.39)	5.90(2.09–24.7)	2.52(1.07–5.56)	[174]
CCAT2	Up	135	2.04(1.05–3.84)	2.55(1.19–5.31)	2.40(1.22–4.59)	2.39(1.10–5.08)	[174]
CCAT1+CCAT2	Up	135	NA	NA	8.38(2.68–37.0)	2.60(1.04–6.06)	[174]
DANCR	Up	104	2.491(1.335–7.264)	2.614(1.572–7.715)	2.131(1.157–7.058)	2.397(1.385–7.279)	[175]
SBDSP1	Up	171	2.673(1.562–5.632)	2.523(1.429–5.013)	2.137(1.238–4.672)	2.132(1.123–4.239)	[177]
BANCR	Up	106	2.74(1.52–4.96)	NA	2.24(1.22–4.16)	NA	[178]
UCA1	Up	715	2.65(1.93–3.63)	NA	2.39(1.80–3.16)	NA	[179]
AFAP1-AS1	Up	52	2.780(1.308–5.909)	2.389(1.162–4.913)	2.358(1.110–5.008)	2.120(1.032–4.353)	[182][183]
NR_029373	Down	94	NA	0.477(0.261–0.870)	NA	0.457(0.246–0.848)	[168]
NR_034119	Down	94	NA	0.561(0.316–0.996)	NA	0.517(0.277–0.964)	[168]

4.3. LncRNAs as primary targets for new anti-cancer drug development

Considering the crucial role of lncRNAs in cancer biology, it is reasonable to identify appropriate lncRNAs as primary targets for developing novel anti-cancer drugs. Several experimental attempts have been made in other cancers based on advancing biomaterial techniques. Nanoliposomes delivering Let-7b into esophageal cancer stem-like cells significantly inhibited the Wnt signaling-mediated self-renewal and enhanced cell sensitivity to 5-fluorouracil or docetaxel[185]. An engineered nanoparticle platform that delivered lncAFAP1-AS1 siRNA effectively reversed radioresistance of triple-negative breast cancer (TNBC)[186]. Similarly, nanoparticles carrying RNA interference of DANCR dramatically inhibited the growth of TNBC xenografts without inducing obvious toxic side-effects[187]. In CRC, 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine nanoparticles carrying lncFLANC siRNA restrained the metastasis of CRC cells in vivo without inducing evident toxic or inflammatory responses[188]. In addition to nanoparticle based drug delivery systems, recent studies have proposed the therapeutic potential of the CRISPR/Cas9 system to edit the expression of LncRNAs in CRC patients and related explorations are starting[189].Despite of these effects, we noted that rare clinical trials are currently available for lncRNA-based targeted drugs, which may be attributed to inherent factors such as lack of well-validated targets and uncertain safety of novel drugs.

5. Challenges and perspectives

In this review, we focused on the role of lncRNAs as hallmarks of CRC including their ability to impact proliferative signaling, EMT-mediated invasion and migration, glycolysis metabolism, tumor angiogenesis and resistance to chemoradiotherapy. The mechanistic investigations have revealed that lncRNAs exert their oncogenic role either through regulation of canonical signaling pathways (such as Wnt/β-catenin and PI3K/Akt) or by serving as ceRNAs for miRNAs. These novel studies have closely linked lncRNAs with cell autophagy, cancer stem phenotype, epigenetic regulation and tumormicroenvironment, where further explorations are still needed. For instance, the regulatory function of lncRNAs in m6A modifications during CRC progression remains poorly studied. Since immune checkpoint inhibitors (ICI) have held the promise for late-stage CRC, the potential correlation of lncRNAs with anti-cancer efficacy of ICIs may be worth studying. Furthermore, accumulating evidence has pointed to the crucial oncogenic role of gut microbiota in CRC and whether the expression of oncogenic lncRNAs is regulated by specific bacteria, which remains unknown. On the other hand, in addition to their oncogenic role, the biological functions of lncRNAs in normal intestinal epithelial development, barrier function protection and homeostasis maintenance should be clarified, which will provide pivotal information for the clinical translation of lncRNAs into CRC diagnosis and treatment.

In this review, we also discussed the clinical potential of lncRNAs as diagnostic or prognostic biomarkers. Encouragingly, a number of circulating lncRNAs have shown superior performance in discriminating CRC patients from healthy controls, implying their attractive potential to serve as noninvasive diagnostic biomarkers. However, several challenges remain to date that must be overcome prior to their clinical translation. First, rare diagnostic lncRNA signatures are currently constructed, which may help improve the performance of a single lncRNA detection. Second, the levels of circulating lncRNAs in patients with stage I-II CRC, advanced adenoma and inflammatory bowel disease should be clarified, which will provide better clues for their potential as early diagnosis and risk population screening. Finally, whether dynamic detection of postoperative circulating lncRNAs for their ability to diagnose CRC recurrence and progression should also be studied. On the other hand, mounting evidence points to their predictive role as tissue-derived lncRNAs in patient survival and disease recurrence, suggesting detection of these lncRNAs may help identify high risk populations from patients within the same TNM stage. However, the prognosis of CRC is affected by various clinical factors and it will be promising to integrate lncRNAs with traditional clinical and pathological indicators to establish and validate multifactorial risk-stratification models. Finally, considering the crucial role of lncRNAs in CRC progression and therapeutic response, it is possible to develop them into therapeutic targets, where selection of most appropriate candidate lncRNAs and developing matched drug delivery systems would be of tremendous clinical significance.

6. Conclusions

In summary, emerging and accumulating evidence has demonstrated that lncRNAs play a crucial role in hallmarks of CRC through various molecular mechanisms and therefore have potential to be utilized for targeted drug development. However, identification and validation of reliable targets, and developing stable and safe drug delivery systems remain to be challenging. In addition, although clinical evidence suggests that lncRNAs can serve as promising diagnostic and prognostic biomarkers in CRC patients, although multicenter validations based on sufficient samples are still necessary. Meanwhile, quality control of samples and standardized detecting techniques of LncRNAs should be emphasized in the following translational work. With our increasing knowledge of lncRNAs in CRC, it is hopeful that some of these may benefit the biomarker-directed precision medicine approaches in improving survival outcomes in CRC patients.

Highlights.

1. LncRNAs extensively participate in various hallmarks of CRC.

2. LncRNAs affect the efficacy of chemotherapy and radiotherapy in CRC.

3. Detecting circulating LncRNAs benefits the early diagnosis of CRC.

4. Tumor-derived LncRNAs act as a reliable prognostic biomarker for CRC.

5. LncRNAs can be utilized as primary targets for anti-CRC drug development.