Background: The CHMG has conducted four meta-analyses using individual patient data (IPD) in recent years. Methods: After defining the planned analyses in a protocol, randomized controlled trials (RCTs) were systematically searched. Once eligible RCTs were identified, authors were asked to provide IPD. Data were collected, standardized and checked for consistency before analysis. Using IPD allows us to address relevant problems in oncology, e.g. identifying prognostic factors, subgroups analyses, investigating secondary endpoints and/or analyzing time-to-event data. These IPD-meta-analyses were mostly publicly funded in Germany. Results: An IPD-meta-analysis with 9312 patients from 37 RCTs was conducted to assess the risk of second malignancies associated with treatment of Hodgkin lymphoma (HL). To identify prognostic factors for progression-free survival in early-unfavorable-stage HL patients, information from 4490 patients in six RCTs were collected. Another IPD-meta-analysis comparing high dose chemotherapy with autologous stem cell support in aggressive non-Hodgkin lymphoma included 2132 patients from 11 RCTs. To determine the effect of recombinant human erythropoiesis-stimulating agents on survival, 13933 cancer patients from 53 RCTs were analyzed. Data for the IPD meta-analyses were contributed both by pharmaceutical companies and independent investigators. Feasibility of an IPD meta-analyses depends on the willingness and the resources of investigators to provide data in standardized formats. The workload required varies during the different phases of meta-analysis and depends on the origin and quality of the data submitted. Conclusions: Even if IPD meta-analyses are time and resource intensive, the CHMG has shown that they can be successfully conducted: IPD meta-analysis provide evidence to questions which are not answerable otherwise. The success of IPD meta-analyses depends on the willingness of investigators to share their clinical trial data and we are grateful that clinical study groups as well as pharmaceutical companies provided data. However, employing standardized definitions for clinical parameters and endpoints in RCTs would facilitate such analyses.