Identifying Novel Epigenetic Therapies for T-Cell Lymphomas By High Throughput Drug Screen

Introduction: Epigenetic alterations are known to contribute to development of T-cell lymphomas (TCL), and numerous epigenetic modifiers have shown safety and clinical efficacy in TCL leading to FDA approval of three histone deacetylase inhibiters (vorinostat, romidepsin, and belinostat). As single-agent therapies in relapsed/refractory TCLs overall response rates are ~30% with median time to response of ~2 months and duration of response typically less than 1 year. Belinostat is generally well tolerated with infrequent high-grade toxicities in relapsed TCL. It would be ideal to identify drugs that could be combined with epigenetic therapies, such as belinostat, to improve efficacy and overcome resistance. Further, identification of highly cytotoxic drugs that could provide rapid tumor debulking while preserving non-malignant T-cells and are broadly applicable across subtypes, would be beneficial to rapidly alleviate symptoms and prevent decline in a generally older population.

Methods: We obtained a drug library containing 729 epigenetic and cytotoxic compounds tested in human clinical trials (or FDA approved) therefore ideal for Phase 2 repurposing, and we screened these compounds' cytotoxic properties against 5 TCL cell lines representative of different TCL subtypes in vitro: HH (CTCL), Hut78 (CTCL), OCI-Ly13.2 (PTCL-NOS), Karpas299 (ALCL), and HDMAR2 (LBL). We treated the five TCL cell lines for 48 hours at four concentrations (1nM, 10nM, 100nM, and 1000nM) for each of the 729 compounds. Following 48 hours of drug, the TCL cell lines were rinsed and given fresh media. After another 17 hours of incubation cell viability was assessed by flow cytometry using a viability dye validated against 7-AAD and Annexin V. From the original 729, we identified the compounds that exhibited the most cytotoxicity and further elucidated their efficacy across a range of 8 concentrations on the 5 prior TCL cell lines; in addition we evaluated cytotoxicity on 5 more TCL cell lines: MAC-1 (CTCL), MAC-2A (ALCL), SMZ-1 (PTCL), T8ML-1 (PTCL-NOS), and DERL-2 (PTCL). We tested the top compounds against 3 unique healthy donor activated T-cells at these 8 concentrations. The same protocol using 48-hour drug exposure followed by 17 hours of incubation was utilized. At the conclusion of this screen, we tested cytotoxicity of our most encouraging compound in combination with other top performing compounds.

Results: We identified BI-847325, bortezomib, camptothecin, CUDC-907, and WAY-118959-A as the most encouraging cytotoxic compounds that are not currently being used clinically. We compared the top 5 compounds to 4 standard of care (SOC) cytotoxic drugs (belinostat, gemcitabine, romidepsin, and vincristine) that are used clinically. We also evaluated effects of the top 5 compounds (and the 4 SOC compounds) on non-malignant healthy donor T-cells. We observed that camptothecin was the most encouraging novel compound, demonstrating high cytotoxicity while sparing healthy non-malignant T-cells. We tested cytotoxicity of camptothecin in combination with 5 compounds (belinostat, bortezomib, BI-847325, romidepsin, and vincristine) that also demonstrated cytotoxicity in the malignant T-cells with relative sparing of healthy T-cells. In vitro and using Bliss modeling, we observed potential synergy when combining belinostat to camptothecin.

Conclusions: Our preclinical studies show that when combined with belinostat, the dose of camptothecin can be decreased by 50%, while still observing increased tumor cell killing and sparing non-malignant T-cells, thereby decreasing risk of dose limiting toxicities observed in prior camptothecin clinical trials. In vivo studies are ongoing to identify the dose and administration schedule that can be studied in a clinical trial and to establish how exatecan is synergistic with belinostat.

No relevant conflicts of interest to declare.