A Targeted Drug Screen Reveals Potent Combinatorial Inhibitors of T-PLL Primary Patient Cells

Background

T-cell prolymphocytic leukemia (T-PLL), the most frequent subgroup of the mature T-cell leukemias/lymphomas (MaTCL), is a rare and aggressive disease with no curative treatment and a median survival of less than two years. T-PLL has a poor treatment prognosis because of resistance to standard chemotherapies and lack of targeted therapies. The most effective treatment option to date is alemtuzumab, an anti-CD52 monoclonal antibody. However, relapses are unavoidable.

Aims

The aim of this study was to find anti-cancer drugs that alone or in combination inhibit T-PLL cell viability in primary patient cells ex vivo, as well as molecular determinants of the drug response differences between the patients.

Methods

We performed a targeted drug screen of eight potentially clinically relevant drugs both as single agents and in 28 combinations, on frozen T-PLL peripheral blood mononuclear cells (PBMCs, >80% leukemic cells) from 20 patients. The tested drugs were idasanutlin (MDM2 inhibitor), cladribine (purine analog), venetoclax (Bcl-2 inhibitor), ruxolitinib (JAK inhibitor), romidepsin (histone deacetylase inhibitor), dinaciclib (CDK inhibitor), ibrutinib (BTK inhibitor) and bendamustine (alkylating agent). PBMCs from two healthy donors were used as controls. After incubating the cells with the drugs for 72 hours, cell viability was measured with a luminescent CellTiter-Glo assay. Drug sensitivity score (DSS) was calculated for the individual drugs and combination synergy scores were determined using the zero-interaction potency (ZIP) model. To investigate transcriptomic changes associated with the drug response profiles, whole-transcriptome RNA sequencing was performed on the same 20 T-PLL patient samples.

Results

Idasanutlin, romidepsin, dinaciclib and venetoclax were the most effective and selective drugs at inhibiting T-PLL cell viability across majority of the patients. Ruxolitinib and cladribine were effective in a subset of the patients (4/20 and 13/20). RNA sequencing analyses revealed that higher expression of protocadherin beta 15 (PCDHB15), a plasma membrane protein functioning in cell adhesion, is linked to the sensitivity to ruxolitinib treatment ex vivo. In addition, male T-PLL patients seemed to be more sensitive to cladribine treatment (7/8), compared to females (4/10, with 2 females excluded as intermediate responders), p=0.07 (Fisher's exact test). Idasanutlin in combination with cladribine resulted in the most synergistic combination, but only in selected patients (8/20).

Conclusion

Transcriptomic changes together with gender may explain variable responses to targeted treatments in T-PLL cells ex vivo, suggesting personalized treatment options.