The Checkpoint Kinase 1 Inhibitor Prexasertib Induces Regression of Preclinical Models of Human Neuroblastoma

Neuroblastoma is derived from neural crest precursor cells of the peripheral sympathetic nervous system, typically developing in the adrenal medulla or paraspinal ganglia (1, 2). These tumors comprise 5% of all childhood malignancies and 10% of pediatric patient deaths, emphasizing the critical need for novel therapies (1). Clinical behavior of the disease ranges from localized tumors which can be cured with surgical excision to invasive and/or metastatic disease which is often refractory to aggressive multimodal treatment regimens (3). Patient risk is stratified on the basis of age at diagnosis; tumor characteristics such as stage, grade, and histology; DNA ploidy; and status of MYCN genomic amplification (4). Nearly half of high-risk neuroblastoma patients experience a relapse, which is usually fatal; those who survive are often faced with damaging complications from intensive therapeutic interventions composed of surgery, chemotherapy, and radiation (1, 3). Therefore, it is essential to evaluate targeted agents to offer these patients more efficacious treatment options.

The serine/threonine kinase checkpoint kinase 1 (CHK1) is critical for replication initiation through licensing of replication forks; furthermore, CHK1 regulates DNA damage response and repair mechanisms following stalled replication forks or single-strand DNA breaks through its modulation of the S-phase and G₂–M cell-cycle checkpoints (5–7). Loss of CHK1 activity due to pharmacologic inhibition or knockdown of total protein by RNA interference in specific tumor types impairs the DNA damage response and the ability to mitigate replication stress, leading to stalled forks, double-strand DNA breaks, and eventual cell death via replication catastrophe (8). Similarly, neuroblastoma cell viability was reduced upon depletion or pharmacologic inhibition of CHK1 and, interestingly, sensitivity to CHK1 inhibition was not influenced by p53 status or baseline level of DNA damage (9).

Prexasertib (LY2606368) is a CHK1 small-molecule inhibitor currently in clinical evaluation as both a single agent and in combination with targeted agents or chemotherapy in adult patients with solid tumors (10). Furthermore, clinical development of the molecule includes a phase I trial in pediatric solid tumors (NCT02808650). A previous preclinical study reported the efficacy of prexasertib in promoting extensive DNA damage in adult carcinoma cell lines and xenograft mouse models, which led to cell death due to replication catastrophe (8). Here, we demonstrate that single-agent prexasertib induced tumor regression in multiple preclinical mouse models of neuroblastoma.

Human neuroblastoma cell lines IMR-32 and SH-SY5Y, pancreatic cell line PANC-1, and primary neonatal epidermal melanocytes were purchased from ATCC. KELLY, MHH-NB-11, and NBL-S were obtained from The German Collection of Microorganisms and Cell Cultures (DSMZ). Cell culture conditions are detailed in Supplementary Table S1. All cells were maintained at 37°C and 5% CO₂ in tissue-culture–treated flasks.

Preclinical studies use 2940930, which is the mesylate monohydrate salt of LY2606368 (prexasertib), and will be henceforth referred to as prexasertib. Prexasertib (LY2606368, Eli Lilly and Company) was dissolved in DMSO at a stock concentration of 10 mmol/L for in vitro use and prepared in 20% Captisol for in vivo experiments.

Profiling of prexasertib in a panel of more than 300 cancer cell lines was performed as previously reported in ref. 11. For comparison with standard of care (SOC), pediatric cancer cell lines were plated in 96-well microtitre plates and treated with prexasertib, doxorubicin, cisplatin, or gemcitabine across a range of concentrations. Cell proliferation was assayed after two doublings by CellTiter Glo Luminescent Cell Viability Assay (Promega, catalog no. G7571). For in-depth evaluation of pediatric neuroblastoma cell lines, PANC-1, and primary melanocytes, cell proliferation was assayed after 72 hours. Luminescence was normalized to the average of the DMSO control for each individual cell line and plotted as the percent of control over the log concentration. EC₅₀ values were calculated from triplicate experiments using GraphPad Prism 6 (GraphPad Software, Inc., Version 7.00).

Cells were lysed in 1% SDS (Fisher BioReagents, catalog no. BP2436-200) supplemented with 1x HALT protease and phosphatase inhibitor (Thermo Fisher Scientific, catalog no. 78440); lysates were briefly sonicated and boiled at 95°C. Protein was quantified using the DC Protein Assay (Bio-Rad, catalog no. 5000116). Whole-cell lysates (30–50 μg of protein per well) were electrophoresed on 4%–20% Tris-Glycine gels (Novex, Thermo Fisher Scientific) and transferred using a semi-dry method to nitrocellulose (Bio-Rad, catalog no. 170-4159). Membranes were blocked in 5% milk diluted in 1× Tris-buffered Saline + 0.1% Tween-20 (1× TBST), probed with primary antibodies diluted in 5% BSA in 1× TBST, and incubated with horseradish peroxidase–conjugated secondary antibodies diluted in 5% milk. Protein was visualized using SuperSignal West Femto Chemiluminescent Substrate (Thermo Fisher Scientific, catalog no. 34095) and imaged with a Bio-Rad ChemiDoc XRS. Antibodies and conditions are listed in Supplementary Table S2.

High-content cell imaging and subsequent analysis were conducted as described previously (12, 13). Briefly, neuroblastoma cells, PANC-1, or primary melanocytes were seeded in clear-bottom black 96-well plates coated with poly-d-lysine. After transfection with siRNA and/or treatment with prexasertib, cells were fixed in 3.7% formaldehyde (Sigma, catalog no. F-1268) or 1× PREFER (Anatech Ltd, catalog no. 414) in D-PBS, permeabilized with 0.1% Triton X-100 in D-PBS, and blocked with 1% BSA in D-PBS. Cells were incubated with primary antibodies overnight, followed by three washes and incubation with secondary antibodies for 1 hour at room temperature. Antibody details are shown in Supplementary Table S2. DNA was stained with Hoescht 33342. TUNEL was performed using the In Situ Cell Death Detection Kit, Fluorescein (Sigma Aldrich, catalog no. 11684795910). Cells were imaged using a CellInsight NXT platform and analyzed by the TargetActivation V.4 Bioapplication (Thermo Fisher Scientific). Percent responders (percent positive for desired marker) were gated on the basis of the DMSO-treated group for each cell line.

Individual and pooled siRNA against CHK1 and nontargeting siRNA were purchased from Dharmacon (Supplementary Table S3). Neuroblastoma and control cells were reverse-transfected with siRNA using the Lipofectamine-RNAiMax according to the manufacturer's protocol (Thermo Fisher Scientific, catalog no. 13778-075). Degree of knockdown was evaluated 72 hours posttransfection by Western blot analysis. For experiments evaluating the effects of prexasertib treatment on CHK1 knockdown cell lines, treatment commenced 48 hours posttransfection for an additional 24 hours.

In vivo studies were approved by the Eli Lilly and Company Animal Care and Use Committee. To evaluate the effects of prexasertib on neuroblastoma xenograft growth, cells were harvested during log phase growth and resuspended in Hank balanced salt solution (HBSS). Suspended cells were diluted 1:1 with BD Matrigel Matrix (catalog no. 356234) and 5 × 10⁶ cells (0.2-mL cell suspension) were injected subcutaneously into the right flank of female CB-17 SCID beige mice. Tumors were monitored beginning 7 days after injection. When tumor volumes averaged 200 or 500 mm³, mice were randomized into treatment groups (n = 6/group) based on tumor volume and body weight. Animals were given vehicle (20% Captisol in water, pH 4) or 10 mg/kg prexasertib by subcutaneous injection twice daily for 3 days, followed by 4 days of no treatment for a total of 4 weeks [(twice daily × 3 days, rest × 4days) × 4 weeks)], unless otherwise indicated in the figure legend. For combination studies, two study arms were added and animals were given 5 mg/kg doxorubicin intravenously with or without 10 mg/kg prexasertib [(twice daily × 3 days, rest × 4 days)× 4 weeks].

To evaluate the effects of prexasertib on tumor health and angiogenesis, cells were injected and tumors monitored as described above. Animals were sacrificed 6 days after the beginning of treatment and tumors were promptly excised and fixed in 10% neutral buffered formalin. Multiplexed immunohistochemical analysis was performed on vehicle- and prexasertib-treated tumors as described previously (14). TUNEL was performed using the Roche kit (catalog no. 12156792910) and antibodies are described in Supplementary Table S2.

Prexasertib was evaluated across a panel of well-characterized, commercially available cancer cell lines encompassing a wide spectrum of adult and pediatric malignancies. The EC₅₀s of several pediatric tumor types, including neuroblastoma, fell below the average plasma concentration [C_avg at 24 hours postinfusion (46.9 ng/mL)] reported in a phase I trial in adult carcinoma patients following the dosing schedule of 105 mg/m² via infusion on Day 1 every 14 days (Fig. 1; Supplementary Table S4; ref. 10). When compared with a series of SOC in vitro, single-agent prexasertib was more potently antiproliferative in 19 pediatric cancer cell lines (Table 1). Five neuroblastoma cell lines were selected for detailed characterization with respect to CHK1 inhibition in this tumor type. Primary neonatal melanocytes share the neural crest lineage with neuroblastoma and are derived from infant foreskin, providing an age-appropriate normal cell control (15). The adult pancreatic cancer cell line PANC-1, previously reported to have an EC₅₀ of prexasertib above 1 μmol/L, served as an intrinsically prexasertib-resistant cell line (8). Consistent with the results of the cancer cell line sensitivity panel, additional cellular assays confirmed the antiproliferative activity of prexasertib in the low nanomolar range (Fig. 2A).In addition, prexasertib induced apoptosis, as evidenced by increased activation of capases 3 and 7, within 24 hours of treatment in the majority of neuroblastoma lines evaluated (Supplementary Fig. S1).

Upon replication stress and/or DNA damage, CHK1 is directly phosphorylated on serines 317 and 345 (S317 and S345) by the DNA damage sensing kinase ataxia telangiectasia and Rad3-related (ATR) in an ATM-independent or -dependent manner, depending on the nature of the genotoxic stress (16–19). Phosphorylation at both S317 and S345 is required for autophosphorylation of CHK1 at serine 296 (S296), resulting in full activation of the kinase (20, 21). As increased transcriptional activity can cause replication stress, neuroblastoma cell lines with high MYC levels due to gene amplification or increased expression were selected for further analysis. Endogenous levels of C-MYC and N-MYC protein were confirmed via Western blot analysis of KELLY, NBL-S, and SH-SY5Y lysates (Fig. 2A). Baseline phosphorylation of CHK1 at S296 has been reported in KELLY cells (9); in addition, endogenous CHK1 S296 was detected in NBL-S and SH-SY5Y as well as PANC-1 (Fig. 2B). Prexasertib treatment over 24 or 48 hours reduced phosphorylation at CHK1 S296 (Fig. 2B) and led to an accrual of γH2AX-positive double-strand DNA breaks in neuroblastoma cell lines, primary melanocytes, and PANC-1 cells (Fig. 3; Supplementary Table S5). In response to CHK1 inhibition and the resulting double-strand breaks, activated DNA damage sensors ataxia telangiectasia mutated (ATM; phosphorylated at serine 1981) and DNA protein kinase-catalytic subunit (DNA-PKcs; phosphorylated at serine 2056) localized to the nucleus in KELLY and NBL-S posttreatment (Supplementary Fig. S2A; Supplementary Table S5). Activation of the DNA damage response was confirmed by concomitant increases in phosphorylation of CHK2 at threonine 68 and CHK1 at S345, which are target sites for ATM and ATR (Fig. 2B). Furthermore, both total and phosphorylated replication protein A 32/2 (RPA32/2), necessary for stabilizing stalled replication forks and coating single-strand DNA, were elevated within 24 hours of prexasertib treatment (Supplementary Fig. S3). Interestingly, although prexasertib led to double-strand DNA breaks in all of the cell lines, apoptosis was observed in neuroblastoma cells, but not PANC-1 or primary melanocytes, as measured by increased cleaved PARP, cleaved caspase-3, and TUNEL (Figs. 2B and 3; Supplementary Fig. S2B; Supplementary Table S5). This dichotomous outcome was not linked to the expression of cyclin-dependent kinase 2 (CDK2) or the phosphatase CDC25A, two proteins previously shown to be necessary for prexasertib efficacy (Supplementary Fig. S3.)

While prexasertib preferentially binds to and inhibits the activity of CHK1, prexasertib also inhibits checkpoint kinase 2 (CHK2) with an IC₅₀ of <10 nmol/L in vitro, while inhibition of CHK2 autophosphorylation was achieved with an IC₅₀ of <31 nmol/L in cellular assays (8). To further validate CHK1 as the functional target of prexasertib in neuroblastoma, CHK1 and CHK2 were transiently knocked down using siRNA in KELLY, which was the neuroblastoma cell line most sensitive to prexasertib and the intrinsically resistant cell line PANC-1, either individually or in combination, and then treated with DMSO or prexasertib (Fig. 4; Supplementary Fig. S4). Depletion of CHK1, but not CHK2, increased γH2AX and cleaved PARP in the absence of prexasertib in KELLY cells (Fig. 4A). Furthermore, knockdown of CHK2 did not alter the antiproliferative response of KELLY cells to prexasertib treatment, while siCHK1 drastically reduced proliferation relative to the siNT control (Fig. 4B). Prexasertib treatment of KELLY cells with siCHK1 did not alter the observable effects on PARP cleavage or γH2AX levels. As expected due to its observed intrinsic resistance to prexasertib, PANC-1 cell proliferation was not affected by CHK1 or CHK2 knockdown.

As prexasertib reduced neuroblastoma cell viability in vitro, prexasertib was evaluated in vivo using IMR-32 or KELLY subcutaneous xenograft mouse models. To examine the effects of prexasertib in the context of primary lesion size, treatment was initiated when tumors reached an average volume of 200 or 500 mm³. Importantly, clinically relevant plasma concentrations of prexasertib were achieved in these mouse models using the 3 days on twice daily treatment, 4 days rest dosing schedule (8, 10).

Durable complete regressions were observed in prexasertib-treated animals harboring KELLY xenografts, regardless of initial tumor size (Fig. 5A). Similarly, IMR-32 xenografts initially and completely regressed upon administration of prexasertib. However, unlike KELLY xenografts which did not regrow after the end of treatment, IMR-32 tumor regrowth was observed 4 weeks after treatment cessation (Fig. 5C). The reemergent tumors retained some sensitivity to prexasertib as evidenced by an ensuing partial regression; however, complete regression was not achieved following rechallenge with drug. No significant changes in body weight were observed during the treatment period (Supplementary Fig. S5).

First-generation CHK1 inhibitors have been used as chemopotentiators, enhancing the genotoxic stress triggered by DNA-damaging chemotherapies via abrogation of cell-cycle checkpoints (22). Recent studies have demonstrated that single agent CHK1 inhibition in the absence of exogenous DNA damage is sufficient to cause tumor cell death (8, 23). While single-agent prexasertib was sufficient to induce KELLY xenograft regression in mouse models, this was not the case in IMR-32 and the C-MYC–driven neuroblastoma model SH-SY5Y, suggesting combination treatment may yield synergistic antitumor effects. To evaluate the combinatorial effects of prexasertib plus SOC, IMR-32 and SH-SY5Y xenograft models were treated with doxorubicin, which has been used in multimodal pediatric neuroblastoma treatment regimens for more than 40 years (24–26), with/without prexasertib (Supplementary Fig. S6). No superior effects on tumor growth inhibition were observed with combination treatment when compared with the single-agent prexasertib arms, indicating that using this dose and schedule doxorubicin treatment neither enhances prexasertib efficacy nor prevents tumor recurrence in these preclinical neuroblastoma models.

Because of the rapid regression of tumors upon initial treatment with prexasertib, two additional study arms were sacrificed 6 days after dosing commenced. These tumors were subjected to multiplexed IHC to assess tumor health and angiogenesis (Fig. 5B and D; Supplementary Fig. S7). Cell proliferation, as measured by Ki67, significantly decreased with prexasertib treatment in both tumor models, although there was more Ki67 after treatment in IMR-32 cells (Fig. 5D). The number of TUNEL–positive tumors cells did not change within the first 6 days of treatment. Surprisingly, increased coincident expression of vascular markers MECA-32, CD31, and smooth muscle actin (SMA) was detected in prexasertib-treated xenografts compared with the control, indicating an enrichment of pericyte-covered vessels in both models (Supplementary Fig. S7). Furthermore, significantly fewer small vessels were observed in prexasertib-treated IMR-32 xenografts, while an increase in the number of large vessels was noted (Supplementary Fig. S7C).

The increase in vascular markers and large vessels combined with fewer smaller vessels following prexasertib treatment of neuroblastoma xenografts suggested that prexasertib may inhibit the formation of new vessels while leaving the established vasculature relatively unaffected. Therefore, we utilized an in vitro endothelial cord formation assay to model the potential activity of prexasertib on the vascular endothelium. Treatment with increasing concentrations of prexasertib strongly reduced VEGF-driven cord formation, as measured by total tube area, back to basal levels but did not significantly alter established cords (Supplementary Fig. S8). In addition, we concluded that the prexasertib-dependent decrease in cord formation was not due to treatment effects on the feeder layer. The ability of neuroblastoma cell lines to support cord formation was also investigated in a tumor-driven cord formation assay; the total tube area of neuroblastoma-driven cords was comparable with VEGF treatment (Supplementary Fig. S9A and S9B). Furthermore, IMR-32, KELLY, and SH-SY5Y cells were found to produce VEGF, VEGF-C, and VEGF-D in coculture conditions, suggesting that these neuroblastoma cell lines could promote neovascularization in vitro and in vivo (Supplementary Fig. S9C).

CHK1 is a master regulator of replication fork licensing and of cell-cycle checkpoints in response to genotoxic stress and DNA damage. Inhibition of its activity by either therapeutic agents or RNA interference leads to excessive replication origin firing, exposing single-strand DNA to endonucleases and resulting in double-strand DNA breaks. In addition, abrogation of the intra-S-phase and G₂–M cell-cycle checkpoints allows for unbridled progression through the cell cycle regardless of the integrity of the genome. Prexasertib (LY2606368), a second-generation small-molecule inhibitor of CHK1, promotes cell death through replication catastrophe in solid tumor models, leading to clinical evaluation of the drug in adult cancer patients (8, 10). In this study, we report that prexasertib is potently antiproliferative in models of pediatric cancer, especially neuroblastoma. Furthermore, prexasertib induces regression of neuroblastoma xenografts in mouse models. High-risk neuroblastoma is a devastating pediatric malignancy responsible for approximately 10% of pediatric cancer deaths annually and leaves survivors with debilitating long-term side effects following severe combinatorial therapeutic regimens.

Previously, an RNAi screen identified CHK1 as a therapeutic target in neuroblastoma cell lines. Furthermore, recent studies have shown that treatment with CHK1 inhibitors, either alone or in combination with systemic or targeted agents, was sufficient to kill neuroblastoma cells (9, 27). As a single agent, prexasertib was more potent in pediatric cancer cell lines than several current SOC agents. Although efficacy was not directly compared with other targeted agents, prexasertib treatment reduced cell proliferation in pediatric cell lines at low nanomolar concentrations, approximately 10- to 100-fold less than EC₅₀ values reported for first-generation CHK1/2 inhibitors in neuroblastoma models (9, 27, 28). Prexasertib can also inhibit CHK2, though it is unlikely that blockade of CHK2 kinase activity contributed to the rapid cell death observed following drug treatment. CHK2 is a potential tumor suppressor gene in adult tumors due to its role in the DNA damage response and regulation of p53; however, oncogenic alterations or mutations of CHEK2 are rare in neuroblastoma and other pediatric tumor types, suggesting that CHK2 activity is not essential for neuroblastoma tumorigenesis (29). Indeed, knockdown of CHK2 in KELLY cells did not increase DNA double-strand DNA breaks, induce the DNA damage response, or cause cell death, while depletion of CHK1 recapitulated the effects of prexasertib treatment, validating that CHK1 is the functional target of prexasertib in neuroblastoma. Importantly, treatment of CHK1-depleted cells with prexasertib did not enhance PARP cleavage or γH2AX levels, indicating that off-target effects of prexasertib resulting in DNA damage and cell death are unlikely and furthering the notion that the efficacy of prexasertib in neuroblastoma is primarily a result of CHK1 inhibition.

The tumor suppressor p53 is a key mediator of the G₁ cell-cycle checkpoint in response to DNA damage (30, 31). In p53-deficient tumors, cells are more reliant on the intra-S and G₂–M checkpoints regulated by CHK1 to maintain a level of genomic stability necessary for continued proliferation and survival. In primary neuroblastoma, TP53 mutations are rare (32); however, the p53 pathway has been shown to be repressed due to direct inhibition by the ubiquitin ligase MDM2 and/or loss of key nodes within the pathway which could prevent activation of the G₁ checkpoint (33, 34). In addition, mutant TP53 and/or inactivation of the p53 pathway have been identified in some cell lines derived from recurrent tumors (35–37). All neuroblastoma cell lines evaluated in this study have intact wild-type p53 (33, 38, 39) and yet are highly sensitive to prexasertib treatment. Although other small molecule inhibitors of CHK1 have been shown to be particularly effective in combination with SOC in p53-deficient tumors due to more permissive S-phase entry of cells with damaged DNA (28, 40, 41), previous data along with this study suggests that p53 status does not influence prexasertib efficacy as a monotherapy (8).

In addition to its well-characterized role in enforcing the DNA-damage response through monitoring the intra-S and G₂–M phase cell-cycle checkpoints, CHK1 is also responsible for regulating replication fork licensing to minimize replication stress and maintain genome integrity. Therefore, it is possible that inhibition of CHK1 may lead to DNA damage and cell death due to replication catastrophe in some tumor types where replication stress is inherently high due to elevated rates of cellular proliferation or expression of specific oncogenes. For example, augmented replication origin firing and increased replication stress have been reported in mammalian cells with increased expression of the MYC family of transcription factors (42). Genomic amplification of MYCN is observed in approximately 20% of neuroblastoma cases and denotes poor prognosis (4). MYC family expression is prevalent in established neuroblastoma cell lines, with the amplicon present in KELLY, IMR-32, and MHH-NB-11, and high C-MYC was expression detected in SH-SY5Y, which may contribute to the sensitivity of neuroblastoma to CHK1 inhibition. However, prexasertib has not yet been evaluated in neuroblastoma cell lines with endogenously low levels of N-MYC and C-MYC expression with or without the MYCN amplicon to fully understand the contribution of MYC activity to prexasertib sensitivity. Interestingly, a reduction in N-MYC or C-MYC protein was observed in neuroblastoma and PANC-1 cells following 48-hour treatment with prexasertib (Fig. 2B), suggesting that inhibition of CHK1 activity may potentially influence MYC protein expression or stability.

An unexpected finding of our study is the effect of prexasertib on xenograft vasculature, whereby drug treatment resulted in an influx of vascular markers and an increase in large vessels, while the number of smaller vessels was diminished. In vitro investigations revealed that neuroblastoma cell lines can support endothelial cord formation, most likely due to the production and secretion of VEGF ligands. Furthermore, although prexasertib treatment blocked new cord formation, it was unable to alter established cords, suggesting that prexasertib blocks neovascularization in vivo. Therefore, it is possible that the observed increase in the number of large vessels is a consequence of rapidly regressing xenografts, resulting in the same number of mature vessels within a smaller tumor area. Our data indicate that single-agent prexasertib, a potent CHK1 inhibitor, promotes DNA damage leading to tumor cell death and xenograft regression in preclinical models of neuroblastoma. Inhibitors of CHK1 have traditionally been used in combination with SOC, to enhance the genotoxic effects of these chemotherapeutic agents (5, 41, 43). However, these data add to the growing body of literature that indicates that CHK1 inhibitors also have activity as a single agent (5, 10, 44, 45). In this study, cotreatment with doxorubicin did not enhance the efficacy of prexasertib nor prevent tumor regrowth in two neuroblastoma mouse models. However, combination with doxorubicin in a preclinical model of alveolar rhabdomyosarcoma, a pediatric malignancy driven by a fusion transcription factor, was sufficient to block tumor regrowth and acquired resistance to prexasertib (46). Therefore, efficacious combination of prexasertib with chemotherapeutics may be tumor-type dependent. Selection and validation of different genotoxic therapies as combination partners as well as alterations in dosing schedule (e.g., CHK1 inhibition following administration of SOC) may improve the degree or duration of response in neuroblastoma. In addition, prexasertib efficacy may be enhanced through combination with other targeted agents, such as an inhibitor against WEE1, which has shown synergistic drug interactions with first-generation CHK1 inhibitors in preclinical neuroblastoma models (27). In preliminary studies conducted by our group, WEE1 phosphorylation decreased with prexasertib treatment in a time-dependent manner in KELLY cells, while treatment did not affect WEE1 phosphorylation in prexasertib-resistant PANC-1 cells (data not shown). Further investigation into potential biomarkers of prexasertib sensitivity as well as effective combination therapies to improve efficacy and prevent tumor recurrence in preclinical models of neuroblastoma are warranted.

R.P. Beckmann and A.B. Lin hold ownership interest (including patents) in Eli Lilly and Company. No potential conflicts of interest were disclosed by the other authors.

Conception and design: C.D. Lowery, M. Dowless, J. Stephens, R.P. Beckmann, L.F. Stancato

Development of methodology: A.B. VanWye, J. Stewart, L.F. Stancato

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): C.D. Lowery, M. Dowless, W. Blosser, B.L. Falcon, J. Stewart, J. Stephens

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): C.D. Lowery, A.B. VanWye, M. Dowless, B.L. Falcon, J. Stewart, J. Stephens, A.B. Lin, L.F. Stancato

Writing, review, and/or revision of the manuscript: C.D. Lowery, A.B VanWye, M. Dowless, B.L. Falcon, J. Stephens, R.P. Beckmann, A.B. Lin, L.F. Stancato

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C.D. Lowery, W. Blosser

Study supervision: L.F. Stancato

The authors would like to thank Dr. Sean Buchanan and Dr. Yue Webster for their assistance with the Cancer Cell Sensitivity Profile database.

This study was funded by Eli Lilly and Company, Lilly Corporate Center.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.