Are current pharmacotherapeutic strategies effective in treating OCD?

KEYWORDS:

• Obsessive-compulsive disorder
• pharmacotherapy
• SSRI
• clomipramine
• augmentation
• antipsychotics
• glutamatergic drugs

1. Introduction

Obsessive-compulsive disorder (OCD) is a common mental illness that has a lifetime prevalence of 2–3% and often runs a chronic course with waxing and waning of symptoms [1]. OCD is associated with significant psychosocial impairment, poorer quality of life, and disability. There have been significant advances in the treatment of OCD in the last three decades. First-line treatments for OCD include specific serotonin-reuptake inhibitors (SSRIs) and cognitive-behavior therapy (CBT) [2]. Subsequent to the advent of serotonergic drugs in the treatment of OCD, other classes of drugs such as the ones that modulate dopaminergic and glutamatergic neurotransmitter systems have also been studied. In this article, we examine critically whether pharmacotherapeutic strategies are effective in treating OCD.

2. SSRIs and clomipramine

Clomipramine, an SSRI, was the first drug shown to be effective in treating OCD and remains the gold standard. Subsequently, SSRIs (fluoxetine, fluvoxamine, sertraline, paroxetine, citalopram, and escitalopram) were examined for their efficacy in treating OCD. Meta-analyses of randomized, double-blind, and placebo-controlled studies have demonstrated that SSRIs and clomipramine are effective in treating OCD with a mean difference in the Yale–Brown Obsessive Compulsive Severity Scale (YBOCS) scores ranging from 3.49 (1.81–5.12) to 4.72 (2.60–5.85), respectively, as compared to drug placebo [3]. Although earlier meta-analyses claimed superiority of clomipramine over the SSRIs [4,5], a recent network meta-analysis has demonstrated similar effect sizes for SSRIs and clomipramine after adjusting for publication year and other characteristics of the population [3]. Given the similar efficacy and higher adverse effect burden, clomipramine is generally not recommended as a first-line treatment [6]. There is also no evidence to support the superiority of one SSRI over the other [3], although it must be mentioned that head-to-head comparisons are few. The SSRIs and clomipramine are usually administered in higher doses and for a much longer period (about 10–12 weeks) than in depression to achieve an optimum response.

Evidence supporting the efficacy of medications also comes from the placebo-controlled discontinuation studies. A large proportion of patients who respond to medications tend to relapse upon discontinuation, mostly within 6 months, suggesting anti-obsessional property of medications [6]. Naturalistic follow-up studies of patients mostly treated and maintained on SSRIs/clomipramine alone also show high rates of remission, implying continued anti-obsessional property of these medications and their beneficial effect on the course of illness (for detailed review, refer [7]). Viewed collectively, SSRIs/clomipramine confer protection against relapse as long as they are continued.

Although both SSRIs and CBT are considered as first-level options in treating OCD, there are no data to support the superiority of the one over the other as monotherapies in patients who have never been treated. Over 80% of CBT trials included patients who continued to receive SSRIs/clomipramine [3]. This suggests a possibility of a synergistic or additive effect of medications and CBT, although some studies suggest no additional benefit of adding medication to CBT [8].

Although SSRIs and clomipramine are shown to be effective in many well-designed, randomized, double-blind controlled studies, it is now widely accepted that 40–60% of patients do not show satisfactory response [9]. Poor insight into symptoms, presence of symmetry/hoarding, and certain personality disorders such as obsessive-compulsive and schizotypal personality disorders may predict poor response to SSRIs [10]. Response is typically defined as a reduction in the total YBOCS score of at least 25–35% from pretreatment score. Even with this liberal definition of response, the number needed to treat is 6 compared to placebo [11]. Many patients who show response to medications continue to have residual symptoms or symptoms severe enough to still qualify for a clinical diagnosis of OCD. For example, a patient with severe OCD with a total YBOCS score of 32 (out of a total of 40) may become a responder to an SSRI as per the standard definition of response but may continue to have troublesome symptoms. While many patients may respond to medications, most fail to achieve remission (YBOCS <12). Since many patients classified as responders continue to have clinically significant symptoms and fail to achieve remission, efforts have been underway to look at strategies to augment the effect of SSRIs both pharmacologically and psychotherapeutically. Patients who achieve partial remission (rather than those who achieve full remission) are also vulnerable to relapse [12].

3. Is there a role for a serotonin–norepinephrine reuptake inhibitor (SNRI)?

Venlafaxine, an SNRI, has shown some promise in treating OCD [13,14], including in treatment-resistant samples [15]. We examined the effectiveness of venlafaxine from the large database of our OCD clinic in which 65 patients continued to be clinically ill despite adequate treatment with at least two SSRIs/clomipramine [15]. A majority of them (83%) had even failed to respond to antipsychotic augmentation and about a third to even CBT. After 4 months of treatment with venlafaxine (mean dose, 225 mg/day), 45% were classified as responders (35% or more reduction on the YBOCS and the Clinical Global Impressions-Improvement Scale rating of ‘very much improved’ or ‘much improved’). Venlafaxine may be a useful option in patients who do not respond to SSRIs, but more systematic research is needed to establish its role in SSRI-resistant OCD.

4. Is antipsychotic augmentation of SSRIs effective?

Interest in antipsychotic augmentation of SSRIs stems from the dopaminergic overactivation theory in OCD, and they are recommended as the first-line augmenting agents in OCD. Several meta-analyses [16–18] have concluded that antipsychotics are effective augmentation agents, but they differ in their conclusion as to which of them are effective and how effective they are. Two influential meta-analyses concluded that aripiprazole and risperidone significantly outperform placebo [16,17], but the study by Veale et al. was very cautious in its recommendation of antipsychotics as augmenting agents because of small effect sizes and long-term side effects. A recent network meta-analysis concluded that antipsychotics as a class were effective as augmenting agents if added after at least 12 weeks of SSRI treatment and in those without comorbid depression and tics. The study could not recommend which drugs are most effective because of limited number of studies and heterogeneity across studies. Interestingly, they found quetiapine and olanzapine to be significantly superior to placebo, but risperidone lost significance after adjusting for baseline symptom severity [18]. Enthusiasm in the efficacy of antipsychotics as augmenting agents is dampened to an extent because of the superiority of CBT over risperidone and pill placebo in a randomized clinical trial that compared the efficacy of CBT vs. risperidone for augmenting SSRIs [19]. In fact, risperidone was not different from placebo. In two real-world effectiveness studies of risperidone (n = 93) and aripiprazole (n = 23) as augmenting agents, both were found to be effective in 25–30% of the patients [20,21]. Clozapine, an antipsychotic with predominantly anti-serotonergic activity, may induce OCD symptoms in patients with schizophrenia [22]. Hence, it is not recommended as an augmenting agent for OCD. Overall, it appears that antipsychotics may be effective as augmenting agents in a proportion of patients but may not be superior to CBT. However, due consideration should be given to their long-term adverse effects that include metabolic, endocrinal, and extrapyramidal side effects.

5. Glutamate-modulating drugs as potential therapeutic options

Glutamate is the primary neurotransmitter within the cortico-striato-thalamic circuits implicated in OCD. In view of the implication of the glutamatergic signaling in the pathophysiology of OCD, there has been great interest in the role of glutamate-modulating drugs in the treatment of OCD (riluzole, memantine, N-acetylcysteine, ketamine, glycine, topiramate, lamotrigine, minocycline, and D-cycloserine). The literature on the efficacy of glutamate-modulating drugs is limited, and studies have mostly small sample sizes [23]. A recent network meta-analysis found memantine, topiramate, and lamotrigine to be significantly superior to placebo [18]. Despite some inconsistent evidence [24], memantine [18] and N-acetylcysteine [25] have the best evidence base as augmenting agents. The literature on riluzole is limited [26]. Ketamine remains an experimental treatment in view of its unproven efficacy and adverse effect profile. The antibiotic D-cycloserine (DCS), a partial agonist at N-Methyl-D-Aspartate (NMDA) receptors in the amygdala, has been studied for pharmacological augmentation of learning of fear extinction, a central mechanism in exposure-based CBT. It was thought to enhance fear extinction when administered close to exposure therapy in low doses and a limited number of times, but a recent meta-analysis indicated that the overall effect of DCS is very small raising the issue of the clinical utility of DCS augmentation of exposure therapy [27]. Overall, of the glutamate-modulating drugs, memantine has the best evidence base, but larger controlled trials are required to establish their efficacy as augmenting agents.

6. Other augmenting agents

The 5-HT3 receptor antagonists (ondansetron and granisetron) that are primarily used to reduce nausea and vomiting in cancer chemotherapy have been studied as augmenting agents in OCD. Through 5-HT3 blockade, these drugs have a weak downstream inhibitory effect on dopaminergic neurotransmission. Ondansetron has been found to be effective in both uncontrolled and placebo-controlled trials [28,29], but the experience with ondansetron in our clinic has not been encouraging [30]. At best, 5HT-3 antagonists may be viewed as experimental treatments needing confirmation in larger studies. Other drugs such as pregabalin and methylphenidate have also been examined in a few randomized controlled studies but need replication. There is preliminary evidence on immunotherapeutic interventions such as antibiotics, immunoglobulins, and anti-inflammatory agents [31]. There is a need for large-scale studies to understand their role in the treatment of OCD. Table 1 summarizes the level of evidence for pharmacotherapeutic strategies in treating OCD.

Table 1. Evidence-based pharmacotherapeutic strategies for obsessive-compulsive disorder.

Drug	Recommended daily dose	Level of evidence^a
Primary agents^b
I. Selective serotonin-reuptake inhibitors
Escitalopram	20–30 mg	A
Fluoxetine	60–80 mg	A
Fluvoxamine	200−300 mg	A
Paroxetine	40–60 mg	A
Sertraline	150–200 mg	A
Citalopram	40–60 mg	A
II. Tricyclic antidepressants
Clomipramine	150–250 mg	A
III. Serotonin–norepinephrine reuptake inhibitors
Venlafaxine	225–300 mg	B
Augmenting agents
I. Dopaminergic agents
Aripiprazole	5–10 mg	A
Risperidone	1–2 mg	A
Quetiapine	100–300 mg	B
Olanzapine	5–10 mg	B
Haloperidol	2–4 mg	B
II. Glutamatergic agents
Memantine	20 mg	B
N-acetylcysteine	1800–3000 mg	B
Lamotrigine	100–200 mg	B
III. 5-HT3 antagonists
Ondansetron	4–8 mg	B
Granisetron	1–2 mg	B

^aBased on modified Strength of Recommendation Taxonomy (SORT) [35].

A – Consistent, good-quality patient-oriented evidence.

B – Inconsistent or limited-quality patient-oriented evidence.

C – Consensus, disease-oriented evidence, usual practice, expert opinion, or case series.

^bEvidence for use as monotherapy in OCD.

7. Expert opinion

The SSRIs are consistently shown to be effective in treating OCD in both children and adults, and they are the first-line pharmacological options. However, there are several unresolved clinical issues that need to be addressed for optimum utilization of these medications. First, although they all seem to be equally efficacious in meta-analyses, there are not many head-to-head trials to conclusively say that they are not differentially efficacious. Second, we do not know the relative efficacy of medications vs. CBT vs. their combination in never-treated patients. A network meta-analysis found the combination treatment to be the most efficacious but with high uncertainty [3]. Guidelines seem to prefer CBT or a combination of CBT and medications over medications alone as first-line treatment [2]. This recommendation is based mostly on the data derived from patients who are already on medications [3]. Third, there is a need to identify early predictors of response to medications rather than wait for a full trial of 12 weeks. There is preliminary evidence that early response to medications predicts later response [32]. Fourth, SSRIs/clomipramine do not induce persistent and full remission in most patients. Those who respond to these medications continue to remain symptomatic, albeit to a somewhat lesser degree [33]. Persistence of symptoms contributes to poorer quality of life and psychosocial dysfunction. Added to this, since most clinical trials use symptomatic improvement as an outcome measure, there is insufficient information as to what proportion of patients show functional recovery and improved quality of life. It is also known that partially remitted patients are at a higher risk of relapse despite being on medications. In essence, for all practical purposes, SSRIs and clomipramine are only partially effective in about half of the patients, and the remaining patients fail to respond. Further, several pharmacological augmenting strategies for SSRIs have been employed, but findings are rather disheartening. Even the first-line augmenting agents such as atypical antipsychotics are only modestly effective in about 25–30% of the patients. We also need data on their long-term efficacy and safety. Much of the enthusiasm related to glutamate-modulating drugs as augmenting agents has not yielded consistently favorable results. Many other agents such as pregabalin, tolcapone, psilocybin, cannabinoids, and nutraceuticals are considered as promising agents in treating OCD, but additional systematic data are needed. The efficacy of multimodal antidepressants with serotonin-reuptake properties such as vortioxetine and vilazodone has to be evaluated. There is also emerging evidence for various brain stimulation techniques as augmenters [34]. There is a need to evaluate their dynamic interaction with ongoing pharmacological and psychotherapeutic treatments.

What is most needed now is to identify robust, clinical, neurobiological, and genetic predictors of response to SSRIs/clomipramine and to augmentation strategies. A first step toward this is to recognize that OCD is not a unitary disorder and deconstruct the complex clinical phenotype into homogeneous dimensions and examine the phenotypic characteristics that may contribute to response or lack of response to various pharmacological options. There is emerging evidence for such sub-characterization based on clinical phenotypes (e.g. age at onset, symptom dimensions, insight), presence of comorbidity (e.g. tic disorder), and other biological characteristics (e.g. genetic markers, neuropsychological deficits). Future treatment strategies should capitalize on emerging knowledge on genetics (including epigenetics, pharmacogenomics), environmental risk factors, and neural basis of various symptom dimensions in OCD. These approaches may pave the way for personalized medicine in treating OCD.

Declaration of interest

YCJ Reddy has received funding support from the National Institute of Mental Health (NIMH) USA, the Department of Biotechnology (DBT), the Department of Science and Technology (DST), and the Indian Council of Medical Research (ICMR) of the Government of India and the Canadian Institute of Health Research (CIHR). Furthermore, SS Arumugham has received funding support from Wellcome Trust/DBT India Alliance. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has not been funded.