Identification of Tipifarnib Sensitivity Biomarkers in T-Cell Tumor Cell Lines

Background

T-cell leukemia and lymphoma (TCL) is a very aggressive and heterogeneous group of hematological malignancies with poor prognosis and inadequate response to current therapies. The standard first-line treatments have resulted in unsatisfactory patient outcomes and unfortunately, targeted treatments are still at a preliminary phase. The RAS/MAPK pathway is crucial for TCR signaling of T-cells and it is deregulated in TCL. We aimed to determine the therapeutic value of farnesyl transferase inhibitor (FTIs) tipifarnib in TCL cell lines. Effects on cell viability, apoptosis, cell cycle and gene expression were evaluated. Tipifarnib blocks the localization of some farnesylated proteins, including some RAS family members, to cellular membranes, thereby inhibiting their activation. Tipifarnib is a potent and specific FTI with prominent anti-proliferative effects.

Methods

We tested tipifarnib in 18 T-cell leukemia and 4 T-cell lymphoma cell lines for in vitro sensitivity and for biomarker discovery, both genomic and immunohistochemical. We selected cell lines with available genomic data from COSMIC, CCLE or generated by our group. The MAPK, NFAT, NFKB and JAK/STAT pathways were tested by immunohistochemistry analysis under basal conditions. The range of drug concentrations to perform IC₅₀ analysis was established between 0-10,000 nM (ten points), using CellTiter-Glo® Luminescent Cell Viability Assay kit (Promega, Madison, WI, USA) following manufacturer´s instructions at 0h, 48h and 96h. All experiments were done in sextuplet and all numerical data were expressed as the average of the values ± the standard error of the mean. IC₅₀ analyses were performed with GraphPad Prism v5. Clinically-relevant drug sensitivity was defined as IC₅₀ <100nM at 96h (defined as sensitive). RNAseq was performed in 4 sensitive cell lines with tipifarnib and DMSO (vehicle) in three biological replicates at 72h. Differential gene expression (DE) analysis was carried out under these two conditions. Pathway analysis was done using Gene Set Enrichment Analysis (GSEA). Induction of apoptosis and cell viability were tested by flow cytometry in sensitive cell lines.

Results

59.1% (n=13) of cell lines were sensitive to tipifarnib at concentrations which are readily achievable in the clinic. RAS, RAS-guanine nucleotide exchange factors (GEFs) and RAS-GTPase activating proteins (GAPs) genes were mutated in 45.5%, 50% and 27.3%, of cell lines respectively, but their mutational status was not associated with drug sensitivity. We found TP53 and DNMT3 mutated in 80% and 33.3% in sensitive cell lines respectively, and in 63.6% and 0% in resistant cell lines. The mutational state of NOTCH1 was associated with tipifarnib sensitivity; 66.7% in sensitive cell lines versus 9.1% in resistant cell lines (p=0.005). ERK activation (MAPK pathway) was significantly associated with drug sensitivity (p=0.046). Conversely, RelB (NFκB pathway) was associated with drug resistance (p=0.007). GSEA analysis showed that tipifarnib downregulated cell cycle, protein localization to membranes, metabolism, and ribosome and mitochondrial activity. The effect of tipifarnib on cell cycle progression and apoptosis was validated by flow cytometry assays. Tipifarnib decreased cellular viability, increased apoptosis and blocked cell cycle progression in G1 phase. GSEA of molecular pathways regulated by tipifarnib identified downregulation of MYC targets and mTOR and MAPK pathways. MYC is a target of NOTCH1, mutated in sensitive cases and associated with tipifarnib sensitivity.

Conclusions

In vitro response to tipifarnib in T-cell lymphoma and leukemia cell lines can be predicted using several biomarkers. The mutational NOTCH1 status, together with p-ERK (MAPK activation) were associated with increased sensitivity while RelB expression was associated with resistance.