CRD summary

This review compared dexamethasone with prednisone during induction chemotherapy in children with acute lymphoblastic leukaemia and found that dexamethasone was more efficacious than prednisone but was also associated with more toxicity. Therefore, it was unclear whether this short-term superiority would translate to superior overall survival with longer follow-up. These conclusions are likely to be reliable.

Authors' objectives

To compare the efficacy and toxicity of dexamethasone with prednisone during induction chemotherapy in children with acute lymphoblastic leukaemia.

Searching

MEDLINE (1950 to September 2010), EMBASE (1980 to September 2010) and Cochrane Central Register of Controlled Trials (CENTRAL; to the third quarter of 2010) were searched and search terms were reported. World Health Organization ICTRP and National Institutes of Health Register were also searched for unpublished studies. Conference proceedings were searched from 2007 to 2010 using Web of Science and Scopus. References of relevant articles were also examined for further studies.

Study selection

Randomised controlled trials (RCTs) that compared dexamethasone with prednisone during induction therapy in childhood acute lymphoblastic leukaemia were eligible for inclusion. Studies had to report at least one of the following outcomes: event rate; relapse rate for central nervous system (CNS), isolated bone marrow, testicular or a combination of these; or mortality rate. Secondary outcomes of interest were: death during induction therapy; osteonecrosis; withdrawal from the study following randomisation to corticosteroid; sepsis; fungal infection; diabetes; neuro-psychiatric events; pancreatitis; and myopathy. Any study site/country, follow-up period, dose, length of induction therapy, concurrent chemotherapy and frequency or method of drug administration was eligible for inclusion.

Most included studies had both high-risk and standard-risk patients, while two studies excluded high-risk patients. The prednisone to dexamethasone ratio varied between studies. Some trials used prednisolone. Three drug induction was used in three trials, the remainder used four-drug induction.

Two reviewers independently selected studies for inclusion and disagreements were resolved with a third reviewer.

Assessment of study quality

Methodological quality was assessed in terms of randomisation sequence generation, allocation concealment, blinding, incomplete outcome data and intention-to-treat analysis. The number of reviewers that assessed study quality was not reported.

Data extraction

Two reviewers independently extracted risk ratios (RR) and associated 95% confidence intervals (CI) for dichotomous outcomes. The authors of primary studies were contacted for missing data where necessary.

Methods of synthesis

Risk ratios and 95% confidence intervals were pooled using a random-effects model. Heterogeneity was investigated using forest plots, Ι² and Chi-squared tests. Subgroup analyses were planned to investigate the effects of dexamethasone dose ratio (below seven versus seven and above), age, gender and immunophenotype. Sensitivity analyses were used to investigate the effects of prednisone versus prednisolone; three versus four drug induction; and corticosteroid randomisation restricted to induction versus corticosteroid randomisation in induction plus other protocol phases. Publication bias was not assessed using a funnel plot as all analyses involved less than 10 RCTs.

Results of the review

Eight RCTs were included in the review (13,568 participants; range 359 to 3,655). Four were only available as conference abstracts. One trial reported generation of allocation sequence; two reported concealment of allocation, withdrawal was reported by four trials and intention-to-treat analysis for three trials. Blinding was not reported by any of the trials. Median follow-up ranged from 4.3 to 10 years.

Dexamethasone was associated with a significantly reduced event rate compared with prednisone (RR 0.80, 95% CI 0.68 to 0.94; five RCTs), although this was associated with substantial heterogeneity (Ι²=60%). Dexamethasone was also associated with decreased risk of CNS relapse (RR 0.53, 95% CI 0.44 to 0.65; six RCTs) compared with prednisone. There was no significant difference between drugs for the outcomes risk of bone marrow relapse (six RCTs), testicular relapse (two RCTs) and overall mortality (three RCTs).

Dexamethasone was associated with significantly greater risk of death during induction (RR 2.31, 95% CI 1.46 to 3.66; four RCTs), neuro-psychiatric adverse events (RR 4.55, 95% CI 2.45 to 8.46; three RCTs), myopathy (RR 7.05, 95% CI 3.00 to 16.58; two RCTs) and a significantly higher risk of withdrawal due to adverse events (RR 121.70, 95% CI 16.34 to 906.64; two RCTs). There was no significant difference between drugs in rates of osteonecrosis, sepsis, fungal infection, diabetes and pancreatitis.

The results of sensitivity and subgroup analyses were also reported.

Authors' conclusions

Dexamethasone was more efficacious than prednisone when given during induction therapy, but it was associated with more toxicity so it was unclear whether this short-term superiority would translate to superior overall survival with longer follow-up.

CRD commentary

The review question was supported by clear inclusion criteria. Three relevant databases and sources of unpublished studies were searched and grey literature was included. Study selection and data extraction were performed by two reviewers, which reduced the possibility of error and bias. It was unclear whether similar steps were taken for quality assessment. Study quality was assessed using appropriate criteria. Meta-analysis was performed, which appeared appropriate. Where heterogeneity occurred this was investigated. The authors’ conclusions were supported by the data presented and were likely to have been reliable.

Implications of the review for practice and research

Practice: The authors did not state any implications for practice.

Research: The authors stated that further analysis could better define whether the effects of dexamethasone were restricted to specific subpopulations and individual patient data meta-analysis should be considered when current studies have matured after longer follow-up. The effect of dexamethasone in induction only, using prednisone for the remaining treatment, should be investigated due to the toxicity profile of dexamethasone. The impact of different dose schedules on the side-effects of dexamethasone should also be investigated.

Funding

Swiss Cancer League; The Canadian Institute of Health Research.

Bibliographic details

Teuffel O, Kuster SP, Hunger SP, Conter V, Hitzler J, Ethier MC, Shah PS, Beyene J, Sung L. Dexamethasone versus prednisone for induction therapy in childhood acute lymphoblastic leukemia: a systematic review and meta-analysis. Leukemia 2011; 25(8): 1232-1238. [PubMed: 21527934]

Original Paper URL

http://www.nature.com/leu/journal/v25/n8/abs/leu201184a.html

Indexing Status

Subject indexing assigned by NLM

MeSH

Child; Dexamethasone /therapeutic use; Humans; Precursor Cell Lymphoblastic Leukemia-Lymphoma /drug therapy /mortality; Prednisone /therapeutic use

AccessionNumber

12011005422

Database entry date

18/06/2012

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.