Vorinostat and fenretinide synergize in preclinical models of T-cell lymphoid malignancies.

T-cell lymphoid malignancies (TCLMs) are in need of novel and more effective therapies. The histone deacetylase (HDAC) inhibitors and the synthetic cytotoxic retinoid fenretinide have achieved durable clinical responses in T-cell lymphomas as single agents, and patients who failed prior HDAC inhibitor treatment have responded to fenretinide. We have previously shown fenretinide synergized with the class I HDAC inhibitor romidepsin in preclinical models of TCLMs. There exist some key differences between HDAC inhibitors. Therefore, we determined if the pan-HDAC inhibitor vorinostat synergizes with fenretinide. We demonstrated cytotoxic synergy between vorinostat and fenretinide in nine TCLM cell lines at clinically achievable concentrations that lacked cytotoxicity for non-malignant cells (fibroblasts and blood mononuclear cells). In vivo, vorinostat + fenretinide + ketoconazole (enhances fenretinide exposures by inhibiting fenretinide metabolism) showed greater activity in subcutaneous TCLM xenograft models than other groups. Fenretinide + vorinostat increased reactive oxygen species (ROS, measured by 2',7'-dichlorodihydrofluorescein diacetate dye), resulting in increased apoptosis (via transferase dUTP nick end labeling assay) and histone acetylation (by immunoblotting). The synergistic cytotoxicity, apoptosis, and histone acetylation of fenretinide + vorinostat was abrogated by the antioxidant vitamin C. Like romidepsin, vorinostat combined with fenretinide achieved synergistic cytotoxic activity and increased histone acetylation in preclinical models of TCLMs, but not in non-malignant cells. As vorinostat is an oral agent and not a P-glycoprotein substrate it may have advantages in such combination therapy. These data support conducting a clinical trial of vorinostat combined with fenretinide in relapsed and refractory TCLMs.

Cytotoxicity and molecular activity of fenretinide and metabolites in T-cell lymphoid malignancy, neuroblastoma, and ovarian cancer cell lines in physiological hypoxia.

OBJECTIVE: All-trans-N-(4-hydroxyphenyl)retinamide or fenretinide (4-HPR) acts by reactive oxygen species (ROS) and dihydroceramides (DHCers). In early-phase clinical trials 4-HPR has achieved complete responses in T-cell lymphomas (TCL) and neuroblastoma (NB) and signals of activity in ovarian cancer (OV). We defined the activity of 4-HPR metabolites in N-(4-methoxyphenyl)retinamide (MPR), 4-oxo-N-(4-hydroxyphenyl)retinamide (oxoHPR), and the 4-HPR isomer 13-cis-fenretinide (cis-HPR) in NB, OV, and TCL cell lines cultured in physiological hypoxia. METHODS: We compared the effect of 4-HPR, cis-HPR, oxoHPR, and MPR on cytotoxicity, ROS, and DHCers in a panel of TCL, NB, and OV cell lines cultured in bone marrow level physiological hypoxia (5% O2), utilizing a fluorescence-based cytotoxicity assay (DIMSCAN), flow cytometry, and quantitative mass spectrometry. RESULTS: 4-HPR (10 µmol/l) achieved more than three logs of cell kill in nine of 15 cell lines. Cytotoxicity of 4-HPR and oxoHPR was comparable; in some cell lines, cis-HPR cytotoxicity was lower than 4-HPR, but additive when combined with 4-HPR. MPR was not cytotoxic. ROS and DHCers were equivalently increased by 4-HPR and oxoHPR in all cell lines (P<0.01), to a lesser extent by cis-HPR (P<0.01), and not increased in response to MPR (P>0.05). Mitochondrial membrane depolarization, caspase-3 cleavage, and apoptosis (TUNEL) were all significantly increased by 4-HPR and oxoHPR (P<0.01). CONCLUSION: Cytotoxic and pharmacodynamic activity was comparable with 4-HPR and oxoHPR, lower with cis-HPR, and MPR was inactive. Neither MPR or cis-HPR antagonized 4-HPR activity. These data support focusing on achieving high 4-HPR exposures for maximizing antineoplastic activity.

Cytotoxic activity of difluoromethylornithine compared with fenretinide in neuroblastoma cell lines.

BACKGROUND: Maintenance therapy with 13-cis-retinoic acid and immunotherapy (given after completion of intensive cytotoxic therapy) improves outcome for high-risk neuroblastoma patients. The synthetic retinoid fenretinide (4-HPR) achieved multiple complete responses in relapse/refractory neuroblastoma in early-phase clinical trials, has low systemic toxicity, and has been considered for maintenance therapy clinical trials. Difluoromethylornithine (DFMO, an irreversible inhibitor of ornithine decarboxylase with minimal single-agent clinical response data) is being used for maintenance therapy of neuroblastoma. We evaluated the cytotoxic activity of DFMO and fenretinide in neuroblastoma cell lines. PROCEDURE: We tested 16 neuroblastoma cell lines in bone marrow-level hypoxia (5% O2 ) using the DIMSCAN cytotoxicity assay. Polyamines were measured by HPLC-mass spectrometry and apoptosis by transferase dUTP nick end labeling (TUNEL) using flow cytometry. RESULTS: At clinically achievable levels (100 µM), DFMO significantly decreased (P < 0.05) polyamine putrescine and achieved modest cytotoxicity (<1 log (90% cytotoxicity). Prolonged exposures (7 days) or culture in 2% and 20% O2 did not enhance DFMO cytotoxicity. However, fenretinide (10 µM) even at a concentration lower than clinically achievable in neuroblastoma patients (20 µM) induced ≥ 1 log cell kill in 14 cell lines. The average IC90 and IC99 of fenretinide was 4.7 ± 1 µM and 9.9 ± 1.8 µM, respectively. DFMO did not induce a significant increase (P > 0.05) in apoptosis (TUNEL assay). Apoptosis by fenretinide was significantly higher (P < 0.001) compared with DFMO or controls. CONCLUSIONS: DFMO as a single agent has minimal cytotoxic activity for neuroblastoma cell lines.