Discussion

To mimic the initial avascular stages of solid tumors in vivo, spheroids need to develop hypoxia and subsequent necrosis in their center. Indeed, due to the presence of quiescent cells, spheroids are more resistant to certain compounds compared to cells grown as conventional monolayers and therefore considered a more relevant model for drug screening. Widely used methods to obtain multicellular spheroids involve culturing malignant cells in anchorage‐independent growth conditions, generating a single spheroid per well and have poor compatibility with high throughput and HCS approaches.
We have developed a simple method to grow 9 spheroids per well in 96 well plate format. Spheroids are attached to the glass bottom and are directly compatible with conventional biochemical readouts as well as High Content Analysis without troublesome plate transfer. Our method is fully compatible with automation and allows spheroids to develop up to 500 µm in diameter with interwell CVs of 5%.
Using biochemical readouts and a customized image analysis script, basic characteristics of the spheroids were validated including i) growth curves, ii) kinetics of necrotic center formation, iii) presence of dividing cells at the periphery and iv) response to standard compounds. In addition, we were able to characterize differences in the MOA of diverse chemotherapeutic drugs and their dose-response profiles.
Overall, our results show that multiple highly reproducible spheroids per well can contribute to improved throughput, reliability and convenience compared to other methods, and improve characterization of drug effects, thanks to complete compatibility with HCA approaches.