AFAP1L1 is indicative for a grim prognosis and immune microenvironment in gastric cancer

doi:10.21203/rs.3.rs-4823876/v1

Download PDF

Article

AFAP1L1 is indicative for a grim prognosis and immune microenvironment in gastric cancer

https://doi.org/10.21203/rs.3.rs-4823876/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The biological role of the actin filament associated protein 1 like 1(AFAP1L1) has been investigated in human malignancies, but its function in gastric cancer (GC) is unclear. This research sought to elucidate more about AFAP1L1's biological function in GC and its prognostic relevance by analyzing its expression profiles and prognostic significance using bioinformatic and immunohistochemical analysis based on large-scale databases and clinical samples. The comparative analysis of normal and tumor tissues indicated that the latter had elevated levels of AFAP1L1 expression level, which was linked to dismal survival in GC patients. Multivariate Cox regression analysis showed that elevated AFAP1L1 expression was an independent factor for poor prognosis in GC patients. Functional enrichment analysis including GO, KEGG and GSEA illustrated that AFAP1L1 could act as an oncogene by regulating gene expression in essential functions and pathways of tumorigenesis, such as cell junction, protein kinase activity, angiogenesis-associated pathways, and immune response-associated pathways. Furthermore, immune cell infiltration results showed that AFAP1L1 was associated with the immune infiltration of macrophages and their polarization. In addition, AFAP1L1 was negatively related to the sensitivity of chemotherapy drug oxaliplatin, while positively with dabrafenib, indicating that AFAP1L1 could be used as a predictive marker of the curative effect of GC patients. In conclusion, AFAP1L1 may be employed as a diagnostic and prognostic biological marker, and it also offers more in-depth insights into the establishment of therapies and prognoses in GC individuals.

Biological sciences/Cancer

Health sciences/Biomarkers

Health sciences/Oncology

AFAP1L1

gastric cancer

prognosis

immune infiltration

Gastric cancer (GC) is among the most prevalent cancers of the digestive tract and ranks as the fourth predominant contributor to deaths associated with cancer, which seriously endangers human health. In the year 2020, there were over one million newly diagnosed cases of GC and nearly 80,000 deaths attributed to the disease^[1]. However, the diagnosis is frequently made at a late stage during the disease progression course, when only palliative therapy options exist. Even though there have been advancements in GC treatment, such as surgery, radiation treatment, immunotherapy, and chemotherapy, the prognosis is still not promising. Stage IV GC has a dismal 5-year overall survival (OS) rate of under 10%^[2]. Improving GC prognosis requires prompt diagnosis and therapy. As a result, it is essential to investigate credible biological markers that can forecast the prognosis with high precision and pinpoint therapeutic targets.

Actin filament-associated protein 1-like 1(AFAP1L1) is an adaptor protein of the actin-filament-associated protein (AFAP) family that has a strong interaction with cortactin and is mostly found in invadosomes^[3]. Recently, extensive research has primarily focused on the biological roles of AFAP1L1 in human cancers and demonstrated that AFAP1L1 is a driving regulator for invasion and progression in different types of cancers. After analyzing the gene expression patterns of 65 cases of spindle cell sarcoma, Furu et al^[4] identified AFAP1L1 as a putative predictive marker for metastasis. In sarcoma cells, knocking down AFAP1L1 contributed to the suppression of cell invasion. AFAP1LI regulated by specificity protein 3 (Sp3)^[5] influences the potential of sarcoma cells to migrate and invade through a phosphotyrosine-dependent pathway^[6]. Takahashi et al^[7] discovered that rectal cancer recurrences were significantly and independently influenced by AFAP1L1. Alterations in cell shape and motility are regulated by AFAP1L1, which has a function in the advancement and metastasis of colorectal cancer. Furthermore, non-small cell lung cancer cells' ability to proliferate and survive is mediated by AFAP1L1. Silencing AFAP1L1 resulted in a dramatic increase in P38 and caspase 3 activation and suppression in the activation of PRAS40^[8]. These research reports uncovered AFAP1L1's inherent significance in malignancies and may point the way toward clarifying its exact function and molecular basis. Nevertheless, there are currently only a few research reports emphasizing AFAP1L1's involvement in GC.

In the first part of our investigation, we evaluated the levels of AFAP1L1 expression in normal and cancerous tissues based on the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases. The next step was to scrutinize the association of AFAP1L1 expression with the clinical features and survival outcomes of GC patients, to better understand the clinical importance of AFAP1L1 in this population. In addition, the probable mechanism and link between AFAP1L1 mRNA expression and tumor immune microenvironment (TIME) were investigated. Ultimately, we probed the link between AFAP1L1 expression and a patient's responsiveness to treatment. The findings shed light on the function of AFAP1L1 in GC tumor growth and the associated immune response against GC.

The expression of AFAP1L1 in GC

Initially, using data gathered from the TCGA database, we compared the levels of AFAP1L1 gene expression found in various types of human cancer tissues with those found in matched healthy controls. The messenger RNA level of AFAP1L1 was shown to be considerably greater in a variety of cancerous tissues compared to the matched normal controls, including head and neck squamous cell carcinoma (HNSCC), cholangiocellular carcinoma (CHOL), stomach adenocarcinoma (STAD), colon adenocarcinoma (COAD), etc, while its expression was downregulated in lung squamous carcinoma (LUSC), breast cancer (BRCA), and lung adenocarcinoma (LUAD), etc (Fig. 1A, B). Specifically, AFAP1L1 was meaningfully up-regulated in GC (Fig. 1C, D). Furthermore, additional validation of the expression of AFAP1L1 was achieved via the use of three separate datasets (GSE19826, GSE54129, and the ACRG Gastric cohort). Across the three datasets, AFAP1L1 overexpressed in tumors relative to normal tissues (Fig. 1F-H). The ROC curve illustrated that the AFAP1L1 expression possessed a remarkable capacity to anticipate outcomes, with an area under the curve (AUC) of 0.741 (95% confidence interval [CI] = 0.0.644–0.838) to differentiate cancerous tissues from healthy tissues (Fig. 1E). Afterward, we employed an IHC assay to assess AFAP1L1 expression in 48 paired tumors and adjacent normal tissue obtained from the Peking University People’s Hospital (PKUPH cohort). As shown by the data, AFAP1L1 IHC scores were considerably elevated in tumors as opposed to normal tissues (Fig. 1I, J). Overall, it was evident that the AFAP1L1 expression level was remarkably elevated in GC.

Association of AFAP1L1 expression with clinical features of GC patients

Additionally, we assessed whether there was a link between the levels of AFAP1L1 expression and the clinicopathological features of the 362 patients who were included in the TCGA database and 48 patients from PKUPH cohort. Based on TCGA, it was confirmed that AFAP1L1 expression was substantially linked to the T stage, OS event, DSS event, and PFI event using Welch one-way ANOVA followed by the Bonferroni correction (Fig. 2A-I). Additionally, the median AFAP1L1 expression value was taken as the dividing line for establishing high- and low-expression groups (categories). The clinicopathologic factors below were then correlated with AFAP1L1 expression in malignant cells: sex, age when first diagnosed, clinical stage, distance metastasis, T stage, and N stage. Table 1 summarizes the associations of clinicopathological characteristics with the expression of AFAP1L1 in 644 GC patients. There was a positive link between AFAP1L1 overexpression and advanced T stage (p = 0.025) relative to low expression. Other clinicopathologic variables were insignificantly correlated with AFAP1L1 expression profiles. Once the relationship of clinicopathological factors with GC was verified via logistic regression analysis, the tumor stage (p = 0.010) was shown to be the final factor affecting AFAP1L1 expression (Table 2). As for PKUPH cohort, overexpression of AFAP1L1 was associated with advanced T stage (p = 0.043) and clinical stage (p = 0.008) compared to low expression (Table 3).

Table 1

The correlation of clinicopathological characteristics and AFAP1L1 expression based on TCGA cohort.
Characteristic	AFAP1L1		P value
Characteristic	Low expression	High expression	P value
n	187	188
T stage, n (%)			0.025
T1	14 (3.8%)	5 (1.4%)
T2	47 (12.8%)	33 (9%)
T3	82 (22.3%)	86 (23.4%)
T4	42 (11.4%)	58 (15.8%)
N stage, n (%)			0.723
N0	59 (16.5%)	52 (14.6%)
N1	49 (13.7%)	48 (13.4%)
N2	38 (10.6%)	37 (10.4%)
N3	33 (9.2%)	41 (11.5%)
M stage, n (%)			0.942
M0	168 (47.3%)	162 (45.6%)
M1	12 (3.4%)	13 (3.7%)
Pathologic stage, n (%)			0.139
Stage I	33 (9.4%)	20 (5.7%)
Stage II	57 (16.2%)	54 (15.3%)
Stage III	66 (18.8%)	84 (23.9%)
Stage IV	18 (5.1%)	20 (5.7%)
Gender, n (%)			0.945
Female	66 (17.6%)	68 (18.1%)
Male	121 (32.3%)	120 (32%)
Age, n (%)			1.000
<=65	82 (22.1%)	82 (22.1%)
> 65	104 (28%)	103 (27.8%)

Table 2

Logistics association between AFAP1L1 expression and clinicopathologic variables.
Characteristics	Total(N)	Odds Ratio (OR)	P value
T stage (T3&T4 vs. T1&T2)	367	1.864 (1.169–3.003)	0.010
N stage (N1&N2&N3 vs. N0)	357	1.191 (0.761–1.870)	0.445
M stage (M1 vs. M0)	355	1.123 (0.495–2.568)	0.779
Pathologic stage (Stage III&Stage IV vs. Stage I&Stage II)	352	1.506 (0.989–2.298)	0.057
Age (> 65 vs. <=65)	371	0.990 (0.657–1.493)	0.963
Gender (Male vs. Female)	375	0.963 (0.630–1.469)	0.860

Table 3

The correlation of clinicopathological characteristics and AFAP1L1 expression based on PKUPH cohort.
Characteristics	AFAP1L1		P value
Characteristics	Low expression	High expression	P value
n	14	34
T stage, n (%)			0.043
T4	4 (8.3%)	20 (41.7%)
T3	4 (8.3%)	9 (18.8%)
T2	2 (4.2%)	4 (8.3%)
T1	4 (8.3%)	1 (2.1%)
N stage, n (%)			0.168
N3	2 (4.2%)	10 (20.8%)
N1	1 (2.1%)	5 (10.4%)
N0	9 (18.8%)	10 (20.8%)
N2	2 (4.2%)	9 (18.8%)
M stage, n (%)			0.165
M0	14 (29.2%)	27 (56.2%)
M1	0 (0%)	7 (14.6%)
Pathologic stage, n (%)			0.009
Stage III	3 (6.2%)	12 (25%)
Stage II	5 (10.4%)	13 (27.1%)
Stage I	6 (12.5%)	2 (4.2%)
Stage IV	0 (0%)	7 (14.6%)
Age, n (%)			0.412
< 65	6 (12.5%)	19 (39.6%)
> 65	8 (16.7%)	15 (31.2%)
Gender, n (%)			1
female	3 (6.2%)	8 (16.7%)
male	11 (22.9%)	26 (54.2%)

Prognostic value of AFAP1L1 in GC

Here, we explored the predictive significance of AFAP1L1 for OS, DSS, and PFI across all GC patients. Patients having low AFAP1L1 expression, as determined by KM survival analysis, fared better than those with high expression in terms of OS (Fig. 3A), while there was no statistical difference for DSS and PFI (Fig. 3B, C). Furthermore, subgroup analyses showed that elevated AFAP1L1 level was correlated with poorer OS in T3 + 4, N1 + 2 + 3, M0, TNM I + II, and III + IV patients (Fig. 3E, G, H, J, K) rather than T1 + 2 (Fig. 3D) and N0 (Fig. 3F). Although the p-value of the log-rank test did not attain a significance threshold, low survival rates were seen among M1 patients with elevated AFAP1L1 expression (Fig. 3I). Both univariate and multivariate analyses were conducted to ascertain if AFAP1L1 expression and the clinicopathological features may independently serve as risk variables for GC patients. Accordingly, the univariate analysis confirmed that age > 65 years old (p = 0.005), T3 and T4 stage(p = 0.011), N1, N2, and N3 stage(p = 0.002), distant metastasis (p = 0.004), TNM III/IV stage (p < 0.001), and AFAP1L1 overexpression (p = 0.009) were substantially linked to a decrease in OS. On the other hand, multivariate analysis proved that age (p = 0.002), M stage (p = 0.017) and AFAP1L1 levels (p = 0.026) independently functioned as prognosticators for OS (Table 4, Fig. 4A, B). Time-dependent ROC analysis was also used to appraise AFAP1L1's predictive ability for OS over 2, 4, and 6 years. The ROC analysis of AFAP1L1 supported the predictive accuracy (Fig. 3L).

Table 4

The univariate and multivariate analysis of OS.
		Univariate analysis		Multivariate analysis
Characteristics	Total(N)	HR (95% CI)	P value	HR (95% CI)	P value
T stage	362
T1&T2	96	Reference
T3&T4	266	1.719 (1.131–2.612)	0.011	1.140 (0.677–1.920)	0.622
N stage	352
N0	107	Reference
N1&N2&N3	245	1.925 (1.264–2.931)	0.002	1.538 (0.857–2.763)	0.149
M stage	352
M0	327	Reference
M1	25	2.254 (1.295–3.924)	0.004	2.109 (1.143–3.891)	0.017
Age	367
<=65	163	Reference
> 65	204	1.620 (1.154–2.276)	0.005	1.795 (1.237–2.603)	0.002
Gender	370
Female	133	Reference
Male	237	1.267 (0.891–1.804)	0.188
Pathologic stage	347
Stage I& Stage II	160	Reference
Stage III &Stage IV	187	1.947 (1.358–2.793)	< 0.001	1.253 (0.721–2.179)	0.424
AFAP1L1	370
Low	185	Reference
High	185	1.560 (1.116–2.181)	0.009	1.505 (1.049–2.159)	0.026

The independent factors of OS were used to develop a nomogram that may be used to anticipate GC patients' prognoses. A poorer prognosis was linked to a greater overall score on the nomogram (Fig. 4C). The nomogram's capacity to make an accurate prediction was further evaluated using calibration curves (Fig. 4D). Nomogram C-index equaled 0.624 (95% CI = 0.599–0.649), illustrating that the model was moderately accurate in anticipating OS for GC patients. The nomogram was appropriate as supported by these findings.

Functional inference of AFAP1L1 in GC

The correlation between AFAP1L1 and the 50 leading genes with significant relationships was examined by spearman analysis and demonstrated in a heat map (Fig. 5A), in which cadherin-5 (CDH5) was the gene with the strongest correlation with AFAP1L1 expression (Fig. 5B). Interestingly, similar to AFAP1L1, patients having low CDH5 expression recorded superior OS results in contrast with cases with high CDH5 expression (Fig. 5C). GO term annotation indicated that genes that are co-expressed and have a correlation value that is higher than 0.7 of AFAP1L1 joined mainly in the regulation of vasculature development, regulation of angiogenesis, cell-cell junction, collagen-containing extracellular matrix, growth factor binding, and protein kinase activity, etc. (Fig. 5D). The KEGG pathway findings illustrated predominant enrichment in the focal adhesion, PI3K-Akt signaling pathway, etc. (Fig. 5D). The GSEA was executed to identify the various signaling pathways linked to GC. The Hallmark results illustrated a differential enrichment of the apical junction, epithelial-mesenchymal transition (EMT), KRAS signaling, apoptosis, and angiogenesis in the phenotype that was positively linked to AFAP1L1 expression (Fig. 5E). Similarly, the KEGG results highlighted the pathways in cancer, cell adhesion molecules (CAM) focal adhesion, MAPK signaling pathway, and tight junction were enriched (Fig. 5F). Interestingly, we found that immune response-associated pathways particularly inflammatory response, IL2-STAT5 signaling, TNFA signaling via NFKB, TGF-BETA signaling, and IL6-JAK-STAT3 signaling in Hallmark and TGF-BETA signaling pathway, chemokine signaling pathway, and cytokine-cytokine receptor interaction in KEGG were also enriched (Fig. 5G, H), suggesting that AFAP1L1 not only implicated in the cancer cell invasiveness but also immune regulation in the tumor microenvironment (TME).

AFAP1L1 expression correlates with infiltration levels of macrophages

The ESTIMATE, stromal, and immune scores were all determined utilizing the ESTIMATE method. The GC samples with high AFAP1L1 expression had higher ESTIMATE, stromal, and immune scores than those with low AFAP1L1expression (Fig. 6A-C). Spearman correlation was conducted to examine the link between AFAP1L1 expression and the multitude of distinct types of immune cells infiltrating a tissue, as measured by the TIMER database, including B cells, dendritic cells, CD4 positive T cells, macrophages, neutrophils, and CD8 positive T cells. Notably, AFAP1L1 expression was most strongly correlated with macrophages (r = 0.52, p < 0.001), which is an integral part of the TME and contributes to the regulation of angiogenesis, remodeling of the extracellular matrix (ECM), the proliferation of cancerous cells, metastases, immunosuppressive function, chemoresistance, and insensitivity to immune checkpoint blockade (ICB) therapy (Fig. 6D-I)^[9]. We also explored how AFAP1L1 expression relates to cell type-specific macrophage markers. Our findings illustrated the correlation between AFAP1L1 and M2 macrophages (CD163, r = 0.426, p < 0.001, CD206: r = 0.419, p < 0.001) (Fig. 6J-L). These findings suggested a possible function for AFAP1L1 in the polarization from M1 to M2 in GC.

Based on our GSEA results indicating that an elevated AFAP1L1 level may enhance immune cell migration through chemokines and chemokine receptors, our research has established a correlation between the expression of AFAP1L1 and chemokines and receptors based on TISIDB database^[10] (Fig. 7A-C). Notably, CXCL12 (r = 0.601, p < 0.001) and CXCR41 (r = 0.403, p < 0.001) were identified as the most relevant chemokine and receptor, respectively (Fig. 7D, E). Intriguingly, patients with low CXCL12 and CXCR4 expression exhibited superior overall survival outcomes compared to those with high expression, similar to the findings for AFAP1L1 (Fig. 7F, G). Moreover, CXCL12 and CXCR4 were also strongly correlated with macrophages and the markers of M2 macrophages (Fig. 7H-J).

In addition, a comparison was made between the expression of AFAP1L1 and genes associated with immune suppression. The results revealed that AFAP1L1 exhibited a robust association with the majority of targets within the immunoinhibitory microenvironment, as identified by TISIDB (Fig. 8A). Notably, the vascular endothelial growth factor receptor 2 (VEGFR2), also known as KDR, displayed the most significant correlation with AFAP1L1 (Fig. 8B). Furthermore, KDR was found to be a prognostic indicator for poor outcomes and macrophage infiltration, and was strongly linked to markers of M2 macrophages (Fig. 8C-G). This may be linked to the mechanism through which upregulated AFAP1L1 contributes to a dismal prognosis of GC.

Subsequently, we proceeded to verify the association between the expression of AFAP1L1 and the M2 macrophage marker CD163 by utilizing surgically excised paraffin samples of GC. Additionally, the employment of immunofluorescence staining revealed a significant upregulation of CD163 in GC tissues exhibiting elevated levels of AFAP1L1 expression (p < 0.05) (Fig. 8H, I).

Sensitivity analysis of drugs

The link between AFAP1L1 expression and pharmacological sensitivity was established after obtaining gene expression and drug sensitivity data from CellMiner and filtering out pharmaceuticals that are not approved by the FDA. We selected the topmost 12 drugs associated with AFAP1L1(Fig. 9A). Notably, elevated AFAP1L1 level was linked to lower tumor cells’ sensitivity to oxaliplatin, a common drug for chemotherapy of gastric cancer with the most significant correlation of -0.411. These results supported the finding that AFAP1L1 upregulation might be implicated in the formation of chemoresistance in GC and highlighted the potential use of AFAP1L1 as a determinant of chemosensitivity.

In order to test the aforementioned hypothesis, a total of nine surgical samples were obtained from GC patients who underwent neoadjuvant chemotherapy based on oxaliplatin. The efficacy of the treatment was assessed by means of tumor regression grading (TRG). Out of the nine samples, three were assigned a TRG grading of 1, 2, and 3, respectively. Immunohistochemical analysis revealed that the expression of AFAP1L1 was lower in tumor tissues of patients who responded well to the treatment, suggesting that AFAP1L1 expression may contribute to chemotherapy resistance in patients with GC (Fig. 9B).

What’s more, we found that AFAP1L1 expression was favorably linked to the sensitivity of dabrafenib. Such findings indicated that the patients exhibiting elevated AFAP1L1 levels may benefit from dabrafenib.

Gastric cancer is among the most commonly occurring malignancies globally and ranked 4th among the major contributors to deaths associated with cancer^[1]. Stomach adenocarcinoma (STAD), is mainly accountable for 95% of GC cases, as determined by the pathological staging of the disease^[11]. The dismal survival rates of individuals with GC are attributed to interfering confounding variables associated with the disease etiology. Patients who are detected at an early stage, have an illness that is restricted to the mucosa and submucosa, and who receive surgical therapy have superior survival rates at the five-year mark. Patients who have advanced stages of the disease, on the other hand, are frequently accompanied by recurrence and metastasis following surgical resection, and this is the most prevalent contributor to the grim prognosis observed among this population^[12]. Hence, finding reliable biological markers that may assist in detecting GC at an earlier stage and that can facilitate the identification of effective treatment targets is of utmost significance for determining the outcomes of the patient's condition.

Recently, accumulating evidence has revealed that AFAP1L1 may be a driving force for carcinogenesis. Many studies demonstrated the high expression of AFAP1L1 in sarcomas^[4], colorectal cancer^[7], and NSCLC^[8]. Recent research discovered that AFAP1L1 facilitates the progression of gastric cancer by inducing epithelial-to-mesenchymal transition^[13]. However, the study had yet to elucidate the function of AFAP1L1 in the immune microenvironment of gastric cancer, as well as its significance in the treatment of the disease. Furthermore, additional verification is required to determine the expression and prognostic value of AFAP1L1 in gastric cancer.

We began our investigation by assessing AFAP1L1 expression and measuring its mRNA levels in a TCGA pan-cancer sample. Notably, we discovered that AFAP1L1 mRNA levels were remarkably higher in GC, as well as many other malignancies, as opposed to normal tissues. The elevated AFAP1L1 levels in GC were additionally corroborated in three separate GEO datasets. In particular, AFAP1L1 overexpression was linked to a dismal survival rate, especially overall survival. Moreover, elevated AFAP1L1 level was substantially linked to the T stage of GC patients. Importantly, AFAP1L1 expression independently served as one of the predictors for OS in GC. Results like these highlighted the potential function of AFAP1L1 as a predictive and diagnostic marker in GC tumor growth.

Additionally, per the GO/KEGG enrichment and GSEA analysis, the potential mechanism of AFAP1L1 might be involved in cell junction^[14], protein kinase activity^[15], and angiogenesis^[16] associated pathways in Hallmark and KEGG, which were verified as driving forces of tumorigenesis, invasion, and metastasis. Notably, CDH5, the molecule with the strongest correlation with AFAP1L1, is linked to unfavorable survival in human gastric cancer^[17] and associated with neovasculogenesis^[18] and metastatic cancer^[19], indicating the potential mechanism via which AFAP1L1 contributes to the dismal prognosis of GC. Interestingly, GSEA illustrated that immune response and chemokine signaling pathways were implicated in the AFAP1L1 expression phenotype, indicating that AFAP1L1 might involve in the immune microenvironment.

Tumor progression is highly dependent on immune microenvironment^[20]. A worse prognosis in cancer patients is strongly linked to an immune system that lacks a balanced composition of cell types^[21]. Furthermore, the immune cells may be employed for prognostic prediction across different types of malignancies^[22]. In this work, we evaluated the AFAP1L1 expression and GC immune microenvironment. We utilized the TIMER database to observe tumor-infiltrating immune cells relative to AFAP1L1 expression. The results of the research imply that AFAP1L1 expression is intrinsically linked to macrophages.

Macrophages, which are among the most critical components of the immune microenvironment, participate in modulating innate immunity, tissue homeostasis, and inflammatory processes^{[23, 24]}. Macrophages can evolve into two distinct polarization states, either of which may be achieved by a process of targeted differentiation: classically activated M1 (pro-inflammatory) and, alternatively, activated M2 macrophages (anti-inflammatory)^[25]. It has been shown that M1 macrophages may enhance the immune system's capacity to fight against tumors. Contrastingly, M2 macrophages aid tumor growth by directly stimulating angiogenesis and/or indirectly eliciting immunosuppressive processes^[26]. In this investigation, we discovered that AFAP1L1 strongly correlated with M2 macrophages. Based on these findings, AFAP1L1 may enhance GC progression by activating M2 macrophages.

In order to investigate the mechanism by which AFAP1L1 governs the immune microenvironment, we conducted an analysis of the relationship between AFAP1L1 expression and chemokines and chemokine receptors. Our attention was particularly drawn to the chemokine CXCL12 and its receptor CXCR4 due to their notable correlation with AFAP1L1. According to the report, the binding of CXCL12 to CXCR4 resulted in the activation of signal transduction pathways that had extensive impacts on chemotaxis, cell proliferation, migration, and gene expression. Consequently, this axis functioned as a conduit for communication between tumor and stromal cells, thereby establishing a conducive microenvironment for tumor survival, development, angiogenesis, and metastasis^[27]. Additionally, the CXCL12-CXCR4 axis played a role in the recruitment of M2-polarized macrophages, which corroborated our prior findings^{[28, 29]}.

In our study, we investigated the correlation between AFAP1L1 and immunoinhibitory genes. Notably, a significant association was observed between AFAP1L1 and the immunoinhibitory gene KDR, also referred to as VEGFR2, the primary receptor of VEGF. The VEGF-VEGFR2 signaling cascade regulates angiogenesis, which in turn modulates tumor growth, invasion, the cancer-promoting microenvironment, and therapeutic resistance.^{[30, 31]}. The present study had revealed a significant association between KDR and unfavorable prognosis in patients with gastric cancer, as well as M2 macrophage infiltration. These findings suggested that KDR might serve as a crucial regulatory factor in the progression of gastric cancer involving AFAP1L1.

Finally, we examined the relationships between the sensitivity of different drugs and AFAP1L1. We discovered that AFAP1L1 was negatively related to the sensitivity of the chemotherapy drug oxaliplatin, suggesting that AFAP1L1 might be involved in the induction of chemoresistance in patients with GC. Oxaliplatin is an essential component of many of the chemotherapy treatment modalities for patients with GC^[32]. The 2016 National Comprehensive Cancer Network (NCCN) recommendations designated treatment modalities using oxaliplatin as the first-line chemotherapeutic agent for treating gastrointestinal cancerous tumors^[33]. The problem is that many individuals with GC do not improve with chemotherapy^[34], and insight into the mechanistic explanation behind oxaliplatin resistance is currently lacking. It is of great importance to shed light on the basic molecular processes behind oxaliplatin resistance in GC patients. Our research uncovered a possible function for AFAP1L1 in the modulation of oxaliplatin resistance, which suggests that this gene could function as a viable pharmacological target for managing chemoresistance in GC.

Furthermore, we confirmed that the expression of AFAP1L1 had a favorable link to the sensitivity of dabrafenib, a BRAF inhibitor approved by the United States Food and Drug Administration (FDA) used in combination treatment of malignant melanoma and non-small cell lung carcinoma^[35]. We demonstrated that the patients with AFAP1L1 overexpression might benefit from dabrafenib. Kang et al^[36] found that through the LIMK1-ADF/cofilin pathway, the anti-cancer drug dabrafenib inhibited the potential of GC cells to migrate and invade. Overall, we inferred that GC patients with elevated AFAP1L1 levels could choose dabrafenib instead of oxaliplatin to improve the treatment response, which is helpful to achieve precise treatment of GC.

Undeniably, additional research has to be done on both the biological significance of AFAP1L1 in enhancing GC progression and the exact modulatory mechanisms. Specifically, additional investigations, both basic and clinical, are required for a comprehensive understanding of the biological role that AFAP1L1 plays in GC.

In summary, our research showed that AFAP1L1 is expressed at a high level in GC and that elevated AFAP1L1 level is strongly linked to dismal survival outcomes and the sensitivity of patients to chemotherapy. Therefore, AFAP1L1 might function as a promising marker for predicting GC patients’ prognoses.

Collection and Processing of Data

We accessed the gene expression, relevant clinical data, and RNA-Seq data of 375 GC patients together with 32 normal tissues from the TCGA database (https://portal.gdc.cancer.gov/) for stomach adenocarcinoma (STAD). GSE19826^[37], GSE54129 and ACRG Gastric cohort (GSE62254, GSE66222, GSE66229)^{[38, 39]} were selected and culled from the GEO (http://www.ncbi.nlm.nih.gov/geo) database.

Patients and Clinical Specimens

Our research was granted approval by the ethical committees of the Peking University People's Hospital. The research was conducted in adherence to the guidelines outlined in the Declaration of Helsinki, and all patients provided written informed consent. The study included 57 eligible patients who underwent surgical treatment for GC at Peking University People's Hospital between May 2022 and January 2023, with nine cases receiving neoadjuvant chemotherapy with an oxaliplatin-based SOX regimen. In order to conduct an immunohistochemistry analysis, samples of tumor tissues and adjacent normal tissues, which were surgically excised and located at a distance greater than 5 cm from the tumor, were fixed in a 4% formalin solution and subsequently embedded in paraffin blocks.

Survival Analysis

The median AFAP1L1 expression value was taken as the criterion for survival analysis via the Kaplan-Meier (KM) technique and the log-rank test. We also analyzed the link between AFAP1L1 expression and the prognosis of individuals with different tumor stages. The predictive significance of AFAP1L1 in patients with GC was scrutinized via a time-dependent receiver operating characteristic (ROC) curve. Clinical factors' impact on patients' outcomes was examined via univariate and multivariate Cox regression models. In the subsequent multivariate Cox regression analysis, the prognostic factors that had a p-value < 0.1 in the univariate model were imported. Using a set of validated independent prognostic indicators identified from multivariate analysis, a nomogram was developed to estimate the likelihood of survival. The power of the nomogram was tested through calibration plots. Additionally, the nomogram's discriminating capacity was measured using the concordance index (C-index). R package RMS (version 5.1–4) was utilized to generate the nomogram and calibration plots.

Enrichment Analysis

The ClusterProfiler tool in R (3.6.3) was employed to conduct functional enrichment analysis of gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), along with gene set enrichment analysis (GSEA). The cellular components (CC), molecular functions (MF), and biological processes (BP) are all included in the GO analysis. GSEA, a computer approach, may be applied to test a priori-determined collection of genes for statistical significance as well as the concordance between two distinct biological states ^[40]. To additionally categorize the enriched pathways for each phenotype in both hallmark pathways and KEGG pathways, we employed the adjusted p-value and normalized enrichment score (NES). Significantly enriched gene sets were designated as those having a false discovery rate (FDR) of < 0.25 and an adjusted p-value of < 0.05.

Analysis of Immune Infiltration

The ESTIMATE^[41] (Estimation of Stromal and Immune cells in Malignant Tumours using Expression data) algorithm was employed to calculate the stromal and immune scores of GC. We examined the association of gene expression with the infiltration status of immune cells, like dendritic cells, macrophages, CD4 + T cells, CD8 + T cells, neutrophils, and B cells, using data from the Tumor Immune Estimation Resource (TIMER) database^{[42, 43]}. Spearman's correlation analysis was executed to probe the link between gene expression and the above immune cells. The Wilcoxon rank-sum test was adopted to make a comparison of the infiltration status between the different groups.

Prediction of Drug Sensitivity

The CellMiner^{[44, 45]} is a platform that is mainly focused on the sixty different kinds of cancerous cells that are identified by the Center for Cancer Research at the National Cancer Institute (NCI). The process of data retrieval includes visiting the site of the CellMiner database, clicking on the Download Data Sets, entering the interface for data download, and choosing RNA expression data (RNA: RNA-seq) and medication data (Compound activity: DTP NCI-60); Currently, the NCI-60 cell line is the most frequently used one for providing a large sampling size of cancer cells for use in evaluating new anti-cancer agents. The gene expression patterns for AFAP1L1 were retrieved from the dataset, and the R program was used to compute the drug-AFAP1L1 correlation coefficients.

Immunohistochemistry Staining

A gradient concentration of ethanol was used to rehydrate sections of paraffin-fixed normal tissue and GC sections after they had been dewaxed in dimethylbenzene. Antigen retrieval by microwave heating was performed using sodium citrate solution at 95℃. Afterwards, the endogenous peroxidase was inhibited by applying 3% H2O2 for 10 minutes. A primary antibody, AFAP1L1, was added to the slices (1:200; orb624536), and the slices were incubated for a whole night at 4°C. Subsequently, the tissue slices were transferred to a light-restricted chamber and subjected to treatment with a universal anti-rabbit secondary antibody labeled with horseradish peroxidase (HRP). Following counterstaining with hematoxylin, the tissue slices were air-dried and affixed to a mount, subsequent to the detection of immunostaining with diaminobenzidine (DAB). The level of AFAP1L1 expression was evaluated by two different pathologists. Among the positive cells, the percentages were as follows:0(0%-10%), 1 (10–40%), 2 (40–70%), and 3 (> 70%), whereas the intensity of the staining was divided into 3 categories: 0 (weak), 1 (moderate), and 2(strong). By adding the proportion of positive cells to the intensity of staining, immunohistochemistry (IHC) scores for AFAP1L1 were calculated. A low expression score of 0 to 2 points, and a high score of 3 to 5 points, is the ultimate definition of AFAP1L1 expression.

Immunofluorescence Staining

A total of 10 GC tissue samples were selected for immunofluorescent staining. The sections were first dewaxed, hydrated, and the endogenous peroxide activity was inactivated. A primary antibody for AFAP1L1 (1:200; orb624536) and CD163 (1:100; ab156769) was then incubated with the processed samples for 48 hours after antigen repair. We then washed the sections in TBS and incubated them overnight at 4°C with the appropriate secondary antibodies. The results were confirmed by two pathologists after counting the percentage of positive cells.

Statistical analysis

The R 3.6.3 software was employed to analyze statistical data derived from the TCGA. Wilcoxon rank-sum test and Wilcoxon signed-rank test were adopted to compare AFAP1L1 expression patterns between the tumor and normal samples. Welch one-way ANOVA with Bonferroni correction or t-test was utilized to examine the association of AFAP1L1 expression with clinicopathological characteristics. Pearson's chi-square test or Yates' correction was employed to investigate how clinicopathological variables influenced AFAP1L1 expression. The link between clinicopathological parameters and GC was subsequently evaluated via logistic regression analysis. The progression-free interval (PFI), disease-specific survival (DSS), and overall survival (OS) for AFAP1L1 were explored utilizing the KM curve. The predictive effectiveness of AFAP1L1 expression and other clinicopathological features on OS was also assessed via univariate and multivariate Cox regression analyses. In the multivariate analysis, we incorporated all of the factors that had statistical significance in the univariate model. With the use of the pROC program, we conducted a receiver operating characteristic (ROC) study of AFAP1L1. Area under the curve (AUC) values ranging between 0.5 and 1.0 were determined to have a discriminatory power of between 50% and 100%. A two-tailed p ≤ 0.05 indicated a significance level.

Ethics approval

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of Peking University People’s Hospital.

Author Contributions Statement

All authors contributed to the study conception and design. L. G., C. Y. and L. Z. performed material preparation, data collection and analysis. L.G. written the first draft of the manuscript and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Competing Interests

The authors declare no competing interests.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Consent to publish

The authors affirm that human research participants provided informed consent for publication of the images in Figure(s) 1I and 8I.

Funding

This research was funded by Beijing Xisike Clinical Oncology Research Foundation, grant number Y-xsk2021-0004.

Data Availability

Declaration of Data Availability StatementThe data in this study mainly come from public databases, including:Xiantao academic online analysis tool: https://www.xiantaozi.com/ (accessed on October 2022).RNA-Seq data were deposited into the Gene Expression Omnibus database (accessed on October 2022) and The Cancer Genome Atlas database (accessed on October 2022) under accession number GSE19826, GSE54129, GSE62254, GSE66222, GSE66229 and are available at the following URL:https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE19826.https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE54129.https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE62254.https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE66222.https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE66229.https://portal.gdc.cancer.gov/.Example from:https://link.springer.com/article/10.1007/s12032-010-9766-y.https://pubmed.ncbi.nlm.nih.gov/11782383/.https://www.nature.com/articles/nm.3850.https://www.nature.com/articles/s41467-018-04179-8.The Tumor Immune Estimation Resource (TIMER): http://cistrome.shinyapps.io/timer (accessed on October 2022).CellMiner: https://discover.nci.nih.gov/cellminer/home.do (accessed on October 2022).The data of cases from our hospital cannot be disclosed temporarily.

Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71, 209–249, doi:10.3322/caac.21660 (2021).
Ooki, A. & Yamaguchi, K. The beginning of the era of precision medicine for gastric cancer with fibroblast growth factor receptor 2 aberration. Gastric Cancer 24, 1169–1183, doi:10.1007/s10120-021-01235-z (2021).
Snyder, B. N. et al. AFAP1L1 is a novel adaptor protein of the AFAP family that interacts with cortactin and localizes to invadosomes. Eur J Cell Biol 90, 376–389, doi:10.1016/j.ejcb.2010.11.016 (2011).
Furu, M. et al. Identification of AFAP1L1 as a prognostic marker for spindle cell sarcomas. Oncogene 30, 4015–4025, doi:10.1038/onc.2011.108 (2011).
Kajita, Y. et al. The transcription factor Sp3 regulates the expression of a metastasis-related marker of sarcoma, actin filament-associated protein 1-like 1 (AFAP1L1). PLoS One 8, e49709, doi:10.1371/journal.pone.0049709 (2013).
Tie, S. R. et al. Regulation of sarcoma cell migration, invasion and invadopodia formation by AFAP1L1 through a phosphotyrosine-dependent pathway. Oncogene 35, 2098–2111, doi:10.1038/onc.2015.272 (2016).
Takahashi, R. et al. AFAP1L1, a novel associating partner with vinculin, modulates cellular morphology and motility, and promotes the progression of colorectal cancers. Cancer Med 3, 759–774, doi:10.1002/cam4.237 (2014).
Wang, M. et al. Actin Filament-Associated Protein 1-Like 1 Mediates Proliferation and Survival in Non-Small Cell Lung Cancer Cells. Med Sci Monit 24, 215–224, doi:10.12659/msm.905900 (2018).
Mantovani, A., Allavena, P., Marchesi, F. & Garlanda, C. Macrophages as tools and targets in cancer therapy. Nat Rev Drug Discov, 1–22, doi:10.1038/s41573-022-00520-5 (2022).
Ru, B. et al. TISIDB: an integrated repository portal for tumor-immune system interactions. Bioinformatics 35, 4200–4202, doi:10.1093/bioinformatics/btz210 (2019).
Ferlay, J. et al. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 127, 2893–2917, doi:10.1002/ijc.25516 (2010).
Seeneevassen, L. et al. Gastric Cancer: Advances in Carcinogenesis Research and New Therapeutic Strategies. Int J Mol Sci 22, doi:10.3390/ijms22073418 (2021).
Sun, B., Ding, B., Chen, Y., Peng, C. & Chen, X. AFAP1L1 promotes gastric cancer progression by interacting with VAV2 to facilitate CDC42-mediated activation of ITGA5 signaling pathway. J Transl Med 21, 18, doi:10.1186/s12967-023-03871-8 (2023).
Mukherjee, S. et al. Cross-talk between mutant p53 and p62/SQSTM1 augments cancer cell migration by promoting the degradation of cell adhesion proteins. Proc Natl Acad Sci U S A 119, e2119644119, doi:10.1073/pnas.2119644119 (2022).
Roskoski, R., Jr. Properties of FDA-approved small molecule protein kinase inhibitors: A 2022 update. Pharmacol Res 175, 106037, doi:10.1016/j.phrs.2021.106037 (2022).
Lawler, J. Counter regulation of tumor angiogenesis by vascular endothelial growth factor and thrombospondin-1. Semin Cancer Biol 86, 126–135, doi:10.1016/j.semcancer.2022.09.006 (2022).
Higuchi, K. et al. Cadherin 5 expression correlates with poor survival in human gastric cancer. J Clin Pathol 70, 217–221, doi:10.1136/jclinpath-2016-203640 (2017).
Mao, X. G. et al. CDH5 is specifically activated in glioblastoma stemlike cells and contributes to vasculogenic mimicry induced by hypoxia. Neuro Oncol 15, 865–879, doi:10.1093/neuonc/not029 (2013).
Fry, S. A., Robertson, C. E., Swann, R. & Dwek, M. V. Cadherin-5: a biomarker for metastatic breast cancer with optimum efficacy in oestrogen receptor-positive breast cancers with vascular invasion. Br J Cancer 114, 1019–1026, doi:10.1038/bjc.2016.66 (2016).
Chen, Y., Zheng, X. & Wu, C. The Role of the Tumor Microenvironment and Treatment Strategies in Colorectal Cancer. Front Immunol 12, 792691, doi:10.3389/fimmu.2021.792691 (2021).
Li, Y. et al. Identification of a novel immune prognostic model in gastric cancer. Clin Transl Oncol 23, 846–855, doi:10.1007/s12094-020-02478-5 (2021).
Giraldo, N. A. et al. The clinical role of the TME in solid cancer. Br J Cancer 120, 45–53, doi:10.1038/s41416-018-0327-z (2019).
Liu, Y. & Cao, X. The origin and function of tumor-associated macrophages. Cell Mol Immunol 12, 1–4, doi:10.1038/cmi.2014.83 (2015).
Kim, J. & Bae, J. S. Tumor-Associated Macrophages and Neutrophils in Tumor Microenvironment. Mediators Inflamm 2016, 6058147, doi:10.1155/2016/6058147 (2016).
Italiani, P. & Boraschi, D. From Monocytes to M1/M2 Macrophages: Phenotypical vs. Functional Differentiation. Front Immunol 5, 514, doi:10.3389/fimmu.2014.00514 (2014).
Wang, H., Tian, T. & Zhang, J. Tumor-Associated Macrophages (TAMs) in Colorectal Cancer (CRC): From Mechanism to Therapy and Prognosis. Int J Mol Sci 22, doi:10.3390/ijms22168470 (2021).
Bocchi, M., de Sousa Pereira, N., de Oliveira, K. B. & Amarante, M. K. Involvement of CXCL12/CXCR4 axis in colorectal cancer: a mini-review. Mol Biol Rep, doi:10.1007/s11033-023-08479-1 (2023).
Shi, T. et al. Increased SPRY1 expression activates NF-κB signaling and promotes pancreatic cancer progression by recruiting neutrophils and macrophages through CXCL12-CXCR4 axis. Cell Oncol (Dordr), doi:10.1007/s13402-023-00791-z (2023).
Wu, J. et al. CXCL12 derived from CD248-expressing cancer-associated fibroblasts mediates M2-polarized macrophages to promote nonsmall cell lung cancer progression. Biochim Biophys Acta Mol Basis Dis 1868, 166521, doi:10.1016/j.bbadis.2022.166521 (2022).
Liang, R. et al. Dihydroartemisinin inhibits the tumorigenesis and invasion of gastric cancer by regulating STAT1/KDR/MMP9 and P53/BCL2L1/CASP3/7 pathways. Pathol Res Pract 218, 153318, doi:10.1016/j.prp.2020.153318 (2021).
Zhuo, Y. J., Shi, Y. & Wu, T. NRP-1 and KDR polymorphisms are associated with survival time in patients with advanced gastric cancer. Oncol Lett 18, 4629–4638, doi:10.3892/ol.2019.10842 (2019).
Noh, S. H. et al. Adjuvant capecitabine plus oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): 5-year follow-up of an open-label, randomised phase 3 trial. Lancet Oncol 15, 1389–1396, doi:10.1016/s1470-2045(14)70473-5 (2014).
Ajani, J. A. et al. Gastric Cancer, Version 3.2016, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 14, 1286–1312, doi:10.6004/jnccn.2016.0137 (2016).
Joshi, S. S. & Badgwell, B. D. Current treatment and recent progress in gastric cancer. CA Cancer J Clin 71, 264–279, doi:10.3322/caac.21657 (2021).
Sorf, A., Vagiannis, D., Ahmed, F., Hofman, J. & Ceckova, M. Dabrafenib inhibits ABCG2 and cytochrome P450 isoenzymes; potential implications for combination anticancer therapy. Toxicol Appl Pharmacol 434, 115797, doi:10.1016/j.taap.2021.115797 (2022).
Kang, X. et al. LIMK1 promotes peritoneal metastasis of gastric cancer and is a therapeutic target. Oncogene 40, 3422–3433, doi:10.1038/s41388-021-01656-1 (2021).
Wang, Q. et al. Upregulated INHBA expression is associated with poor survival in gastric cancer. Med Oncol 29, 77–83, doi:10.1007/s12032-010-9766-y (2012).
Cristescu, R. et al. Molecular analysis of gastric cancer identifies subtypes associated with distinct clinical outcomes. Nat Med 21, 449–456, doi:10.1038/nm.3850 (2015).
Oh, S. C. et al. Clinical and genomic landscape of gastric cancer with a mesenchymal phenotype. Nat Commun 9, 1777, doi:10.1038/s41467-018-04179-8 (2018).
Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 102, 15545–15550, doi:10.1073/pnas.0506580102 (2005).
Yoshihara, K. et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 4, 2612, doi:10.1038/ncomms3612 (2013).
Li, T. et al. TIMER: A Web Server for Comprehensive Analysis of Tumor-Infiltrating Immune Cells. Cancer Res 77, e108-e110, doi:10.1158/0008-5472.Can-17-0307 (2017).
Li, T. et al. TIMER2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res 48, W509-w514, doi:10.1093/nar/gkaa407 (2020).
Shankavaram, U. T. et al. CellMiner: a relational database and query tool for the NCI-60 cancer cell lines. BMC Genomics 10, 277, doi:10.1186/1471-2164-10-277 (2009).
Reinhold, W. C. et al. CellMiner: a web-based suite of genomic and pharmacologic tools to explore transcript and drug patterns in the NCI-60 cell line set. Cancer Res 72, 3499–3511, doi:10.1158/0008-5472.Can-12-1370 (2012).

No competing interests reported.

Download PDF

Reviewers invited by journal
24 Oct, 2024
Editor assigned by journal
22 Oct, 2024
Editor invited by journal
12 Aug, 2024
Submission checks completed at journal
12 Aug, 2024
First submitted to journal
29 Jul, 2024

You are reading this latest preprint version

AFAP1L1 is indicative for a grim prognosis and immune microenvironment in gastric cancer

Status:

Version 1

Abstract

Figures

Introduction

Results

The expression of AFAP1L1 in GC

Association of AFAP1L1 expression with clinical features of GC patients

Prognostic value of AFAP1L1 in GC

Functional inference of AFAP1L1 in GC

AFAP1L1 expression correlates with infiltration levels of macrophages

Sensitivity analysis of drugs

Discussion

Materials and methods

Collection and Processing of Data

Patients and Clinical Specimens

Survival Analysis

Enrichment Analysis

Analysis of Immune Infiltration

Prediction of Drug Sensitivity

Immunohistochemistry Staining

Immunofluorescence Staining

Statistical analysis

Declarations

Ethics approval

Author Contributions Statement

Competing Interests

Funding

Data Availability

References

Additional Declarations

Status:

Version 1