The extracellular matrix (ECM) is a macromolecular structure consisting of the basement membrane (BM) and intercellular substance (Padhi and Nain, 2020). The ECM constitutes a complex network structure that supports and connects tissue structures and regulates tissue formation and cellular physiological activities (Valdoz et al., 2021).

The ECM acts as an important bio-barrier for tumour local or distant metastasis. Tumour cells secrete proteolytic enzymes to degrade the ECM, thereby creating a local lysis zone for cancer cell intravasation and extravasation (Najafi et al., 2019). Generally, cancer cells with a high degree of malignancy have a strong proteolytic effect, which can erode and destroy the membrane and promote metastasis. The enzymes involved in this process are mainly serine proteases, such as plasminogen activator (PA) and metalloproteinase (MP), which include collagenase IV, matrix-degrading enzymes, and hyaluronidase (Mohan et al., 2020). Various secreted proteins, including cross-linkers, modifying enzymes, and proteases and their inhibitors, regulate ECM homoeostasis. Thus, the activation or inhibition of the aforementioned proteins promotes the proliferation and invasion of cancer cells, which suggests their potential as therapeutic targets in cancer treatment (Walker et al., 2018).

In addition to degradation, the turnover of the matrix is also significantly important for ECM homoeostasis (Huang et al., 2021). The substrate turnover of the matrix is regulated by various enzyme families. The “normal” stroma degradation followed by an increase in the stromal turnover of the tumour promotes the replacement by the tumour stroma, reinforces aggressive features, and removes physical barriers (e.g., the basement membrane), consequently favouring the malignancy and metastasis of tumours (Nissen et al., 2019). As an important ligand pool of growth factors, the degradation and turnover of the ECM could activate the binding growth factors and induce intracellular signalling responses. Moreover, the ECM not only includes enzymes such as growth factors, chemokines, and cytokines but also is a repository for inorganic molecules (Girigoswami et al., 2021). During matrix remodelling, divalent cations like calcium ions are activated to facilitate calcium transport, which further induces the activation of matrix metalloproteinases (MMPs), a disintegrin and metalloproteinases (ADAMs), and ADAMs with thrombospondin motifs (ADAMTSs), which constitute the calcium-dependent, zinc-containing thyroxine superfamily endopeptidases (Theret et al., 2021).

The ADAM family can be divided into trans-model ADAM and secretory ADAM components based on structural differences (Saha et al., 2019). The genomic distribution of ADAMs is presented in Table 1. The pre-control region ensures that the metalloprotease domain is inactive, while the catalytic site is activated by the activation of the cysteine switch mechanism. The furin recognition site (RXXR sequence) is located between the pre-control region and the metalloprotease domain, which is involved in the intracellular activation of many ADAM family members in a trans-Golgi network by a furin-like preprotein convertase. The active site of the metalloprotease domain of ADAM molecule with proteolytic activity contains a common “HEXGH” conserved sequence, whose alteration will result in the loss of proteolytic activity (Malemud, 2019). Although the ADAM family has been extensively studied, little is known about the specificity of ADAM substrates, which are hypothesised to be determined by disintegrin and cysteine domains, especially the interaction of a substrate and enzyme (Heib et al., 2020). The main function of ADAM molecules is to mediate “extracellular domain shedding,” a post-translational mechanism. ADAMs can induce proteolytic processing to release the membrane-attached proteins and activate the cleaved molecules, which are involved in growth factor signalling, cell migration, cell adhesion, and other aspects (Camodeca et al., 2019). Studies report that ADAM is abnormally upregulated and downregulated in various malignant tumour tissues, such as lung cancer, liver cancer, and colon cancer (Li et al., 2018; Jin et al., 2020; Du et al., 2022). Such abnormal expression promotes the proliferation of tumour cells and participation in tumour angiogenesis by regulating intercellular adhesion and degrading the intercellular substance.

In this study, we aimed to investigate the potential effects and mechanisms of ADAM families for pan-cancer analysis across 33 distinct tumours by integrating bulk RNA-seq, tumour mutation burden (TMB), and clinicopathological parameters in The Cancer Genome Atlas (TCGA) datasets. Moreover, we further used ESTIMATE and CIBERSORT to explore the correlation of ADAMs with immune cell infiltration and TME status. Moreover, the sensitivity of distinct FDA-approved drugs to target cancers was also examined through the CellMiner database. Furthermore, as ADAMs were significantly correlated with TME infiltration, the sensitivity of immune checkpoint inhibitor (ICI)-based immunotherapy and the association between the expression of ADAMs and the outcomes of patients treated with ICIs were explored in IMvigor210 and GSE78220 cohorts, and we found that ADAM19 was strikingly correlated with ICI immunotherapy response. Moreover, we also found that ADAMs, especially ADAM9, were strikingly correlated with the sensitivity of FDA-approved drugs. Considering the correlation of all ADAMs with drug sensitivity and immunotherapy response, we finally verified that ADAMs were significantly associated with all types of cancers and might be the novel targets for chemotherapy and immunotherapy.

Cancer is the leading cause of death globally (Shams-White et al., 2021; Soerjomataram and Bray, 2021). In 2022, approximately 4.82 million new cancer cases and 3.21 million cancer-related deaths have been reported in China (Xia et al., 2022). Since 2000, the number of cancer-related morbidity and mortality along with their incidence has been increasing annually in China (Wei et al., 2020). Accumulating evidence indicated that proteolytic enzymes, such as MMPs and closely related ADAMs and ADAMTSs, play key roles in cancer initiation and progression (Rocks et al., 2008). Wei et al. (2019) reported that the over-expression of ADAM28 in pancreatic cancer was closely correlated with the regulation of gemcitabine resistance. Similarly, Wang et al. (2011) reported that ADAM10 was significantly associated with lymph node and distant metastasis in gastric cancer. In this study, we focused on the pan-cancer analysis of all ADAMs and explored the differential expression of ADAMs in various types of cancer, as shown in Table 1; we found that both ADAM8 and ADAM12 were located at 10q26. In our results, we also found a significant positive correlation between ADAM8 and ADAM12. Similar results were also observed between ADAM21 and ADAM22. Shimura et al. (2015) revealed that urinary MMP-9/NGAL and urinary ADAM12 are potential non-invasive biomarkers for gastric cancer, including early-stage diseases. Xiao et al. (2012) reported that ADAM17 was an important contributor to prostate cancer invasion according to the shedding of the EGFR ligand TGF-α, which subsequently activates the EGFR–MEK–ERK signalling pathway and induces the over-expression of MMP-2 and MMP-9. Furthermore, Karan et al. (2003) indicated that an inverse expression pattern of ADAM17/TACE and TIMP3 and the regulation of ADAMs with DHT could play an important role in the pathogenesis of prostate cancer. Currently, the most specific therapeutic monoclonal antibody against ADAM17 is D1 (A12), which binds to both the catalytic and disintegrin/cysteine-rich domains of ADAM17 (Tape et al., 2011). It has been proven to be effective in ovarian cancer cells (Richards et al., 2012), breast cancer (Caiazza et al., 2015), and head and neck cancer (Huang et al., 2014). Additionally, ADAM inhibitors can also effectively assist the therapeutic effect of existing monoclonal antibodies. ADAM10 inhibitors have been reported to be protective against HER2-positive breast cancers, which are resistant to Herceptin (trastuzumab) (Rimawi et al., 2015).

The role of ADAMs in tumours remains limited, despite the increasing knowledge of the overall role of ADAMs. In our study, significant differential expression of ADAM8, ADAM9, ADAM10, ADAM11, ADAM12, ADAM15, ADAM17, ADAM19, ADAM22, ADAM23, and ADAM33 was observed among almost all types of cancer. ADAM8 was significantly highly expressed in tumour tissue compared with para-cancerous tissue in all types of cancer, except for KIRC, PRAD, and THCA, indicating ADAM8 could be a significant biomarker for tumours. Notably, the correlation of ADAM8 with tumour progression, metastasis, and chemoresistance in various invasive cancers, including pancreatic cancer (Yu et al., 2019), breast cancer (Conrad et al., 2018), and lung cancer (Ishikawa et al., 2004), has been previously reported. Furthermore, ADAM12 was also generally highly expressed in almost all types of tumour tissues. Wang et al. (2021) reported that ADAM12 inhibition, which was induced by a hypoxia-inducible factor, could effectively downregulate migration and invasion in breast cancer cell lines and also in immuno-deficient mice.

Recently, researchers have been increasingly focusing on the surroundings of solid cancers instead of the tumour itself (Quail and Joyce, 2013; Wu and Dai, 2017; Hinshaw and Shevde, 2019). The surroundings of the cancers are collectively known as the TME which includes immune and stromal cells, and these factors synergistically formed an inflammatory, tumour-promoting, and immuno-suppressive environment which helps cancer cells escape from immune surveillance (Wu and Dai, 2017). In this study, the ESTIMATE algorithm was used to calculate the immune score and stromal score as well as explore the correlation between these scores and ADAM expression. As shown in Figures 5A, B, ADAM8, 19, and 28 showed a significantly positive correlation with both the immune score and stromal score in pan-cancers, indicating that these genes may play an important role in TME development and could be potential immunotherapy targets. Furthermore, we assessed tumour stemness among different ADAMs in cancers using RNAss and DNAss. These results highlighted the negative effect of ADAMs on characterising cancer cells. Furthermore, as shown in Figures 6, 7, a significant positive correlation was observed between ADAM expression, including ADAM8, 9, 12, 19, 23, 28, and 33, whereas ADAM10, 11, and 22 were negatively correlated with the stromal score and immune score.

Although most studies related to ADAM9 focused more on its effect on tumour proliferation, migration, and invasion, various studies have pointed out the value of ADAM9 for tumour drug therapy. In the current study, ADAM9 was significantly associated with 82 different drug sensitivities, including tamoxifen, cyclophosphamide, oxaliplatin, and bafetinib. Liu et al. (2021) revealed that the cisplatin treatment of non-small-cell lung cancer could be promoted by ADAM9 while being negatively regulated by miR-126-5p. Slapak et al. (2021) reported a novel ADAM9-responsive, protease-dependent, drug delivery system for patients with pancreatic ductal adenocarcinoma that can reduce the cytotoxicity of systemic chemotherapy. As for the different results of ADAM9 of PCR and immunohistochemistry in BLCA, it happens occasionally. The transcription and translation of eukaryotic genes were differentially expressed; hence, it is also necessary to determine how the protein level changes.

In terms of cancer treatment, monoclonal antibodies also have their niche population. Monoclonal antibodies have a more specific mode of action and thus cause fewer cytotoxic side effects than low-molecular-weight antibodies. In addition, some monoclonal antibodies can exert anti-cancer activity by inducing antibody-dependent cytotoxicity (Mullooly et al., 2016). Furthermore, in the current study, ADAM19 was negatively correlated with immunotherapy outcomes and strongly correlated with chemotherapy drugs, thus providing a solid theoretical foundation for further research.

The expression profile of ADAMs in pan-cancers has been demonstrated to correlate with prognosis, tumour microenvironment, treatment outcome, and stemness score. Furthermore, the expression levels of ADAMs in tumour cells are also related to the efficacy of different chemotherapy-related drugs and their response to immunotherapy. These results thus provide a reference for future research on ADAM family genes as potential pan-cancer targets.