Tissue Factor Pathway Inhibitor 2 (TFPI2) is a Serum Biomarker for Clear Cell Renal Carcinoma

doi:10.21203/rs.3.rs-5001340/v1

Tissue factor pathway inhibitor 2 (TFPI2), a serine protease inhibitor, has emerged as a promising serum biomarker for ovarian clear cell carcinoma. We evaluated the efficacy of TFPI2 as a biomarker for renal cell carcinoma (RCC). This single-center study enrolled patients with RCC who underwent radical nephrectomy or tumor biopsy, and healthy volunteers. Preoperative serum samples were collected from patients, and some patients underwent postoperative sampling. Serum TFPI2 levels were measured using automated enzyme-linked immunosorbent assay. Expression of TFPI2 in each cell type was evaluated using single-cell RNA sequencing. Survival analyses according to TFPI2 expression levels were performed based on publicly available databases. Serum TFPI2 was significantly elevated in patients with RCC compared to healthy volunteers, particularly those with clear cell histology. Metastatic RCC tumors exhibited higher TFPI2 levels than localized RCCs. Moreover, higher TFPI2 levels correlated with higher Fuhrman grades in clear cell RCC. Analyses of publicly available databases further supported this finding, showing an association between TFPI2 expression and overall survival, particularly in clear cell RCC. Single-cell RNA sequencing confirmed TFPI2 expression in clear cell RCC and normal kidney tubular epithelial cells, supporting its relevance. TFPI2 has emerged as a potential serum biomarker for RCC, offering avenues for improved detection and prognostication, similar to its utility in ovarian clear cell carcinoma. However, its clinical utility warrants further exploration in routine diagnostic and monitoring practices for patients with RCC.

TFP2

biomarker

renal cell carcinoma

clear cell

Globally, renal cell carcinoma (RCC) is the sixth and tenth most commonly diagnosed cancer in men and women, respectively, accounting for 5% and 3% of all tumors, respectively [1]. This prevalence has been fueled by an increase in tumors incidentally diagnosed on imaging studies such as ultrasonography and computed tomography. A critical issue in RCC is that no specific serum biomarker has yet been established for its detection.

The predominant pathological subtype of RCC is clear cell carcinoma (CCC), which accounts for approximately 75–80% of RCC tumors [2, 3]. CCC is also found in ovarian malignancies, with an incidence of 10–27% of all epithelial ovarian cancers [4, 5]. Ovarian CCC often exhibits resistance to standard chemotherapies, such as paclitaxel and carboplatin, which leads to lower survival rates in patients with ovarian CCC compared with patients with chemosensitive ovarian serous carcinoma [6]. Similarly, renal CCC is also normally resistant to chemotherapy, and we hypothesized that ovarian and renal CCCs may have some similarities in biological characteristics.

A serine protease inhibitor, tissue factor pathway inhibitor 2 (TFPI2; also known as placental protein 5) [7], has been identified as a highly specific serum biomarker for predicting ovarian CCC [7-9]. A modified proteomics technique, “secretome,” was used to identify TFPI2 in media conditioned by CCC-derived cell lines [7]. We recently developed a highly efficient automated enzyme-linked immunosorbent assay for TFPI2 detection and determined an adequate cutoff level of serum TFPI2 to differentiate between patients with CCC and those with other epithelial ovarian cancers and borderline tumors or benign ovarian lesions, including endometriosis [8]. Therefore, we hypothesized that TFPI2 could be an efficient biomarker for identifying renal CCC. Here, we validated the performance of TFPI2 as a specific serum biomarker for the preoperative prediction of renal CCC in a single-center study.

Patients

TFPI2 levels were measured in patients with RCC and healthy volunteers at a single-center (Yokohama City University Hospital). All patients with RCC underwent radical nephrectomy between 2001 and 2023 for suspected renal masses and were diagnosed with renal malignancies based on pathological findings. Healthy voluntary samples were collected anonymously by the Tosoh Corporation (Tokyo, Japan) or the Biobank at Yokohama City University. This study was performed in accordance with the Declaration of Helsinki and the Ethical Guidelines for Medical and Health Research Involving Human Subjects, after approval by the Institutional Ethics Committee of Yokohama City University (B181100031, B200800009, and B210300038).

Measurement of serum TFPI2 values in patients with RCC and healthy volunteers

Preoperative serum samples were obtained and stored for the analysis of TFPI2 serum concentration. To compare the preoperative and postoperative values of TFPI2, some patients were prospectively enrolled and blood sampling was performed preoperatively.

Blood samples were drawn within one month of surgery and collected in Venoject II serum separator tubes (VPAS109K60, Terumo, Tokyo, Japan). The tubes were stored for 2–3 h at 4 °C or 30 min at room temperature, and then centrifuged at 1000–1500g for 10 min. Serum aliquots were stored between −40 to 80 °C. TFPI2 concentrations in each serum sample were measured at the Department of Clinical Laboratory at Yokohama City University Hospital or Tosoh Corporation using reagents provided by the Tosoh Diagnostics Product Division (Tosoh Corporation, Tokyo, Japan). The measurements were performed by clinical laboratory technologists who were blinded to the sample information. TFPI2 concentration was measured by the direct assay method using an automated immunoassay analyzer system (Tosoh Corporation), as described in our previous study [7].

Expression analysis of TFPI2 in renal tumor tissue and generation of Kaplan-Meier survival curves derived from the publicly available databases

The UCSC Xena database (https://xena.ucsc.edu/) provided the bulk RNA-seq data. For comparison of TFPI2 gene expression levels in normal tissues and renal CCC tissues, "GDC TCGA Kidney Clear Cell Carcinoma (KIRC)" from the UCSC Xena was used. TFPI2 expression in normal tissues (72 samples) and primary tumors (534 samples) was compared.

The UCSC Xena data was also used to perform prognostic analysis according to the expression levels of TFPI2 in TCGA database. “GDC TCGA Kidney Clear Cell Carcinoma (KIRC)”, ”GDC TCGA Kidney Papillary Cell Carcinoma (KIRP)”, and ”GDC TCGA Kidney Chromophobe (KICH)” were used for clear cell, papillary, and chromophobe RCCs, respectively.

The primary tumors in each dataset were analyzed. The samples were divided into two groups based on median TFPI2 expression levels.

Identification of RNA expression of TFPI2 using single-cell RNA sequencing

We previously performed single-cell RNA sequencing of twelve surgically resected specimens from seven patients, including one Birt-Hogg-Dubé (BHD)-associated hybrid oncocytic chromophobe tumor (HOCT), one BHD-associated chromophobe RCC, one primary lesion, one lymph node metastasis from hereditary leiomyomatosis and renal cell cancer (HLRCC)-associated kidney cancer, two von Hippel-Lindau (VHL)-associated kidney cancers, one sporadic renal CCC, three intratumoral samples from a second sporadic renal CCC, and two normal kidney tissues [10]. We obtained the single-cell transcriptomes of 108,342 cells from these 12 tissues and divided them into 46,890 immune and 61,452 nonimmune cells using CD45, an immune cell marker. Nonimmune cells were annotated into cell clusters using previously reported marker genes for intercalated or principal cells of the collecting duct, distal tubules, loops of Henle, proximal tubules, glomerulus/vascular, and kidney cancers.

We analyzed this dataset following the methodology used in previous studies [10]. A total of 61,452 nonimmune cells were analyzed using the R package “Seurat” (version 3.1.2) [11]. The FindNeighbors (dims = 1:10) and FindClusters functions (resolution = 0.8) were used as parameters, and default settings were used for all other parameters. Each cluster was annotated based on the expression of existing marker genes as described in previous studies. Finally, the expression of the TFPI2 in each cluster was visualized using UMAP and violin plots.

Statistical analyses

Continuous variables were reported as means and standard deviations. Comparisons between two groups were performed using Student’s t, chi-square, or Welch's tests. For multiple comparisons of TFPI2, the Kruskal-Wallis H test was used. To compare survival curves, a log-rank test was performed between the two groups. The selection thresholds were as follows: log-rank P value<0.05 (two-sided) and hazard ratio (HR) within the 95% confidence interval (CI). Spearman’s coefficient values were used for the correlation analysis. All statistical analyses were two-sided, and statistical significance was set at P<0.05. All analyses were performed using SPSS software (version 28.0, Armonk, NY, IBM Corp.).

Background of enrolled patients with RCC and healthy volunteers

The characteristics of patients with localized (N=42) and metastatic (N=12) RCC are summarized in Tables 1 and 2, respectively. The average ages at diagnosis were 64.0 ± 10.7 and 64.8 ± 10.9 years in patients with localized and metastatic renal carcinoma, respectively. In both groups, the predominant histological finding was CCC (88.1% of localized and 75% of metastatic tissue samples). In the patients with metastatic renal carcinoma, the International Metastatic RCC Database Consortium (IMDC) risk classification was good (41.7%), intermediate (16.7%), or poor (41.7%). The predominant metastatic lesion sites were the lymph nodes and lungs (66.7%).

A total of 241 healthy volunteers, consisting of 139 women and 102 men, were enrolled to serve as controls, and their average age was 43.5 ± 9.7 years. Average serum TFPI2 value in the control group was 130.4 ± 35.4 pg/mL, and there was no significant sex differences between female (131.9 ± 35.2 pg/mL) and male (128.4 ± 35.7 pg/mL) volunteers (P=0.440; Figure S1A). Spearman’s coefficient values between TFPI2 concentrations and age were 0.074 (P=0.389; Figure S1B) in female and 0.036 (P=0.717, Figure S1C) in male volunteers.

Serum TFPI2 levels in patients with RCC and healthy volunteers

Serum TFPI2 levels were significantly higher in patients with RCC (P<0.001; Figure 1A), especially in those with CCC (P<0.001; Figure 1B). Metastatic RCC samples showed higher TFPI2 expression than localized RCC (P<0.001) and localized RCC levels were higher than those in healthy volunteers (P=0.042; Figure 1C). When focusing only on CCC, Fuhrman grades 3–4 showed higher serum TFPI2 levels than grades 1 (P=0.042) and 2 (P=0.007; Figure 2A). The Spearman’s coefficient between the tumor size of localized CCC and serum TFPI2 was 0.113 (P=0.500; Figure 2B). The cutoff point for TFPI2 to distinguish between patients with localized CCC and healthy controls was 170.0 pg/mL with 59.46% sensitivity and 88.36% specificity based on receiver operating characteristic (ROC) curve analysis (Figure 2C).

Protein and mRNA expressions of TFPI2 in RCC and normal kidney tissue

The GDC data set showed that mRNA expression of TFPI2 indicated no significant difference between RCC and normal kidney tissues at the bulk level of mRNA-seq (Figure 3A). However, the Kaplan-Meier curve derived from UCSC Xena data suggested that higher expression of TFPI2 was associated with a significantly shorter overall survival (OS) of patients than lower TFPI2 expression (P<0.001; data not shown). With respect to histological types, worse OS with higher expression of TFPI2 was prominent in CCC (P<0.001), but not in papillary (P=0.258) or chromophobe (P=0.176; Figure 3B) tumors.

RNA expression of TFPI2 in single-cell RNA sequencing

Using single-cell RNA sequencing, the expression of TFPI2 was identified in CCC using a UMAP plot (Figure 4A). The expression of TFPI2 in CCC was significantly higher than that in normal kidneys, including the loop of Henle, distal tubule, and glomerular vasculature as shown in the violin plot (P<0.001; Figure 4B).

The current study demonstrates that the measurement of serum TFPI2 concentration is a promising biomarker for renal CCC, a predominant malignant subtype of RCC. In this study, serum TFPI2 could discriminate patients with RCC, especially CCC with 170.0 pg/mL as possible cutoff value and could be a possible prognostic marker based on the analysis of publicly available data which showed that the mRNA expression of TFPI2 was associated with oncological outcome. Seeking biological evidence for the predictive and prognostic potential of serum TFPI2 in RCC, single-cell transcriptome analysis revealed significantly higher TFPI2 expression in CCC of RCC than in normal kidney cells.

The most common histological types of renal cell carcinoma are CCC (75–80% of RCC), papillary (10–15%), chromophobe (5%), and other rare forms, such as collecting duct carcinoma (< 1%), comprise the remainder [2, 3]. To date, the early detection of RCC has been hindered by the absence of effective serum biomarkers [12], although plasma cytokines or circulating proteins have been detected to predict the efficacy of pharmacological treatment for metastatic RCC [13–15]. Indeed, the precise prediction of RCC, especially CCC, within a real world clinical setup, including serum or urine tests, is a key unmet need.

This study demonstrated that serum TFPI2 could be a novel biomarker for renal CCC similar to its prior discovery for ovarian CCC [8]. Renal CCC-specific elevation of serum TFPI2 was consistent with the mRNA expression pattern observed in single-cell RNA sequencing. Thus, TFPI2 expression was significantly higher in renal CCC tumor cell clusters than in all normal kidney cell clusters. Although there was no difference between normal and tumor cells in the bulk data, there was a difference between normal kidney cells and renal CCC tumor cells at the single-cell level. Furthermore, high TFPI2 expression in tumor tissues was associated with worse OS in publicly available data ( UCSC Xena). These findings suggest that the upregulation of TFPI2 in tumor tissues is responsible for the high levels of serum TFPI2 in patients with renal CCC.

Human RCC tumors are thought to arise from a variety of specialized cells located along the length of the nephrons. Both CCC and papillary RCC are thought to arise from the epithelium of the proximal tubules [3]. Chromophobe RCC, oncocytoma, and collecting duct RCC are believed to arise from the distal nephrons, probably from the collecting tubule epithelium. In this study, single-cell transcriptome analysis indicated TFPI2 expression in the renal tubular epithelium of normal kidneys, supporting the oncological origin theory of TFPI2 expression in CCC.

TFPI2, also known as placental protein 5 (PP5) [16, 17], is abundantly produced in the placenta and significantly elevated in the serum of pregnant women [18]. However, this study found no difference in serum TFPI2 values between female and male volunteers. In addition, this study showed no evidence of an effect of age on serum TFPI2 levels in either female or male volunteers. In fact, TFPI2 is now utilized to predict ovarian CCC in clinics without any correction for patient background [8]. Thus, we believe that TFPI2 could be used as a robust serum biomarker for the simple screening of patients with renal CCC, potentially monitoring tumor recurrence and even predicting OS in daily clinical use.

Although this study provides valuable insights into the potential of TFPI2 as a serum biomarker for renal CCC, this study had several limitations. First, the study was conducted at a single-center, which may have limited the generalizability of the findings. Second, the sample size, particularly for certain subgroups such as patients with metastatic RCC, was relatively small, limiting the statistical power and robustness of the results. This study primarily focused on preoperative serum TFPI2 levels and their association with RCC diagnosis. Third, long-term follow-up data, including recurrence rates, disease progression, and OS, was not collected in this study and may provide a more comprehensive assessment of TFPI2's prognostic value in RCC. Finally, TFPI2's diagnostic and prognostic utility should be externally validated in independent cohorts in future studies.

TFPI2 has emerged as a potential serum biomarker for renal CCC, offering avenues for improved detection and prognostication, similar to its utility in ovarian CCC. However, the clinical utility of TFPI2 warrants further exploration in routine diagnostic and monitoring practices for patients with RCC. Single-cell transcriptome analysis further elucidated the TFPI2 expression patterns, confirming its relevance in renal CCC.

Data availability

The data that support the findings of this study are available on request from the corresponding author, [H.I.].

Acknowledgements

This study was supported by:

Takeda Science Foundation (Hiroki Ito)

Grants-in-Aid for Scientific Research (No. 19K09676) was received from the Ministry of Education, Science, Sports, and Culture of Japan. (Noboru Nakaigawa)

Competing Interests: None to declare.

Personal Financial Interests: None

Funding: Joint research funds from Tosoh Corporation (Masahiro Yao, Noboru Nakaigawa, Kazuhide Makiyama)

Employment: None

Ethics Approval:

This study was performed in accordance with the Declaration of Helsinki and the Ethical Guidelines for Medical and Health Research Involving Human Subjects, after approval by the Institutional Ethics Committee of Yokohama City University (B181100031, B200800009, and B210300038).

Consent: Written informed consent was obtained from all patients for their data to be used for research purposes.

Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;68(1):7-30. Epub 20180104. doi: 10.3322/caac.21442. PubMed PMID: 29313949.
Cairns P. Renal cell carcinoma. Cancer Biomark. 2010;9(1-6):461-73. doi: 10.3233/CBM-2011-0176. PubMed PMID: 22112490; PubMed Central PMCID: PMCPMC3308682.
Kovacs G, Akhtar M, Beckwith BJ, Bugert P, Cooper CS, Delahunt B, et al. The Heidelberg classification of renal cell tumours. J Pathol. 1997;183(2):131-3. doi: 10.1002/(SICI)1096-9896(199710)183:2<131::AID-PATH931>3.0.CO;2-G. PubMed PMID: 9390023.
Haruta S, Furukawa N, Yoshizawa Y, Tsunemi T, Nagai A, Kawaguchi R, et al. Molecular genetics and epidemiology of epithelial ovarian cancer (Review). Oncology reports. 2011;26(6):1347-56. Epub 20110912. doi: 10.3892/or.2011.1456. PubMed PMID: 21922146.
Machida H, Matsuo K, Yamagami W, Ebina Y, Kobayashi Y, Tabata T, et al. Trends and characteristics of epithelial ovarian cancer in Japan between 2002 and 2015: A JSGO-JSOG joint study. Gynecologic oncology. 2019;153(3):589-96. Epub 20190321. doi: 10.1016/j.ygyno.2019.03.243. PubMed PMID: 30905436; PubMed Central PMCID: PMCPMC7526703.
Reid BM, Permuth JB, Sellers TA. Epidemiology of ovarian cancer: a review. Cancer Biol Med. 2017;14(1):9-32. doi: 10.20892/j.issn.2095-3941.2016.0084. PubMed PMID: 28443200; PubMed Central PMCID: PMCPMC5365187.
Arakawa N, Miyagi E, Nomura A, Morita E, Ino Y, Ohtake N, et al. Secretome-based identification of TFPI2, a novel serum biomarker for detection of ovarian clear cell adenocarcinoma. J Proteome Res. 2013;12(10):4340-50. Epub 20130829. doi: 10.1021/pr400282j. PubMed PMID: 23805888.
Arakawa N, Kobayashi H, Yonemoto N, Masuishi Y, Ino Y, Shigetomi H, et al. Clinical Significance of Tissue Factor Pathway Inhibitor 2, a Serum Biomarker Candidate for Ovarian Clear Cell Carcinoma. PloS one. 2016;11(10):e0165609. Epub 20161031. doi: 10.1371/journal.pone.0165609. PubMed PMID: 27798689; PubMed Central PMCID: PMCPMC5087914.
Miyagi E, Arakawa N, Sakamaki K, Yokota NR, Yamanaka T, Yamada Y, et al. Validation of tissue factor pathway inhibitor 2 as a specific biomarker for preoperative prediction of clear cell carcinoma of the ovary. Int J Clin Oncol. 2021;26(7):1336-44. Epub 20210519. doi: 10.1007/s10147-021-01914-y. PubMed PMID: 34009487; PubMed Central PMCID: PMCPMC8213588.
Jikuya R, Murakami K, Nishiyama A, Kato I, Furuya M, Nakabayashi J, et al. Single-cell transcriptomes underscore genetically distinct tumor characteristics and microenvironment for hereditary kidney cancers. iScience. 2022;25(6):104463. Epub 20220525. doi: 10.1016/j.isci.2022.104463. PubMed PMID: 35874919; PubMed Central PMCID: PMCPMC9301876.
Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM, 3rd, et al. Comprehensive Integration of Single-Cell Data. Cell. 2019;177(7):1888-902 e21. Epub 20190606. doi: 10.1016/j.cell.2019.05.031. PubMed PMID: 31178118; PubMed Central PMCID: PMCPMC6687398.
Choueiri TK, Motzer RJ. Systemic Therapy for Metastatic Renal-Cell Carcinoma. The New England journal of medicine. 2017;376(4):354-66. doi: 10.1056/NEJMra1601333. PubMed PMID: 28121507.
Tran HT, Liu Y, Zurita AJ, Lin Y, Baker-Neblett KL, Martin AM, et al. Prognostic or predictive plasma cytokines and angiogenic factors for patients treated with pazopanib for metastatic renal-cell cancer: a retrospective analysis of phase 2 and phase 3 trials. Lancet Oncol. 2012;13(8):827-37. Epub 20120702. doi: 10.1016/S1470-2045(12)70241-3. PubMed PMID: 22759480.
Harmon CS, DePrimo SE, Figlin RA, Hudes GR, Hutson TE, Michaelson MD, et al. Circulating proteins as potential biomarkers of sunitinib and interferon-alpha efficacy in treatment-naive patients with metastatic renal cell carcinoma. Cancer Chemother Pharmacol. 2014;73(1):151-61. Epub 20131113. doi: 10.1007/s00280-013-2333-4. PubMed PMID: 24220935; PubMed Central PMCID: PMCPMC3889677.
Zurita AJ, Jonasch E, Wang X, Khajavi M, Yan S, Du DZ, et al. A cytokine and angiogenic factor (CAF) analysis in plasma for selection of sorafenib therapy in patients with metastatic renal cell carcinoma. Ann Oncol. 2012;23(1):46-52. Epub 20110404. doi: 10.1093/annonc/mdr047. PubMed PMID: 21464158; PubMed Central PMCID: PMCPMC3276320.
Kisiel W, Sprecher CA, Foster DC. Evidence that a second human tissue factor pathway inhibitor (TFPI-2) and human placental protein 5 are equivalent. Blood. 1994;84(12):4384-5. PubMed PMID: 7994054.
Miyagi Y, Koshikawa N, Yasumitsu H, Miyagi E, Hirahara F, Aoki I, et al. cDNA cloning and mRNA expression of a serine proteinase inhibitor secreted by cancer cells: identification as placental protein 5 and tissue factor pathway inhibitor-2. J Biochem. 1994;116(5):939-42. doi: 10.1093/oxfordjournals.jbchem.a124648. PubMed PMID: 7896752.
Bohn H, Winckler W. [Isolation and characterization of the placental protein pp5 (author's transl)]. Arch Gynakol. 1977;223(3):179-86. doi: 10.1007/BF00667387. PubMed PMID: 579296.

Table 1: Clinicopathologic characteristics of patients with localized renal cell carcinoma

Number of patient			42
Age, yr, median (range)			64 (40-80)
Female/Male, n			9/33
Tumor size, cm, mean±SD			5.7 ± 2.6
Histology, n (%)	Clear cell		37 (88.1)
	Non-clear	Papillary	2 (4.8)
		Chromophobe	1 (2.4)
		Chromophobe & collecting duct	1 (2.4)
		Neuroendocrine	1 (2.4)
	lyv (+)		22 (52.4)
pT stage, n (%)	1a		13 (31.0)
	1b		7 (16.7)
	2a		3 (7.1)
	2b		1 (2.4)
	3a		11 (26.2)
	3b		4 (9.5)
	4		3 (7.1)

Table 2: Clinicopathologic characteristics of patients with metastatic renal cell carcinoma

Number of patient			12
Age, yr, median (range)			65 (40-80)
Female/Male			2/10
Tumor size, cm, mean±SD			7.6 ± 3.5
Radical nephrectomy/Biopsy			7/5
Histology, n (%)	Clear cell		9 (75)
	Non-clear	Papillary	2 (16.7)
		Mucinous tubular spindle cell	1 (8.3)
	*Lymph vascular invasion positive		7 (100)
IMDC classification, n (%)		Good	5 (41.7)
		Intermediate	2 (16.7)
		Poor	5 (41.7)
Metastatic lesions, n (%)	Lymph node		8 (66.7)
	Lung		8 (66.7)
	Bone		5 (41.7)
	Liver		5 (41.7)
	Adrenal gland		1 (8.3)
	Contralateral Kidney		1 (8.3)
	Spleen		1 (8.3)
	Pleural		1 (8.3)
	Salivary gland		1 (8.3)
	Parotid gland		1 (8.3)

* Evaluated in radical nephrectomy cases only

Competing interest reported. Masahiro Yao, Noboru Nakaigawa, and Kazuhide Makiyama reports financial support was provided by Tosoh Corporation. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

SupplementaryFigure1.pptx
Supplementary Fig. 1 Dot plot analysis comparing serum TFPI2 levels in female and male control (A), and correlation between serum TFPI2 concentration and age in female and (B) male (C) control

Tissue Factor Pathway Inhibitor 2 (TFPI2) is a Serum Biomarker for Clear Cell Renal Carcinoma

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Materials and methods

Results

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1