(1) New marketing authorizations continue to be granted for treatments of multiple myeloma, and new trials and meta-analyses continue to be published. This review re-examines our previous conclusions based on data published between 2003 and 2008. We again used the standard Prescrire methodology to review the latest data; (2) In patients who are symptomatic but who do not qualify for haematopoietic stem cell transplantation (especially people aged over 65), the results of five comparative trials suggest that adding thalidomide to the melphalan-prednisone combination delays myeloma progression by an additional 5 to 10 months. There is possibly also an increase in overall survival time. This triple combination is therefore a first-line treatment option. There is no firm evidence that replacing thalidomide with bortezomib in this combination provides an advantage; (3) In symptomatic patients aged under 65, two meta-analyses have compared high-dose chemotherapy followed by autologous stem cell transplantation with conventional chemotherapy. Unlike early trials, these meta-analyses showed no overall survival benefit but only a delay in myeloma progression. Initial treatment with two successive transplantation procedures has a negative risk-benefit balance. (4) The optimal chemotherapy regimen prior to autologous stem cell transplantation is controversial. It is unclear which combination (vincristine + doxorubicin + dexamethasone, cyclophosphamide + dexamethasone, cyclophosphamide + dexamethasone, or bortezomib + dexamethasone, etc.) has a better risk-benefit balance in terms of survival and quality of life; (5) According to a meta-analysis of three clinical trials, thalidomide maintenance therapy appears to improve overall survival after Autologous stem cell transplantation; (6) Despite their inadequate evaluation, lenalidomide and pegylated liposomal doxorubicin are licensed for use in patients who relapse or who are refractory to initial treatment. In view of their major adverse effects, we consider that these drugs should only be used in clinical trials.