Reboxetine: a Randomized Controlled Open-Label Study in Children and Adolescents with Major Depression

Home , Reboxetine

IsrIsr JJ PsychiatryPsychiatry -- Vol. 56 - No 1 (2019) Reboxetine: A Randomized Controlled Open-label Study in Children and Adolescents with Major Depression

Paz Toren, MD,1,2 Guy Goldstein, MA,1,3 Galit Ben-Amitay, MD,1,2 Joseph Ben-Sheetrit, MD,1,4 and Michelle Slone, PhD3

1 Tel Aviv Brüll Community Mental Health Center, Clalit Health Services, Tel Aviv, Israel 2 Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel 3 School of Psychological Sciences, Tel Aviv University, Tel Aviv, Israel 4 Geha Mental Health Center, Petah Tikva, Israel

Introduction Abstract Objective: This study is the first to examine the efficacy General rates of adolescents diagnosed as having major of reboxetine in the treatment of children and adolescents depressive disorder (MDD) or dysthymia are approxi- with depression. mately 10% (1), with an annual prevalence of 8%-20% for adolescents and 5% for children (2, 3). Youth depres- Method: In this open-label randomized controlled study, 14 sion may lead to impairment in social and educational youths (aged 9-18 years) with major depression received functioning and can impact mental health through to either reboxetine (4mg/day) or fluoxetine (20mg/day) adulthood (4, 5). Individuals who suffered from depres- in a 2:1 ratio. Primary outcome measures included the sion in their youth are twice or three times more likely to Children’s Depression Rating Scale-Revised (CDRS-R), develop major depression as young adults compared to the Clinical Global Impression-Improvement scale (CGI-I), children and adolescents who did not suffer from depres- and the Children’s Depression Inventory (CDI). Response sion (6, 7). Children and adolescents with MDD are also at was defined by either a 20% reduction in CDRS-R or a increased risk for suicidality compared to non-depressed 2-point reduction in CGI-I. youth (8). Reboxetine is an antidepressant of the selective norepi- Results: Improvement in depression with reboxetine nephrine reuptake inhibitors (NaRIs). Reboxetine’s main was noted within 4 weeks. A further non-significant action is in raising the synaptic levels of norepinephrine improvement was noted after 8 weeks, proceeding through by binding to the norepinephrine transporter and block- week 12. Response was achieved by six subjects (66.7%) ing its reuptake (9). Reboxetine’s selectivity was found by week 4 and eight subjects (88.9%) by week 12. Both to be equal or superior to desipramine and superior to reboxetine and fluoxetine were safe with tolerable adverse imipramine (10). Reboxetine was found to be effective effects, although several non-significant differences in for treating adult depression compared to placebo and efficacy and overall frequency of side effects were found other antidepressants. Reboxetine as effective as imipra- in favor of fluoxetine. mine and fluoxetine, and at least as effective as desipra- Conclusions: Reboxetine was safe and effective in mine (11-14). Recent meta-analyses have addressed the treating youth with major depressive disorder. Further question of the efficacy of reboxetine compared to the studies with long-term follow up are needed in order to selective serotonin reuptake inhibitors (SSRIs) in adults. replicate our findings and determine the relative efficacy Eyding et al. (15) concluded that reboxetine showed no of reboxetine compared to fluoxetine. differences from placebo and was inferior to SSRIs on both response and remission rates, whereas Papakostas et al. (16) concluded that the efficacy of the SSRIs and

Address for Correspondence: Joseph Ben-Sheetrit, MD, Tel Aviv Brüll Community Mental Health Center, 9 Hatzvi St., Tel Aviv 6719709, Israel [email protected]

26 Paz Toren et al. reboxetine were similar, and that the major differences or fluoxetine (20 mg/day) treatment. All participants between them were only in their adverse effects profiles. were intended to receive concomitant dynamic psycho- Reboxetine has a high safety profile (10). Adverse therapy. By the end of the 8th week, the medication dosage effects were found to subside spontaneously with no was gradually reduced until termination. When clinical cardiotoxic potential (13). No differences were found in improvement was noted, the pharmacological treatment the rates of adverse effects of fluoxetine and reboxetine could be continued beyond the 8th week, depending on (15). A previous study reported that the adverse effects clinical considerations. of the NaRIs, including reboxetine are as tolerable as Assessments were made at baseline, after 2 and 4 weeks those of the SSRIs (17). on drug treatment, and every 4 weeks thereafter, and In a previous 6-week open-label study in the Tel Aviv included: Community Mental Health Center clinic (18), we assessed a. A DSM-IV-based psychiatric interview of the child the efficacy of reboxetine (4 mg/day) in the treatment of 31 and an interview with his or her parents, by a senior children and adolescents with attention deficit/hyperactivity child and adolescent psychiatrist (baseline only). disorder (ADHD). A significant improvement was shown b. Clinician-rated (see Study Tools 1-3) and self-report in ADHD symptoms, as well as a decrease in comorbid (see Study Tools 4-6) questionnaires. depressive symptoms, but not in anxiety symptoms. Adverse c. Monitoring of adverse events. effects were generally mild and tolerable, appearing within the first days of treatment and subsiding spontaneously Sample Size within a few days. The most common adverse effects were Considering a strict criterion of being able to demonstrate drowsiness or sedation and gastrointestinal complaints a significant difference for depression of mild baseline (18). The reduction in depression symptoms and the safety severity (e.g., CDRS-R=55)(20) and assuming an expected profile of the drug suggested that reboxetine may be an standard deviation of SD=9.9 (based on a previous study appropriate antidepressant treatment for this age group. (21)), in order to detect a 20% difference in CDRS-R scores Surprisingly, no data have yet been published regarding (considering α=.05 and 1-β=.80), a total sample size of the efficacy of reboxetine in the treatment of depression N=30 would be required in a 2:1 ratio (i.e., n=20 for the in youth. The present randomized controlled open-label reboxetine group and n=10 for the fluoxetine group), study aimed to assess the efficacy of reboxetine as an anti- which would then yield an effect size of d=1.1. depressant in children and adolescents, and the tolerability of reboxetine compared to fluoxetine as antidepressants Subjects in children and adolescents. Seventeen subjects (11 girls and 5 boys, aged 15.6±2.24 years, range 9-18 years) gave their consent to participate in the study and matched the enrollment criteria. Two Methods subjects (11.8%, a 15.5-year-old male and a 17-year-old Inclusion & Exclusion Criteria female) did not receive the study medication due to All subjects met the Diagnostic and Statistical Manual of baseline ECG abnormalities. Ten participants were ran- Mental Disorders, 4th edition, text revision (DSM-IV-TR) domized to receive reboxetine treatment and 5 to receive (19) criteria for major depressive disorder. The diagnosis fluoxetine treatment. One subject in the reboxetine group was made by a senior child and adolescent psychiatrist. (a 17-year-old male) dropped out before the 2nd week of All participants were drug-naïve or without any medi- treatment due to noncompliance. In the fluoxetine group, cation for at least one month. Exclusion criteria were no dropouts occurred before the 2nd week of the study. a diagnosis of autistic spectrum disorder, psychosis or Fourteen of the 17 (82.3%) subjects who initially signed bipolar disorder. Other mental disorders were allowed to up for the study completed at least 4 weeks of treatment. be documented and included in the study. Children with A full treatment period of 12 weeks was completed by 6 intellectual disability, substance use or chronic medical (66%) subjects of the reboxetine and 5 (100%) of the fluox- illnesses were excluded. etine group. Two subjects (22%) in the reboxetine group completed only 4 weeks of treatment: non-compliance Design & Intervention attributed to symptom improvement (1 subject) and Following baseline assessment, participants were ran- non-compliance due to mild side effects, i.e., mild loss domly assigned to receive either reboxetine (4 mg/day) of appetite and mild insomnia (1 subject).

27 reboxetine in Children

There were no significant differences in the CDRS-R gastrointestinal complaints, headaches and dizziness). baseline scores between of subjects who completed the Weight and height, heart rate, blood pressure and elec- entire treatment period, subjects who completed more trocardiograms (ECGs) were also assessed. Complete than two weeks of treatment, and those who did not have blood counts (CBCs) and liver and kidney functions a minimum of 2 weeks of treatment (F2=0.13, ns). Further, were assessed at baseline and after 4 weeks of treatment, no significant correlations were found between treatment and according to clinical judgement thereafter. length and baseline CDRS-R scores for either of the study groups. The analysis of dropout subjects was made accord- Ethical Approval ing to the intention-to-treat method. Two subjects (22%) The study (Clinicaltrials.gov NCT00426946) was approved from the reboxetine group continued to receive reboxetine by the Institutional and the Ministry of Health Review treatment after study termination. One was treated for 6 Boards. All children gave their assent to participate in the months and the other was treated for a year. study and their parents signed an informed consent form.

Measures Data analysis Main outcomes Data analysis was made according to the “intention to 1. The Children’s Depression Rating Scale-Revised treat” principle. Baseline age and gender distribution (CDRS-R) (22) is a semi-structured clinician-rated were analyzed by the two tailed independent sample interview for the assessment of youth depression. It T test and Fisher exact test, respectively. Although the combines information from several areas with particu- final sample size for analysis was small (N=14), the data lar emphasis on the physiological aspects of depression. were normally distributed (assessed both visually using 2. The Clinical Global Impression-Improvement scale Q-Q Plots and calculated using Skewness and Kurtosis Z (CGI-I) (23) is a clinician-rated assessment of clinical scores, which were all well within ±2.58). Therefore, we improvement or deterioration of depressive symptoms preferred to use parametric statistical analyses in order compared to baseline assessment (pre-test), rated from to increase statistical power (28). 1 (“very much improved”) to 7 (“very much worse”). In order to compare symptoms at various assessment 3. The Children’s Depression Inventory (CDI) (24) is a points between the two treatments, a mixed-model analy- 27-item, self-rated questionnaire for the evaluation sis of variance (ANOVA) was used with repeated measures of affective, cognitive, and behavioral symptoms of (5 measures) as an independent within-subjects variable, depression. and the drug (reboxetine/fluoxetine) as a between-subject independent variable. The dependent measures were the Additional measures different outcome assessment measures described above. 4. The Clinical Global Impression-Severity Scale (CGI- Greenhouse-Geisser adjustment was applied given a S) (23) is a clinician-rated scale used to assess overall significant Mauchly’s test of sphericity, and the corrected severity and impairment due to depressive symptoms, values were reported appropriately, with the degrees rated from 1 (“not ill”) to 7 (“extremely ill”). In the of freedom rounded to the nearest integer. Post-hoc current study, this scale was used to determine the pairwise comparisons were used to assess the relation severity of the depressive clinical condition at baseline. between measures. Statistical significance was set as 5. The Revised Children’s Manifest Anxiety Scale p<.05. Non-significant results are marked as “ns.” The (RCMAS) (25, 26) is a 37-item self-report inven- Bonferroni correction for multiple comparisons was tory for evaluating physiological, apprehensive, and employed, as appropriate. oversensitivity-of- concentration factors of anxiety. Response was defined as an improvement equal to or 6. The Suicide Ideation Questionnaire-Short Version greater than 2 points of the CGI-I score at T4 (week 4) and (SIQ) (27) is an 8-item questionnaire that measures T12, compared to score at T0 (baseline), or a reduction severity and frequency of suicide ideation that was equal to or greater than 20% in CDRS-R score between used in the current study to assess the effects of the baseline (T0) and weeks 4 and 12. Reduction in symptoms medication on suicide ideation tendencies. of depression between baseline (T0) and measurement 7. Adverse events monitoring, assessed using both spon- points T4 - T12 was measured by calculating the per- taneous reports made by the patient and directive ques- centage of CDRS-R total score change, while subtracting tioning by the physician (regarding sleep, drowsiness, the CDRS-R minimal score (17) from the denominator:

28 Paz Toren et al.

Figure 1. Changes in CDRS-R and CDI scores across time points in our study (N=14)

80 Reboxetine 25 Reboxetine Fluoxetine Fluoxetine 75.17 22.6 74.4

70 20 65.67 19.33 64.67 63.2* 15.56 16.67 16.89 63.4 60 15 CDI Scores

CDRS-R Scores 54.6 50 10 49.6

6.6 6.0 6.0

0 4 8 12 0 4 8 12 WEEKSWeeks WEEKSWeeks

The difference between the reboxetine and fluoxetine was not statistically significant, see text for further details; CDRS - Children’s Depression Rating Scale-Revised; CDI – Children’s Depression Inventory (CDI); * - The difference from week 8 was not statistically significant.

(100x[CDRS-R.T0 – CDRS-R.T4]/[CDRS-R.T0-17]) (29). Depressive Symptoms - Primary Outcome In order to assess suicidal tendencies, a mixed-model Measures analysis-of-variance was applied, with two measures CDRS–R and CDI (see Figure 1). CDRS scores signifi- for the within–subjects independent variable, the drug cantly decreased during the study (F[4,48]=7.7, p<.001). between-subjects variable and the SIQ scores mean as Although improvement in depressive symptoms was the dependent variable. greater with fluoxetine, the difference from reboxetine Reboxetine Long-Term Follow-up: Participants who was not statistically significant. Regarding the CDI, a wished to continue reboxetine treatment due to symptom significant treatment-by-time interaction was found improvement beyond the 12th week were able to do so in (F[4,48]=4.16, p<.01), but the change in CDI scores accordance with the clinician’s assessment. CGI-I scores was not statistically significant for the reboxetine group. were assessed at 6 months and 12 months from baseline. CGI-S and CGI-I. Overall clinical severity and impair- Adverse effects were assessed by the subjects’ spontane- ment due to depression symptoms were measured by ous reports and directive questioning by the clinician. CGI-S scores at baseline (T0). Mean CGI-S scores at T0 were 5.67±0.7 for the reboxetine and 5.8±0.45 for the fluoxetine group (ns). Improvement in clinical severity was Results measured by CGI-I (a linear transformation was performed Baseline Measures [-X - 4], in order to be able to present “no change” as “0”). The reboxetine and fluoxetine study groups did not show CGI-I scores increased along the treatment period and a statistically significant differences in age and gender significant improvement from baseline (0 denoting no distribution (age: T4=0.71, ns; gender: Fisher exact test, change) was noted at the 4th week of the study (F13=6.62 ns). No significant differences were found between the p<0.001). For the reboxetine group, the mean improve- reboxetine and the fluoxetine groups on all baseline ment score was 1.33±0.87 at T4 and 2.0±0.23 at T8, with measures of depression, i.e., CDRS-R, CDI and baseline a further significant improvement to 2.2±0.8 at week 12 CGI-S [T12=-.53, T12=.14, T12=-.38, respectively, all p (T12). For the fluoxetine group, the mean improvement values ns). score was 2.2±0.87 at T4 with a significant improvement

29 reboxetine in Children to 3.0±0.0 at T12 through 2.4±0.34 at T8. A main effect ferent between the reboxetine and the fluoxetine groups was found for the time variable between week 4 and 12 (F1=0.13, ns). SIQ scores were not significantly different (F1=13.19, p<0.05), and no difference was found between between the two drugs throughout the study (i.e., F1<4.0, the two drugs (F1=0.05, ns). ns, for all analyses). Response rate. Response rate was measured by CDRS-R Other adverse effects: Reboxetine and fluoxetine treat- score reduction and CGI-I score, see Table 1. Although ment were both safe without any moderate or severe response rates for fluoxetine were higher, the difference adverse effects (adverse effects are presented in Table 2). was not statistically significant. In the reboxetine group, 6 subjects (66.6 %) experienced Although the subjects in our study were intended to at least one adverse effect. Most adverse effects were mild receive psychotherapy along with the drug treatment, and tolerable and subsided within a period of 5-14 days. in practice, most of them (57.1%, n=8) did not receive One subject reported a decrease in appetite which did not psychotherapy during the study. The difference in the improve. She eventually left the study during the th4 week. proportions of subjects who underwent psychotherapy This subject discontinued treatment at a time point of between the reboxetine and fluoxetine groups was not significant improvement of her symptoms. Three (33.3%) 2 statistically significant (33.3% vs. 60.0%, χ (1)=0.93, subjects reported headaches, dizziness or drowsiness. All p=.33). Therefore, we were unable to assess the effects of these symptoms disappeared spontaneously by the 2nd of psychotherapy in youth with depression who received week of the treatment. Decreased appetite or nausea were reboxetine or fluoxetine. reported by 3 (33%) subjects. Two (22%) subjects reported dry mouth and one subject (11%) reported palpitations Anxiety Symptoms and nocturnal awakenings. In the fluoxetine group, 2 Anxiety symptoms were measured by RCMAS, where a subjects (40%) reported adverse effects: akathisia (40%), treatment-by-time interaction effect was found (F[2,48]= drowsiness (40%) and headaches (20%). All adverse 3.27, p<0.05). RCMAS decreased only for the fluoxetine effects subsided spontaneously within a week. group, where a significant change of scores was noted between baseline and week 4, with a further significant Reboxetine Follow-up decrease in the 12th week [T0: 16.0±2.52, T4: 9.2±2.9, Reboxetine Long-Term Efficacy. Two subjects treated T8: 8.3±3.2, T12: 5.0±2.96). beyond the 12th week remained in remission. At 6 months on reboxetine treatment, one (50%) of the subjects was Tolerability and Safety rated as “very much improved,” and the other (50%) Suicide ideation: Changes in suicide ideation, as mea- as “much improved.” After 12 months on reboxetine sured by the SIQ Sum score, were not significantly dif- treatment, the only participant who still continued the treatment was rated as “much improved,” and his CDRS-R score was reduced by 46% as compared to his baseline Table 1. Proportions of subjects meeting response criteria in the reboxetine and fluoxetine groups in our study score (CDRS-R score was 81 at baseline and 46 at 12 Week 4 Week 12 months). Criteria Group N (%) N (%) Sig. CDRS-R reduction Reboxetine 4 (44.4) 5 (55.5) Table 2. Frequency of adverse effects in the reboxetine and ns* fluoxetine groups in our study (N=14) ≥ 20% Fluoxetine 4 (80.0) 4 (80) CGI-I ≥ 2 Reboxetine 4 (33.3) 7 (77.8) Fluoxetine Reboxetine ns* Fluoxetine 4 (80) 5 (100) Adverse Effects N (%) Adverse Effects N (%) Either† Reboxetine 6 (66.7) Any adverse effect 2 (40) Any adverse effect 6 (66.6) ns** Drowsiness/ sedation 2 (40) Decreased appetite 3 (33) Criteria Fluoxetine 0 (0.0) Akathisia 2 (40) Drowsiness/sedation 3 (33) Both‡ Reboxetine 1 (11.0) Headaches 1 (20) Appetite-loss 3 (33) Criteria ns** Fluoxetine 4 (80) Headaches 2 (22) Sig. – Significance; ns – non-significant, p>.05; CDRS-R – Children’s Dizziness 2 (22) Depression Rating Scale – Revised; CGI-I – Clinical Global Impression, Dry mouth 2 (22) Improvement Scale; Nausea 1 (11) * – Assessed by Fischer exact test. ** – Assessed by Chi squared test; † – Either CDRS-R reduction ≥ 20% or CGI-I ≥ 2; ‡ – Both CDRS-R Nocturnal awakenings 1 (11) reduction ≥ 20% and CGI-I ≥ 2. Palpitations 1 (11)

30 Paz Toren et al.

Reboxetine follow-up adverse effects. Subjects continu- be addressed in future studies. Currently, the choice ing reboxetine beyond the 12th week did not report any between these two drugs, as with other antidepressants in adverse effects during the follow-up period. general, is heavily influenced by their side effects profile. Adverse effects of reboxetine in the present study were similar to existing data regarding its safety profile in children Discussion with depression (17, 31) and ADHD (18). Reboxetine was General Discussion well tolerated, with mild and transient adverse effects. The In this randomized controlled small-sample open-label main adverse effects were headaches, dizziness or drowsi- study of reboxetine and fluoxetine treatment in children ness, decreased appetite, nausea and dry mouth. The rate and adolescents with major depression, we found that of reboxetine discontinuation due to adverse effects in the both reboxetine and fluoxetine were effective treatments. present study matches existing data from meta-analyses of However, there was a clear yet non-statistically signifi- reboxetine use in adults, in which 12.6%-16% discontinu- cant differences suggesting better outcomes in the fluox- ation rates due to adverse effects were found (16, 31). An etine group. Subject of the fluoxetine group had greater issue of major concern with antidepressant use in chil- improvements in CDRS-R, CDI (in fact, the latter remained dren and adolescents is the potential of the medication to unchanged in the reboxetine group), and CGI-I (despite a increase suicide ideation and attempts. In general, a debate rather similar baseline CGI-S) scores. Moreover, response exists regarding whether antidepressant use increases the rates were higher in the fluoxetine group, and only fluox- risk for suicidal thoughts and actions (32) or not (33, 34). etine significantly ameliorated anxiety symptoms. Finally, Regulatory black box warnings require clinical follow up the fluoxetine group had a lower percentage of side effects, (35). In the present study, no increase in suicide ideation or although the profile was markedly different (see Table attempts were observed during the study or the follow-up 2). Due to the small sample of our study we are unable periods, and reboxetine did not differ from fluoxetine in to determine whether these differences represent true the suicide ideation score. differences in drug responses and safety profiles. Clearly, The differences between reboxetine and fluoxetine further studies are indicated in this regard. regarding the prevalence of adverse effect were not statistically significant, but as noted above, this should be Reboxetine taken with caution given the small number of our study Concerning reboxetine, improvement in depressive symp- participants and the possibility for a type 2 error. Two toms was achieved within 4 weeks of reboxetine treatment, as meta-analyses assessing adults regarding depression (15, evidenced by both CDRS-R and CGI but not by CDI scores. 16) also did not find a difference between fluoxetine and Considering that the CDRS-R is clinician-rated and the CDI reboxetine in the rate of subjects with at least one adverse is self-reported, a positive response in the CDRS-R but not effect. However, in the present study, fluoxetine showed the CDI suggests that improvement with reboxetine was a lower adverse-effect related withdrawal rate (0%) than more pronounced when assessed objectively than subjec- reboxetine (11%). This is in line with studies assessing tively. At the 4th week of the study, 66.7% of the reboxetine treatment of adults with depression (15, 36). group participants met clinical response criteria by either a positive change in CGI score or by a decrease of 20% or Anxiety symptoms more in their CDRS-R score. Rate of symptom improvement Anxiety symptoms decreased significantly for subjects is a good indicator of acute treatment response. According treated with fluoxetine but did not change in those treated to previous studies, significant symptom reduction by week with reboxetine. The anxiolytic effect of fluoxetine in 4, desirably around 50%, correlates highly with remission the current study was not surprising since SSRIs are at the end of the acute treatment (29). recommended as first-line treatment in anxiety (37) and were found to be safe and effective in youth with Adverse Effects anxiety disorders as well (38). As for reboxetine, some Although fluoxetine is presently the drug of choice in researchers suggest noradrenergic dysfunction as a pos- children and adolescents (30), reboxetine, with its norad- sible mechanism for anxiety and have raised the hypoth- renergic profile, may offer a viable alternative. However, esis that NaRIs may increase anxiety. Some studies on our study is not powered to adequately assess between- adult populations suggest that NaRIs do not exacerbate drug differences and, as mentioned above, this should anxiety, as noted in the current study, and that reboxetine

31 reboxetine in Children might even be effective in treating panic disorder (39- Conflicts of Interest and Sources of Funding: 41). Unfortunately, data on the anxiolytic use of NaRIs All authors declare that they have no conflict of interest regarding this study. in general and reboxetine in particular are limited, and further studies are clearly needed. Overall, the results of the present study support those of our previous study References on reboxetine for ADHD (18), in which a significant 1. Merikangas KR, He JP, Burstein M, et al. Lifetime prevalence of mental disorders in US adolescents: Results from the National Comorbidity decrease in comorbid depressive symptoms was found, Survey Replication–Adolescent Supplement (NCS-A). J Am Acad Child but not in anxiety symptoms. Adolesc Psychiatry 2010;49:980-989.‏ 2. Kessler RC, Avenevoli S, Costello E, et al. Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Limitations and Future Directions Comorbidity Survey Replication Adolescent Supplement. Arch Gen The current study has several limitations. Most impor- Psychiatry 2012; 69:372-380. tantly, the small sample size and the lack of a placebo 3. Rohde P, Lewinsohn PM Klein DN, et al. Key characteristics of major depressive disorder occurring in childhood, adolescence, emerging arm limit the interpretation of the efficacy data: Our adulthood, and adulthood. Clin Psychol Sci 2013;1:41-53.‏ study was planned for a sample size of N=30, but final 4. Fletcher JM. Adolescent depression: Diagnosis, treatment, and analysis was conducted for 14 subjects. Several differences educational attainment. Health Econ 2008;17(11):1215-1235.‏ were found in favor of fluoxetine that raise the question 5. Pine DS, Goldstein RB, Wolk S, Weissman MM. The association between childhood depression and adulthood body mass index. Pediatrics whether fluoxetine is indeed superior to reboxetine, a 2001;107:1049-1056. question that cannot be answered by the current study and 6. Weissman MM, Wolk S, Goldstein RB, et al. Depressed adolescents necessitates further investigation of the matter. Moreover, grown up. J Am Med Assoc 1999;281:1707-1713. 7. Pine DS, Cohen P, Gurley D, et al. The risk for early adulthood anxiety we were unable to assess the effects of psychotherapy and depressive disorders in adolescents with anxiety and depressive in youth with depression who received reboxetine or disorders. Arch Gen Psychiatry 1998;55:56-64. fluoxetine, and this too should be addressed by future 8. Hawton K, Saunders KE, O’Connor RC. Self-harm and suicide in studies. The lack of a placebo arm in an open-study poses adolescents. Lancet 2012;379: 2373-2382. 9. Page ME. The promises and pitfalls of reboxetine. CNS Drug Review a limit concerning the extent to which between-drug 2003;9: 327-342. differences were drug-related and not placebo-related. 10. Wong EHF, Sonders MS, Amara SS, et al. Reboxetine: A pharmacologically In particular, subjects knew that reboxetine was experi- potent, selective, and specific norepinephrine reuptake inhibitor. Biol Psychiatry 2000;47:818-829. mental for youth depression while fluoxetine was the 11. Kent JM. SNaRIs, NaSSAs, and NaRIs: New agents for the treatment of standard treatment, and we cannot rule out that this depression. Lancet 2000;355:911-918. may have affected response rates. The safety data are 12. Massana J. Reboxetine versus fluoxetine: An overview of efficacy and also limited by the small sample size and the lack of a tolerability. J Clin Psychiatry 1998;59:8-10. 13. Montgomery SA. Chairman’s overview. The place of reboxetine in placebo arm. It should also be noted that the study was antidepressant therapy. J Clin Psychiatry 1989;59:26-29. randomized yet open-labeled and participants were fully 14. Versiani M, Amin M, Chouinard G. Double-blind, placebo-controlled aware about reboxetine being a newly-assessed drug as study with reboxetine in inpatients with severe major depressive disorder. compared to fluoxetine as the golden standard comparator J Clin Psychopharmacol 2000;20:28-34. 15. Eyding D, Lelgemann M, Grouven U, et al. Reboxetine for acute treatment (e.g., through the detailed informed consent procedure). of major depression: Systematic review and meta-analysis of published and unpublished placebo and selective serotonin reuptake inhibitor controlled trials. Brit Med J 2010; 341:c4737. Conclusions and clinical significance 16. Papakostas GI, Nelson JC, Kasper S, Möller HJ. A meta-analysis of Reboxetine (4 mg/day) may be a safe and effective treat- clinical trials comparing reboxetine, a norepinephrine reuptake inhibitor, ment of major depression in children and adolescents. with selective serotonin reuptake inhibitors for the treatment of major Several differences in favor of fluoxetine were found, but depressive disorder. Eur Neuropsychopharmacol 2008;18:122-127. 17. Whiskey E, Taylor D. A review of the adverse effects and safety of we were unable to assess whether fluoxetine was indeed noradrenergic antidepressants. J Psychopharmacol 2013;27:732-739. superior to reboxetine due to out small sample size. The 18. Ratner S, Laor N, Bronstein Y, et al. Six-week open-label reboxetine current study is the first to assess reboxetine’s efficacy in treatment in children and adolescents with attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 2005;44: 428-433. the treatment of youth depression. Placebo-controlled 19. American Psychiatric Association: Diagnostic and Statistical Manual studies are needed to validate or refute the preliminary of Mental Disorders, 4th edition (DSM-IV-TR). Washington, D.C.: results of the present study, in drug naive and in SSRIs- American Psychiatric Association, 2000. 20. Jain S, Carmody TJ, Trivedi MH, et al. A psychometric evaluation of resistant youth. Long-term placebo-controlled trials are the CDRS and MADRS in assessing depressive symptoms in children. needed to assess the recommended maintenance period. J Am Acad Child Adolescent Psychiatry 2007, 46:1204-1212.‏

32 Paz Toren et al.

21. Plener PL, Grieb J, Sproeber N, et al. Convergence of children’s depression drug treatment: A meta-analysis. J Child Adolesc Psychopharmacol rating scale-revised scores and clinical diagnosis in rating adolescent 2006;16:25-32. depressive symptomatology. Ment Illness 2012;4:29-31. 33. Cooper WO, Callahan ST, Shintani A, et al. Antidepressants and suicide 22. Poznanski EO, Freeman LN, Mokros HB. Childrens’ Depression Rating attempts in children. Pediatrics 2014; 133:204-210. Scale Revised. Psychopharmacol Bull 1985;21:979-989. 34. Bridge JA, Iyengar S, Salary CB, et al. Clinical response and risk for 23. Guy W. ECDEU Assessment Manual for Psychopharmacology. Rockville, reported suicidal ideation and suicide attempts in pediatric antidepressant MD: U.S. Department of Health, Education and Welfare, 1976. treatment. J Am Med Assoc 2007;297:1683-1696.‏ 24. Kovacs M. The Children’s Depression Inventory. Psychopharmacol 35. Kurian BT, Ray WA, Arbogast PG, Fuchs DC, Dudley JA, Cooper WO. Bull. 1985;21:995-998. Effect of regulatory warnings on antidepressant prescribing for children 25. Reynolds CR, Richmond BO. Manual for the revised children’s manifest and adolescents. Arch Pediatr Adolesc Med 2007;161: 690-696. anxiety scale. Los Angeles: Western Psychological Services (WPS), 1985. 36. Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and 26. Reynolds CR, Richmond BO. What I think and feel: A revised measure acceptability of 12 new-generation antidepressants: A multiple-treatments of children’s manifest anxiety. J Abnorm Child Psychol 1997;25:15-20. meta-analysis. Lancet 2009; 373:746-758. 27. Reynolds WM. Suicidal Ideation Questionnaire (SIQ). Odessa, Florida: 37. Bandelow B, Sher L, Bunevicius R, et al. Guidelines for the pharmacological Psychological Assessment Resources, 1987. treatment of anxiety disorders, obsessive-compulsive disorder and 28. Vickers, AJ. Parametric versus non-parametric statistics in the analysis posttraumatic stress disorder in primary care. Int J Psychiatry Clin of randomized trials with non-normally distributed data. BMC Med Pract 2012;16:77-84. Res Methodol 2005, 5:35.‏ 38. Birmaher B, Axelson DA, Monk K, et al. Fluoxetine for the treatment 29. Tao R, Emslie G, Mayes T, et al. Early prediction of acute antidepressant of childhood anxiety disorders. J Am Acad Child Adolesc Psychiatry treatment response and remission in pediatric major depressive disorder. 2003;42:415-423. J Am Acad Child Adolesc Psychiatry 2009;48:71-78. 39. Kalk NJ, Nutt DJ, Lingford-Hughes AR. The role of central noradrenergic 30. March JS, Silva S, Petrycki S, et al. The Treatment for Adolescents with dysregulation in anxiety disorders: Evidence from clinical studies. Depression Study (TADS): Long-term effectiveness and safety outcomes. J Psychopharmacol 2011;25:3-16.‏ Arch Gen Psychiatry 2007;64:1132-43. 40. Stahl SM, Mendels M, Schwartz GE. Effects of reboxetine on anxiety, 31. Hajós M, Fleishaker JC, Filipiak‐Reisner JK, et al.‏ The elective agitation, and insomnia: Results of a pooled evaluation of randomized norepinephrine reuptake inhibitor antidepressant reboxetine: clinical trials. J Clin Psychopharmacol 2002;22:388-392.‏ Pharmacological and clinical profile. CNS Drug Review 2004;10: 23-44. 41. Schosser A, Kasper S. The role of pharmacogenetics in the treatment of 32. Mosholder AD, Willy M. Suicidal adverse events in pediatric randomized, depression and anxiety disorders. Int Clin Psychopharmacol 2009;24: controlled clinical trials of antidepressant drugs are associated with active 277-288.