Identification and Validation of Cuproptosis-Related LncRNA Signatures as a New Biologic Prognostic Model for Head and Neck Squamous Cell Carcinoma

doi:10.21203/rs.3.rs-1887605/v1

Download PDF

Research Article

Identification and Validation of Cuproptosis-Related LncRNA Signatures as a New Biologic Prognostic Model for Head and Neck Squamous Cell Carcinoma

https://doi.org/10.21203/rs.3.rs-1887605/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Cuproptosis is a novel, distinct form of regulated cell death that has a pivotal role in cancer metabolism. However, relatively little si known about the role of cuproptosis-related lncRNAs (CRlncRNAs) in head and neck squamous cell carcinoma (HNSCC). This study aimed to identify a CRlncRNAs signature, evalute its prognostic value and explore its molecular functions in HNSCC.

Methods

RNA-seq data and relevant clinical data were downloaded directly from The Cancer Genome Atlas (TCGA) database, and cuproptosis-related genes were identified from a search of the relevant candidate-gene literature. Analysis of differentially expressed lncRNAs (DElncRNAs) was performed using the ‘edgeR’ package in R software. The intersection of the lncRNAs between DElncRNAs and CRlncRNAs was obtained using the R package “Venn Diagram”. By implementing co-expression analysis, we identified the link betwwen the prognostic lncRNAs and their corresponding mRNAs. Univariate Cox regression was used to identify cuproptosis-related prognostic lncRNAs. The range of cuproptosis-related prognostic lncRNAs was narrowed via the LASSO-Cox method, and CRlncRNAs with the greatest significance were chosen for constructing a prognostic model to predict patient survival. Kaplan‒Meier (KM) survival analysis was carried out, and the receiver operating characteristic (ROC) curve and the area under the ROC curve (ROC-AUC) were also calculated to evaluate the predictive ability of the model. Moreover, the relationships between the prognostic model and immune cell subpopulations, related functions and pathways and drug sensitivity were explored.

Results

A CRlncRNAs signature that included three lncRNAs was constructed. According to the risk score calculated by this signature, two risk groups were established. In HNSCC patients, the risk score was a better predictor of survival than traditional clinicopathological features. In addition, significant differences in immune cells such as B cells, T cells and macrophages were observed between the two groups. Finally, the high-risk group had a lower IC50 for certain chemotherapeutic agents, such as cisplatin and cetuximab.

Conclusion

This 3 CRlncRNAs signature is a powerful prognostic biomarker for predicting clinical outcomes and therapeutic responses in HNSCC patients.

lncRNAs

cuproptosis

head and neck squamous cell carcinoma

prognostic models

immune microenvironment

Head and neck squamous cell carcinoma (HNSCC), accounting for nearly 95% of head and neck cancer cases, is the sixth most common cancer worldwide and it arises in the lip, oral cavity, pharynx, larynx, and paranasal sinuses (Pfister et al., 2020; Siegel et al., 2022; Sung et al., 2021). According to the latest reports, there were more than 0.87 million new HNSCC cases and 0.34 million HNSCC deaths globally in 2020 (Sung et al., 2021). HNSCC patients within the early stage have a better prognosis after surgery or radiotherapy (Liu et al., 2022). However, approximately 64% of patients with HNSCC present with locoregionally advanced disease at diagnosis (Gatta et al., 2015), and a high rate of recurrence and metastasis impede the effective treatment of HNSCC, so the prognosis of advanced patients is often dissatisfactory (Specenier and Vermorken, 2018). Therefore, along with improvements in the main treatment regimens for HNSCC, including surgical resection, adjuvant and neoadjuvant chemotherapy, radiation therapy, and chemoradiotherapy, it is also of great interest to find effective biomarkers for the early diagnosis of HNSCC as well as therapeutic targets for this malignancy.

The term “cuproptosis” was pioneered by Tsvetkov et al. in their seminal publication in Science earlier this year. Cuproptosis is a nonapoptotic cell death pathway that induces excess intracellular copper and mediated by lipoylated TCA cycle proteins (Tsvetkov et al., 2022). The novel form of regulated cell death is different from oxidative stress-related cell death (e.g., apoptosis, ferroptosis, and necroptosis) (Tang et al., 2022). Intracellular copper directly binds to lipoylated components of the tricarboxylic acid (TCA) cycle, and then triggers the aggregation of mitochondrial lipoylated proteins and the destabilization of Fe–S cluster proteins, subsequently leading to cuproptosis (Li et al., 2022). Cuproptosis has been associated with life-threatening disease carcinogenesis, including genetic diseases and a variety of human tumors (Tsvetkov et al., 2022). Tsvetkov et al. demonstrated that the ES–Cu complex could induce ABC1 and NCIH2030 lung cancer cell death via the cuproptosis pathway (Tsvetkov et al., 2022). Cuproptosis has opened the door to the exploration of novel aspects of mitochondrial biology in cancers and normal cellular physiology (Cobine and Brady, 2022). Recently, Bian et al. identified key cancer cuproptosis-related genes implicated in clear cell renal cell carcinoma through integrated bioinformatics analysis (Bian et al., 2022). However, at present, there is a lack of research based on cuproptosis and its related mechanisms in HNSCC, so it is very important to seek new approaches to the diagnosis and treatment of HNSCC based on cuproptosis-related genes.

Long noncoding RNAs (lncRNAs) are a large class of transcripts from nonprotein coding regions of the genome, with a length of > 200 nucleotides (Wang and Chang, 2011). Accumulating evidence strongly implicates a pivotal role of lncRNAs in the pathogenesis of tumor, such as tumorigenesis, proliferation, metastasis, drug resistance and programmed death in cancers (Fang and Fullwood, 2016; Peng et al., 2017; Schmitt and Chang, 2016; Wu et al., 2020). For example, the lncRNA MIR31HG can promote cell proliferation and cell cycle progression and inhibit cell apoptosis in HNSCC cells via HIF1A and p21 in vitro and in vivo (Wang et al., 2018). Recently, Zhang et al. found that cuproptosis-related lncRNAs (CRlncRNAs) signature could accurately predict the prognoses of hepatocellular carcinoma (Zhang et al., 2022), however, the involvement of CRlncRNAs in HNSCC patient treatment or prognosis is still ambiguous; accordingly, it is of great importance to identify prognostic CRlncRNAs in HNSCC patients.

In our study, we downloaded RNA-sequencing data for HNSCC from the TCGA data portal (http://cancergenome.nih.gov/) and ultimately screened three differentially expressed CRlncRNAs and constructed a prognostic model. Then, the functions and underlying mechanisms of CRlncRNAs action in HNSCC were further explored by gene set enrichment analysis (GSEA), immunoinfiltration analysis and chemotherapy drug sensitivity analysis.

The aim of this study was to recognize prognostic CRlncRNAs in HNSCC to offer vital signaling and mechanistic aspects of this process and provide references for clinical chemotherapy and immunotherapy.

The flow diagrams for this study are presented in Fig. 1.

Data Acquisition

The RNA-Seq data and clinical data of HNSCC were downloaded from The Cancer Genome Atlas (TCGA) website (https://portal.gdc.cancer.gov/projects/TCGA- HNSCC) in July 2020. Then, 44 normal samples, 504 HNSCC samples and 528 HNSCC patients' clinical information were obtained. To analyze the correlation of lncRNA expression signatures with the prognosis of HNSCC patients, we filtered out samples without survival information. Clinical information without complete survival information was also removed. A total of 501 HNSCC patients' complete clinical information and corresponding RNA-Seq data were selected for the subsequent clinical significance analysis. LncRNA expression profiles of TCGA HNSCC cases were obtained based on the lncRNA annotation of the GENCODE project (Frankish et al., 2019).

Differential Expression Analysis

We obtained differentially expressed lncRNAs (DElncRNAs) between normal samples and HNSCC samples, where |logFC| ≥ 1.0 and adj. P < 0.05 was used as the cutoff by using the R package “edgeR” (Robinson et al., 2010).

Identification of Cuproptosis-related LncRNAs

Cuproptosis-related genes were identified from a search of the relevant candidate-gene literature (Bian et al., 2022; Tsvetkov et al., 2022), and we obtained 19 validated human cuproptosis-related genes. Subsequently, the correlation of the expression pattern between lncRNAs and genes linked to cuproptosis was scrutinized by Pearson correlation coefficients with P < 0.001 and |R| > 0.4. Subsequently, overlapping lncRNAs between DElncRNAs and CRlncRNAs were used to identify potential prognostic lncRNAs and they are shown in a Venn diagram using the R package “VennDiagrams” (Chen and Boutros, 2011).

Construction of the LncRNA-mRNA Co-expression Network

To scrutinize the link between the potential prognostic lncRNAs and their corresponding mRNAs, a lncRNA‒mRNA co-expression network was constructed with Cytoscape software (version 3.7.2, http://www.cytoscape.org/). Further probing of this aspect of the predictive lncRNAs entailed Sankey diagram construction using the R package “ggalluvial” (Brunson, 2020).

Construction and Validation of the Cuproptosis-related Prognostic Signature

The overall dataset was randomly divided into training and validation sets at an approximate ratio of 7:3. Univariate Cox regression analysis of overall survival (OS) was performed to identify prognostic lncRNAs that were associated (p < 0.05) with OS in the training set. Subsequently, to prevent overfitting of the cuproptosis-related prognostic lncRNAs, least absolute shrinkage and selection operator (LASSO) Cox regression was employed to narrow the number of lncRNAs. Then, multivariate Cox regression analysis was performed to construct the prognostic model based on the Akaike information criteria (AIC) value. The optimal prognostic model was constructed based on 3 robust lncRNAs linked to cuproptosis. Then, the prognostic risk score was calculated by the following: risk score\(={\sum }_{i=1}^{n}\left(\text{e}\text{x}\text{p}\text{i}\text{*}{\beta }\text{i}\right)\). Here, Exp and β are representative of the expression level of the identified lncRNA and its corresponding coefficient, respectively. The patients in the training set were categorized into low-risk or high-risk groups based on their median score in the training set. The signature was corroborated by the following tests: Kaplan‒Meier (KM) survival analysis was carried out by using the R package “survminer” with the log-rank test to compare overall survival (OS) between the high and low-risk groups. The receiver operating characteristic (ROC) curve and the area under the ROC curve (ROC-AUC) were calculated to evaluate the predictive power of the prediction model using the R package “timeROC” (Kamarudin et al., 2017). Principal component analysis (PCA) was employed to observe the expression pattern of the two groups. Then, the predictive value of the risk-scoring model in the validation set and the overall set was further examined.

Gene Set Enrichment Analysis

In terms of the functional enrichment analysis, differential expression analysis was first conducted on all lncRNAs in the samples with high- and low-risk scores by using the R package “limma” (|logFC| ≥ 1.0, FDR < 0.05) (Ritchie et al., 2015). To explore the potential signaling pathways in the two groups, GSEA was conducted using the R package “clusterProfiler”. The annotated gene set c2.kegg.v7.5.1.symbols.gmt was chosen for the reference gene sets. Gene sets with p < 0.05 and FDR < 0.25 were considered significantly enriched.

Assessment of Immune Infiltration and Landscape of the Tumor Immune Microenvironment

To explore the differences in immune cell infiltration between the two groups, the CIBERSORT algorithm was used to compare the high/low risk assessment groups (Chen et al., 2018). CIBERSORT is an analytical algorithm for analyzing immune cell composition from normalized gene expression profiles in complex tissues (Newman et al., 2015). The relative proportion of twenty-two different types of infiltrated immune cells in HNSCC was estimated using the online analytical platform CIBERSORT (http://CIBERSORT.stanford.edu/). Then, a single sample gene set enrichment analysis (ssGSEA) was used to calculate 16 immune cell-associated infiltration scores and 13 immune related pathways in the R package “gsva” (Rooney et al., 2015). The R package “estimate” was used to analyze the microenvironment, immunity, stroma, and estimated score of the tumor samples (Yoshihara et al., 2013). Finally, TIDE (http://tide.dfci.harvard.edu/) algorithms were used to predict immune checkpoint response inhibitors in the low- and high-rated groups (p < 0.05) (Jiang et al., 2018).

Chemotherapeutic Response Prediction

Since chemotherapy improves survival in HNSCC patients, the chemotherapeutic response of each patient, which was determined by the half-maximal inhibitory concentration (IC50), was estimated using the R package “pRRophetic” based on the Genomics of Drug Sensitivity in Cancer (GDSC) (https://www.cancerrxgene.org/) (Geeleher et al., 2014).

Statistical Analysis

Student's t-test was used to compare the gene expression between tumor tissues and adjacent normal tissues, and differences in the proportions were compared using the χ2-test. The relative proportion of infiltrated immune cells in the tumor was compared using the Wilcoxon test. Cox univariate regression analysis and multivariate Cox regression analysis were performed to select independent prognostic factors. The area under the receiver operating characteristic (ROC) curve was used to assess the predictive ability of the prognostic model for OS. All statistical analyses were conducted using R software (version 4.13). A two-tailed P value ≤ 0.05 was considered statistically significant.

Identification of Cuproptosis-related Differentially Expressed LncRNAs in HNSCC

The complete clinical information and corresponding RNA-Seq data of 501 HNSCC patients were downloaded from the TCGA database, and 16876 lncRNAs and 19938 mRNAs were used to obtain 448 DElncRNAs with adj. P < 0.05 and |logFC| ≥ 1 (Fig. 2A). We obtained 19 validated human cuproptosis-related genes from the literature. Then, the correlation between lncRNAs in the TCGA database and the cuproptosis-related mRNAs was analyzed by the Pearson correlation coefficient method, and 783 CRlncRNAs were identified by Pearson correlation coefficients with P < 0.001 and |R| > 0.4. Finally, we identified 39 cuproptosis-related DElncRNAs as potential prognostic lncRNAs (Fig. 2B).

The LncRNA–mRNA Co-Expression Network

To evaluate the relationship between these 39 cuproptosis-related DElncRNAs and the representative cuproptosis-related genes, we established a co-expression network using Cytoscape. There were 10 mRNAs associated with 39 lncRNAs (P < 0.001, |R| > 0.4; Fig. 2C). A Sankey diagram was established to explore the relationship between cuproptosis-related DElncRNAs and cuproptosis-related genes (Fig. 2D).

Identification of Prognostic Cuproptosis-related Differentially Expressed lncRNAs

In this study, we randomly segregated these HNSCC cases into a training set of 351 cases and a validation set of 150 cases at an approximate ratio of 7:3. The epidemiological and clinical characteristics of each HNSCC patient in the two sets are shown in Table 1. These cuproptosis-related DElncRNAs with prognostic capacity were evaluated using Cox univariate regression analysis of the OS data of HNSCC patients in the training set, and, 8 prognostic cuproptosis-related DElncRNAs in HNSCC were identified (Fig. 3A). Six cuproptosis-related DElncRNAs were “protective” genes, while AC092115.3 and LINC01063 could be treated as “risk” genes.

Table 1

The clinical characteristics of HNSCC patients in the training, validation and overall sets.
Characteristics	Overall	Validation	Training	P
Age
≤ 65	326(65.07%)	93(62.00%)	233(66.38%)	0.401
> 65	175(34.93%)	57(38.00%)	118(33.62%)	0.401
Gender
female	133(26.55%)	35(23.33%)	98(27.92%)	0.3399
male	368(73.45%)	115(76.67%)	253(72.08%)	0.3399
Grade
G1	61(12.18%)	18(12.00%)	43(12.25%)	0.9277
G2	299(59.68%)	87(58.00%)	212(60.40%)
G3	119(23.75%)	36(24.00%)	83(23.65%)
G4	2(0.40%)	1(0.67%)	1(0.28%)
unknow	20(3.99%)	8(5.33%)	12(3.42%)
T classification
T1	47(9.38%)	15(10.00%)	32(9.12%)	0.6357
T2	151(30.14%)	49(32.67%)	102(29.06%)
T3	118(23.55%)	37(24.67%)	81(23.08%)
T4	185(36.93%)	49(32.66%)	136(38.74%)
T classification
N0	214(42.71%)	65(43.34%)	149(42.45%)	0.9677
N1	76(15.17%)	21(14.00%)	55(15.67%)
N2	198(39.52%)	59(39.33%)	139(39.60%)
N3	9(1.80%)	3(2.00%)	6(1.71%)
unknow	4(0.80%)	2(1.33%)	2(0.57%)
M classification
M0	486(97.00%)	145(96.67%)	341(97.15%)	0.4521
M1	4(0.80%)	0(0.00%)	4(1.14%)
unknow	11(2.20%)	5(3.33%)	6(1.71%)

Construction and Validation of a Prognostic Model

To reduce multicollinearity and further identify the 8 prognostic cuproptosis-related DElncRNAs that were significantly correlated with the prognosis of HNSCC patients, through LASSO Cox regression analysis, three cuproptosis-related DElncRNAs (AL109936.2, MIR9-3HG and LINC01063) were used to establish a risk score to predict the OS of HNSCC patients in the TCGA training set. The cvfit and lambda curves are shown in Figs. 3B, C. In this model, the risk score of each patient was determined based on the following formula: Risk score = AL109936.2 * (-0.30836) + MIR9-3HG * (-0.20797) + LINC01063 * 0.29029. Then, the correlation heatmap further summarized the correlations between these 3 CRlncRNAs and 20 representative cuproptosis-related genes (Fig. 3D).

Next, we continued to estimate the prognostic values of the model. The risk score for all HNSCC patients in each set was calculated according to the formula mentioned above and ranked in ascending order. Then, HNSCC patients in each set were divided into high-risk and low-risk groups based on their high and low risk scores in the training set, respectively. Each patient’s survival outcome, risk status and lncRNA expression levels in the training set were shown in Fig. 4A, D and G, and we found that the mortality rate of patients gradually increased with increasing risk scores. Kaplan–Meier survival analysis demonstrated that the OS for the high-risk group was evidently worse than that for the low-risk group in the training set (Fig. 4J). ROC curve analysis was used to measure the sensitivity and specificity of this model and it demonstrated an AUC of 0.646 (training set), which was more than that of the other factors scrutinized demonstrating the robustness of our model in predicting HNSCC survival (Fig. 4M). Moreover, we found that the area under the ROC curves (AUC) of the prognostic model for OS was 0.646 at 1 year, 0.696 at 3 years, and 0.633 at 5 years (Fig. 4P). To better assess its predictive accuracy, the above results were replicated in both the validation set (Fig. 4B, E, H, K, N, Q) and the overall set (Fig. 4C, F, I, K, N, Q). In addition, the results of PCA (Fig. 4S, T, U) also suggested heterogeneity between the high- and low-risk subgroups. These results guarantee the credibility and practicality of our data.

To assess whether the risk score could be used as an independent risk factor for HNSCC, we performed Cox univariate and multivariate regression analyses in combination with clinical factors. Univariate and multivariate analyses of risk scores based on the training set (Fig. 5A, B), the validation set (Fig. 5C, D) and the overall set (Fig. 5E, F) all showed that the risk score of this model could be independent of other clinical factors and effectively predict the survival of patients (P < 0.05). In addition, we constructed a nomogram based on the risk score and clinical factors, which is more practical for predicting the survival of patients in clinical practice (Fig. 6A). We built calibration curves to evaluate the performance of the nomogram, and the results showed good calibration (Fig. 6B).

Gene Set Enrichment Analysis

According to the median risk score, GSEA was conducted to sort out significantly enriched pathways between the high-risk and low-risk groups. The GSEA results showed that there were 43 enriched pathways that were significantly enriched in the low-risk group, while there was no considerable enrichment of pathways in the high-risk group. Among these pathways were immune-associated and cancer-associated pathways, including T-cell receptor signaling pathway, B-cell receptor signaling pathway, Fc epsilon RI signaling pathway, leukocyte transendothelial migration, natural killer cell-mediated cytotoxicity, chemokine signaling pathway, cell adhesion molecules (CAMs), tight junction and apoptosis (Fig. 7).

Analysis of the Tumor Environment and Immune Infiltration

While studying the immune cell infiltration of all HNSCC patients in the TCGA database, our results showed that the main components of the immune environment were T lymphocytes and macrophages as obtained by the CIBERSORT algorithm (Fig. 8A). The stromal score (substrate cells in tumor tissue) and immune score (immune cell infiltration in tumor tissue) were determined with the ESTIMATE algorithm to explore differences in immune cell infiltration between the low- and high-risk groups. Surprisingly, we found differences in the immune score, but not the stromal score or ESTIMATE score (Fig. 8B). Comparing the proportion of immune cells between the high-risk and low-risk groups, we found significant differences in B cells, CD8 + T cells, immature dendritic cells, macrophages, mast cells, follicular helper T cells, T helper cell type 2 (Th2) cells, and tumor infiltrating lymphocytes between the two groups (Fig. 8C). Moreover, the association between immune cells in HNSCC indicated interactions within immune microenvironments (Fig. 8D).

ssGSEA results revealed a conspicuous alteration between the low- and high-risk groups in terms of T-cell functions. We found that there were significant differences in immune function scores including checkpoint, HLA, inflammation promoting, T-cell co-stimulation and Type II IFN response scores between the two groups (Fig. 9A). Subsequently, we checked the expression changes of the immune checkpoint genes in the high-risk and low-risk groups. As shown in Fig. 9B, a total of 31 tested common checkpoint genes displayed significant differences between the two groups. Many tested immunotherapy targets, such as PDCD1 (PD-1), IDO1, LAG3 and CTLA4 were highly expressed in the low-risk group.

Analysis of the Correlation Between the Risk Score and Chemotherapeutics

Cisplatin, cetuximab, docetaxel and paclitaxel are chemotherapy drugs commonly used for the treatment of HNSCC(Pfister et al., 2020), so we attempted to identify correlations between the risk score and the effectiveness of four common anticancer drugs. We found that high-risk patients were more sensitive to cisplatin and cetuximab than low-risk patients, while they were more resistant to docetaxel and paclitaxel than low-risk patients (Fig. 10A, B, C, D). These results can be helpful for the precise treatment of HNSCC.

HNSCC is a devastating disease with a high mortality rate. With the emergence of individualized precision therapy, including radical resection, radiotherapy and chemotherapy, alone or in combination, the prognosis of HNSCC patients has improved; however, HNSCC has not shown an increase in the overall or long-term survival rates worldwide, and advanced HNSCC patients have a poor prognosis and high recurrence rates (Kampel et al., 2021; Lacas et al., 2021). Therefore, identifying new molecular targets is a key step in improving the diagnosis, prognosis, and treatment of HNSCC. Cuproptosis is a form of cell death only discovered in 2022, characterized by the copper-dependent accumulation of lipoylated mitochondrial enzymes and the loss of Fe–S cluster proteins, which might indicate novel approaches to exploiting copper toxicity for treating tumors (Tsvetkov et al., 2022). Many studies have shown that lncRNAs play vital regulatory roles in tumor occurrence and development (Chen et al., 2017). However, there are still no bioinformatics studies of HNSCC based on cuproptosis-related lncRNAs.

In this study, based on the analysis of 501 HNSCC patient samples in the TCGA dataset, we constructed a 3-lncRNAs prognostic model to reliably predict the survival of patients with HNSCC utilizing LASSO Cox regression analysis in the training set. HNSCC patients were divided into high- and low-risk groups according to their risk scores derive from this prognostic model. KM survival analysis indicated significantly different survival between patients with higher risk scores and lower scores. The AUC suggests the good predictive ability of the lncRNA signature, and the PCA also suggests that our prognostic model is effective in distinguishing the high-risk group from the low-risk group. Univariate analysis and multivariate analysis indicated that the risk score was an independent factor for overall survival. The forest plot of HR value (HR > 1) also indicated that the risk score is representative of a harmful factor. In addition, this model was validated in the validation set and the overall set. Altogether, the above analyses, compared with conventional clinical factors, confirmed that the lncRNA signature, including AL109936.2, MIR9-3HG and LINC01063, was highly reliable and effective for predicting the survival of HNSCC patients.

Associations between a cuproptosis-related model and tumor immunity have not been reported in previous studies, so GSEA was then conducted. The GSEA results demonstrated the enrichment of immunoregulatory pathways against cancer in the low-risk group, which may indicate that cuproptosis is closely associated with tumor immunity. Furthermore, GSEA revealed several significantly enriched immunoregulatory pathways, and the proportion of different types of tumor-infiltrating immune cells in HNSCC from the TCGA database was calculated using CIBERSORT. We found that the immune score of the low-risk group was significantly higher than that of the high-risk group, which indicated greater immune cell infiltration within the low-risk group. Furthermore, patients in the high-risk group also presented relatively higher expression levels of macrophages and antitumor infiltrating immune cells (NK cells, B cells, T-helper cells, CD4 + T cells, and CD8 + T cells) and five immune-related pathways were highly expressed in the low-risk group. Macrophages generally play a role of protumor, and high infiltration of macrophages is associated with a poor prognosis (Noy and Pollard, 2014). CD4 + T cells and CD8 + T cells are the main cells that involved in immune responses and are active in the anti-tumor response, and their high infiltration improves the prognosis of cancer patients (Cho et al., 2003; Schreibelt et al., 2013). These reasons may explain the difference in prognosis between the two groups.

Immunotherapy is currently viewed as a breakthrough in cancer treatment that has demonstrated promising results in improving the prognosis of cancer patients across the globe (Kruger et al., 2019; Ribas and Wolchok, 2018). Immunotherapy targeting immune checkpoints is now the most attractive therapy in the cancer field (Postow et al., 2015). Targeted therapy for the PD-1/PD-L1 axis (pembrolizumab and nivolumab) has already been approved for the first-line treatment setting of patients with recurrent or metastatic HNSCC (Ferris et al., 2018; Shen et al., 2020). Compared to the high-risk group, the expression of many immune checkpoints, such as PDCD1 (PD-1), IDO1, LAG3 and CTLA4, was upregulated in the low-risk group, which means that low-risk patients might achieve improved responses to treatment with anti- PD-L1/PD-1 therapy. Therefore, HNSCC patients in the low-risk group would have better survival after receiving immunotherapy based on our cuproptosis-related lncRNA signature. These results further confirmed the potential of the CRlncRNA signature in predicting the suitability of immune checkpoint blockade in patients.

Finally, we endeavored to estimate an association between the CRlncRNA signature and chemosensitivity in HNSCC patients based on IC50. We found that HNSCC patients in the high-risk group showed greater sensitivity to cisplatin and cetuximab treatment than those in the low-risk group, indicating that cuproptosis induced by chemotherapeutic drugs was more frequently observed in high-risk patients and that the CRlncRNA signature might serve as a potential prognosticator for chemosensitivity in HNSCC.

However, our study has limitations. First, HNSCC encompasses several types of cancers, each of which requires separate, detailed analyses. Moreover, both our training and validation sets were obtained from the TCGA database. Despite the large sample size of HNSCC patients, our findings would be strengthened if there were external validation cohorts for analysis available. Finally, in vitro and in vivo experiments to clarify the biological mechanisms of CRlncRNAs have not been conducted; therefore, further experimental validation is needed in the future.

A robust predictive model was constructed based on three cuproptosis-related lncRNAs. This novel model may provide new research strategies for exploring the mechanisms of cuproptosis, and may provide a new treatment strategy for HNSCC patients.

Data Availability Statement

The dataset analyzed in this study is available from the TCGA database. These data can be found here: https://portal.gdc.cancer.gov/.

Acknowledgments

We acknowledge the TCGA database for providing their platforms and the contributors for uploading their meaningful datasets.

Authors’ Contributions

JL and QS designed the study and drafted the manuscript. XQ, JZ, and TG provided the statistical software and performed the data analysis. YX and GC arranged the figures and tables. GB and ZG reviewed and revised the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by the Hospital-level Project of Qingpu Branch of Zhongshan Hospital Affiliated to Fudan University (QYT2021-02) and the Project of Health Commission (W2021-09, Qingpu District, Shanghai, P.R. China).

Conflicts of interest

No potential conflicts of interest were reported by the authors.

Bian Z, Fan R, Xie L (2022) A Novel Cuproptosis-Related Prognostic Gene Signature and Validation of Differential Expression in Clear Cell Renal Cell Carcinoma. Genes (Basel) 13(5). https://doi.org/10.3390/genes13050851
Brunson JC (2020) Ggalluvial: layered grammar for alluvial plots. Journal of Open Source Software 5(49): 2017
Chen B, Khodadoust MS, Liu CL, Newman AM, Alizadeh AA (2018) Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods Mol Biol 1711243–1711259. https://doi.org/10.1007/978-1-4939-7493-1_12
Chen H, Boutros PC (2011) VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics 1235. https://doi.org/10.1186/1471-2105-12-35
Chen QN, Wei CC, Wang ZX, Sun M (2017) Long non-coding RNAs in anti-cancer drug resistance. Oncotarget 8(1):1925–1936. https://doi.org/10.18632/oncotarget.12461
Cho Y, Miyamoto M, Kato K et al (2003) CD4 + and CD8 + T cells cooperate to improve prognosis of patients with esophageal squamous cell carcinoma. Cancer Res 63(7):1555–1559
Cobine PA, Brady DC (2022) Cuproptosis: Cellular and molecular mechanisms underlying copper-induced cell death. Mol Cell 82(10):1786–1787. https://doi.org/10.1016/j.molcel.2022.05.001
Fang Y, Fullwood MJ (2016) Roles, Functions, and Mechanisms of Long Non-coding RNAs in Cancer. Genomics Proteom Bioinf 14(1):42–54. https://doi.org/10.1016/j.gpb.2015.09.006
Ferris RL, Blumenschein G Jr, Fayette J et al (2018) Nivolumab vs investigator's choice in recurrent or metastatic squamous cell carcinoma of the head and neck: 2-year long-term survival update of CheckMate 141 with analyses by tumor PD-L1 expression. https://doi.org/10.1016/j.oraloncology.2018.04.008. Oral Oncol 8145-51
Frankish A, Diekhans M, Ferreira AM et al (2019) GENCODE reference annotation for the human and mouse genomes. Nucleic Acids Res 47(D1):D766–d773. https://doi.org/10.1093/nar/gky955
Gatta G, Botta L, Sanchez MJ et al (2015) Prognoses and improvement for head and neck cancers diagnosed in Europe in early 2000s: The EUROCARE-5 population-based study. Eur J Cancer 51(15):2130–2143. https://doi.org/10.1016/j.ejca.2015.07.043
Geeleher P, Cox N, Huang RS (2014) pRRophetic: an R package for prediction of clinical chemotherapeutic response from tumor gene expression levels. PLoS ONE 9(9):e107468. https://doi.org/10.1371/journal.pone.0107468
Jiang P, Gu S, Pan D et al (2018) Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat Med 24(10):1550–1558. https://doi.org/10.1038/s41591-018-0136-1
Kamarudin AN, Cox T, Kolamunnage-Dona R (2017) Time-dependent ROC curve analysis in medical research: current methods and applications. BMC Med Res Methodol 17(1):53. https://doi.org/10.1186/s12874-017-0332-6
Kampel L, Dorman A, Horowitz G et al (2021) Surgically Treated Advanced Cutaneous Squamous Cell Carcinoma of the Head and Neck: Outcome Predictors and the Role of Adjuvant Radiation Therapy. Ann Otol Rhinol Laryngol 130(9):1016–1023. https://doi.org/10.1177/0003489421990182
Kruger S, Ilmer M, Kobold S et al (2019) Advances in cancer immunotherapy 2019 - latest trends. J Exp Clin Cancer Res 38(1):268. https://doi.org/10.1186/s13046-019-1266-0
Lacas B, Carmel A, Landais C et al (2021) Meta-analysis of chemotherapy in head and neck cancer (MACH-NC): An update on 107 randomized trials and 19,805 patients, on behalf of MACH-NC Group. https://doi.org/10.1016/j.radonc.2021.01.013. Radiother Oncol 156281-293
Li SR, Bu LL, Cai L (2022) Cuproptosis: lipoylated TCA cycle proteins-mediated novel cell death pathway. Signal Transduct Target Ther 7(1):158. https://doi.org/10.1038/s41392-022-01014-x
Liu J, Zhao J, Xu J et al (2022) SPINK5 is a Prognostic Biomarker Associated With the Progression and Prognosis of Laryngeal Squamous Cell Carcinoma Identified by Weighted Gene Co-Expression Network Analysis. https://doi.org/10.1177/11769343221077118. Evolutionary Bioinformatics 181 – 13
Newman AM, Liu CL, Green MR et al (2015) Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 12(5):453–457. https://doi.org/10.1038/nmeth.3337
Noy R, Pollard JW (2014) Tumor-associated macrophages: from mechanisms to therapy. Immunity 41(1):49–61. https://doi.org/10.1016/j.immuni.2014.06.010
Peng WX, Koirala P, Mo YY (2017) LncRNA-mediated regulation of cell signaling in cancer. Oncogene 36(41):5661–5667. https://doi.org/10.1038/onc.2017.184
Pfister DG, Spencer S, Adelstein D et al (2020) Head and Neck Cancers, Version 2.2020, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 18(7):873–898. https://doi.org/10.6004/jnccn.2020.0031
Postow MA, Callahan MK, Wolchok JD (2015) Immune Checkpoint Blockade in Cancer Therapy. J Clin Oncol 33(17):1974–1982. https://doi.org/10.1200/jco.2014.59.4358
Ribas A, Wolchok JD (2018) Cancer immunotherapy using checkpoint blockade. Science 359(6382):1350–1355. https://doi.org/10.1126/science.aar4060
Ritchie ME, Phipson B, Wu D et al (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43(7):e47. https://doi.org/10.1093/nar/gkv007
Robinson MD, McCarthy DJ, Smyth GK (2010) edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26(1):139–140. https://doi.org/10.1093/bioinformatics/btp616
Rooney MS, Shukla SA, Wu CJ, Getz G, Hacohen N (2015) Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell 160(1–2):48–61. https://doi.org/10.1016/j.cell.2014.12.033
Schmitt AM, Chang HY (2016) Long Noncoding RNAs in Cancer Pathways. Cancer Cell 29(4):452–463. https://doi.org/10.1016/j.ccell.2016.03.010
Schreibelt G, Bol KF, Aarntzen EH et al (2013) Importance of helper T-cell activation in dendritic cell-based anticancer immunotherapy. Oncoimmunology 2(6):e24440. https://doi.org/10.4161/onci.24440
Shen LF, Zhou SH, Guo Y (2020) Role of GLUT-1 in the Upregulation of PD-L1 Expression After Radiotherapy and Association of PD-L1 with Favourable Overall Survival in Hypopharyngeal Cancer. Onco Targets Ther 1311221–1311235. https://doi.org/10.2147/ott.S269767
Siegel RL, Miller KD, Fuchs HE, Jemal A (2022) Cancer statistics, 2022. CA Cancer J Clin 72(1):7–33. https://doi.org/10.3322/caac.21708
Specenier P, Vermorken JB (2018) Optimizing treatments for recurrent or metastatic head and neck squamous cell carcinoma. Expert Rev Anticancer Ther 18(9):901–915. https://doi.org/10.1080/14737140.2018.1493925
Sung H, Ferlay J, Siegel RL et al (2021) Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71(3):209–249. https://doi.org/10.3322/caac.21660
Tang D, Chen X, Kroemer G (2022) Cuproptosis: a copper-triggered modality of mitochondrial cell death. Cell Res 32(5):417–418. https://doi.org/10.1038/s41422-022-00653-7
Tsvetkov P, Coy S, Petrova B et al (2022) Copper induces cell death by targeting lipoylated TCA cycle proteins. Science 375(6586):1254–1261. https://doi.org/10.1126/science.abf0529
Wang KC, Chang HY (2011) Molecular mechanisms of long noncoding RNAs. Mol Cell 43(6):904–914. https://doi.org/10.1016/j.molcel.2011.08.018
Wang R, Ma Z, Feng L et al (2018) LncRNA MIR31HG targets HIF1A and P21 to facilitate head and neck cancer cell proliferation and tumorigenesis by promoting cell-cycle progression. Mol Cancer 17(1):162. https://doi.org/10.1186/s12943-018-0916-8
Wu P, Mo Y, Peng M et al (2020) Emerging role of tumor-related functional peptides encoded by lncRNA and circRNA. Mol Cancer 19(1):22. https://doi.org/10.1186/s12943-020-1147-3
Yoshihara K, Shahmoradgoli M, Martínez E et al (2013) Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 42612. https://doi.org/10.1038/ncomms3612
Zhang G, Sun J, Zhang X (2022) A novel Cuproptosis-related LncRNA signature to predict prognosis in hepatocellular carcinoma. Sci Rep 12(1):11325. https://doi.org/10.1038/s41598-022-15251-1

No competing interests reported.

Download PDF

Editorial decision: Major revision
15 Oct, 2022
Reviews received at journal
15 Oct, 2022
Reviewers agreed at journal
15 Oct, 2022
Reviewers invited by journal
04 Aug, 2022
Editor assigned by journal
01 Aug, 2022
Submission checks completed at journal
01 Aug, 2022
First submitted to journal
23 Jul, 2022

You are reading this latest preprint version

Identification and Validation of Cuproptosis-Related LncRNA Signatures as a New Biologic Prognostic Model for Head and Neck Squamous Cell Carcinoma

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Materials And Methods

Statistical Analysis

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1