SY-1425, a Potent and Selective RARα Agonist, in Combination with Azacitidine Demonstrates a High Complete Response Rate and a Rapid Onset of Response in RARA-Positive Newly Diagnosed Unfit Acute Myeloid Leukemia

Introduction: A novel genomically defined subset of non-APL AML patients (pts) with an actionable target characterized by overexpression of RARA has been identified (McKeown, Cancer Discovery 2017). RARA-positive (RARA+) pts can be detected by a novel blood-based biomarker test that predicts sensitivity to SY-1425, an oral potent and selective RARα agonist (Vigil, ESH 2017). Approximately 30% of newly diagnosed (ND) unfit AML pts are RARA+ (de Botton, ESH 2019), and recent data demonstrates RARA+ AML is correlated with venetoclax resistance, which may suggest a new poor prognostic risk (Fiore, ASH 2020). Preclinical synergy of SY-1425 with azacitidine (aza) supported clinical development of the combination (McKeown, Haematologica 2018), which is currently being evaluated in RARA+ ND unfit AML (NCT02807558). Early data of SY-1425 + aza demonstrated high response rates and a rapid onset of responses in the RARA+ ND unfit AML cohort (Cook, ASH 2018; de Botton, ESH 2019). Updated data for the fully enrolled cohort are presented here.

Methods: Both RARA+ and RARA-negative (RARA-) ND unfit AML pts were enrolled, and all pts received aza at 75 mg/m² IV/SC daily on days 1-7 followed by SY-1425 at 6 mg/m²/day PO in divided doses twice daily on days 8-28 of each 28-day cycle. The primary objective was characterization of activity by the overall response rate (ORR) per IWG. Analyses of the composite complete response rate, time to response, comparison of responses across RARA+ and RARA- cohorts, transfusion independence, and safety were also included.

Results: A total of 51 pts were treated, 22 RARA+ and 29 RARA-, with data available through 27 May 2020 reported. Baseline characteristics included 32 (63%) male; median age 76 (60-91); 17 (33%) poor and 33 (65%) intermediate cytogenetic risk; 29 (57%) de novo and 22 (43%) secondary AML; and 33 (65%) with marrow blasts > 30%. Time on treatment was up to 18.2 months, with discontinuation of treatment primarily due to AE (33%), the majority being not treatment-related, and disease progression (22%).

Among 18 response-evaluable RARA+ pts, analysis for best response showed an ORR of 67% (12/18), with complete responses in 61% (11/18; 9 CR, 2 CRi, 0 CRh) and MLFS in 6% (1 pt). CRs showed a high rate (89%, 8/9) of complete molecular or cytogenetic remissions (3 molecular, 5 cytogenetic). Median time to initial complete response was rapid at 1.2 months. Transfusion independence (TI) was achieved in 13 (72%) of 18 TI evaluable pts, including conversion to TI in all 6 (100%) who were transfusion dependent at baseline.

Of the 28 response-evaluable RARA- pts, the ORR and complete response rate were lower and time to initial response was longer relative to RARA+ pts. ORR was 39%, with complete responses in 29% (8/28; 6 CR, 2 CRi, 0 CRh). Median time to initial complete response was 2.9 months.

The AE profile of the combination is consistent with that previously reported for single-agent SY-1425 or aza in AML. Most common AEs (all grades/causality) included decreased appetite (41%; 21/51 pts), hypertriglyceridemia (41%), nausea (39%), constipation (39%), fatigue (35%), peripheral edema (35%), pyrexia (33%) and febrile neutropenia (33%). Hematologic AEs ≥ grade 3 included febrile neutropenia (33%), thrombocytopenia (24%), anemia (20%) and neutropenia (16%). The majority of non-hematologic AEs were low grade. The most frequent SAEs included febrile neutropenia (27%), pyrexia (12%), pneumonia (12%), and sepsis (10%).

Conclusion: SY-1425 + aza was generally well-tolerated without evidence of increased toxicity of the combination relative to either as a single agent. Importantly, rates of myelosuppression were comparable to single-agent aza in this population. There was evidence of clinical activity with a high CR rate and rapid time to initial response, with evidence of clinical benefit associated with TI in a majority of RARA+ ND unfit AML pts. Use of the novel blood-based biomarker test to predict sensitivity to SY-1425 in AML pts was supported by clinical outcomes, with higher response rates and a more rapid onset in RARA+ vs RARA- pts, whose responses were consistent with single-agent aza. SY-1425 + aza shows potential as a novel targeted regimen for the treatment of RARA+ ND unfit AML and warrants further development in this genomically defined subset of AML pts, as well as other related RARA+ pt populations. Updated data, including mature duration of response and survival, will be presented.