Onvansertib Induces Anti-Leukemic Activity at Higher Doses in Relapsed/Refractory Acute Myeloid Leukemia

June 16, 2020
Danielle Ternyila

“While the trial is still ongoing, we are encouraged by the efficacy we are seeing thus far in patients with relapsed/refractory AML, particularly the durability of response observed in some patients."

Higher dose levels of onvansertib in combination with decitabine (Dacogen) demonstrated durable responses as treatment of patients with relapsed/refractory acute myeloid leukemia (AML), and the agent also had anti-leukemic activity and appeared well-tolerated, according to final phase 1b and preliminary phase 2 results presented during the virtual 25th Congress of the European Hematology Association (EHA).1

“While the trial is still ongoing, we are encouraged by the efficacy we are seeing thus far in patients with relapsed/refractory AML, particularly the durability of response observed in some patients,” stated Amer Zeidan,MBBS, MHS, lead investigator and associate professor of Medicine at Yale School of Medicine, and medical director of Hematology Early Therapeutics Research at Yale Cancer Center. “As we continue with enrollment and assessment of efficacy in the Phase 2 portion of the trial, I look forward to seeing additional clinical benefit with the combination of onvansertib and decitabine in our patients in an indication that is in dire need of new safe and effective treatment options.”

Twenty-one out of 23 patients completed at least 1 cycle of treatment in the phase 1b portion of the study. Seven patients had an objective response, and 5 patients had a complete response (CR) or a complete response with incomplete hematologic recovery (CRi). The median time to achieve a CR or CRi was 2.5 months (range, 0.7-8.0 months), and 3 patients remained on treatment since the initial response at 6, 12, and 15 months, respectively. One patient received a transplant after achieving a CR, and 1 patient progressed after 2.5 months.2

In the phase 2 portion, 2 out of 13 patients achieved an objective response (28%). One patient had a CRi at cycle 1 and a CR at cycle 2, and this patient remained on treatment after 3 months in response. Another patient had a partial response at cycle 1 and remained on treatment.

Of the patients who achieved a CR/CRi (n = 6), the median age was 72 years old (range, 51-76 yrs.), and 4 of the patients had relapsed or were refractory to intensive chemotherapy while 2 patients were treatment-naïve for AML but had received hypomethylating agents for myelodysplastic syndrome. The cytogenetic risk among these patients was intermediate in 4 patients and adverse in 2. Two patients had an NRAS mutation, 2 had an SRSF2, 2 had FLT-ITD, 1 had SF3B1, 1 had FLT3-TKD, 1 had DNMT3A, and 1 had IDH2.

The recommended phase 2 dose was determined as 60 mg/m2, and treatment was well tolerated for the first 5 dose levels, which ranged from 12 to 60 mg/m2. Two of the 6 patients who received 90 mg/m2 onvansertib had a dose-limiting toxicity, which was grade 3 mucositis and grade 4 rash.

The most common any-grade treatment-related adverse events (TRAEs) occurring in greater than 10% of patients were febrile neutropenia (39%), rash (36%), anemia (33%), fatigue (28%), nausea (28%), thrombocytopenia (25%), stomatitis (25%), and neutropenia (22%). Most AEs were grades 1 or 2.

The most common grade 3 to 5 AEs included hematological AEs like neutropenia, thrombocytopenia, and white blood cell decrease; febrile neutropenia; and skin toxicities, including rash and stomatitis, which were observed mostly in the higher onvansertib doses of 60 and 90 mg/m2. The only grade 5 AE was febrile neutropenia (3%).

Four patients had to discontinue treatment due to an AE, which included grade 3 mucositis (n = 2), grade 3 rash (n = 1), and grade 4 rash (n = 1). A total of 59 serious AEs occurred in 25 patients, 14 of which were possibly related to the study drug. Five patients died while on treatment, and all deaths were related to either AML or complications of the disease, such as progressive disease, pneumonia, septic shock, infection, or sepsis.

In the biomarker analysis, circulating tumor DNA (ctDNA), a noninvasive biomarker was used to monitor tumor heterogeneity, response to treatment, minimal residual disease, and disease progression in AML and MDS. A driver mutation was identified in 17 patients by targeted next-generation sequencing, and this was confirmed with droplet digital PCR. The mutant allele frequency (MAF) was measured at baseline and the end of cycle 1, and changes in MAF were calculated as Log2.

A decrease n ctDNA MAF at cycle 1 was observed in all 6 patients who had a CR or CRi, and only 1 of the 11 non responders had a similar decrease. Investigators noted that measuring the changes in plasma over the first cycle was highly predictive of clinical response, with 100% sensitivity, 91% specificity, 86% positive predictive value, and 100% negative predictive value.

Phosphorylated translational control tumor protein (pTCTP) was identified as a specific marker for PLK1 activity in vivo in preclinical models, the authors noted. Capillary Western-Blot was used to assess pTCTP and TCTP, which is a substrate of PLK1, in mononuclear cells that were isolated from blood samples on day 1 prior to treatment and 3 hours after dosing in patients with ≥10% circulating blasts (n = 15).

According to the investigators, the target engagement was not associated with the study drug dose, but an association was observed with higher bone marrow responses. Four of the 6 patients with target engagement (67%) had a ≥20% reduction in blast compared with 3 of the 9 patients who were non-target engagement (33%). Additionally, 2 patients in the target engagement group had a CR or CRi (33%) compared with 1 of the non-target engagement patients who were in a morphologic leukemia-free state.

Overall, target engagement in circulating blasts was associated with a greater decrease in bone marrow blasts, according to the biomarker analysis. Decreases in mutant ctDNA after 1 cycle of treatment with onvansertib was highly predictive of clinical response.

In the phase 1b, onvansertib was given at 12 mg/m2 in 4 patients, 18 mg/m2 in 3 patients, 27 mg/m2 in 3 patients, 40 mg/m2 in 4 patients, 60 mg/m2 in 3 patients and 90 mg/m2 in 6 patients. The study drug was administered at 60 mg/m2 for the phase 2 dose, which included 13 patients who have been treated for the preliminary analysis. The median age of patients was 66 years (range, 33-77) in the phase 1b and 73 years (range, 23-85) in the phase 2. The majority of patients were men in both portions of the study, so men made up 65% of the cohort in phase 1b and 62% in the phase 2. The majority of patients also had an ECOG performance status of 1 in the phase 1b (range, 0-2) and phase 2 (range, 0-2).

Overall, 13 patients in the phase 1b had received 1 prior line of therapy (57%), 6 received 2 or more lines (26%), and 4 received no prior treatment (17%). All patients in the phase 2 portion had received 1 prior line of therapy (100%). Thirteen patients in the phase 1b had adverse cytogenetic risk (57%), 9 had intermediate risk (39%), and 1 had favorable risk (4%), compared with the phase 2 arm where 6 patients had adverse risk (46%), 6 had intermediate risk (46%), and 1 had favorable risk (8%).

Patients with relapsed/refractory AML who had received up to 3 prior lines of therapy were included in the phase 1b portion of the study, while patients who received 1 prior line of therapy were included in the phase 2 portion. Patients who had not received prior treatment or had treatment-related AML or APL were excluded from the study.

The study drug was given at a starting dose of 12 mg/m2 on days 1 through 5, and decitabine was administered at 20 mg/m2 on days 1 through 5 in a 28-day cycle. The purpose of the phase 1b portion of the trial, which was a dose-escalation phase with 3+3 design, was to determine the maximum tolerated dose or recommended phase 2 dose. Investigators used a 50% incremental dose increase in 3 cohorts of patients, and dose-limiting toxicities were evaluated during the first cycle.

The phase 2 portion of the study used the recommended dose determined in the phase 1b. The primary objective of the expansion phase of the study was to evaluate the safety, tolerability, and anti-leukemic activity of the recommended phase 2 dose. Other end points included pharmacokinetics, evaluation of predictive biomarkers associated with response to treatment, and target engagement in circulating leukemic cells.

Patients with relapsed/refractory AML have limited treatment options, especially patients who do not harbor a FLT3 or IDH1/2 mutation. The outcomes for this patient population remain dismal, with an expected median overall survival of less than 6 months.

Onvansertib is an oral, highly-selective inhibitor of Polo-like kinase 1 (PLK1), which is a serine/threonine kinase that is a master regulator of mitosis and appears to be overexpressed in patients with AML. This third generation PLK1 inhibitor had a half-life of 24 hours. Preclinical models demonstrated that the agent had potent anti-tumor activity as a single agent and in combination with other treatments.

The phase 2 portion of the study is currently ongoing and enrolling patients. This study will assess the safety and efficacy of onvansertib, as well as biomarker strategies. The maximum tolerated dose was established as 60 mg/m2, which is the recommended phase 2 dose, and this dose showed no dose-limiting toxicities.


  1. Cardiff Oncology Data Continues to Demonstrate Efficacy, Durability and Safety of Onvansertib in Patients with Difficult-to-Treat Relapsed/Refractory AML. Press Release. June 16, 2020. Accessed June 16, 2020. https://bit.ly/2UPqIPe
  2. Zeidan AM, Lin TL, Becker PS, et al. Phase 1b/2 study of onvansertib, a PLK1 inhibitor, in combination with decitabine in patients with relapsed or refractory acute myeloid leukemia. Poster presented at: 2020 European Hematology Association Annual Congress; June 11-21, 2020.