Multiple myeloma is an incurable hematological malignancy that relies on drug combinations as first and secondary lines of treatment. The inclusion of proteasome inhibitors, such as bortezomib, into these drug combination regimens has improved median survival. Resistance to bortezomib, however, is a common occurrence that ultimately contributes to treatment failure. Thus, there remains a need to identify improved drug combinations that may serve as later lines of treatment.
We have developed the quantitative parabolic optimization platform (QPOP) to optimize drug combinations against bortezomib-resistant multiple myeloma. By mapping phenotypic output data to parabolic response surfaces, QPOP is able to deterministically optimize drug combinations as well as drug dosages.
We have successfully identified potential optimal drug combinations against bortezomib-resistant RPMI 8226 (P100v) cell line as projected via QPOP, with these combinations exhibiting synergistic response surface maps. The drug combinations displayed lower half-maximal inhibitory concentrations (IC50) in vitro as compared to single drug administration. While QPOP does not rely on molecular mechanism prediction, the identified optimal drug combinations can reverse DNA hypermethylation and silencing of tumor suppressors that occurs following acquired bortezomib-resistance. Prolonged survival of P100v tumor-bearing mice was observed when these optimized combinations were validated in vivo, further highlighting the importance of treating in vitro and in vivo as two separate entities. Moreover, the drug combination is broadly effective across a range of primary multiple myeloma patient samples.
These results collectively show that QPOP is a robust platform that is able to eradicate the bortezomib-resistant clones of multiple myeloma. Beyond bortezomib-resistant multiple myeloma, global optimization of drug combinations by QPOP can serve to improve drug combination design across a range of other cancers and diseases through a continuous optimization process across the entire drug development pipeline.
Clinical trial identification
Legal entity responsible for the study
National Research Foundation Cancer Science Institute of Singapore RCE Main Grant
All authors have declared no conflicts of interest.