Phase 1 study of zavondemstat (TACH101), a first-in-class KDM4 inhibitor, in patients with advanced solid tumors

Abstract

Background

This was a first-in-human, phase I, dose-escalation study evaluating the safety, pharmacokinetics, and preliminary efficacy of zavondemstat (TACH101), an epigenetic targeting inhibitor of KDM4 histone demethylase, in patients with heavily pre-treated advanced/metastatic cancers.

Patients and Methods

Patients received zavondemstat orally on a weekly schedule in 28-day cycles. Dose escalation followed a Bayesian optimal interval design and explored both intermittent and continuous dosing. The primary objectives were to assess safety, dose-limiting toxicities (DLTs), maximum tolerated dose (MTD), and recommended phase II dose (RP2D). Secondary objectives included pharmacokinetics and radiographic response per Response Evaluation Criteria in Solid Tumors, version 1.1.

Results

Thirty patients were enrolled across 6 dose cohorts. MTD was not reached at the maximum dose tested. The most common treatment-related adverse events (TRAEs) were diarrhea (12%), fatigue (7%), decreased appetite (7%), nausea (7%), and hyponatremia (7%). All TRAEs were grade 1 or 2. No serious TRAEs or DLTs were reported. Of 23 response-evaluable patients, 10 (44%) achieved stable disease (SD). Two patients (9%) had SD ≥ 6 months, including a patient with castration-resistant prostate cancer and a patient with leiomyosarcoma. A third patient (leiomyosarcoma) receiving ongoing treatment with zavondemstat under compassionate use has had SD for 6+ months. Zavondemstat demonstrated a dose-proportional exposure profile with a half-life of about 1.5 hours. There was no to minimal drug accumulation observed.

Conclusions

Zavondemstat was very well tolerated and showed encouraging preliminary clinical benefit in heavily pretreated patients with advanced cancer. Continued evaluation of zavondemstat is warranted.

Implications for Practice.

This is the first evaluation of KDM4 inhibition in a clinical setting. In this phase I study, zavondemstat, a pan-isoform KDM4 inhibitor, demonstrated a favorable benefit/risk profile in patients with heavily pretreated advanced cancers. The safety profile for zavondemstat was excellent, with only grade 1 and 2 adverse events reported, including at the highest dose evaluated. No serious adverse events were observed. Preliminary antitumor activity showed a stable disease rate of 44%, with SD ≥ 6 months in 3 patients. These findings suggest that KDM4 inhibition is a promising novel therapeutic strategy in heavily pre-treated patients with advanced cancer.

Introduction

KDM4 (histone lysine demethylase 4) is part of the Jumonji C (JmjC) family of epigenetic regulators that play a crucial role in embryonic development, gene transcription, cell cycle progression, and DNA repair.^1-5 KDM4 regulates gene transcription and expression by removing methyl groups from histones and controlling the accessibility of DNA to transcription factors, influencing which genes are turned on or off as well as their level of expression. KDM4 overamplification or dysregulation has been observed in various cancers and is known to promote critical tumorigenic processes, including replicative immortality, apoptosis evasion, metastasis, DNA repair deficiency, and genomic instability^4,6-13 (Figure 1). Thus, dysregulation of KDM4 has been linked to the activation of various oncogenic pathways correlating with heightened cancer aggressiveness and poorer clinical outcomes.^12-15

Figure 1.

KDM4 dysregulation contributes to cancer progression by promoting multiple oncogenic pathways. Specifically, KDM4 inhibits apoptosis, leading to increased survival of abnormal cells. It enhances androgen receptor (AR) and estrogen receptor (ER) activity, driving hormone-dependent cancer progression. KDM4 promotes genomic instability through DNA damage, facilitates metastasis by enabling cancer cell migration, and stimulates uncontrolled cell proliferation. Collectively, these mechanisms underscore the role of KDM4 as a key driver of tumorigenesis. “Created with Biorender.com.”

KDM4 consists of four primary isoforms: KDM4A, KDM4B, KDM4C, and KDM4D. Functional redundancy and cross-activity have been observed across the KDM4 isoforms; therefore, pan-inhibition targeting all KDM4 isoforms is required to fully suppress KDM4 activity.^1,16 Zavondemstat (TACH101) is the first KDM4 inhibitor to enter clinical development. It selectively and potently inhibits KDM4 isoforms A-D by competing with KDM4’s co-factor, alpha-ketoglutarate (α-KG), for binding to KDM4’s catalytic domains.¹⁷ Preclinical studies demonstrated robust anti-proliferative effects and significant inhibition of tumor growth across numerous cell-line-derived and patient-derived xenograft models.^17-19 Zavondemstat exhibited favorable oral bioavailability and dose-dependent exposure, accompanied by manageable toxicities, after oral administration in rats and dogs (data on file, Tachyon Therapeutics, Inc.).

This first-in-human, phase I, dose-escalation study (ClinicalTrials.gov: NCT05076552) enrolled patients at 6 sites in the United States to characterize the safety, tolerability, pharmacokinetics (PK), and preliminary efficacy of single-agent zavondemstat in patients with heavily pre-treated advanced metastatic cancers. The study completed enrollment of 6 dose cohorts; however, it did not fully explore higher-dose cohorts due to the sponsor’s business decision to discontinue the study.

Patients and methods

Study design

In this open-label, phase I, dose-escalation study, patients were treated with a single dose of zavondemstat during a lead-in period, followed by repeated dosing on a weekly schedule. Dose escalation followed a Bayesian optimal interval design, with up to 6 patients enrolled per cohort. The primary endpoints were the safety and tolerability of zavondemstat and the maximum tolerated dose (MTD) and the recommended phase II dose (RP2D). The secondary objectives were to evaluate the PK and preliminary antitumor activity of zavondemstat per the Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST v1.1).²⁰ Treatment was administered in 28-day cycles. Tumor measurements were performed every 8 weeks.

The study was conducted in accordance with the International Council for Harmonisation Guideline on Good Clinical Practice, with applicable local regulations, and the Declaration of Helsinki. The protocol was reviewed and approved by independent ethics committees at each study site. All patients were required to provide written informed consent prior to enrollment.

Patient selection

Eligible patients were ≥18 years of age and had metastatic or advanced metastatic cancers that had progressed or were non-responsive to available therapies and for which no standard therapy exists. The study inclusion criteria were measurable disease according to RECIST (v1.1), an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1, adequate hematologic function (absolute neutrophil count ≥1500/µL, platelets ≥100 000/µL, and hemoglobin >9.0 mg/dL), adequate renal function (estimated creatinine clearance ≥60 mL/min according to the Cockcroft-Gault formula), adequate hepatic function (bilirubin ≤ 1.5 × upper limit of normal [ULN], alanine transaminase and aspartate transaminase ≤ 3 × ULN, and alkaline phosphatase ≤ 2.5 × ULN), and adequate cardiovascular function (no myocardial infarction within 6 months and no uncontrolled angina within 3 months prior to study entry, no history of congestive heart failure [New York Heart Association class III or IV], no QT interval corrected using the Fridericia formula >450 millisecond, no clinically significant ventricular arrhythmias, and no uncontrolled hypertension). Exclusion criteria included brain metastases. Patients were not pre-selected on the basis of gene mutational status. Full eligibility criteria are available in the protocol (Supplementary Material).

Drug treatment

Zavondemstat was administered orally (capsule formulation) on a weekly regimen (3 days on/4 days off, 5 days on/2 days off, or 7 days on) following a 28-day cycle until documented progression, unacceptable toxicity, or withdrawal of consent. Patients were treated with escalating doses of zavondemstat, ranging from 5 mg daily, 3 days on/4 days off to 25 mg BID daily.

Dose escalation

Decisions to escalate or de-escalate the dose were guided by the Bayesian optimal interval design,²¹ which was chosen to minimize the probability of inappropriate dose assignments and to identify the MTD sooner, while allowing flexibility in enrollment to the cohorts during dose escalation by considering data across all cohorts. The target DLT rate was 30% with an escalation boundary of 0.236 and a de-escalation boundary of 0.359 (to prevent subtherapeutic and overly toxic dose levels). Generally, 3 patients were treated in each cohort until an MTD was defined. Decisions to escalate or de-escalate the dose were made by the Safety Review Committee (SRC) after all patients in a cohort had completed ≥28 days of dosing (the DLT assessment period, starting from day 1) and after all the safety data including vital sign measurements, physical examination findings, 12-lead electrocardiography results, treatment-related adverse events (TRAEs), clinical laboratory test results, concomitant medications, and available PK data had been reviewed by the SRC. Dose escalation was conducted cohort by cohort until an intolerable dose was reached. In the event that an intolerable dose was not reached, the sponsor and investigators were to define an RP2D. New patients were treated in the next-higher dose cohort if the previous dose level had been established as safe (no TRAEs greater than Grade 2) by the SRC. Intra-patient dose escalation was allowed if the patient had completed at least 4 weeks of treatment at the originally assigned dose, had not experienced a TRAE greater than Grade 2, and the safety of the dose to be escalated to had been established (no greater than Grade 2 TRAEs) in patients who had received treatment for at least 4 weeks.

Assessments

Routine safety assessments included evaluations of adverse events, laboratory tests (including hematology, coagulation, biochemistry, and urinalysis), vital signs, and 12-lead electrocardiograms. Adverse events were categorized based on the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. A DLT was defined as an adverse event or abnormal laboratory finding that was unrelated to the disease, disease progression, intercurrent illness, or concurrent medications; occurring within 28 days of the first zavondemstat dose; and meeting any protocol-specified DLT criteria.

Tumor assessments were conducted at baseline and every 8 weeks using computed tomography (CT) or magnetic resonance imaging (MRI) until disease progression or study discontinuation. Local investigators evaluated tumor responses based on RECIST version 1.1 criteria. A response was considered confirmed if it was verified within 4 to 8 weeks following the initial observation.

Pharmacokinetic assessments

Blood samples were collected at various time points before treatment and up to 24 hours after treatment to measure zavondemstat concentrations. The PK analysis dataset included all participants who received at least one dose of the study drug and had at least one evaluable PK concentration. The PK parameters assessed included maximum observed plasma concentration (C_max), time to C_max (T_max), the area under the plasma concentration-time curve from the time of dosing extrapolated to infinity (AUC_inf), the area under the concentration-time curve during a dosing interval at steady-state (AUCτau), terminal elimination half-life (t_1/2), apparent clearance (CL/F), apparent volume of distribution during the terminal elimination phase (Vz/F), accumulation ratio calculated from AUCτ (day 15), and AUC₀₋₂₄ h for once-daily dosing or AUC₀₋₁₂ h for twice-daily dosing (PK lead-in period or day 1).

Statistical analysis

Continuous data were reported using descriptive statistics (ie, number, mean, standard deviation, median, minimum, and maximum). Categorical data were reported using the number and percentage in each category. No formal power calculation was performed, but simulations of the chosen Bayesian optimal interval design estimated mean sample sizes for different scenarios during the dose escalation. No formal interim analysis was conducted for the study; however, the data for each dosing cohort were reviewed prior to each dose escalation. PK parameters were estimated using noncompartmental methods with Phoenix WinNonlin, version 8.4 (Certara USA, Inc., Princeton, NJ).

Results

Patient characteristics

Between March 2023 and July 2024, a total of 30 patients with advanced metastatic cancers were enrolled in this phase I study and received treatment with zavondemstat. Baseline demographic, tumor, and treatment characteristics are shown in Table 1. Nineteen patients (63%) were male, and 11 patients (37%) were female. The median age was 58 years (range: 37-78 years). The majority of patients (80%) had an ECOG performance status of 1. The most frequent tumor type of enrolled patients was colorectal cancer (40%), followed by prostate (13%) and pancreatic (10%) cancers. The median number of prior lines of therapy was 6 (range: 1-13), which included chemotherapy (93%), targeted therapy (73%), radiation therapy (53%), immunotherapy (50%), and other investigational agents (40%). The median time from the date of diagnosis to the date of initiation of treatment was 3.8 years (range, 0.5-12.6 years; Figure 2).

Table 1.

Patient demographics and baseline characteristics.

Characteristics	N = 30
Median age (range), years	58 (37-78)

Sex, n (%)
Male	19 (63%)
Female	11 (37%)

Race and ethnic groups, n (%)
Non-Hispanic White	21 (70%)
Hispanic or Latino	4 (14%)
Asian	3 (10%)
Native American	1 (3%)
Black or African American	1 (3%)

ECOG performance status at screening, n (%)
0	6 (20%)
1	24 (80%)

Tumor types
Colorectal	12 (40%)
Prostate	4 (13%)
Pancreatic	3 (10%)
Leiomyosarcoma	2 (7%)
Other*	9 (30%)

Prior therapy
Median no. lines of prior therapy (range)	6 (1-13)
Lines of therapy, n (%)
1	1 (3%)
2	1 (3%)
3	4 (13%)
≥ 4	24 (80%)
Type of prior therapy, n (%)
Chemotherapy	28 (93%)
Targeted therapy	22 (73%)
Radiation therapy	16 (53%)
Immunotherapy	15 (50%)
Investigational agent	12 (40%)

ECOG, Eastern Cooperative Oncology Group.

^*Other tumor types: appendiceal cancer, breast cancer, uveal melanoma, esophageal cancer, endometrial cancer, thyroid cancer, sarcoma, mesothelioma, cancer of unknown origin.

Figure 2.

Twenty-three of the thirty patients had at least one post-baseline scan and were evaluable for response. * 5 × 5 mg zavondemstat capsules. ** 7 patients were not evaluated for response due to withdrawal of consent (n = 2) or early study discontinuation due to clinical/disease progression (n = 5). Abbreviations: CRPC, castration-resistant prostate cancer; Unk, unknown; CRC, colorectal cancer; QD, once daily; BID, twice daily.

Dose escalation

Patients were enrolled across 6 dose cohorts (Table 2): Cohort 1 (n = 3) received zavondemstat at 5 mg QD on 3 days on/4 days off regimen; Cohort 2 (n = 3) received zavondemstat at 10 mg QD on 3 days on/4 days off regimen; Cohort 3 (n = 3) received zavondemstat at 25 mg QD (using a 25mg capsule) on 3 days on/4 days off regimen. PK analyses from patients in Cohort 3 showed lower-than-expected drug exposure which may be due to issues with capsule dissolution. Therefore, to fully assess the exposure and safety of the 25 mg dose, 3 additional subjects were enrolled into Cohort 3 (designated as Cohort 3a) and were administered 5 × 5 mg capsules (for the same total dose of 25 mg).

Table 2.

Zavondemstat dose cohorts.

Cohort	Zavondemstat dose (daily total)	Dosing frequency	Weekly regimen	# Patients enrolled
1	5 mg*	QD	3 days on/4 days off	3
2	10 mg*	QD	3 days on/4 days off	3
3	25 mg**	QD	3 days on/4 days off	3
3a	25 mg*	QD	3 days on/4 days off	6
4	25 mg*	QD	5 days on/2 days off	6
5	15 mg*	BID		4
6	25 mg*	BID		5
Total				30

QD = once daily; BID = twice daily.

^*using 5-mg capsules.

^**using 25-mg capsules.

The same dose was used in Cohort 3 and 3a. The difference between these cohorts was in the capsule strength.

In order to determine an optimal balance between efficacy, safety, and tolerability, Cohorts 4-6 continued dose escalation (using 5 mg capsule strength) and evaluated zavondemstat at different dosing regimens. Cohort 4 (n = 6) received zavondemstat at 25 mg QD administered on 5 days on/2 days off regimen. Cohort 5 (n = 4) received zavondemstat at 15 mg BID administered on a continuous (7 days on) regimen. Cohort 6 (n = 5) received zavondemstat at 25 mg BID on a continuous (7 days on) regimen. No dose-limiting toxicities (DLTs) occurred at any of the doses explored during dose escalation. Due to the sponsor’s business decision to discontinue the study, higher-dose cohorts were not fully explored and the MTD for zavondemstat was not reached.

Safety and tolerability

The median duration of treatment with zavondemstat across all dose levels was 2 months (range: 0.3-9.5 months). The most common TRAEs were diarrhea (12%), fatigue (7%), decreased appetite (7%), nausea (7%), and hyponatremia (7%; Table 3). The incidence of hematologic toxicity was as follows: leukopenia: G1, 16.7%; G2, 3.3%; neutropenia: G1, 3.3%; G2, 6.7%; anemia: G1, 36.7%; G2, 23.3%; G3, 6.7%; and thrombocytopenia: G1, 13.3%, and G2, 3.3% (Supplementary Table S1). All TRAEs were Grade 1 or 2. No AEs ≥ Grade 3 (either treatment-related or not) were reported and no treatment-related serious adverse events were reported. No dose modifications or permanent study discontinuation was required due to TRAEs.

Table 3.

Treatment-related adverse events (TRAEs).

TRAE, n (%)	Total number of subjects treated with zavondemstat (N = 30)
	Grade 1	Grade 2	≥ Grade 3	Total
Any TRAE	29 (67)	14 (33)	0	43 (100)
Diarrhea	3 (7)	2 (5)	0	5 (12)
Fatigue	1 (2)	2 (5)	0	3 (7)
Decreased appetite	2 (5)	1 (2)	0	3 (7)
Nausea	2 (5)	1 (2)	0	3 (7)
Hyponatremia	2 (5)	1 (2)	0	3 (7)
Pruritus	1 (2)	1 (2)	0	2 (5)
ALP increase	1 (2)	1 (2)	0	2 (5)
Abdominal pain	2 (5)	0	0	2 (5)
Tremor	2 (5)	0	0	2 (5)
Dyspnea	0	1 (2)	0	1 (2)
Worsening hyperkalemia	0	1 (2)	0	1 (2)
Weight loss	0	1 (2)	0	1 (2)
Dry eye	0	1 (2)	0	1 (2)
Elevated WBC	1 (2)	0	0	1 (2)
Edema (left ankle)	1 (2)	0	0	1 (2)

Abbreviations: ALP, alkaline phosphatase; WBC, white blood cell count.

Antitumor activity

Twenty-three (77%) of 30 patients were evaluable for tumor response (had at least one post-baseline scan). SD was noted in 10 patients (44%), including 2 patients (9%) with SD lasting ≥ 24 weeks (castration-resistant prostate cancer [CRPC], n = 1; leiomyosarcoma, n = 1) and 1 patient with SD ≥ 32 weeks (CRPC; Figure 2). Another patient with leiomyosarcoma is continuing to receive zavondemstat under compassionate use, demonstrating SD ≥ 24 weeks to date (SD ≥ 16 weeks at the time of data cutoff; October 10, 2024). Due to the difficulty in obtaining optional tumor biopsies, the KDM4 status of these patients was not evaluated.

Pharmacokinetics

Concentration-time profiles of zavondemstat in plasma were generally well characterized following single oral doses of 5 to 25 mg. Mean plasma concentrations increased with higher dose levels. Peak concentrations were reached between 1 and 4 hours post-dose across all doses studied (Figure 3A). Both overall exposure (AUC) and peak exposure (C_max) of zavondemstat increased with increasing dose levels. The median half-life (t_1/2) was similar across all doses, ranging from approximately 1.4 to 2.6 hours (Table 4). Following multiple oral doses of 5 to 25 mg once or twice daily, peak concentrations of zavondemstat were reached between 1.3 and 3.5 hours post-dose (Figure 3B). Both overall exposure (AUC_tau) and peak exposure (C_max) increased with increasing dose levels, with the accumulation ratios for C_max and AUC_tau ranging from 0.7 to 2.9 (Table 4).

Figure 3.

Serial blood samples were collected to measure concentrations of zavondemstat (TACH101) before and up to 24 hours after treatment. Mean (± SD) plasma concentration-time plots of zavondemstat are shown for the (A) single-dose lead-in period (day 1) and (B) multiple-dose period (day 15).

Table 4.

Summary of pharmacokinetic parameters of zavondemstat.

	Parametera (unit)	Cohort 1 5 mg (n = 3)	Cohort 2 10 mg (n = 3)	Cohort 3 25 mg (n = 3)	Cohort 3A 5 × 5 mg (n = 6)	Cohort 4 5 × 5 mg (n = 6)	Cohort 5 3 × 5 mg BID (n = 4)	Cohort 6 5 × 5 mg BID (n = 5)
Day 1	AUC0_24 (h*ng/mL)	51.74 (20.53)	173.88 (NA)	NA	295.54 (32.68)	287.61 (29.86)	431.37 (26.46)	361.82 (35.8)
	AUCinf (h*ng/mL)	51.84 (20.63)	173.99 (NA)	NA	295.98 (32.6)	290.71 (29.18)	455.27 (22.44)	365.21 (34.68)
	AUC%Extrap (%)	9.85 (24.56)	2.75 (NA)	NA	2.88 (59.52)	3.58 (52.48)	6.34 (73.49)	3.51 (71.83)
	Cmax (ng/mL)	19.4 (45.54)	40.4 (60.71)	6.76 (68.29)	140.37 (45.78)	129.28 (60.01)	108.43 (73.49)	158.24 (38.53)
	Clast (ng/mL)	1.7 (38.13)	3.92 (40.39)	2.99 (9.53)	4.18 (79.22)	3.53 (46.57)	7.53 (64.53)	4.37 (60.49)
	Tmax (h)	2 (1, 2)	4 (2, 4)	4 (2, 24)	1 (1, 2)	1.5 (1, 2)	3 (2, 4)	2 (1, 2)
	t1/2 (h)	2.63 (2.16, 3.1)	1.49 (1.49, 1.49)	NA	1.42 (1.05, 2.25)	1.54 (1.23, 6.28)	1.52 (1.12, 14.33)	1.63 (1.44, 2.07)
	Cl/F (L/h)	98.54 (20.63)	57.48 (NA)	NA	95.67 (45.12)	92.52 (30.17)	34.14 (21.08)	75.04 (32.53)
	Vz/F (L)	364.49 (4.72)	123.26 (NA)	NA	194.19 (35.14)	307.9 (85.6)	240.3 (142.99)	187.11 (45.66)
Day 15	AUCtau (h*ng/mL)	53.36 (NA)	108.24 (23.69)	27.69 (NA)	416.14 (70.35)	359.43 (61.72)	410.14 (64.1)	517.41 (17.85)
	Cavg (ng/mL)	2.22 (NA)	4.51 (23.69)	1.15 (NA)	17.34 (70.35)	14.98 (61.72)	34.18 (64.1)	43.12 (17.85)
	Cmax (ng/mL)	34.47 (49.65)	40.4 (12.87)	11.55 (15.31)	100.05 (85.46)	170.1 (87.54)	99.25 (48.21)	202 (11.2)
	Cmin (ng/mL)	0 (NA)	0 (NA)	0 (NA)	0 (NA)	0 (NA)	10.49 (104.1)	1.04 (141.42)
	Fluctuation% (%)	1065.91 (NA)	971.06 (15.29)	892.89 (NA)	995.86 (76.21)	1019.13 (29.49)	293.38 (47.2)	469.18 (7.39)
	Tmax (h)	1.67 (34.64)	2.33 (65.47)	3 (47.14)	3.17 (80.92)	1.33 (43.3)	3.5 (54.71)	1.5 (47.14)
	CLss/F (L/h)	93.7 (NA)	95.06 (23.69)	903 (NA)	77.58 (43.98)	86.33 (48.11)	45.32 (40.91)	49.1 (17.85)
	Vz/F (L)	288.86 (NA)	186.67 (27.93)	4512.78 (NA)	357.73 (79.65)	183.85 (58.28)	148.64 (66.6)	107.06 (64.1)
	ARCmax	1.76 (4.66)	1.28 (61.94)	2.85 (14.89)	0.89 (92.51)	0.88 (51.3)	1.32 (74.75)	1.29 (31.94)
	ARAUC24	NA	0.73 (0.73, 0.73)	NA	1.41 (0.82, 2.15)	1.04 (0.78, 1.68)	1.05 (0.97, 1.69)	2.23 (1.32, 3.15)

^aArithmetic mean (%CV) statistics presented; T_max and t_1/2 are presented as median (min, max).

BID, twice daily; AUC_0-24, area under the plasma concentration-time curve from 0 to 24 hours; NA, not applicable; AUC_inf, AUC from zero to infinity; AUC_%Extrap, extrapolated area percentage calculated by AUC_t-inf/AUC_inf; C_max, maximum concentration; C_last, the last concentration which can be measured; T_max, time to reach Cmax; t_1/2, apparent terminal elimination half-life; CL/F, apparent clearance; V_z/F, apparent volume distribution; AUC_tau, area under the concentration-time curve during a dosing interval at steady-state; C_avg, average concentration; C_min, minimum concentration; Fluctuation%, percentage of concentration fluctuation; CL_ss/F, apparent total plasma clearance at steady state; AR_Cmax, accumulation ratio for Cmax; AR_AUC24, accumulation ratio calculated by AUC24 Day 15/AUC24 Day 1.

Discussion

This is the first report on the clinical safety, tolerability, and pharmacokinetics—as well as an initial assessment of tumor efficacy—of zavondemstat, a novel epigenetic drug targeting KDM4. In this phase I dose-escalation study, zavondemstat demonstrated a favorable benefit/risk profile in patients with advanced cancers. Considering this heavily pretreated patient population (median number of prior therapies = 6), zavondemstat demonstrated an excellent safety profile, with only Grade 1 and 2 AEs reported, including at the highest dose evaluated (50 mg: 25 mg twice daily, 7 days on). No treatment-related serious adverse events were observed; no patient required dose modification or dose reduction; and study discontinuation due to any safety concerns did not occur.

Because of the favorable safety profile of zavondemstat, the MTD was not reached, and no DLTs were observed at the tested dose levels within the protocol-defined observation period. As this was a first-in-human study for a novel targeted therapeutic agent, the planned dosing strategy was conservative, with incremental changes being made to dose increases, dosing frequency, and dosing regimen; this approach was chosen to ensure patient safety and minimize the risk of severe adverse events and unexpected toxicity. In addition, the absence of any significant toxicity suggests a wide therapeutic window for zavondemstat, allowing for safe administration at higher doses. The study did not explore daily doses higher than 50 mg because the sponsor decided to halt the study at this stage. Therefore, the dose range has not yet been fully explored. The PK data demonstrated a dose-linear increase in exposures. In order to establish the optimal PK and safety profile of zavondemstat, a modification of the dosing regimen from 3 days on/4 days off to 5 days on/2 days off; and to daily dosing was further investigated. Furthermore, in the absence of major drug–drug interactions or overlapping toxicity, higher doses of zavondemstat can be used in combination therapy with other anticancer agents, although it is difficult to predict the optimal dose. As zavondemstat is a highly potent drug with very low nanomolar or even picomolar IC50s, animal PK data and results of the current clinical trial indicate that even low doses of zavondemstat may have antitumor activity. These findings highlight the need for further studies to explore the upper dose limits and optimize the therapeutic dosing of zavondemstat.

Zavondemstat showed encouraging preliminary antitumor activity (44% rate of SD) in heavily pretreated patients with a variety of advanced metastatic cancers including colorectal cancer (CRC), CRPC, esophageal cancer, endometrial cancer, and leiomyosarcoma. Two patients (1 with CRPC and 1 with leiomyosarcoma) achieved SD ≥ 6 months with zavondemstat despite having had ≥ 4 prior lines of standard therapy. Among the 10 patients who achieved SD, KDM4 status was unknown due to difficulty in obtaining optional tumor biopsies; however, these patients had documented mutations in KRAS (n = 2; CRC), TP53 (n = 2; CRC and endometrial), PI3K (n = 1; CRC), and mTOR (n = 1; endometrial), showing that KDM4 inhibition was effective despite the presence of known drivers of cancer-resistant pathways.

Due to the exploratory nature of the study, there are some limitations, including the small number of patients treated and the absence of correlative biomarkers associated with response. Biomarker analysis of patients with alterations in KDM4 or its associated pathways would be of particular interest and may shed light on other mechanisms involved in response to zavondemstat in this patient population.

Our data indicate that KDM4 histone demethylases are emerging as exciting therapeutic targets in cancer. These targets play a pivotal role in epigenetic regulation and chromatin dynamics. These enzymes are responsible for demethylating specific lysine residues on histone H3, particularly H3K9me3/me2 and H3K36me3/me2, which are critical markers of transcriptional repression and activation.^7,22 Dysregulation of KDM4 enzymes has been implicated in various cancers, where they drive oncogenesis and resistance pathways by promoting aberrant gene expression, genomic instability, and uncontrolled cell proliferation.^4,6-13 Furthermore, KDM4 enzymes are often overexpressed in cancers with poor prognoses, such as breast, prostate, and colorectal cancers.^12-15 Targeting KDM4 offers the potential to reprogram the epigenetic landscape of cancer cells, restoring normal gene expression patterns and sensitizing tumors to other treatments, including chemotherapy and immunotherapy. The druggability of KDM4, combined with its cancer-specific roles, makes it a promising avenue for novel therapeutic strategies.

Conclusions

In conclusion, zavondemstat was safe and well-tolerated and showed encouraging preliminary anti-tumor activity in heavily pretreated patients. This suggests that further study of zavondemstat, either as a single agent or in combination, is warranted, particularly in patients who may have exhausted all prior therapies. These studies should also include the exploration of biomarkers for response to zavondemstat.

Author Contribution

Apostolia M. Tsimberidou: Conceptualization, Investigation, Supervision, Data Curation, Writing (review & editing). Farshid Dayyani: Investigation, Supervision, Data Curation, Writing (review & editing). David Sommerhalder: Investigation, Supervision, Data Curation, Writing (review & editing). Andrae L. Vandross: Investigation, Supervision, Data Curation, Writing (review & editing). Meredith S. Pelster: Investigation, Supervision, Data Curation, Writing (review & editing). Jason T. Henry: Investigation, Supervision, Data Curation, Writing (review & editing). Cesar A. Perez: Investigation, Supervision, Data Curation, Writing (review & editing). Chandtip Chandhasin: Methodology, Data Curation, Writing (original draft). Yiyun Dai: Methodology, Data curation, Formal analysis, Writing (review & editing). Shirley Tu: Data curation, Project Administration, Resources, Writing (review & editing). Ivan King: Methodology, Data curation, Formal analysis, Writing (review & editing). Frank Perabo: Conceptualization, Supervision, Methodology, Writing (original draft), Funding acquisition. Abhijit Chakraborty: writing—review & editing. Mehmet A. Baysal: writing—review & editing, prepared Figure 1. All authors have read and approved the manuscript.

Funding

This study was supported by the California Institute for Regenerative Medicine (CIRM) grant award CLIN2-14232 and Tachyon Therapeutics, Inc. This work was supported in part by Mr. and Mrs. Steven Mckenzie’s Endowment, Katherine Russell Dixie’s Distinguished Professorship Endowment, and donor funds from Jamie’s Hope and Mrs. and Mr. James Ritter for Dr. Tsimberidou’s Personalized Medicine Program. This work was in part also supported by the National Institutes of Health/National Cancer Institute award number P30 CA016672 (University of Texas MD Anderson Cancer Center).

Conflict of Interest

A.M. Tsimberidou reports grants from OBI Pharma, Agenus, Vividion, Mac- roGenics, AbbVie, Immatics, Novocure, Tachyon, Parker Institute for Cancer Immunotherapy, Tempus, and Tvardi Therapeutics and personal fees from Avstera Therapeutics, BioEclipse, BrYet, Diaccurate, MacroGenics, NEX-I, and Vincerx during the conduct of the study. F. Dayyani reports research funding (inst) from AstraZeneca, Bristol-Myers Squibb, Exelixis, Genentech, Ipsen, Merck, Tachyon, and Taiho Pharmaceuticals. He acts as a consultant/advisor for AstraZeneca, Eisai, Jazz and Taiho Pharmaceuticals and reports participating in speakers’ bureau for Astellas, Beigene, Ipsen, Sirtex Medical, and Takeda. D. Sommerhalder reports research funding (inst) from Abbvie, Acrivon Therapeutics, ADC Therapeutics, Aprea Therapeutics, Ascentage Pharma Group, Astellas, Biomea Fusion, Boehringer Ingelheim, BJ Bioscience, BioNTech, Bristol-Myers Squibb, Compugen, Day One Biopharma, Dicerna/Novo Nordisk, Exelixis, Fate Therapeutics, Gilead Sciences, GlaskoSmithKline, Haihe Pharmaceutical, Iconovir Bio, Immuneering, Impact Therapeutics, Incendia, Kura Oncology, MediLink Therapeutics, Mirati Therapeutics, ModeX Therapeutics, Monopteros Therapeutics, Navire Pharma Inc, Nimbus Therapeutics, NGM Biopharmaceuticals, OBI Pharma, OncoResponse Inc, Pfizer, Revolution Medicines, Step Pharmaceuticals, Symphogen, Tachyon Therapeutics, Teon Therapeutics, Tyligand Bioscience, Vincerx Pharma, Vividion Therapeutics, ZielBio Inc, Zymeworks Inc. He acts as a consultant/advisor for Guidepoint, Revolution Medicines and Nimbus Therapeutics and receives honoraria from Syneos. A. L. Vandross reports research funding (inst) from Abbvie, Adcentrix, Ascentage Pharma, Asana Bio(Kirilys), BioOneCure, Deciphera, Impact therapeutics, Nanjing Immunophage Biotech, Chugai Pharmaceutical Co, Gilead, Lyvgen Biopharma, Medikine, Mbrace Therapeutics, Nikang, Nitto Therapeutics, NGMBio, OncoResponse, PMV Pharmaceuticals, Pasithea Pharmaceutical, Shuhai Yufan Biotechnologies, Iambic Immunomedics, Institute de Recherches Internationales Servier (I.R.I.S), siRNAomics, Tachyon, Teon Therapeutics, Kura, Exelixis, Xilio Development, ZielBio, Vincerex, MBRACE Therapeutics, Bolt Therapeutics, Pliant Therapeutics, Vividion Therapeutics, Nested Therapeutics, and Zumutor Biologics. He acts as a consultant/advisor for Abbvie, Boxer Capital, Nikang, and Pliant Pharmacuticals. M. S. Pelster reports research funding (inst) from Abbvie, Actuate Therapeutics, Affini-T Therapeutics, Agenus, Arcus Biosciences, Artios, Astellas, BeiGene, BioNTech, Bristol-Myers Squibb, Codiak Biosciences, Compass Therapeutics, CytomX, Eisai, Elevation Oncology, Elicio, Exelixis, Fate Therapeutics, Fog Pharmaceuticals, Gilead, GlaxoSmithKline, HiberCell, Immune-Onc Therapeutics, Impact Therapeutics, Jazz Pharmaceuticals, Kura Oncology, Leap Therapeutics, Neogene, Novartis, OncXerna Therapeutics, Panbela Therapeutics, Revolution Medicines, Roche, SeaGen, SQZ Biotechnologies, Surface Oncology, Tachyon, Takeda, Translational Genomics, TransThera Sciences, ZielBio, and 1200 Pharma. She acts as a consultant/advisor for Arcus Biosciences, AstraZeneca, Curio Science, CytomX, Elevation Oncology, EMD Serono, Ipsen Biopharmaceuticals, Jazz Pharmaceuticals, Kura Oncology, Pfizer, and Takeda. J.T. Henry reports research funding to his institution from Abbiscko Therapeutics, ABL Bio, Accutar Biotech, Acepodia, Agenus, ALLink Biotherapeutics, Alterome, Amgen, Artios, Astellas, AstraZeneca, Avistone, Bayer, Bicycle Therapeutics, BioAlta, BioInvent Pharma, Biomea Fusion, Biosplice Therapeutics, Black Diamond Therapeutics, Boehringer Ingelheim, Boundless Bio, Centessa Pharmaceuticals, Context Therapeutics, Cyteir, CytomX, D3 Bio, Daiichi Sankyo, Deciphera, Dynamicure Biotechnology, Eli Lilly & Company, Epizyme, Erasca, Exelixis, FujiFilm, Genentech, GSK, GV20 Therapeutics, Halda Therapeutics, Harbour Biomed, Hutchison MediPharma, ICON plc, IDEAYA Biosciences, IGM Biosciences, Ikena, Immuneering Corporation, Immunitas, Immunogen, Incyte, Ingelheim, InSilico, Inspirna, iTEOS Therapeutics, Jazz Pharmaceuticals, Jacobio Pharmaceuticals, Jounce Pharma, Jubilant Therapeutics, Kineta, Kumquat Biosciences, Kura Oncology, Loxo Oncology, Marengo Therapeutics, MediLink Therapeutics, Merck & Co., Metabomed, Mirati, ModeX Therapeutics, Molecular Templates, Nammi Therapeutics, Navire Pharma, Nested Therapeutics, Nikang Pharmaceuticals, NGM Bio, Oncorus, OnKure, Pfizer, Phanes, Pierre Fabre Medicament, Poseida, Prelude Therapeutics, PureTech, Pyxis, Pyramid Biosciences, Quanta Therapeutics, Rascal Therapeutics, Regeneron, Rein Therapeutics, Relay Therapeutics, Roche, Ribon Therapeutics, Sapience, Sarah Cannon Development Innovations, SCRI, Seagen, Simcha Therapeutics, Siranomics, Stingthera, Sumitomo Pharma, Synthorx, Systimmune, Tachyon Therapeutics, Tango Therapeutics, Tarus Therapeutics, Takeda Pharmaceuticals, Tallac Therapeutics, Tarveda, Tesaro, TORL Biotherapeutics, Turning Point Therapeutics, Vista Therapeutics, and Xencor during the conduct of the study. He also reports personal financial interests as an employee of SCRI and stock ownership in HCA. C.A. Perez reports research funding (inst) from Artios, Kinnate Biopharma, Ribon Therapeutics, Relay Therapeutics, Kura Oncology, Seagen, Xilio Therapeutics, Mirati Therapeutics, Hyamab, Elpiscience, Tallac Therapeutics, Inc., Accutar Biotech, Zhuhai Yufan Biotechnologies, ADANATE Biotechnology, Elevation Oncology, Genentech, Merus, BeiGene, Corvus Pharmaceuticals, DualityBio, Compass Therapeutics, Systimmune, Pfizer, Tachyon, Coherent Biopharma, Pierre Fabre, and Loxo/Lilly. He acts as a consultant/advisor for EMD Serono, BeiGene, and Pfizer. A. Chakraborty has no financial relationships to disclose. M. A. Baysal has no financial relationships to disclose. C. Chandhasin, Y. Dai, S. Tu, I. King and F. Perabo were employees of Tachyon Therapeutics, Inc.

Data Availability

The data supporting the findings of this study are available within the article. Deidentified data can be shared upon reasonable request to the corresponding author.