LncRNA Polymorphism Affects the Prognosis of Gastric Cancer

doi:10.21203/rs.3.rs-1431892/v1

Background: Previous studies have found that lncRNA polymorphisms are associated with the prognosis of gastric cancer (GC), but the specific roles of many lncRNA polymorphism sites in gastric cancer are still unclear. The aim of our study is to deeply explore the relationship between genetic polymorphism of lncRNA and the prognosis of GC.

Methods: The genotypes of candidate SNPs locus were detected by Sequenom Mass ARRAY SNP. We deeply analyzed the association of lncRNA polymorphisms with GC prognosis by univariate and multivariate COX regression, stratified analysis, conjoint analysis, and log-rank test.

Results: We found that mutations at rs2579878 and rs10036719 loci reduced the risk of poor prognosis of GC. Stratified analysis showed that rs2795025, rs10036719 and rs12516079 polymorphisms were all associated with tumor prognosis. In addition, conjoint analyses showed that the interaction between these two polymorphic sites, rs2795025 and rs12516079, could increase the risk of poor prognosis. Multivariate analysis also found that AG/AA genotype of rs10036719 and AG genotype of rs12516079 were independent prognostic factors. Moreover, we also found that the high express of both CCDC26 and LINC02122 were shown to be associated with the poor survival status of GC patients.

Conclusions: We found that the genetic polymorphism of lncRNA plays a role in the development of GC and is closely related to the survival time of patients. Our results could serve as a predictor to the prognosis of GC.

GC

LncRNA

Gene

Polymorphism

Prognosis

Gastric cancer (GC) is one of the fatal digestive tract tumors worldwide and is responsible for over one million new cases in 2020 and an estimated 769,000 deaths[1]. GC is a complex heterogeneous disease and is closely related to genetic alterations [2]. Increasing studies have found that single nucleotide polymorphism (SNP) is closely related to the occurrence, progression and metastasis of GC [3–6], and is expected to become a powerful marker for its diagnosis and prognosis.

Long-chain non-coding RNA (lncRNA) has become the focus of cancer research due to its high specificity and easy detection in tissue, serum, plasma, urine and saliva. LncRNA polymorphism can affect the outcome of many biological processes and then affect the entire occurrence and development of cancer.

Previous studies have shown that lncRNA HOX transcript antisense RNA (HOTAIR) gene rs17720428 SNP is correlated with the risk and prognosis of GC in the Chinese Han population[7]. Similarly, other research also found that specific lncRNA (HOTTIP and MALAT1) SNPs have potential to be biomarkers for hepatocellular cancer (HCC) risk and prognosis[8].The lncRNA growth arrest-specific 5 (GAS5) plays an important role in the development of digestive system tumors[9],its polymorphic site rs145204276 might induce the promoter activity of lncRNA GAS5 to protect against the development of breast cancer[10]. However, other scholars also found that the polymorphic site rs145204276 may contribute to hepatocarcinogenesis by affecting methylation status of the GAS5 promoter and subsequently its transcriptional activity[11].In addition, lncRNA GAS5 rs145204276 can affect the prognosis of prostate cancer by regulating the expression of HMGB1[12]. So far, there have been many researches involved to the association of lncRNA polymorphisms with cancer[13–15], but the specific role of lncRNA polymorphisms in GC and the corresponding mechanism are still unclear.

Our research aims to explore whether lncRNA polymorphism affects the prognosis of GC, we selected 10 lncRNA polymorphism sites based on the previous results of the research group[16].First of all, the candidate SNPs genotypes are divided into four models: codominant model, dominant model, recessive model and allele model to initially explore its relationship with the prognosis of GC. Then, we combined Stratified analysis to further explore its association with the prognosis of gastric cancer. Besides, we also explored the association between the combined effects of SNPs and the prognosis of GC. Further, we did a multivariate analysis and constructed a risk model for poor prognosis of GC based on its results. Finally, we explored the relationship between lncRNA expression and prognosis of GC in the TCGA database. Our results could serve as a new avenue to the development of personalized therapy for the treatment of GC.

Study populations and specimens

GC patient was derived from new cases in Xianyou County Hospital of Fujian Province, China. The inclusion criteria for patients are as follows: 1) tissue sample obtained by operation or endoscopy, new cases confirmed by pathology; 2) confirmed date from April 2013 to November 2017; 3) living in Xianyou for more than 10 years. We also applied the following exclusion criteria: 1) patients with gastric inflammation or benign lesions; 2) patients with critical conditions or inability to clearly answer questions; 3) recurrent and relapse cases.

Our study adopts a prospective case follow-up study design, and obtains its complete survival information and clinical data through annual data on all causes of death and excerpts of case information, and follow-up by village doctors. Finally, a total of 344 people were included in this study for follow-up analysis.

5ml of fasting peripheral venous blood was collected from the patients. The blood samples were placed in EDTA anticoagulant tubes, centrifuged at 3,000 g for 10 min, then packed into plasma, leukocyte and erythrocytes, and stored in − 80 ℃ refrigerator. All subjects gave their consent for inclusion before they participated in the study. All procedures involving human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Bioethics Committee of the Medical University of Fujian (Fu Medical Ethics Review No. 97).

Genotyping lncRNA-related SNPs

The genotypes of candidate SNPs locus were detected by Sequenom Mass ARRAY SNP. The PCR amplification primers and single-base extension primers of the SNP site to be detected were designed using Genotyping Tools of Sequenom Company and MassARRAY Assay Design software. The relative molecular mass of the extension product was detected by matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS), and the genotyping of SNP was detected by judging the different.

Quality Control

Quality controls were carried out according to the following standards: 1) Dish QC values can be calculated from the signal values of thousands of non-polymorphic probes, which can be evaluated the difference between the distribution of signal value and background signal value. Samples with DQC lower than 0.82 were not included in the subsequent typing; 2) A phenotype-blind genotyping during genotyping was pursued.

Statistical analyses

The 1-, 3- and 5-year survival rates of different genotypes at the same polymorphism site were obtained by life table method. The median survival time (MST) of patients was obtained by Kaplan-Meier method. Univariate and multivariate COX regression analysis were used to analyze the relationship between lncRNA polymorphism sites and prognosis of GC and calculate the hazard ratios (HRs) and its 95% confidence interval (confidence intervals, CIs). Nomogram were used to visualize the results of multivariate analysis. Log-rank test was used to explore the relationship between lncRNA expression and prognosis of GC. SPSS 18.0 and R 4.0 software package was used to complete the above analysis. All P values were based on bilateral test, and the statistical test level was α = 0.05.

Screening of GC-related lncRNA SNP

A total of 10 lncRNA polymorphic loci and 344 patients were included in this study for analysis. The detailed information of these 10 polymorphic loci is shown in Table 1, and characteristics of the patients are presented in Table 2.

Table 1

Basic information of 10 candidate lincRNA SNPs locus
NO.	SNP ID	Gene name	Num. of transcripts	Chrosome	Cytoband	H-W P	MAF
1	rs10036719	LINC02122	1	5	q13.3	0.834	0.358
2	rs12516079	LINC02122	1	5	q13.3	0.988	0.349
3	rs56093317	LINC01137	1	1	p34.3	0.513	0.316
4	rs61894277	LINC02553	1	11	q21	0.250	0.345
5	rs2795025	LINC00687	3	20	p12.2	0.961	0.253
6	rs11617815	LINC00327	3	13	q12.12	0.975	0.324
7	rs1348758	LINC00927	4	15	q25.1	0.874	0.385
8	rs2579878	CCDC26	3	8	q24.21	0.673	0.358
9	rs5829142	LINC00298	2	2	p25.1	0.283	0.349
10	rs9809325	LINC00879	5	3	q11.2	0.722	0.315

Table 2

Relationship between basic characteristics and prognosis of patients
Variables		N	MST (M)	Survival rate(%)			HR (95%CI)	P
Variables		N	MST (M)	1-year	3-year	5-year	HR (95%CI)	P
Gender	Male	253	29.00	72.67	44.35	37.75	1	0.953
	Female	91	35.00	75.69	48.31	43.71	0.917(0.667–1.261)
Age(year)	≦ 65	120	55.00	84.10	58.86	47.04	1	0.003*
	65-	224	23.00	67.79	38.29	35.17	1.600(1.179–2.173)
Marriage status	Married	320	31.00	74.30	46.81	40.83	1	0.020*
Marriage status	others	24	17.00	62.50	26.17	17.45	1.753(1.078–0.851)
educational level	Primary and below	272	28.00	72.38	44.98	38.92	1	0.172
educational level	junior high	52	28.00	76.70	38.52	33.02	1.151(0.794–1.667)
	Senior high and above	20	73.00	80.00	69.44	62.14	0.560(0.275–1.139)
Occupation	Farmers	255	28.00	71.32	41.67	36.53	1	0.107
	Others	89	55.00	79.66	56.42	47.37	0.764(0.548–1.065)
Tumor location	Non-cardia	179	28.00	72.07	44.68	39.76	1	0.594
Tumor location	Cardia	165	31.00	75.00	46.21	39.01	0.927(0.701–1.225)
TNM stage	Early and middle stage	217	81.00	92.17	66.93	60.16	1	< 0.001*
	Late stage	127	10.00	41.27	8.50	4.64	6.590(4.889–8.881)
Operation	No	76	8.00	35.53	3.80	1.90	1	< 0.001*
	Yes	268	64.00	84.27	57.17	49.92	0.175(0.129–0.238)
Chemotherapy	No	140	22.00	65.00	40.48	36.49	1	0.036*
	Yes	204	33.00	79.31	48.81	41.23	0.742(0.561–0.981)
Radiotherapy	No	243	29.00	73.25	44.61	38.47	1	0.528
	Yes	101	33.00	74.00	47.40	41.52	0.905(0.665–1.233)
* P < 0.05. When the MST cannot be calculated, it is replaced by the average survival time.

Health lifestyle and prognosis of patients with GC

In order to assess the factors affecting the survival of patients with GC, we analyzed the association between the health lifestyle and the survival statue of GC patients. The results are shown as Table 3. Certain habits including smoking, alcohol or tea consumption, and frequent mental depression, are the risk factors for poor prognosis of GC. Self-care, sleep for more than 5 hours, and regular exercise, are the protective factors for the prognosis of GC. (P < 0.05)

Table 3

Relationship between the health habits and lifestyle and prognosis of patients with GC after diagnosed 1 year
Variable		N	MST (M)	Survival rate(%)			HR (95%CI)	P
Variable		N	MST (M)	1-year	3-year	5-year	HR (95%CI)	P
Drinking changes	No-No	286	30.00	73.33	44.10	38.22	1	0.134
Drinking changes	No-Yes	16	20.00	75.00	31.25	25.00	1.853(1.035–3.318)	0.038*
	Yes-No	35	90.80	77.14	59.60	55.02	0.815(0.468–1.419)	0.469
	Yes-Yes	7	36.00	57.14	57.14	40.82	1.362(0.496–3.743)	0.549
Drinking tea changes	No-No	286	30.00	74.91	45.27	38.06	1	0.003*
Drinking tea changes	No-Yes	39	21.00	66.67	35.60	29.67	2.24(1.456–3.447)	< 0.001*
	Yes-No	27	143.12	66.67	55.33	55.33	1.08(0.589–1.979)	0.804
	Yes-Yes	10	81.00	80.00	60.00	60.00	0.750(0.301–1.865)	0.536
Drinking	No	321	31.00	73.75	45.84	40.10	1	0.030*
	Yes	23	21.00	69.57	39.13	29.92	1.759(1.056–2.93)
Drink tea	No	295	31.00	74.15	46.24	39.94	1	0.008*
	Yes	49	22.00	69.39	40.46	35.40	1.701(1.148–2.521)
Self-care	No	62	12.00	51.61	7.60	7.60	1	< 0.001*
	Yes	282	41.00	78.29	53.92	46.46	0.452(0.322–0.634)
Depressed	None	117	41.00	75.21	54.50	47.85	1	0.087
	Seldom	131	64.00	79.31	53.23	50.03	1.078(0.752–1.544)	0.684
	Often	96	19.00	63.35	23.35	15.17	1.458(1.017–2.089)	0.040*
sleep time (hours/day)	Less 5	108	17.00	60.00	30.71	24.43	1	0.001*
	5–7	168	33.00	77.91	46.90	42.09	0.681(0.503–0.922)	0.013*
	More 8	68	155.71	83.82	65.09	57.99	0.461(0.293–0.724)	0.001*
Rehabilitation exercise (times/week)	None	168	20.00	65.37	34.88	27.97	1	0.211
	1–3	102	35.00	77.45	49.67	46.66	1.042(0.741–1.467)	0.813
	3–5	36	50.00	83.10	56.92	44.15	0.940(0.555–1.592)	0.818
	> 5	38	81.00	89.47	70.20	66.40	0.548(0.306–0.983)	0.044*

LncRNA-related SNP and prognosis of patients with GC

In this study, a total of 10 lncRNA polymorphic loci were examined by univariate analysis. The results are shown in Table 4 and Table 5. Interestingly, at CCDC26 rs2579878 locus, we found that the survival rate of GC patients with C allele was higher than that of patients with T allele, which could also be seen in the codominance model. Similarly, at LINC02122 rs10036719 locus, the survival rate of patients with A allele of was higher than that of G allele, which was also reflected in its recessive model.

Table 4

The relations between polymorphism site of *CCDC26* rs2579878 And prognosis of patients with GC
rs2579878	N	MST (M)	Survival rate(%)					HR (95%CI)	P
rs2579878	N	MST (M)	1-year		3-year	5-year		HR (95%CI)	P
Codominance
TT	141	27.00	70.92	40.59			35.25	1	0.087
TC	159	31.00	73.50	46.77			40.22	0.830(0.612–1.124)	0.228
CC	44	136.05	81.61	57.08			49.71	0.581(0.354–0.954)	0.032*
Allele gene
T	441	28.00	71.85	42.82			37.08	1	0.030*
C	247	36.00	76.38	50.19			43.52	0.790(0.638–0.977)
Dominant model
TT	141	27.00	70.92	40.59			35.25	1	0.080
TC + CC	203	34.00	75.25	48.82			42.22	0.771(0.576–1.032)
Recessive model
TT + TC	300	29.00	72.29	43.87			37.93	1	0.066
CC	44	136.05	81.61	57.08			49.71	0.645(0.404–1.03)
*P was adjusted according to age, sex, TNM stage, operation and chemotherapy.

Table 5

The relations between polymorphism site of *LINC02122* rs10036719 And prognosis of patients with GC
rs10036719	N	MST (M)	Survival rate(%)			HR (95%CI)	P
rs10036719	N	MST (M)	1-year	3-year	5-year	HR (95%CI)	P
Codominance
GG	45	19.00	66.67	45.67	41.70	1	0.093
AG	155	28.00	75.41	42.03	36.89	0.851(0.555–1.303)	0.458
AA	144	34.00	73.52	48.99	41.31	0.654(0.422–1.012)	0.057
Allele gene
G	245	28.00	72.19	43.27	38.54	1	0.029*
A	443	31.00	74.18	46.59	39.79	0.796(0.649–0.977)
Dominant model
GG	45	19.00	66.67	45.67	41.70	1	0.161
AG + AA	299	30.00	74.50	45.42	39.05	0.749(0.5-1.122)
Recessive model
GG + AG	200	28.00	73.43	42.78	37.88	1	0.042*
AA	144	34.00	73.52	48.99	41.31	0.742(0.556–0.989)
*P was adjusted according to age, sex, TNM stage, operation and chemotherapy.

Stratified analysis of SNP

After combined stratification of age and TNM stage, we found 3 polymorphic loci out of 10 lncRNA-related SNPs to be associated with the prognosis of GC. From both codominant and recessive models, we found that the CC genotype at LINC00687 rs2795025 to be a risk factor for poor prognosis in patients younger than 65 years of age with advanced GC. The AA or AG genotype at LINC02122 rs10036719 and the GG genotype at LINC02122 rs12516079 are protective factors for the prognosis of patients older than 65 years of age with early or middle stage GC. (Table 6–8)

Table 6

Stratified Analysis of LINC00687 Polymorphism rs2795025 and prognosis of GC
rs2795025	≦ 65						> 65
	Early and middle stage			Late stage			Early and middle stage			Late stage
	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)
Codominance
TT	46	73.67	1	19	10.00	1	81	157.29 1		56	9.00	1
TC	35	106.04 1.157		9	21.00	0.696	38	75.00	0.901	32	8.00	0.860
			(0.525–2.554)			(0.282–1.717)			(0.475–1.709)			(0.538–1.374)
CC	9	73.00	1.446	2	4.00	5.911	8	27.00	1.936	9	11.00	1.087
			(0.406–5.151)			(1.103–31.671)			(0.675–5.553)			(0.503–2.348)
Dominant model
TT	46	73.67	1	19	10.00	1	81	157.29 1		56	9.00	1
TC + CC	44	73.00	1.209	11	19.00	0.872	46	75.00	1.029	41	9.00	0.901
TC + CC			(0.576–2.540)			(0.380–2.002)			(0.572–1.850)			(0.584–1.391)
Recessive model
TT + TC	81	103.76 1		28	13.00	1	119	81.00	1	88	8.00	1
CC	9	73.00	1.364	2	4.00	6.611	8	27.00	1.996	9	11.00	1.148
			(0.400-4.648)			(1.257–34.775)			(0.707–5.634)			(0.541–2.437)

Table 7

Stratified Analysis of LINC02122 Polymorphism rs10036719 and prognosis of GC
rs10036719	≦ 65						> 65
	Early and middle stage			Late stage			Early and middle stage			Late stage
	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)
Codominance
GG	13	73.00	1	4	6.00	1	17	33.00	1	11	8.00	1
AG	38	105.24	1.206	13	10.00	0.720	61	81.00	0.592	43	9.00	0.935
			(0.384–3.783)			(0.202–2.560)			(0.279–1.256)			(0.457–1.914)
AA	39	72.25	1.206	13	19.00	0.554	49	115.44	0.377	43	10.00	0.676
			(0.387–3.752)			(0.148–2.081)			(0.165–0.859)			(0.333–1.374)
Dominant model
GG	13	73.00	1	4	6.00	1	17	33.00	1	11	8.00	1
AG + AA	77	101.82	1.206	26	12.00	0.643	110	147.15	0.490	86	9.00	0.780
			(0.414–3.512)			(0.189–2.181)			(0.241–0.997)			(0.398–1.530)
Recessive model
GG + AG	51	96.94	1	17	10.00	1	78	75.00	1	54	8.00	1
AA	39	72.25	1.050	13	19.00	0.727	49	115.44	0.562	43	10.00	0.713
			(0.503–2.192)			(0.317–1.668)			(0.302–1.046)			(0.461-1.100)

Table 8

Stratified Analysis of LINC02122 Polymorphism rs12516079 and prognosis of GC
rs12516079	≦ 65						> 65
	Early and middle stage			Late stage			Early and middle stage				Late stage
	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)	N	MST (M)	HR (95%CI)		N	MST (M)	HR (95%CI)
Codominance
AA	11	75.08	1	2	4.00	1	15	75.00	1		10	8.00	1
AG	35	105.48	1.665	13	13.00	0.511	57	46.00	0.997		42	8.00	1.022
			(0.459–6.031)			(0.103–2.528)			(0.432–2.297)				(0.483–2.164)
GG	44	71.86	1.420	15	12.00	0.522	55	117.32		0.535	45	10.00		0.729
			(0.400-5.035)			(0.107–2.541)			(0.217–1.32)				(0.350–1.519)
Dominant model
AA	11	75.08	1	2	4.00	1	15	75.00	1		10	8.00	1
AG + GG	79	101.38	1.522	28	12.00	0.517	112	81.00	0.767		87	9.00	0.837
			(0.452–5.127)			(0.112–2.391)			(0.343–1.715)				(0.414–1.692)
Recessive model
AA + AG	46	73.00	1	15	13.00	1	72	46.00	1		52	8.00	1
GG	44	71.86	0.965	15	12.00	0.941	55	117.32 0.536			45	10.00	0.717
			(0.462–2.016)			(0.416–2.132)			(0.291–0.987)				(0.465–1.106)

Combinatory effects of GC-related lncRNA polymorphism

Based on the above analysis, the TT genotype of rs2579878, the CC genotype of rs2795025, the GG genotype of rs10036719, and the AG genotype of rs12516079, all lead to poor prognosis of GC. Interestingly, patients carrying both rs2795025 CC and rs12516079 AG alleles have higher risk of poor prognosis, while other gene polymorphisms have no significant combinatory effects, as shown in Table 9. (P > 0.05)

Table 9

Combined action of gene polymorphism loci
SNPs loci	Num. of bad genotypes	N	MST(M)	HR(95%CI)	P
rs2579878༊rs2795025	0	183	36.00	1	0.062
	1	98	24.00	1.386(1.036–1.855)	0.028*
	2	8	36.00	1.749(0.704–4.343)	0.228
rs2579878༊rs10036719	0	178	35.00		0.087
	1	146	24.00	1.349(1.004–1.814)	0.047*
	2	20	33.00	1.546(0.854–2.798)	0.150
rs2579878༊rs12516079	0	197	34.00	1	0.247
	1	84	30.00	1.206(0.859–1.692)	0.280
	2	63	23.00	1.334(0.926–1.921)	0.122
rs2795025༊rs10036719	0	279	31.00	1	0.121
	1	57	18.00	1.393(0.973–1.996)	0.070
	2	8	73.00	1.764(0.645–4.824)	0.269
rs2795025༊rs12516079	0	180	33.00	1	0.054
	1	153	30.00	1.227(0.915–1.644)	0.171
	2	11	11.00	2.122(1.089–4.138)	0.027*
rs10036719*rs12516079	0	160	34.00	1	0.056
	1	176	28.00	1.310(0.981–1.748)	0.067
	2	8	9.00	2.194(0.979–4.917)	0.056

Multifactor analysis

The 10 lncRNA polymorphic loci were included in multivariate COX regression analysis, as shown in Table 10. In model 1, none of the 10 polymorphic loci were associated with the prognosis of GC. After adjusting for possible confounding factors, 2 of 10 polymorphic loci were associated with the prognosis of GC patients. Among them, AG and AA genotypes of rs10036719 were shown to be protective factors for the prognosis of GC, while AG genotype of rs12516079 was shown to associate with poor prognosis of GC patients.

Table 10

COX regression Analysis of Survival in patients with GC
SNP loci	Genotype	Model 1^a		Model 2^b
SNP loci	Genotype	HR (95%CI)	P	HR (95%CI)	P
rs2579878	TT	1	0.253	1	0.448
rs2579878	TC	0.864(0.638–1.171)	0.346	0.910(0.65–1.272)	0.580
	CC	0.667(0.403–1.102)	0.114	0.712(0.419–1.21)	0.210
rs5829142	INs	1	0.822	1	0.795
	DEl/INS	1.180(0.677–2.054)	0.560	1.085(0.589–1.998)	0.794
	DEL	1.194(0.679–2.097)	0.538	0.971(0.524–1.799)	0.927
rs11617815	AA	1	0.949	1	0.681
	GA	0.946(0.526–1.702)	0.853	0.858(0.458–1.606)	0.631
	GG	0.921(0.51–1.664)	0.785	0.778(0.401–1.508)	0.457
rs1348758	GG	1	0.851	1	0.485
	TG	0.975(0.624–1.523)	0.911	0.873(0.542–1.404)	0.574
	TT	0.898(0.561–1.437)	0.653	0.754(0.461–1.232)	0.260
rs2795025	TT	1	0.453	1	0.314
	TC	0.917(0.668–1.259)	0.593	0.975(0.701–1.357)	0.882
	CC	1.304(0.776–2.193)	0.316	1.484(0.872–2.525)	0.145
rs9809325	AA	1	0.925	1	0.227
	AG	0.976(0.725–1.314)	0.874	1.065(0.785–1.446)	0.686
	GG	0.894(0.508–1.572)	0.697	1.684(0.930–3.049)	0.085
rs10036719	GG	1	0.210	1	0.017*
	AG	0.503(0.235–1.078)	0.077	0.282(0.116–0.680)	0.005*
	AA	0.520(0.184–1.468)	0.217	0.226(0.067–0.767)	0.017*
rs61894277	TT	1	0.330	1	0.809
	TC	1.251(0.923–1.696)	0.150	0.902(0.650–1.251)	0.536
	CC	1.029(0.603–1.755)	0.917	0.893(0.512–1.557)	0.690
rs56093317	GG	1	0.471	1	0.086
	AG	1.167(0.86–1.585)	0.321	1.197(0.864–1.660)	0.280
	AA	0.895(0.524–1.528)	0.683	0.628(0.355–1.112)	0.110
rs12516079	AA	1	0.137	1	0.075
	AG	2.330(0.991–5.478)	0.052	2.999(1.155–7.785)	0.024*
	GG	1.943(0.643–5.876)	0.239	3.541(0.955–13.126)	0.059
a. Model 1 does not adjust. b. Model 2 was adjusted according to the basic characteristics and the health habits and lifestyle after illness.

Nomogram Prediction model

We incorporated the polymorphic sites with statistical significance from the above multivariate analysis to establish a nomogram and further evaluate their prediction performance. As shown in Fig. 1, TNM staging accounts for the largest proportion in the chart and has the greatest impact on the prognosis. The C index of the whole nomogram is 0.762, indicating that the predictive ability of the model is moderate, as show as the calibration curve (Fig. 2).

The relationship between expression of the lncRNA and the prognosis of GC

In the above studies, we found that four polymorphism loci of three lncRNA were associated with the survival outcome of GC patients. SNPs often affect disease by affecting gene expression, thus we downloaded the RNA-seq data of Asian gastric adenocarcinoma patients from the TCGA database to further explore the relationship between gene expression and GC prognosis. The results showed that the high express of both CCDC26 and LINC02122 were shown to be associated with the poor survival status of GC patients (Fig.3).

LncRNA is an important class of non-coding RNA with a potential carcinogenic role affecting cell proliferation and apoptosis[17–20]. It can also affect the prognosis of cancer by acting on key signaling pathways and alter the invasiveness of cancer cells [21–24]. In this study, we deeply explored the relationship between lncRNA polymorphisms and GC prognosis. Through univariate COX analysis and stratified analysis, we found four lncRNA polymorphism loci associated with gastric cancer prognosis. Afterwards, we further explored their combined effects through conjoint analysis, and evaluated their predictive performance through multivariate analysis. Finally, we explored the association between lncRNA expression and gastric cancer prognosis in the TCGA database.

In this study, we found that polymorphism of CCDC26 was associated with the prognosis of GC. CCDC26, or coiled-coil domain-containing 26, is a long non-coding RNA located on 8q24 chromosome. Previous studies have shown that lncRNA CCDC26 levels were correlated with tumor size, tumor number, and reduced overall survival in pancreatic cancer[25]. CCDC26 participates in cancer cell growth and apoptosis by regulating the expression of PCNA and Bcl2[26]. CCDC26 promotes thyroid cancer malignant progression via miR-422a/EZH2/Sirt6 axis [27]. Silencing of CCDC26 can strongly reduce the wound closing rate and the number of invasive cells, and further regulates the growth and metastasis of gliomas [28]. CCDC26 can affect the drug sensitivity in gastrointestinal stromal tumors and affect the prognosis[29]. Besides, scholars also found the polymorphism of CCDC26 related to cancer risk [30–32]. However, the correlation between the polymorphism of CCDC26 and GC prognosis has not been found yet. Our research found that the patients with C mutation at CCDC26 rs2579878 locus have higher survival probability and the expression of CCDC26 can affect the survival of patients. It is possible that the CCDC26 polymorphism in GC inhibits its expression, and thereby reducing the number of invasive cells and improve the prognosis. However, mechanisms underlying CCDC26 polymorphism and its clinical significance in GC remain to be further investigated.

LINC02122 is located on chromosome 5q13.3. The gene IQGAP2 located on the same site is reported to be a tumor suppressor gene for prostate cancer[33]. Previous studies found 5q13.3 deletions in myeloid tumors [34]. Other studies have also showed that loss of heterozygosity in 5q13.3 is related to the progression and metastasis of colon cancer [35]. Genome-wide DNA copy number analysis showed that focal recurrent genomic losses were observed in chromosome regions 5q13.3 of desmoplastic infantile ganglioglioma (DIG) and desmoplastic infantile astrocytoma (DIA)[36]. Furthermore, Losses of copy number at 5q13.3-q35.3 is correlated with higher histological grade of urothelial carcinomas (UCs)[37]. In our study, we found LINC02122 polymorphisms sites rs10036719 and rs12516079 to be associated with the prognosis of GC and the expression of LINC02122 also plays a role in patient survival. In addition, LINC02122 rs10036719 A mutant allele and rs12516079 G mutant allele are beneficial to the survival of GC patients. Multivariate analysis also showed that LINC02122 rs10036719 A mutant allele to be an independent factor for the prognosis of GC. However, there is no report about the relationship between the polymorphism of chromosome 5q13.3 and GC. Our research suggests for the first time that it is related to the development of GC. More research is needed to further explore the its role and mode of action in GC in the future.

We found that polymorphic site rs2795025 of LINC00687 is a risk factor for poor prognosis of GC. A study based on weighted gene co-expression network analysis (WGCNA) and the linear models for microarray data analysis (LIMMA) found that LINC00687 could be one of the important hub nodes involved in the pathogenesis of periodontitis [38]. LINC00687 is located in 20p12.2. Genome-wide analysis of genetic variants suggested that this locus could influence the effectiveness of platinum-based chemotherapy of small-cell lung cancer (SCLC) [39]. Although no research report on the relationship between LINC00687 polymorphism and GC has been found, our research has shown that it is involved in the development process of GC, which is conducive to further research on the complex regulatory mechanism of GC.

Consistent with previous research, we found that age, TNM stage, operation and chemotherapy were significantly related to the prognosis of GC[40–42]. In addition, changes in drinking habits are adverse factors in the prognosis of GC. Other factors associated with the survival time of GC patients include self-care, depression, sleep duration, and exercise.

Although the findings of our study can provide clues for the study of GC mechanism and the exploration of regulatory networks, there are ethnic and regional differences in gene polymorphisms, so the generalization of the conclusions of this study needs to be considered. Since different detection techniques and methods in the genetic testing process may also cause different detection results and the sample size of our research is still small, its conclusion needs to be further verified in a large-sample, multi-center study.

In conclusion, we found four lncRNA-related polymorphisms that are closely related to the prognosis of GC by multivariate and stratified analyses. In addition, we found that the interaction between polymorphous loci rs2795025 and rs12516079 could increase the risk of poor prognosis of GC. In order to further visualize the results of multivariate analysis, we included gender, age, TNM staging, surgery, chemotherapy and statistically significant polymorphic sites rs10036719 and rs12516079 in multivariate analysis to draw a nomogram. According to the nomogram, we calculated the total score according to various indexes of patients with GC, and then speculate their 1-, 3-, and 5-year survival rates. The nomogram map proved to be able to successfully predict the prognosis of patients with GC, and therefore could become one of the prognostic markers for future clinical studies. Our study is helpful to understand the development trend of the GC, predict the prognosis of patients, help clinicians make corresponding treatment decisions, and ultimately achieve the purpose of prolonging the life of patients and improving the quality of life of patients. At present, the molecular epidemiological research is still the focus of current research, we expect that with the continuous expansion and deepening of research, the prognostic factors of GC will continue to be clarified, so as to make a more accurate judgment on the prognosis of patients and make individualized treatment possible. Due to the differences in race, region, technology, and detection methods, it is desirable to verify our current results with a larger population.

GC: Gastric cancer; SNP: single nucleotide polymorphism; lncRNA: Long-chain non-coding RNA; MALDI-TOF-MS: matrix-assisted laser desorption ionization time-of-flight mass spectrometry; H-W: Hardy Weinberg genetic equilibrium; CCDC26: The coiled-coil domain-containing 26.

Ethics approval and consent to participate: Ethical approval for the study was obtained from the Biomedical Research Ethics Committee of Fujian Medical University, China (No. 97,2014).

Consent for publication: Not applicable.

Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests：The authors declare that they have no competing interests.

Funding: This work was supported by Fujian Natural Science Foundation (grant number 2015J01673); Fujian Natural Science Foundation (grant number 2017J01811) and Fujian Medical Innovation Project (grant number 2016-CX-41). The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Authors' contributions: Y.P.L.：formal analysis and writing—original draft preparation；S.F.Y.：investigation；X.J.L.: methodology; Y.L.W.：Writing - Review & Editing; S.M.J.：visualization；S.L.K.：visualization；Y.J.: Software and Project administration; J.J.X.：Project administration; C.Z.H.：Project administration; P.X.L.：Project administration; B.Y.L.：Supervision; C.C.W.: Writing - Review & Editing, Supervision. All authors read and approved the final manuscript.

Acknowledgements: Not applicable.

Sung, H.; Ferlay, J.; Siegel, R. L., et al., Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021, 71 (3), 209-249.
Bonelli, P.; Borrelli, A.; Tuccillo, F. M., et al., Precision medicine in gastric cancer. World J Gastrointest Oncol 2019, 11 (10), 804-829.
Gonzalez-Hormazabal, P.; Romero, S.; Musleh, M., et al., IL-8-251T>A (rs4073) Polymorphism Is Associated with Prognosis in Gastric Cancer Patients. Anticancer Res 2018, 38 (10), 5703-5708.
Ananiev, J.; Manolova, I.; Aleksandrova, E., et al., Impact of TGF-beta1 expression and -509C>T polymorphism in the TGF-beta1 gene on the progression and survival of gastric cancer. Pol J Pathol 2017, 68 (3), 234-240.
Zhang, L.; Liu, Q.; Wang, F., Association Between miR-149 Gene rs2292832 Polymorphism and Risk of Gastric Cancer. Archives of medical research 2018, 49 (4), 270-277.
Jiang, J.; Jia, Z.; Cao, D., et al., Association of the miR-146a rs2910164 polymorphism with gastric cancer susceptibility and prognosis. Future oncology (London, England) 2016, 12 (19), 2215-26.
Li, Y.; Zhang, Z.; Yang, Y., et al., Long Noncoding RNA Transcript Antisense RNA Gene rs17720428 Single Nucleotide Polymorphism Is Associated with Gastric Cancer Risk and Prognosis. Genet Test Mol Biomarkers 2020, 24 (1), 38-46.
Wang, B.-G.; Xu, Q.; Lv, Z., et al., Association of twelve polymorphisms in three onco-lncRNA genes with hepatocellular cancer risk and prognosis: A case-control study. World J Gastroenterol 2018, 24 (23), 2482-2490.
Lu, S.; Su, Z.; Fu, W., et al., Altered expression of long non-coding RNA GAS5 in digestive tumors. Biosci Rep 2019, 39 (1).
Tang, Y.; Wang, Y.; Wang, X., et al., A Genetic Variant of rs145204276 in the Promoter Region of Long Noncoding RNA GAS5 Is Associated With a Reduced Risk of Breast Cancer. Clin Breast Cancer 2019, 19 (3), e415-e421.
Tao, R.; Hu, S.; Wang, S., et al., Association between indel polymorphism in the promoter region of lncRNA GAS5 and the risk of hepatocellular carcinoma. Carcinogenesis 2015, 36 (10), 1136-1143.
Deng, Z.-H.; Yu, G.-S.; Pan, B., et al., Rs145204276 and rs4759314 affect the prognosis of prostate cancer by modulating the GAS5/miR-1284/HMGB1 and HOTAIR/miR-22/HMGB1 signalling pathways. Artif Cells Nanomed Biotechnol 2020, 48 (1), 435-442.
Yang, M.-L.; Huang, Z.; Wang, Q., et al., The association of polymorphisms in lncRNA- with hepatocellular cancer risk and prognosis. Biosci Rep 2018, 38 (5).
Dong, X. Y.; Gao, W. Y.; Lv, X. L., et al., Association between lncRNA GAS5, MEG3, and PCAT-1 Polymorphisms and Cancer Risk: A Meta-Analysis. Dis Markers 2020, 2020.
Hong, J. H.; Jin, E. H.; Chang, I. A., et al., Association Between lncRNA HULC rs7763881 Polymorphism and Gastric Cancer Risk. Pharmacogen Pers Med 2020, 13, 121-126.
Fuzhi, L. Association study of lincRNA SNPs and gastric cancer based on gene chip and construction of risk prediction model. Master, Fujian Medical University 2018.
Bai, N.; Ma, Y.; Zhao, J., et al., Knockdown of lncRNA HCP5 Suppresses the Progression of Colorectal Cancer by miR-299-3p/PFN1/AKT Axis. Cancer Manag Res 2020, 12, 4747-4758.
Dai, G.; Huang, C.; Yang, J., et al., LncRNA SNHG3 promotes bladder cancer proliferation and metastasis through miR-515-5p/GINS2 axis. J Cell Mol Med 2020.
Duan, D. M.; Zhang, L.; Hua, F., LncRNA UCA1 inhibits proliferation and promotes apoptosis of cervical cancer cells by regulating beta-catenin/TCF-4. Eur Rev Med Pharmacol Sci 2020, 24 (11), 5963-5969.
Duan, Q.; Cai, L.; Zheng, K., et al., lncRNA KCNQ1OT1 knockdown inhibits colorectal cancer cell proliferation, migration and invasiveness via the PI3K/AKT pathway. Oncol Lett 2020, 20 (1), 601-610.
Pucci, P.; Venalainen, E.; Alborelli, I., et al., LncRNA HORAS5 promotes taxane resistance in castration-resistant prostate cancer via a BCL2A1-dependent mechanism. Epigenomics 2020.
Guo, C.; Shi, H.; Shang, Y., et al., LncRNA LINC00261 overexpression suppresses the growth and metastasis of lung cancer via regulating miR-1269a/FOXO1 axis. Cancer Cell Int 2020, 20, 275.
Lu, C.; Xie, T.; Guo, X., et al., LncRNA DSCAM-AS1 Promotes Colon Cancer Cells Proliferation and Migration via Regulating the miR-204/SOX4 Axis. Cancer Manag Res 2020, 12, 4347-4356.
Peperstraete, E.; Lecerf, C.; Collette, J., et al., H19Enhancement of Breast Cancer Cell Aggressiveness by lncRNA and its Mir-675 Derivative: Insight into Shared and Different Actions. Cancers 2020, 12 (7).
Peng, W.; Jiang, A., Long noncoding RNA CCDC26 as a potential predictor biomarker contributes to tumorigenesis in pancreatic cancer. Biomed Pharmacother 2016, 83, 712-717.
Peng, W.; Jiang, A., Long noncoding RNA CCDC26 as a potential predictor biomarker contributes to tumorigenesis in pancreatic cancer. Biomed Pharmacother 2016, 83, 712-717.
Ma, X.; Li, Y.; Song, Y., et al., Long Noncoding RNA CCDC26 Promotes Thyroid Cancer Malignant Progression via miR-422a/EZH2/Sirt6 Axis. Onco Targets Ther 2021, 14, 3083-3094.
Wang, S.; Hui, Y.; Li, X., et al., Silencing of lncRNA CCDC26 Restrains the Growth and Migration of Glioma Cells In Vitro and In Vivo via Targeting miR-203. Oncol Res 2018, 26 (8), 1143-1154.
Cao, K.; Li, M.; Miao, J., et al., CCDC26 knockdown enhances resistance of gastrointestinal stromal tumor cells to imatinib by interacting with c-KIT. Am J Transl Res 2018, 10 (1), 274-282.
Wibom, C.; Späth, F.; Dahlin, A. M., et al., Investigation of established genetic risk variants for glioma in prediagnostic samples from a population-based nested case-control study. Cancer Epidemiol Biomarkers Prev 2015, 24 (5), 810-816.
Adel Fahmideh, M.; Lavebratt, C.; Schüz, J., et al., CCDC26, CDKN2BAS, RTEL1 and TERT Polymorphisms in pediatric brain tumor susceptibility. Carcinogenesis 2015, 36 (8), 876-882.
Wang, X.; Luo, T.; Ruan, M., et al., Association of the CCDC26 rs4295627 polymorphism with the risk of glioma: Evidence from 7,290 cases and 11,630 controls. Mol Clin Oncol 2016, 4 (5), 878-882.
Xie, Y.; Zheng, L.; Tao, L., Downregulation of IQGAP2 Correlates with Prostate Cancer Recurrence and Metastasis. Transl Oncol 2019, 12 (2), 236-244.
Castro, P. D.; Liang, J. C.; Nagarajan, L., Deletions of chromosome 5q13.3 and 17p loci cooperate in myeloid neoplasms. Blood 2000, 95 (6), 2138-43.
Zhang, F.; Zhou, C.; Ling, Y., et al., [Allelic analysis on chromosome 5 in sporadic colorectal cancer patients]. Zhonghua Zhong Liu Za Zhi 2002, 24 (5), 458-60.
Gessi, M.; Zur Muhlen, A.; Hammes, J., et al., Genome-wide DNA copy number analysis of desmoplastic infantile astrocytomas and desmoplastic infantile gangliogliomas. J Neuropathol Exp Neurol 2013, 72 (9), 807-15.
Nishiyama, N.; Arai, E.; Nagashio, R., et al., Copy number alterations in urothelial carcinomas: their clinicopathological significance and correlation with DNA methylation alterations. Carcinogenesis 2011, 32 (4), 462-469.
Jin, S. H.; Zhou, R. H.; Guan, X. Y., et al., Identification of novel key lncRNAs involved in periodontitis by weighted gene co-expression network analysis. J Periodontal Res 2020, 55 (1), 96-106.
Wu, C.; Xu, B.; Yuan, P., et al., Genome-wide examination of genetic variants associated with response to platinum-based chemotherapy in patients with small-cell lung cancer. Pharmacogenet Genomics 2010, 20 (6), 389-95.
McTiernan, A.; Friedenreich, C. M.; Katzmarzyk, P. T., et al., Physical Activity in Cancer Prevention and Survival: A Systematic Review. Med Sci Sports Exerc 2019, 51 (6), 1252-1261.
Minami, Y.; Kanemura, S.; Oikawa, T., et al., Associations of cigarette smoking and alcohol drinking with stomach cancer survival: A prospective patient cohort study in Japan. Int J Cancer 2018, 143 (5), 1072-1085.
Okada, E.; Ukawa, S.; Nakamura, K., et al., Demographic and lifestyle factors and survival among patients with esophageal and gastric cancer: The Biobank Japan Project. J Epidemiol 2017, 27 (3), S29-S35.

No competing interests reported.

LncRNA Polymorphism Affects the Prognosis of Gastric Cancer

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusions

List Of Abbreviations

Declarations

References

Competing Interests

Status:

Version 1