Non-Hodgkin’s lymphoma (NHL) is a heterogenous malignancy with many different subtypes. Although patients initially respond to treatments, approximately 40% of this cohort will subsequently relapse. After exhausting several lines of treatment, determining the next option is largely empirical. There is a need to develop approaches that are able to identify personalized therapies. There are existing ex vivo approaches that use single-drug sensitivity assays but lack the ability to truly evaluate drug combinations. In this clinical study, we aim to evaluate the potential of an ex vivo combinatorial drug sensitivity platform, Optim.AI™, to identify therapeutic options for relapsed/refractory (RR) NHL. Repeatability, between-run and intermediate precision parameters were used to evaluate the analytical precision of the approach. Optim.AI™ was carried out using K422 and SU-DHL-4 cell lines, either on same or different days by two operators. The cell viability was quantified post-drug treatment and used to calculate the standard deviation (SD) and coefficient of variation (CV) between runs. 98 patients with RR-NHL of both B-NHL and NK/T-NHL origin were analyzed. Single cell suspensions from tumor biopsies or blood aspirates were treated with a panel of drugs with known efficacy. Post-drug treatment cell viability was used as phenotypic input for Optim.AI™ analysis, mapping experimental data points to a second-order quadratic function to predict all other cell killing efficacies. Optim.AI™ results were shared with treating physicians and Optim.AI™-guided therapy was offered in the absence of standard options. For the analytical precision validation, the percentage of replicates with CV ≤ 30% followed the same trend for both cell lines, where 92.2%, 96.1% and 92.6% for repeatability, between-run and intermediate precision parameters respectively were observed for K422 cell line. Out of the 98 successfully generated reports from the processed samples, 35 patient results were evaluable upon analysis where 19 patients were given Optim.AI™ top ranked therapies while the rest were given non-top ranked options, as per physicians’ discretion. Approximately 67% of those with guided therapies achieved clinical benefit, with 7 patients observing complete response while 6 patients attained partial response. For the 35 patients analyzed, the sensitivity of the study was found to be 85% (95% CI, 64-94.8%) while the specificity was at 86.7% (95% CI, 62.1-96.3%). This prospective study demonstrated the precision utility of Optim.AI™, evident from the high percentage of replicates falling within the defined CV. More importantly, it highlights the clinical utility of Optim.AI™ in identifying patient-specific therapies that improved clinical outcomes. The high sensitivity and specificity scores provide confidence in the reliability of the approach to accurately predict efficacious options. The results of this study show the potential of functional precision medicine in identifying patient-specific therapy, while improving precision and personalized cancer treatment.

Citation Format: Masturah Mohd Abdul Rashid, Jhin Jieh Lim, Lisa Chow, Edward Kai-Hua Chow. Analytical and clinical evaluation of a functional combinatorial precision medicine platform [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A128.