Identification of randomized clinical trials (RCT) in sarcoma

Article type
Authors
Bramwell V
Abstract
Objective: Identification of completed RCT evaluating management strategies in soft tissue (STS) and bone sarcomas, to create a register for the Cochrane Cancer Network. Trials on this register would be available for inclusion in Cochrane systematic reviews.

Methods: Computer databases were searched, and relevant references down-loaded and reviewed. For MEDLINE (1966-1995) a Cochrane optimal search strategy (Dickerson et al "Systematic Reviews", Ed. Chalmers and Antman, p 17-36, BMJ Publishing Group 1995) was used. For EMBASE (1979-1995) and SCISEARCH (1981-1995) keywords Randomi* and sarcoma* were used. All references to Kaposi sarcoma/AIDS were subsequently excluded.

Results: Phase l/II (specific) of the Cochrane optimal search strategy yielded 18-164 references/yr. All were reviewed, and revealed approx 100 RCT in sarcoma. Phase III (less specific) yielded up to 500 references/yr, but in 1994 none of these were RCT in sarcoma. EMBASE yielded 267 references, but all RCT were on MEDLINE, with the exception of a 1982 Spanish trial. Similarly there were no RCT on SCISEARCH not on MEDLINE, except for one 1995 reference not yet indexed. Breakdown by disease and trial type showed soft tissue sarcoma (adjuvant) 23 RCT; soft tissue sarcoma (metastatic) 26 RCT; osteosarcoma (adjuvant) 16 RCT; Ewing's sarcoma 6 RCT; paediatric rhabdomyosarcoma 12 RCT, miscellaneous 12 RCT.

Discussion: A meta-analysis of 17 RCT of adjuvant chemotherapy in STS has been completed. There is scope for systematic reviews in treatment of metastatic STS and osteosarcoma, as well as specific treatments such as radiotherapy. However, this computerised database search will have to be supplemented by handsearches currently being conducted through the Cochrane Cancer Network.