Elimusertib has Antitumor Activity in Preclinical Patient-Derived Pediatric Solid Tumor Models

Abstract

The small-molecule inhibitor of ataxia telangiectasia and Rad3-related protein (ATR), elimusertib, is currently being tested clinically in various cancer entities in adults and children. Its preclinical antitumor activity in pediatric malignancies, however, is largely unknown. We here assessed the preclinical activity of elimusertib in 38 cell lines and 32 patient-derived xenograft (PDX) models derived from common pediatric solid tumor entities. Detailed in vitro and in vivo molecular characterization of the treated models enabled the evaluation of response biomarkers. Pronounced objective response rates were observed for elimusertib monotherapy in PDX, when treated with a regimen currently used in clinical trials. Strikingly, elimusertib showed stronger antitumor effects than some standard-of-care chemotherapies, particularly in alveolar rhabdomysarcoma PDX. Thus, elimusertib has strong preclinical antitumor activity in pediatric solid tumor models, which may translate to clinically meaningful responses in patients.

Figure 1.

Elimusertib shows antitumor activity in a broad spectrum of pediatric cancer cell lines. A–C, Dose–response curves of the cell viability for ARMS (A), ERMS (A), Ewing sarcoma (B), MNA NB (C), and NMNA NB cell lines (C) treated with the ATR inhibitor elimusertib compared with non-cancer cell lines BJ and RPE (n = 3; 50% inhibitory concentrations, IC₅₀, and area under the curve, AUC, values are listed in Supplementary Table S1). D, AUC corresponding to the graphs in (A–C; unpaired, two-sided Student t test, P = 0.0096, 0.0410, 0.0761, 0.1488, 0.8260 for MNA NB vs. Control, Ewing sarcoma vs. Control, ARMS vs. Control, NMNA_NB vs. Control, ERMS vs. Control, respectively). E, Representative pictures of a colony formation assay in 5,838 cells treated at corresponding IC₅₀ value for 48 hours and cultivated for 7 days. F, Bar plot showing decrease in colony formation upon elimusertib treatment at the corresponding IC₅₀ value for 48 hours and cultivation for 7–10 days (P = 0.0022, 0.0821, 0.4753, 0.0028, 0.0786, 0.0121, 0.0466, 0.0124, 0.1685, 0.2402, respectively; n = 3). G, Representative photomicrographs of micronuclei (white arrow) in cells treated with elimusertib. H, Fraction of micronucleated cells after treatment with elimusertib (20 nmol/L) for 72 hours (P = 0.0242, 0.0014, 0.0033, 0.0002, 0.0108, 0.0065, 0.520, 0.0061, 0.0312, 1.30×10⁻⁵, 0.0072, 0.0008, 0.0014, 0.0026, 0.0088, 0.1448, 0.0013, 0.3740, 0.0030, 0.0042, 0.0008, respectively; n = 3, with 50 cells per replicate). I, Representative gating for TUNEL labeling in 5,838 cells. J, Quantification of TUNEL signal in a set of pediatric solid tumor cell lines treated with elimusertib (20 nmol/L) for 72 hours. (P = 2.08×10⁻⁵, 0.0232, 0.0002, 0.0018, 0.0045, 6.38×10⁻⁷, respectively; n = 3). K, Representative gating for EdU and PI co-staining in 5,838 cells. L, Quantification of the fraction of cells in each cell-cycle phase in a set of pediatric solid tumor cell lines after elimusertib treatment (20 nmol/L) for 72 hours (n = 3; error bars represent standard deviation). M, Table of mutations (including translocations, single-nucleotide variants, and copy-number alterations) affecting genes associated with ATR inhibitor sensitivity in a subset of cell lines tested. *t test resulting in P < 0.05; **t test resulting in P < 0.01.

Figure 2.

Elimusertib treatment induces heterogeneous response in a large cohort of patient-derived xenografts of pediatric solid tumors. A, Schematic representation of the preclinical study in PDX models. A total of 39 PDX models were established from 134 patients. 32 of those PDXs received 40 mg/kg body weight elimusertib twice daily per oral gavage, on a 3 days-on/4 days-off schedule. B, Dot plot showing the relative tumor volume at the end of the treatment for all PDXs treated with elimusertib or vehicle control (n and P values are listed in Supplementary Table S3). C, Dot plot showing the relative tumor volume at the end of the treatment for all tumor entities treated with elimusertib or vehicle control (n and P values are listed in Supplementary Table S3). D, Waterfall plot representing tumor volume change in mice receiving elimusertib. Tumors were classified according to the RECIST criteria (55) as progressive disease (red), stable disease (yellow), partial response (light green), and complete response (dark green). For statistical comparison, an unpaired, two-sided Student t test was performed; error bars represent standard deviation. *P < 0.05; **P < 0.01.

Figure 3.

Elimusertib treatment extends the progression-free survival of preclinical pediatric solid tumor models. A–H, Kaplan–Meier curves showing the progression-free survival, defined as time until the tumor was classified as progressive disease, PD, according to the RECIST criteria, in mice treated with elimusertib (red) or vehicle (black), across tumor types (A, n_total = 195, P < 0.0001), ARMS (B, n_total = 44, P < 0,0001), ERMS (C, n_total = 22, P = 0.0001), Ewing sarcoma (D, n_total = 53, P = < 0.0001), MNA NB (E, n_total = 30, P = 0.0064), NMNA NB (F, n_total = 23, P < 0.0001), osteosarcoma (G, n_total = 18, P = 0.0033) and CIC-DUX sarcoma (H, n_total = 5, P = 0.0389). Log-rank tests were performed for statistical comparison.

Figure 4.

Elimusertib reduces the proliferation rate in PDX models of pediatric solid tumors. A, Exemplary hematoxylin and eosin and Ki-67 stainings of Ewing sarcoma, ARMS, ERMS, MNA NB, and NMNA NB PDXs treated with elimusertib or vehicle control. B–I, changes in the fraction of Ki-67–expressing cells for all PDXs combined (B), PDXs responding to elimusertib as defined per RECIST (OR, C) and PDXs with progressive disease (PD, D), Ewing sarcoma (E), ARMS (F), ERMS (G), MNA NB (H), and NMNA NB (I). (n = 10; paired, two-sided Student t test; error bars represent standard deviation, P = 0.0371, 0.0216, 0.4764, 0.9394, 0.4935, 0.2945, 0.7005, and 0.0933 for all PDXs combined, responding PDXs, PDXs with progressive disease, Ewing sarcoma, ARMS, ERMS, MNA NB, and NMNA NB, respectively); scale bar, 40 μm.

Figure 5.

Elimusertib treatment shows that a progression-free survival benefit in a subset of preclinical pediatric solid tumors models compared with SoC treatment. A, Representation of the tumor volume after elimusertib treatment (top) and response to commonly used chemotherapeutic agents in our cohort of PDX models according to the RECIST criteria in a heat map (bottom, progressive disease, red; stable disease, yellow; partial response, light green; complete response, dark green;). In dark blue, PDX derived from patients that previously received SoC treatment are marked. B–F, Kaplan–Meier curves comparing the response of tumors with elimusertib, vehicle control treatment, or treatment with standard-of-care chemotherapeutic agents for ARMS (B, n_total = 110, P < 0.0001), ERMS (C, n_total = 79, P < 0.0001), Ewing sarcoma (D, n_total = 132, P < 0.0001), MNA NB (E, n_total = 104, P < 0.0001), and NMNA NB (F, n_total = 88, P = 0.0003). Log-rank tests were performed for statistical comparison. Single comparisons between elimusertib/SoC and vehicle treatment can be found in Supplementary Table S7.

Figure 6.

Genomic tumor evolution reveals mutations that are associated with altered response to elimusertib. A–F, Oncoplot showing mutations and CNVs present in PDX models for ARMS (A), ERMS (B), Ewing sarcoma (C), osteosarcoma (D), MNA NB (E), and NMNA NB (F). G, Timeline and chemotherapy treatment of a patient with ERMS and tumor response to elimusertib of the corresponding PDXs. The first PDX was established from a primary tumor. The patient received a cycle of vincristine, actinomycin D, and cyclophosphamide (VAC) complemented with low dosage of doxorubicin. A second line of treatment with irinotecan and temozolomide was added later on. Six months after the first biopsy, a biopsy from a relapsed tumor was used to establish a second PDX, and a new relapse after one month was used for the third PDX. H, Timeline and chemotherapy treatment of a patient with Ewing sarcoma and tumor response to elimusertib of the corresponding PDXs. The first PDX was established from a tumor biopsy used for diagnosis. The patient received a cycle of vincristine, ifosfamide, doxorubicin, and etoposide (VIDE) complemented with low dosage of doxorubicin. Four months after the initial biopsy, a biopsy from a relapsed tumor was used to establish a second PDX.