Artikel
Reproducibility of in vitro glioma research exemplified by the U-87 MG model
Reproduzierbarkeit von in vitro Hirntumorforschung am Beispiel des U-87 MG Modells
Suche in Medline nach
Autoren
Veröffentlicht: | 25. Mai 2022 |
---|
Gliederung
Text
Objective: The reproducibility of research results marks one of the foundations of translational preclinical science and receives rising awareness by major stakeholders of biomedical research such as the National Institutes of Health and Nature. Considering this development, our project investigates the reproducibility of a commonly performed in vitro drug test assay and aims to identify drivers of irreproducibility. Thereby, we focus on preclinical in vitro research on glioblastoma, a malignant tumour with largely unmet need for improvement in therapeutic outcome.
Methods: We conducted a systematic search on PubMed, Embase and Web of Science for literature that describes the effect of the first-line chemotherapeutic agent Temozolomide on the viability of U-87 MG glioblastoma cells. Accompanying, we extracted a series of experimental parameters that in theory could affect drug sensitivity of U-87 MG cells, e.g., cell concentration, age, and culture medium components. In a subsequent meta-analysis, we quantified the reporting of these parameters and evaluated associations between parameter characteristics, reporting, and the reproducibility of TMZ sensitivity.
Results: In total, 137 of 1158 unique articles found in the search were included into the review. The overall reporting of experimental features was deficient (e.g., cell concentration reported in only 16.8% of the articles). However, recent but slight improvement in reporting habits was observed (p = 0.011). Interestingly, the experimental results of the same experiment (U-87 MG cell viability reduction caused by Temozolomide treatment) were highly heterogeneous across the included studies (p < 0.001). At this, the cell culture mediums glucose level was identified as a significant driver of irreproducibility (p = 0.016).
Conclusion: Based on our findings, we strongly support the establishment of consensus minimum reporting standards for preclinical research including parameters being able to influence the comparability of results across studies. Concretely, for in vitro cell viability assays, the mediums glucose level should be explicitly reported. In perspective, we hope to increase scientists' awareness of the importance of reproducible preclinical research results as a potential contribution to higher translation rates of promising in vitro findings into neuro-oncological care, not only for glioblastoma.