Causal associations between plasma ADAMTS-5 protein level and cancer risk: a bi-directional mendelian randomization analysis

doi:10.21203/rs.3.rs-3397054/v1

Download PDF

Research Article

Causal associations between plasma ADAMTS-5 protein level and cancer risk: a bi-directional mendelian randomization analysis

https://doi.org/10.21203/rs.3.rs-3397054/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

According to relevant research, ADAMTS-5 was associated with some cancers. However, the causal relationship between ADAMTS-5 and different types of cancers is still unclear.

Methods

The genome-wide summary statistics of plasma ADAMTS-5 protein level and 17 differential types of cancers were acquired for the deCODE database, the IEU Open GWAS project and FinnGen database. The estimated causal effect was given by the Wald ratio for each variant, the inverse-variance weighted model was used for two or more genetic instruments. Sensitivity analyses were conducted to assess the robustness of MR results. The Bonferroni corrected significance was set at P < 0.0015 (0.05/34) to account for multiple testing, and a lenient threshold P < 0.05 was considered to suggestively relationship. To strengthen our findings in MR analysis, we conducted the Bayesian co-localization analysis for validation analyses apart from using an independent cohort.

Results

After Bonferroni correction, we only detected significant evidence for genetic prediction of the causal relationship between ADAMTS-5 and oropharyngeal cancer (OR: 0.62, 95%CI: 0.47–0.81, P = 0.0007). ADAMTS-5 had suggestive associations with esophagus cancer (OR: 0.73, 95%CI: 0.55–0.98, P = 0.034). There had no statistical effect on other cancers. There was also no evidence of the reverse causal relationships. Our findings also were found in independent cohort. Furthermore, we detected the presence of a shared variant for the association between ADAMTS-5 and oropharyngeal cancer by Bayesian co-localization analysis (PP4 > 0.8), strengthening our results.

Conclusions

MR analysis reveal that plasma ADAMTS-5 level had significantly causal association with oropharyngeal cancer. It also existed suggestive associations with esophagus cancer. ADAMTS-5 was a potential drug target for oropharyngeal cancer, thus providing guidance for further clinical research.

ADAMTS-5

Cancer

Mendelian randomization

co-localization

ADAMTS-5 (A disintegrin and metalloprotease with motif 5) was first discovered in the 1990s as a functional protein encoded by the Adamts5 gene, located in human 21q21.3, encoding a multi domain of 930 amino acids. It was the enzyme with the highest activity in degrading proteoglycans[1, 2]. ADAMTS-5 was involved in the pathological processes of many diseases, with research on arthritis being the most typical. It was a key enzyme in the pathological process that functions by degrading proteoglycans[3, 4]. With the progress of research, attention had been paid to the pathophysiological regulation of ADAMTS-5 in malignant tumors [5]. Interestingly, the expression and role of ADAMTS-5 in malignant tumors were bidirectional. In human non-small cell lung cancer[6] and malignant gliomas[7], ADAMTS-5 was highly expressed and could promote tumor invasion and metastasis; However, in human hepatocellular carcinoma[8] and colorectal cancer[9], ADAMTS-5 was low expressed and played a role as a tumor suppressor gene. This suggested that ADAMTS-5 had diversity and complexity in the regulation of malignant tumors, and further research is needed.

Mendelian randomization (MR) was an alternative method for inferring potential causal relationships, using single nucleotide polymorphism (SNP) as an instrumental variable for exposure factors to estimate the causal relationship between exposure and outcomes[10–12]. Due to the random allocation of alleles for specific SNPs at birth, the biggest advantage of Mendelian randomization was that it could avoid the impact of residual confounding factors on the accuracy of association results, making the strength of association result argumentation very reliable[13].

In this study, we conducted two sample bi-directional MR and co-localization analysis to explore the causal relationship between plasma ADAMTS-5 level and risk of multiple cancers by using publicly available GWAS summary statistics, including ovarian cancer, breast cancer, lung cancer, pancreatic cancer, prostate cancer, rectal cancer, colon cancer, thyroid cancer, bladder cancer, Hodgkin lymphoma, non- Hodgkin lymphoma, esophageal cancer, gastric cancer, kidney cancer, melanoma, oral cancer and oropharyngeal cancer. Our findings would offer genetically informative evidence for studies on novel biomarkers and drug target on cancers.

Data source

The cancer-related GWAS summary level data derived from the public database comes from the IEU Open GWAS project[14]. The population selected for GWAS only includes European ancestry. The GWAS summary level data was obtained from the Pancreatic Cancer Cohort Consortium (PanScan) [15], the International Lung Cancer Consortium (ILCCO)[16], Ovarian Cancer Association Consortium (OCAC)[17], the Prostate Cancer Association Group to Investigate Cancer Associated Alterations in the Genome (PRACTICA-L) [18], the Breast Cancer Association Consortium (BCAC) [19],and the Oncoarray oral cavity and oropharyngeal cancer [20], respectively. Genetic variation and rectal malignancies, colon malignancies, thyroid malignancies, bladder cancer, Hodgkin's lymphoma, non-Hodgkin's lymphoma, esophageal malignancies, gastric cancer, kidney cancer, melanoma and oral malignancies are respectively from publicly available summary statistics of FinnGen database [21]. These 17 cancer-related detailed data was shown in Table S1. A detailed analysis methods was presented in Fig. 1.

ADAMTS-5 on risk of differential cancers

Plasma ADAMTS-5 protein level GWAS dataset was obtained from deCODE and validated by using the IEU Open GWAS project[22]. SNPs significantly related to plasma ADAMTS-5 level was selected with P < 5e-08 to serve as IVs. Next, the clumping process (r2 < 0.001 and clump distance = 10000 kb) to reduce influence from strong linkage disequilibrium (LD). Third, we remove weak instrumental variables with F < 10 by using the formula F = [(n – k – 1)/k] * [R2/ (1 – R2)]. Finally, if SNPs was not identified in outcome data, SNPs with high LD (r2 > 0.85) was selected out to serve as proxy SNPs. Supplement S1 displayed all SNPs included in the MR analysis of each exposure and pair outcome.

Differential cancers on risk of ADAMTS-5

To explore whether these differential cancers existed causal impact on plasma ADAMTS-5 level, we conducted a reverse MR analysis. SNPs with P value less than 5e-08 were selected out to serve as IVs. To conform to the 3 assumptions for MR analysis, we subsequently selected independent SNPs by standard clumping parameters (r2 < 0.001 and clump distance = 10000 kb). Accounted for having no IVs with P < 5e-08 in the exposure of kidney, Hodgkin, esophagus and oral cavity cancer, we broaden the P threshold to 5e-06 for acquiring exposure associated variants as instrumental SNPs. Finally, we removed the IVs with F > 10 and excluded the outliers through MR-PRESSO test. Supplement S2 displayed all SNPs included in the MR analysis of each exposure and pair outcome.

MR analyses

Before performing MR analysis to investigate the causal relationship between plasma ADAMTS-5 protein level and cancer, we remove intermediate allele frequency IVs and detect outliers by using MR-PRESSO to improve the accuracy of our analysis. If only one instrumental variable was available, the Wald ratio [23] method was used. When two or more genetic instruments were available, inverse variance weighted MR (ivw) was used. Furthermore, we constructed multiple testing to produce a more conservative P value by using Bonferroni correction. P < 0.0015 (0.05/34) was identified as strongly associated, P greater than 0.0015 and less than 0.05 were deemed suggestively associated.

Pleiotropy and sensitivity analyses

To evaluate the robustness of causal effect estimation, we performed sensitivity analysis, including Cochran's Q statistics, Horizontal pleiotropy test and Steiger directionality test. Cochran's Q statistics was carried out to test heterogeneity. Horizontal pleiotropy test was applied to examine horizontal heterogeneity. Steiger directionality test was used to examine whether exposure was directionally causal for outcome.

Bayesian co-localization analysis

Bayesian co-localization analysis were important in strengthening the findings in MR analysis. Thus, we performed co-localization analysis to furtherly verify the significant casual association between ADAMTS-5 and risk of pharyngeal cancer in MR analyses through IEU Open GWAS project. We used the genomic region extending 1Mb on both sides of the lead ADAMT-5 variant. Results with a posterior probability > 0.8 within the gene locus of the putative causal ADAMTS-5 and pharyngeal cancer pair were deemed as evidence for colocalization.

Statistical analysis

In our study, all statistical analyses were performed through “TwoSampleMR" (https://github.com/MRCIEU/TwoSampleMR) and MR-PRESSOO R package [24].

ADAMTS-5 on risk of differential cancers

Seven SNPs associated with plasma ADAMTS-5 level was severed as IVs for our analysis in deCODE dataset (Table S2). Figure 1 showed the result of MR analysis. It declared that ADAMTS-5 existed suggestive associations with esophagus cancer after the Bonferroni correction (OR: 0.73, 95%CI: 0.55–0.98, P = 0.034). Only oropharyngeal cancer shown statistically significance (OR: 0.62, 95%CI: 0.47–0.81, P = 0.0007). It showed that ADAMTS-5 was causally associated with oropharyngeal cancer. No statistical significance was found in the remaining cancer (Fig. 2).

In sensitivity analyses (Table S3), Cochrane Q statistics showed no significant heterogeneity (P > 0.05). We also found no evidence of horizontal pleiotropy for ADAMTS-5 in cancers with p > 0.05. MR-PRESSO analysis revealed no outliers in our results. In addition, Steiger directionality test revealed the influence of the ADAMTS-5 on risk of differential cancers was the correct causal direction.

Differential cancers on risk of ADAMTS-5

To evaluate any reverse causation association, we used differential cancers as exposure and plasma ADAMTS-5 level as outcome. Our findings revealed that any cancers had no statistically significance on risks of plasma ADAMTS-5 levels (Fig. 3, P > 0.05). Considering that breast cancer existed heterogeneity by Cochrane Q statistics (P = 0.01), so we used the random effect IVW method to correct for heterogeneity. No evidence of horizontal pleiotropy was found in differential cancers on risk of ADAMTS-5. In addition, there is no outliers through MR-PRESSO (Table S4).

Validation of ADAMTS-5 on risk of differential cancers

To improve the accuracy of results, we used an independent exposure data came from the IEU Open GWAS project to verify our findings. Only one SNP (rs2830585) was strongly correlated with exposure in validation cohort. MR analysis also found the same causal relationship (Fig. 4). Plasma ADAMTS-5 level existed suggestive associations with esophagus cancer after the Bonferroni correction (OR: 0.83, 95%CI: 0.71–0.98, P = 0.02). It also shown significant causally relationship between ADAMTS-5 and oropharyngeal cancer in validation project (OR: 0.77, 95%CI: 0.67–0.88, P = 0.000214). No statistical significance was found in the remaining cancers.

Bayesian co-localization analysis

To furtherly ensure the causality between plasma ADAMTS-5 level and oropharyngeal cancer, we perform Bayesian co-localization analysis between ADAMTS-5 and oropharyngeal cancer. Our findings found evidence (posterior probability for shared variant > 0.8) to support the presence of a shared variant for the association between ADAMTS-5 and oropharyngeal cancer in both two cohorts (Fig. 5, Table S5). That means the causal relationship between ADAMTS-5 and oropharyngeal cancer were more reliable.

In our studied, we used the publicly available GWAS summary statistics to carry out a bi-directional MR analysis to investigate the causal relationship between plasma ADAMTS-5 protein level and risks of a variety of cancers, including esophageal cancer, oral cancer, gastric cancer, thyroid cancer, breast cancer, lung cancer, pancreatic cancer, Hodgkin's lymphoma, non-Hodgkin's lymphoma, kidney cancer, bladder cancer, rectal cancer, colon cancer, melanoma, ovarian cancer and ovarian cancer. We only detected significant evidence for genetic prediction of the causal relationship between ADAMTS-5 and oropharyngeal cancer after Bonferroni correctio. ADAMTS-5 had suggestive associations with esophagus cancer. There had no statistical effect on other cancers.

Previous studies had shown that cancer cell metastasis required a mechanism mediated by microenvironment, and extracellular matrix (ECM) was necessary for cancer cell migration. Metastatic cancer cells typically had high fluidity and used various ECM degrading proteases, including matrix metalloproteinases (MMPs) to achieve invasion[25–28]. ADAM (a deintegrin and metalloproteinase) belonged to the metalloproteinase family of MMP, including membrane anchored ADAM and secretory ADAM. Secretory ADAM, also known as proteases and metalloproteinases (ADAMTS) with thrombin responsive protein motifs, had been found to degrade ECM, promoting invasion and metastasis[29, 30]. This indicated that ADAMTS-5 had potential association with cancers. It consistent with our study on relationship of plasma ADAMTS-5 level and oropharyngeal or esophagus cancer.

ADAMTSs was particularly typical in the study of osteoarthritis, as their disintegrin domain has intercellular adhesion function, allowing them to directly adhere to the extracellular matrix and participate in regulating intracellular and extracellular signaling pathways during tumor cell invasion and migration, degrading the extracellular matrix and basement membrane[31]. Although the mechanism of ADAMTS in carcinogenesis had not been fully elucidated, recent studies had shown that differential expression of ADAMTS-1, -8, -9, and − 18 family members was associated with the occurrence and development of various tumors, including cell proliferation, migration, invasion, apoptosis, and angiogenesis[32–35]. The progress in the field of arthritis provided valuable new insights into ADAMTS-5 biology from developmental biology and biochemistry perspectives, and its emerging role in pathology, such as cancer, was becoming increasingly apparent[1]. Although research on the relationship between ADAMTS-5 and tumors were still in its early stages, there was some evidence that ADAMTS5 might regulate the growth, migration, and invasion of tumor cells. ADAMTSs was a family of proteases that have at least one thrombospondin type 1 repeat (TSR). ADAMTS-5 had TSR1that inhibits angiogenesis, and TSR2, but the TSR2 domain does not have this function[36]. ADAMTS-5 inhibited tumor metastasis and angiogenesis inhibiting ets1 mediated microvessel density changes, and its low expression was associated with poor overall survival in gastric cancer patients[37]. The results of in vitro experiments on colorectal cancer showed that hypermethylation of the ADAMTS-5 promoter leads to epigenetic silencing. Overexpression of ADAMTS-5 can inhibit the migration and invasion ability of tumor cells, and did not affect the growth, cycle, and apoptosis of normal cells[9]. In liver cancer cell lines, ADAMTS-5 might inhibit angiogenesis, thereby inhibiting tumor growth[38]. Other experiments showed that the lack of ADAMTS-5 protein expression was negatively correlated with the progression and prognosis of hepatocellular carcinoma. Lentivirus-mediated ADAMTS-5 expression significantly inhibited tumor angiogenesis by downregulating in vitro expression of vascular endothelial growth factor and inhibiting migration and tube formations[8]. All above previous studies have found that the effect of ADAMTS-5 closely related to cancer type. However, our analysis had evidence to support causal relationship between plasma ADAMTS-5 and these cancers. It might be related to tissue specific factors. In addition, there were no literature reports on association between ADAMTS-5 and esophagus and oropharyngeal cancers. Thus, further research on it would be a new insight and expected to make breakthrough.

Our result was based on genetic instrumental variables, and causal inference by using MR analysis. The findings were robust and accuracy. Moreover, we conducted the Bayesian co-localization analysis for strengthening our findings apart from using an independent cohort. It was the advantage of our studies. Nevertheless, our study had some limitations. First, our study was based on European, which cannot be extended to other ethnic group and affect universality. Second, even when we performed multiple sensitivity analyses, horizontal pleiotropy cannot be fully assessed. Finally, our analysis cannot find the specific molecular biological mechanism of between ADAMTS5 and esophagus or oropharyngeal cancer, and further studies were needed to implement it.

We used a bi-directional mendelian randomization analysis and Bayesian co-localization analysis to explore the causal associations between plasma ADAMTS-5 and risk of 17 cancers. Our result supported that ADAMTS-5 had potential association with oropharyngeal cancer and esophagus cancer, though further studies would need to explore the underlying mechanism and assess the potential of ADAMTS-5 as a drug target for certain cancer.

ADAM: a deintegrin and metalloproteinase;

ADAMTSs: a disintegrin and metalloproteinase with thrombospondin motifs;

ADAMTS-5: A disintegrin and metalloprotease with motif 5;

BCAC: Breast Cancer Association Consortium;

CI: Confidence Internal;

ECM: Extracellular matrix;

GWAS: Genome-Wide Association Study;

ILCCO: International Lung Cancer Consortium;

IVW: Inverse variance weighted;

IVs: Instrumental variants;

LD: Linkage disequilibrium;

MMPs: Matrix metalloproteinases;

MR: Mendelian randomization;

MR-PRESSO: Mendelian randomization pleiotropy residual sum and outlier;

OCAC: Ovarian Cancer Association Consortium;

OR: Odds ratio;

PanScan: Pancreatic Cancer Cohort Consortium;

PRACTICA-L: Prostate Cancer Association Group to Investigate Cancer Associated Alterations in the Genome;

SNP: Single nucleotide polymorphism;

TSR: Thrombospondin type 1 repeat.

Ethics approval and consent to participate

Ethical approval and consent to participate were waived since we used only

publicly available data and materials in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Data Availability

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Further inquiries can be directed to the corresponding author.

Finding

This work was supported by the Guangxi Natural Science Foundation project

(Grant Number 2023GXNSFAA026037 to LJ) and Guangxi Medical and Health

Appropriate Technology Development and Application (Grant Number S2022062 to

LJ).

Author’s contributions

Designed analysis, LJ and RH; Analysis, RH and SL; visualization, RH and CL;

writing—original draft, TL and JQ; writing—review and editing, RH and XF. All the

authors have read and approved the final submitted version

Acknowledgements

The authors thank the participants of all GWAS cohorts included in the present work and the investigators of the IEU Open GWAS project, UK Biobank, BCAC, OCAC, MRC-IEU, PRACTICA-L, ILCCO, FINNGEN and Oncoarray oral cavity and oropharyngeal cancer consortium for sharing the GWAS summary statistics.

Author information

Authors and Affiliations

Department of Gastrointestinal Gland Surgery, The First Affiliated Hospital of

Guangxi Medical University, Nanning, The Guangxi Zhuang Autonomous Region,

530021, China.

Rongzhi Huang, Jihao Qin, Xiaowen Fang, Chenxi Liang, Siyu Li, Jiehua Li

The First People's Hospital of Qinzhou, Qinzhou, Guangxi 535000, China

Tanli Lu

Corresponding authors

Correspondence to Jiehua Li.

Kintakas, C. and D. McCulloch, Emerging roles for ADAMTS5 during development and disease. Matrix biology : journal of the International Society for Matrix Biology, 2011. 30: p. 311-7.
Durigova, M., et al., MMPs are less efficient than ADAMTS5 in cleaving aggrecan core protein. Matrix biology : journal of the International Society for Matrix Biology, 2011. 30(2): p. 145-53.
Mladenovic, Z., et al., Potential role of hyaluronic acid on bone in osteoarthritis: matrix metalloproteinases, aggrecanases, and RANKL expression are partially prevented by hyaluronic acid in interleukin 1-stimulated osteoblasts. The Journal of rheumatology, 2014. 41(5): p. 945-54.
Meng, P., et al., ADAMTS4 and ADAMTS5 may be considered as new molecular therapeutic targets for cartilage damages with Kashin-Beck Disease. Medical hypotheses, 2020. 135: p. 109440.
Cal, S. and C. López-Otín, ADAMTS proteases and cancer. Matrix biology : journal of the International Society for Matrix Biology, 2015: p. 77-85.
Gu, J., et al., Overexpression of ADAMTS5 can regulate the migration and invasion of non-small cell lung cancer. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine, 2016. 37(7): p. 8681-9.
Nakada, M., et al., Human glioblastomas overexpress ADAMTS-5 that degrades brevican. Acta neuropathologica, 2005. 110(3): p. 239-46.
Li, C., et al., Lost expression of ADAMTS5 protein associates with progression and poor prognosis of hepatocellular carcinoma. Drug design, development and therapy, 2015. 9: p. 1773-83.
Li, J., et al., Epigenetic silencing of ADAMTS5 is associated with increased invasiveness and poor survival in patients with colorectal cancer. Journal of cancer research and clinical oncology, 2018. 144(2): p. 215-227.
Boef, A., O. Dekkers, and S. le Cessie, Mendelian randomization studies: a review of the approaches used and the quality of reporting. International journal of epidemiology, 2015. 44(2): p. 496-511.
Lawlor, D., et al., Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Statistics in medicine, 2008. 27(8): p. 1133-63.
Zheng, J., et al., Recent Developments in Mendelian Randomization Studies. Current epidemiology reports, 2017. 4(4): p. 330-345.
Lee, K. and C. Lim, Mendelian Randomization Analysis in Observational Epidemiology. Journal of lipid and atherosclerosis, 2019. 8(2): p. 67-77.
Hemani, G., et al., The MR-Base platform supports systematic causal inference across the human phenome. eLife, 2018. 7.
Amundadottir, L., et al., Genome-wide association study identifies variants in the ABO locus associated with susceptibility to pancreatic cancer. Nature genetics, 2009. 41(9): p. 986-90.
Wang, Y., et al., Rare variants of large effect in BRCA2 and CHEK2 affect risk of lung cancer. Nature genetics, 2014. 46(7): p. 736-41.
Phelan, C., et al., Identification of 12 new susceptibility loci for different histotypes of epithelial ovarian cancer. Nature genetics, 2017. 49(5): p. 680-691.
Schumacher, F., et al., Association analyses of more than 140,000 men identify 63 new prostate cancer susceptibility loci. Nature genetics, 2018. 50(7): p. 928-936.
Michailidou, K., et al., Association analysis identifies 65 new breast cancer risk loci. Nature, 2017. 551(7678): p. 92-94.
Lesseur, C., et al., Genome-wide association analyses identify new susceptibility loci for oral cavity and pharyngeal cancer. Nature genetics, 2016. 48(12): p. 1544-1550.
Kurki, M., et al., FinnGen provides genetic insights from a well-phenotyped isolated population. Nature, 2023. 613(7944): p. 508-518.
Suhre, K., et al., Connecting genetic risk to disease end points through the human blood plasma proteome. Nature communications, 2017. 8: p. 14357.
Burgess, S., D. Small, and S. Thompson, A review of instrumental variable estimators for Mendelian randomization. Statistical methods in medical research, 2017. 26(5): p. 2333-2355.
Verbanck, M., et al., Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature genetics, 2018. 50(5): p. 693-698.
Parsons, J., A. Horwitz, and M. Schwartz, Cell adhesion: integrating cytoskeletal dynamics and cellular tension. Nature reviews. Molecular cell biology, 2010. 11(9): p. 633-43.
Sung, S., et al., Tumor microenvironment promotes cancer progression, metastasis, and therapeutic resistance. Current problems in cancer, 2007. 31(2): p. 36-100.
Friedl, P. and K. Wolf, Tumour-cell invasion and migration: diversity and escape mechanisms. Nature reviews. Cancer, 2003. 3(5): p. 362-74.
Hoshino, D., et al., The phosphoinositide-binding protein ZF21 regulates ECM degradation by invadopodia. PloS one, 2013. 8(1): p. e50825.
Arribas, J., J. Bech-Serra, and B. Santiago-Josefat, ADAMs, cell migration and cancer. Cancer metastasis reviews, 2006. 25(1): p. 57-68.
Shao, S., et al., ADAM-12 as a diagnostic marker for the proliferation, migration and invasion in patients with small cell lung cancer. PloS one, 2014. 9(1): p. e85936.
Wang, Q., et al., Immunomodulatory role of metalloproteases in cancers: Current progress and future trends. Frontiers in immunology, 2022. 13: p. 1064033.
Wei, J., C. Liu, and Z. Li, ADAMTS-18: a metalloproteinase with multiple functions. Frontiers in bioscience (Landmark edition), 2014. 19(8): p. 1456-67.
Tan, I.A., C. Ricciardelli, and D. Russell, The metalloproteinase ADAMTS1: a comprehensive review of its role in tumorigenic and metastatic pathways. International journal of cancer, 2013. 133(10): p. 2263-76.
Kilic, M., et al., Differentially regulated ADAMTS1, 8, and 18 in gastric adenocarcinoma. Bratislavske lekarske listy, 2017. 118(2): p. 71-76.
Kılıç, M., et al., Differentially regulated ADAMTS1, 8, 9, and 18 in pancreas adenocarcinoma. Przeglad gastroenterologiczny, 2017. 12(4): p. 262-270.
Sharghi-Namini, S., et al., The first but not the second thrombospondin type 1 repeat of ADAMTS5 functions as an angiogenesis inhibitor. Biochemical and biophysical research communications, 2008. 371(2): p. 215-9.
Huang, J., et al., ADAMTS5 acts as a tumor suppressor by inhibiting migration, invasion and angiogenesis in human gastric cancer. Gastric cancer : official journal of the International Gastric Cancer Association and the Japanese Gastric Cancer Association, 2019. 22(2): p. 287-301.
Turner, S., et al., Expression of ADAMTS-1, ADAMTS-4, ADAMTS-5 and TIMP3 by hepatocellular carcinoma cell lines. International journal of oncology, 2012. 41(3): p. 1043-9.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Causal associations between plasma ADAMTS-5 protein level and cancer risk: a bi-directional mendelian randomization analysis

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Data source

ADAMTS-5 on risk of differential cancers

Differential cancers on risk of ADAMTS-5

MR analyses

Pleiotropy and sensitivity analyses

Bayesian co-localization analysis

Statistical analysis

Result

ADAMTS-5 on risk of differential cancers

Differential cancers on risk of ADAMTS-5

Validation of ADAMTS-5 on risk of differential cancers

Bayesian co-localization analysis

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1