Cofilin1 Is a Poorer Prognostic Biomarker and potential Immunotherapy in Systematic Pan-Cancer

doi:10.21203/rs.3.rs-1401384/v1

Background: Cofilin1(CFL1) is an actin-binding protein that can impact the prognosis of patients with malignant cancer through different mechanisms. However, the CFL1 and pan-cancer correlation are unclear. The current research aims studying the link between CFL1 expression and tumor microenvironment and exploring this mechanism in pan-cancer.

Methods: To assess the CFL1 expression in diverse cancer types when compared to normal tissues, the Oncomine database was utilized. The link between CFL1 expression and clinical prognosis was estimated by Kaplan-Meier Plotter. To evaluate the connection between CFL1, immune infiltration and related gene maker sets, the TIMER database was utilized. CFL1 was further committed to enrichment analysis for the KEGG pathway.

Results: The expression level of CFL1 gene is heterogeneous in different cancer cell lines. The high expression of CFL1 makes pan-cancer patients have poor survival benefits, excluding THYM(Thymoma) patients. The correlation between CFL1 and immune cells was different among cancer types, as proved. The CFL1 expression was linked to TMB and MSI within 14 and six cancer types, respectively. The majority of genes related to immunity and co-expressed with CFL1 has a positive correlation in most tumors. Finally, GSEA analysis shows that CFL1 impact the pathway of metabolism.

Conclusion: The study results showed that CFL1 could be a latent prognostic biomarker as well as possible immunotherapy in pan-cancers.

CFL1

Pan-cancer

Biomarker

Prognosis

Immunity

The main mortality cause among patients is cancer. It is also a factor that affects the patients’ quality of life (1). Immunotherapy has now become another effective therapy following surgery, radiotherapy and chemotherapy, and targeted therapy and has been widely used in clinics (3, 4). Tumor immunotherapy is based on a full understanding of the body’s immune principles, applying immunological methods that improve the tumor cells’ immunogenicity and the sensitivity to killer cells while enhancing the body’s anti-tumor immune response (2). In recent years, immune checkpoint inhibitors have received widespread attention due to their excellent efficacy and safety in clinical trials. Currently, clinical trials have shown that the programmed death-ligand 1 (PD-L1) expression (5) and tumor mutation burden (TMB) (6) are both having certain defects. They cannot accurately identify the real benefit population of immunotherapy. Due to the spatial and temporal heterogeneity of tumors, it is imminent to find potential biomarkers.

Cofilin1 (CFL1), an actin-binding protein, is widely expressed in eukaryotes. CFL1 have a vital role in cytoskeletal remodeling and actin function, including depolymerization and polymerization (7). According to many previous studies, CFL1 has promiscuous physiological effects, including cell motility, cell apoptosis and lipid metabolism (8). Currently, many researches showed the pivotal role of CFL1 in the development of tumor (9, 10). CFL1 was higher in tumor than paired-normal tissue, including pancreatic cancer, vulvar squamous cell carcinoma, hepatocellular carcinoma, esophagus cancer, and prostate cancer (7, 11–14). CFL1, as a promising novel biomarker of cancer, has received wide attention. However, the link between tumor immunity, CFL1, and prognosis in pan-cancer remains unclear.

The link between CFL1 expression, tumor prognosis, tumor infiltrating immune cells (TIICs), and associated immunological markers were thoroughly analyzed based on a database of multiple data mining analysis methods and further visualized the prognostic prospect of generalized cancer.

2.1 ONCOMINE database analysis

To analyze the gene expression differences, the Oncomine database (15) was used. It is a large database of cancer genes. It includes 65 datasets from gene chips, 4700 gene chips, and 480 million gene expression data. Oncomine can perform differential expression analysis of genes by integrating the RNA and DNA data obtained from GEO, TCGA, and published sources. The Oncomine was utilized for the expression analysis, such as CFL1 expression in 33 tumors. The threshold is set to (P value = 0.001) and fold change (1.5).

2.2 Analysis of the link between CFL1 and clinical prognosis

To study the link between CFL1 gene expression and 33 cancer types’ prognosis, we utilized the TCGA database. In addition, survival analysis was conducted by Cox regression and Kaplan-Meier methods. The forest maps and survival analysis curves were used, respectively, to visualize the results. The overall survival (OS), disease-free interval (DFI), progression-free interval (PFI), and disease-specific survival (DSS) were analyzed. P < 0.05 was statistically significant.

2.3 Association between the CFL1 expression and immune

To predict the tumor tissues’ immune cell infiltration, The Tumor Immune Estimate Resource (TIMER) was utilized. The TIMER database estimated the immune infiltrates’ abundance in 10,897 samples from 32 tumor samples using The Cancer Genome Atlas (TCGA). We estimated the CFL1 expression in different cancers and the immune correlation.

The ESTIMATE database focuses on the tumor tissue’s non-tumor components, primarily stromal cells and immune cells.

To assess the non-tumor components’ ratio in tumor tissue, we applied the ESTIMATE algorithm. Stromal and immune cells and the matrix and immune scores were utilized to predict the matrix and immune cells’ infiltrating levels. Immune, Stromal and ESTIMATE Scores are three immune matrix components forms in the tumor microenvironment, positively correlated with Immunity, Stromal Scores, and their sum, respectively.

2.4 Relationship between CFL1, Tumor Mutation Burden and Microsatellite instability

To predict the immune checkpoint inhibitor therapy’s response, TMB is utilized as a marker. MSI occurs due to DNA mismatch repair in tumor tissue. By the Pearman rank correlation coefficient usage, we counted each tumor sample’s TMB and analyzed the link between the CFL1, TMB and MSI expression. We utilized the R-packages, “RColorBrewer and “reshape2” to make heat maps to present the data.

2.5 Gene set enrichment analysis

GSEA (16) is a database containing 1325 gene sets, including multiple functional gene sets based on the location, function, and biological significance of existing genes. We downloaded and analyzed the Kyoto Encyclopedia of Genes and Genomesn KEGG) gene sets and HALLMARK pathway using GSEA. According to the CFL1 expression, the gene sets were classified with high and low expression levels. All significant KEGG and HALLMARK pathways were visualized with a ggplot package.

2.6 Statistical analysis

Using the PrognoScan and Kaplan-Meier plots, the survival curves were analyzed. The Oncomine output data are in the form of P-values, fold changes and ranks. To assess the gene expression correlation, Spearman's correlation was utilized. P = 0.05 indicated a significant difference.

1. The CFL1 expression level in different human cancers

By the Oncomine database usage, we speculated the CFL1 expression in human cancers and normal tissues. Figure 1A showed, CFL1 is higher in bladder, head and neck, lung, pancreatic, breast, ovarian, kidney, liver cancers, and leukemia, myeloma, lymphoma. The detailed p-value, fold change, and reference information of CFL1 expression in different cancers was listed in Supplemental Table 1.

We analyzed the CFL1 expression in tumor and normal tissues combining two databases, TCGA and GTEx. Figure 1B showed, the CFL1 expression levels in BLCA (bladder urothelial carcinoma), CHOL (cholangiocarcinoma), KIRP (kidney renal papillary cell carcinoma), KIRC (kidney renal clear cell carcinoma), HNSC (head and neck cancer), BRCA (breast invasive carcinoma), STAD (stomach adenocarcinoma), GBM (glioblastoma), LIHC (liver hepatocellular carcinoma), OV (ovarian serous cystadenocarcinoma), LUSC (lung squamous cell carcinoma), PAAD (pancreatic adenocarcinoma), and UCEC (uterine Corpus Endometrial Carcinoma) are higher than their peritumoral normal tissues (P < 0.001). On the contrary, CFL1 was significantly lower expressed in ACC (adrenocortical carcinoma), ESCA (esophageal carcinoma), LUAD (lung adenocarcinoma), LAML (acute myeloid leukemia), SKCM (skin cutaneous melanoma), and THCA (thyroid carcinoma) (P < 0.001). In addition, the CFL1 expression in CESC (cervical squamous cell carcinoma and endocervical adenocarcinoma), COAD (colon adenocarcinoma), LGG (low-grade glioma), READ (rectum adenocarcinoma), and UCS (uterine carcinosarcoma) were equal with adjusted normal tissues.

2. CFL1 prognostic potential in different cancer types

A relevant survival analysis on pan-cancer patients was done for assessing the CFL1 expression and prognosis correlation. It included overall survival (OS), progression-free interval (PFI), progression-free survival (PFS), and disease-specific survival (DSS). To evaluate the CFL1 and OS link in 33 tumor types, univariate cox regression had been utilized. As shown in Fig. 2A, the CFL1 high expression affects the survival of ACC (p = 0.000006), BLCA (p = 0.017), CESC (p = 0.014), HNSC (p = 0.036), KICH (p = 0.0016), LAML (p = 0.00005), LGG (p = 0.033), LIHC (p = 5.9*10^− 8), LUAD (p = 1.8*10^− 6), MESO (p = 0.000002), PAAD (p = 0.005), THYM (p = 0.037) and UVM (p = 0.013). The CFL1 high levels were related to a poor OS.

Between patients with ACC (F. 2B;p<0.0001), BLCA (F. 2C;p = 0.004), CESE (F. 2D;p<0.0001), HNSC (F. 2E;p = 0.00096), KICH (F. 2F;p = 0.0013), LAML (F. 2G;p<0.0001), LGG (F. 2H;p = 0.0051), LIHC (F. 2I;p<0.0001), LUAD (F. 2J;p<0.0001), MESO (F. 2K;p<0.0001), PAAD (F. 2L;p = 0.0025), UVW (F. 2N;p = 0.002), the Kaplan-Meier survival analysis showed, the patients who had high CFL1 levels, had poorer OS, yet patients with THYM (F. 2M;p = 0.018) and high CFL1 levels had longer OS. Considering the influence of non-tumor factors during follow-up, we analyzed in 33 tumors the connection between CFL1 expression and disease-specific survival (DS-S). The CFL1 high expression in tumor samples is correlated to prognosis, in ACC (p = 0.000022), BLCA(p = 0.012), BRCA(p = 0.044), CESC(p = 0.049), HNSC(p = 0.017), KICH(p = 0.0028), KIRC(p = 0.0016), KIRP(p = 0.00044), LIHC(p = 0.00051), LUAD(p = 0.00009), MESO(p = 0.00011), PAAD(p = 0.0081) and UVM(p = 0.01), as the univariate analysis showed. (F. 3A). The Kaplan-Meier analysis shows that patients with ACC(p<0.0001),BLCA(p<0.0001), BRCA(p = 0.025), CESE(p = 0.0021),HNSC(p = 0.00011), KICH(p = 0.00081),KIRP(p = 0.00051), LIHC(p = 0.018), LUAD(p = 0.00019), MESO(p = 0.0018), PAAD(p = 0.00071), UVM(p = 0.002), as shown from Fig. 3B-Figure3N.

Notably, Univariate Cox analysis revealed the prognostic impact of CFL1 in ACC(p = 6.4*10^− 5), HNSC(p = 0.021), KICH(p = 0.01), KIRC(p = 0.0041), KIRP(p = 0.0071), LUAD(p = 0.026), LUSC(p = 0.021), MESO(p = 0.0063), PAAD(p = 0.0078), PCPG(p = 0.046), PRAD(p = 0.0038), TGTC(p = 0.008), UVM(p = 0.026) (Fig. 4A). The Kaplan-Meier analysis shows the same trend, CFL1 expression significantly impacts 13 type cancers in the analysis of PFI, including ACC (F. 4B; p<0.0001), HNSC (F. 4C; p<0.0001), KICH (F. 4D; p = 0.006), KIRC (F. 4E; p = 0.0028), KIRP (F. 4F; p = 0.0077), LUAD (F. 4G; p = 0.0037), LUSC (F. 4H; p = 0.0031), MESO (F. 4I; p = 0.0014), PAAD (F. 4J; p = 0.0034), PCPG (F. 4K; p = 0.035), PRAD (F. 4L; p<0.0001), TGCT (F. 4M; p = 0.063), UVM (F. 4N; p = 0.024).

The results gained using Cox regression analyses are presented as forest maps (F. 5A), which show a link between CFL1 expression, prognosis and patients’ survival with ACC (p = 0.0076), CESE (p = 0.042), CHOL (p = 0.023), KIRP (p = 0.038), MESO (p = 0.039), PRAD (p = 0.034). Higher CFL1 expression is significantly linked with increased DFI in ACC (F. 5B; p = 0.00051), CESC (F. 5C; p = 0.0049), CHOL (F. 5D; p = 0.011), ESCA (F. 5E; p = 0.019), KIRP (F. 5F; p = 0.006), MESO (F. 5G; p = 0.01) and PRAD (F. 5H; p = 0.00026).

3. CFL1 Expression and Immune Infiltration Level correlation in pan-cancers

In the cancer microenvironment, two main non-tumor components should be considered in tumor diagnosis and prognosis, the immune and stromal cells. Tumor infiltrating lymphocytes are white blood cells that enter the tumor in the bloodstream. The body initiates an immune response against tumors when there are a large number of tumors infiltrating lymphocytes. We assessed the CFL1 expression and the immune infiltration level within 32 cancers detected using TIMER (Table 1). There is a significant link between the CFL1 expression and CD4⁺ T cells infiltrating levels within 14 different cancers, CD8⁺ T cells within 10 different cancers, neutrophils within 14 different cancers, macrophages within 17 different cancers, B cell within 17 different cancers, and dendritic cells within 17 different cancers. KIRC, KIRP and LIHC are the three cancer types most related to CFL1 expression at the immune infiltration level (Fig. 6). In KIRC, CFL1 expression was positively related to B cell (r = 0.312, p = 1.87e^− 13), CD4⁺T cell (r = 0.319, p = 5.22e^− 14), CD8⁺T cell (r = 0.319, p = 5.22e^− 14), Dendrific (r = 0.499, p = 1.02e^− 34), Macrophage (r = 0.327, p = 1.01e^− 14), Neutrophil (r = 0.355, p = 2.93e^− 17). In KIRP, the level of CFL1 expression was positively related to B cell (r = 0.187, p = 0.00143), CD4⁺T cell (r = 0.203, p = 0.000532), CD8⁺T cell (r = 0.183, p = 0.00178), Dendrific (r = 0.369, p = 1.22e^− 10), Macrophage (r = 0.176, p = 0.00262), Neutrophil (r = 0.266, p = 5.05e^− 6). In LIHC, CFL1 expression was positively related to B cell (r = 0.454, p = 2.09e^− 20), CD4⁺T cell (r = 0.419, p = 2.64e^− 17), CD8⁺T cell (r = 0.385, p = 1.29e^− 14), Dendrific (r = 0.498, p = 8e^− 25), Macrophage (r = 0.504, p = 1.81e^− 25), Neutrophil (r = 0.411, p = 1.14e^− 16).

Table 1

Relationship between CFL1 expression and immune cell infiltration in pan-cancer
		CD4 T cell (P-value/cor)	CD8 T cell (P-value/cor)	Neutrophil (P-value/cor)	Marcophage (P-value/cor)	Dendritic (P-value/cor)
ACC	***/0.496	0.015	-0.046	**/0.304	0.12	***/0.376
BLCA	-0.069	***/0.184	***/0.209	***/0.325	***/0.171	0.41
BRCA	*/0.077	0.041	**/-0.096	0.037	-0.059	**/0.095
CESC	-0.019	-0.004	-0.081	0.033	-0.036	0.083
CHOL	0.079	0.003	0.057	0.213	0.218	0.216
COAD	-0.065	0.017	-0.048	0.016	-0.019	0.004
DLBC	0.202	-0.249	-0.117	-0.009	0.067	-0.088
ESCA	*/-0.195	-0.034	-0.09	-0.05	*/-0.169	-0.066
GBM	-0.095	-0.09	-0.022	-0.118	***/-0.337	0.131
HNSC	***/-0.151	0.041	**/-0.118	***/0.165	*/0.089	0.078
KICH	**/0.355	0.215	0.233	-0.03	***/0.54	***/0.497
KICR	***/0.312	***/0.319	***/0.343	***/0.355	***/0.327	***/0.499
KIRP	**/0.187	***/0.203	**/0.183	***/0.266	***/0.176	***/0.369
LGG	*/0.096	***/0.159	-0.034	*/0.097	0.073	***/0.232
LIHC	***/0.454	***/0.419	***/0.358	***/0.411	***/0.504	***/0.498
LUAD	***/-0.222	0.025	-0.064	**/0.137	0.064	0.055
LUSC	*/-0.1	*/0.105	-0.061	*/0.107	0.016	*/0.1
MESO	0.211	**/0.326	0.086	-0.185	0.023	***/0.431
OV	0.048	*/0.133	-0.002	***/0.186	*/0.121	**/0.142
PAAD	0.137	0.109	0.016	0.145	0.003	***/0.26
PCPG	***/0.332	***/0.274	0.005	0.128	***/0.321	0.495
PRAD	-0.007	-0.088	*/-0.11	-0.052	*/-0.112	0.002
READ	-0.058	0.028	-0.104	-0.041	0.1	0.041
SARC	***/0.239	**/0.173	0.074	0.023	***/0.218	***/0.398
SKCM	0.032	0.036	-0.074	-0.034	-0.004	0.07
STAD	***/-0.311	***/-0.228	0.033	0.01	***/-0.183	0.012
TGCT	-0.071	0.023	-0.21	-0.013	0.005	-0.002
THCA	**/0.143	**/0.122	*/0.096	***/0.197	-0.003	***/0.204
THYM	***/0.628	***/0.769	***/0.49	***/-0.305	***/0.465	0.815
UCEC	-0.056	-0.039	0.018	***/0.155	**/-0.137	0.04
USC	0.206	0.174	-0.031	-0.016	-0.037	0.16
UVM	***/-0.479	0.106	0.186	***/-0.372	0.205	**/0.311
p < 0.05,p < 0.01, and**p < 0.001.

The ESTIMATE algorithm uses the transcription profile of cancer patient samples to infer infiltrating immune cells and stromal cells. The BLCA, LIHC, THCA, LGG, KIRC, LAML, PCPG, SARC, KIRP, UVM and OV had a positive correlation with the Immune, Stromal, and ESTIMATE Scores (Table 2). The top three cancers with a significant correlation to CFL1 expression were KIRC, THCA, and BLCA (Stromal Score), KIRC, LIHC and LGG (Immune Score), and KIRC, THCA and BLCA (ESTIMATE score) (Fig. 7).

Table 2

ImmuneScore、StromalScore and ESTIMATEScore relate with CFL1 expression
	ImmuneScore (p-value/cox)	StromalScore (p-value/cox)	ESTIMATEScore (p-value/cox)
ACC	-0.086	-0.006	-0.045
BLCA	***/0.264	***/0.264	***/0.279
BRCA	***/0.125	-0.005	*/0.074
CESE	-0.037	0.078	0.016
CHOL	0.269	0.301	0.31
COAD	-0.035	0.019	-0.002
DLBC	-0.063	0.166	0.084
ESCA	-0.033	-0.117	-0.075
GBM	-0.132	-0.115	-0.12
HNSC	0.023	**/0.143	*/0.097
KICH	*/0.313	0.239	*/0.299
KIRC	***/0.271	***/0.387	***/0.368
KIRP	*/0.146	**/0.161	***/0.165
LAML	***/0.419	*/0.201	***/0.344
LGG	***/0.237	***/0.14	***/0.204
LIHC	***/0.304	***/0.179	***/0.204
LUAD	0.002	0.057	0.033
LUSC	-0.021	0.008	0.009
MESO	0.111	*/0.267	0.182
OV	**/0.153	**/0.155	**/0.167
PAAD	-0.002	0.027	0.014
PCPG	***/0.268	***/0.227	***/0.265
PRAD	0.011	-0.064	-0.022
READ	-0.073	0.103	0.031
SARC	***/0.319	**/0.16	***/0.276
SKCM	0.01	0.009	0.008
STAD	0.039	-0.064	-0.022
TGCT	0.012	***/0.321	*/0.174
THCA	***/0.227	***/0.296	***/0.265
THYM	***/0.355	***/-0.358	0.018
UCEC	-0.083	-0.043	0.069
UCS	0.161	0.024	0.083
UVM	***/0.46	***/0.497	***/0.489
p < 0.05, p < 0.01, and**p < 0.001.

4. Relationship between Immune Marker Sets, TMB, and MSI with CFL1 expression in pan-cancer

To clarify the CFL1 and immunotherapy efficacy’s correlation, a CFL1 gene co-expression analysis was conducted. The analyzed 44 genes encoded Immune checkpoint proteins. Almost all genes involved in the immune system were co-expressed with CFL1 (Fig. 8A), and most of them had a positive correlation with CFL1 in different cancers, except COAD, DLBC, ESCA, GBM, HNSC, LUSC, PRAD, UCEC, STAD, READ, SKCM, UCS, TGCT, SARC, MESO, OV, and UVM (* p < 0.05, ** p < 0.01, *** p < 0.001).

The CFL1, tumor mutation burden (TMB), and microsatellite instability (MSI) correlation with immunotherapy sensitivity were assessed. The CFL1 expression and TMB were linked in 14 different cancers, including breast, colorectal, lung, kidney cancers, and glioma (Fig. 8B). In another six different cancers, CFL1 was negatively correlated with microsatellite instability in gliomas, while positively related with Melanoma, UCEC (Uterine Corpus Endometrial Carcinoma, kidney cancer and liver cancer (Fig. 8C).

5. Gene set enrichment analysis

We investigated the CFL1 expression’s biological role in tumor tissue. We conducted GSEA to analyze KEGG and HALLMARK pathways, as shown in Fig. 9. The result of KEGG indicates that CFL1 positively regulates TASTE Transduction, ABC Transporters, Nitrogen Metabolism and LINOLEIC Acid Metabolism. Then, the data of the HALLMARK pathway indicates that CFL1 positively regulates UV Response, PANCREAS Beta, Bile Acid Metabolism and KRAS Signaling. In contrast, CFL1 is predicated as a negative regulator of PYRIMIDINE METABOLISM, PATHOGENIC ESCHERICHIA COLQ_INFECTION and PROTEASOME by KEGG. Furthermore, CFL1 is also predicated negative regulator of GLYCOLYSIS, DNA REPAIR and MTORC1 SIGNALING.

Previous studies about CFL1 had focused on autoimmune diseases, T-cell differentiation, Alzheimer's disease, and tumorigenesis among the biological function of CFL1 (8, 17–19). The CFL1 expression was higher in pancreatic cancer, NSCLC, hepatocellular carcinoma, and esophageal squamous cell carcinoma (7, 12, 13, 20, 21). In epithelial ovarian cancer cells, phosphorylation of Cofilin-1 enhances Paclitaxel resistance by preventing the apoptosis of cancer cells (22). In human mutant glioma, CFL1 was identified as novel immunogenic tumor-associated antigens by mass spectrometry (23). Nicola Ternette used nano-ultra performance liquid chromatography-tandem mass spectrometry (nUPLC-MS²) for analyzing the immunopeptidomes from HLA-A2-positive triple-negative breast cancer (TNBC) samples, and the result shows that the potential of CFL1 as an immunotherapeutic target for breast cancer can be further evaluated (24). The exact CFL1 gene function in pan-cancer is not fully indicated. Therefore, our study used biological information analysis to determine the CFL1 role in pan-cancer and its effect on the tumor microenvironment.

The CFL1 was highly expressed in BLCA, BRCA, CHOL, GBM, HNSC, KIRC, KIRP, LIHC, LUSC, OV, PAAD, STAD, UCEC. However, the CFL1 high expression and poor prognosis are correlated in ACC, BLCA, CESE, HNSC, KICH, LAML, LGG, LIHC, LUAD, MESO, PAAD and UVM.

Stimulating the immune system to attack tumors is a promising cancer treatment. Current research focuses on stimulating the immune system by removing the immune system suppression by tumor cells and using immune drugs to stimulate the immune system. The clinical success of immune checkpoint inhibitors has given new life to the immunotherapy of cancer. After analyzing the link between CFL1 expression and immune checkpoint gene expression, we found the CFL1 expression positive correlation with CD276 in most cancers. B7 homolog 3 protein, also known as CD276, was identified as an attractive and promising target for cancer immunotherapy (25). This study results showed, the CFL1 expression was significantly linked to CD276, which we defined as immunosuppression in 27 kinds of tumors. This may indicate that CFL1 affect the therapeutic effect of immunotherapy in pan-cancer by affecting CD276.

Tumorigenesis is a gradual process that has many response levels and the mutations’ accumulation, in which the cancerous cell lines become increasingly independent of the regulatory mechanisms in vivo and gradually invade normal tissues. Following a malignant transformation, the cancer cell continues accumulating mutations giving them new characteristics that make them more dangerous. TMB was a biomarker to predict ICI outcomes (26). A phenomenon defined as microsatellite instability (MSI) is related to the mismatch repair (MMR) genes’ function loss. A previous study shows that MSI was an important biomarker in immune-checkpoint inhibitors (27). Our results showed, the CFL1 expression was related with 14 types of cancer in TMB and six kinds of cancer in MSI. These findings showed that the CFL1 expression levels affect the TMB and MSI in patients and thus affect the efficacy of immune checkpoint inhibitors. The specific mechanism of the correlation between CFL1, TMB and MSI still need more research.

The high CFL1 expression was linked with proteasomes, as the enrichment analysis showed. The proteasome function was the generation of antigenic peptides presented in complex with class I MHC molecules. And proteasomes can mediate activation of antitumor T cells (28). Therefore, our study clarifies the CFL1 role in tumor immunology and the prognostic significance in pan-cancer.

To summarize, this study shows that CFL1 can be considered an independent prognostic factor for ACC. The poor prognosis is related to its high levels. The CFL1 expression and TMB, MSI and immune cell infiltration were linked in different cancers. This study findings clarify the CFL1 role in tumorigenesis and development. It can be used as guidance to develop immunotherapies in the future.

Acknowledgements

We acknowledge Oncomine, TCGA,and TIMER databases for providing the platform and contributors to uploadea their meaningful datasets.

Ethics approval and consent to participate

All procedures were performed in accordance with relevant guidelines and regulations.

Consent for publication

All authors read and approved the final manuscript.

Availability of data and materials

The datasets in the current study come from Oncomine database (https://www.oncomine.org/resource/login.html), TCGA database(https://www.cureline.com/the-cancer-genome-atlas.html), and the Tumor Immune Estimate Resource(TIMER https:// cistrome.shinyapps.io/timer/).

Competing interests

All authors declare that they have no competing interests.

Funding

This work was supported by grants from the Science and Technology Commission of Shanghai Municipality (21ZR1439900).

Authors’ contributions

Jin Yang: writing - original draft. Cheng-Jie Qiu: conceptualization, visualization. Shan Zhong: statistical analysis. Yi Xiang: writing - review & editing. Si-Jian Pan: project administration.

Berger MF, Mardis ER. The emerging clinical relevance of genomics in cancer medicine. Nat Rev Clin Oncol. 2018;15(6):353–65.
Kraehenbuehl L, Weng CH, Eghbali S, Wolchok JD, Merghoub T. Enhancing immunotherapy in cancer by targeting emerging immunomodulatory pathways. Nat Rev Clin Oncol. 2021.
Ambrosini M, Fuca G, Duca M, Damian S, De Santis F, Corti F, et al. Targeted protein degraders from an oncologist point of view: the Holy Grail of cancer therapy? #N/A. 2021:103532.
Fadhil I, Alkhalawi E, Nasr R, Fouad H, Basu P, Camacho R, et al. National cancer control plans across the Eastern Mediterranean region: challenges and opportunities to scale-up. Lancet Oncol. 2021;22(11):e517-e29.
Sun C, Mezzadra R, Schumacher TN. Regulation and Function of the PD-L1 Checkpoint. Immunity. 2018;48(3):434–52.
Rebuzzi SE, Banna GL, Murianni V, Damassi A, Giunta EF, Fraggetta F, et al. Prognostic and Predictive Factors in Advanced Urothelial Carcinoma Treated with Immune Checkpoint Inhibitors: A Review of the Current Evidence. Cancers (Basel). 2021;13(21).
Zhang L, Chai Z, Kong S, Feng J, Wu M, Tan J, et al. Nujiangexanthone A Inhibits Hepatocellular Carcinoma Metastasis via Down Regulation of Cofilin 1. Front Cell Dev Biol. 2021;9:644716.
Howard J, Goh CY, Gorzel KW, Higgins M, McCann A. The potential role of cofilin-1 in promoting triple negative breast cancer (TNBC) metastasis via the extracellular vesicles (EVs). Transl Oncol. 2021;15(1):101247.
Noun M, Akoumeh R, Abbas I. Cell and Tissue Imaging by TOF-SIMS and MALDI-TOF: An Overview for Biological and Pharmaceutical Analysis. Microsc Microanal. 2021:1–26.
Yao B, Li Y, Chen T, Niu Y, Wang Y, Yang Y, et al. Hypoxia-induced cofilin 1 promotes hepatocellular carcinoma progression by regulating the PLD1/AKT pathway. Clin Transl Med. 2021;11(3):e366.
Lu LI, Fu NI, Luo XU, Li XY, Li XP. Overexpression of cofilin 1 in prostate cancer and the corresponding clinical implications. Oncol Lett. 2015;9(6):2757–61.
Wang WS, Zhong HJ, Xiao DW, Huang X, Liao LD, Xie ZF, et al. The expression of CFL1 and N-WASP in esophageal squamous cell carcinoma and its correlation with clinicopathological features. Dis Esophagus. 2010;23(6):512–21.
Wang L, Xiong L, Wu Z, Miao X, Liu Z, Li D, et al. Expression of UGP2 and CFL1 expression levels in benign and malignant pancreatic lesions and their clinicopathological significance. World J Surg Oncol. 2018;16(1):11.
Wu Q, Jiang Y, Cui S, Wang Y, Wu X. The role of cofilin-l in vulvar squamous cell carcinoma: A marker of carcinogenesis, progression and targeted therapy. Oncol Rep. 2016;35(5):2743–54.
Rhodes DR, Kalyana-Sundaram S, Mahavisno V, Varambally R, Yu J, Briggs BB, et al. Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. Neoplasia. 2007;9(2):166–80.
Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102(43):15545–50.
Cui L, Elzakra N, Xu S, Xiao GG, Yang Y, Hu S. Investigation of three potential autoantibodies in Sjogren's syndrome and associated MALT lymphoma. #N/A. 2017;8(18):30039–49.
Bamburg JR, Bernstein BW. Actin dynamics and cofilin-actin rods in alzheimer disease. Cytoskeleton (Hoboken). 2016;73(9):477–97.
Salz A, Gurniak C, Jonsson F, Witke W. Cofilin1-driven actin dynamics controls migration of thymocytes and is essential for positive selection in the thymus. J Cell Sci. 2020;133(5).
Werle SD, Schwab JD, Tatura M, Kirchhoff S, Szekely R, Diels R, et al. Unraveling the Molecular Tumor-Promoting Regulation of Cofilin-1 in Pancreatic Cancer. Cancers (Basel). 2021;13(4).
Chang CY, Chang SL, Leu JD, Chang YC, Hsiao M, Lin LT, et al. Comparison of cofilin1 and Twist1 protein expression in human nonsmall cell lung cancer tissues. Oncol Rep. 2019;42(2):805–16.
Li M, Dong XD, Lyu QB, Zhang W, Huang S, Yang CX, et al. Phosphorylation of Cofilin-1 Enhances Paclitaxel Resistance of Epithelial Ovarian Cancer Cells by Inhibiting Apoptosis. Biomed Environ Sci. 2021;34(6):465–77.
Dettling S, Stamova S, Warta R, Schnolzer M, Rapp C, Rathinasamy A, et al. Identification of CRKII, CFL1, CNTN1, NME2, and TKT as Novel and Frequent T-Cell Targets in Human IDH-Mutant Glioma. Clin Cancer Res. 2018;24(12):2951–62.
Ternette N, Olde Nordkamp MJM, Muller J, Anderson AP, Nicastri A, Hill AVS, et al. Immunopeptidomic Profiling of HLA-A2-Positive Triple Negative Breast Cancer Identifies Potential Immunotherapy Target Antigens. Proteomics. 2018;18(12):e1700465.
Zhou WT, Jin WL. B7-H3/CD276: An Emerging Cancer Immunotherapy. Front Immunol. 2021;12:701006.
Fumet JD, Truntzer C, Yarchoan M, Ghiringhelli F. Tumour mutational burden as a biomarker for immunotherapy: Current data and emerging concepts. Eur J Cancer. 2020;131:40–50.
Sahin IH, Akce M, Alese O, Shaib W, Lesinski GB, El-Rayes B, et al. Immune checkpoint inhibitors for the treatment of MSI-H/MMR-D colorectal cancer and a perspective on resistance mechanisms. Br J Cancer. 2019;121(10):809–18.
Tundo GR, Sbardella D, Oddone F, Kudriaeva AA, Lacal PM, Belogurov AA, Jr., et al. At the Cutting Edge against Cancer: A Perspective on Immunoproteasome and Immune Checkpoints Modulation as a Potential Therapeutic Intervention. Cancers (Basel). 2021;13(19).

No competing interests reported.

SupplementaryTable1.docx

Cofilin1 Is a Poorer Prognostic Biomarker and potential Immunotherapy in Systematic Pan-Cancer

Status:

Version 1

Abstract

Figures

Introduction

Methods

2.1 ONCOMINE database analysis

2.2 Analysis of the link between CFL1 and clinical prognosis

2.3 Association between the CFL1 expression and immune

2.4 Relationship between CFL1, Tumor Mutation Burden and Microsatellite instability

2.5 Gene set enrichment analysis

2.6 Statistical analysis

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1