STEPS for cost-effective prescribing of serotonin-norepinephrine reuptake inhibitors (SNRIs) in major depressive disorder: Focus on venlafaxine (Effexor®) versus duloxetine (Cymbalta®)
Selection of an antidepressant medication is based on various factors, such as safety, tolerability, and cost. This article reviews the cost-effectiveness of the different serotonin-norepinephrine reuptake inhibitors (SNRIs).
BACKGROUND/INTRODUCTION
Between 2005–2008, antidepressants were the third most common prescription drugs taken by Americans of all ages and were also the most frequently used prescription drug class by middle-aged adults.1 There are substantial direct and indirect healthcare costs associated with Major Depressive Disorder (MDD), with depression costing the U.S. an estimated $44 billion each year.23 Selection of the most cost-effective antidepressant(s) for each individual patient in this uncertain economic climate should be of continued importance not only for health care organizations, but to decrease out-of-pocket expenses for patients as well.
In general, effectiveness of antidepressant medications for MDD is comparable between classes and within classes of medications.4 Therefore, initial selection of an antidepressant is based on other factors such as safety, tolerability, and cost.4 Because of their improved safety and tolerability profiles over older agents, the selective serotonin reuptake inhibitors (SSRIs), serotonin norepinephrine reuptake inhibitors (SNRIs), bupropion, and mirtazapine are usually preferred treatment options over tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs).4 Of these, SSRIs are usually utilized first line given their favorable safety-tolerability profile and generic availability. Compared to SSRIs, SNRIs are still costly, with limited generic availability. There are five SNRIs approved for use in the US which include venlafaxine (Effexor®), desvenlafaxine (Pristiq®), duloxetine (Cymbalta®), milnacipran (Savella®), and recently approved (July 2013) levomilnacipran (Fetzima®). Of these, duloxetine is one of the top ten prescribed medications in the US and had total sales in 2011 of $3,666,405,000.56 Given variation between agents in the class regarding secondary pharmacologic properties, level of evidence for different indications (FDA approved and off-label), and cost, we will review the STEPS (Safety, Tolerability, Effectiveness, Price, and Simplicity) to help select the most cost-effective SNRI based on patient specific factors.7
SAFETY
SNRIs are generally safe and well tolerated. Some key differences in safety are in overdose, patients at risk of QT prolongation, potential for hepatotoxicity, and effects on blood pressure.
Overdose
In studies, venlafaxine has been shown to confer lower risk of fatal outcomes from overdose than TCAs, but higher risk than SSRIs.89 Within the SNRI class, venlafaxine appears to have the highest risk in overdose, possibly related to cardiotoxic effects.10 Clinical experience with overdose of venlafaxine's active metabolite, desvenlafaxine, in humans is limited, with no cases of fatal overdose reported in pre-clinical trials.9 Fatal outcomes of duloxetine overdose have been primarily reported with co-ingestants, but there are also case reports when fatal duloxetine overdose was the sole agent consumed.611 Significant toxicity from milnacipran overdose is rare. In clinical trials there were no fatalities reported with acute ingestions of up to 1000 mg daily of milnacipran, alone or in combination with other drugs.1213 Clinical experience with overdose of the newest SNRI levomilnacipran is limited in humans.14 Ingestions of up to three times the maximum recommended daily dose (360 mg) of levomilnacipran were reported with no fatalities in clinical studies.14
QT Prolongation
There has been growing concern regarding antidepressant medications and risk of sudden cardiac death.15 Venlafaxine can cause prolongation of the QT interval and is the only SNRI listed to avoid in patients with congenital long QT syndrome on Azcert.org (a resource for the safe use of medications with focus on drugs that prolong the QT interval and cause sudden death).1617 Case reports have shown that QT prolongation with venlafaxine can occur with overdose or when multiple additional risk factors are present.18 Desvenlafaxine has low potential to prolong the QT interval.19 In a study, desvenlafaxine at doses of 200 mg and 600 mg did not cause prolongation of the QT interval.19 Generally, duloxetine is thought to have little to no effect on the corrected QT interval (QTc).920 In a study of 117 healthy females, doses of 200 mg BID of duloxetine did not significantly prolong the QTc interval.20 A double-blind study looking at the effect of milnacipran on the QTc interval did not find a clinically significant change.13 Levomilnacipran was not found to significantly prolong the QT interval at 2.5 times the maximum recommended dose.14
Hepatotoxicity
Post-marketing reports of hepatic failure, sometimes fatal, have been reported in patients treated with duloxetine. These cases have presented as hepatitis with abdominal pain, hepatomegaly, and elevation of transaminase levels to more than twenty times the upper limit of normal.61121 These reports led Eli Lilly, the manufacturer of duloxetine, in October 2005 to issue a letter regarding updated safety information on possible hepatotoxicity with duloxetine to healthcare professionals.21 In addition, some of these reports indicated patients with preexisting liver disease may have an increased risk for further liver damage with duloxetine use.21 Overall, duloxetine is not recommended to be used in patients with any hepatic insufficiency.61122 Milnacipran also carries a warning for possible hepatotoxicity and is not recommended for use in chronic liver disease as it can exacerbate the condition.613 Milnacipran can possibly increase liver enzymes, with post-marketing reports of severe liver injury, particularly in patients with significant underlying clinical conditions and/or use of multiple concurrent medications.613 There is one published non-fatal case of hepatitis with venlafaxine, with complete resolution of liver function abnormalities after discontinuation.23 The current FDA prescribing information for desvenlafaxine, and levomilnacipran do not mention risk for hepatotoxicity.1419
Blood Pressure
Given SNRIs can cause an increase in blood pressure, the patient's vital signs (blood pressure and heart rate) should be monitored at baseline and during treatment.9 Consider a dose reduction or discontinuation of the SNRI in any patient experiencing sustained blood pressure increase after initiation.46 Venlafaxine can cause a dose-dependent increase in blood pressure, especially at doses above 150 mg/day.4 This effect may be less frequent with other SNRIs such as desvenlafaxine, duloxetine and milnacipran.423 Levomilnacipran can also increase blood pressure and has the same precautions as other SNRIs approved for MDD.614
TOLERABILITY
Tolerability is a key factor to consider. Even if an antidepressant is effective, the inability of a patient to tolerate it can lead to poor compliance, which affects remission rates, cost, and puts the patient at risk of relapse.24 Generally all SNRIS are well-tolerated, with most adverse effects of mild-to-moderate severity improving with continued treatment.23
A study which combined data from two multicenter, randomized, double-blind studies looked at global benefit–risk (GBR; assessment allows the simultaneous evaluation of both efficacy and adverse events) in patients with major depressive disorder who were randomized to either duloxetine 60 mg/day or venlafaxine extended-release (XR) 150 mg/day for 6-weeks at fixed doses, followed by an additional 6 weeks of treatment in which the dose could be increased up to 120 mg/day for duloxetine and 225 mg/day for venlafaxine. They found significantly more venlafaxine-treated patients (74.5%) completed 12 weeks of treatment compared with duloxetine treated patients (64.8%, P =.006).25 Nausea was the most common treatment-emergent adverse event (TEAE) for both drugs, and was significantly higher with duloxetine 60 mg/day compared to venlafaxine 150 mg/day during the first 6 weeks of treatment (43.6% vs. 35.0%, P = 0.05).25 Significantly more venlafaxine treated patients reported discontinuation-emergent adverse events (DEAEs) during the taper period and sustained elevations of systolic blood pressure during the fixed dosing period than duloxetine treated patients.25 The study used extended-release venlafaxine, which is usually the preferred formulation over immediate-release due in part to a better tolerability profile.4
EFFECTIVENESS
Major Depressive Disorder
The SNRIs that are FDA approved for treatment of MDD include venlafaxine, desvenlafaxine, duloxetine, and levomilnacipran (see Table 1).6 Milnacipran is only FDA approved in the US for treatment of fibromyalgia, but has been studied for MDD.6 In a 24 week randomized, double-blind exploratory study looking at milnacipran and venlafaxine at flexible doses (up to 200 mg/day) in the outpatient treatment of adults with moderate to severe major depressive disorder, the authors found the overall safety/tolerability and efficacy profiles to be similar.26 The newest SNRI, levomilnacipran is the more active enantiomer of milnacipran. Levomilnacipran was found to be effective for MDD in three 8-week, randomized, double-blind, placebo-controlled studies. In all three studies, levomilnacipran demonstrated superiority over placebo in improvement of depressive symptoms as measured by the Montgomery-Asberg Depression Rating Scale (MADRS) total score and the Sheehan Disability Scale (SDS) functional impairment total score.14 In vitro studies show levomilnacipran may be the most noradrenergically active of the SNRI class of antidepressant drugs FDA Indicated for MDD.2728 In general, levomilnacipran is thought to have a 2 fold greater potency for norepinephrine (NE) relative to serotonin (5-HT) and 17–27 times higher selectivity for NE reuptake inhibition compared to venlafaxine and duloxetine.28
A Cochrane review, which included a total of three studies with venlafaxine and one with desvenlafaxine compared to duloxetine, found no statistically significant differences in efficacy between them and duloxetine. However, when compared to venlafaxine, there was a higher rate of drop out due to any cause in the patients randomized to duloxetine. The authors concluded that duloxetine did not seem to provide a significant advantage in efficacy over other antidepressants for the acute-phase treatment of major depression. No differences in terms of efficacy were found, even though duloxetine was worse than some newer antidepressants (like venlafaxine) in terms of acceptability and tolerability.29
Co-Occurring Pain
Depression and pain are two of the most frequent co-occurring conditions. Pain can lead to insufficient antidepressant response and, in turn, depression can cause poor treatment outcome and prognosis in pain patients.30 Alterations in serotonergic and noradrenergic neurotransmissions in the central nervous system (CNS) are thought to have a role in the pathophysiology of both chronic pain and depression.30 According to the American Psychiatric Association (APA) depression treatment guidelines, SNRIs and TCAs may be preferable to other antidepressants in patients with co-occurring depression and chronic pain.4
Fibromyalgia is frequently associated with symptoms of depression.30 Duloxetine, milnacipran, and venlafaxine all work on both NE and 5HT and have been studied in fibromyalgia. Of these, duloxetine and milnacipran have FDA approved indications for use in fibromyalgia and have been studied in multiple randomized trials.631 The literature on the use of venlafaxine for this indication is limited. Dwight et al., in an open-label, small study looking at venlafaxine for the treatment of fibromyalgia found that 55% (6 of 11) of the patients that completed the study experienced a ≥ 50% decrease in their fibromyalgia symptoms. Also, the presence of chronic psychiatric disorders (in particular, anxiety and depressive disorders) predicted a positive response to venlafaxine.32 In another small (N=15) open-label study looking at patients with fibromyalgia before and after treatment with venlafaxine, the authors found significant improvement in pain intensity and disability caused by fibromyalgia.33 Although, in contrast to the Dwight et al. study, they found that neither past nor current psychiatric disorders predicted treatment response with venlafaxine use in fibromyalgia.3233
In evidence-based guidelines, TCAs and SNRIs are antidepressants used for treatment of neuropathic pain.4
In a Cochrane review for neuropathic pain, the number needed to treat was found to be comparable between the SNRI venlafaxine and TCAs.3435 In general, TCAs are first-line treatments for diabetic peripheral neuropathy.31 Given their tolerability profile, a SNRI may sometimes be chosen before a TCA for a patient with co-occurring depression and neuropathic pain.4 Of these, duloxetine and venlafaxine have both been studied. The use of duloxetine for this indication is FDA approved and its efficacy was established in 2 randomized, 12-week, double-blind, placebo-controlled, fixed-dose studies.11 Venlafaxine ER at doses of 75mg and 150–225 mg, studied in a multicenter, double-blind, randomized, placebo-controlled trial in patients with painful diabetic neuropathy, was found to have significant benefit at higher doses (150–225 mg), but not lower doses.36 This may be explained by venlafaxine having more of a balanced noradrenergic-to-serotonergic effect at higher doses.35 Therefore, titration of venlafaxine ER up to higher doses (if tolerated) for efficacy in diabetic peripheral neuropathy for dual serotonin-norepinephrine reuptake inhibition is likely to be required.
PRICE
Given the lack of clinical superiority of one antidepressant over another for MDD, initial selection of an antidepressant should be made in part, based on cost considerations. Generic antidepressant medications create an opportunity to lower cost for patients (i.e., brand versus generic co-pay) and health maintenance organizations (HMOs).37 Venlafaxine ER is available as a generic and is the least costly per dosage unit of the SNRIs that are FDA approved for MDD. Its active metabolite, desvenlafaxine (Pristiq) is the only other SNRI available as a generic, but is still more costly per dosage unit than venlafaxine ER.6 These costs are listed in Table 2 and are based on average wholesale price (AWP) from the Red Book.6 The AWP is often higher than pharmacies, institutions, and/or consumers pay for a drug. Therefore depending on contracting, there may be more or less of a cost-savings between SNRIs. In addition, duloxetine is expected to be available as a generic in December 2013 and therefore the cost may dramatically decrease with generic availability.
Pharmacoeconomic Outcome Studies
A meta-analysis looking at the cost effectiveness of duloxetine compared with venlafaxine XR for MDD was conducted using a decision tree modeling outpatient treatment over 6 months. Pharmacoeconomic outcomes from the perspective of the payer (Ministry of Health of Ontario) for expected cost, expected cost per success, and expected cost per symptom-free day tended to favor treatment with venlafaxine XR. From a societal perspective, modest differences in all pharmacoeconomic outcomes tended to favor treatment of MDD with venlafaxine XR.38
Benedict and colleagues looked at the cost-utility of duloxetine versus venlafaxine ER, commonly used SSRIs, and mirtazapine in primary and secondary care in Scotland. The study used a decision analytic Markov model from the National Health Service (NHS) perspective. Model outcomes included total treatment costs and quality adjusted life-years (QALYs). QALYs are a measure of benefit/effect used in cost-effectiveness studies making it possible to compare results across different types of interventions within a therapeutic area and across different illnesses. The authors stated that duloxetine has the probability of being the cost-effective treatment of choice when compared to SNRIs, mirtazapine, and SSRIs for moderately severe MDD in primary care in Scotland as long as health care authorities are willing to pay £6000 or more per QALY gained from duloxetine use compared to the others. They concluded that duloxetine with similar efficacy, is marginally more effective and less costly than venlafaxine ER.2 The acquisition cost of duloxetine in the study may have been less than venlafaxine XR given it was conducted outside of the US. The study was funded by the manufacturer of duloxetine, Eli Lilly.
SIMPLICITY
Usually venlafaxine ER, desvenlafaxine and levomilnacipran are dosed once daily, duloxetine once or twice daily, and milnacipran twice daily.6 Once daily dosing of venlafaxine ER can simplify the patient's regimen over 2–3 times daily dosing of venlafaxine IR.
Dose consolidation not only can help improve patient compliance, but can be a cost-savings tool as well. Even though duloxetine has a mean plasma half-life of 12 hours, it can be dosed once daily given the central nervous system half-life may be very different from the plasma half-life.39 The package insert also states that duloxetine can be given once daily for both psychiatric and pain indications.611 Consolidating from duloxetine 30 mg twice daily to one 60 mg capsule once daily can decrease the cost of the regimen and give the same equivalent dose per day. Some patients may need to take duloxetine twice daily for tolerance (i.e., decrease nausea).7
In general, the 30 mg and 60 mg strength capsules of duloxetine may be more cost-effective than the 20 mg strength if all strengths are the same cost. Patients may be started at 30 mg once daily for 1 week then increased to 60 mg once daily for tolerability.11 Even in a study of elderly patients with medical comorbidity randomized to start duloxetine at a higher dose of 60 mg/day or placebo, the overall discontinuation rates for any reason did not significantly differ from placebo.40 In contrast, patients initiated on the 20 mg strength of duloxetine can end up after dose titration with increased pill burden and cost (i.e., three 20 mg capsules of duloxetine). Then, if the provider titrates duloxetine further by 20 mg increments the patient could end up on 4 or 5 capsules (80 mg or 100 mg) daily which can't be consolidated to an equivalent dose with other strengths.
CONCLUSION
Given that the efficacy of antidepressants for MDD is considered to be equivalent between and within classes, in general venlafaxine appears to be the most cost-effective SNRI currently. This might change with generic availability of duloxetine at the end of 2013. When choosing a venlafaxine product, the generic extended-release of venlafaxine is usually clinically preferred over immediate-release given its increased tolerability profile and simplicity of once daily dosing. In addition, when looking at patient specific factors, duloxetine may be preferred over venlafaxine in patients at risk of QT prolongation, with uncontrolled HTN (use cautiously as all SNRIs can increase blood pressure), and for treatment of fibromyalgia. If the patient is not a candidate for venlafaxine, and duloxetine is initiated, dose consolidation can help minimize cost and improve adherence for the patient. Desvenlafaxine is an alternative option to duloxetine if not a candidate for venlafaxine. There is little clinical experience with levomilnacipran and it is only available as a brand name product. Consider levomilnacipran if the patient is not a candidate for venlafaxine, desvenlafaxine, or duloxetine and needs an SNRI. Although milnacipran is FDA approved for fibromyalgia, it is not FDA approved in the US for MDD and is also only available as a brand name product. Consider milnacipran as an alternative SNRI to duloxetine if needed for fibromyalgia.