We're working on meta-analysis of studies that provide median OS data. We conducted other meta-analyses using objective response rates (ORR) and used the inverse-variance method. But we're not sure what method to use for OS. I'm comfortable using R on packages like "meta" or "metafor" and I can use a new package if necessary.
Second problem is that some studies did not provide a confidence interval, and I read a systematic review (attached below) and they dealt with this problem by "repeating the analysis 1000 times on a bootstrap sample", I need some explanation on how we can do this.
Thanks a lot!
Article Cisplatin/gemcitabine or oxaliplatin/gemcitabine in the trea...
Dear Sebawe Syaj
1) For the time-to-event data likes overall survival, I believe that the pooled estimates by using the median OS are not an appropriate way:
- "It is not appropriate to analyze time-to-event data using methods for continuous outcomes (e.g., using mean times-to-event), as the relevant times are only known for the subset of participants who have had the event. The most appropriate way of summarizing time-to-event data is to use methods of survival analysis and express the intervention effect as a hazard ratio." Ref: https://training.cochrane.org/handbook/current/chapter-06#section-6-8
- Moreover, "using median survival times or survival rates at a particular point in time are not reasonable surrogate measures for meta-analyses of survival outcomes." Ref: https://www.cambridge.org/core/journals/international-journal-of-technology-assessment-in-health-care/article/metaanalysis-when-only-the-median-survival-times-are-known-a-comparison-with-individual-patient-data-results/CEADCE539D51BD9EBF68FA58EA072745
2) For the bootstrap sampling, I think you can find useful data in http://www.metafor-project.org/doku.php/tips:bootstrapping_with_ma
To Nguyen-Phong Vo ,
Dr. Vo, thanks a lot for answering my question,
I'm aware that using HR to express time-to-event data is the best way.
However, I forgot to specify my problem; we're doing a single-arm meta-analysis
So unfortunately, there was no way that we could extract an HR. Nevertheless, we conducted a meta-analysis for median OS using the weighted median of medians methods described by McGrath et al* . I'd like to have your opinion on it if you know whether we should proceed with this or not, thank you.Article One-sample aggregate data meta-analysis of medians
Dear Sebawe Syaj,
For your specific problems, I think you could consider these below papers:
1) Combining median survival estimates in the Appendix of this article: Article Confronting publication bias: A cohort design for meta-analysis
2) In case your single-arm studies provided the survival curve, you can consider another approach: Article Meta-analysis of single-arm survival studies: A distribution...
PS: I need to say something:
- The pooled results from single-arm studies maybe not make sense in some journals due to the absence of the control group.
- I don't know that the methods described by McGrath et al. are suitable for time-to-event data or not because the time-to-event data is different than standard continuous data.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R