ABSTRACT
Intro: Deregulated protein synthesis is a common trait across solid and hematologic malignancies and an attractive target for cancer therapy. Rocaglates compounds that inhibit eukaryotic initiation factor 4A1 (eIF4A1), the essential DEAD-box RNA helicase that resolves mRNA 5'UTR secondary structures during cap-dependent translation initiation. Rocaglates' unique mechanism of action causes sequence-selective mRNA binding by eIF4A1, clamping the inactive helicase onto the transcript. This suppresses translation globally and affects many oncogenic and pro-survival transcripts in particular. Zotatifin, the first-in class synthetic rocaglate, is currently in Phase I clinical trials for the treatment of solid tumors and as an antiviral against SARS-CoV2. Currently, eIF4A1 and DDX3 are the only reported targets of rocaglate-mediated RNA clamping. Employing unbiased proteomic approaches, we have discovered that rocaglates, thought to act as pure eIF4A/translation inhibitors, extensively remodel the translation machinery and translatome. Additionally, mass-spec interrogation for proteins interacting with specific RNA sequences reveals novel targets of rocaglate-mediated, sequence-specific RNA clamping. Methods: We conducted original mass-spectrometry analyses of translational reprogramming by rocaglates. TMT-pSILAC assessed acute changes in protein production, while MATRIX, which captures high-resolution profiles of the translation machinery, revealed translation factors that drive reprogramming in response to rocaglate exposure. We validated results biochemically, in cellulo, and in vivo using patient-derived xenograft (PDX) mouse models. To probe existing and novel rocaglate RNA-clamping targets, we developed unbiased “clampome” assays - in cellulo protein-RNA-pull downs followed by mass-spec analysis of proteins with increased binding to RNA in the presence of rocaglates. Results: We find rocaglates, including zotatifin, have effects far more complex than simple “translational inhibition” as currently defined. Indeed, translatome analysis by TMT-pSILAC revealed myriad up-regulated proteins that drive hitherto unrecognized cytotoxic mechanisms. The GEF-H1 guanine exchange factor, for example, drives anti-survival RHOA/JNK activation, suggesting novel candidate biomarkers of rocaglate clinical outcomes. Translation-machinery analysis by MATRIX identifed rocaglate-induced dependence on specific translation factors including eEF1ϵ1 that drive remodeling. Novel rocaglate RNA-binding targets revealed by clampome studies remain under detailed evaluation as mediators of drug activities. Discussion: Our original proteome-level interrogation revealed that the complete cellular response to these historical “translation inhibitors” is mediated by comprehensive translational landscape remodeling. Effects on a broader suite of RNA binding proteins than eIF4A1 alone we suggest mediate the potent antitumor activities of these unique compounds, elucidation of which permits development of novel precision approaches to targeted translational deregulation in cancer.
ABSTRACT
BACKGROUND Patients with peripheral T-cell lymphoma (PTCL) lack good treatment options, particularly in the relapsed and refractory setting (Mak V et al. J Clin Oncol 2013). The development of the targeted therapies in PTCL has been lagging behind those developed for B cell lymphomas. Our work suggested that combinations of epigenetic therapies can be a safe and effective approach for patients with PTCL, particularly those with T-cell lymphomas with a follicular helper phenotype (Marchi E et al. Br. J Haematol 2015;O'Connor O.A. et al;Blood 2019;Falchi L et al. Blood 2020). While the reason for this is not clear, it is thought recurrent mutations in epigenetic factors, including Ten-Eleven Translocation-2 (TET2), DNA methyl transferase-3A (DNMT3A) and isocitrate dehydrogenase-2 (IDH2) may contribute for their increased vulnerability (Couronné L. et al. N Eng J Med 2012;Lemonnier F et al. Blood 2012). Despite these presumptions, a direct explanation for the sensitivity to epigenetic based treatment remains to be established. OBJECTIVES To evaluate the merits of romidepsin plus subcutaneous azacitidine in patients with PTCL when administered in a ‘real-world’ scenario. METHODS We retrospectively identified PTCL patients that were treated with azacitidine and romidepsin outside of a clinical trial based upon queries regarding off study use. The study was reviewed and approved by each Medical Center Institutional Review Board. We have identified 13 patients world-wide whose pretreatment characteristics are shown in Table 1. These patients were treated using 3 different schedules: Schedule A: azacitidine 75mg/m2 s.c. on days 1-7, romidepsin 14 mg/m2 on day 1, 8 and 15 of a 28 day cycle (total of 6 patients);Schedule B: azacitidine 75mg/m2 s.c. on days 1-5, romidepsin 14 mg/m2 on day 8, 15 and 22 of a 35 day cycle;and Schedule C (total of 2 patients): azacitidine 75mg/m2 s.c. on days 1-7, romidepsin 12-14 mg/m2 on day 8, 15 and 22 of a 28 day cycle (total of 5 patients). RESULTS We retrospectively identified 13 patients that were treated with romidepsin and azacitidine off study. Ten patients had angioimmunoblastic lymphoma (AITL), 2 had adult T-cell leukemia/lymphoma (ATLL) and 1 had PTCL-NOS. Eight of the 13 patients had next generation sequencing performed. Most common mutations found were those of TET2 (5 pts), RHOA (4pts), IDH2 (3pts) and DNMT3A (1 pt). One ATLL patient had mutations in TRAF3, FAT1 and MED12. Among these 13 patients, overall response rate (ORR) was 84% and the complete response rate (CR) was 61%. Median number of cycles was 3 (range 1-12). Treatment was well tolerated but notable adverse effects included nausea, fatigue, rash, neutropenia and thrombocytopenia. One patient experienced febrile neutropenia while another had pulmonary infiltrates (differential diagnosis included drug toxicity versus infection). Thrombocytopenia was the most common reason for dose reduction of romidepsin (to 12mg/m2) or its omission on day 8, 15 or 22. In 3 patients, azacitidine and romidepsin were used to achieve remission prior to allogeneic transplant (range of cycles 1-3), with all 3 patients were in CR at their last disease assessment. One patient died of transplant related mortality 8 months after his allogeneic stem cell transplant. There was 1 patient with AITL (treatment naïve) noted to have progression of disease at first imaging following 2 cycles of romidepsin and azacitidine. On the day of her PET/CT, she was however diagnosed with symptomatic Covid19 infection and was hospitalized. A repeat PET/CT 6 weeks later (without any additional lymphoma treatment) revealed PR. CONCLUSIONS Subcutaneous azacitidine and romidepsin administered in a ‘real-world’ situation is highly effective in patients with relapsed PTCL with tolerable toxicity, and can be used to successfully bridge patients to stem cell transplant. Notably, the efficacy was similar to the one reported on a clinical study with oral azacitidine and romidepsin. [Formula presented] Disclosures: Kalac: Astra Zeneca: Consultancy;Kyowa Kirin Consultancy;Gilead: Consultancy;Johnson and Johnson: Research Funding;Guidepoint: Consultancy;GLG: Consultancy. Tam: Beigene: Research Funding;Janssen: Research Funding;Abbvie: Research Funding;Loxo: Honoraria;Beigene: Honoraria;Janssen: Honoraria;Abbvie: Honoraria. Montanari: Seattle Genetics: Research Funding. O'Connor: Servier: Research Funding;Mundipharma: Honoraria;Myeloid Therapeutics: Current equity holder in publicly-traded company, Honoraria, Membership on an entity's Board of Directors or advisory committees;Kymera: Current equity holder in publicly-traded company, Honoraria, Membership on an entity's Board of Directors or advisory committees;Astex: Research Funding;BMS: Research Funding;Merck: Research Funding;TG Therapeutics: Current Employment, Current equity holder in publicly-traded company. Marchi: BMS: Research Funding;Astex: Research Funding;Merck: Research Funding;Myeloid Therapeutics: Honoraria;Kyowa Kirin: Honoraria;Kymera Therapeutics: Other: Scientific Advisor.