Background:
Functional precision medicine evaluates drug response of patient's tumor cells to help guide treatment. This is particularly useful for T-cell lymphoma (TCL), which has limited treatment options. In a recent prospective study, Quadratic Phenotypic Optimization Platform (QPOP/Optim.AITM) was evaluated in relapsed/refractory non-Hodgkin's lymphoma . This study highlights the key results of the prior study and the real-world performance of Optim.AITM in clinical settings to guide TCL treatment.
Methods:
Optim.AI™ applies a hybrid experimental–analytical approach to predictively rank all clinically actionable treatments within a 12-drug panel, using fewer than one million patient-derived cells. Patient reports are provided to clinicians to support treatment selection and response to Optim.AI™-guided therapy is subsequently assessed.
Results:
Optim.AI™ demonstrated clinical concordance with an overall accuracy of 74.5% in predicting treatment outcomes. For heavily pre-treated patients who received Optim.AI™-guided therapies, there was an overall response rate of 59% and a hazard ratio of 0.56 for progression-free survival (PFS), as compared with physician's choice treatment. In the real-world setting, Optim.AITM had a 86% report generation rate and a median 4 days turnaround time. An Optim.AI™-guided PTCL patient, with four prior lines of treatments, was shown to respond to a sequential, three-drug combination; Bortezomib, Gemcitabine and Venetoclax. This resulted in a near complete response, which lasted 2.2 times longer than prior line treatment.
Conclusion:
These findings demonstrate the clinical feasibility of Optim.AITM , offering effective, personalized treatments for TCL. Future work includes combining results from Optim.AITM with genomics and expanding the evaluable classes of drugs.
