The Combination of PARP and Topoisomerase 1 Inhibitors Improves Radiation Therapy for Ewing Sarcoma

Abstract

Although primary tumor control rates after surgery and/or radiation therapy (RT) are generally high in patients with Ewing sarcoma (EWS), those with unresectable tumors have failure rates approaching 30% and experience poorer outcomes. Additionally, although metastatic site irradiation is associated with improved survival, dose, and volume effects influence the long-term toxicity risk. Consequently, it is important to identify novel systemic agents to enhance the therapeutic ratio of RT. Given the reported DNA damage response deficits in EWS, we hypothesized that PARP inhibitors (PARPis) would preferentially potentiate radiation relative to standard-of-care (SOC) chemotherapeutics. We investigated primary and recurrent SOC drugs and PARPis with varied trapping potential in combination with radiation in EWS cell lines. At physiologically relevant concentrations, the strong PARP trapper talazoparib (TAL) potentiated radiation to a greater extent than did SOC or other PARPis, although the magnitude of the effect was modest. The radiosensitizing effect of TAL was mediated through the induction of DNA double-strand breaks, rather than through the catalytic inhibition of PARP1. Drug + RT combinations were further tested in vivo by using orthotopic xenograft models of EWS treated with image-guided fractionated radiation. The addition of RT to the combination of TAL plus irinotecan (IRN), a recently evaluated clinical regimen for relapsed pediatric solid tumors, significantly prolonged survival and reduced tumor burden in all EWS-treated mice. This triplet therapy (TAL + IRN + RT) was feasible and yielded responses in several patients with EWS and may represent a useful salvage strategy in recurrent or progressive disease.

Keywords: DNA damage; Ewing sarcoma; PARP inhibitors; radiation therapy; topoisomerase I inhibitors.

FIGURE 1

In vitro potentiation of radiation by SOC and PARPi in EWS cells. (A) Linear quadratic model fits from colony forming assay (CFA) experiments in ES8 cells treated with radiation and vehicle (DMSO) or the indicated compound at the highest non‐cytotoxic concentration. (B) BRAID response surface model fits of the full CFA experiments from (A). (C) Summary of the BRAID parameters κ and IAE calculated for all CFA experiments. n ≥ 2 biological replicates.

FIGURE 2

Mechanistic evaluation of PARPis in EWS cells. (A) Summary of the DNA repair efficiency assay in U2OS‐EJ‐DRs cells and representative images. The intensity of the green and red signals is proportional to the level of HR and mNHEJ repair, respectively, in these cells. n ≥ 3 biological replicates. (B, C) Representative immunofluorescence images (top) and quantification (bottom) of γH2A.X foci formation in ES8 cells after treatment with the indicated compound or with 2 Gy of radiation for 1 h (B) or 24 h (C). Blue: DAPI; red: γH2A.X (images of irradiated cells have been pseudo‐colored to red for clarity). The γH2A.X signal induced by 2 Gy of radiation is plotted for reference purposes.

FIGURE 3

Radiosensitization by TAL is driven by PARP1 trapping. (A) Immunoblot confirmation of PARP1 KD in EWS cells engineered to express shPARP1. Treatment with 1 μg/mL doxycycline reduced PARP1 expression in ES8, A673, and EW8 engineered cells by 57%, 77%, and 75%, respectively. (B) Linear quadratic model fits from CFA experiments in EWS cells with or without PARP1 KD. n ≥ 2 biological replicates. (C) The surviving fraction from CFA experiments of EWS cells with or without stable PARP KD after treatment with TAL for 8–14 days. n ≥ 2 biological replicates. (D–F) The DMF₁₀ (calculated at indicated concentration of TAL) (D), IAE (E), and representative images (F) from CFA experiments in EWS cells with or without PARP1 KD after treatment with TAL and radiation for 8–14 days. (G–H) Representative images (G) and DMF₁₀ results (H) from CFA experiments in ES8 cells treated with TAL and VEL. (I) Dose–response results of CellTiter‐Glo assays of ES8 cells treated for 72 h with TAL or SN‐38 in combination with vehicle (DMSO), VEL, or OLA.

FIGURE 4

In vivo efficacy of drug alone and drug + RT in ES8 orthotopic xenograft models. (A) Treatment schedule for the in vivo RT dose–response experiment using 2 Gy fractions. (B) Kaplan–Meier survival curves for the ES8 xenografts from the experiment in (A). (C) Treatment schedule for the in vivo efficacy studies of drug alone and drug + RT. (D) Line plot of tumor burden over time as measured by bioluminescence for ES8 xenografts treated as indicated. Each line represents a different mouse. (E) Kaplan–Meier survival curves for the ES8 xenografts from the experiments in (D). (F) Summary of the responses at the end of treatment as determined by bioluminescence for all drug‐alone and drug + RT studies. (G) Summary of median survival for all drug‐alone and drug + RT studies.

FIGURE 5

In vivo efficacy of triple combination therapy. (A) Treatment schedule for each combination group. (B) Line plot of tumor burden over time as measured by bioluminescence for ES8 xenografts treated as indicated. Each line represents a different mouse. (C) Kaplan–Meier survival curves for the ES8 xenografts from the experiments in (B). (D) Representative images of mice at enrollment and progression/end of study from triplet combination therapy treatment. (E) CT images in the axial, coronal (cor), and sagittal (sag) dimensions demonstrating a mediastinal mass (red) prior to concurrent TAL + IRN + RT (upper) and 4 months post‐treatment (middle) in a patient with relapsed EWS with a mixed radiographic response to TAL + IRN alone and enlarging mediastinal mass. A 95% reduction in tumor volume was observed following triple therapy. RT consisted of 35 Gy delivered over 5 daily fractions of 7 Gy with stereotactic body radiotherapy (lower; yellow, target volume accounting for respiration; radiation dosimetry displayed in colorwash from 20% to 95% of the prescribed dose with labeled beams).