Moderate evidence exists for four microRNAs as potential biomarkers for tendinopathies and degenerative tendon ruptures at the upper extremity in elderly patients: conclusion of a systematic review with best-evidence synthesis

Purpose The aim of this systematic review was to investigate tendon-specific microRNAs (miRNAs) as biomarkers for the detection of tendinopathies or degenerative tendon ruptures. Also, their regulatory mechanisms within the tendon pathophysiology were summarized. Methods A systematic literature research was performed using the PRISMA guidelines. The search was conducted in the Pubmed database. The SIGN checklist was used to assess the study quality of the included original studies. To determine the evidence and direction of the miRNA expression rates, a best-evidence synthesis was carried out, whereby only studies with at least a borderline methodological quality were considered for validity purposes. Results Three thousand three hundred seventy studies were reviewed from which 22 fulfilled the inclusion criteria. Moderate evidence was found for miR-140-3p and miR-425-5p as potential biomarkers for tendinopathies as well as for miR-25-3p, miR-29a-3p, miR-140-3p, and miR-425-5p for the detection of degenerative tendon ruptures. This evidence applies to tendons at the upper extremity in elderly patients. All miRNAs were associated with inflammatory cytokines as interleukin-6 or interleukin-1ß and tumor necrosis factor alpha. Conclusions Moderate evidence exists for four miRNAs as potential biomarkers for tendinopathies and degenerative tendon ruptures at the upper extremity in elderly patients. The identified miRNAs are associated with inflammatory processes.


Introduction
Tendons are a key element in the musculoskeletal system for the generation of movements due to their ability to transmit and withstand forces [1].However, pathological tendon conditions such as tendinopathies are prevalent in the entire population with incidences of up to 10.52 per 1,000 persons per year [2].Tendinopathies are characterized by persistent tendon pain and loss of function associated with mechanical loading [3] and could cause a reduced life quality [4], impairments of work and sportive performances [5], and underestimated high socio-economic costs [6].The pathogenesis is understood as a continuum model with the end stage of degenerative tendinopathy [7], where symptoms may persist for decades [8].Since associated degenerative changes are present in 97% of all ruptured tendons [9], it is assumed that tendinopathies can cause such acute severe tendon injuries [10].However, high-quality evidence for effective preventive measures for tendinopathies is lacking [11,12] and early clinical management is challenging due to asymptomatic early stages [13] as well as often ignored minor symptoms [14].In this context, established clinical routine diagnostics such as anamnesis, clinical examination, and tendon imaging [15] are suitable for the diagnosis of manifested tendinopathies, but inappropriate for asymptomatic early stages.Thus, more research is needed to evaluate diagnostic tools for the early diagnosis of tendinopathies and associated degenerative tendon ruptures, including the identification of potential biomarkers.
MicroRNAs (MiRNAs) are short noncoding RNA molecules that bind to complementary messenger-RNAs to regulate their activity [16].In humans, miRNAs are expressed in a cell-and tissue-specific manner [17,18].They can be detected in a variety of different body fluids including blood, tears, or saliva [19].MiRNAs are suitable diagnostic biomarkers [20], because they are protected from endogenous RNAse activity [21] and can endure freeze-thaw cycles [22].In this context, miRNAs have been evaluated as non-or minimal-invasive biomarkers for numerous diseases including Alzheimer [23], multiple sclerosis [24], heart failure [25], or various cancer types [26][27][28], but little is known with respect to degenerative tendon conditions yet.
MiRNAs have been associated with the tendon tissue pathophysiology.It has been demonstrated that miR-NAs could reduce adhesion, enhance remodeling, and promote angiogenesis in the context of tendon healing [29].Also, miRNAs are known to regulate a variety of different genes related to tendon healing and tenogenesis [30].To date, there are two systematic reviews investigating the relationship between the expression rates of miRNAs and tendon tissue functions.Dubin et al. [31] investigated the effect of miRNAs on tenocytes and tendon-related gene expression.They show that miRNAs have both positive and negative effects on the tendon tissue homeostasis.Giordano et al. [32] examined the therapeutic potential of miRNAs in the context of tendon healing.The authors conclude that miRNAs could serve as useful therapeutic targets due to their influence on the expression of cytokines and differentiation and proliferation of stromal cell lines involved in the composition of the extracellular matrix.However, there is no systematic review questioning, if miRNAs can be used as biomarkers for pathological tendon conditions.Therefore, the aim of this systematic review was to investigate tendonspecific miRNAs as biomarkers for the detection of tendinopathies or degenerative tendon ruptures.Also, the regulatory mechanisms of miRNAs within the tendon pathophysiology were summarized.

Research design
The systematic review was conducted using the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA) [33].The inclusion and exclusion criteria were determined using a PICO(S) scheme: i.e., population (P), intervention (I), comparison (C) outcome (O), and study design (S) [34].Additionally, the item "other" was included to account for further criteria (Table 1).The inclusion criteria were: (i) human studies including patients with tendinopathies or degenerative tendon ruptures; (ii) tendon-specific miRNAs quantified in the tissue and/or circulation; (iii) primary data published in original investigations; (iv) publication language in English or German; and (v) full text availability.Studies were excluded, when the miRNAs were not specified.All methodological steps were conducted by one author and a second validated them.In terms of uncertainties, it was discussed until a consensus was reached.Due to the noninvasive character, no ethical approval was considered.

Literature search strategy and study selection
The search was performed in the meta-database Pubmed on 04/25/2022 and was not restricted to a specific time period.To find relevant studies, a search line was elaborated using the inclusion and exclusion criteria.The search line included the following terms: (micro RNA OR miR OR miRNA OR microRNA OR circRNA OR circulating RNA OR ciRNA) AND (tendon OR tendinopathy OR tendinosis OR tendinitis OR tendosynovitis OR tenocytes OR ruptures OR connective tissue) AND (physiology OR pathology OR pathophysiology OR maladaptation OR load OR intervention OR adaptation OR baseline OR timepoint OR pre-post OR comparison).Additionally, the reference list of two previous systematic [31,32] and five previous narrative reviews [29,30,[35][36][37] within the particular research field were screened for further suitable studies.After duplicates were removed, the abstracts and full texts of the remaining studies were checked for their fit by taking the eligibility criteria into account.

Risk of bias assessment
The study quality and associated risk of bias was determined using the Scottish Intercollegiate Guidelines Network (SIGN) checklist [38].Therefore, the particular checklist for randomized controlled trials, cohort studies, case-control studies, and diagnostic and economic studies was used.The checklists consisted of 10-15 items to test the internal validity of the studies.The items were rated as "Yes" (Y), "No" (N), "Can't say" (CS), or "not applicable" (NA).The overall rating of the studies involved the following outcomes: "high quality", "acceptable quality", "borderline quality", or "unacceptable quality", as described in detail elsewhere [39].

Data extraction
The data extraction of the studies was conducted according to the PICO(S) scheme.For validity, studies with an unacceptable quality were not considered, as conducted previously [39].Due to the found heterogeneity in terms of the methodologies and results of the studies, no metaanalysis was performed.Instead, a best-evidence synthesis was conducted to clarify the evidence and direction of the miRNA expression rates [39].The expression rates and their associations with tendinopathies or degenerative ruptures were classified as: upregulated (↑), downregulated (↓), or neutral ( →), which means that no clear pattern was given.To increase the validity, only miRNAs that were found, at least in part, twice in different studies were considered in the best-evidence synthesis.An exception was made for the study by Thankam et al. [40], where only the 10 most up-and down-regulated miRNAs were included to reduce the amount of data from this comprehensive microarray study including more than 235 miRNAs.Nevertheless, miRNAs that occurred more than two times were matched to the study by Thankam et al. [40], if they were not already included in the 10 most up-or down-regulated in this study.Table 2 summarizes the applied criteria for the best-evidence synthesis according to Asker et al. [39], whereby the final ratings were as follows: "strong evidence", "moderate evidence", "limited evidence", and "no evidence".
MiR-140-3p and miR-425-5p could serve as potential biomarkers for tendinopathies (Table 6).For both miRNAs, significantly decreased expression levels were observed in tendinopathic tendons in the circulation, when compared to healthy tendons [41].In addition, the study by Thankam et al. [40] found that both miRNAs were significantly decreased in tendon injuries with glenohumeral arthritis compared to healthy control tendons.However, it is important to emphasize that these were not tendinopathic, but tendons with massive tears.Moreover, miR-140-3p was significantly decreased in tendinopathic tendons with glenohumeral arthritis compared to tendinopathic tendons without glenohumeral arthritis [54].Thus, miR-140-3p and miR-425-5p may be potential diagnostic biomarkers for tendinopathies, but the results should be taken with caution due to the association found with further diseases.
MiR-25-3p, miR-29-a-3p, miR-140-3p, and miR-425-5p could serve as potential biomarkers for the detection of degenerative tendon ruptures (Table 6) due to the significant downregulation in the circulation in degenerative ruptured tendons compared with healthy tendons [41].Here, miR-29a-3p and miR140-3p were shown to be significantly downregulated in both tissue and circulation in degenerative ruptured tendons [41].Additionally, the study by Thankam et al. [40] demonstrated that miR-25-3p, miR-29a-3p, miR-140-3p, and miR-425-5p were also significantly downregulated in tendon ruptures of the bicep tendon compared with healthy control tendons.In the study by Thankam et al. [54], it was shown that miR-25-3p, miR-29a-3p, and miR-140-3p were also significantly downregulated in tendinopathic tendons with glenohumeral arthritis compared with tendinopathic biceps tendons.Furthermore, miR-29a-3p was downregulated in tissue in tendinopathic supraspinatus tendons compared with healthy subscapularis tendons [43].In a study by Leal et al. [44], miR-29a-3p was inversely correlated with various matrix metalloproteinases (MMPs), but there were no significant differences in the expression rates of miR-29a-3p between healthy and ruptured supraspinatus tendons.Thus, there seems to be a relationship between miR-25-3p, miR-29a-3p, miR-140-3p, and miR-425-5p with degenerative tendon ruptures.MiR-140-3p and miR-425-5p were significantly downregulated in both tendinopathic tendons and degenerative tendons in the circulation compared with healthy control tendons.A progressive decrease in expression levels was also observed for the two miRNAs in relation to the severity of tendon degeneration [41].This suggests that miR-140-3p and miR-425-5p may contribute to the pathogenesis and/or progression of degenerative rotator cuff diseases in elderly patients, requiring further validation.
Different regulatory mechanisms of miRNAs in tendon tissue are discussed in the literature.Briefly, miR-25-3p can be considered as a potential tumor biomarker in breast cancer [62] or osteosarcoma [63].In both cases, cytokines such as interleukin-6 (Il-6) influence tumor genesis [64,65] and Il-6 also plays a role in tendon ruptures [66].For miR-29a-3p, it has been shown to be an eligible biomarker in colorectal cancer [21] and tuberculosis [67], among others.In both tuberculosis and carciogenesis, Il-6 play an important role again [68,69].Regarding the miR-140-3p, it is evident that this miRNA is also significantly down-regulated in human chondrocytes in glenohumeral arthritis [70], among others.MiR-140-3p was shown to reduce the concentration of interleukin-1ß (IL-1ß) induced inflammatory factors [70].IL1-ß plays a crucial role mainly in the inflammatory phase of tendon healing [71], but it has also been shown that it has a significant role in arthritis [72].Gu et al. [73] demonstrated that miR-425-5p is associated with both tumor necrosis factor alpha (TNF-alpha) and IL-1ß, which also plays a role in tendinopathies [71].Overall, all miRNAs for which moderate evidence was found are associated with specific inflammatory cytokines.Therefore, it is unclear, if these miRNAs can serve as potential biomarkers for tendon diseases or significantly alter their expression patterns tissue-independently due to inflammatory processes.More experimental high-quality research is needed to validate miR-25-3p, miR-29a-3p, miR-140-3p, and miR-425-5p as tendon-specific biomarkers.
Although this systematic review increased the knowledge on miRNAs as potential biomarkers for tendon diseases, there are few limitations.While Pubmed can be regarded as the most comprehensive database, it has to be noted that it was the only platform used for the literature search.Additionally, all methodological steps of our review were conducted only by one author.However, a second author carefully validated the entire proceed and all outcomes independently, which is not fully compliant with the PRISMA guidelines.More experimental high-quality studies are needed to investigate miRNAs in both the tissue and circulation to validate them as biomarkers for tendinopathies or degenerative tendon ruptures; especially, at the lower extremity and in younger individuals.Also, more basic research is required to better understand the regulatory mechanisms of miRNAs within the tendon pathophysiology.

Conclusion
Our systematic review based on a best-evidence synthesis suggests that moderate evidence exists for four miRNAs as potential biomarkers for tendinopathies and degenerative tendon ruptures at the upper extremity in elderly patients.The identified miRNAs are associated with inflammatory processes.More experimental highquality research to validate the four miRNAs is required.

Figure 1
Figure 1 shows the results of the literature search strategy and study selection.3,345 and 25 articles were

Table 1
PICO(S) scheme for the definition of the inclusion and exclusion criteria

Table 3
Results of the 22 studies checked for the risk of bias assessment using the SIGN checklist SIGN Scottish Intercollegiate Guidelines Network, Y Yes, N No, NA Not applicable, CS Can't say

Table 4
Characteristics of the included studies according to the PICO(S) scheme

Table 5
Overview of the regulatory mechanisms of the miRNAs of the included 22 studies