Introduction

Cancer has caused social and public problems that cannot be ignored, with huge economic losses and mental burdens to people all over the world every year [1, 2]. Based on the 2021 Cancer Statistics Report, the numbers of individuals newly diagnosed with cancer and dying from cancer in 2020 will be approximately 19.84 million and 10 million, respectively [3, 4]. Despite the well-being and satisfaction of cancer patients having improved to a certain extent alongside advancements in molecular biology technology and medical care, the 5-year survival rate for cancer patients has remained unsatisfactory [5, 6]. The main reason is that classic treatment methods such as chemotherapy, radiation therapy, hormone therapy and targeted therapeutics are already in their prime, and it is difficult for these methods to improve survival [7,8,9,10]. Therefore, novel therapeutic targets aimed at improving the prognosis of cancer patients are urgently needed [11,12,13].

Over the past few years, the employment of high-throughput sequencing techniques and advancements in molecular biology have gradually revealed an increasing number of genes that are intricately linked to cancer. Additionally, a growing body of evidence has confirmed the involvement of numerous noncoding RNAs in the initiation and progression of cancer [14, 15]. Noncoding RNAs are a class of small molecular compounds that lack the ability to encode proteins [16,17,18]. They are called the "useless product" of genetic material and account for 95–98% of human genetic material [19]. Although noncoding RNAs do not perform the biological function of encoding proteins, there is substantially more evidence that noncoding RNAs can exert their control over the biological behaviour of cells by influencing the expression of cell-coding genes at various levels, encompassing gene transcription, post-transcriptional translation, and epigenetic regulation [20,21,22]. Noncoding RNAs can affect the proliferation, invasion and apoptosis of tumour cells, thereby affecting the progression of tumours. For example, Professor Yang showed that low expression of lncRNA-BANCR can significantly stimulate the growth and motility of lung cancer cells and suppress programmed cell death, thereby contributing to the initiation and progression of lung cancer [23]. Professor Yan stated that high expression of lnc-SNHG6 can significantly suppress the programmed cell death (apoptosis) of gastric cancer cells, stimulate their proliferation, migration, and invasion, and exhibit a strong correlation with an unfavourable prognosis [24]. Noncoding RNA are being considered by more and more researchers as promising targets and prognostic indicators for future cancer therapies.

Lnc-SNHG5 is a category of low-molecular-weight compounds characterized by more than 200 nucleotide units that lack protein coding ability [25]. An increasing number of studies have shown that SNHG5 is differentially expressed across various tumour cell types and is implicated in the aetiology and progression of diverse cancers [26, 27]. For example, Professor Wei showed that SNHG5 is abundantly expressed in oesophageal cancer tissues and that elevated SNHG5 levels can enhance the proliferative and migratory potential of oesophageal squamous cell carcinoma (ESCC) cells, suppress cellular apoptosis, and consequently promote the progression of ESCC [28]. Professor Kang reported that SNHG5 is overexpressed in lung cancer, where its high expression facilitates the movement and infiltration of lung cancer cells while suppressing their apoptosis. Furthermore, numerous studies have increasingly indicated a significant association between elevated SNHG5 expression and poor lung cancer prognosis [29]. Inconsistencies across different studies involving relatively small numbers of patients have made conclusions questionable; even though some outcome indicators have suggested that SNHG5 is a poor prognostic marker, the differences did not reach statistical significance. Hence, the objective of this investigation was to conduct a meta-analysis to investigate the potential correlation between the expression level of SNHG5 and cancer prognosis.

Materials and methods

Inclusion of appropriate literature

Utilizing the reporting guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) as the foundation for our reporting format, a comprehensive search was performed by browsing relative databases such as PubMed, Embase, the Web of Science, the Cochrane Library, Google Scholar, the China National Knowledge Infrastructure (CNKI) and the Wanfang Database from the establishment of the database to January 1, 2023. The retrieval strategies used in this study were as follows: “Small nucleolar RNA host gene 5” OR “lncRNA Small nucleolar RNA host gene 5” OR “lncRNA SNHG5” OR “SNHG5” OR “lncSNHG5”) AND “cancer” OR “carcinoma” OR “prognosis” OR “survival” OR “survival prognosis.” We also consulted the references of the included publications in detail to obtain useful and appropriate publications.

Inclusion and exclusion criteria

Publications that fulfilled the following criteria were deemed appropriate for inclusion in this meta-analysis: (1) the fundamental purpose of the literature review was to evaluate the relationship between SNHG5 expression levels and cancer prognosis; (2) patients were divided into two distinct groups based on their expression level (high or low); (3) the research subjects were limited to human patients; (4) provided enough raw data to be extracted. Studies with the following characteristics were considered unsuitable for inclusion in this meta-analysis: (1) lacked sufficient data; (2) the participants were animals; (3) literature reviews, meta-analyses, case reports, conference summaries, and research not officially published; (4) non-English language publications.

NOS score of included studies

The Newcastle-Ottawa Scale (NOS) score, which includes eight items, was utilized for evaluating the overall quality of the studies included [30]. Two researchers meticulously and independently assessed the quality of every included article, taking into consideration the aforementioned NOS score. If the evaluation results were not consistent, an agreement was reached through discussion or a third researcher was consulted for discussion and confirmation. The total NOS score ranged from 0 to 9. Articles with a score of 6 or more were considered high-quality documents and suitable for inclusion in this study. Articles with a score less than 6 points were considered low-quality studies and were excluded from this investigation.

Data extraction

The useful raw data were obtained by two researchers independently, and the detailed information included the name of the primary author, the year the article was published, the total number of patients involved, the reference gene, the cutoff value, and the country to which the patient belonged. We also extracted the main outcome indicators, including overall survival (OS) and relapse-free survival (RFS). Furthermore, secondary outcome indicators, including TNM stage, lymph node metastasis (LNM) status, distant metastasis(DM) status, tumour dimensions, histological grade, depth of infiltration, chronological age and sex, were obtained. If the survival data provided in the publication included multivariate analysis and univariate analysis, only multivariate analysis was obtained; for example, if the study only included the survival curve, then the detailed survival data were obtained utilizing Engauge Digitizer V 4.1 software, and the study included the time-dependent survival rates of both the high- and low-expression groups of SNHG5 [31].

Statistics and analysis

RevMan V 5.4 software and STATA V 12.0 software were used to perform the statistical analysis of this meta-analysis. Patients were categorized into either a high-expression group or a low-expression group according to the original literature reports. The combination of the odds ratio (OR) with 95% confidence interval (CI) was used to evaluate the associations between SNHG5 expression and TNM stage, LNM, DM, tumour size, etc. The combination of the hazard ratio (HR) and 95% confidence interval (CI) was used to evaluate the relationship between SNHG5 expression and cancer prognosis, including OS and RFS. If I2 (I-square) < 50% and P > 0.05, the result was considered to indicate insignificant heterogeneity, and a fixed effects model was used. If I2 > 50% and P < 0.05 were significant, significant heterogeneity was considered. A random effects model was used, and subgroup analysis was conducted based on cancer type (digestive system and nondigestive tract), number of patients (fewer than 60 patients and no less than 60 patients), follow-up month (fewer than 60 and no less than 60), cutoff value (mean and median), and data analysis method (multivariate analysis and univariate analysis). Sensitivity analysis was conducted using STATA software to assess whether the results of individual studies had a significant impact on the overall findings. Additionally, Begg's analysis was employed to detect any significant publication bias in the original study.

Results

Characteristics of the enrolled publications

After the systematic and detailed searches, 127 publications were initially obtained, 32 duplicate documents were found, 68 articles were excluded for not exploring the relationship between SNHG5 expression and cancer prognosis, 6 publications were excluded for involving animal experimentation, 8 articles lacked sufficient data, and 2 non-English papers were also excluded. Finally, 11 suitable investigations involving 721 patients were included in this meta-analysis (Fig. 1). The cancer types included bladder cancer [32], hepatocellular carcinoma [33], cervical cancer [34], osteosarcoma [35], nasopharyngeal carcinoma [36], non-small cell lung cancer [29], oesophageal cancer [28], diffuse large B-cell lymphoma [37], and gastric cancer [38, 39]. All the patients were Chinese, and the number of patients in each study varied between 32 and 90; seven [7] papers provided survival data (Table 1). According to the NOS score, the enrolled studies ranged from 7 to 8, indicating that all the papers were suitable for enrolment and analysis (Table 2).

Fig. 1
figure 1

Literature search and inclusion process

Full size image
Table 1 Basic features of the publications included in this meta-analysis (n = 11)
Full size table
Table 2 Quality assessment of eligible studies Newcastle–Ottawa scale (NOS) score
Full size table

The correlation between SNHG5 expression and survival outcome

Seven studies involving a total of 425 patients were obtained to evaluate the correlation between SNHG5 expression and cancer prognosis. A positive correlation was revealed between elevated SNHG5 expression and poor survival outcome (HR = 1.90, 95% CI 0.87–4.15) (Fig. 2), but the correlation was not statistically significant. Considering the inconsistent results of SNHG5 expression in patients with different tumour types, a subgroup analysis was conducted, taking into account cancer type, sample size, cutoff value, follow-up duration, hazard ratio (HR) statistics, and analysis method. The findings indicated that elevated SNHG5 expression was associated with a poorer cancer prognosis, specifically in patients with nondigestive system cancers (HR = 2.54, 95% CI 1.18–5.48); multivariate analysis was also performed (HR = 4.84, 95% CI 2.42–9.70), and the mean cutoff value was 2.82, 95% CI 1.13–7.04. Moreover, there was an insignificant relationship between SNHG5 expression and OS in the digestive system subgroup according to cancer type (HR = 1.45, 95% CI 0.42–4.96); univariate analysis was also performed (HR = 1.31, 95% CI 0.58–2.99), and the median cutoff value was 1.79, 95% CI 0.74–4.30 (Table 3). In addition, the results of one original study demonstrated that increasing SNHG5 expression might contribute to poor DFS (hazard ratio (HR): 3.690, 95% CI 1.229–11.082) (Table 3).

Fig. 2
figure 2

Forest plot of SNHG5 expression and survival outcome in cancers. A OS; B DFS

Full size image
Table 3 Pooled HRs of overall survival of patients with increased SNHG5 expression
Full size table

The correlation between SNHG5 expression and TNM stage

Eight original studies comprising 527 patients were included in this study to explore the correlation between SNHG5 expression and TNM stage. A pooled OR with 95% CI was used to determine the strong positive correlation between high SNHG5 expression and advanced TNM stage (OR = 1.988, 95% CI 1.205–3.278) (Fig. 3). Due to inconsistencies in cancer types between different primary studies, subgroup analyses were also performed. The results of subgroup analysis demonstrated that increasing SNHG5 expression predicted advanced TNM stage in the nondigestive system subgroup (OR = 2.617, 95% CI 1.686–4.061), and an insignificant correlation was observed in the digestive system subgroup (OR = 1.237, 95% CI 0.560–2.733) (Table 4).

Fig. 3
figure 3

Forest plot of SNHG5 expression and TNM stage in cancers

Full size image
Table 4 Pool effects of clinicopathologic characteristics in cancer patients with abnormal SNHG5 expression
Full size table

The correlation between SNHG5 expression and LNM

Four papers with 270 cases were obtained in this analysis to assess the correlation between SNHG5 expression and LNM. Combining the odds ratio (OR) with the 95% confidence interval (CI) indicated that SNHG5 expression was positively related to LNM (OR: 1.443, 95% CI 0.709–2.939) (Fig. 4); however, the correlation was nonsignificant. The results of the subgroup analysis revealed a noteworthy correlation between SNHG5 expression and easier-to-lymph node metastasis in the subgroup of patients with a nondigestive system invasion (OR = 1.992, 95% CI 1.204–3.295) and a nonsignificant correlation in the subgroup of patients with a digestive system invasion (OR = 0.467, 95% CI 0.175–1.245) (Table 4).

Fig. 4
figure 4

Forest plot of SNHG5 expression and LNM in cancers

Full size image

Correlations between SNHG5 expression and various other clinicopathological factors

Analysis of the pooled ORs with 95% CIs revealed that elevated SNHG5 expression was associated with increased tumour size (OR: 1.571, 95% CI 1.090–2.264) (Fig. 5); moreover, there were no significant associations between SNHG5 expression and DM (OR: 0.449, 95% CI 0.077–2.630) (Fig. 6A), histological grade (OR: 2.098, 95% CI 0.910–4.838) (Fig. 6B), depth of invasion (OR: 1.106, 95% CI 0.376–3.248) (Fig. 6C), age (OR: 0.946, 95% CI 0.718–1.247) or sex (OR: 0.762, 95% CI 0.521–1.115) (Table 4).

Fig. 5
figure 5

Forest plot of SNHG5 expression and tumor size in cancers

Full size image
Fig. 6
figure 6

Forest plot of SNHG5 expression and DM, histological grade and depth of invasion in cancers. A DM; B histological grade; C depth of invasion

Full size image

Sensitivity analysis and publication bias

The results of the sensitivity analysis of the overall survival rate showed that after removing the results of any one study, the overall survival was not affected at any time, suggesting that the overall survival rate was reliable and robust (Fig. 7). The results of Beeg's test of OS were as follows: Pr >|z|= 0.368, TNM stage = Pr >|z|= 0.108, LNM >|z|= 0.734, tumour size >|z|= 0.452, histological grade >|z|= 0.308, and DM >|z|= 1.000. These findings indicate that publication bias or other bias was not present in the original study (Fig. 8).

Fig. 7
figure 7

Sensitivity analysis for SNHG5 expression with overall survival (OS) in various cancers. HR: hazard ratio, CI confidence interval

Full size image
Fig. 8
figure 8

Beeg’s test about the relationship between SNHG5 expression and survival outcome in various cancers. A OS; B TNM stage; C LNM; D DM; E Tumor size; F Histological grade

Full size image

Discussion

Cancer has consistently posed a profound threat to the wellbeing of humanity [1, 3]. Over the years, diverse treatment modalities have been incrementally employed to address this menace, leading to important advancements [5, 40]. Nevertheless, the survival outcomes in numerous cancers have reached a plateau, rendering further progress challenging. Consequently, there is an urgent need to explore innovative therapeutic strategies. Long noncoding RNAs have been confirmed to play a role in the emergence and progression of diverse ailments, including cardiovascular disease [41], metabolic disease [42], nervous system disease [43], rheumatic immune system disease and cancer [44]. Because cancer is the number one killer in human health, an increasing number of researchers have begun to uncover the underlying pathogenic mechanisms by which long noncoding RNAs (lncRNAs) contribute to cancer progression. Mounting evidence suggests that these RNAs can modulate key processes in tumour cells, including proliferation, migration, invasion, and apoptosis, and influence the response of these cells to chemotherapy and radiotherapy [45]. Long noncoding RNAs (lncRNAs) regulate stem cell transformation and epithelial-to-mesenchymal transition (EMT), and numerous such RNAs have been identified as potential targets for cancer therapeutics; these RNAs notably affect tumour progression and markedly predict tumour prognosis. Therefore, long noncoding RNAs are promising potential tumour therapeutic targets and prognostic markers.

This analysis included 11 original studies, and the scores assigned based on the NOS indicated a high level of research quality across all 11 documents. When the hazard ratio (HR) was combined, it was revealed that elevated expression of SNHG5 could predict poor cancer prognosis, but the results were not statistically significant, and additional relevant high-quality original studies are needed to further support the results and conclusions of this investigation. Furthermore, combined OR data indicated that elevated SNHG5 expression was a predictor of advanced TNM staging, larger tumour size, easier distant metastasis, and poor histological grade. However, the correlations of SNHG5 expression with cancer LNM, invasion depth, age and sex were not statistically significant. In summary, the number of studies included in this meta-analysis was small, and the insufficient sample size may explain the reason for the effect of some prognostic indicators not reaching statistical significance. The results of the sensitivity analysis suggested that the overall survival results were reliable and robust. Furthermore, Begg's test results indicated the absence of any significant publication bias or other biases in the original studies.

SNHG5 was first revealed to be highly expressed as an oncogene in gastric cancer, and successive researchers subsequently reported that SNHG5 was differentially expressed in bladder cancer, lung cancer, liver cancer, cervical cancer, osteosarcoma, laryngeal cancer, oral cancer, and lymphoma. An increasing number of researchers have explored the oncogenic mechanism of SNHG5 (Fig. 9 and Table 5). Ma et al. reported that SNHG5 contributes to proliferation and inhibits the apoptosis of bladder cancer cells by downregulating p27 and caspase-3 and caspase-9 and upregulating CDK2 expression [32]. Wang et al. reported that SNHG5 induces gefitinib resistance by upregulating its expression via the competitive sponging of miR-377 [46]. Li et al. discovered that SNHG5 contributed to the proliferation and migration of hepatocellular carcinoma (HCC) cells through regulating GSK3β and the Wnt/β-catenin signalling pathway by competitively binding miR-26a-5p [33]. Yan et al. suggested that SNHG5 could promote the proliferation and migration of HCC cells by upregulating spermatogenesis-associated serine-rich 2 (SPATS2) expression (47). Zhang et al. indicated that SNHG5 accelerates the proliferation, migration and invasion of cervical cancer cells through upregulating sex-determining region Y-Box 4 (SOX4) expression via competitive sponging of miR-132 [34]. Wang et al. demonstrated that SNHG5 could accelerate the migration, invasion and proliferation and inhibit the apoptosis of osteosarcoma cells through targeting and positively regulating Rho-associated coiled coil-containing protein kinase 1 (ROCK1) expression by sponging and negatively regulating miR-26a [35]. Liu et al. reported that SNHG5 accelerated the proliferation, migration and invasion of nasopharyngeal carcinoma (NCC) cells by positively regulating high mobility group Box 3 (HMGB3) expression by downregulating miR-1179 expression [36]. Zhang et al. showed that SNHG5 contributed to the proliferation—and inhibited apoptosis—of AML cells through accelerating sex-determining region Y-Box 4 (SOX4) expression by competitively binding to miR-489-3p [34]. Wei et al. discovered that SNHG5 could inhibit the epithelial–mesenchymal transition (EMT) process in oesophageal cancer cells by downregulating the expression of metastasis-associated protein 2 (MTA2) [28]. Xing et al. revealed that SNHG5 could promote proliferation, invasion, migration, and inhibited apoptosis of diffuse large B-cell lymphoma (DLBC) cells through the positive regulation of X-linked inhibitor of apoptosis protein (XIAP) expression via the competitive sponging of miR-181-5p [37]. Li et al. reported that high SNHG5 expression indicated poor GC prognosis, but the detailed biological mechanism was not revealed [38]. Zhao et al. reported that SNHG5 suppressed the proliferation, migration and invasion of gastric cancer cells by upregulating MAT2 expression [39].

Fig. 9
figure 9

SNHG5 regulates some molecular biological mechanisms of tumor cells

Full size image
Table 5 Some molecular biological mechanisms of SNHG5 regulating cancer progression
Full size table

This study has several limitations. First, all patients in this meta-analysis were from China, and thus, the conclusions of this study are representative of only Asians. Second, the number of patients included in this study was insufficient, preventing some positive conclusions from reaching statistical significance. Third, some of the included studies did not provide hazard ratios (HRs) or 95% confidence intervals (95% CIs); therefore, we had to use the Engage software to analyze the survival data. This result is inconsistent with the data from the original SPSS. In addition, this study collected and explored only the relationship between SNHG5 and the prognosis of some cancers, which may bias the results. However, additional high-quality studies with larger sample sizes are needed to support the conclusions of this study. Finally, several studies revealed that SNHG5 is highly expressed in tumour cells, while others revealed that SNHG5 is expressed at low levels in tumour cells, leading to inconsistent experimental conclusions.

Conclusion

SNHG5 is abundantly expressed across numerous tumour tissues, and elevated SNHG5 levels are significantly positively associated with poorer tumour prognosis. Additionally, high SNHG5 expression predicts advanced TNM staging, increased susceptibility to distant metastasis, increased tumour diameter, and decreased histological grade. These findings suggest that SNHG5 could emerge as a potential therapeutic target and a promising prognostic marker for tumour management. However, the conclusions drawn from this study require further validation through high-quality original research.