Cohort | Treatment | Patients | R | NR | PMID |
Gide et al. [42] | anti-PD-1 | 41 | 19 | 22 | 30753825 |
Gide et al. [42] | anti-CTLA-4 + PD-1 | 51 | 35 | 16 | 30753825 |
GSE91061 | anti-CTLA-4 + PD-1 | 105 | 23 | 82 | 29033130 |
In recent years, significant progress has been made in video-based person re-identification (Re-ID). The key challenge in video person Re-ID lies in effectively constructing discriminative and robust person feature representations. Methods based on local regions utilize spatial and temporal attention to extract representative local features. However, prior approaches often overlook the correlations between local regions. To leverage relationships among different local regions, we have proposed a novel video person Re-ID representation learning approach based on a graph transformer, which facilitates contextual interactions between relevant region features. Specifically, we construct a local relation graph to model intrinsic relationships between nodes representing local regions. This graph employs the architecture of a transformer for feature propagation, iteratively refining region features and considering information from adjacent nodes to obtain partial feature representations. To learn compact and discriminative representations, we have further proposed a global feature learning branch based on a vision transformer to capture the relationships between different frames in a sequence. Additionally, we designed a dual-branch interaction network based on multi-head fusion attention to integrate frame-level features extracted by both local and global branches. Finally, the concatenated global and local features, after interaction, are used for testing. We evaluated the proposed method on three datasets, namely iLIDS-VID, MARS, and DukeMTMC-VideoReID. Experimental results demonstrate competitive performance, validating the effectiveness of our proposed approach.
Citation: Hai Lu, Enbo Luo, Yong Feng, Yifan Wang. Video-based person re-identification with complementary local and global features using a graph transformer[J]. Mathematical Biosciences and Engineering, 2024, 21(7): 6694-6709. doi: 10.3934/mbe.2024293
[1] | Tongmeng Jiang, Pan Jin, Guoxiu Huang, Shi-Cheng Li . The function of guanylate binding protein 3 (GBP3) in human cancers by pan-cancer bioinformatics. Mathematical Biosciences and Engineering, 2023, 20(5): 9511-9529. doi: 10.3934/mbe.2023418 |
[2] | Linqian Guo, Qingrong Meng, Wenqi Lin, Kaiyuan Weng . Identification of immune subtypes of melanoma based on single-cell and bulk RNA sequencing data. Mathematical Biosciences and Engineering, 2023, 20(2): 2920-2936. doi: 10.3934/mbe.2023138 |
[3] | Jing Liu, Xuefang Zhang, Ting Ye, Yongjian Dong, Wenfeng Zhang, Fenglin Wu, Huaben Bo, Hongwei Shao, Rongxin Zhang, Han Shen . Prognostic modeling of patients with metastatic melanoma based on tumor immune microenvironment characteristics. Mathematical Biosciences and Engineering, 2022, 19(2): 1448-1470. doi: 10.3934/mbe.2022067 |
[4] | Zehao Niu, Yujian Xu, Yan Li, Youbai Chen, Yan Han . Construction and validation of a novel pyroptosis-related signature to predict prognosis in patients with cutaneous melanoma. Mathematical Biosciences and Engineering, 2022, 19(1): 688-706. doi: 10.3934/mbe.2022031 |
[5] | Xiaowei Zhang, Jiayu Tan, Xinyu Zhang, Kritika Pandey, Yuqing Zhong, Guitao Wu, Kejun He . Aggrephagy-related gene signature correlates with survival and tumor-associated macrophages in glioma: Insights from single-cell and bulk RNA sequencing. Mathematical Biosciences and Engineering, 2024, 21(2): 2407-2431. doi: 10.3934/mbe.2024106 |
[6] | Jiaping Wang . Prognostic score model-based signature genes for predicting the prognosis of metastatic skin cutaneous melanoma. Mathematical Biosciences and Engineering, 2021, 18(5): 5125-5145. doi: 10.3934/mbe.2021261 |
[7] | Ernesto A. B. F. Lima, Patrick N. Song, Kirsten Reeves, Benjamin Larimer, Anna G. Sorace, Thomas E. Yankeelov . Predicting response to combination evofosfamide and immunotherapy under hypoxic conditions in murine models of colon cancer. Mathematical Biosciences and Engineering, 2023, 20(10): 17625-17645. doi: 10.3934/mbe.2023783 |
[8] | Bin Ma, Lianqun Cao, Yongmin Li . A novel 10-gene immune-related lncRNA signature model for the prognosis of colorectal cancer. Mathematical Biosciences and Engineering, 2021, 18(6): 9743-9760. doi: 10.3934/mbe.2021477 |
[9] | Chunpei Ou, Qin Peng, Changchun Zeng . An integrative prognostic and immune analysis of PTPRD in cancer. Mathematical Biosciences and Engineering, 2022, 19(6): 5361-5379. doi: 10.3934/mbe.2022251 |
[10] | Yong Luo, Xiaopeng Liu, Jingbo Lin, Weide Zhong, Qingbiao Chen . Development and validation of novel inflammatory response-related gene signature to predict prostate cancer recurrence and response to immune checkpoint therapy. Mathematical Biosciences and Engineering, 2022, 19(11): 11345-11366. doi: 10.3934/mbe.2022528 |
In recent years, significant progress has been made in video-based person re-identification (Re-ID). The key challenge in video person Re-ID lies in effectively constructing discriminative and robust person feature representations. Methods based on local regions utilize spatial and temporal attention to extract representative local features. However, prior approaches often overlook the correlations between local regions. To leverage relationships among different local regions, we have proposed a novel video person Re-ID representation learning approach based on a graph transformer, which facilitates contextual interactions between relevant region features. Specifically, we construct a local relation graph to model intrinsic relationships between nodes representing local regions. This graph employs the architecture of a transformer for feature propagation, iteratively refining region features and considering information from adjacent nodes to obtain partial feature representations. To learn compact and discriminative representations, we have further proposed a global feature learning branch based on a vision transformer to capture the relationships between different frames in a sequence. Additionally, we designed a dual-branch interaction network based on multi-head fusion attention to integrate frame-level features extracted by both local and global branches. Finally, the concatenated global and local features, after interaction, are used for testing. We evaluated the proposed method on three datasets, namely iLIDS-VID, MARS, and DukeMTMC-VideoReID. Experimental results demonstrate competitive performance, validating the effectiveness of our proposed approach.
Skin cutaneous melanoma (SKCM) is one of the most aggressive skin cancers, which has become a significant public health problem [1,2]. Its highly metastatic and invasive nature has led to a poor patient prognosis [3]. Despite significant advancements in clinical treatment approaches, including targeted therapies and immunotherapy, SKCM remains a severe clinical challenge [4]. Therefore, understanding the molecular mechanisms of SKCM is crucial for the effective prevention and improvement of patient prognosis [5].
Inflammation is a vital physiological response of the immune system, playing an essential role in both immune regulation and cancer [6]. The inflammasome, which is a multiprotein complex, serves as a critical mediator in the inflammatory process [7]. It is involved in various biological processes, including immune responses and cancer development [8]. Aberrant inflammasome activation can contribute to tumor development, progression, and metastasis [9,10]. NLRP3, CASP1, GSDMD, IL1B, and IL18 are interconnected components of inflammasome signaling, working together to regulate the inflammatory response [11]. Upon activation, NLRP3 recruits CASP1 and cleaves GSDMD, releasing IL1B and IL18 [12,13]. In turn, IL1B can promote the expression of NLRP3 and CASP1, thereby establishing a positive feedback loop in the inflammatory signaling cascade [14,15,16]. This intricate interplay between inflammation features highlights their coordinated involvement in mediating inflammation and suggests their potential as therapeutic targets for cancer [8]. Despite the recognized importance of inflammasome in cancer, there is a lack of comprehensive characterization of the inflammasome complex in SKCM.
Immune checkpoint blockade (ICB) therapy has become a breakthrough in cancer treatments, which involves targeting key regulatory checkpoints, including PD-1/PD-L1 and CTLA-4, to restore and enhance anti-tumor immune responses [17,18]. ICB therapy has demonstrated remarkable efficacy across multiple malignancies, including SKCM [19,20]. ICB therapy has revolutionized treatment options for SKCM patients, leading to durable clinical benefits and an improved overall survival in patients [21]. Despite the remarkable achievements of ICB therapy, the emergence of immune resistance poses a significant challenge to its effectiveness [22,23,24]. Therefore, there is a pressing need to elucidate the mechanisms underlying ICB therapy resistance. The inflammasome complex represents an intriguing target for immune checkpoint therapy [25,26]. Modulating the inflammasome pathway has the potential to augment antitumor immune responses and enhance the effectiveness of immune checkpoint inhibitors [27]. Exploring the interplay between inflammasome activation and ICB in melanoma could provide valuable insights into novel therapeutic strategies and improve patient outcomes.
In this study, we quantified the expression level of six inflammasome features, including CASP1, GSDMD, NLRP3, IL1B, IL18, and inflammasome complexes (IC) features, based on transcriptomic data, to comprehensively characterize the role of inflammasome features in tumorigenesis and progression. These inflammasome features exhibited a strong correlation with clinical characteristics. They were all dysregulated and served oncogenic roles in SKCM samples. The unsupervised clustering method identified four immune clusters and two SKCM subtypes, and significant overlaps were found between immune clusters and SKCM subtypes. Lastly, inflammasome features were implicated in the ICB immunotherapy response. In conclusion, our study highlighted the significant role of inflammasomes in SKCM, shedding light on their potential implications for SKCM progression, therapeutic strategies, and survival outcomes of patients. Furthermore, investigating the interplay between inflammasomes and ICB resistance is crucial for unraveling the underlying mechanisms in SKCM treatment.
Six gene sets representing distinct inflammasome features were collected from previous studies: 1) 30 NLRP3-related genes [28]; 2) 15 IC-related genes [14]; 3) 34 CASP1-related genes [29] 4) 72 IL1B-related genes [14]; 5) 8 IL18-related genes [30]; and 6) 13 GSDMD-related genes [30]. Each gene set represents different steps of inflammasome signaling.
Fragments Per Kilobase of transcript per Million mapped reads upper quartile (FPKM-UQ) of RNA-seq data from the Cancer Genome Atlas data portal (TCGA) of SKCM patients and non-cancerous tissues from the Genotype-Tissue Expression (GTEx) consortium [31] were downloaded from the University of California, Santa Cruz (UCSC) Xena platform [32] (https://xenabrowser.net/). Approximately 19,724 protein-coding genes were obtained. Expression profile was log-transformed for analysis. Clinical patient information, including age, gender, TNM staging, Clark level, and cancer type (metastatic/primary), were obtained from UCSC Xena.
A single-sample gene-set enrichment analysis (ssGSEA) was used to quantify the expression level of each inflammasome feature across various samples. We applied the R "limma" v3.54.2 package [33] to identify dysregulated inflammasome features and protein-coding genes. Features with a Benjamini-Hochberg (BH) adjusted p-value < 0.01 were defined as dysregulated; genes with a BH adjusted p-value < 0.01 and fold change > 2 were defined as dysregulated.
The tumor purity, stromal and immune cell abundances describing the TIME of SKCM were inferred by the Estimation of Stromal and Immune Cells in Malignant Tumours using Expression data (ESTIMATE) [34]. Four TIME classifications defined by 29 transcriptomic-based gene expression signatures, including angiogenesis fibrosis, anti-tumor microenvironment, malignant cell properties, and pro-tumor microenvironment, were obtained from a previous study [35]. A ssGSEA was used to quantify the enrichment level of each TIME signature. Distinct immune subtypes of SKCM patients were identified by an unsupervised hierarchical clustering method.
The R 'Consensus ClusterPlus' v1.62.0 package was applied to perform unsupervised consensus clustering on SKCM samples [36]. The clustering procedure using the inflammasome features matrix was repeated 5,000 times, thereby subsampling 80% of the samples. The optimal number of clusters was determined by the relative change in the area under the cumulative distribution function (CDF) curves and the Calinski criterion.
Patients' overall survival (OS) and progression-free survival (PFS) information were downloaded for the USCS Xena platform. The R "survival" v3.5.5 and "survminer" v0.4.9 packages were applied to conduct uni and multi-variate Cox proportional hazards regression models. Inflammasome features with a Cox p-value < 0.05 were defined as significant. A hazard ratio (HR) < 1 was defined as a protective feature, while an HR > 1 was defined as a risky feature. Then, Kaplan-Meier curves were utilized to compare the survival status of different groups. A log-rank p-value < 0.05 was identified as a prognosis-related feature. A nomogram was constructed by the R "rms" v6.6.0 package and evaluated by a calibration plot.
Gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, and Reactome gene sets enrichment were conducted by the Metascape platform (https://metascape.org/) [37]. Ten oncogenic signaling pathways were obtained from a previous study [38]. Cancer hallmark gene sets were obtained from a previous study [39]. GO Biological Process (BP) and functional pathway gene sets were downloaded from MSigDB [40,41] (https://www.gsea-msigdb.org/gsea/msigdb). The R "fgsea" v1.24.0 package was applied to conduct a Gene Set Enrichment Analysis (GSEA).
Three cohorts of SKCM patients who received ICB therapy were collected from previous studies [42,43] (Table 1). Patients who either partially or completely responded (P/CR) to ICB therapy were considered as responders, while patients with either a stable disease (SD) or progressive disease (PD) status were considered as non-responders. The R "pROC" v1.18.0 package was used to calculate the area under the curve (AUCs) of the receiver operating characteristic (ROC) curve for discriminating responders vs non-responders. Then, we used the R "e1071" v1.7.13 package to train the support vector machines (SVM) classifier based on the Gide cohort to predict the response of TCGA SKCM patients.
Cohort | Treatment | Patients | R | NR | PMID |
Gide et al. [42] | anti-PD-1 | 41 | 19 | 22 | 30753825 |
Gide et al. [42] | anti-CTLA-4 + PD-1 | 51 | 35 | 16 | 30753825 |
GSE91061 | anti-CTLA-4 + PD-1 | 105 | 23 | 82 | 29033130 |
A Mann-Whitney U test was used to compare the ssGSEA level in different groups. The Kruskal-Wallis test was used to compare ssGSEA levels among multiple groups. Spearman's correlation test was used to perform a correlation analysis. A P-value < 0.05 was considered as statistically significant. The statistical analysis in this study was performed by the R 4.2.3 software.
The overall workflow of this study is displayed in Figure S1. To quantify the expression level of six inflammasome features, ssGSEA was performed on TCGA and GTEx samples. A differential expression analysis showed that CASP1, GSDMD, and NLRP3 features were upregulated in cancer samples, while IL18, IL1B, and IC features were downregulated in cancer samples (Figure 1(A), Figure S2(A)). Next, we investigated the associations between the clinical characteristics and the inflammasome features. Results showed that the expression level of the CASP1, GSDMD, NLRP3, and IC features decreased with the progression of the T stage (Figure 1(B) and (C), Figure S2(B) and (C)).
These four features were upregulated in the metastatic samples compared to the primary tumor samples. The CASP1, GSDMD, NLRP3, and IL1B features demonstrated significant differences across pathological stages. Regarding the Clark level, we found that the GSDMD, NLRP3, and IC features first increased and then decreased with the progression of the Clark level. As for any gender differences, the GSDMD, IL1B, and IC features exhibited higher expression levels in female samples compared to male samples. The only age-related difference was found in the CASP1 feature, whose expression level decreased with age. No significant differences were found in the M and N stages. Patients with different survival statuses exhibited significant differences in the levels of inflammasome features, whether OS or PFS. In conclusion, the expression levels of the inflammasome features were significantly associated with various clinical characteristics of SKCM.
To explore the associations between patient prognosis and the expression level of inflammasome features, we performed a survival analysis on TCGA SKCM patients. As shown in the K-M curves, patients with high expression of the CASP1, GSDMD, NLRP3, IL1B, and IC features exhibited improved prognoses (Figure 2(A)). The univariate Cox regression model results showed that these five features could serve as favorable factors for the patients' OS (Figure S3). Only the IL1B and IC features were associated with the patients' PFS. Next, we performed a multivariate Cox regression model to identify independent prognostic factors. Among the six inflammasome features, only the CASP1 and IC features remained significantly associated with overall survival (Figure 2(B)). Therefore, we constructed a nomogram based on these two independent prognostic factors and the age of patients. The nomogram was used to predict three- and five-year survival probabilities and the median survival time of SKCM patients (Figure 2(C)). Then, we used a calibration plot to assess the predictive accuracy of the nomogram. Results showed that only minor differences were found between the predicted and observed overall survival, suggesting the excellent predictive accuracy of the nomogram (Figure 2(D)). In summary, high levels of inflammasome features indicated a prolonged survival in patients.
Here, we calculated the overall inflammasome feature score (IM score) using ssGSEA based on all the inflammasome-related genes in the aforementioned six gene sets. All samples were divided into two subgroups according to the median value of the IM score (Figure 3(A)). K-M curves showed that patients with higher levels of an IM score presented better OS and PFS than those with lower scores (Figure 3(B)). Therefore, we sought to explore the differences between the two subgroups.
First, we quantified the enrichment level of ten oncogenic pathways and ten cancer hallmark gene sets using ssGSEA (Figure 3(C), Figure S4). As a result, the NRF2 pathway was upregulated in the high IM score subgroup, while the Hippo, MYC, and PI3K pathways were downregulated in the high IM subgroup. As for the cancer hallmark gene sets, except for Reprogramming Energy Metabolism, all hallmarks exhibited significant expression differences between the two subgroups. The Genome Instability and Mutation and Limitless Replicative Potential gene set showed a higher level in the low IM subgroup, while other hallmarks showed an opposing pattern. Next, we performed a differential expression analysis on all protein-coding genes to further investigate the differences between the two subgroups (Figure 3(A)). As a result, 643 dysregulated genes were identified (13 downregulated and 630 upregulated genes). We performed a functional enrichment analysis on upregulated and downregulated genes. As a result, downregulated genes were enriched in melanin- and pigment-related biological processes (Figure 3(E)). The upregulated genes were associated with various immune response-related biological processes (Figure 3(D)). In conclusion, our analysis revealed the oncogenic role of inflammasome features in SKCM.
Given the enrichment analysis result, we sought to investigate the relationship between the inflammasome features and TIME. A previous study collected 29 knowledge-based gene expression signatures to describe the functional and cellular components of TIME. Here, we quantified the enrichment level of 29 signatures using ssGSEA on SKCM samples. Then, we performed an unsupervised hierarchical clustering method on TIME signatures. As a result, four immune clusters were identified (Figure 4(A)).
As shown in the figure, the IM score exhibited the highest level in cluster 4 and the lowest level in cluster 2 (Figure 4(D)). Additionally, anti-and pro-tumor microenvironment-related signatures were higher in cluster 4 compared to other clusters (Figure 4(A)). Spearman's correlation analysis showed that all TIME signatures were positively correlated with the IM score, except for EMT and tumor proliferation rate-related signatures (Figure 4(B)). However, the expression levels of all TIME signatures exhibited significant differences across four immune clusters (Figure S5). As for the ESTIMATE-generated TIME patterns, the immune, stromal, and ESTIMATE scores showed positive correlations with the IM score (Figure 4C) and exhibited the highest level in cluster 4. However, tumor purity was negatively correlated with the IM score and showed the lowest level in cluster 4 (Figure 4(D)). In summary, the different clusters defined by TIME signatures were closely associated with inflammasome features.
Here, we applied the consensus clustering method on TCGA SKCM patients using inflammasome feature scores. Based on the Calinski criterion and the relative change in the area under the CDF curves, two SKCM subtypes were identified: subtype 1 included 313 patients (66.88%), and subtype 2 included 155 patients (33.12%) (Figure 5(A), (C) and (D)). First, a survival analysis revealed distinct survival outcomes between two subtypes: patients from subtype 2 had a better overall survival compared to those from subtype 1. Regarding PFS, subtype 2 patients only showed better survival outcomes within the 3000–6000-day interval. However, there was no significant difference in PFS across the entire time range (Figure 5(B)). Next, a sample overlap analysis suggested that immune cluster 2 was entirely classified with subtype 1, while immune cluster 4 was classified with subtype 2 (Figure 5(E) and (F)). The majority of immune cluster 3 was in subtype 1, whereas immune cluster 1 was almost evenly distributed between the two subtypes. Furthermore, subtype 1 demonstrated relatively lower levels of the IM score, while subtype 2 exhibited relatively higher levels of the IM score (Figure 5(E)). TCGA defined three melanoma subtypes [44]; here, we found that most keratin and MTF-low samples belonged to subtype 1, while most immune samples were in subtype 2 (Figure 5(F)). The results from the overlap analysis suggested potential correlations between the subtypes defined by the IM score and the SKCM immunity. In addition, we performed a differential expression analysis between two SKCM subtypes and identified 692 dysregulated genes (Figure 5(E)). Almost all dysregulated genes exhibited significantly higher expression in subtype 2 (8 upregulated genes and 684 downregulated genes in subtype 1). Lastly, we performed a GSEA enrichment on dysregulated genes between two subtypes. As a result, dysregulated genes were enriched in essential immune functions, including the immune response, cytokine-cytokine receptor interaction [45], the apoptotic process, and the T cell receptor signaling pathway [46] (Figure 5(G)). These pathways represent the core functions and crucial regulatory processes of the immune system. The results of the functional enrichment analysis further validated the correlations between the IM score-defined subtypes and the immune response of SKCM patients. This provided additional support for the notion that inflammasome features could potentially serve as predictive indicators of the patients' responses to ICB therapy.
Here, we collected three cohorts of SKCM patients who received ICB therapy (Table 1). First, we found higher expression levels of inflammasome features in responders among all three cohorts compared to non-responders (Figure 6(A), (H) and (I)). More specifically, in the Gide (anti-PD-1) cohort, the expression levels of the GSDMD, IL1B, IL18, and NLRP3 features were significantly correlated with the ICB response (Figure 6(A)). A similar pattern was found in the Gide (anti-CTLA-4 + PD-1) cohort (Figure 6(H)). Moreover, in the GSE91061 cohort, the expression level of the CASP1 feature showed significant differences between the two types of patients (Figure 6(I)).
Next, we performed a ROC analysis on the features with expression differences between responders and non-responders of the three cohorts. As a result, all four features showed high AUC values in the Gide (anti-PD-1) cohort, with the IL18 feature achieving the highest AUC of 0.837, suggesting its excellent predictive ability (Figure 6(B)). Additionally, the IL1B, IL18, and NLRP3 features demonstrated a strong discriminatory ability among patients in the Gide (anti-PD-1 + CTLA4) cohort and the GSE91061 cohort (Figure 6(F) and (G)). Moreover, we trained an SVM classifier using the Gide (anti-PD-1) cohort to predict the ICB therapy response of TCGA SKCM patients. The expression level of inflammasome features could effectively discriminate between the predicted responders and non-responders (Figure 6(C)). As shown in the K-M curve, the predicted responders had better overall survival than the predicted non-responders (Figure 6(D)), further suggesting the predictive ability of the inflammasome features. Lastly, among the three TCGA SKCM subtypes, the immune subtype exhibited the highest possibility of response, consistent with the predicted results (Figure 6(E)). In summary, inflammasome features could predict the response to ICB therapy in SKCM patients.
Our study provided a systematic insight into the dysregulation of the inflammasomes in aggressive SKCM. We observed elevated expression levels of inflammasome activation markers, suggesting increased inflammasome activity in SKCM. Our findings also revealed a strong correlation between inflammasome activation and the immune cell composition within the tumor microenvironment, highlighting its role in shaping the immune response in SKCM. Importantly, we found a significant association between inflammasome activation and the response to immune checkpoint blockade therapy in SKCM. Overall, our study emphasized the critical role of the inflammasomes in SKCM, from tumor development and immune microenvironment modulation to therapeutic response.
SKCM is one of the most aggressive forms of skin cancer, characterized by its high metastatic potential and resistance to conventional therapies [1,2]. The emerging link between inflammation and SKCM has gained significant attention [47], with previous studies demonstrating the involvement of inflammatory processes in the progression and metastasis of SKCM [48]. The inflammasome, which is a vital complex composed of multiple proteins, exerts its role during inflammatory immune reactions [49]. The inflammasome detects pathogens and triggers inflammatory responses. The coordinated actions in inflammasome signaling collectively contribute to regulating inflammation and the immune response mediated by the inflammasome.
In this study, we comprehensively investigated the expression patterns of the inflammasome features and their correlations with clinical characteristics in SKCM patients, revealing significant dysregulation of the inflammasome in SKCM. Specifically, we observed elevated expression levels of CASP1 and NLRP3, which are responsible for inflammasome activation, suggesting increased inflammasome activity in SKCM (Figure 1(A)). Additionally, we found that the expression level of GSDMD, which is a downstream effector of inflammasome activation, was lower than other features, indicating a potential disruption of pyroptosis in SKCM. These findings highlighted the perturbation of inflammasome components in SKCM and provided insights into the potential involvement of the inflammasome in the pathogenesis and progression of this aggressive skin cancer. In addition, we observed a decrease in the expression of IL-1β and IL-18 in the cancer samples, indicating a potential disruption of their pro-inflammatory functions in SKCM [29]. The decreased expression of IL-1β and IL-18 suggests a potential impairment in the activation of downstream inflammatory signaling pathways, which could impact immune responses and contribute to the tumor microenvironment's immune evasion mechanisms. It is worth noting that in Result 4, we found various correlations between inflammasome features and TIME signatures. Future studies should investigate the impact of TIME components disturbances in the inflammasome.
The inflammasome has been shown to play a crucial role in shaping the tumor microenvironment by modulating the immune response [29]. Activation of the inflammasome can induce programmed cell death in cancer cells and inhibit tumor growth. Dysregulation of the inflammasome has been associated with developing immune escape mechanisms, leading to tumor progression and poor clinical outcomes [25]. In this study, we employed the TIME signature derived from a previous study [35] to perform unsupervised clustering of SKCM patients. We investigated the association between immune clusters and inflammasome feature scores. The results revealed that immune cluster 4 exhibited the highest inflammasome feature scores and TIME signature levels. This finding suggested a strong correlation between the activation of the inflammasome and the immune cell composition within immune cluster 4, as indicated by the TIME signature (Figure 4(A)). These results provided valuable insights into the potential role of inflammasome activation in shaping the immune microenvironment of SKCM. They highlighted the importance of further investigating the functional implications of this association. The identification of two IM score-defined subtypes further proved the association between the inflammasome and SKCM immunity, as the results showed significant overlaps between IM subtypes and distinct SKCM immune clusters. Collectively, these results implied the potential of using inflammasome features to predict the patients' ICB therapy response.
Immunotherapy, particularly ICB, has emerged as a promising treatment approach for SKCM [50]; however, only a limited number of patients benefit from the treatment [43,51,52]. There is growing evidence suggesting a significant association between inflammasome activation and the response to ICB therapy in SKCM [53], highlighting the potential of targeting inflammasome pathways to enhance the efficacy of immunotherapy [54]. Our study revealed a close association between the inflammasome and immune function in SKCM (Figures 4 and 5). Consequently, we further explored the correlation between the inflammasome and the patients' response to ICB therapy. In the ICB cohorts, we identified significant upregulation of inflammasome expression in responders. This finding suggested that inflammasome activation may play a crucial role in enhancing the efficacy of ICB by promoting immune activation [55]. Furthermore, the increased expression of the inflammasome in responders highlights the potential of targeting inflammasome components as a promising strategy to enhance the clinical benefits of immunotherapy in SKCM patients. In conclusion, our study revealed the significant association of dysregulated inflammasome expression with tumor development, the immune microenvironment, and immune therapy response in SKCM. These findings underscored the critical role of the inflammasome in SKCM. Additionally, they lay the foundation for developing potential therapeutic strategies and prognostic markers in SKCM.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
This work was supported by the Natural Science Fund of Heilongjiang Province [LH2020H038], and Research and Innovation Fund of the First Affiliated Hospital of Harbin Medical University [2019M12].
The authors declare there is no conflict of interest.
[1] |
H. Li, K. Xu, J. Li, Z. Yu, Dual-stream reciprocal disentanglement learning for domain adaptation person re-identification, Knowl. Based Syst., 251 (2022), 109315. https://doi.org/10.1016/j.knosys.2022.109315 doi: 10.1016/j.knosys.2022.109315
![]() |
[2] |
S. Yan, Y. Zhang, M. Xie, D. Zhang, Z. Yu, Cross-domain person re-identification with pose-invariant feature decomposition and hypergraph structure alignment, Neurocomputing, 467 (2022), 229–241. https://doi.org/10.1016/j.neucom.2021.09.054 doi: 10.1016/j.neucom.2021.09.054
![]() |
[3] |
H. Li, J. Xu, Z. Yu, J. Luo, Jointly learning commonality and specificity dictionaries for person re-identification, IEEE Trans. Image Process., 29 (2020), 7345–7358. https://doi.org/10.1109/TIP.2020.3001424 doi: 10.1109/TIP.2020.3001424
![]() |
[4] | Y. Zhang, Y. Wang, H. Li, S. Li, Cross-compatible embedding and semantic consistent feature construction for sketch re-identification, in Proceedings of the 30th ACM International Conference on Multimedia (MM'22), (2022), 3347–3355. https://doi.org/10.1145/3503161.3548224 |
[5] |
H. Li, M. Liu, Z. Hu, F. Nie, Z. Yu, Intermediary-guided bidirectional spatial-temporal aggregation network for video-based visible-infrared person reidentification, IEEE Trans. Circuits Syst. Video Technol., 33 (2023), 4962–4972. https://doi.org/10.1109/TCSVT.2023.3246091 doi: 10.1109/TCSVT.2023.3246091
![]() |
[6] | A. Subramaniam, A. Nambiar, A. Mittal, Co-segmentation inspired attention networks for video-based person re-identification, in Proceedings of the IEEE/CVF International Conference on Computer Vision, IEEE, (2019), 562–572. https://doi.org/10.1109/ICCV.2019.00065 |
[7] |
D. Wu, M. Ye, G. Lin, X. Gao, J. Shen, Person re-identification by context-aware part attention and multi-head collaborative learning, IEEE Trans. Inf. Forensics Secur., 17 (2021), 115–126. https://doi.org/10.1109/TIFS.2021.3075894 doi: 10.1109/TIFS.2021.3075894
![]() |
[8] | A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, et al., An image is worth 16x16 words: Transformers for image recognition at scale, in 9th International Conference on Learning Representations (ICLR), (2021). https://doi.org/10.48550/arXiv.2010.11929 |
[9] | T. Zhang, L. Wei, L. Xie, Z. Zhuang, Y. Zhang, B. Li, et al., Spatiotemporal transformer for video-based person re-identification, preprint, arXiv: 2103.16469. |
[10] | X. Liu, P. Zhang, C. Yu, H. Lu, X. Qian, X. Yang, A video is worth three views: Trigeminal transformers for video-based person re-identification, preprint, arXiv: 2104.01745. |
[11] |
D. Cheng, Y. Gong, X. Chang, W. Shi, A. Hauptmann, N. Zheng, Deep feature learning via structured graph laplacian embedding for person re-identification, Pattern Recognit., 82 (2018), 94–104. https://doi.org/10.1016/j.patcog.2018.05.007 doi: 10.1016/j.patcog.2018.05.007
![]() |
[12] | A. Barman, S. K. Shah, Shape: A novel graph theoretic algorithm for making consensus-based decisions in person re-identification systems, in Proceedings of the IEEE International Conference on Computer Vision, IEEE, (2017), 1115–1124. https://doi.org/10.1109/ICCV.2017.127 |
[13] | Y. Shen, H. Li, S. Yi, D. Chen, X. Wang, Person re-identification with deep similarity-guided graph neural network, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2018), 486–504. https://doi.org/10.48550/arXiv.1807.099757 |
[14] | D. Chen, D. Xu, H. Li, N. Sebe, X. Wang, Group consistent similarity learning via deep crf for person re-identification, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, (2018), 8649–8658. https://doi.org/10.1109/CVPR.2018.00902 |
[15] | Y. Yan, Q. Zhang, B. Ni, W. Zhang, M. Xu, X. Yang, Learning context graph for person search, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, (2019), 2153–2162. https://doi.org/10.1109/CVPR.2019.00226 |
[16] | M. Ye, A. J. Ma, L. Zheng, J. Li, P. C. Yuen, Dynamic label graph matching for unsupervised video re-identification, in Proceedings of the IEEE International Conference on Computer Vision, IEEE, (2017), 5122–5160. https://doi.org/10.1109/ICCV.2017.550 |
[17] | L. Bao, B. Ma, H. Chang, X. Chen, Preserving structural relationships for person re-identification, in 2019 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), IEEE, (2019), 120–125. https://doi.org/10.1109/ICMEW.2019.00028 |
[18] | Z. Zhang, H. Zhang, S. Liu, Person re-identification using heterogeneous local graph attention networks, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, (2021), 12131–12140. https://doi.org/10.1109/CVPR46437.2021.01196 |
[19] | T. N. Kipf, M. Welling, Semi-supervised classification with graph convolutional networks, in International Conference on Learning Representations (ICLR), (2016). https://doi.org/10.48550/arXiv.1609.02907 |
[20] | P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Lio, Y. Bengio, Graph attention networks, preprint, arXiv: 1710.10903. |
[21] |
H. Li, S. Yan, Z. Yu, D. Tao, Attribute-identity embedding and self-supervised learning for scalable person re-identification, IEEE Transactions on Circuits and Systems for Video Technology, 30 (2020), 3472–3485. https://doi.org/10.1109/TCSVT.2019.2952550 doi: 10.1109/TCSVT.2019.2952550
![]() |
[22] |
H. Li, N. Dong, Z. Yu, D. Tao, G. Qi, Triple adversarial learning and multi-view imaginative reasoning for unsupervised domain adaptation person re-identification, IEEE Trans. Circuits Syst. Video Technol., 32 (2022), 2814–2830. https://doi.org/10.1109/TCSVT.2021.3099943 doi: 10.1109/TCSVT.2021.3099943
![]() |
[23] |
H. Li, Y. Chen, D. Tao, Z. Yu, G. Qi, Attribute-aligned domain-invariant feature learning for unsupervised domain adaptation person re-identification, IEEE Trans. Forensics Secur., 16 (2021), 1480–1494. https://doi.org/10.1109/TIFS.2020.3036800 doi: 10.1109/TIFS.2020.3036800
![]() |
[24] | K. Zhou, Y. Yang, A. Cavallaro, T. Xiang, Omni-scale feature learning for person re-identification, in Proceedings of the IEEE/CVF International Conference on Computer Vision, IEEE, (2019), 3701–3711. https://doi.org/10.1109/ICCV.2019.00380 |
[25] | F. Yu, X. Jiang, Y. Gong, S. Zhao, X. Guo, W. S. Zheng, et al., Devil's in the details: Aligning visual clues for conditional embedding in person re-identification, preprint, arXiv: 2009.05250. |
[26] | Z. Zhang, C. Lan, W. Zeng, Z. Chen, Densely semantically aligned person re-identification, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, (2019), 667–676. https://doi.org/10.1109/CVPR.2019.00076 |
[27] |
M. Ye, H. Li, B. Du, J. Shen, L. Shao, S. C. H. Hoi, Collaborative refining for person re-identification with label noise, IEEE Trans. Image Process., 31 (2021), 379–391. https://doi.org/10.1109/TIP.2021.3131937 doi: 10.1109/TIP.2021.3131937
![]() |
[28] | S. He, H. Luo, P. Wang, F. Wang, H. Li, W. Jiang, Transreid: Transformer-based object re-identification, in Proceedings of the IEEE/CVF International Conference on Computer Vision, IEEE, (2021), 14993–15002. https://doi.org/10.1109/ICCV48922.2021.01474 |
[29] | T. Mikolov, M. Karafiàt, L. Burget, J. Cernockỳ, S. Khudanpur, Recurrent neural network based language model, in 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010, (2010), 1045–1048. https://doi.org/10.21437/Interspeech.2010-343 |
[30] |
S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput., 9 (1997), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735 doi: 10.1162/neco.1997.9.8.1735
![]() |
[31] | N. McLaughlin, J. Martinez del Rincon, P. Miller, Recurrent convolutional network for video-based person re-identification, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, (2016), 1325–1334. https://doi.org/10.1109/CVPR.2016.148 |
[32] | D. Chung, K. Tahboub, E. J. Delp, A two stream siamese convolutional neural network for person re-identification, in Proceedings of the IEEE International Conference on Computer Vision, IEEE, (2017), 1992–2000. https://doi.org/10.1109/ICCV.2017.218 |
[33] | Y. Suh, J. Wang, S. Tang, T. Mei, K. M. Lee, Part-aligned bilinear representations for person re-identification, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2018), 418–437. https://doi.org/10.1007/978-3-030-01264-9_25 |
[34] | X. Gu, H. Chang, B. Ma, H. Zhang, X. Chen, Appearance-preserving 3D convolution for video-based person re-identification, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2020), 228–243. https://doi.org/10.1007/978-3-030-58536-5_14 |
[35] | J. Li, S. Zhang, T. Huang, Multi-scale 3d convolution network for video based person re-identification, in Proceedings of the AAAI Conference on Artificial Intelligence, (2019), 8618–8625. https://doi.org/10.1609/aaai.v33i01.33018618 |
[36] |
P. Zhang, J. Xu, Q. Wu, Y. Huang, X. Ben, Learning spatial-temporal representations over walking tracklet for long-term person re-identification in the wild, IEEE Trans. Multimedia, 23 (2020), 3562–3576. https://doi.org/10.1109/TMM.2020.3028461 doi: 10.1109/TMM.2020.3028461
![]() |
[37] | Y. Zhao, X. Shen, Z. Jin, H. Lu, X. S. Hua, Attribute-driven feature disentangling and temporal aggregation for video person re-identification, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, (2019), 4908–4917. https://doi.org/10.1109/CVPR.2019.00505 |
[38] | Z. Chang, Z. Yang, Y. Chen, Q. Zhou, S. Zheng, Seq-masks: Bridging the gap between appearance and gait modeling for video-based person re-identification, in 2021 International Conference on Visual Communications and Image Processing (VCIP), IEEE, (2021), 1–5. https://doi.org/10.1109/VCIP53242.2021.9675368 |
[39] |
T. Chai, Z. Chen, A. Li, J. Chen, X. Mei, Y. Wang, Video person re-identification using attribute-enhanced features, IEEE Trans. Circuits Syst. Video Technol., 32 (2022), 7951–7966. https://doi.org/10.1109/TCSVT.2022.3189027 doi: 10.1109/TCSVT.2022.3189027
![]() |
[40] |
L. Wu, Y. Wang, L. Shao, M. Wang, 3-d personvlad: Learning deep global representations for video-based person reidentification, IEEE Trans. Neural Networks Learn. Syst., 30 (2019), 3347–3359. https://doi.org/10.1109/TNNLS.2019.2891244 doi: 10.1109/TNNLS.2019.2891244
![]() |
[41] | V. Dwivedi, X. Bresson, A generalization of transformer networks to graphs, preprint, arXiv: 2012.0969. |
[42] | T. Wang, S. Gong, X. Zhu, S. Wang, Person re-identification by video ranking, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2014), 688–703. https://doi.org/10.1007/978-3-319-10593-2_45 |
[43] | E. Ristani, F. Solera, R. Zou, R. Cucchiara, C. Tomasi, Performance measures and a data set for multi-target, multi-camera tracking, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2016), 17–35. https://doi.org/10.1007/978-3-319-48881-3_2 |
[44] | L. Zheng, Z. Bie, Y. Sun, J. Wang, C. Su, S. Wang, et al., Mars: A video benchmark for large-scale person re-identification, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2016), 868–884. https://doi.org/10.1007/978-3-319-46466-4_52 |
[45] | Y. Liu, Z. Yuan, W. Zhou, H. Li, Spatial and temporal mutual promotion for video-based person re-identification, in Proceedings of the AAAI Conference on Artificial Intelligence, (2019), 8786–8793. https://doi.org/10.1609/aaai.v33i01.33018786 |
[46] |
R. Zhang, J. Li, H. Sun, Y. Ge, P. Luo, X. Wang, et al., Scan: Self-and-collaborative attention network for video person re-identification, IEEE Trans. Image Process., 28 (2019), 4870–4882. https://doi.org/10.1109/TIP.2019.2911488 doi: 10.1109/TIP.2019.2911488
![]() |
[47] | G. Chen, Y. Rao, J. Lu, J. Zhou, Temporal coherence or temporal motion: Which is more critical for video-based person re-identification?, in Proceedings of the European Conference on Computer Vision (ECCV), Springer, (2020), 660–676. https://doi.org/10.1007/978-3-030-58598-3_39 |
[48] | X. Liu, P. Zhang, C. Yu, H. Lu, X. Yang, Watching you: Global-guided reciprocal learning for video-based person re-identification, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, IEEE, (2021), 13329–13338. https://doi.org/10.1109/CVPR46437.2021.01313 |
[49] | A. Aich, M. Zheng, S. Karanam, T. Chen, A. K. Roy-Chowdhury, Z. Wu, Spatio-temporal representation factorization for video-based person re-identification, in Proceedings of the IEEE/CVF International Conference on Computer Vision, IEEE, (2021), 152–162. https://doi.org/10.1109/ICCV48922.2021.00022 |
[50] | D. Chen, H. Li, T. Xiao, S. Yi, X. Wang, Video person re-identification with competitive snippet-similarity aggregation and co-attentive snippet embedding, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, (2018), 1169–1178. https://doi.org/10.1109/CVPR.2018.00128 |
[51] | X. Liu, C. Yu, P. Zhang, H. Lu, Deeply coupled convolution–transformer with spatial–temporal complementary learning for video-based person re-identification, IEEE Trans. Neural Networks Learn.Syst., (2023), 1–11. https://doi.org/10.1109/TNNLS.2023.3271353 |
[52] |
Z. Tang, R. Zhang, Z. Peng, J. Chen, L. Lin, Multi-stage spatio-temporal aggregation transformer for video person re-identification, IEEE Trans. Multimedia, 25 (2023), 7917–7929. https://doi.org/10.1109/TMM.2022.3231103 doi: 10.1109/TMM.2022.3231103
![]() |
[53] |
X. Zang, G. Li, W. Gao, Multidirection and multiscale pyramid in transformer for video-based pedestrian retrieval, IEEE Trans. Ind. Inf., 18 (2022), 8776–8785. https://doi.org/10.1109/TII.2022.3151766 doi: 10.1109/TII.2022.3151766
![]() |
1. | Ramona Marrapodi, Barbara Bellei, The Keratinocyte in the Picture Cutaneous Melanoma Microenvironment, 2024, 16, 2072-6694, 913, 10.3390/cancers16050913 | |
2. | Zaidong Deng, Jie Liu, Yanxun V. Yu, Youngnam N. Jin, Machine learning-based identification of an immunotherapy-related signature to enhance outcomes and immunotherapy responses in melanoma, 2024, 15, 1664-3224, 10.3389/fimmu.2024.1451103 |