Home > DARE Reviews > The treatment of depression in cancer...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

The treatment of depression in cancer patients: a systematic review

G Rodin, G Lloyd, M Katz, E Green, JA Mackay, and RK Wong. Supportive Care Guidelines Group of Cancer Care Ontario Program in Evidence-based Care.

Review published: 2007.

Link to full article: [Journal publisher]

CRD summary

This review evaluated the effects of treatment for depressive disorders in cancer patients. There was little evidence to support the effectiveness of either pharmacological or non-pharmacological interventions, and no evidence that either treatment modality was superior. The conclusions are likely to be reliable, despite some limitations in the reporting of the review and language restrictions in the literature search.

Authors' objectives

To assess the effects of pharmacological and non-pharmacological treatments for major depression and other depressive disorders in cancer patients.


MEDLINE, EMBASE, CINAHL, PsycINFO, and the Cochrane Library were searched to 2005; the search terms were reported. In addition, selected conference proceedings were searched for abstracts and selected websites were searched for evidence-based guidelines. The reference lists of retrieved articles were also examined. Both published and unpublished studies were included. The search was limited to studies reported in English.

Study selection

Study designs of evaluations included in the review

Systematic reviews, meta-analyses, controlled studies, prospective cohort studies, cross-sectional studies and case-control studies were eligible for inclusion. The included studies were blinded and non-blinded randomised controlled trials (RCTs) and a cohort study.

Specific interventions included in the review

Studies that compared a pharmacological or non-pharmacological treatment with another active treatment, placebo or observation were eligible. The pharmacological interventions evaluated in the review included mianserin (daily dose 20 to 40 mg), fluoxetine (20 to 60 mg), desipramine (100 to 150 mg), alprazolam (4.5 mg), paroxetine (20 to 40 mg) and amitriptyline (75 to 150 mg); these were compared head-to-head, or with progressive muscle relaxation or placebo. Non-pharmacological interventions included multi-component nurse-led management, psychological therapy (of unspecified type), group psychotherapy plus relaxation, and an orientation programme; these were compared with usual care, relaxation alone or no therapy. The duration of the interventions was not specified, but study length ranged from 1 week to 1 year.

Participants included in the review

Studies that included adult patients with cancer diagnosed with major or minor depression, dysthymia or adjustment disorder were eligible. Patients were diagnosed with major depression by a ‘gold-standard’ structured diagnostic interview (category A), or had depressive symptoms greater than a pre-defined cut-off score when using a validated assessment scale (category B). The included studies evaluated patients with various types of cancer; where reported, some included patients receiving chemotherapy and/or radiotherapy.

Outcomes assessed in the review

Studies that reported symptomatic response to treatment using a standardised measure, quality of life, treatment discontinuation rate or adverse effects were eligible. The outcome measures used to determine symptom improvement varied across the studies.

How were decisions on the relevance of primary studies made?

Two reviewers selected studies for the review. The authors did not state whether these reviewers worked independently, nor how any disagreements were resolved.

Assessment of study quality

Validity was assessed by consideration of the following aspects of quality: randomisation method, blinding, explanation of withdrawals, sample size, power calculation, whether power was achieved, intention-to-treat analysis and funding source. The authors did not state how the validity assessment was performed.

Data extraction

The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction. It appears that data were extracted as reported in the individual studies. To determine symptomatic improvement, for continuous data, the mean end score or change from baseline with standard deviation and/or the p-value were extracted. For dichotomous data, the proportion of patients who achieved a predetermined response was extracted, with its associated p-value. Withdrawal and adverse event rates were extracted as percentages, with associated p-values where available.

Methods of synthesis

How were the studies combined?

The authors noted the heterogeneity between the interventions and the variability of findings in the included studies and reported that this precluded the pooling of results.

The studies were combined in a narrative, grouped by intervention (pharmacological and non-pharmacological). Studies with positive findings were reported first.

How were differences between studies investigated?

Differences were discussed in the narrative. The authors noted how differences in study findings might relate to study quality and the way in which patients were screened for depression. Differences were also apparent from the tabulated results.

Results of the review

Eleven studies were included: 10 RCTS (n=1,199) and one cohort study (n=60).

The authors noted that the included studies had methodological shortcomings, including small sample sizes and high rates of discontinuation from treatment, and that few used structured diagnostic interviews to establish eligibility for inclusion or measure response to treatment. The tables indicated that nearly all of the included studies were randomised (10 out of 11) and most (8 out of 11) reported results for at least 85% of participants. Most of the pharmacological studies (6 out of 7) were double-blinded. Among the non-pharmacological studies, one was not randomised and the others were non-blinded. Few of the RCTs (2 out of 10) gave details of their randomisation method.

Symptomatic improvement (11 studies).

Among RCTs evaluating antidepressants, two (n=128) reported that mianserin was associated with a statistically significant reduction in symptoms compared with placebo, and one (n=174) had inconsistent findings when alprazolam was compared with progressive muscle relaxation. No statistically significant difference was reported between fluoxetine and placebo (2 RCTs, n=254). Two RCTs of head-to-head comparisons (fluoxetine versus desipramine, n=38; paroxetine versus amitriptyline, n=179) reported improvements from baseline with both antidepressants, but no statistically significant difference between the antidepressants.

Among non-pharmacological comparisons, multi-component nurse-led management (1 RCT, n=60) and an orientation programme (1 RCT, n=180) both led to a statistically significant reduction in symptoms compared with usual care. No statistically significant difference was reported in studies of adjuvant psychological therapy (1 RCT, n=174) or psychotherapy with relaxation (1 RCT, n=303) compared, respectively, with no therapy or relaxation alone.

Quality of life (3 RCTs).

Two RCTs (n=254) compared fluoxetine with placebo. One (n=91) reported no statistically significant difference between the groups. The other (n=163) had mixed results, with a statistically significant benefit from fluoxetine when the data were analysed longitudinally but not when change scores were compared between groups. One RCT (n=179) comparing paroxetine with amitriptyline found no statistically significant difference between the groups.

Discontinuation rate from treatment (11 studies).

In the 4 placebo-controlled pharmacological studies, discontinuation rates varied from 19 to 54% in the intervention groups and from 7 to 56% in the placebo groups. There were statistically significant differences between the groups in three of these studies, with an inconsistent direction of effect: 2 studies favoured antidepressants (p=<0.05) and one favoured placebo (p=0.04). In the head-to-head comparisons of antidepressants, the discontinuation rate was significantly higher in the amitriptyline arm than in the paroxetine arm (24% versus 19%, p=0.49), while rates were not significantly different between the two arms in the small study comparing fluoxetine with desipramine. When alprazolam was compared with progressive muscle relaxation, significantly more patients discontinued treatment in the alprazolam arm (20% versus 11%, p=0.21). Adverse effects were the most commonly reported reason for discontinuation in the pharmacological studies.

Where reported, discontinuation rates ranged from 9 to 20% in the non-pharmacological studies, with no statistically significant difference between the groups.

Adverse effects (7 RCTs).

Only the pharmacological studies reported data on adverse effects. In 3 of the 4 RCTs that compared antidepressants with placebo, overall adverse effects were more frequently reported in patients who received antidepressant, but only one RCT of fluoxetine reported a statistically significant higher frequency of adverse effects (emesis) in the fluoxetine group compared with the placebo group. In head-to-head comparisons, a dry mouth was significantly more common with fluoxetine than with desipramine. There were no other statistically significant findings for this outcome.

Authors' conclusions

In cancer patients with depression or significant depressive symptoms, there is little evidence to support the effectiveness of pharmacological or non-pharmacological interventions and no evidence as to which treatment modality is superior.

CRD commentary

The review question and inclusion criteria were clear and adequate details of the primary studies were provided. Several relevant sources were searched for relevant studies, although the restriction to studies reported in English could have introduced language bias. The methods used in the review were not clearly defined: it is unclear to what extent the two reviewers worked independently while undertaking the study selection and data extraction processes, and thus whether steps were taken to minimise bias and reviewer error. Aspects of validity were discussed and considered in the review findings. Adequate details of the primary studies were reported; these indicate that the decision not to formally combine the results of the studies was appropriate. Overall, the authors’ conclusions are appropriately cautious, but the lack of reporting of the review process and the potential for language bias must be taken into consideration.

Implications of the review for practice and research

Practice: The authors stated that the relative effectiveness of different treatments for depression in cancer patients is unclear. The authors’ clinical practice guideline can be found on the Cancer Care Ontario’s Program in Evidence-Based Care website (accessed 30/05/08). See Web address at end of abstract.

Research: The authors stated that further research is needed to determine the relative effectiveness of pharmacological, non-pharmacological and combined treatments in cancer patients. Newer pharmacological interventions such as escitalopram, citalopram, mirtazapine, venlafaxine, duloxetine and bupropion should be evaluated.


Cancer Care Ontario; Ontario Ministry of Health and Long-term Care.

Bibliographic details

Rodin G, Lloyd G, Katz M, Green E, Mackay J A, Wong R K, Supportive Care Guidelines Group of Cancer Care Ontario Program in Evidence-based Care. The treatment of depression in cancer patients: a systematic review. Supportive Care in Cancer 2007; 15(2): 123-136. [PubMed: 17058100]

Other publications of related interest

This paper is based on a Practice Guideline produced by Cancer Care Ontario Practice Guidelines Initiative. The series is published on the Internet and regularly updated. To ensure that you are viewing the most up to date version, go the Cancer Care Ontario website at: http://www.cancercare.on.ca/english/toolbox/qualityguidelines/pebc/

1. Browman GP, Levine MN, Mohide EA, Hayward RS, Pritchard KI, Gafni A, et al. The practice guidelines development cycle: a conceptual tool for practice guidelines development and implementation. J Clin Oncol 1995;13:502-12.

Indexing Status

Subject indexing assigned by NLM


Antidepressive Agents /therapeutic use; Clinical Trials as Topic; Depressive Disorder /etiology /therapy; Humans; Neoplasms /complications /psychology; Psychotherapy; Treatment Outcome



Database entry date


Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 17058100


PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...