Drug Repurposing for the Management of Depression: Where Do We Stand Currently?

A slow rate of new drug discovery and higher costs of new drug development attracted the attention of scientists and physicians for the repurposing and repositioning of old medications. Experimental studies and off-label use of drugs have helped drive data for further studies of approving these medications. A deeper understanding of the pathogenesis of depression encourages novel discoveries through drug repurposing and drug repositioning to treat depression. In addition to reducing neurotransmitters like epinephrine and serotonin, other mechanisms such as inflammation, insufficient blood supply, and neurotoxicants are now considered as the possible involved mechanisms. Considering the mentioned mechanisms has resulted in repurposed medications to treat treatment-resistant depression (TRD) as alternative approaches. This review aims to discuss the available treatments and their progress way during repositioning. Neurotransmitters’ antagonists, atypical antipsychotics, and CNS stimulants have been studied for the repurposing aims. However, they need proper studies in terms of formulation, matching with regulatory standards, and efficacy.


Introduction
Despite the high rate of technological progress and improvements in knowledge of different diseases, the discovery of new medications demonstrated a lower speed [1].
As the regulatory requirements for bringing a new medication to the market is becoming more challenging to meet, medication cost increases globally [2].
Drug repositioning, also called drug repurposing or re-tasking, is a promising strategy to introduce new indications for other therapeutic goals for an available drug in the market [3]. Since the safety profile of these medications was studied thoroughly before, the development of their formulation has been analyzed, and the medicines successfully passed the preclinical and clinical steps, the risk of failure decreases significantly [4]. To apply a drug repositioning strategy, three main steps are needed to be concluded; first, a molecule/substance should be suggested for the mentioned indication; second, preclinical models, including animal and computational models, should be assessed and last the efficacy of medication should be analyzed [5].
As a significant mental disorder, major depressive disorder (MDD) affects approximately 264 million globally [6]. Due to the insufficient therapeutic response of patients to the available medications, the need for new medicines has attracted scientists worldwide [7]. Various mechanisms have been associated with the prevalence of MDD, and these mechanisms directly influence medication selection. Changes in inflammatory biomarkers, neurotransmitters, age-related, and genetic factors are among the mechanisms constituting the indications of medications [8]. Moreover, new antidepressant medications usually act on multiple intra-and extra-cellular markers, indicating their poly-pharmacology indications [9]. This review aims to provide a complete insight into the drug repositioning strategy, especially the candidates that could be beneficial in managing MDD.

History
Drug repurposing is the procedure of finding new indications for approved or investigational medications [10,11]. Investigational medicines may have had a desirable safety profile in phase I/II clinical trials [12] but never reached the market [11,12] due to reasons unrelated to safety [12], such as lack of efficacy [11]. Drug repurposing can emerge in different forms like repositioning, reformulation, and combination [13,14]. Drug repurposing was serendipitous and accidental in the past; whenever a medication had shown an off-target or a new on-target effect, it was investigated for commercial exploitation. [10]. So far, the most successful repurposed drugs have been found accidentally, and no systematic approach has been involved in the process [10,15]. Sildenafil citrate was an antihypertensive medication that got repurposed as retrospective clinical data analysis showed its positive effect on erectile dysfunction [10,16]. The year sildenafil got marketed by Pfizer for erectile dysfunction under the name Viagra ® [17], it held a 47% share of this problem's market, and the total sales were USD 2.05 billion worldwide [10]. Thalidomide is another well-known instance causing many severe skeletal birth defections in children whose mothers had taken this medication during the first trimester of pregnancy [1,10]. In consequence, thalidomide was withdrawn for four years [10]. In 1964, thalidomide was fortuitously recognized to be effective in erythema nodosum leprosum (ENL) [18]. Decades later, in 1999, thalidomide was discovered to be effective in multiple myeloma [19]. The positive outcome in multiple myeloma led to further derivative developments like lenalidomide [10]. Bupropion, an antidepressant medication got approved by the United States Food and Drug Administration (US-FDA) for smoking cessation [13], botulinum toxin A, the compound used for eye muscle disorders with cosmetic impacts [18] and minoxidil, the antihypertensive medication, became established for pattern hair loss in male and female [17,20] are some instances of well-known repurposed drugs [12]. Iproniazid, an antitubercular compound, was the first medication got reported for its antidepressant effect. This compound showed euphoria, psychostimulation, increased appetite, and improved sleep as the side effects [21]. The story of finding D-lysergic acid diethylamide (LSD) psychedelic effects is another exciting example of serendipitous discovery in the field of psychiatry. LSD was first synthesized in 1938, but it did not show considerable physiological effects in animal testing. LSD's strong and extraordinary influences on the mind were accidentally discovered for the first time in 1943 [22]. Thirteen medications have been repurposed for depression or bipolar depression treatment by 2017 [13].

Different Types of Drug Repurposing
Drug reformulation, repositioning, and combination are counted as different drug repurposing/repositioning [13,14].

Drug Repositioning
It finds new indications for a medication that already has other therapeutic indications [13,14]. For instance, mifepristone, an anti-progesterone drug with an initial indication for abortion, was experimentally effective in psychotic depression [3,23].

Drug Reformulation
It is about using a medication in a new dosage form [13,14]. The new formulation can both be taken via the same old route or a different route of administration. An example of drug reformulation is formulating ketamine for intranasal and sublingual routes to treat MDD [13].

Drug Combination
It refers to using two or more medications together [13,14] to improve efficacy and safety [13]. For example, using quetiapine and antidepressants in combination leads to the increased effect of antidepressant medications in the elderly suffering from MDD and cerebrovascular deterioration. In addition, taking anti-inflammatory medicines with antidepressants enhances responses to first-line antidepressants [13].

Common Approaches
Before the development stage in drug repositioning, three levels should be considered. The first level is discovering appropriate molecules for the desired indication, wherein most drug repurposing approaches are related to it. The second one is evaluating effects in preclinical experiments. The last is the appraisal of efficacy in phase II clinical trials, given that safety has been approved in phase I clinical trials for the original use [10].
Repositioning approaches are divided into two major groups: computational and experimental strategies [10,12,16]. These are growingly being used together [10,16] and will be separately discussed as follows. Explanations, pros, and cons of the drug repurposing approaches are summarized in Table 1. In addition, examples of each separate approach are indicated in Table 2.

Computational Approaches
Computational approaches consist of data analysis. These data can be obtained from different resources. For example, gene expression, chemical structure [10,12], or electronic health records (EHRs) can all be kinds of data resources [10]. In comparison with experimental approaches, computational approaches have lower expenses and fewer barriers [3].

Signature Matching
This method compares exclusive features (signature) of medication with another medication or disease. Three different types of data could be used as resources for extracting medication characteristics: chemical structure, adverse event profiles, proteomics, transcriptomic, and metabolomics, which are explored aptly in the following [10]: Transcriptomic: This technique compares gene expression in a healthy state, diseaseassociated state, and medication-using state. If a medication can reverse the expression pattern of the genes related to disease phenotype, it will probably also revert the disease phenotype itself [10,12,16,[23][24][25]. An example of this approach is that ketamine improves mood by modulating miRNAs like miR-598-5p and miR-451 [13]. Histone deacetylase (HDAC) inhibitors like vorinostat are promising drug repositioning targets for depression, anxiety and schizophrenia treatment due to their role in affecting gene expression [26]. Peroxisome proliferator-activator receptor (PPAR-γ) agonists, especially pioglitazone, have significant antidepressant outcomes in MDD and major depressive episodes of bipolar disorder due to their role in adjusting responsible gene expressions [13]. HMG-CoA reductase inhibitors (statins) are PPARα ligands that increase the expression of some neuronal growth factors. Randomized controlled trials have suggested that they possess beneficial effects in combination with selective serotonin reuptake inhibitors (SSRIs) [21]. Metabolomics: Metabolomics is the study of all chemical procedures in the body [27]. A drug can be shared between two different disease treatments with similar pathophysiology [5]. This approach helps us to gain a comprehensive idea about the molecular processes involved in disease pathophysiology and finding how close our preclinical models to reality are [28]. Nuclear magnetic resonance and mass spectroscopy are two methods for analyzing the metabolome [29]. Proteomics: Most medications apply their therapeutic effects by interacting with protein targets, and it is crucial to understand these interactions for drug development [12]. Chemical Structure: In this method, networks are made based on the shared chemical features [10] as similarity in chemical structure may lead to the same biological activity [10,16]. As an example, chlorcyclizine belongs to the phenylpiperazine class. This class includes many antipsychotic and antidepressant medications and chemical structure similarities between these medications and chlorcyclizine make it likely to possess the same effects [26]. Adverse Event Profiles: A hypothesis suggests that two different medications showing the same adverse effects might affect a shared target, protein, or pathway [10,16]. As well, a medication's adverse effect resembling a disease phenotype can imply a shared pathway or physiology between the drug and the illness [10,15,16]. In addition, if two treatments for one disease with different mechanisms demonstrate the same uncommon adverse effect, there may be a shared underlying mechanism that links adverse events and therapeutic effects [30]. Side effects are more helpful in predicting drug indications than chemical structure or protein targets. It is possible to extract adverse events data from chemical structures if a drug has not reached the clinical trial level [31].
Cabergoline, an ergot derivative dopamine agonist, showed delusion adverse events and got recognized to have antidepressant-like effects. Pergolide, another dopamine agonist, demonstrated antidepressant effects in Parkinson's disease. Modafinil is a narcolepsy medication that may be effective in depression treatment in combination with fluoxetine. Phenytoin, the famous anticonvulsant medicine, can be efficient in depression. This effectiveness is a result of hyperacusis, the phenytoin adverse effect [15].

Computational Molecular Docking
The basis of this approach is complementarity between ligand and target [10,23]. In conventional docking, the target involved in the disease is already known, and different medications get tested. In inverse docking, a set of targets are studied to check if they match particular medicines [10]. Computational molecular docking indicates that dextromethorphan shows an antidepressant effect with rapid onset of action during the first administration days due to involving glutamatergic receptors [13]. Cyproheptadine was hypothesized to improve depression based on its potential ability to be a serotonin receptor (5-HT2) antagonist [26]. Computational molecular docking suggests that mecamylamine, a nicotinic receptor antagonist, might be effective in depression treatment [23].

Genome-Wide Associated Studies (GWAS)
This method is proceeded on finding genetic variants associated with common diseases and understanding the biology of disease. In addition, these data can result in recognition of shared targets between conditions [10].

Pathway or Network Mapping
This approach is building networks based on signature matching data, protein interactions [10,25,32], gene expression pattern [10,25], disease pathology, or GWAS data [10] to find similarity or relation between medication and disease [25]. Some disease-associated genes are not appropriate druggable targets. Therefore, constructing and analyzing such networks could be a way to find upstream or downstream genes which can be used for drug repurposing [10]. Pathway mapping based on disease pathogenesis showed that nimodipine, a calcium channel blocker, makes antidepressants effective in old patients suffering from vascular depression. Scopolamine, the muscarinic antagonist, possesses rapid antidepressant effects since the cholinergic system is responsible for the pathogenesis of mood disorders [13]. Cannabidiol is a propitious agent for MDD due to its effects on involved pathways in this disorder. Sho-saiko-to, a traditional Chinese medicine, showed antidepressant effects in mice upon its influence on the serotonergic system in the central nervous system. Medications that inhibit p38 mitogen-activated protein (p38-MAPK) signaling pathways such as neflamapimod may have desirable effects on depression as the p38-MAPK pathway is engaged in many cellular processes, especially neuro-inflammation. Spermine is useful in treatment-resistant depression (TRD), as it is a glutamatergic receptor modulator. N-acetyl-l-cysteine (NAC), the glutathione precursor, positively affects the different mechanisms involved in depression and is a beneficial nutraceutical for adding to antidepressant medications in MDD treatment [21].

Experimental Approaches Retrospective Clinical Analysis
The main idea of this approach is reviewing and extracting valuable data from different resources such as EHRs, post-marketing surveillance, and clinical trials. This process would result in repositioning and using a medication for a dissimilar indication or finding an indication for a drug that had failed for its initial purpose [10].
EHRs: EHRs data are subdivided into structured (diagnosis and pathophysiology data, laboratory test results, and medication prescriptions) and unstructured (patients' symptoms reports and imaging data) groups [10]. Analyzing data gained from the national health insurance of Taiwan research database showed metformin is a promising target for drug repositioning for depression and anxiety [26]. The results of observational or case-control studies indicated a lower risk of MDD in patients using angiotensin-converting enzyme inhibitors (ACEIs) like telmisartan in comparison with other antihypertensive medications. Case reports also demonstrated that pramipexole, a relatively new dopamine receptor agonist, has potential effects in treating MDD. An observational study on 82,643 women revealed the relation between higher flavonoid intake and lower risk of MDD. Last but not least, a meta-analysis of 17 observational studies suggested the association between depression and zinc deficiency [21]. Post-Marketing Surveillance and Clinical Data: Retrospective clinical data analysis showed that using anti-inflammatory medications, especially celecoxib, with antidepressants improves the responses to first-line antidepressants. Clinical data claims that valproic acid enhances the effects of antidepressants in resistant depression patients. Based on clinical data, quetiapine co-administered with antidepressants improves the outcomes in patients with MDD and cerebrovascular deterioration [13]. Reviewing prior literature has shown that phenothiazines have anti-depressive effects resembling tricyclic antidepressants. Atypical antipsychotics can also help treat depression as adjunctive or primary therapy based on a meta-analysis [28]. Taking antidepressants combined with zinc caused decreased depressive symptoms than antidepressants alone in randomized control trials [21].

Novel Sources
This approach consists of three methods. The first one is using immortalized human cancer cell lines (CCLs) for screening different compounds to examine if pharmacological and genetic data match. The second method links EHR data to DNA biobanks; thereby, identifying the association between the patient's genome and the patient's illness. The third novel resource is patients' online self-reported data about their condition while taking medicine [10].

Binding Assays
Binding assessments help us realize target and ligand interactions. For example, affinity chromatography, mass spectroscopy, and cellular thermos ability assay (CETSA) techniques are three methods used in this approach [10].

Phenotypic Screening
This approach attempted to identify compounds showing effects of disease consequences in model systems without any earlier information about targets they affect [10].

Proteomics
Applying interaction between medication and proteome Pros: gaining information about safety, probable medication toxicity, mode of action of small molecule medications [12], involving more genetic and/or molecular level mechanisms in comparison to knowledge-based methods, finding new mechanisms of action [25,32] Chemical structure Developing networks based on shared chemical features Cons: the difference between actual results and expectations, variety of physiological effects despite structure resemblance [10], possibility of happening alterations in structure due to biological activity [16] Adverse event profile Finding shared targets, proteins, or pathways affected by different medications showing the same adverse effects Pros: unlike animal models, both therapeutic and adverse effects are observable in humans [15] Cons: problems in extracting information on medication package inserts, lack of proper adverse event profile and causality assessment [10] Computational molecular

Novel sources
Using immortalized human CCLs for screening different compounds, linking EHR data to DNA bank to identify an association between patients' genome and patients' illness, using patients' online self-reported data while taking medicine Pros: development of sequencing technologies, which helps ones collect more thorough information on each patient's genetics (2), acceleration in the drug discovery process, reduction in research costs, increase in patient involvement, ability to assess the effectiveness of the in-use medication (3) Cons: Happening alterations that make in vitro results better (1), challenges in using big data and technology for analysis (2), bias in collecting data, threat in patients' safety in case of self-prescription (3) [10] Binding assays Realizing interactions between target and ligand by using different methods as chromatography and mass spectroscopy -

Phenotypic screening
Identifying compounds showing disease consequences related effects in model systems, without any prior information about targets Pros: testing many medications for a therapeutic effect over a complete range of concentration [34], high flexibility for administration to numerous drugs or diseases [32] Cons: not reaching a complete picture by in vitro assays [33] CCLs: Cancer cell lines: FAERS: Food and Drug Administration adverse event reporting system; HGP: Human genome project; Ref: Reference; WHO: World Health Organization.

Advantages
Drug discovery is a high-cost and lengthy process. Bringing a new medication to market takes 13-15 years and costs USD 2-3 billion [10,21,25]. Although the expenses are increasing, the number of approved medications has remained constant or even decreased through the past years [22,25]. Moreover, demands in therapeutic fields are growing, and traditional drug discovery cannot answer these needs [13]. In the case of psychiatric medications, it should be noted that this field has not been developed enough over time [26]. Drug repurposing is cost-effective and reduces the time taken to get a new medication to market [10,11,21] as it costs on average USD 300 million and takes about 6.5 years [10,21]. The preclinical tests [10,16] and phase I and II clinical trials can be skipped in drug repurposing if these steps have already passed for other indications and safety has been approved [10]. This is why drug repurposing may shorten the time needed and expenses as mentioned above [1,10].
Furthermore, if the formulation is appropriate for the new indication, there is no need for formulation development. This can also be another helping hand for reaching the stated aims. Another great pro of drug repurposing is the lowered risk of failure due to approved sufficient safety [10,16]. At last, drug repurposing may uncover novel targets or pathways in treating a disease, which can be exploited further [10].

Barriers
Although toxicity and safety are not obstacles in drug repurposing, some barriers lead to failures, such as patent consideration, regulatory issues, and organization hurdles. In brief, many of the repurposed uses are already mentioned in the prior scientific literature or clinical data leading to limitations in patent protection. In addition, when an available generic formulation gets repurposed for a new indication, profitability reduces. This reduction happens due to off-label using the medication for novel indications. Governments make some rules for collaboration on patents that are near expiring to save public benefit. Creating a new formulation or dosage forms, developing new derivatives with the same therapeutic effects, or presenting medication in a new geographic region market are strategies for making a profit from the repurposed drugs [10]. Another trouble is that the effect of the medication is dependent on its dose. Therefore, it is necessary to identify the appropriate dose for novel indications during clinical trials [16]. Investments might be another obstacle in repurposing medications that have already failed during the drug development process. This trouble happens due to investors' unwillingness as they see medication's failure. In addition, medicines that failed in later stages of drug development have less time until patent expiration for repurposed try. Designing parallel development processes for different indications can lower the risk of failure [1] (Figure 1). Table 2. Examples of repurposed/suggested repurposing medications for different types of depression with their repurposing approach.

Management of Depression
In order to manage a patient diagnosed with MDD, two or three main options, including psychotherapy, pharmacotherapy, and somatic interventions, exist. Some guidelines suggest that those with moderate to severe depression would benefit from both psychotherapy and pharmacotherapy. In a mildly depressed person, treatment could be initially based on psychotherapy, and if needed, switching to medication could be applied after weeks [17]. Physical activity and exercise, balanced nutritional habits, improved sleep patterns, etc., can impact mental health and might be beneficial towards depressive disorders. Training like meditation, yoga, Tai chi, or daily journaling events is another helpful way to reduce stress, leading to the improved mental condition [53] (Figure 2).
In mild-to-moderate depression, psychotherapy has proved to be adequate and comparable to pharmacological therapies. Many experts suggest different types of psychotherapy like cognitive-behavioral therapy, behavioral activation therapy, and interpersonal psychotherapy. However, for severe forms of depressive disorder, antidepressant drugs have appeared to be much more effective by possessing a more rapid onset of action. Furthermore, psychotherapy is often used in those who have shown a response to antidepressants to prevent its relapse. Overall, the combination of both therapies is suggested and has been demonstrated to be more effective to either alone [48]. It is shown that the beneficial effects of different psychotherapy methods can last for at least one year after the treatment process. However, many people refuse this choice due to its high costs, lack of time, and the recent issue of Coronavirus disease-2019 (COVID-19). As solutions, attending some group psychotherapies to reduce the costs, setting online or over the phone sessions are valued, although all patients still do not believe in such ways [17]. Somatic intervention is another non-pharmacological option for the treatment of MDD. Electroconvulsive therapy (ECT) and repetitive transcranial magnetic stimulus (rTMS) are some noninvasive examples suggested to be beneficial for patients who have already failed at least one antidepressant trial. Vagus nerve stimulation is a US-FDA-approved surgical and invasive procedure for the management of TRD, which of course, carries its risks [53].

Pharmacotherapy
After a patient is diagnosed with MDD, a psychiatrist can start pharmacologic treatment to symptom remission. Antidepressants take approximately 3-4 weeks to exert their effects, although there is always the risk of relapse or recurrence of mood episodes even after the therapy. Choosing the first-line treatment for a patient depends on multiple factors, including age, concurrent medical conditions or psychiatric state, adverse effect profiles of the drug and its interactions, ease of access, cost, convenience and patient's preference, safety in overdose, etc. Another essential issue is the patient's initial responses to antidepressants (if taken) and the family history [53].

Repurposed Drugs for MDD
Plenty of repurposed agents for depression are studied or have gotten US-FDA approval for use in the clinic from different pharmacological categories. These categories vary from some central nervous system (CNS)-related medications like the second generation (atypical) antipsychotics, NMDA receptor antagonists and anesthetics, GABA receptor modulators, dopamine agonists, anticholinergic agents, CNS stimulants, anticonvulsant agents, histamine antagonists and ergot derivatives to even some unrelated ones such as thyroid products, antidiabetic agents, anti-inflammatory agents, antibiotics, HMG-CoA reductase inhibitors, calcium channel blockers, angiotensin-converting enzyme inhibitors, antineoplastic agents and some nutritional supplements. These agents exert their antidepressant effects through various pathways due to their pharmacological category. The complete list of repurposed drugs for MDD with a particular focus on their mechanism of action, significant adverse effects, contraindications, and dosages (if available) are provided in Table 3. Moreover, related clinical trials studying their effects on MDD in real-world settings are summarized in Table 4.

Conclusions
MDD, as a severe mental disorder declining the quality of life of the patient, requires on-time screening and management. Psychotherapy, pharmacotherapy, and somatic interventions are among the suggested managements. However, due to the incomplete response of the patients to the approved medications, physicians tend to prescribe medications on their off-label use. Hence, a great need for a drug repositioning method took place for the MDD management medications. Drug repositioning is a cost-effective method that decreases the required time to introduce medicine to the market.
Moreover, as there are available data on the safety profile of the medications, the risk of failure decreases significantly, and this method is capable of uncovering novel targets to treat a disease. Various pharmacological categories, including neurotransmitters' antagonists, atypical antipsychotics, and CNS stimulants, have been studied for the repurposing aims. However, proper studies on the formulation, regulatory, and efficacy of the medication are required to better this approach.