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

Abstract

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.

Abbreviations

• BRAID
• (Bivariate Response to Additive Interacting Doses) model
• CFA
• colony forming assay
• DDR
• DNA damage response
• ETO
• etoposide
• EWS
• ewing sarcoma
• IRN
• irinotecan
• OLA
• olaparib
• PARPis
• poly (ADP-ribose) polymerase inhibitors
• RT
• radiation therapy
• RUC
• rucaparib
• SOC
• standard-of-care chemotherapies
• TAL
• talazoparib
• TMZ
• temozolomide
• VCR
• vincristine
• VEL
• veliparib

Introduction

Ewing sarcoma (EWS) is a radiosensitive tumor [1]. In North America, definitive radiation therapy (RT) is used for both primary and metastatic tumors that cannot be surgically resected with acceptable morbidity, and RT is added to surgery in cases of incomplete resection. Primary tumor control is required for cure, and the local control rates on modern Children's Oncology Group studies are approximately 90% [2]. However, local control rates vary significantly by tumor location, with unresectable pelvic and extremity tumors faring the worst, with failure rates approaching 30% [2]. Primary tumor size may also play a prognostic role [3, 4]. Therapeutic approaches to improve outcomes in these high-risk patients have included RT dose escalation [5] or combination approaches with cytotoxic chemotherapy [6]. However, few studies have systematically compared the potentiation of radiation when administered concomitantly with standard-of-care (SOC) chemotherapeutics such as vincristine, cyclophosphamide, and etoposide [7] to that of radiation administered with emerging or repurposed radiosensitizing agents.

Ewing sarcoma is characterized by a reciprocal translocation involving breakpoints in the EWSR1 gene and in genes of the ETS family of transcription factors [8]. This translocation initiates a positive feedback loop that maintains the expression of poly(ADP-ribose) polymerase 1 (PARP1) [9]. We and others have shown that EWS has a defect in the DNA damage response (DDR) that can be exploited therapeutically in model systems by using PARP inhibitors (PARPis) [9–11]. The anti-tumor activity of PARPis can be attributed to at least two distinct mechanisms: (a) the catalytic inhibition of PARP-mediated poly-ADP-ribosylation (parylation), which hinders single-strand DNA break repair and induces synthetic lethality in tumors with impaired homologous recombination (HR), such as those with mutations in BRCA1 or BRCA2 [12, 13]; and (b) the trapping of PARP protein on DNA to generate complexes that interfere with replication and transcription in a manner analogous to the interference induced by topoisomerase I inhibitors [14]. Unlike the inhibition of parylation, trapping is reduced when PARP protein is depleted. Trapping potential varies greatly among clinically advanced PARPis, being high in talazoparib (TAL), intermediate in olaparib (OLA), and low in veliparib (VEL) [14, 15]. In xenograft models of BRCA-deficient tumors, single-agent PARPis with different trapping potentials showed similar levels of efficacy [16].

Although earlier work showed that OLA potentiated the cytotoxicity of radiation in EWS [17], the benefit of PARPis, as compared to SOC drugs, and the role of PARP trapping in the context of radiosensitization have not been explored. Here, we investigated SOC drugs for treating primary and recurrent EWS and several clinically advanced PARPis, both as single agents and in combination with radiation, in human EWS cell lines. The molecular mechanisms of the potentiation of radiation by each drug were investigated through quantitative assessments of DNA damage and repair. Drug combinations were tested in vivo by using orthotopic xenograft models of EWS treated with image-guided fractionated radiation, and tumor burden was monitored via bioluminescence. Finally, we report results from patients who received TAL + IRN + RT on the clinical trial BMNIRN (NCT02392793) to support the broader clinical evaluation of this drug-RT combination strategy as an effective salvage therapy for EWS.

Material and Methods

Cell Culture, Drugs, and Antibodies

Details regarding the cell models, shRNA lentiviral particles, and other reagents used in this study are reported in Table S1. All cell lines were authenticated using short tandem repeat analysis and tested for mycoplasma. Drugs were purchased from commercial sources or obtained from the SJCRH clinical pharmacy. The purity of all compounds was confirmed to be ≥ 95% by using liquid chromatography–mass spectrometry (LC/MS) coupled with ultraviolet total-wavelength chromatogram evaporative light-scattering (UVTWC/ELSD) detection, and concentrations were verified by chemiluminescent nitrogen detection (CLND) if nitrogen was present in the compound.

Cell Viability and Colony Forming Assays (CFAs)

Cell viability was assessed using CellTiter-Glo, with DMSO acting as the negative control and 10 μM staurosporine acting as the positive control. For CFA experiments, adherent cells were plated overnight and treated with the drug (if applicable) 1–3 h prior to radiation. Radiation was administered using orthovoltage x-rays. Colonies were fixed with 4% paraformaldehyde, stained with 0.05% crystal violet, and scored 8–14 days after treatment using a GelCount colony counter (Oxford Optronix). Colonies were defined as groups of ≥ 50 cells. The surviving fraction (SF) was normalized to the plating efficiency of each cell line [18]. Cell survival data was fitted to the linear quadratic model [19] or the BRAID (Bivariate Response to Additive Interacting Doses) model [20]. BRAID kappa (κ) measures the presence and magnitude of synergy or antagonism: κ < 0 indicates antagonism, κ = 0 implies additivity, and κ > 0 reflects synergy. The BRAID index of achievable efficacies (IAE) measures the degree to which drug combinations in a concentration range achieve a desired level of efficacy (≤ 1 μM and 50% efficacy in this study). When comparing drug combinations, the one with the highest IAE is more efficacious. Two radiosensitization metrics, the radiation enhancement factor (REF, calculated at 2 Gy) and the dose modulation factor (DMF, calculated at 10% survival), were assessed [21] at the highest drug dose that induced little or no cytotoxicity on its own. The timeline for the CFA experiments with PARP1 shRNA cell lines was as follows: (1) induce with doxycycline on day 1; (2) plate cells for the CFA and re-dose with doxycycline on day 4; (3) add drug and irradiate on day 5; (4) re-dose with doxycycline on days 7 and 10; and (5) fix and stain the cells on day 13.

DNA Damage Assessment

To assess γ-H2A.X foci, cells were incubated for 1 h or 24 h after drug or radiation treatment, then fixed with 4% PFA, stained with γ-H2A.X antibody, counterstained with Hoechst, and then analyzed using a Lionheart-FX high-content microscope (BioTek). Gen5 software (BioTek) was used to quantify the number and intensity of the γ-H2A.X foci. The DNA repair efficiency assay was conducted as previously described [22] and quantified with the Lionheart-FX and Gen5 software. The percentage of red or green cells relative to the total number of cells for each condition was determined, and the relative HR and mNHEJ activity was calculated by normalization to the DMSO-treated control.

Mice and Orthotopic Xenograft Models

Athymic nude immunodeficient mice were purchased from Charles River Laboratories (strain code 553). ES8 orthotopic xenografts were created by injecting luciferase-labeled cells into athymic nude mice as described previously [11]. Mice were screened weekly by Xenogen imaging, and the bioluminescence was measured. Mice were enrolled in the study after a target bioluminescence signal of 10⁷ photons/s/cm² or a palpable tumor was obtained, and chemotherapy was started on the following Monday. Mice received four courses of chemotherapy (3 weeks per course), and bioluminescence was monitored weekly and at the end of therapy. For Xenogen assessment, mice were injected intraperitoneally with firefly D-luciferin (3 mg/mouse). Bioluminescent images were acquired 5 min later with a Xenogen IVIS 200 imaging system. The identical ROI was used to determine the average radiance (in photons/s/cm²/sr) for all xenografts. Responses were defined by theXenogen signal as follows: complete response (≤ 10⁵ photons/s/cm², similar to background); partial response (10⁵–10⁷ photons/s/cm²); stable disease (10⁷–10⁸ photons/s/cm², similar to the enrollment signal); and progressive disease (> 10⁸ photons/s/cm²). Mice whose tumor burden exceeded 20% of their body weight at any time were also classified as having progressive disease.

In Vivo Radiation Dose–Response and Drug–Radiation Survival Studies

Image-guided fractionated radiation was administered daily, Monday–Friday, using a Small Animal Radiation Research Platform (SARRP, Xstrahl Inc.) as described previously [23]. The radiation dose response was evaluated with a single fraction, five fractions (10 Gy total), or 10 fractions (20 Gy total) of 2 Gy, and mice were followed until moribund, whereas drug–radiation survival studies used 30 daily fractions of 0.5 Gy (15 Gy total).

In Vivo Pharmacodynamics of DNA Damage and Cell Death

Nude mice were injected with 1 × 10⁶ ES-8 cells intrafemorally as previously described. Following engraftment, mice were monitored weekly until a medium-sized tumor was formed. Mice were then randomized to the following treatment groups and treated for 5 days of therapy: control, TAL + IRN + TMZ, or RT + TAL + IRN. The following doses and schedules were used for each drug: radiation (RT, 0.5 Gy once daily), talazoparib (TAL, 0.1 mg/kg oral gavage twice daily), irinotecan (IRN, 1.25 mg/kg intraperitoneal once daily), and temozolomide (TMZ, 10 mg/kg oral gavage once daily). Following treatment, the mice were euthanized, and the tumors were harvested. Tissues were fixed in 10% neutral-buffered formalin and processed as paraffin-embedded samples. Tissues were sectioned at a 5-μm thickness and mounted onto positively charged glass slides (Superfrost Plus; ThermoFisher Scientific, Waltham, MA) and then HE stained and coverslipped using the HistoCore SPECTRA Workstain (Lecia Biosystems) or stained for markers of cellular proliferation. A ZEISS AxioScan microscope slide scanner (Zeiss) was used to create whole slide images to a 20× scalable magnification. The HALO v3.6.4134.137 software program (Indica Labs) was used to make images, quantify the area of immunolabeling for each marker, and generate the percentages of cells that were immunolabeled for the markers of interest (Multiplex IHC macro v3.4). Isotype controls were used to confirm the specificity and sensitivity of the staining for γ-H2A.X and Ki67. Additional details regarding the methodology for the immunohistochemical and histochemical assays are reported in Table S1.

Statistical Analysis

GraphPad Prism (GraphPad Software) was used for all other statistical analyses. Data is presented as the mean ± standard deviation unless otherwise indicated. Pairwise comparisons were analyzed using Student's t-test. Kaplan–Meier survival curves were compared using the Mantel–Cox log-rank test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; n.s., not significant.

Results

PARPis Outperform SOC Drugs as Radiosensitizers in EWS Cells

To compare the potentiation of radiation when combined with single-agent SOC agents or PARPis, we performed CFAs in ES8, EW8, and A673 cells by testing both the drug and radiation in dose–response (Figure 1A–C; Figure S1A). Cell survival data was fitted to (a) the linear quadratic model [19], which was then used to calculate radiobiologic metrics (REF_2Gy, and DMF₁₀), and (b) the BRAID response surface model [20] to obtain the κ and IAE parameters. BRAID models the entire two-dimensional response surface that results from testing both drug and radiation in dose–response and provides a global interpretation of the combination, whereas the radiobiologic metrics examine the drug–radiation interaction at specific points on the response surface. To improve the accuracy of the radiobiologic metrics, we assessed them at the highest dose possible for each drug that, by itself, induced only limited cytotoxicity. We observed statistically significant correlations between κ and REF_2Gy and DMF₁₀, with Spearman correlations of 0.68 and 0.71, respectively (Figure S1B).

[IMAGE OMITTED. SEE PDF]

PARPis as a class generally showed stronger potentiation with radiation (radiobiologic metrics > 1.1 and κ > 0) when compared to SOC drugs (Figure 1C). In contrast, when assessing the total cytotoxic effect of the combination by using the IAE, SN-38 (the active metabolite of IRN), VCR, and the strong PARP trapper TAL were the three most effective agents in combination with radiation. We found no statistically significant correlation between the IAE and the radiobiologic metrics, respectively (Figure S1C), indicating that the most synergistic combinations were not the most efficacious. We confirmed synergy in EW8 and A673 cells between radiation and TAL (κ > 0), but not VCR, ETO (etoposide), or SN-38 (Figure 1C). In these two cell lines, VCR and SN-38 outperformed TAL with respect to combined efficacy with radiation. Taken together, these results show that PARPi treatment led to more robust potentiation of radiation than that was observed with conventional chemotherapy in EWS cells. Additionally, SN-38 and VCR did not strongly potentiate radiation but were independently cytotoxic and demonstrated the highest efficacy when combined with radiation in vitro.

PARPis Induce DNA Damage but Do Not Impair ds-DNA Repair

To interrogate the effect of PARPis on ds-DNA repair, we evaluated the impact of drug treatment using a U2OS cell line engineered to express the inducible DR-GFP HR and EJ-RFP mNHEJ reporter systems [22]. To validate the assay, we treated the cells with AZD0156 and KU60648, potent and selective inhibitors of the ATM and DNA-PKcs, respectively. As expected, AZD0156 inhibited HR and increased mNHEJ repair, while KU60648 enhanced both mNHEJ repair and HR [24–26] (Figure 2A; Figure S2A). We then assayed three clinically advanced PARPis (TAL, OLA, and VEL) at three different concentrations and found that none substantially altered the level of HR or mNHEJ.

[IMAGE OMITTED. SEE PDF]

Next, we evaluated the extent and kinetics of DNA damage in ES8 cells after treatment with SN-38 and three PARPis that induced a variable response in the CFA (TAL, OLA, VEL), using the γH2A.X foci count as a marker for ds-DNA breaks (Figure 2B,C; Figure S2B) [27]. TMZ was included as a negative control and, consistent with its function as an alkylating agent, γH2A.X foci were largely undetectable in TMZ-treated cells. SN-38 induced the strongest and most rapid DNA damage—comparable to that obtained with 2 Gy—after only 1 h of exposure to a 1.4 nM concentration of drug. However, the DNA damage caused by PARPis took longer to evolve and was correlated with PARP trapping potential at 24 h [14, 15]. OLA and VEL required 10–100-fold and > 1000-fold higher concentrations, respectively, relative to TAL to achieve the same levels of DNA damage. Finally, we observed that the rank order of compounds by potency at 24 h in this assay (SN-38 ≈ TAL>OLA>VEL>TMZ) closely matched the combined efficacy as assessed by IAE in the CFA.

TAL Cytotoxicity Alone and in Combination With Radiation Is Mediated by PARP Trapping

To further delineate mechanisms of PARPi cytotoxicity and distinguish between enzymatic inhibition and PARP trapping, we used lentiviral transduction to engineer ES8, EW8, and A673 cells to express a doxycycline-inducible shRNA against PARP1 (Figure 3A; Figure S3A). PARP1 depletion did not increase the sensitivity of EWS cells to radiation (Figure 3B; Figure S3B). However, a reduction in PARP1 expression significantly reduced sensitivity to single-agent TAL (Figure 3C) and decreased the radiation potentiation (DMF₁₀) and combined efficacy (IAE) induced by TAL + RT (Figure 3D–F; Figure S3C). To further interrogate this mechanism, we treated wild-type ES8 cells with either TAL alone or TAL combined with the weaker PARP trapper VEL in the CFA (Figure 3G,H). Adding 1 μM VEL rescued EWS cells compared to those treated with 1 nM TAL alone. We also treated wild-type ES8 cells with either TAL or SN-38 alone or combined with weaker PARP trappers OLA or VEL and monitored cell viability with the CellTiter-Glo assay (Figure 3I). Combining TAL with 100 nM OLA or 10 μM VEL reduced toxicity by a factor of > 10 compared to TAL alone. In contrast, adding 100 nM OLA or 10 μM VEL to SN-38 increased cytotoxicity relative to that observed with SN-38 alone. Collectively, these observations suggest that the cytotoxicity and potentiation of radiation induced by TAL is driven primarily through PARP trapping as opposed to catalytic inhibition.

[IMAGE OMITTED. SEE PDF]

TAL + IRN + RT Shows Significant Efficacy In Vivo

We next sought to explore the in vivo efficacy of combination therapy with radiation in ES8 orthotopic xenografts. We first conducted a pilot in vivo radiation dose–response experiment to define an appropriate dose and schedule for RT. Xenografts were treated with 1, 5, or 10 daily fractions of 2 Gy of radiation, using image-guided focal RT, over a 2-week period (Figure 4A). A subset of mice treated to a cumulative dose of 20 Gy, but none treated to 10 Gy, showed long-term survival (Figure 4B). To induce sufficient DNA damage to detect potentiation by a drug while not curing the mice with RT alone and to enable protracted drug + RT regimens to facilitate clinically relevant drug doses and schedules, ES8 xenograft models were subsequently treated with 0.5 Gy/day for a cumulative dose of 15 Gy.

[IMAGE OMITTED. SEE PDF]

We evaluated the efficacy of RT alone; IRN, TAL, VCR, ETO, or TMZ alone; and the combination of each drug with RT (Figure 4C–E; Table 1; Figure S4A–C; and Table S2A). Chemotherapy was administered using clinically relevant doses and schedules (Table S3). As expected, RT alone increased median survival only marginally (to 12 days vs. 10 days for control mice) and did not significantly improve overall survival (OS) (p = 0.22), although one of the 11 mice in the RT group experienced a complete response (CR) as assessed by bioluminescence (Figure 4F,G). Among the single-drug combination therapies evaluated with RT, IRN + RT achieved the best therapeutic efficacy. Median survival for IRN + RT was 66 days, with two of eight mice experiencing a CR, and OS increased significantly when compared to that observed with RT alone (p = 0.02). One of six mice treated with IRN alone experienced a partial response (PR), and although the median survival for this cohort was 26 days, the improvement in OS did not reach statistical significance when compared to the OS with RT alone (p = 0.08).

TABLE 1 Summary of the in vivo efficacy studies.

Treatment	Median survival (days)	Response	p (vs. RT only)
TAL + IRN + RT	84	3 PR, 5 CR	0.0002
TAL + IRN + TMZ	84	1 PD, 2 PR, 2 CR	0.0049
IRN + RT	66	6 PD, 2 CR	0.024
TAL + RT	42	5 PD, 1 SD	0.36
TAL + TMZ + RT	31	100% PD	0.14
ETO + RT	29	5 PD, 1 PR	0.22
TMZ + RT	26	5 PD, 1 PR	0.0467
IRN	26	5 PD, 1 PR	0.08
TAL	23	100% PD	0.27
TMZ	15	100% PD	0.53
VCR	15	100% PD	0.37
VCR + RT	15	100% PD	0.65
ETO	15	4 PD, 1 CR	0.29
RT	12	10 PD, 1 CR	NA
Control	10	100% PD	NA

TAL + RT was the second most efficacious combination in terms of increased median survival (42 days), although the improvement in OS was not statistically significant compared to RT alone (p = 0.36), and five of six mice showed progressive disease (PD). The combination of VCR or ETO with RT did not improve OS significantly when compared to RT alone (p = 0.65 and p = 0.22, respectively). Median survival was 15 days for VCR + RT and 29 days for ETO + RT. Adding TMZ to RT significantly improved OS (p = 0.047), but median survival was only 26 days.

In previous work, we showed that adding TMZ to the combination of TAL + IRN resulted in prolonged objective responses and significantly improved survival of orthotopic models of EWS [11]. TAL + IRN + TMZ triplet therapy was recently investigated in pediatric patients with recurrent solid tumors (BMNIRN, NCT02392793) [28]. The combination was reported to be feasible and yielded promising responses in several patients with EWS, albeit with a narrow therapeutic window that necessitated reducing the IRN and TMZ doses to 80% and 10% of their standard doses, respectively. Importantly, patients were able to receive focal RT and/or surgery for symptomatic management after the initial 2 cycles of systemic therapy on this trial. Motivated by these earlier findings and by our single-agent in vivo results, we hypothesized that TAL + IRN + RT would be as effective as TAL + IRN + TMZ triplet therapy but more tolerable, given the ability to selectively target RT. To understand the importance of IRN as a driver of efficacy in this combination therapy, we also evaluated TAL + TMZ + RT.

Following the same dose and schedule for IRN, TAL, and TMZ used for combination studies reported in our previous work [11] (Figure 5A), we found that all eight mice treated with TAL + IRN + RT survived, all showed evidence of decreased tumor burden, and five experienced a CR by the end of the study (84 days) (Figure 5B–D; Table 1; and Table S2B). We observed one PD, two PRs, and two CRs in the TAL + IRN + TMZ cohort, and the median survival was at least 84 days. In our earlier work, TAL + IRN alone also resulted in a median survival of at least 84 days [11]. However, outcomes were worse, with 53% of mice maintaining or increasing tumor burden after treatment (six CR, two PR, one SD, and 8 PD).

[IMAGE OMITTED. SEE PDF]

Overall survival in the TAL + IRN + RT and TAL + IRN + TMZ cohorts was superior to that in the RT-alone group (p = 0.0002 and p = 0.0049, respectively). Remarkably, all five mice treated with TAL + TMZ + RT had PD, and median survival was only 31 days. Both the TAL + IRN + RT and TAL + IRN + TMZ treatment arms showed statistically significant improvement in OS when compared to the TAL + TMZ + RT arm (p = 0.0002 and p = 0.002, respectively), underscoring the importance of the topoisomerase I inhibitor IRN as a crucial component of this combination strategy with radiation. Pharmacodynamic assessment using immunohistochemistry confirmed that TAL + IRN + RT treatment induced significantly higher DNA damage as assessed by γH2A.X staining compared to control and TAL + IRN + TMZ treated mice; and that both TAL + IRN + RT and TAL + IRN + TMZ significantly reduced tumor cell proliferation as assessed by Ki67 staining compared to controls (Figure S5A–E).

TAL + IRN + RT May Be a Useful Salvage Strategy

Intriguingly, while response data were censored at the time of radiation and/or surgical resection as part of the response evaluation, marked responses were observed in patients who received TAL + IRN + RT, with or without TMZ, on the BMNIRN clinical trial [28] (Figure 5E), supporting, at least in part, our in vivo findings. Of the 41 pediatric patients with refractory/recurrent solid tumors enrolled in this trial, 6 patients received focal RT for symptomatic management after completing the first response assessment following course 2 of chemotherapy. Two patients were treated with TAL + IRN + RT, while four received TAL + IRN + TMZ + RT; all six patients had progressive Ewing or Ewing-like sarcoma at the time of study enrollment. Radiotherapy consisted of conventionally fractionated external beam RT to 54–55.8 Gy or stereotactic body radiotherapy (SBRT) to 35–40 Gy over 5–8 fractions. Of the three evaluable patients, post hoc analysis of radiotherapy site-specific responses, per protocol-defined radiographic response criteria using RECIST version 1.1, included 2 partial responses and 1 stable disease. Additionally, no new or worsening treatment-related adverse events attributed to RT were noted with a median duration of follow-up of 121 days (range, 24–191 days) following RT.

Discussion

Radiation therapy plays an essential role in the treatment of EWS in patients with incompletely resected, metastatic, or recurrent disease. Although dose-escalated RT may improve local tumor control in EWS, the risk of most radiation-associated toxicities has been correlated with increasing RT dose [29–33]. Therefore, finding agents that better potentiate RT-induced cytotoxicity selectively in tumor cells is an attractive option to improve local control and limit RT-mediated toxicity.

Work by others demonstrated the utility of combining OLA with RT in EWS [17], but the importance of PARP trapping vs. catalytic inhibition and the benefit of this approach as compared to SOC drugs were not explored. Here, we have shown that in vitro, PARPis are better potentiators of radiation than SOC agents. Moreover, the combined efficacy of PARPis and RT in vitro varied by orders of magnitude. Using assays to assess DNA damage and repair, we found that PARPis do not affect the levels of HR or mNHEJ but instead induce DNA damage and cytotoxicity on their own in EWS cells as a function of their PARP-trapping potential. Knockdown of PARP1 protein or competition with weaker PARP-trapping agents reduced the cytotoxicity of TAL as a single agent or in combination with RT, suggesting that catalytic inhibition plays, at most, a minor role in sensitizing EWS cells to radiation. Our observation that efficacy was correlated with stronger PARP-trapping activity may have clinical implications when nominating a potential PARPi for use in combination with RT.

In vitro, the combinations of IRN, TAL, or VCR with RT showed the highest levels of combined efficacy in EWS cells, although only TAL exhibited synergy. However, IRN + RT proved to be the most effective in vivo, as assessed by median survival, followed by the others in the order TAL>ETO>TMZ>VCR. The lack of activity with the VCR + RT combination could be attributed to poor tumor exposure, as the drug was given only 1 day each week in a 3-week cycle. In contrast, both TAL and IRN were given daily for the first 2 weeks in a 3-week cycle. Previous work indicated that TAL and IRN would result in similar tumor exposures with the doses and schedules employed here [11]. However, we showed that SN-38, the active metabolite of IRN, could induce significant DNA damage at low concentrations after only 1 h of drug exposure, whereas more modest levels were observed with PARPis after 24 h of exposure. The greater magnitude and faster kinetics of DNA damage induced by IRN, as compared with TAL, might explain why IRN + RT was more efficacious in vivo.

Combining the top two best-performing drugs from our single-agent study, TAL and IRN, with RT resulted in 100% survival in ES8 orthotopic xenograft models—a benefit comparable to that reported earlier for the triplet combination of TAL, IRN, and TMZ [11]. This prior observation formed the rationale for the BMNIRN clinical trial that evaluated the safety and tolerability of IRN and TAL, with or without added TMZ. Results from this trial suggest that the combination of IRN and TAL was feasible and yielded responses in several patients with EWS [28]. Additionally, responses have been observed in patients necessitating RT concurrently with IRN and TAL, with or without TMZ, and without concerning additional safety signals as part of this trial, as we demonstrate here. Overall, our work suggests that TAL + IRN + RT may be a useful therapeutic strategy, particularly when RT is indicated in the progressive disease setting.

Author Contributions

Jia Xie: investigation, methodology, writing – original draft. Marcia M. Mellado-Lagarde: investigation, methodology. Kaley Blankenship: investigation. Debolina Ganguly: investigation. Nathaniel R. Twarog: investigation, visualization. Brandon Bianski: investigation. Matthew Kieffer: investigation. Stefan Atkinson: investigation. Heather Sheppard: investigation. Jessica Gartrell: investigation. Samuel Cler: investigation. Sara M. Federico: investigation. Elizabeth A. Stewart: conceptualization, funding acquisition, methodology, project administration, supervision. Christopher L. Tinkle: conceptualization, methodology, project administration, supervision, writing – original draft, writing – review and editing. Anang A. Shelat: conceptualization, funding acquisition, methodology, project administration, supervision, writing – original draft, writing – review and editing.

Acknowledgments

We thank Dr. Richard Ashmun and the Flow Cytometry and Cell Sorting Shared Resource (SJCRH) for expert guidance; Lauren Hoffmann (SJCRH) for help with the preclinical studies; Dr. Keith A. Laycock (SJCRH) for scientific editing of the manuscript; and the Center for In Vivo Therapeutics (SJCRH) for preclinical imaging. We would like to acknowledge the contributions of Nancy E. Martinez, who sadly passed away during this research.

Disclosure

The authors have nothing to report.

Ethics Statement

Animal Studies: Approved by the Institutional Animal Care and Use Committee (IACUC) at SJCRH.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

References

J. Ewing, “Classics in Oncology. Diffuse Endothelioma of Bone. James Ewing. Proceedings of the New York Pathological Society, 1921,” CA: A Cancer Journal for Clinicians 22, no. 2 (1972): 95–98.

S. K. Ahmed, R. L. Randall, S. G. DuBois, et al., “Identification of Patients With Localized Ewing Sarcoma at Higher Risk for Local Failure: A Report From the Children's Oncology Group,” International Journal of Radiation Oncology, Biology, Physics 99, no. 5 (2017): 1286–1294.

Y. Arai, L. E. Kun, M. T. Brooks, et al., “Ewing's Sarcoma: Local Tumor Control and Patterns of Failure Following Limited‐Volume Radiation Therapy,” International Journal of Radiation Oncology, Biology, Physics 21, no. 6 (1991): 1501–1508.

C. Rodriguez‐Galindo, F. Navid, T. Liu, C. A. Billups, B. N. Rao, and M. J. Krasin, “Prognostic Factors for Local and Distant Control in Ewing Sarcoma Family of Tumors,” Annals of Oncology 19, no. 4 (2008): 814–820.

A. C. Talleur, F. Navid, S. L. Spunt, et al., “Limited Margin Radiation Therapy for Children and Young Adults With Ewing Sarcoma Achieves High Rates of Local Tumor Control,” International Journal of Radiation Oncology, Biology, Physics 96, no. 1 (2016): 119–126, https://doi.org/10.1016/j.ijrobp.2016.04.001.

H. E. Grier, M. D. Krailo, N. J. Tarbell, et al., “Addition of Ifosfamide and Etoposide to Standard Chemotherapy for Ewing's Sarcoma and Primitive Neuroectodermal Tumor of Bone,” New England Journal of Medicine 348, no. 8 (2003): 694–701, https://doi.org/10.1056/NEJMoa020890.

S. Durer, D. P. Gasalberti, and H. Shaikh, Ewing Sarcoma (StatPearls, 2024).

S. A. Burchill, “Ewing's Sarcoma: Diagnostic, Prognostic, and Therapeutic Implications of Molecular Abnormalities,” Journal of Clinical Pathology 56, no. 2 (2003): 96–102.

J. C. Brenner, F. Y. Feng, S. Han, et al., “PARP‐1 Inhibition as a Targeted Strategy to Treat Ewing's Sarcoma,” Cancer Research 72, no. 7 (2012): 1608–1613.

M. J. Garnett, E. J. Edelman, S. J. Heidorn, et al., “Systematic Identification of Genomic Markers of Drug Sensitivity in Cancer Cells,” Nature 483, no. 7391 (2012): 570–575, https://doi.org/10.1038/nature11005.

E. Stewart, R. Goshorn, C. Bradley, et al., “Targeting the DNA Repair Pathway in Ewing Sarcoma,” Cell Reports 9, no. 3 (2014): 829–841, https://doi.org/10.1016/j.celrep.2014.09.028.

H. E. Bryant, N. Schultz, H. D. Thomas, et al., “Specific Killing of BRCA2‐Deficient Tumours With Inhibitors of Poly(ADP‐Ribose) Polymerase,” Nature 434, no. 7035 (2005): 913–917, https://doi.org/10.1038/nature03443.

H. Farmer, N. McCabe, C. J. Lord, et al., “Targeting the DNA Repair Defect in BRCA Mutant Cells as a Therapeutic Strategy,” Nature 434, no. 7035 (2005): 917–921, https://doi.org/10.1038/nature03445.

J. Murai, S. Y. Huang, B. B. Das, et al., “Trapping of PARP1 and PARP2 by Clinical PARP Inhibitors,” Cancer Research 72, no. 21 (2012): 5588–5599, https://doi.org/10.1158/0008‐5472.CAN‐12‐2753.

J. Murai, S. Y. Huang, A. Renaud, et al., “Stereospecific PARP Trapping by BMN 673 and Comparison With Olaparib and Rucaparib,” Molecular Cancer Therapeutics 13, no. 2 (2014): 433–443, https://doi.org/10.1158/1535‐7163.MCT‐13‐0803.

T. A. Hopkins, W. B. Ainsworth, P. A. Ellis, et al., “PARP1 Trapping by PARP Inhibitors Drives Cytotoxicity in Both Cancer Cells and Healthy Bone Marrow,” Molecular Cancer Research 17, no. 2 (2019): 409–419, https://doi.org/10.1158/1541‐7786.MCR‐18‐0138.

H. J. Lee, C. Yoon, B. Schmidt, et al., “Combining PARP‐1 Inhibition and Radiation in Ewing Sarcoma Results in Lethal DNA Damage,” Molecular Cancer Therapeutics 12, no. 11 (2013): 2591–2600.

R. A. El‐Awady, E. Dikomey, and J. Dahm‐Daphi, “Radiosensitivity of Human Tumour Cells Is Correlated With the Induction but Not With the Repair of DNA Double‐Strand Breaks,” British Journal of Cancer 89, no. 3 (2003): 593–601.

S. J. McMahon, “The Linear Quadratic Model: Usage, Interpretation and Challenges,” Physics in Medicine and Biology 64, no. 1 (2018): 1TR.

N. R. Twarog, E. Stewart, C. V. Hammill, and A. A. Shelat, “BRAID: A Unifying Paradigm for the Analysis of Combined Drug Action,” Scientific Reports 6 (2016): 25523.

Q. Liu, M. Wang, A. M. Kern, et al., “Adapting a Drug Screening Platform to Discover Associations of Molecular Targeted Radiosensitizers With Genomic Biomarkers,” Molecular Cancer Research 13, no. 4 (2015): 713–720.

R. S. Bindra, A. G. Goglia, M. Jasin, and S. N. Powell, “Development of an Assay to Measure Mutagenic Non‐Homologous End‐Joining Repair Activity in Mammalian Cells,” Nucleic Acids Research 41, no. 11 (2013): e115.

S. M. C. Smith, B. M. Bianski, B. A. Orr, et al., “Preclinical Modeling of Image‐Guided Craniospinal Irradiation for Very‐High‐Risk Medulloblastoma,” International Journal of Radiation Oncology, Biology, Physics 103, no. 3 (2019): 728–737.

A. Bothmer, D. F. Robbiani, N. Feldhahn, A. Gazumyan, A. Nussenzweig, and M. C. Nussenzweig, “53BP1 Regulates DNA Resection and the Choice Between Classical and Alternative End Joining During Class Switch Recombination,” Journal of Experimental Medicine 207, no. 4 (2010): 855–865.

K. K. Chiruvella, Z. Liang, and T. E. Wilson, “Repair of Double‐Strand Breaks by End Joining,” Cold Spring Harbor Perspectives in Biology 5, no. 5 (2013): a012757.

A. G. Goglia, R. Delsite, A. N. Luz, et al., “Identification of Novel Radiosensitizers in a High‐Throughput, Cell‐Based Screen for DSB Repair Inhibitors,” Molecular Cancer Therapeutics 14, no. 2 (2015): 326–342.

L. J. Mah, A. El‐Osta, and T. C. Karagiannis, “gammaH2AX: A Sensitive Molecular Marker of DNA Damage and Repair,” Leukemia 24, no. 4 (2010): 679–686.

S. M. Federico, A. S. Pappo, N. Sahr, et al., “A Phase I Trial of Talazoparib and Irinotecan With and Without Temozolomide in Children and Young Adults With Recurrent or Refractory Solid Malignancies,” European Journal of Cancer 137 (2020): 204–213.

P. L. Gawade, M. M. Hudson, S. C. Kaste, et al., “A Systematic Review of Selected Musculoskeletal Late Effects in Survivors of Childhood Cancer,” Current Pediatric Reviews 10, no. 4 (2014): 249–262.

R. Goldsby, C. Burke, R. Nagarajan, et al., “Second Solid Malignancies Among Children, Adolescents, and Young Adults Diagnosed With Malignant Bone Tumors After 1976: Follow‐Up of a Children's Oncology Group Cohort,” Cancer 113, no. 9 (2008): 2597–2604.

K. Jentzsch, H. Binder, H. Cramer, et al., “Leg Function After Radiotherapy for Ewing's Sarcoma,” Cancer 47, no. 6 (1981): 1267–1278.

J. F. Kuttesch, Jr., L. H. Wexler, R. B. Marcus, et al., “Second Malignancies After Ewing's Sarcoma: Radiation Dose‐Dependency of Secondary Sarcomas,” Journal of Clinical Oncology 14, no. 10 (1996): 2818–2825.

M. A. Tucker, G. J. D'Angio, J. D. Boice, Jr., et al., “Bone Sarcomas Linked to Radiotherapy and Chemotherapy in Children,” New England Journal of Medicine 317, no. 10 (1987): 588–593.