This review compared clodronate, pamidronate, and zoledronate with placebo for reducing morbidity and mortality in cancer patients with metastatic bone disease and concluded that skeletal-related events were improved, but mortality was not. Although a well-conducted review, the findings should be interpreted with caution due to a lack of good-quality data, small sample sizes and uncertainty over publication bias.

To compare the efficacy of clodronate, pamidronate and zoledronate against placebo for reducing morbidity and overall mortality in cancer patients with metastatic bone disease.

MEDLINE, EMBASE and Cochrane Central Register of Controlled Trials (CENTRAL) were searched from inception to January 2009 for English, French, Italian, Portuguese and Spanish-language publications; search terms were reported. Reference lists of included and excluded studies and reviews were searched for additional articles.

Placebo-controlled randomised controlled trials (RCTs) that used clodronate, pamidronate or zoledronate in therapeutic doses and assessed skeletal-related events or mortality from any cause in patients with a definite (biopsy-proven) diagnosis of metastatic bone disease were eligible for inclusion. Skeletal-related events were evaluated as either: pathologic fractures; radiation to bone or bone surgery; spinal cord compression; or hypercalcaemia.

The included studies evaluated oral clodronate (range 1,600 to 2,080mg/day), intravenous pamidronate (range 45 to 90mg/every three or four weeks), oral pamidronate (range 150 to 300mg/day) and intravenous zoledronate (range 4 to 8mg/every three or four weeks). Most studies assessed patients with breast cancer and at least one metastatic bone lesion. Additional studies assessed myeloma, prostate cancer, lung cancer and various other solid tumours.

Two reviewers independently selected studies for inclusion in the review.

Two reviewers independently assessed study quality using the 5-point Jadad scale, which assessed randomisation, blinding, withdrawals and dropouts. A high score was defined as at least 3 points and a low score less than 3. Disagreements were resolved by consensus.

Two reviewers independently extracted the number of patients with pathologic fractures, radiation to bone or bone surgery, hypercalcaemia and mortality from any cause to calculate the relative risk (RR) with 95% confidence intervals (CIs) for each study; discrepancies were resolved through consensus.

Pooled RRs and their 95% confidence intervals (CIs) were calculated using a random-effects model. Heterogeneity was assessed using the Χ² and I² tests (a value >20% was considered to be high). Publication bias was assessed using Begg and Mazumdar statistics and funnel plots. Sensitivity analysis was undertaken using study quality (high score versus low score).

A total of 18 RCTs were included in the review; nine were non blinded and nine double-blind. It was not possible to calculate the number of participants in the included studies. The mean quality score was 2.89: two studies scored 1; four studies scored 2; eight studies scored 3; two studies scored 4; and two studies scored 5. The included studies were too small to detect the presence or absence of publication bias.

Skeletal-related events:

Compared with placebo, all three bisphosphonates yielded a significant reduction in skeletal-related events for cancer patients with metastatic bone disease: zoledronate (RR 0.70, 95% CI 0.61 to 0.81; four studies, n=1,211); pamidronate (RR 0.81, 95% CI 0.73 to 0.91; five studies, n=2,251); and clodronate (RR 0.87, 95% CI 0.75 to 1.00; four studies, n=681). Heterogeneity was absent for all comparisons except for pamidronate. Compared with placebo, a significant reduction in pathologic fractures was observed for clodronate (RR 0.76, 95% CI 0.64 to 0.90; six studies, n=1,096) and zoledronate (RR 0.60, 95% CI 0.47 to 0.76; three studies, n=1,165), but not for pamidronate; heterogeneity was absent for all comparisons. In the sensitivity analysis the results for all skeletal-related events for zoledronate and pamidronate were sensitive to the study quality and those for clodronate were not.

Radiation to bone or bone surgery:

Compared with placebo, a significant reduction was observed for pamidronate (RR 0.72, 95% CI 0.62 to 0.84; six studies, n=2,127) and zoledronate (RR 0.67, 95% CI 0.46 to 0.97; three studies, n=1,165); heterogeneity was present for these comparisons.

Hypercalcaemia:

Compared with placebo, a significant reduction was observed for zoledronate (RR 0.27, 95% CI 0.10 to 0.72; two studies, n=743), pamidronate (RR 0.60, 95% CI 0.41 to 0.86; six studies, n=2,127) and for clodronate (RR 0.73, 95% CI 0.56 to 0.97; five studies, n=1,079); heterogeneity was absent for all comparisons.

None of the bisphosphonates significantly improved overall mortality compared with placebo.

Clodronate, pamidronate and zoledronate were associated with reductions in morbidity in cancer patients with metastatic bone disease with regard to preventing skeletal-related events, but were not associated with a reduction in overall mortality.

The review question and inclusion criteria were clear. The authors searched three relevant databases for publications in English, French, Italian, Portuguese and Spanish and it was unclear whether unpublished studies were sought; therefore, language bias could have been present and some studies may have been missed. Publication bias was assessed, but the included studies were too small to detect its presence or absence. All stages of the review process appeared to have been conducted in duplicate, which reduced the potential for error and bias. Appropriate criteria were used to assess the quality of the included studies; most only achieved a Jadad score of 3 or less. The included studies were generally of poor quality. Results of the quality assessment were not reported for each study, so it was unclear on which criteria each study failed. Sample sizes of the included studies were not reported, so a total sample size could not be calculated. The authors stated that the analyses may have been influenced by the small sample sizes of the included studies. Suitable methods were used for the meta-analysis. Heterogeneity was assessed and found to be absent from most analyses for significant outcomes. Generally this is a well-conducted review, but the findings were limited by a lack of good-quality data, small sample sizes and uncertainty over publication bias and should be interpreted with caution.

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

Research: The authors stated that evaluation of tolerability data and comparisons with other bisphosphonates should be considered for future research.

None.

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.