Dysregulated MicroRNAs as Biomarkers or Therapeutic Targets in Cisplatin-Induced Nephrotoxicity: A Systematic Review

The purpose of this systematic review was to map out and summarize scientific evidence on dysregulated microRNAs (miRNAs) that can be possible biomarkers or therapeutic targets for cisplatin nephrotoxicity and have already been tested in humans, animals, or cells. In addition, an in silico analysis of the two miRNAs found to be dysregulated in the majority of studies was performed. A literature search was performed using eight databases for studies published up to 4 July 2021. Two independent reviewers selected the studies and extracted the data; disagreements were resolved by a third and fourth reviewers. A total of 1002 records were identified, of which 30 met the eligibility criteria. All studies were published in English and reported between 2010 and 2021. The main findings were as follows: (a) miR-34a and miR-21 were the main miRNAs identified by the studies as possible biomarkers and therapeutic targets of cisplatin nephrotoxicity; (b) the in silico analysis revealed 124 and 131 different strongly validated targets for miR-34a and miR-21, respectively; and (c) studies in humans remain scarce.


Introduction
Cisplatin is an antineoplastic agent widely used in the treatment of solid tumors because of its extensive cytotoxic activity; however, it is associated with a high incidence of treatment-induced acute kidney injury (AKI) [1]. The kidney plays a crucial role in the excretion of drugs and is therefore highly susceptible to drug use-related injury. The known mechanisms underlying cisplatin-induced nephrotoxicity include apoptosis, autophagy, nuclear and mitochondrial DNA damage and production of reactive oxygen species [2]. Because it is dose-limiting, the dose of cisplatin is often reduced or treatment discontinued following nephrotoxicity, which can increase the risk of disease recurrence or progression.
The biomarkers used routinely to assess kidney function and damage are mainly serum creatinine (CRE), blood urea nitrogen (BUN), and urine output levels. However, these biomarkers have low sensitivity and specificity for detecting early events associated with AKI [3,4] and are significantly altered only after the marked progression of kidney injury [5]. A late diagnosis impairs the timely treatment of patients with AKI, which is also an aggravating factor in several oncological treatments based on cisplatin regimens. Therefore, identification of new nephrotoxicity biomarkers is urgently required.
MicroRNAs (miRNAs) have attracted considerable research attention as novel biomarkers. miRNAs are a class of non-coding RNAs that are approximately 20 bp in length and city in cells, animals, or humans. Only original articles that evaluated the expression of miRNAs following exposure to cisplatin, not concomitantly with another drug, as well as those studies that comprised some nephrotoxicity assessment method, quantified miRNA expression or showed these data, and did not have an exclusive computational approach ("in silico") were included in this review. No language restrictions were imposed.
Duplicate studies retrieved using the search strategy were excluded using EndNote ® . The remaining studies were allocated to the Rayyan ® tool (https://www.rayyan.ai/, accessed on 12 July 2021), where two review steps were performed: screening and analysis of the article titles and abstracts. This last step was performed independently by two reviewers (NGT and JKNP), and any discrepancies were discussed with a third and fourth reviewer (PM and MBV).
The studies selected for the first analysis were moved to the complete and critical article reading stage, which was also performed by two reviewers independently (NGT and JKNP), and the discrepancies were discussed with a third and fourth reviewer (PM and MBV). In addition, references cited in all included articles were reviewed to identify any studies that might have been missed in the search.

Data Extraction and Analysis
Data extracted from articles that met the requirements included author, year of publication, country, aim of the study, study population (cell, animal, or human), cisplatin treatment/exposure (dose/concentration), nephrotoxicity assessment, methods used to identify miRNAs, time of the analysis related to cisplatin exposure, miRNAs significantly differentially expressed in nephrotoxicity, pathophysiological implications of dysregulated miRNAs (only those related to the authors' own results and discussions; results or discussions based on other studies were not collected), and role of miRNAs in nephrotoxicity. The extracted data were entered into a Microsoft Word spreadsheet ® , and this process was again conducted by two independent reviewers (NGT and JKNP), and disagreements were discussed with a third and fourth reviewer (PM and MBV). The included studies were categorized according to the studied population, and the results of this review were presented using the narrative synthesis approach.

Bioinformatics Analysis
To generate the interaction network of selected miRNAs (miR-34a and miR-21), we employed miRTargetLink 2.0, a tool containing experimentally validated interactions of human miRNA-mRNA pairs. The data shown correspond to strongly supported miRNAtarget interactions, that is, those validated experimentally using reporter assays, western blotting, RT-qPCR, microarray, or next-generation sequencing experiments. miRNA annotations were obtained from the latest version of miRBase (v.22.1), while the experimentally validated targets were retrieved from miRTarBase (v.8) and miRATBase. miRTargetLink 2.0 is freely accessible (https://ccbcompute.cs.uni-saarland.de/mirtargetlink2/, accessed on 26 August 2021).

Results and Discussion
The search strategy identified 1002 studies from multiple databases, with 570 overlapping studies. Of the remaining 432 articles, 393 were excluded by reviewing their titles and abstracts. The full text of the remaining 39 articles underwent a review, and 30 studies  that met the eligibility criteria were included. Figure 1 shows a flowchart of the literature search and Supplementary Table S1. ontains a reference list of the studies excluded in the full-text review, along with the reasons for their exclusion. studies  that met the eligibility criteria were included. Figure 1 shows a flowchart of the literature search and Supplementary Table S1. ontains a reference list of the studies excluded in the full-text review, along with the reasons for their exclusion. All studies were published in English, between 2010 and 2021. Most of these studies were conducted in China [24,25,31,32,[36][37][38][39][40][41][42][43][44][47][48][49][50] and the United States of America (USA) [26,27,34,38,46]. The main differences among the 30 included studies were related All studies were published in English, between 2010 and 2021. Most of these studies were conducted in China [24,25,31,32,[36][37][38][39][40][41][42][43][44][47][48][49][50] and the United States of America (USA) [26,27,34,38,46]. The main differences among the 30 included studies were related to population, cisplatin dose, nephrotoxicity assessment, and time of analysis of the miR-NAs. Seven studies were conducted only in cells [24,25,36,[47][48][49][50], 11 only in animal models [26][27][28][29][30][31][32][33][51][52][53], 10 in both in vitro and in vivo models [34,35,[37][38][39][40][41][42][43][44], one in cells and humans [46], and only one in humans [45]. The principal methods used for assessing nephrotoxicity in the studies conducted in cells were viability and apoptosis tests, with 17 studies using at least one of these. Of the 21 studies involving animal models, only four did not assess nephrotoxicity by histological analyses [28,34,38,39], and only three did not measure BUN or CRE levels [28,37,53]. Human studies used two different criteria to assess nephrotoxicity, but both were based on serum CRE levels; one used the Common Toxicity Criteria for Adverse Events (version 4) [45], while the other used the AKI Network criteria [46]. As for the cisplatin treatment used, neither the dose nor the duration of exposure or treatment was similar among the studies. A summary of the characteristics of the 30 studies included in this systematic review is presented in Table 1.

Methods of Included Studies
For miRNA identification, the cellular studies used cells or the supernatants of cell cultures, and only one did not specify the sample used [48]. Animal studies used plasma, urine, or kidney tissue, while one human study used plasma samples [45] and the other used urine samples [46]. Regardless of the population studied, the method chosen for miRNA identification was mainly real-time polymerase chain reaction (RT-PCR), which in some cases was also associated with microarray or miRNA-seq techniques. RT-PCR is the gold standard method for quantifying miRNAs, and the results of studies that did not use RT-PCR should be validated using this technique.
In addition, all studies evaluated the expression of miRNAs after exposure to cisplatin, although the exact time was different for each of them. The study by Quintanilha et al. was the only one to propose miRNA as a predictor of nephrotoxicity, as they identified altered expression of miRNAs in patients with nephrotoxicity before chemotherapy with cisplatin [45]. Table 2 presents the main results of the included studies.
Interestingly, some studies selected miRNAs for validation based on their sequencing or array results, while others selected miRNAs based on the existing literature and on the action of target genes already known for these miRNAs.
Findings from in vitro models are important for discovering potential therapeutic targets; however, as these findings may differ from those of other organisms, it is essential that they be validated in in vivo models. The in vivo model studies included in this review used serum, urine, or renal tissue samples, and those performed in humans used plasma or urine. Blood contains high levels of potentially interfering proteins, and alterations in serum biomarker levels may not be related to renal function, as it can be a systemic response. Instead, urine samples have several advantages, such as non-invasive collection in large quantities and strong specificity for kidney damage [55].
Although extracellular fluids may contain RNA-degrading enzymes, miRNAs remain stable and detectable in blood because they are associated with protein complexes [56]; in addition, they are present inside exosomes [57]. Urinary miRNAs appear to be derived from the kidney and urinary tract cells and are filtered through the glomerulus or secreted by renal tubules [58].

miRNAs as Biomarkers or Therapeutic Agents or Targets of AKI
A total of 115 different miRNAs were found to be differentially expressed in the studies included in this systematic review (Supplementary Table S2), with 90 being dysregulated in one study and only 25 in more than one ( Table 3). The miRNAs that most frequently appeared in the studies were miR-34a and miR-21, appearing in six and five studies, respectively ( Figure 2). miR-34a was upregulated in all six studies included [27,30,34,35,47,52], and miR-21 was upregulated in four studies [33,46,52] and downregulated in one study [25].      CIS upregulated miR-181a expression leading to negative regulation of Bcl-2 (anti-apoptotic gene) and positive regulation of BAX (pro-apoptotic gene). Thus, miR-181a expression is associated with cell apoptosis.
CIS may play a role in tubular epithelial cell apoptosis by suppressing Bcl-2 expression, which is achieved by regulating the target gene of microRNA-181a. These findings pave a novel approach to the enhancement of prevention treatment of CIS-induced nephrotoxicity. Overexpression of miR-500a-3P had a protective role in CIS-induced kidney injury, as it showed: -To limit programmed cell death; -To decreased chemokine MCP-1 and proinflammatory cytokines TNF-a and IL-8; -To decrease phosphorylation and membrane translocation of MLKL, a key index for detecting necroptosis.
Considering the antinecroptotic and anti-inflammatory merits, miR-500a-3P may be a novel therapeutic agent for AKI treatment. HIPK2, a key regulator of kidney fibrosis, was predicted as the common target gene of miR-9-3p and miR-371b-5p.
An integrative network approach encompassing miRNAs, target genes, and bioinformatics analysis showed that miR-9-3p and miR-371b-5p could be critical miRNAs in CIS-induced renal tubular cell injury.
MiR-500a-3P is effective in controlling the AKI and may be an appropriate miRNA therapeutics. The data suggest that there may exist a pro-apoptotic role of miR-449 in CIS-induced AKI via regulating the SIRT1/P53/BAX pathway. Therefore, it is suggested that miR-449 be a potential therapeutic target for treating AKI.  The study hypothesized that the negative regulation of Numb affects Notch signaling via miR-31 in CIS-induced AKI, because Notch signaling is associated with the balance among the cell proliferation and apoptosis that influence the process of various organ injuries.
MiR-31 expression is upregulated in CIS-induced AKI. These miRNAs are associated with pathways, as DNA damage response, apoptosis, cell cycle regulation, and inflammation. The top canonical pathway affected were p53 and PI3K/AKT pathways. Also, mRNAs predicted as targets of the altered miRNAs in the kidney were associated with DNA damage response, apoptosis, and cell cycle regulation.
These miRNAs are potential urinary biomarkers for CIS-induced kidney injury.  This study showed that miR-34a induction during CIS nephrotoxicity was mediated by p53. However, blockage of miR-34a increased cell injury and death. The authors speculated that miR-34a may regulate or repress proapoptotic genes.
There is evidence for a cytoprotective role of induced miR-34a against CIS-induced apoptosis in renal cells. Downregulated-Both cells: miR-30a miR-30b miR-30c miR-30d miR-30e The downregulation of miR-30c induced by CIS positively regulated the expression of its Bnip3L and Hspa5 target genes, which resulted in significant increase on apoptosis.
MiR-30c might be involved in regulating CIS-induced cell apoptosis, and it might supply a new strategy to minimize CIS-induced nephrotoxicity. CIS exposure upregulated miR-140-5p in response to oxidative stress induced by CIS. It was also showed that MnSOD activity and cell vitality were increased, and LDH leakage was reduced in miR-140-5p overexpression. In fact, miR-140-5p directly targets the 3 -UTR of Nrf2 mRNA and increases the Nrf2 expression. The activation of Nrf2 pathway is a mechanism involved in ROS-protection by increased expression of antioxidant genes thus attenuating oxidative stress.
The overexpression of miR-140-5p after exposure to CIS may protect against CIS induced oxidative stress by activating Nrf2-dependent antioxidant pathway and provides a potentially therapeutic target in AKI. Kidney tissue Days 1, 3, 5, 7, and 14.
P53 promoted miR-199a-3p expression both in vivo and in vitro, which subsequently inhibited mTOR signaling. So, it might provide a promising therapeutic target of AKI. Kidney tissue Day 3.

Yang et al./2019 [42]
Cells (HK-2) RT-qPCR 24 h. Downregulated: miR-26a The upregulation of miR-26a (using miR-26a mimics) alleviated the CIS-induced injury via the downregulation of TRPC6. Overexpression of miR-26a could attenuate CIS-induced cell injury. Upregulation of miR-26a could restrain Drp1 expression (an important mediator in regulating mitochondrial fission), which was consistent with the changes in TRPC6 expression. This means that the renoprotective effects of miR-26a against CIS-induced cells injury were inhibited through the mitochondrial apoptosis pathway.
MiR-26a can protect CIS-induced HK2 cell apoptosis via negatively regulating TRPC6 expression and may be targeted for the prevention and treatment of drug-related AKI.
The expression level of miR-182-5p was raised in mouse kidney and HK-2 cells after cisplatin treatment. miR-182-5p was the target gene of PRNCR1. MiR-122 can be a direct suppressor of Foxo3 mRNA translation, while miR-34a activates Foxo3 by suppressing SIRT1. Increased expression and activation of Foxo3 has a role in triggering the p53 signaling pathway, culminating in cell apoptosis. Therefore, miR-122 and miR-34a dysregulation induces and actives Foxo3 contributing to CIS-induced acute tubular injury by fortifying the p53 signaling pathway.
The modulation of miR-122 and miR-34a could be a mechanism with which to prevent or treat AKI-induced by CIS. Overexpressing miR-186 could reverse the effects of cisplatin on NRK-52E cells proliferation and apoptosis. Moreover, inflammatory cytokines (IL-6, IL-1β, TNF-α, and Cox-2) expression was elevated by CIS; the increase of miR-186 reversed it, implying that increase of miR-186 repressed cell inflammatory response induced by CIS. ZEB1 was identified as miR-186 downstream target, which was found to be increased in AKI rat models. Knockdown of ZEB1 increased NRK-52E cell proliferation and restrained the apoptosis induced by CIS.
Loss of miR-186 expression contributed to CIS-induced AKI, partly through targeting ZEB1. MiR-186 might be provided an effective biomarker of AKI and a potential therapeutic target for its treatment.
Serum and kidney Days 0, 1, 3, and 5. Bioinformatics analysis showed that upregulated miR-3168 targeting genes of the ErbB signaling pathway, which target PDK, could downregulate the pathway, leading to CIS-induced apoptosis in renal cells. The regulation of genes involved in the mitochondrial apoptosis pathway may also contribute to higher nephrotoxicity, suggested by a decrease in the activity of the anti-apoptotic protein Bcl-2 by miR-3168 and miR-6125. Genes of the CIS detoxification pathway, which includes the conjugation of CIS with glutathione, are also shown to be target of miR-3168 and miR-6125, which could reduce the content of glutathione S-transferase and reduced glutathione.
The evidence suggests the baseline plasmatic expression of miR-3168, miR-6125, and miR-4718 as potential predictors of CIS-induced nephrotoxicity, with miR-4718 being the most promising marker.

Humans and Cells
Author ( Target prediction analysis of these miRNAs showed that the top pathway and associated pathological condition was found to be MYC-mediated apoptosis signaling and renal necrosis/cell death, respectively. In addition, they have several lapping targets including genes well-known in apoptosis as p21.
MiR-21, miR-200c, and miR-423 can be non-invasive and specific urinary biomarkers for the detection of drug-induced AKI in patients.
By comparing the nephrotoxicity profile presented by different individuals, animals, or cells undergoing the same treatment with cisplatin, it is possible to identify specific miRNA profiles, which can, therefore, be exploited to overcome diagnostic and therapeutic challenges. In this case, it is important to differentiate that some miRNAs were identified by the included studies as possible biomarkers of AKI, and others were identified as possible therapeutic targets (in accordance with the purpose of each study). Despite this, many miRNAs can function as biomarkers as well as potential targets. For example, the detection of miRNAs expressed in blood as AKI biomarkers could serve as a basis for the administration of a mimic miRNA or an anti-miR, regulating the pathway that is dysregulated in AKI, thus, functioning as a potential treatment.
The early detection of cisplatin-induced AKI is essential in the case of patients with cancer receiving cisplatin-based treatments. To this end, new biomarkers with adequate efficiency and sensitivity are needed. As miR-34a and miR-21 were the miRNAs that were the most frequently reported for this possible use in previous studies (six and five citations in different studies, respectively), the current discussion mainly focuses on them.
The use of miRNAs in cisplatin nephrotoxicity therapy aims to provide new approaches for the treatment of AKI through which miRNAs can be upregulated if they are related to anti-necroptotic or anti-inflammatory pathways, or they can be silenced if they activate pro-apoptotic genes. However, the ability of a single miRNA to regulate multiple genes in signaling pathways can be both beneficial and harmful. Caution is needed regarding off-target effects because there is a possibility that an anti-miR can inhibit other miRNAs with a common seed region [59]. Furthermore, since various miRNAs can target the same mRNA, the identification and validation of the downstream targets of miRNAs are unclear. Lastly, this therapeutic tool has some barriers, including the potential for degradation by RNases as well as the need for an efficient in vivo delivery system [22]. miR-34a was upregulated in all of the six included studies, whether it was in cells or animal urine and kidneys [27,30,34,35,47,52]. Unlike other members of the miR-34 family, miR-34a is ubiquitously expressed in normal human tissues. It is a p53-target gene, implying that miR-34a expression is transcriptionally regulated by p53; however, some studies have shown that miR-34a levels can be regulated by p53-independent mechanisms [60].
The induction of this miRNA by p53 targeting during DNA damage promotes apoptosis and cell cycle arrest [60,61]. Moreover, p53 activation is one of the mechanisms that contribute to renal cell death during cisplatin nephrotoxicity [62].
In a study by Lee et al., miR-34a was upregulated in the kidney tissue of mice after cisplatin treatment. This increase promoted the acetylation of Foxo3 by repressing the expression of SIRT1, an NAD-dependent deacetylase involved in the control of pro-apoptotic protein synthesis [63]. It was shown to be a bridge to p53-dependent apoptosis, and miR-34a is involved in this network controlling cisplatin-induced tubular injury. Indeed, the in silico analysis of the current study confirmed the prediction of SIRT1 as one of the most predicted targets of miR-34a.
The study by Lee et al. also showed two antagonistic mechanisms of apoptosis modulation during cisplatin treatment: (1) treatment with an miR-34a mimic enhanced the acetylation of Foxo3 and promoted the expression of p53 and Bax (a mediator of mitochondria-dependent programmed cell death), decreasing cell viability and (2) the miR-34a-antisense oligonucleotide transfection antagonized the ability of cisplatin to increase p53 and Bax levels and improved cell viability compared with the controls [35].
The activation of the miR-34a and p53 networks in response to cisplatin was also reported in a study by Pavkovic et al., where miR-34a was found to be upregulated, along with 17 other miRNAs, in rat urine over 26 days of investigation after cisplatin exposure [52]. The mRNA-predicted target analysis of these miRNAs was associated with various mechanisms known to be involved in cisplatin-induced nephrotoxicity [62,64], such as DNA damage response, apoptosis, cell cycle regulation, and inflammation [52].
Moreover, El Magdoub et al. showed the involvement of another pathway. They found that cisplatin treatment in rats upregulated miRNA-34a and induced the expression of transforming growth factor beta (TGF-β), a mediator of renal fibrosis that affects kidney function by stimulating extracellular matrix protein production [65]. As a consequence, TGFβR-1 was also induced and led to the stimulation of TAK1 [30]. Interestingly, TAK1 regulates cell necroptosis and has already been reported to regulate nuclear factor kappa B production [66], which was also elevated after cisplatin treatment [65].
However, a study by Bhatt et al. showed that blocking the induction of miR-34a with antisense oligonucleotides increased cell death during cisplatin treatment and exacerbated tissue damage, suggesting that this miRNA may play a cytoprotective role [34]. They also confirmed the involvement of p53 in miR-34a upregulation by cisplatin in cells treated with pifithrin-α (a pharmacological inhibitor of p53) as well as in p53-deficient mice. In both cases, the induction of miR-34a expression by cisplatin was completely or largely suppressed [34].
This difference between the findings of these studies (some indicating a protective role and others indicating nephrotoxicity induction) shows that the role of miR-34a in cisplatin-induced AKI is still controversial. One of the challenges in identifying new miRNA biomarkers is that many studies have reported conflicting data. miR-34a was primarily identified by the studies included in this systematic review as a potential AKI biomarker; nevertheless, the therapeutic use of miR-34a has already been investigated in a phase I clinical trial. As it is a known tumor suppressor that is not highly expressed in most malignancies, the trial treated solid malignancies, refractory to standard treatment, with an miR-34a mimic delivered intravenously through liposomes [67]. However, the study reported adverse effects and could not confirm whether they were related to the miR-34a mimic or the liposomal carrier. This raises concerns regarding the effects of miR-34a or other miRNA-mediated modulation strategies on AKI treatment. miR-21 was the second miRNA whose possible use as a biomarker of cisplatin-induced AKI was the most reported in previous studies. miR-21 is considered a promising biomarker for AKI because its expression is conserved in the kidneys [68], an organ in which it performs physiological and pathological functions [69]. Specific to the studies included in this systematic review, miR-21 was upregulated in four studies [33,46,52] and downregulated in one study [25].
In a study by Pavkovic et al., miR-21 levels increased in the cellular medium and decreased in the proximal kidney tubular epithelial cells. This change has been suggested as a hypothesis to explain the presence of this and other miRNAs in urine [46]. This is also the case for miR-15, miR-16, miR-20a, miR-192, miR-193, and miR-210, which were found to be upregulated in urine samples and downregulated in kidney tissue samples [52].
Various studies that did not involve cisplatin but that also evaluated the role of miR-21 as an AKI biomarker have previously been carried out. In a study by Du et al., in adult patients undergoing cardiac surgery, urinary and plasma miR-21 levels were associated with severe AKI [70]. In humans, increased urinary miR-21 expression has been associated with the progression of renal inflammation and fibrosis [71]. A third study using models of diabetic nephropathy showed that animals transfected with miR-21 knockdown plasmids showed improvement in microalbuminuria, fibrosis, and renal inflammation [72].
Contrastingly, some studies have shown that when miR-21 is upregulated in AKI, it plays a protective role by inhibiting the apoptosis and necrosis of renal tubular epithelial cells in response to stress in kidney injury renal ischemia-reperfusion [73,74]. Furthermore, it has been demonstrated to have therapeutic potential for the inhibition of fibrosis in pulmonary tissue by decreasing miR-21 levels [75].
Therefore, the results obtained to date indicate that the role of miR-21 must be associated with a high-precision adjustment mechanism to maintain its physiological equilibrium. Low miR-21 expression results in increased cell death, as miR-21 acts as an anti-apoptotic agent, while the overexpression of miR-21 can lead to severe inflammation and fibrosis [18].
Importantly, neither of the two miRNAs (miR-34a and miR-21) had their results validated in human cohorts. This reveals a gap in the literature, as results found in in vivo and in vitro models will not necessarily be similar to those observed in humans.
In silico analysis revealed 124 and 131 different strongly validated targets for miR-34a and miR-21, respectively (targets validated through robust experiments that, in addition to gene expression, evaluated protein expression and function, using analyses such as luciferase reporter assay and western blotting). Most of the targets were predicted by more than one type of experiment and more than one study.
For miR-34a, the most predicted targets were BCL2, MET, NOTCH1, and SIRT1; for miR-21-5p, they were PDCD4, PTEN, and RECK. Based on the results of this analysis, Supplementary Figures S1 and S2 were prepared for the two miRNAs, showing the relationships of their targets with various possible pathways involved in cisplatin nephrotoxicity.
BCL2 and SIRT1 have been shown to be involved in apoptosis regulation pathways, which are key processes in the pathophysiology of cisplatin nephrotoxicity [62]. In contrast, the MET and NOTCH1 genes, despite having been detected as possible targets, were not related to these selected pathways or the other two genes; they are involved in oncogenesis [76] and the development of numerous cell and tissue types [77].
As for miR-21, the main predicted targets were PDCD4, PTEN, and RECK. The PDCD4 and PTEN genes are involved in the regulation of apoptosis, the mitogen-activated protein kinase cascade, and protein phosphorylation; all of these processes are related to cisplatin nephrotoxicity [78]. The RECK gene, which is not related to any of these pathways, is mainly involved in embryogenesis and vasculogenesis [79].

Future Perspectives
One of the challenges in developing miRNAs biomarkers is the high cost of the techniques for miRNAs detection; therefore, more studies should be carried out on their costeffectiveness and cost-utility, evaluating the possible benefits of early care with the use of these miRNAs, either by improving the quality of life or the outcome of cisplatin treatment.
Furthermore, those miRNAs that showed good prospects in animal trials still have a long way to go before being implemented in the clinic, which includes having their results validated in humans and having sensitivity and specificity at least comparable to those of currently used biomarkers. In addition, a set of miRNAs is more likely to be more robust than a single miRNA [80].
Once these described implementation challenges are overcome, they could be useful in different contexts: they could be used in pretreatment tests (as a predictor of nephrotoxicity induced by cisplatin) or in post-treatment tests (as early markers that would detect nephrotoxicity earlier than currently used biomarkers).

Conclusions
The data presented in the studies included in this review show that miR-34a and miR-21 appear to be promising new biomarkers for cisplatin-induced nephrotoxicity. Despite this, validation studies in humans are still needed for the use of these miRNAs in clinical practice.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.