A phase 1 dose-escalation study of checkpoint kinase 1 (CHK1) inhibitor prexasertib in combination with p38 mitogen-activated protein kinase (p38 MAPK) inhibitor ralimetinib in patients with advanced or metastatic cancer
Johanna C. Bendell • Helge G. Bischoff • Jimmy Hwang • Hans Christian Reinhardt • Thomas Zander • Xuejing Wang • Scott Hynes • Celine Pitou • Robert Campbell • Philip Iversen • Daphne L. Farrington • Katherine Bell-McGuinn • Michael Thomas
1 Sarah Cannon Research Institute, Tennessee Oncology PLLC, 250 25th Avenue N, Suite 200 Nashville TN 37203 USA
2 Department of Thoracic Oncology, Translational Lung Research Center Heidelberg (TLRC-H), Member of the German Center for Lung Research (DZL), Heidelberg University Hospital, 69126 Heidelberg Germany
3 Levine Cancer Institute, Atrium Health, Charlotte NC USA
4 University of Cologne, University Hospital Cologne, Clinic I for Internal Medicine, Weyertal 115B 50931 Cologne Germany
5 University of Cologne, Center for Molecular Medicine Cologne, Rober Koch Str. 21 50931 Cologne Germany
6 Clinic I of Internal Medicine, University Hospital of Cologne, Cologne Germany
7 Eli Lilly and Company, Indianapolis IN USA
8 Present address: Verastem Oncology, Needham MA USA
Summary
Purpose The primary objective was to determine the recommended Phase 2 dose (RP2D) of checkpoint kinase 1 inhibitor, prexasertib, in combination with the p38 mitogen-activated protein kinase inhibitor, ralimetinib, which may be safely adminis- tered to patients with advanced cancer. Methods This Phase 1, nonrandomized, open-label, dose-escalation study of prexasertib+ ralimetinib included patients with advanced and/or metastatic cancer, followed by a planned cohort expansion in patients with colorectal or non-small-cell lung cancer with KRAS and/or BRAF mutations. Intravenous prexasertib was administered at 60 mg/ m2 (days 1 and 15 of a 28-day cycle), together with oral ralimetinib every 12 h (days 1 to 14 at 100 mg [Cohort 1, n = 3] or 200 mg [Cohort 2, n = 6]). Dose escalations for each agent were planned using a model-based 3 + 3 escalation paradigm. Safety was assessed using Common Terminology Criteria for Adverse Events (CTCAE) v4.0X. Tumor response was determined by Response Evaluation Criteria in Solid Tumors (RECIST) v1.1. Results Nine patients were treated; 3 experienced dose-limiting toxicities, all in Cohort 2, prohibiting further dose escalation. The most common ≥Grade 3 adverse event was neutrophil count decreased; other reported ≥Grade 3 hematological toxicities included febrile neutropenia and anemia. The pharmacokinetics of prexasertib+ralimetinib was comparable to the monotherapy population profile for each agent. One patient achieved a best overall response of stable disease (for 2 cycles); there were no complete/partial responses. Conclusions This study did not achieve its primary objective of establishing an RP2D of combination prexasertib + ralimetinib that could be safely administered to patients with advanced cancer.
Introduction
Checkpoint kinase 1 (CHK1) is a serine/threonine kinase es- sential to the DNA damage response pathway whereby cells detect and repair damaged DNA, preserving the fidelity of DNA replication during cellular mitosis [1–3]. Upon sensing DNA damage or replication stress, CHK1 acts as a brake to halt progression through the G2/M phase transition of the cell cycle, allowing time for cellular repair of genotoxic lesions before proceeding through mitosis [4]. Besides governing mi- totic entry and exit, CHK1 has also been shown to more di- rectly promote homologous recombination repair of DNA double-strand breaks, initiation of replication origin firing, stabilization of replication forks, and resolution of replication stress, even in the absence of DNA damage [5–7]. Because of this multifactorial role in the DNA damage response and res- olution of replication stress, CHK1 inhibitors have garnered considerable interest as anti-cancer agents, initially simply as potentiators of DNA-damaging radiation therapy and genotoxic chemotherapy [8, 9], but more recently as single- agent therapies [10].
While the classic DNA damage response pathway consists of 2 arms, CHK2/Ataxia telangiectasia mutated (ATM) and CHK1/Ataxia telangiectasia and Rad3 related (ATR) [2], a non-canonical pathway has also been described, involving p38 mitogen-activated protein kinase (MAPK) and its sub- strate MAPK-activated protein kinase 2 (MK2) [2, 11, 12]. This alternate pathway operates in parallel with the CHK1/ ATR pathway, and notably, tumors with KRAS and/or BRAF mutations become dependent on these pathways to prevent cell cycle progression in cells with genotoxic lesions [12]. Therefore, combining p38 MAPK inhibition, which would block MK2 phosphorylation, with CHK1 inhibition may po- tentiate G2/M progression and hence mitotic catastrophe, and may represent a valid therapeutic strategy, particularly in pa- tients with tumors containing KRAS or BRAF mutations [7].
Prexasertib is an ATP-competitive inhibitor of CHK1 with a half-maximal inhibitory concentration (IC50) of<1 nM in cell-free assays [13]. Preclinical studies con- firmed that prexasertib induces DNA double-stranded breaks, chromosome fragmentation and subsequent rep- lication catastrophe in cell lines, and inhibited tumor growth of cancer xenografts in mice [ 13, 14]. Prexasertib was recently evaluated as a single-agent therapy in patients with advanced cancer; this Phase 1 study reported the first objective responses achieved with a CHK1/CHK2 inhibitor as monotherapy [10]. Dose-limiting toxicities (DLTs) all involved Grade 4 neutropenia and thrombocytopenia and/or leukopenia>5 days, occurring at the highest doses (120 and130 mg/m2 given once every 14 days) [10]. A dose- expansion cohort in patients with advanced squamous cell carcinoma demonstrated a 3 month clinical benefitrate (complete response + partial response + stable dis- ease) of 29% across tumor types [15]. Hematological toxicities were the predominant treatment-emergent ad- verse events (TEAEs) across orts in this study, with the most common being Grade 4 neutropenia (71%); 12% of patients experienced febrile neutropenia. Thus, the maximum tolerated dose (MTD) for prexasertib monotherapy was determined to be 105 mg/m2 once every 14 days [10], which was confirmed in the dose- expansion cohort [15].
Ralimetinib is a selective inhibitor of the α- and β-isoforms of p38 MAPK, with an in vitro IC50 of 5.3 and 3.2 nM, re- spectively [16]. Ralimetinib significantly inhibited phosphor- ylation of the downstream p38 MAPK target MAPKAP-K2 (MK2) (>40% reduction) in tumors of mice transplanted with B16-F10 melanoma, and delayed tumor growth in multiple types of human cancer xenografts [16]. In human clinical tri- als, the recommended phase II dose (RP2D) for ralimetinib monotherapy was determined to be 300 mg every 12 h (Q12H) on days 1 to 14 of a 28-day cycle [17]. The most common TEAEs associated with single-agent ralimetinib were rash (25.8%), fatigue (24.7%), and nausea (18%). DLTs included grade 3 ataxia, grade 2 dizziness, and DLT- equivalent rash, all occurring at the maximum dose explored (560 mg, Q12H). No hematological toxicities were observed with single-agent ralimetinib.
In a large scale cell-line-based screen, combined inhibition of CHK1 and the p38 MAPK substrate MK2 demonstrated synergistic cytotoxic effects in cancer cell lines, specifically those with KRAS and BRAF mutations [12]. Dual inhibition of these kinases also inhibited proliferation in xenograft models and enhanced apoptotic cell death in KRAS- or BRAF-mutant patient-derived tumor cells [12]. While this dependency was demonstrated using CHK1 and MK2 inhibitor combinations, synergy of CHK1 with p38 MAPK inhibition was not directly examined. Therefore, we first evaluated the combination of prexasertib (CHK1 inhibitor) and ralimetinib (p38 MAPK in- hibitor) in vitro using KRAS and BRAF mutant tumor cell lines. Based upon observed synergies with the CHK1 and p38 MAPK kinase inhibitors in vitro (See Electronic Supplementary Material), combination prexasertib and ralimetinib therapy was then explored in patients with ad- vanced and/or metastatic cancer. Part A of this study was a dose-escalation with a primary study objective of determining the RP2D of prexasertib and ralimetinib used in combination that may be safely administered to patients with advanced cancer. Secondary objectives were to evaluate safety and tol- erability, to characterize the pharmacokinetics (PK), and to document any antitumor activity of prexasertib and ralimetinib when used in combination. A cohort expansion of prexasertib + ralimetinib in patients with colorectal cancer or non-small cell lung cancer with mutations in KRAS and/or BRAF was also planned.
Methods
Study design and treatment
This was a Phase 1, multicenter, nonrandomized, open-label, dose-escalation study of prexasertib in combination with ralimetinib in patients with advanced and/or metastatic cancer. Patients with histological or cytological diagnosis of advanced or metastatic non-hematological cancer after available stan- dard therapies failed to provide clinical benefit, or for whom no effective therapies existed, were enrolled. Patients could have measurable or nonmeasurable disease as defined by Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 [18] and must have had a performance status of 0 or 1 on the Eastern Cooperative Oncology Group (ECOG) scale [19]. Patients must have discontinued all previous treatments for cancer; have recovered from the acute effects of previous therapies before enrollment; and have had adequate hepatic, renal, and hematologic function (absolute neutrophil count≥1.5 × 109/L, platelet count ≥100 × 109/L, and hemoglobin≥8 g/dL or ≥ 5 mmol). Written informed consent was obtained from all individual participants included in the study. This study complied with ethics principles derived from interna- tional guidelines, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences, and was reviewed and approved by each institution’s ethical review board.
Dose justification was based on the monotherapy MTD for each compound. When administered as monotherapy on Day 1 of a 14-day cycle, the MTD of prexasertib has been defined to be 105 mg/m2, and for ralimetinib, the MTD has been defined as 300 mg when administered on Days 1 to 14 of a 28-day cycle [10, 17]. However, because the current study was the first evaluation of combination prexasertib and ralimetinib therapy, and since ralimetinib has the potential to increase exposure of prexasertib based on in vitro metabolism data (unpublished; data on file), starting doses were reduced to 60 mg/m2 prexasertib and 100 mg ralimetinib. Subsequent dose escalations of ralimetinib and prexasertib were to be driven by a modified 3 + 3 paradigm using a model-based dose-escalation method, with dosing capped at the MTD of each respective agent. In Part A (dose-escalation), Cohort 1 received prexasertib (60 mg/m2) administered as a 60 min intravenous (IV) infusion on Days 1 and 15 of a 28- day cycle and oral ralimetinib (100 mg) Q12H (±3 h) on Days 1 to 14 of each cycle. Cohort 2 received the same dose of prexasertib (60 mg/m2), and oral ralimetinib was increased to 200 mg Q12H. In the event of a DLT, prexasertib could be reduced from 60 mg/m2 to 40 mg/m2 within or between cycles at the discretion of the investigator. There was a planned cohort-expansion (Part B) of prexasertib in combination with ralimetinib in patients with colorectal cancer or non-small-cell lung cancer with known mutations in KRAS and/or BRAF.
Study endpoints and assessments
For safety assessments, TEAEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0X. A DLT was defined as an adverse event (AE) during Cycle 1 that was considered possibly related to the study drugs and that fulfilled any one of the following criteria: any CTCAE Grade 3 or 4 toxicities including: non-hematological toxicity, such as nausea, vomiting, or diarrhea if persisting for >2 days with maximal supportive intervention; fatigue lasting >5 days; alanine ami- notransferase (ALT) and/or aspartate aminotransferase (AST) elevations if persisting >5 days; hematological toxicity, in- cluding Grade 4 neutropenia of >5 days duration, any febrile neutropenia, Grade 4 thrombocytopenia, or Grade 3 thrombo- cytopenia with clinically significant bleeding; and any other significant toxicity deemed by the primary investigator to be dose limiting.
For PK analysis, blood samples were collected at prespecified times according to the protocol schedule to deter- mine the plasma concentrations of prexasertib and ralimetinib using a validated liquid chromatography-tandem mass spec- trometry (LC/MS/MS) method. PK parameters, including maximum plasma concentration (Cmax), area under the plasma concentration vs. time curve (AUC), apparent systemic clear- ance (CL, CL/F), apparent volume of distribution at steady- state (Vss, Vss/F), and terminal elimination half-life (t1/2) for prexasertib and ralimetinib were calculated by standard noncompartmental methods of analysis using Phoenix™ WinNonlin® 7.0 (Pharsight, A Certara Company; Princeton, New Jersey). The observed prexasertib PK data from this study were compared to the population-based PK model for prexasertib monotherapy determined from a previous Phase 1 monotherapy study [10, 15]. Similarly, the observed ralimetinib PK data from this study were compared to the population-based PK model for ralimetinib monotherapy de- termined from the Phase 1 monotherapy study (data on file). The population–based compartmental PK analysis (unpub- lished; data on file) was conducted using NONMEM version7.2 (ICON Development Solutions, Dublin, Ireland). Radiological tumor response assessment using RECIST,version 1.1 [18] was used to determine efficacy. Best overall response (BOR) was determined for each patient who re- ceived at least 1 dose of study therapy. Tumors were evaluated at baseline, then approximately every 8 weeks (every 2 cycles) starting at the end of Cycle 2.
Statistical analyses
Dose escalations were guided by the model-based Neuenschwander continuous reassessment method (NCRM) approach [20]. This method incorporated the prior expecta- tions of the dose-toxicity curve for the combination of bothagents (based on the monotherapy profiles) into the model to provide a recommendation for the respective doses of each agent at the next level. A sample size of up to 35 patients was estimated in dose escalation.
Results
Synergy of prexasertib and ralimetinib combination in vitro
Previous work by Dietlein et al. [12] revealed synergistic cy- totoxic effects in cancer cell lines treated with inhibitors of CHK1 and the p38 MAPK substrate, MK2, specifically those with KRAS and BRAF mutations [12]. We first explored whether dual inhibition of CHK1 and p38 MAPK itself would produce a similar synergistic effect in vitro (Online Resource: Supplementary Methods). Ralimetinib dimesylate and prexasertib monomesylate monohydrate, alone and in combi- nation, were tested for anti-proliferative activity in vitro using a panel of 22 human tumor cell lines (5 colorectal, 2 intestinal, 9 lung, and 6 pancreatic cancer), 14 of which had known activating mutations in KRAS and/or BRAF, or homozygous deletion of cyclin-dependent kinase inhibitor 2A (CDKN2A). Loewe Combination Analysis [21] revealed that this drug combination is synergistic in >50% of the lung (5 of 9) and pancreatic (4 of 6) tumor lines tested (Online Resource: Supplemental Table S2). Only 1 of 7 colorectal/intestinal tu- mor lines showed synergy for this drug combination in vitro. Concentration-response curves indicated that synergy oc- curred when concentrations of ralimetinib are high and prexasertib are low, suggesting that ralimetinib is sensitizing the tumor cells to prexasertib (Online Resource: Supplemental Table S2). Overall, 50% of cell lines containing KRAS, BRAF, and or/ CDKN2A mutations showed synergy for the combina- tion of prexasertib plus ralimetinib (Online Resource: Supplemental Table S2). Having established that inhibition of CHK1 and p38 MAPK kinase was synergistic in vitro, combination prexasertib and ralimetinib therapy was then ex- plored in patients with advanced and/or metastatic cancer.
Patient demographics and treatment
Between August 2016 and May 2017, 9 patients were treated in the dose-escalation portion of the study. Cohort 1 (n = 3) received prexasertib 60 mg/m2 + ralimetinib 100 mg, and Cohort 2 (n = 6) received prexasertib 60 mg/m2 + ralimetinib 200 mg. All participants had received prior anticancer treat- ments: surgery (100%), radiotherapy (56%), and/or systemic therapies (100%). The median age was 65 years (range: 28– 69), and the most common tumor type was colorectal cancer (67%). Baseline demographics and clinical characteristics are summarized in Table 1.
Safety
No patients in Cohort 1 experienced a DLT. In Cohort 2, 1 of the first 3 patients to be enrolled experienced a DLT (Grade 4 febrile neutropenia in Cycle 1); 2 of 3 additional patients ex- perienced DLTs (Grade 4 neutropenia lasting >5 days, both in Cycle 1). The 3 observed DLTs at the second dosing level (Cohort 2) prohibited further dose escalation.
Serious adverse events (SAEs) requiring hospitalization were reported in 3 patients. Two reported SAEs were consid- ered unrelated to study treatment (Grade 3 gastric ulcer, Cohort 2, n = 1; Grade 3 dyspnea, Cohort 1, n = 1). One pa- tient in Cohort 2 reported febrile neutropenia (Grade 4, DLT), which was deemed related to study treatment. No subject died due to an AE while on study treatment or within 30 days of discontinuation from study treatment.
The most common ≥Grade 3 TEAE was neutrophil count decreased, with Grade 4 events experienced by 1 patient (33.3%) in Cohort 1, and 4 patients (66.7%) in Cohort 2 (me- dian duration, 6 days (range, 2–10 days). Two patients in Cohort 2 (33.3%) experienced Grade 3 anemia. There were no Grade 5 events. The most common ≥Grade 3 TEAEs by cohort are listed in Table 2.
Four patients required at least 1 dose adjustment of prexasertib. One patient in Cohort 1 required a dose omission of prexasertib for an AE of blood creatinine increased (Grade 3), and 2 patients in Cohort 2 had one dose omission each for which the reason was missing. One patient (Cohort 2) required a prexasertib dose reduction at the beginning of Cycle 2 due to an AE of neutrophil count decreased (Grade 4), and 1 patient (Cohort 2) required a dose delay of 13 days due to neutrophil count decreased. One patient (Cohort 2) had treatment interrupted due to an infusion-related reaction (Grade 2) dur- ing Cycle 1. Doses of ralimetinib were withheld in 1 patient in Cohort 1 and 2 patients in Cohort 2, including 1 patient who had 1 oral dose withheld for reasons not reported and 1 patient had the dose withheld for 7 consecutive days due to febrile neutropenia. Only 2 patients had dose adjustments for both drugs (prexasertib, for blood creatinine increased and reasons not reported; ralimetinib, for febrile neutropenia and reasons not reported).
Most patients discontinued treatment due to progressive disease (77.8%; Cohort 1, n = 2; Cohort 2, n = 5). One patient (11.1%) in Cohort 1 discontinued study treatment because of an SAE considered unrelated to study treatment (Grade 3 dys- pnea), and 1 patient (11.1%) in Cohort 2 who had an event of Grade 4 neutropenia (also a DLT) withdrew consent.
Pharmacokinetics for combination therapy
Prexasertib noncompartmental PK parameters are summa- rized in Table 3, including the observed maximum plasma concentration (Cmax) of prexasertib for each Cohort at Day 1and 15 of each cycle. Intracycle and intercycle accumulation was observed, with the mean accumulation ratio (RA) ranging from 1.04 to 1.46. The terminal half-life (t1/2) varied depend- ing on cycle and days of treatment, ranging from 9.55 to19.4 h. Prexasertib PK parameters were similar between Cohorts 1 and 2 across days and cycles independent of ralimetinib dose. After prexasertib administration of 60 mg/ m2 in combination with ralimetinib 100 or 200 mg Q12H, the mean prexasertib AUC(0-72) values on some days during Cycle 1 and Cycle 2 (Table 3) were greater than the median AUC(0-72) (1896 ng·hr./mL) predicted before clinical investi- gation to achieve the maximal tumor response with prexasertib monotherapy [10]. Finally, the observed prexasertib individual PK data from the prexasertib (60 mg/ m2) and ralimetinib (100 or 200 mg) combination in this study coincided relatively well with the prexasertib population PK model-predicted monotherapy profile (Fig. 1). Therefore, the PK data show no evidence of a significant PK drug-drug in- teraction between prexasertib and ralimetinib at the dose levels evaluated.
PK parameters of ralimetinib on Days 1 and 14 when ad- ministered in combination with prexasertib are summarized in Table 4. Median Cmax was reached 1.57 h after dosing on Day 1 of the cycle and 1.02 h after dosing on Day 14. The PK profile of ralimetinib displayed a multicompartmental model
with a large peripheral volume of distribution (average of 1400 L for the apparent volume of distribution) and a long terminal elimination phase (average of 181 h for t1/2), with an apparent total body clearance of 39.2 L/h on average. After administration of ralimetinib in combination with prexasertib, the AUC(0-24,ss) values in Cycle 1 were slightly lower com- pared to the predicted daily exposure of 13,400 ng·hr./mL needed to achieve biologically effective dose (data on file, [17]). The PK profiles of individual patients given 100 mg or 200 mg ralimetinib in combination with 60 mg/m2 prexasertib in this study were comparable with the corre- sponding PK-predicted profiles in the ralimetinib monothera- py population (Fig. 2).
Efficacy
In Cohort 1, 1 patient (33%) achieved a BOR of stable disease after 2 cycles of treatment, but progressed on assessment at Cycle 4; 1 patient had a BOR of progressive disease (33%); and 1 patient was non-evaluable. In Cohort 2, 5 patients had a BOR of progressive disease (83.3%) and 1 patient died before a next scan and was therefore non-evaluable due to withdraw- al from the study in Cycle 1. There were no complete or partial responses.
Discussion
We evaluated the combination of prexasertib and ralimetinib in patients with advanced and/or metastatic cancer, based on the hypothesis that combining parallel pathways of CHK1 inhibition with p38 MAPK inhibition may potentiate the ef- fects on mitotic progression in cells with DNA damage, and thus enhance mitotic catastrophe and tumor cell death. When administered as single agents, prexasertib and ralimetinib ex- hibit acceptable safety, tolerability, and pharmacokinetics inpatients with advanced cancer [15, 17]. The RP2D for mono- therapy are 105 mg/m2 every 14 days for prexasertib and 300 mg Q12H for 14 days of a 28-day cycle for ralimetinib [10, 15, 17]. In the current study, the dosing regimen for the combination therapy (60 mg/m2 prexasertib and 100 mg [Cohort 1] or 200 mg ralimetinib [Cohort 2]) contained doses well below the monotherapy RP2Ds. There were no DLTs observed in Cohort 1. However, 3 observed DLTs at the sec- ond dosing level (prexasertib 60 mg/m2 + ralimetinib 200 mg) prohibited further dose escalation and planned cohortexpansion in patients with tumors containing KRAS or BRAF mutations. In the original Phase 1 dose-escalation study of prexasertib as a single-agent therapy, the lowest dose at which a DLT occurred when prexasertib was administered once ev- ery 14 days was 120 mg/m2 [10], and no participants experi- enced a DLT at a dose of 60 mg/m2 given once every 14 days. The observation in the current study that patients treated with the prexasertib 60 mg/m2 plus ralimetinib 200 mg/kg combi- nation had more severe AEs, including DLTs at this lower dose of prexasertib, than those treated with each agent alone, suggests a potential unanticipated increase in toxicity of the combination.
In this study, the most common observed Grade 4 AE was decreased neutrophil count, experienced by 5 of 9 (55.6%) patients. Of note, initial onset of these Grade 4 decreased neutrophil count events occurred early during Cycle 1 (<28 days), consistent with what has been reported with prexasertib monotherapy. The 1 reported incident of febrile neutropenia (Grade 4) occurred at Day 7 following the start date of study treatment. However, the duration of grade 4 decreased neutrophil count (median 6, range, 2–10 days) was longer than that reported for prexasertib monotherapy. Hematological toxicities were also the most common AEs reported in the 2 previously published prexasertib monother- apy studies [10, 15]. In contrast, no hematological AEs were reported for single-agent ralimetinib [17].
Despite the relatively small number of patients in this study, the plasma concentrations of both prexasertib and ralimetinib used in combination coincided relatively well with the monotherapy population PK model- predicted systemic exposure profile for each drug. The starting dose of 60 mg/m2 was well below the established MTD of 105 mg/m2 for monotherapy prexasertib, based on in vitro metabolism data (unpub- lished; data on file) suggesting that when administered in combination with ralimetinib, exposure of prexasertib could be increased. Indeed, some of the observed prexasertib plasma concentrations with combination therapy were slightly higher compared to the prexasertib monotherapy population PK profile. However, the PKprofiles of the 3 patients who experienced a DLT were consistent and not atypical compared with the prexasertib monotherapy population PK profile. Additionally, the ralimetinib PK profiles of the same 3 patients were also consistent and not atypical compared with the ralimetinib monotherapy population PK profile after a 200 mg dose. Together, the PK data show no evidence of a significant pharmacokinetic drug-drug in- teraction between prexasertib and ralimetinib at the dose levels evaluated. Therefore, the increased toxicity seen with the combination therapy is likely not caused by increased exposure of each agent but could be due to a potential unanticipated interaction of the CHK1 and p38 MAPK pathways in the bone marrow compartment. It is also important to note that p38 MAPK can phos- phorylate other substrates in addition to MK2, including MK3 and possibly MK5 [22]. Inhibition of these, or other, downstream targets of p38 MAPK may contribute to the observed additive toxic effects of this drugcombination. Further study will be needed to determine why blockade of the p38 MAPK pathway together with inhibition of cell cycle checkpoint control has an addi- tive effect on hematological parameters.
Due to DLTs in this dose-escalation part of the study (Part A), the planned cohort expansion in patients with colorectal cancer or non-small-cell lung cancer with known activating mutations in KRAS and/or BRAF did not occur. Our preclinical data supports the hypothesis that combined inhibition of CHK1 and p38 MAPK may have synergistic cytotoxic effects. However, 50% of the mutant cell lines tested did not show synergy in vitro, suggesting that other factors beyond KRAS, BRAF, and/ or CDKN2A may also contribute to synergism. The KRAS and/or BRAF status of patients enrolled in Part A of this study were not available; therefore, no con- clusions can be drawn regarding the possibility of syn- ergy using combined treatment with inhibitors of CHK1 and p38 MAPK pathways in vivo.
In summary, while preclinical efficacy signals of the com- bination of prexasertib and ralimetinib were intriguing, hema- tological toxicity in patients makes this combination not via- ble to explore further in the clinic. Mechanistically, further understanding of the interactions of the p38 MAPK and CHK1 pathways in the bone marrow compartment are needed to explain the toxicity observed in this trial. Caution should be taken when combining inhibitors of these two pathways in the clinic moving forward. Overall, this study did not achieve its primary objective of establishing a RP2D of prexasertib and ralimetinib used in combination that could be safely adminis- tered to patients with advanced cancer.
Conflict of interest Dr. Bendell reports the following: grants (payment to institution for conduct of clinical trials for which Dr. Bendell served as PI) from EMD Serono, Koltan, SynDevRex, Forty Seven,
References
1. Smith J, Tho LM, Xu N, Gillespie DA (2010) The ATM-Chk2 and ATR-Chk1 pathways in DNA damage signaling and cancer. Adv Cancer Res 108:73–112
2. Reinhardt HC, Yaffe MB (2009) Kinases that control the cell cycle in response to DNA damage: Chk1, Chk2, and MK2. Curr Opin Cell Biol 21(2):245–255
3. Reinhardt HC, Yaffe MB (2013) Phospho-Ser/Thr-binding do- mains: navigating the cell cycle and DNA damage response. Nat Rev Mol Cell Biol 14(9):563–580
4. Syljuåsen RG, Hasvold G, Hauge S, Helland Å (2015) Targeting lung cancer through inhibition of checkpoint kinases. Front Genet 6:70
5. Sørensen CS, Hansen LT, Dziegielewski J, Syljuåsen RG, Lundin C, Bartek J et al (2005) The cell-cycle checkpoint kinase Chk1 is required for mammalian homologous recombination repair. Nat Cell Biol 7(2):195–201
6. Syljuåsen RG, Sørensen CS, Hansen LT, Fugger K, Lundin C, Johansson F et al (2005) Inhibition of human Chk1 causes in- creased initiation of DNA replication, phosphorylation of ATR tar- gets, and DNA breakage. Mol Cell Biol 25(9):3553–3562
7. McNeely S, Beckmann R, Bence Lin AK (2014) CHEK again: revisiting the development of CHK1 inhibitors for cancer therapy. Pharmacol Ther 142(1):1–10
8. Thompson R, Eastman A (2013) The cancer therapeutic potential of Chk1 inhibitors: how mechanistic studies impact on clinical trial design. Br J Clin Pharmacol 76(3):358–369
9. Dent P, Tang Y, Yacoub A, Dai Y, Fisher PB, Grant S (2011) CHK1 inhibitors in combination chemotherapy: thinking beyond the cell cycle. Mol Interv 11(2):133–140
10. Hong D, Infante J, Janku F, Jones S, Nguyen LM, Burris H et al (2016) Phase I study of LY2606368, a checkpoint kinase 1 inhibi- tor, in patients with advanced cancer. J Clin Oncol 34(15):1764– 1771
11. Reinhardt HC, Aslanian AS, Lees JA, Yaffe MB (2007) p53- deficient cells rely on ATM- and ATR-mediated checkpoint signal- ing through the p38MAPK/MK2 pathway for survival after DNA damage. Cancer Cell 11(2):175–189
12. Dietlein F, Kalb B, Jokic M, Noll EM, Strong A, Tharun L et al (2015) A synergistic interaction between Chk1- and MK2 inhibitors in KRAS-mutant cancer. Cell. 162(1):146–159
13. King C, Diaz HB, McNeely S, Barnard D, Dempsey J, Blosser W et al (2015) LY2606368 causes replication catastrophe and antitu- mor effects through CHK1-dependent mechanisms. Mol Cancer Ther 14(9):2004–2013
14. Wu W, Bi C, Bence AK, Um SL, Yan B, Starling JJ et al (2012) Abstract 1776: Antitumor activity of Chk1 inhibitor LY2606368 as a single agent in SW1990 human pancreas orthotopic tumor model. Cancer Res 72(8 Suppl):1776
15. Hong DS, Moore K, Patel M, Grant SC, Burris HA III, William WN Jr et al (2018) Evaluation of prexasertib, a checkpoint kinase 1 inhibitor, in a phase Ib study of patients with squamous cell carci- noma. Clin Cancer Res 24(14):3263–3272
16. Campbell RM, Anderson BD, Brooks NA, Brooks HB, Chan EM, De Dios A et al (2014) Characterization of LY2228820 dimesylate, a potent and selective inhibitor of p38 MAPK with antitumor ac- tivity. Mol Cancer Ther 13(2):364–374
17. Patnaik A, Haluska P, Tolcher AW, Erlichman C, Papadopoulos KP, Lensing JL et al (2016) A first-in-human phase I study of the oral p38 MAPK inhibitor, ralimetinib (LY2228820 dimesylate), in pa- tients with advanced cancer. Clin Cancer Res 22(5):1095–1102
18. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R et al (2009) New response evaluation criteria in solidtumours: revised RECIST guideline (version 1.1). Eur J Cancer 45(2):228–247
19. Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, McFadden ET et al (1982) Toxicity and response criteria of the eastern cooperative oncology group. Am J Clin Oncol 5(6):649– 655
20. Neuenschwander B, Branson M, Gsponer T (2008) Critical aspects of the Bayesian approach to phase I cancer trials. Stat Med 27(13): 2420–2439
21. Loewe S (1953) The problem of synergism and antagonism of combined drugs. Arzneimittelforschung. 3(6):285–290
22. Roux PP, Blenis J (2004) ERK and p38 MAPK-activated protein kinases: a family of protein kinases with diverse biological func- tions. Microbiol Mol Biol Rev 68(2):320–344