Abstract
Background
This study aimed to evaluate the safety, pharmacokinetics (PKs), and preliminary activity of LY3405105, a covalent inhibitor of cyclin-dependent kinase 7 (CDK7), in patients with advanced solid tumors.
Materials and Methods
LY3405105 monotherapy was given once daily (QD; part A1) or thrice weekly (TIW; part A2) starting at 1 and 2 mg orally, respectively, and escalated per a Bayesian design in adult patients. The primary endpoint was safety, and secondary endpoints included PKs and antitumor activity.
Results
Fifty-four patients were enrolled: 43 in part A1 and 11 in part A2. Seven patients had dose-limiting toxicities, all in part A1 (45 mg: n = 3; 35 mg: n = 3; 25 mg: n = 1). Thirty-five patients (64.8%) reported at least one treatment-related adverse event (TRAE). TRAEs (≥10%) were diarrhea, nausea, fatigue, vomiting, abdominal pain, anemia, asthenia, and decreased platelet count. QD dosing showed sustained exposure with less peak-trough fluctuation compared to TIW dosing. Median time to maximum concentration was 1-2 hours and half-life was 15-19 hours. CDK7-target occupancy in skin and peripheral blood on day 15 was dose-dependent and reached near maximal occupancy of 75% at ≥15 mg QD. The maximum tolerated dose (MTD) was 20 mg QD. Twelve patients in part A1 (27.9%) and 5 patients in part A2 (45.5%) had a best overall response of stable disease. No complete response or partial response was observed.
Conclusion
The MTD of LY3405105 monotherapy was 20 mg QD. The most common toxicities were gastrointestinal adverse events, myelosuppression, fatigue, and asthenia. Limited clinical activity was observed in this phase I trial, and there are no plans for further development.
ClinicalTrials.gov Identifier
Keywords: solid tumors, cyclin-dependent kinase 7, cyclin-dependent kinase inhibitor, phase I
LY3405105 is the first orally bioavailable and selective covalent CDK7 inhibitor to reach clinical development. In this phase I, dose-escalation study, the optimal dose and regimen of LY3405105 as a single agent in advanced solid tumors were evaluated, as well as safety, tolerability, pharmacokinetics, and CDK7-target occupancy.
Implications for Practice.
Cyclin-dependent kinase 7 (CDK7) is a central regulator of cell cycle and gene transcription. CDK7 inhibition impacts 2%-5% of the transcriptome that is enriched for enhancer-regulated genes. Selective targeting of CDK7 may provide significant therapeutic benefits to patients with cancer, particularly those with tumors addicted to dysregulated transcription. LY3405105 demonstrated CDK7 dose-dependent target occupancy and achieved >75% target occupancy at ≥15 mg once daily; the most common toxicities were gastrointestinal adverse events, myelosuppression, fatigue, and asthenia. No future development is planned due to the narrow therapeutic window and limited clinical activity.
Introduction
Cyclin-dependent kinase 7 (CDK7) is a central regulator of cell cycle and gene transcription.1-4 In the cell, CDK7, cyclin H, and methionine adenosyltransferase form the CDK-activating kinase complex, which promotes cell cycle progression via T-loop phosphorylation of cell cycle CDKs.5 CAK complex is also a component of the general transcription factor IIH, a complex essential for RNA polymerase II-mediated transcription.3,6-12 RNA polymerase II C-terminal domain phosphorylation at the promoters is tightly regulated by a family of kinases (eg, CDK7, CDK9, CDK12, and CDK13) and phosphatases with overlapping and compensatory activities.13-15
CDK7 is considered a drug target due to its dual role in cell cycle and transcription regulation. CDK7 controls the cell cycle by phosphorylating the cell cycle CDKs 1, 2, 4, and 6. CDK7 inhibition prevents CDK2 activation and delays the transition of G1 to S phase, and its inhibition during S/G2 prevents CDK1 activation and delays entry into mitosis.16,17 CDK7 also phosphorates both CDK4 and CDK6.18,19 Inhibition of CDK7 prevents CDK4/6 phosphorylation of Rb, therefore, delaying G1 cell cycle progression.19
CDK7 biology has been implicated in transcription regulation and enhancing cancer growth associated with super-enhancer-linked oncogenes.20-23 Importantly, dysregulation of transcriptional enhancers is a hallmark of cancer that promotes resistance to cell death, sustenance of proliferative signaling, and evasion of growth suppressors.24 As a result, cancer cells become transcriptionally addicted, requiring higher levels of transcription than normal cells to sustain growth.25 CDK7 inhibition impacts the transcriptome which is enriched for enhancer-regulated genes (eg, developmental genes, heat shock, and oncogenes).3,11,26 Inhibition of CDK7 demonstrated disruption of MYCN transcription in neuroblastoma cells and suppression of MYCN-driven global transcriptional amplification, and tumor regression in a mouse model of high-risk neuroblastoma.20 Triple negative breast cancer (TNBC) cells were shown to be highly dependent on CDK7 as compared with hormone receptor-positive breast cancer cells.27 TNBC-specific genes, such as TGF-B, STAT, WNT, MYC, and ETS1, were especially sensitive to CDK7 inhibition, suggesting that CDK7 mediates TNBC transcriptional addiction to this cluster of genes. Therefore, selective targeting of CDK7 may modulate enhancer activity and the associated oncogenic transcriptome, thereby providing significant therapeutic benefit to patients with cancer, particularly those with tumors addicted to dysregulated transcription.25
Although CDKs have emerged as a druggable transcriptional dependence in cancer, their precise evaluation as a therapeutic target has been limited due to the lack of selective inhibitors, associated toxicities, and a narrow therapeutic window.28 LY3405105, a highly selective, orally bioavailable, covalent inhibitor of CDK7, is rationally designed to spare inhibition of other CDKs. LY3405105 demonstrates robust preclinical antitumor efficacy, including durable tumor regressions in cell line and patient-derived xenograft tumor models in multiple tumor types irrespective of their molecular profile.
LY3405105 is the first orally bioavailable and selective covalent CDK7 inhibitor to reach clinical development. In this phase I, dose-escalation study, we evaluated the optimal dose and regimen of LY3405105 as a single agent in advanced solid tumors, in addition to safety, tolerability, pharmacokinetics (PKs), and CDK7-target occupancy.
Materials and Methods
Study Design
JZFA was a phase I, open-label, multicenter, nonrandomized trial (ClinicalTrials.gov identifier: NCT03770494). The ethics committees of all participating centers approved the protocol, and the trial was conducted in accordance with the Declaration of Helsinki. Informed consent was obtained for all patients. The trial was designed with 3 parts, 2 dose escalations (phase Ia, parts A1 and A2), and a phase Ib dose expansion; however, the phase Ib expansion cohorts were not initiated based on the sponsor’s decision and limited efficacy observed as a single agent. Patients were required to have advanced and/or metastatic solid tumors that had progressed after standard treatment or were not appropriate for standard treatment.
The primary endpoint of dose escalation was to assess the safety and tolerability of LY3405105, administered as monotherapy once daily (QD) (part A1) or thrice weekly (TIW) (part A2), and to determine the recommended phase II dose (RP2D) and schedule in patients with solid tumors. The dose escalation followed a Bayesian model-based toxicity method,29 with patients receiving LY3405105 at starting doses of 1 mg QD (part A1) and 2 mg TIW (part A2). These starting doses of LY3405105 were selected based on data from good laboratory practice-compliant, 4-week, oral, daily repeat-dose toxicity studies in rats and dogs, and PK/pharmacodynamic (PD) data modeling. Dose-escalation decisions were based on the safety from the previous dose level, any available PK and CDK7-target occupancy data, and agreed upon by the investigators and the sponsor (see Supplementary Material for details on the selection of starting dose, dose-escalation method, and schedule).
Treatment and Assessments
Patients were administered LY3405105 until confirmed progressive disease, unacceptable toxicity, or discontinuation for any other reason. Safety was monitored throughout the study and at 30 (±7) days after the last dose of treatment by physical examination, electrocardiogram, and blood laboratory testing. Adverse events (AEs) were graded by the investigator according to the National Cancer Institute Common Terminology Criteria for Adverse Events (Version 4.0). A dose-limiting toxicity (DLT) was defined as any of the following events occurring during cycle 1 (28-day cycle) of the dose-escalation phase: total bilirubin (>2 × upper limit of normal [ULN] with alanine aminotransferase/aspartate aminotransferase [ALT/AST] > 3 × ULN in the absence of cholestasis [alkaline phosphatase < 2 × ULN]); grade 4 neutropenia > 5 days duration; febrile neutropenia; grade 4 thrombocytopenia of any duration; grade 3 thrombocytopenia with clinically significant bleeding; grade 3 or above anemia; or any nonhematologic AE of grade ≥ 3 (except fatigue or anorexia, grade 3 or grade 4 fatigue was to be considered a DLT if persisting>5 days, transient grade 3 elevations of ALT/AST, without evidence of other hepatic injury, and controlled nausea, constipation, electrolyte disturbance, vomiting, and diarrhea). Any other significant toxicity deemed by the primary investigator and the sponsor but not specified earlier could also be considered as dose limiting (eg, any toxicity that was possibly related to the study medication that required the withdrawal of the patient from the study during cycle 1, or certain grade 2 AEs affecting vital organs).
PK Parameter Estimates
Intensive PK samples were collected post-LY3405105 dose on cycle 1 day 1 and cycle 1 day 15 for up to 24 and 48 hours in part A1 and part A2, respectively. PK parameter estimates, such as maximum drug concentration and area under the concentration versus time curve from time 0 to infinity (AUC[0-Inf]) for LY3405105, were derived from plasma concentrations by standard noncompartmental methods of analysis. Tumor assessments by computed tomography or magnetic resonance imaging were performed according to Response Evaluation Criteria in Solid Tumors, version 1.1, during screening and every 8 weeks for the first 6 months after cycle 1 day 1 and then every 12 weeks until disease progression, death, or study completion.
CDK7-Target Occupancy Assay
CDK7-target occupancy is a measure of the percentage of CDK7 protein covalently bound by LY3405105. Skin biopsies for evaluation of CDK7-target occupancy analysis were collected at baseline (any time within the 14 days prior to the first dose of LY3405105 on day 1 cycle 1) and 6- to 8-hours postdose on days 1 and 15. CDK7-target occupancy was also evaluated in peripheral blood mononuclear cells (PBMCs) isolated from whole blood samples collected predose on day 1 and 8 hours postdose on days 1 and 15.
The ratio of unbound and total CDK7 was determined with a sandwich-like Meso Scale Diagnostics (MSD, Rockville, MD) assay at BioAgilytix (Durham, NC). Protein lysates were prepared from frozen skin and PBMC samples and labeled with biotinylated LY3405105 probe. To detect unbound CDK7, rabbit polyclonal anti-cyclin H antibody (Cell Signaling #2927, Danvers, MA) was applied to an MSD goat anti-rabbit immunoglobulin G (IgG) plate to capture CDK7/cyclin H complexes. Probe-bound CDK7 was detected with MSD SULFO-TAG-labeled Streptavidin and read on an MSD Sector Imager 600. To detect total CDK7, CDK7/cyclin H complexes were captured as described earlier, followed by detection with anti-CDK7 monoclonal mouse antibody (Clone# 45H1, Thermo Fisher #MA5-17210, Waltham, MA) and MSD SULFO-TAG goat anti-mouse IgG.
CDK7-target occupancy in skin samples at each timepoint was calculated with the following formula: % CDK7-target occupancy = 100 × [1 − (unbound CDK7 at timepoint/total CDK7 at timepoint)/(unbound CDK7 at baseline/total CDK7 at baseline)]. For the majority of PBMC samples, the concentration of total CDK7 was below the limit of quantitation for the MSD assay. Therefore, the amount of unbound CDK7 could not be normalized to total CDK7, and the following formula was used: % CDK7-target occupancy = 100 × [1 − (unbound CDK7 at timepoint)/(unbound CDK7 predose on cycle 1 day 1)].
Statistical Methods
Safety and PK analyses were conducted in all patients who received at least one dose of LY3405105. Efficacy analyses included all patients enrolled in part A1 and part A2. Disease control rate was calculated as the proportion of patients who achieved a complete response, partial response, or stable disease. Descriptive summary statistics were provided for patient characteristics, safety, and PK variables.
Results
Baseline Characteristics and Disposition
From February 11, 2019 to February 04, 2021, 54 patients were enrolled in the study. In part A1 (QD group), 43 were enrolled at 10 dose levels (3 cohorts due to de-escalation) of LY3405105: 1 mg (n = 4), 2 mg (n = 4), 4 mg (n = 4), 8 mg (n = 4), 15 mg (n = 5), 30 mg (n = 5), 45 mg (n = 5), 35 mg (n = 3), 25 mg (n = 3), and 20 mg (n = 6), and 11 patients were enrolled in part A2 at 3 TIW dose levels of LY3405105: 2 mg (n = 4), 4 mg (n = 4), and 8 mg (n = 3). At the data cutoff, all 54 patients (100.0%) were off study treatment.
The baseline characteristics of enrolled patients are summarized in Table 1. Briefly, the majority of patients were female (76%), median age was 61 years (range: 29-87 years), and most were White (81%). The most common tumor types were ovarian cancer (n = 14), breast cancer (n = 8), colorectal cancer (n = 5), and leiomyosarcoma (n = 5). Of the reported prior therapies in the enrolled population, 49 patients (91%) received prior surgical procedures, 21 patients (39%) received prior radiotherapy (14 patients [26%] in the advanced/metastatic setting), and 53 patients (98%) received prior systemic therapies. Most of these patients were heavily pretreated, with a median of 5 (range: 1-9) systemic therapies. The most common reason for treatment discontinuation was progressive disease (46 patients [85%]).
Table 1.
Patient baseline characteristics.
| Characteristic, n (%) |
Dose escalation | ||
|---|---|---|---|
| QD (N = 43) |
TIW (N = 11) |
Total (N = 54) |
|
| Female | 33 (77) | 8 (73) | 41 (76) |
| Median age (min, max), years | 63 (29, 87) | 55 (30, 70) | 61 (29, 87) |
| Pooled age group <65 years | 24 (56) | 8 (73) | 32 (59) |
| Race | |||
| White | 34 (79) | 9 (82) | 43 (80) |
| Black or African American | 3 (7) | 1 (9) | 4 (7) |
| Asian | 4 (9) | 1 (9) | 5 (9) |
| Native Hawaiian or other Pacific Islander | 1 (2) | 0 | 1 (2) |
| Missing | 1 (2) | 0 | 1 (2) |
| ECOG PS (baseline) | |||
| 0 | 26 (61) | 6 (55) | 32 (59) |
| 1 | 17 (40) | 5 (46) | 22 (41) |
| Prior anticancer therapy | |||
| Surgical procedure | 41 (95) | 8 (73) | 49 (91) |
| Radiotherapy | 19 (44) | 2 (18) | 21 (39) |
| Systemic therapy | 42 (98) | 11 (100) | 53 (98) |
| Prior lines of systemic therapya | |||
| Median | 5 | 4 | 5 |
| Minimum | 1 | 1 | 1 |
| Maximum | 9 | 8 | 9 |
| Tumor type | |||
| Ovarian cancer | 13 (30) | 1 (9) | 14 (26) |
| Breast cancer | 7 (16) | 1 (9) | 8 (15) |
| Colorectal cancer | 2 (5) | 3 (27) | 5 (9) |
| Cholangiocarcinoma | 0 | 2 (18) | 2 (4) |
| Endometrial cancerb | 4 (9) | 0 | 4 (7) |
| Leiomyosarcomac | 5 (12) | 0 | 5 (9) |
| Pancreatic carcinomad | 3 (7) | 1 (9) | 4 (7) |
| Othere | 9 (21) | 3 (27) | 12 (22) |
aSubject 1053 from QD did not report any prior regimen and is not included in the analysis (QD N = 42; total N = 53).
bIncludes uterine cancer.
cIncludes all sarcoma types.
dIncludes all pancreatic cancers.
eAnal cancer, chordoma, desmoplastic small round cell tumor, gallbladder cancer, head and neck cancer, malignant melanoma, malignant neoplasm of ampulla of Vater, mucinous adenocarcinoma of appendix, esophageal adenocarcinoma, rhabdoid tumor, squamous cell carcinoma of the cervix, and transitional cell carcinoma were 1% of each tumor type (n = 1).
Abbreviations: ECOG PS: Eastern Cooperative Oncology Group performance status; n: number of patients in each characteristic; N: number of patients in each group; QD: once daily; TIW: thrice weekly.
Safety
In part A1, no DLT was observed at 1, 2, 4, 8, 15, or 30 mg QD. Three patients had a DLT at 45 mg QD (n = 5) (Table 2). The DLTs included a grade 3 mucosal inflammation (serious AE [SAE]; lasted for 4 days) in one patient (resulted in dose omission), a grade 2 mucosal inflammation (lasted for 5 days) and grade 3 anemia (lasted for 2 days) in one patient (resulted in dose omission/discontinuation), and grade 2 nausea and fatigue (lasted for 1 day) in one patient (resulted in discontinuation). The dose level was de-escalated to 35 mg QD (n = 3), and all 3 patients had a DLT. The DLTs included grade 4 decreased platelet count (lasted for 3 days) that resulted in dose reduction in one patient, grade 3 decreased platelet count (lasted for 6 days), nausea (SAE; lasted for 3 days), and fatigue (lasted for 2 days) that resulted in dose reduction in one patient, and grade 3 diarrhea (SAE; lasted for 8 days) in one patient. The dose level was further de-escalated to 25 mg (n = 3), and one patient had a DLT of grade 3 increased ALT (SAE; lasted for 5 days) that resulted in dose omission. Additional patients were enrolled to 20 mg QD, and no DLT was observed. The maximum tolerated dose (MTD) and RP2D were determined to be 20 mg QD. In part A2 (TIW), no DLT was observed in 2 mg, 4 mg, or 8 mg; no additional dose levels were tested when the study was terminated.
Table 2.
Summary of dose-limiting toxicities during the dose-escalation phase.
| Event, n (%) | Dose escalation | Dose de-escalation | ||||
|---|---|---|---|---|---|---|
| 45 mg QD N = 5 |
35 mg QD N = 3 |
25 mg QD N = 3 |
||||
| Any grade | ≥Grade 3 | Any grade | ≥Grade 3 | Any grade | ≥Grade 3 | |
| Fatigue | 1 (20) | 0 | 1 (33) | 1 (33) | 0 | 0 |
| Mucosal inflammation | 2 (40) | 1 (20) | 0 | 0 | 0 | 0 |
| Nausea | 1 (20) | 0 | 1 (33) | 1 (33) | 0 | 0 |
| Platelet count decreased | 0 | 0 | 2 (67) | 2 (67) | 0 | 0 |
| Anemia | 1 (20) | 1 (20) | 0 | 0 | 0 | 0 |
| Alanine aminotransferase increased | 0 | 0 | 0 | 0 | 1 (33) | 1 (33) |
| Diarrhea | 0 | 0 | 1 (33) | 1 (33) | 0 | 0 |
| Patients (n) with DLT events | 3 | 3 | 1 | |||
Abbreviations: DLT: dose-limiting toxicity; n: number of patients in each characteristic; N: number of patients in each group; QD: once daily.
Twenty-nine (67%) patients in the QD group and 6 patients (55%) in the TIW group experienced treatment-emergent AEs (TEAEs) related to study treatment (TRAEs) (Table 3). The most common (≥10%) TRAEs in all patients were diarrhea, nausea, fatigue, vomiting, abdominal pain, anemia, asthenia, and decreased platelet count. Grade 3 TRAEs occurred in 9 patients (21%) in the QD group, with none reported in the TIW group. Grade 3 TRAEs were diarrhea, nausea, fatigue, anemia, increased ALT, dyspnea, mucosal inflation, acute kidney injury, decreased platelet count, increased blood alkaline phosphatase, ophthalmic herpes zoster, and platelet count decreased. Grade 4 decreased platelet count occurred in one patient at the 35 mg QD dose level. Serious adverse events related to study treatment were reported in the QD group (7 patients [16%] with any grade and 6 patients [14%] with grade ≥ 3). A total of 20 deaths (37%) occurred in this study, and all deaths were attributed to study disease. Three patients discontinued study treatment due to AEs (grade 2 anemia, grade 2 nausea, or grade 2 illness).
Table 3.
The most common (≥10% in all patients) treatment-related adverse events and all treatment-related serious adverse events.
| Event, n (%) | Dose escalation | |||||
|---|---|---|---|---|---|---|
| QD (N = 43) |
TIW (N = 11) |
Total (N = 54) |
||||
| Any grade | ≥Grade 3 | Any grade | ≥Grade 3 | Any grade | ≥Grade 3 | |
| Patients with ≥1 TRAE | 29 (67) | 10 (23) | 6 (55) | 0 | 35 (65) | 10 (19) |
| Diarrhea | 18 (42) | 2 (5) | 0 | 0 | 18 (33) | 2 (4) |
| Nausea | 9 (21) | 1 (2) | 1 (9) | 0 | 10 (19) | 1 (2) |
| Fatigue | 8 (19) | 1 (2) | 0 | 0 | 8 (15) | 1 (2) |
| Vomiting | 8 (19) | 0 | 0 | 0 | 8 (15) | 0 |
| Abdominal pain | 6 (14) | 0 | 0 | 0 | 6 (11) | 0 |
| Anemia | 5 (12) | 2 (5) | 1 (9) | 0 | 6 (11) | 2 (4) |
| Asthenia | 2 (5) | 0 | 4 (36) | 0 | 6 (11) | 0 |
| Platelet count decrease | 6 (14) | 3 (7) | 0 | 0 | 6 (11) | 3 (6) |
| Patients with ≥1 SAE related to study treatment | 7 (16) | 6 (14) | 0 | 0 | 7 (13) | 6 (11) |
| Diarrhea | 2 (5) | 2 (5) | 0 | 0 | 2 (4) | 2 (4) |
| Abdominal pain | 1 (2) | 0 | 0 | 0 | 1 (2) | 0 |
| Acute kidney injury | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Alanine aminotransferase increase | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Dyspnea | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Hypotension | 1 (2) | 0 | 0 | 0 | 1 (2) | 0 |
| Mucosal inflammation | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Nausea | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Ophthalmic herpes zoster | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
| Platelet count decrease | 1 (2) | 1 (2) | 0 | 0 | 1 (2) | 1 (2) |
Other ≥grade 3 TRAEs are aspartate aminotransferase increase (4.0%), dyspnea (2%), mucosal inflammation (2%), acute kidney injury (2%), alkaline phosphatase increase (2%), ophthalmic herpes zoster (2%), and peripheral edema (2%).
Abbreviations: n: number of patients in each TRAE; N: number of patients in each group; QD: once daily; SAE: serious adverse events; TIW: thrice weekly; TRAE: treatment-related adverse events.
TEAEs occurred in 42 patients (98%) in the QD group and 10 patients (91%) in the TIW group, of which 19 patients (44%) in QD and 3 (27%) patients in TIW had grade 3/4 TEAEs (Supplementary Table S1).
Study Treatment Duration and Antitumor Activity
The median duration of therapy was 56 days (range: 9-336 days) and 6 patients (11%) received 6 or more cycles of treatment. No complete or partial responses were observed in the QD or TIW groups. The disease control rates, calculated as the proportion of patients who achieved a complete response, partial response, or stable disease measured at least 4 weeks following the initial observation, were 28% in the QD group and 46% in the TIW group, with stable disease as the only best overall response (12 [28%] patients in the QD group and 5 [46%] patients in the TIW group). One patient with metastatic chondrosarcoma achieved stable disease and received study treatment for 336 days.
Pharmacokinetics
After oral dosing, absorption of LY3405105 was rapid, with median time to maximum concentration occurring approximately 1 to 2 hours postdose. Subsequently, plasma concentrations of LY3405105 appeared to decline in a biphasic manner with a geometric meaN-terminal elimination half-life of approximately 15 to 19 hours (assessed in the TIW group). Overall, mean plasma concentration time profiles across the dose range studied showed that exposures increased with dose across 1 to 45 mg QD day 1, 1 to 35 mg QD day 15, and 2 to 8 mg TIW (day 1 and day 15) dose ranges (Fig. 1). QD dosing provides sustained exposure with less peak-trough fluctuation compared to TIW dosing (results not shown). Variability in steady-state exposures, as assessed by percentage of coefficient of variation, was approximately 25% to 90% after QD dosing in part A1 and 20% to 200% after TIW dosing in part A2.
Figure 1.
Mean LY3405105 plasma concentration-time profiles following single or multiple doses. (A) One to 45 mg QD after single dose on day 1. (B) One to 45 mg QD after multiple doses on day 15. (C) Two to 8 mg TIW after single dose on day 1. (D) Two to 8 mg TIW after multiple doses on day 15. Abbreviations: QD: daily; TIW: thrice weekly.
Target Occupancy Assessment in PBMCs and Skin
The relationship between CDK7-target occupancy and exposure is shown in Fig. 2. Evaluation of CDK7-target occupancy on day 1 was minimally informative, but assessment at steady state on day 15 demonstrated LY3405105 dose- and plasma concentration-dependent increases in occupancy. Near maximal occupancy was achieved at doses of 15 mg QD and above. For this reason, collection of skin biopsies and evaluation of target occupancy was discontinued after the completion of the 30 mg QD dose-escalation cohort. Compared with CDK7-target occupancy data derived from skin biopsies, the data from PBMCs were more variable due to an inability to normalize to total levels of CDK7. Concordance between data from skin biopsies and PBMCs is shown in Supplementary Fig. S1.
Figure 2.
CDK7-target occupancy relationships in skin (A and C) or PBMC (B and D) on study day 15. Reference lines: 40% CDK7 occupancy (blue-dotted line) = level below which a signal cannot be differentiated from noise; 75% CDK7 occupancy (black-dotted line) = level above which represents the start of the potential efficacious range. Negative % CDK7 occupancy values were substituted with zero for visualization. Each datapoint represents one patient where paired % CDK7 occupancy values and dose or concentration values were available. Abbreviations: CDK7: cyclin-dependent kinase 7; PBMC: peripheral blood mononuclear cells.
Relationship Between Exposure, Target Occupancy, and DLT
The relationship between LY3405105 steady-state exposure, target occupancy, and DLT in part A1 of the study is shown in Fig. 3. Some patients were not included, because PK or PD were not collected. Maximum skin target occupancy occurred above a steady-state area under the curve (AUC) of approximately 500 ng h/mL, which corresponds to doses at 15 mg QD. The start of the steady-state AUC range, where DLTs were observed was approximately 1000 ng h/mL. This AUC range corresponds to doses within the range of 30 to 45 mg QD.
Figure 3.
CDK7-target occupancy, dose-limiting toxicities, and LY3405105 exposure relationships (Skin, day 15). AUC(0-Inf) day 1 values were only used for patients whose AUC(0-24) day 15 values were not available. Negative % CDK7 occupancy values were substituted with zero for visualization. Each datapoint represents one patient where paired % CDK7 occupancy values, and AUC values were available. The patients represented in the DLT plot include those in part A1, cycle 1 of the study with AUC values available, including patients without CDK7-target occupancy data; each mark represents one patient (n = 35); DLT events in red (patients with PK or PD data only). Abbreviations: AUC(0-24): area under the concentration versus time curve from 0 to 24 hours; AUC(0-Inf): area under the concentration versus time curve from time 0 to infinity; CDK7: cyclin-dependent kinase 7; DLT: dose-limiting toxicity.
Discussion
This first-in-human study demonstrated that LY3405105, an orally bioavailable covalent inhibitor of CDK7, can be safely administered in patients with advanced cancer at doses up to 20 mg QD. The MTD and RP2D were determined to be 20 mg QD based on safety, tolerability, PK, and target occupancy results. The DLTs and most frequent AEs observed were consistent with preclinical toxicology studies. These results are also consistent with that of SY-5609, a selective non-covalent inhibitor of CDK7, whose most frequent TEAEs included gastrointestinal toxicity, fatigue, thrombocytopenia, and anemia; DLTs included cytopenia, gastrointestinal toxicity, and hypotension.30 The similarities suggest the side effects observed with LY3405105 are on-target. In the SY-5609 phase I study, intermittent schedules (7 days on/7 days off) allowed dose escalation beyond the continuous dosing MTD. In the current study, TIW was only dose escalated to 8 mg, and the target occupancy and escalation to higher dose levels were not pursued because of early termination of the compound due to lack of efficacy at tolerable doses.
The best overall response observed in this study was stable disease, with 28% in the QD group and 46% in the TIW group. Several reasons could have accounted for the limited clinical activity from this agent. First, the MTD/recommended dose of LY3405105 is 20 mg QD, where 75%-90% CDK7-target occupancy was observed in skin biopsies; however, the expression of target genes, such as POLR2A, MYC, and other super-enhancers, was not investigated in this study. It remains unknown whether the expression of these target genes was maximally inhibited at 20 mg QD. A recent study demonstrated that there was a linear increase of POLR2A fold change (up to 4- to 5-fold change) with the CDK7-target occupancy increased from 75% to 90%-100%.31 The POLR2A expression in PBMCs had approximately 2.5-fold changes when SY-5609 achieved 75% CDK7-target occupancy. With the improvement of CDK7-target occupancy (up to 100%), POLR2A showed more pronounced changes (4-5 fold) as compared with baseline. In this study of LY3405105, dose levels>20 mg QD were not deemed tolerable, and therefore, CDK7-target occupancy in the 90%-100% range was not feasible. The TIW dosing strategy was stopped early and was unable to address whether the alternative dosing schedule can achieve maximal target occupancy and downstream gene inhibition while allowing recovery from gastrointestinal and hematological toxicities. A second potential reason for lack of observed efficacy is that CDK7 inhibition impacts only 2%-5% of the transcriptome. Consistent with this hypothesis is the limited single-agent activity observed in the phase I study of SY-5609. Preclinical work has shown that models with specific oncogenic alterations in KRAS, BRAF, MYCN, and the Rb pathway were especially sensitive to CDK7 inhibition, likely due to associated changes in super enhancers and increased transcriptional dependence.20,32 Therefore, it might be critical to have a molecularly selected patient population with super enhancer-associated gene alterations to maximize LY3405105’s clinical activity. Phase Ia (part A1 and part A2) did not set up a patient selection strategy for enrichment of these molecular alterations. Molecularly selected patient populations were planned in the phase Ib part of the trial; however, this was not initiated based on the limited efficacy observed as a single agent and the sponsor’s decision not to continue further development.
Conclusions
In conclusion, this phase I trial of LY3405105 monotherapy identified a maximum tolerated/recommended phase II dose of 20 mg QD. The commonly observed side effects were gastrointestinal toxicities and myelosuppression, likely on-target effects. CDK7-target occupancy was observed up to 90%; however, limited clinical activity led to early termination of the study and discontinuation of the compound.
Supplementary Material
Acknowledgments
We thank all the patients and their caregivers for their participation in this trial. We thank all the investigators and their support staff who generously participated in this work. Ira Ayene, MS, PhD, MPA, and Gillipsie Minhas of Syneos Health provided medical writing and editorial assistance, funded by Eli Lilly and Company. A.M.S. would like to acknowledge the ASCO Conquer Cancer Foundation CDA, NCI P30CA008748 CCITLA, and Memorial Sloan Kettering Cancer Center Support Grant (P30 CA008748).
Contributor Information
Elena Garralda, Department of Medical Oncology, Vall d’Hebron University Hospital, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain.
Alison M Schram, Department of Medicine, Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College, New York, NY, USA.
Philippe L Bedard, Division of Medical Oncology and Hematology, Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada; Department of Medicine, University of Toronto, Toronto, Ontario, Canada.
Gary K Schwartz, Columbia University Vagelos School of Medicine, Herbert Irving Comprehensive Cancer Center, New York, NY, USA.
Eunice Yuen, Eli Lilly and Company, Indianapolis, IN, USA.
Samuel C McNeely, Eli Lilly and Company, Indianapolis, IN, USA.
Silvia Ribeiro, Eli Lilly and Company, Indianapolis, IN, USA.
Jason Cunningham, Eli Lilly and Company, Indianapolis, IN, USA.
Yi Wang, Eli Lilly and Company, Indianapolis, IN, USA.
Arantxa Urunuela, Eli Lilly and Company, Indianapolis, IN, USA.
Xiaojian Xu, Eli Lilly and Company, Indianapolis, IN, USA.
Patricia LoRusso, Yale Cancer Center, New Haven, CT, USA.
Funding
This work was supported by Eli Lilly and Company. Employees of Eli Lilly and Company participated in the study design, collection, analysis, and interpretation of the data, writing of this report, and decision to submit this article for publication.
Conflict of Interest
Elena Garralda is a paid consultant for Roche/Genentech, Ellipses Pharma, Boehringer Ingelheim, Janssen, AstraZeneca, Seattle Genetics, Alkermes, and Neomed Therapeutics, reports receiving commercial research grants from Novartis and Roche, and reports receiving speakers bureau honoraria from Bristol-Myers Squibb, Merck, Roche, and Thermo Fisher. Alison M. Schram has received advisory board compensation from Relay Therapeutics and Mersana, and research funding paid to institution from AstraZeneca, ArQule, BeiGene/Springworks, Black Diamond Therapeutics, Elevation Oncology, Kura, Eli Lilly and Company, Merus, Northern Biologics, Pfizer, PMV Pharma, Relay, Repare Therapeutics, Revolution Medicine, and Surface Oncology. Philippe L. Bedard reports uncompensated consulting/advisory relationships with Bristol‐Myers Squibb, Merck, Pfizer, SeaGen, Gilead, and Sanofi, and reports receiving research funding from Amgen, Bristol‐Myers Squibb, Sanofi, Novartis, GlaxoSmithKline, AstraZeneca, Merck, Seattle GeneticsSeaGen, Nektar, Immunomedics, Mersana, Zymeworks, Sanofi, Bicara Therapeutics, and Servier. Gary K. Schwartz reports consulting/advisory relationships and stock and other ownership relationships with Bionaut Labs LTD, GenCirq Inc., Concarlo, OnCusp, Shanghai Parma, Astex, Epizyme, Sellas Life Sciences Group, Puretech Health, Kirilys Therapeutics, and reports research funding from AstexGenCirq, Bionaut Labs, January Therapeutics, reports consulting/advisory relationships with Bionaut Labs, Ellipses Pharma, GenCirq, Epizyme, Array BioPharma, Apexigen, Oncogenuity, Oncusp, Concarlo, Shanghai Pharma, Astex, January Therapeutics, Sellas Life Sciences, Purtech, and reports research funding from Astex Pharmaceuticals, Incyte, Calithera Biosciences, Eli Lilly and Company, Daiichi Sankyo, Fortress, Karyopharm Therapeutics, Oxford BioTherapeutics, Astex Pharmaceuticals, TopAlliance BioSciences Inc, Adaptimmune, SpringWorks Therapeutics, TRACON Pharma. Eunice Yuen is an employee of Eli Lilly and Company. Samuel C. McNeely is an employee of Eli Lilly and Company. Silvia Ribeiro is an employee of Eli Lilly and Company. Jason Cunningham is an employee of Eli Lilly and Company. Eric Wang is an employee of Eli Lilly and Company. Arantxa Urunuela is a former employee of Eli Lilly and Company. Xiaojian Xu is an employee of Eli Lilly and Company. Patricia LoRusso reports advisory relationships with AbbVie, GenMab, Genentech, CytomX, Takeda, Cybrexa, Agenus, IQVIA, TRIGR, Pfizer, ImmunoMet, Black Diamond, GSK, QED Therapeutic, AstraZeneca, EMD Serono, Shattuck, Astellas, Salarius, Silverback, MacroGenics, Kyowa Kirin Pharmaceutical Development, Kineta, Inc., Zentalis Pharmaceuticals, Molecular Templates, ABL Bio, STCube Pharmaceuticals, Bayer, I-Mab, Seagen, imCheck, Relay Therapeutics, Stemline, Compass BADX, Mekanist, Mersana Therapeutics, BAKX Therapeutics, Scenic Biotech, Qualigen, Neuro Trials, reports data safety monitoring relationships with Agios, Five Prime, Halozyme, and Tyme, and reports consultant relationships with SOTIO, SK Life Science, I-Mab, Roivant Sciences, and Roche-Genentech (imCore Alliance).
Author Contributions
Conception/design: E.G., A.M.S., P.L.B., G.K.S., X.X., and P.L. Methodology: X.X., E.Y., S.C.M., S.R., and J.C. Validation: E.G., A.M.S., P.L.B., G.K.S., X.X., P.L., E.Y., S.C.M., S.R., Y.W., and J.C. Formal analysis: X.X., E.Y., S.C.M., S.R., Y.W., and J.C. Investigation: E.G., A.M.S., G.K.S., P.L., P.L.B., and X.X. Manuscript writing (original draft): E.G., A.M.S., G.K.S., P.L., P.L.B., X.X., E.Y., S.C.M., S.R., Y.W., and J.C. Manuscript writing (review and editing): E.G., A.M.S., G.K.S., P.L., P.L.B., X.X., E.Y., S.M., S.R., Y.W., and J.C. Final approval of manuscript: all authors.
Data Availability
All preliminary data analyzed are included in this report. The anonymized data are not publicly available due to privacy restrictions and will be made available by request from any qualified investigator. Eli Lilly and Company’s policy is to provide access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. No expiration date of data requests is currently set, once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, and blank or annotated case report forms, will be provided in a secure data sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.
References
- 1. Devault A, Martinez AM, Fesquet D, et al. Mat1 (“menage à trois”) a new ring finger protein subunit stabilizing cyclin H-CDK7 complexes in starfish and xenopus CAK. EMBO J. 1995;14(20):5027-5036. 10.1002/j.1460-2075.1995.tb00185.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Devos M, Mommaerts E, Migeot V, van Bakel H, Hermand D.. Fission yeast CDK7 controls gene expression through both its CAK and C-terminal domain kinase activities. Mol Cell Biol. 2015;35(9):1480-1490. 10.1128/MCB.00024-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Glover-Cutter K, Larochelle S, Erickson B, et al. TFIIH-associated CDK7 kinase functions in phosphorylation of C-terminal domain Ser7 residues, promoter-proximal pausing, and termination by RNA polymerase II. Mol Cell Biol. 2009;29(20):5455-5464. 10.1128/MCB.00637-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kelso TW, Baumgart K, Eickhoff J, et al. Cyclin-dependent kinase 7 controls mRNA synthesis by affecting stability of preinitiation complexes, leading to altered gene expression, cell cycle progression, and survival of tumor cells. Mol Cell Biol. 2014;34(19):3675-3688. 10.1128/MCB.00595-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lolli G, Johnson LN.. CAK-cyclin-dependent activating kinase: a key kinase in cell cycle control and a target for drugs?. Cell Cycle. 2005;4(4):572-577. [PubMed] [Google Scholar]
- 6. Abdulrahman W, Iltis I, Radu L, et al. ARCH domain of XPD, an anchoring platform for CAK that conditions TFIIH DNA repair and transcription activities. Proc Natl Acad Sci USA. 2013;110(8):E633-E642. 10.1073/pnas.1213981110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Adamczewski JP, Rossignol M, Tassan JP, et al. MAT1, CDK7 and cyclin H form a kinase complex which is UV light-sensitive upon association with TFIIH. EMBO J. 1996;15(8):1877-1884. [PMC free article] [PubMed] [Google Scholar]
- 8. Busso D, Keriel A, Sandrock B, et al. Distinct regions of MAT1 regulate CDK7 kinase and TFIIH transcription activities. J Biol Chem. 2000;275(30):22815-22823. 10.1074/jbc.M002578200. [DOI] [PubMed] [Google Scholar]
- 9. Chen F, Gao X, Shilatifard A.. Stably paused genes revealed through inhibition of transcription initiation by the TFIIH inhibitor triptolide. Genes Dev. 2015;29(1):39-47. 10.1101/gad.246173.114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Feaver WJ, Svejstrup JQ, Henry NL, Kornberg RD.. Relationship of CDK-activating kinase and RNA polymerase II CTD kinase TFIIH/TFIIK. Cell. 1994;79(6):1103-1109. 10.1016/0092-8674(94)90040-x. [DOI] [PubMed] [Google Scholar]
- 11. Serizawa H, Mäkelä TP, Conaway JW, et al. Association of CDK-activating kinase subunits with transcription factor TFIIH. Nature. 1995;374(6519):280-282. 10.1038/374280a0. [DOI] [PubMed] [Google Scholar]
- 12. Shiekhattar R, Mermelstein F, Fisher RP, et al. CDK-activating kinase complex is a component of human transcription factor TFIIH. Nature. 1995;374(6519):283-287. 10.1038/374283a0. [DOI] [PubMed] [Google Scholar]
- 13. Buratowski S. Progression through the RNA polymerase II CTD cycle. Mol Cell. 2009;36(4):541-546. 10.1016/j.molcel.2009.10.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Egloff S, Murphy S.. Cracking the RNA polymerase II CTD code. Trends Genet. 2008;24(6):280-288. 10.1016/j.tig.2008.03.008. [DOI] [PubMed] [Google Scholar]
- 15. Greenleaf AL. Human CDK12 and CDK13, multi-tasking CTD kinases for the new millenium. Transcription. 2019;10(2):91-110. 10.1080/21541264.2018.1535211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Larochelle S, Merrick KA, Terret ME, et al. Requirements for CDK7 in the assembly of CDK1/cyclin B and activation of CDK2 revealed by chemical genetics in human cells. Mol Cell. 2007;25(6):839-850. 10.1016/j.molcel.2007.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Patel H, Periyasamy M, Sava GP, et al. ICEC0942, an orally bioavailable selective inhibitor of CDK7 for cancer treatment. Mol Cancer Ther. 2018;17(6):1156-1166. 10.1158/1535-7163.MCT-16-0847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Bisteau X, Paternot S, Colleoni B, et al. CDK4 T172 phosphorylation is central in a CDK7-dependent bidirectional CDK4/CDK2 interplay mediated by p21 phosphorylation at the restriction point. PLoS Genet. 2013;9(5):e1003546. 10.1371/journal.pgen.1003546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Schachter MM, Merrick KA, Larochelle S, et al. A CDK7-CDK4 T-loop phosphorylation cascade promotes G1 progression. Mol Cell. 2013;50(2):250-260. 10.1016/j.molcel.2013.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Chipumuro E, Marco E, Christensen CL, et al. CDK7 inhibition suppresses super-enhancer-linked oncogenic transcription in MYCN-driven cancer. Cell. 2014;159(5):1126-1139. 10.1016/j.cell.2014.10.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sharifnia T, Wawer MJ, Chen T, et al. Small-molecule targeting of brachyury transcription factor addiction in chordoma. Nat Med. 2019;25(2):292-300. 10.1038/s41591-018-0312-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Zhang J, Liu W, Zou C, et al. Targeting super-enhancer-associated oncogenes in osteosarcoma with THZ2, a covalent CDK7 inhibitor. Clin Cancer Res. 2020;26(11):2681-2692. 10.1158/1078-0432.CCR-19-1418. [DOI] [PubMed] [Google Scholar]
- 23. Zhang Z, Peng H, Wang X, et al. Preclinical efficacy and molecular mechanism of targeting CDK7-dependent transcriptional addiction in ovarian cancer. Mol Cancer Ther. 2017;16(9):1739-1750. 10.1158/1535-7163.MCT-17-0078. [DOI] [PubMed] [Google Scholar]
- 24. Hanahan D, Weinberg RA.. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-674. 10.1016/j.cell.2011.02.013. [DOI] [PubMed] [Google Scholar]
- 25. Sava GP, Fan H, Coombes RC, Buluwela L, Ali S.. CDK7 inhibitors as anticancer drugs. Cancer Metastasis Rev. 2020;39(3):805-823. 10.1007/s10555-020-09885-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kanin EI, Kipp RT, Kung C, et al. Chemical inhibition of the TFIIH-associated kinase CDK7/KIN28 does not impair global mRNA synthesis. Proc Natl Acad Sci USA. 2007;104(14):5812-5817. 10.1073/pnas.0611505104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Wang Y, Zhang T, Kwiatkowski N, et al. CDK7-dependent transcriptional addiction in triple-negative breast cancer. Cell. 2015;163(1):174-186. 10.1016/j.cell.2015.08.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Hamilton E, Infante JR.. Targeting CDK4/6 in patients with cancer. Cancer Treat Rev. 2016;45:129-138. 10.1016/j.ctrv.2016.03.002. [DOI] [PubMed] [Google Scholar]
- 29. Neuenschwander B, Branson M, Gsponer T.. Critical aspects of the Bayesian approach to phase I cancer trials. Stat Med. 2008;27(13):2420-2439. 10.1002/sim.3230. [DOI] [PubMed] [Google Scholar]
- 30. Sharma M, Bashir B, Hamilton E, et al. Tolerability and preliminary clinical activity of SY-5609, a highly potent and selective oral CDK7 inhibitor, in patients with advanced solid tumors. Ann Oncol. 2021;32(Suppl 5):S587-S588. 10.1016/j.annonc.2021.08.1040. [DOI] [Google Scholar]
- 31. Kyriakos P.Papadopoulos MRS, Hamilton E, et al. Early Evidence of Dose-Dependent Pharmacodynamic Activity Following Treatment With SY-5609, a Highly Selective and Potent Oral CDK7 Inhibitor, in Patients With Advanced Solid Tumors. Department of Medical Oncology Posters; 2020; 12. [Google Scholar]
- 32. Johannessen L, Ke N, Sawant P, et al. Activity of SY-5609, an oral, noncovalent, potent, and selective CDK7 inhibitor, in preclinical models of colorectal cancer. J Clin Oncol. 2020;38(15_suppl):3585-3585. 10.1200/jco.2020.38.15_suppl.3585. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All preliminary data analyzed are included in this report. The anonymized data are not publicly available due to privacy restrictions and will be made available by request from any qualified investigator. Eli Lilly and Company’s policy is to provide access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. No expiration date of data requests is currently set, once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, and blank or annotated case report forms, will be provided in a secure data sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.