Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 May 6.
Published in final edited form as: Cell Rep. 2014 Oct 23;9(3):829–841. doi: 10.1016/j.celrep.2014.09.028

Targeting the DNA Repair Pathway in Ewing Sarcoma

Elizabeth Stewart 1, Ross Goshorn 1, Cori Bradley 1, Lyra M Griffiths 1, Claudia Benavente 1, Nathaniel R Twarog 2, Gregory M Miller 2, William Caufield 3, Burgess B Freeman III 3, Armita Bahrami 4, Alberto Pappo 5, Jianrong Wu 6, Amos Loh 1, Åsa Karlström 1, Chris Calabrese 7, Brittney Gordon 7, Lyudmila Tsurkan 2, M Jason Hatfield 2, Philip M Potter 2, Scott Snyder 2, Suresh Thiagarajan 8, Abbas Shirinifard 8, Andras Sablauer 8, Anang A Shelat 2,*, Michael A Dyer 1,9,*
PMCID: PMC4386669  NIHMSID: NIHMS667197  PMID: 25437539

Abstract

Ewing sarcoma (EWS) is a tumor of the bone and soft-tissue that primarily affects adolescents and young adults. With current therapies, 70% of patients with localized disease survive, but patients with metastatic or recurrent disease have a poor outcome. We found that EWS cell lines are defective in DNA break repair and are sensitive to PARP inhibitors (PARPis). PARPi-induced cytotoxicity in EWS cells was 10- to 1,000-fold higher after administration of the DNA-damaging agents irinotecan or temozolomide. We developed an orthotopic EWS mouse model and performed pharmacokinetic and pharmacodynamic studies using 3 different PARPis that are in clinical development for pediatric cancer. Irinotecan administered on a low-dose, protracted schedule previously optimized for pediatric patients was an effective DNA-damaging agent when combined with PARPis; it was also better tolerated than combinations with temozolomide. Combining PARPis with irinotecan and temozolomide gave complete and durable responses in more than 80% of the mice.

Introduction

Ewing sarcoma (EWS) is the second most common bone tumor in children and adolescents; approximately 250 new cases are diagnosed each year in the U.S. (Howlader et al., 2013). Most EWS tumors have a translocation involving the EWS gene on chromosome 22 and the FLI gene on chromosome 11 (Delattre et al., 1992). The EWSFLI translocation is an important driver of tumorigenesis in EWS (Lessnick and Ladanyi, 2012). Patients with recurrent or metastatic disease have a poor outcome (Granowetter et al., 2009; Stahl et al., 2011). Recent clinical trials in relapsed disease have shown that the combination of irinotecan (IRN) and temozolomide (TMZ) is active in EWS, and these drugs are now used in combination with other agents such as temsirolimus (Bagatell et al., 2011).

EWS cell lines are sensitive to the poly-ADP ribose polymerase inhibitor (PARPi) olaparib, and this sensitivity is selective for EWS cell lines (Brenner et al., 2012; Garnett et al., 2012). Olaparib sensitivity depends on the EWS-FLI translocation, suggesting a direct mechanistic connection between PARP inhibition by olaparib and the mechanism of transformation by EWS-FLI (Brenner et al., 2012; Garnett et al., 2012). Brenner et al. also showed that the EWS-FLI1 fusion protein interacts with PARP1 and proposed a positive feedback loop involving both proteins (Brenner et al., 2012). High levels of PARP1 expression are correlated with increased sensitivity to PARPis (Liu et al., 2009; Byers et al., 2012; Pettitt et al, 2013; Bajrami et al, 2014), consistent with a ‘trapping’ mechanism whereby the inhibitor acts as a poison to stabilize a PARP-DNA complex (Murai et al., 2012). Furthermore, EWS cell lines express high levels of Schlafen-11 (SLFN11), a putative DNA/RNA helicase whose expression is positively correlated with increased sensitivity to Topoisomerase I inhibitors (Topo1i) and other DNA-damaging agents, but not protein kinase inhibitors or tubulin poisons (Barretina et al., 2012; Zoppoli et al., 2012). For BRCA-deficient breast cancer and ovarian cancer, PARPis are combined with DNA-damaging agents to potentiate selective tumor-cell killing (Ashworth, 2008; Bryant et al., 2005; Farmer et al., 2005; Fong et al., 2009; Tutt et al., 2005).

In this study, we tested the cytotoxic activity and in vivo efficacy of 3 different PARPis (BMN-673, olaparib, veliparib) in combination with IRN and TMZ. Both TMZ and IRN potentiated PARPi-mediated killing of EWS cells, but at least a 1,000-fold higher concentration of TMZ was required to achieve the same level of potentiation as that achieved with IRN. We performed in vivo plasma and tumor pharmacokinetic (PK) experiments in parallel with pharmacodynamics (PD) studies for each PARPi to determine the murine-equivalent dose (MED) and whether sufficient levels of drug can be reached in the tumor to mediate tumor-cell killing. The data were used to design preclinical phase I, II, and III studies to test 15 drug combinations incorporating all 3 PARPis with IRN, TMZ, or both.

RESULTS

EWS Cells Are Defective in DNA-Break Repair

An analysis of the Cancer Cell Line Encyclopedia indicates that EWS behaves as an outlier and shows high levels of both PARP1 and SLNF11 compared to other cancers, including osteosarcoma (OS), another form of bone cancer (Barretina et al., 2012). Since high levels of SLFN11 expression can increase sensitivity to DNA-damaging agents (Zoppoli et al., 2012), we reasoned that EWS might be deficient in DNA-damage repair. We performed quantitative PCR using TaqMan probes for 46 DNA-repair genes and confirmed the downregulation of BRCA1, GEN1, and ATM; in addition, several other genes in EWS cell lines were downregulated relative to their levels in OS cell lines and OS orthotopic xenografts (Figures S1). Gene expression array data from primary EWS and OS tumors was consistent with our QPCR data, and confirmed higher levels of PARP1 and SLFN11 in EWS (Table S1).

To monitor DNA damage in individual cells, we performed a single-cell alkali gel electrophoresis (comet) assay using an OS cell line (U2OS) and ES-8 EWS cells (Figure 1A). Thirty minutes after 10 Gy IR exposure, the tail moment significantly increased for both cells (Figures 1B,C). However, 11 hours after IR exposure, the tail moment in the OS line was restored to basal levels, but that ES-8 EWS was not (Figures 1B,C). To analyze the contribution of PARP, we performed a similar experiment in the presence of olaparib, veliparib, and BMN-673 at 2 concentrations (1 or 10 μM). At 10-μM olaparib and 1- or 10-μM BMN-673, the basal level of DNA damage after 12 hours of exposure to the PARPi was elevated in ES-8, especially for BMN-673 (Figure 1D-G). This DNA-repair defect was even more pronounced when the cells were exposed to 10-Gy IR (Figures 1D-G).

Figure 1. EWS cell lines are defective in dsDNA-break repair.

Figure 1

Open in a new tab

(A) Representative image of single-cell alkali electrophoresis with genomic DNA is shown in red; the tail moment is indicated. Comet data for U2OS cells (B) or ES-8 cells (C) prior to exposure to 10-Gy IR (untreated), 30 minutes after treatment, or 11 hours after treatment. The red line indicates the mean. (D-G) Comet data for cells treated with 1 or 10 μM PARPi before and 12 hours after exposure to 10-Gy IR. (H) Micrographs of U2OS and ES-8 nuclei (blue) stained for γ-H2AX (red). Scale bars: 1 μm. The proportion of γ-H2AX+ cells (>20 foci/nucleus) are shown in the histograms to the right of the micrographs. Each bar represents the mean ± SD of duplicate scoring.

To independently validate these data, we performed γ-H2AX immunostaining analysis on ES-1, ES-6, ES-8, EW-8, U2OS, and SAOS cells. The basal proportion of cells that were γ-H2AX+ and the distribution of γ-H2AX+ foci per nucleus were similar across the EWS and OS cell lines (Figure S1). All cell lines showed rapid γ-H2AX localization to foci with double-strand DNA (dsDNA) breaks after exposure to 5-Gy IR (Figure S1). To determine whether any defect in the repair of the dsDNA breaks occurred after IR exposure, we performed a timecourse experiment. Cells were exposed to 5-Gy IR, and then at 5 minutes, 2 hours, 8 hours, and 24 hours, the proportions of γ-H2AX+ cells (>20 foci/cell) were scored. In the OS cell line, the proportion of γ-H2AX+ decreased at 2 hours after IR exposure; by 8 hours, the cells were indistinguishable from the original cell population. The resolution of γ-H2AX+ foci was significantly slower for the EWS cell line (p<0.01; Figure 1H). To determine whether exposure to PARPis further delays the repair of IR-induced dsDNA breaks, we performed a similar experiment in the presence of olaparib, veliparib, or BMN-673 (Figure S1). The number of γ-H2AX+ cells was significantly increased at 24 hours after IR exposure in the presence of PARPis (p< 0.01; Figure S1). Together, these data suggest that EWS cell lines are defective in DNA repair.

DNA-Damaging Agent and PARPi Cytotoxicity in EWS Cells

IRN and TMZ are used in combination to treat recurrent EWS (Casey et al., 2009) and both drugs induce DNA damage through distinct mechanisms (Figure 2A) (Hsiang et al., 1989; Quiros et al., 2010). We performed dose-response experiments measuring the cytotoxicity of SN-38 (the active metabolite of IRN), TMZ, BMN-673, olaparib, and veliparib in 8 Ewing sarcoma cell lines (ES-1, ES-4, ES-6, ES-7, ES-8, EW-8, RD-ES, and SK-ES1), 3 OS cell lines (SAOS2, SAOS2LM7, and U2OS), and 5 OS xenografts (MAST22, MAST38, OS39R, OS43, and OS45).The EWS cell lines were chosen to reflect diversity in EWS-FLI1 translocation type, p53-status, and STAG2 status (Supplemental Information). At 72 hours of exposure, EW-8, ES-8, and ES-1 cells were sensitive to BMN-673 and olaparib, and at 144 hours, all EWS cell lines except ES-6 were sensitive to all 3 PARPis (Figures 2B-E, S2, and Table S2), though veliparib activity was marginal. ES-6 cells, which have a nonfunctional EWS-FLI1 translocation, were reported to be resistant to olaparib at 72 hours but had some sensitivity at later timepoints in colony assays (Garnett et al., 2012). In our analysis, ES-6 cells were resistant to all 3 PARPis at 72 hours and to BMN-673 and veliparib at 144 hours, and had significantly lower potency for olaparib and SN-38 compared to other EWS cell lines (Table S2). At 144 hours, all EWS cell lines tested excluding ES-6 had EC50< 10nM for SN-38 and >100μM for TMZ. In contrast to EWS cells, OS cells showed little sensitivity to any of the compounds tested. Only one compound had an EC50 < 100nM: SN-38 in SAOS2. (Table S2).

Figure 2. Potentiation of PARPi cytotoxicity with IRN and TMZ.

Figure 2

Open in a new tab

(A) Model of DNA damage and synthetic lethality for the combination of PARPis with TMZ or IRN. (B-D) Dose response for EW-8, ES-6, and SAOS cells 72 hours after exposure to each indicated PARPi. Curves fit using data pooled from two biological replicates, each with at least three technical replicates. (E) A similar experiment was performed at 144 hours. (F) Immunoblot with quantification (G) of the knock down of PARP1 in EW-8 cells transfected with a PARP1 siRNA. (H-J) Dose response for BMN-673, olaparib, and veliparib in EW-8 cells at 72 hours with (solid line) and without (dashed line) knock down of PARP1. Each data point is the mean ± SD of triplicate wells. (K-P) Potentiation of PARPi in the presence of increasing concentrations of TMZ and SN-38. Curves were generated by taking horizontal slices through the efficacy surface estimated using the response surface model (RSM) approach.

To determine whether sensitivity to PARPis depends on the expression of PARP1, we knocked down PARP1 protein in ES-8 cells with an siRNA (Murai et al., 2012)(Figure 2F-J). For each drug, PARPi sensitivity was reduced when PARP1 expression was knocked down relative to a control siRNA, consistent with the PARP trapping mechanism (Murai et al., 2012; Murai et al., 2014). Indeed, PARP trapping potential followed the same trend observed in the cytotoxicity assay: BMN-673 > olaparib > veliparib (Figure S2).

Development of an Orthotopic EWS Xenograft Model

To generate an EWS orthotopic xenograft, we developed a method for injecting EWS cell lines or primary human tumor cells into the bone marrow of the femur of immunocompromised mice (Supplemental Information). Briefly, cells are resuspended in Matrigel at 100,000 cells/μL and drawn up in a Hamilton syringe with a 25-G needle. The patella and ligaments of the knee are laterally displaced, and the needle is inserted into the intercondylar fossa by using a closed technique (Figure 3A-F). Early signs of engraftment include periosteal elevation on x-ray images, followed by local tumor cell invasion, and hair-on-end appearance (Figure 3G), as seen in patients with EWS. The soft-tissue features of EWS orthotopic xenografts can be visualized by MR imaging (Figure 3H), and tumor calcification and accompanying alterations in femoral architecture can be monitored via micro-CT (Figure 3I). We also tested 2 PET tracers (F18-deoxyglucose and C11-methionine); C11-methionine provided superior sensitivity and signal-to-noise for the orthotopic EWS xenografts (Figure 3J-M and data not shown). Tumor growth surrounding the femur (Figure 3N) and tumor histology (Figure 3O,P) of the orthotopic tumors were very similar to EWS tumors in patients.

Figure 3. Development and characterization of an orthotopic EWS tumor model.

Figure 3

Open in a new tab

(A) Diagram of the injection procedure. (B-D) X-ray images of a mouse leg showing the injection procedure before, during, and after injection. (E, F) Hematoxylin and eosin staining of mouse femur and bone marrow after injection of EWS cells in Matrigel (yellow dashed line). Arrows indicates the injection site. (G) X-ray image of an orthotopic tumor with bony extensions (arrow). (H) Transverse view of the soft-tissue and bony component of the orthotopic EWS xenograft in an MR image. (I-M) Micro-PET/CT scans using 11C-methionine. The tumor (arrow) shows accumulation of the radiotracer. (N) Photograph of the femur removed from a mouse with a large mass from the orthotopic xenograft. (O, P) High- and low-power images of the orthotopic tumor showing its extension from the bone to the surrounding soft tissue.

In Vivo PK and PD of Veliparib, Olaparib, and BMN-673

To determine the murine equivalent dose (MED) and the level of tumor penetration of the 3 PARPis, we measured drug concentration in the plasma and tumors of CD1-nude mice with orthotopic xenografts at timepoints ranging from 30 minutes to 24 hours (Figure S3). We estimated the plasma area under the concentration-time curve (AUC) for each drug and compared those data to the AUC from plasma PK data in patients (Figure S3 and Supplemental Information). Our data suggest that the most appropriate dose for twice-daily oral administration in mice is 12.5 mg/kg veliparib, 50 mg/kg olaparib, and 0.125 mg/kg BMN-673 (Supplemental Information). At 12 hours after dosing, concentration of at least 0.05 μM veliparib, 0.09 μM olaparib, and 0.015 μM BMN-673 were achieved in the orthotopic tumor (Figure S3).

Next, we combined the in vivo tumor pharmacokinetic data with the cytotoxicity data for each PARPi in combination with SN-38 or TMZ using a modified response surface model (RSM) approach (Figure S3 and Supplemental Information). A total of 7 drug pairs were examined (3 PARPi + SN-38, 3 PARPi + TMZ, and SN-38 + TMZ) in four EWS cell lines. We used non-linear regression to fit the observed two-dimensional assay response surface as a function of the hill equation parameters for each compound, the overall efficacy, and kappa (κ) – a measure of the interaction between the two drugs (κ<0 indicates antagonism, κ =0 indicates Loewe additivity, and κ>0 indicates synergy). In every EWS cell line tested, the fitted κ value indicated additivity or synergy for all PARPi + SN-38 or TMZ combinations. Overall, the RSM analyses suggest that combining PARPi with either TMZ or SN-38 will be synergistic in EWS cell lines, with a greater degree of synergy observed for PARPi + TMZ drug pairs.

Although the sign and magnitude of the interaction between drug pairs is important, the combined efficacy at physiologically-reasonable concentrations is more relevant when determining the potential utility of a drug combination for in vivo application. Using the RSM, one can predict the combined efficacy of a drug combination at any concentration. In translocation-positive EWS cell lines, SN-38 begins to significantly potentiate PARPi between 1-10 nM, whereas 10-100uM TMZ is required for the same level of potentiation (Figure 2K-P). In all translocation-positive EWS cell lines, we predict that the most efficacious drug combinations are: BMN 673 + SN-38 > Olaparib + SN-38 > BMN 673 + TMZ > all other drug pairs. In summary, our in vitro synergy study using RSM suggests that although PARPi + TMZ pairs tend to be more synergistic than PARPi + SN-38 pairs, less SN-38 is needed to achieve a desirable level of efficacy.

To verify that the drugs penetrated the tumor and inhibited PARP in the tumor cells in vivo, we performed a PD assay II (Supplemental Information). At the MED, each of the PARPis reduced PARP activity within 1 hour, and PARP activity was restored over the next 6 to 12 hours (Figure S3). BMN-673 and olaparib sustained PARP inhibition longer than did veliparib in vivo.

Preclinical Phase I Study

Having established that sufficient levels of the 3 PARPis can be achieved in the orthotopic xenograft in vivo to inhibit PARP at the MED, we initiated a preclinical phase I trial to test the tolerability of PARPis as single agents and in combination with IRN, TMZ, or both. Our preclinical phase I trials were performed in 3 to 5 female CD1-nude mice per treatment group for a total of 4 to 6 courses of therapy (12-18 weeks; Figure 4A). The PARPis were administered orally twice daily to match the dosing in patients. Importantly, we used a low-dose, protracted schedule of IRN (i.e., 1.25 mg/kg twice daily for 5 days (d×5×2) equivalent to 20 mg/m2 d×5×2 in children ((Furman et al., 1999) and Supplemental Information). The pediatric dose using this d×5×2 schedule is much lower than that used in adults with cancer, who receive 100 to 350 mg/m2 IRN 1 to 3 times per course (Kummar et al., 2011; Samol et al., 2012). TMZ was initially administered at 33 mg/kg on a d×5 schedule, which is equivalent to 100 mg/m2 in children (Horton et al., 2007).

Figure 4. Preclinical phase I/II studies.

Figure 4

Open in a new tab

(A) Drug-combination schedules. Yellow circles represent the PARPi, green bars represent the TMZ and red stars represent daily IP dosing of IRN. (B-D) Survival of mice in the preclinical phase I trial of each PARPi combined with IRN (I) or TMZ (T). In some groups, TMZ was reduced by 50% (T50), 64% (T36), or 70% (T30). For BMN-673, the dose was reduced by 20% in 1 group (P80). (E-G) Preclinical phase II data for IRN+TMZ alone or in combination with veliparib (blue), olaparib (red), or BMN-673 (green). Tumor burden was monitored by Xenogen imaging. (H-L) Representative Xenogen images for each treatment group and photographs of the tumors (arrows) or femurs at the end of the study.

For veliparib, nearly all combinations were well tolerated for 4 courses (12 weeks), except the veliparib+IRN+TMZ combination (Figure 4B). In patients who do not tolerate the combination of TMZ+IRN, the dose is often reduced 50% from 100 mg/m2 to 50 mg/m2 (Wagner et al., 2009). Veliparib+IRN+TMZ(50%) was well tolerated for 4 courses (Figure 4B). Similar results were obtained with olaparib (Figure 4C). The tolerability of TMZ was even less in combination with BMN-673 (Figure 4D). The TMZ dose had to be reduced by 70% (10 mg/kg in mice and 30 mg/m2 in children), and the BMN-673 dose by 20% of the MED (0.1 mg/kg) to make this combination tolerable (Figure 4D and Table S3).

Preclinical Phase II Study

To test if any of the PARPis reduce orthotopic EWS tumor growth in vivo, we performed a preclinical phase II study. Phase II studies are designed to provide rapid efficacy data by using a randomized, placebo-controlled study design (Supplemental Information). 5 groups comprising 5 mice per group with EW-8 orthotopic xenografts were used in this study. In addition to the placebo group, we included IRN+TMZ(50%) as a control group. The placebo group showed rapid tumor progression, and all mice were off study by 14 days after enrollment (data not shown). The mice in the IRN+TMZ(50%) group had stable disease (SD) or progressive disease (PD) but no partial response (PR) or complete response (CR) (Figure 4E-G and Supplemental Information). The veliparib+IRN+TMZ(50%) group included 1 mouse with PR; the rest had SD or PD (Figure 4E). The olaparib+IRN+TMZ(50%) group had 2 mice with CR, 1 with PR, and 1 with SD (Figure 4F). The BMN-673(80%)+IRN+TMZ(30%) group had 4 mice with CR (Figure 4G). Overall, the bioluminescence from the Xenogen imaging correlated with tumor weight and histopathologic evaluation (Figures 4H-L, S4 and Supplemental Information). As long as 12 weeks after cessation of treatment, mice with a CR that received olaparib+IRN+TMZ(50%) or BMN-673(80%)+IRN+TMZ(30%) have no evidence of tumor recurrence, but those with PD, SD, or PR have all progressed rapidly (data not shown). ES-1 cells had poor engraftment and IRN+TMZ completely eliminated the tumors (data not shown). Mice with orthotopic ES-8 xenografted cells had similar response to the EW-8 cells (Figure S5). The ES-6 cells were similar to ES-1 in their poor engraftment efficiency and slow growth but the tumors showed response to olaparib+IRN+TMZ(50%) and BMN-673(80%)+IRN+TMZ(50%) (Figure S5).

The Cmax and AUC of SN-38 in mouse plasma has been reported to be slightly higher than those in human plasma because mice have higher levels of the plasma carboxylesterases that convert IRN to SN-38 (Morton et al., 2005; Morton et al., 2000). To determine whether the efficacy of IRN+ PARPis in vivo was caused by the higher Cmax or AUC of SN-38 in mice, we performed plasma PK and efficacy studies in carboxylesterase-deficient mice that more closely recapitulate the PK profile of SN-38 in humans (Morton et al., 2005). As shown previously, the plasma SN-38 Cmax and AUC were slightly reduced in the carboxylesterase-deficient mice after intraperitoneal injection of IRN (Fig. S5 and Supplemental Information) but there was no effect on tumor response (Fig. S5).

Preclinical Phase III Study

Our in vitro RSM predicted that BMN-673+IRN or BMN-673+TMZ would be significantly cytotoxic, olaparib+IRN may be more efficacious than olaparib+TMZ, and that veliparib-combination chemotherapy would probably not achieve significant in vivo efficacy at the dose and schedule used for these experiments. To test these predictions and directly compare the efficacy of each PARPi in combination with TMZ, IRN, or both, we performed a double-blind, randomized, placebo-controlled preclinical phase III trial. Briefly, we performed 350 intrafemoral injections of luciferase-labeled ES-8 cells into female CD1-nude mice. Over 5 weeks, the trial enrolled 274 mice and randomized them to 15 treatment groups (Table S4). Ten mice each were assigned to the placebo, IRN+TMZ, or IRN+TMZ(50%) groups; 25 mice each were assigned to the BMN-673(80%)+IRN+TMZ(30%) group; and 20 mice each were assigned to the other 11 treatment groups (Table S4). All mice were assigned a mouse medical record number at enrollment, which allowed their data to be linked via an OpenClinica database (Supplemental Information).

Overall survival of mice in the placebo group did not significantly differ from that in the single-agent PARPi-treatment groups (Figure 5A). As predicted by the RSM, of the groups that received combinations of PARPis with TMZ, only those that received BMN-673+TMZ(50%) had significantly improved overall survival (p=0.0004, Fig. 5B). The groups that received PARPi+IRN tolerated the combinations well; responses were significantly better in the olaparib+IRN or BMN-673+IRN groups than in the veliparib+IRN or TMZ+IRN groups (p=0.0001, Figure 5C). All but 1 mouse in the olaparib+IRN+TMZ(50%) or BMN-673(80%)+IRN+TMZ(30%) groups were alive at the end of the 12-week study (Figure 5D). All mice that came off study due to tumor growth were classified as having PD. Those that completed the 4 courses of therapy were classified as having CR, PR, SD, or PD based on the Xenogen thresholds used in the preclinical phase II study. The following percentages of each group showed CR: 15% (3/20) in the veliparib+IRN+TMZ(50%) group, 71% (12/17) in the olaparib+IRN+TMZ(50%) group, and 88% (14/16) in the BMN-673(80%)+IRN+TMZ(30%) group. The Xenogen data correlated with tumor burden and histopathology (Figure 5I-K).

Figure 5. Preclinical phase III study.

Figure 5

Open in a new tab

(A-D) Survival curves for each of the 15 treatment groups. (E-G) Tumor response for individual mice in the TMZ+IRN group and the triple-drug combinations for veliparib (blue), olaparib (red), and BMN-67e (green). The cutoffs for progressive disease (PD), stable disease (SD), partial response (PR), and complete response (CR) are indicated by gray shading. (H) Histogram of the proportion of CRs seen in each triple-drug treatment group. (I) Representative Xenogen images of single-agent PARPis and corresponding triple-drug treatment groups. (J) Representative photographs of tumors from placebo, IRN+TMZ, single-agent PARPi groups and corresponding triple-drug treatment groups. (K,L) Representative micrographs of H&E stained tissue sections from placebo, IRN+TMZ, single-agent PARPi groups and corresponding triple-drug treatment groups. Scale bars K, 500μm; L, 100μm.

In cell culture, IRN and TMZ showed strong potentiation of veliparib-mediated killing of EWS cells (Table S2). However, at the MED of that used in pediatric brain tumor phase II studies of veliparib with temozolomide (Su et al., 2014), our model predicted and our preclinical phase II and III studies validated that the levels of veliparib in the tumors were not sufficient to achieve high rates of CR. The pediatric brain tumor combination phase I trial with veliparib and temozolomide was designed to maximize the temozolomide dose and escalate the veliparib. However, our data presented here suggest that the opposite may be advantageous for Ewing sarcoma. Therefore, we performed additional phase I/II studies with a higher dose of veliparib (62.5 mg/kg BID x5 x2) in combination with IRN and TMZ(50%). The higher dose veliparib was well tolerated in combination with IRN and TMZ and had efficacy similar to that of olaparib+IRN+TMZ(50%) and BMN-673+IRN+TMZ(30%) (Figure S5).

DISCUSSION

Three PARPis in combination with DNA-damaging agents showed that PARPi-mediated cytotoxicity can be potentiated by IRN and TMZ in vitro and in vivo. In vivo PK of veliparib, olaparib, and BMN-673 in plasma and tumor were used to calculate the MEDs and demonstrate tumor penetration and PARP inhibition at clinically relevant doses in orthotopic EWS xenografts. A preclinical phase I study revealed that combinations of PARPis with IRN were better tolerated than those with TMZ. Furthermore, the low-dose, protracted schedule of IRN used to treat pediatric patients with solid tumors (20 mg/m2 d×5×2) was well tolerated with all 3 PARPis, but the TMZ dose had to be reduced (50%-70%). The levels of all 3 PARPis in the tumor were sufficient to potentiate cytotoxicity with DNA-damaging agents but were insufficient to induce cytotoxicity on their own.

A preclinical phase II study demonstrated the efficacy of all 3 PARPis in combination with IRN and TMZ, and this provided justification for a subsequent in vivo efficacy study. In a double-blind, randomized, placebo-controlled preclinical phase III trial, we found significant improvement in overall survival and outcome when olaparib or BMN-673 was combined with TMZ and IRN. Most of the mice had a CR and did not exhibit tumor recurrence as long as 12 weeks after cessation of therapy. These data suggest that the combination of PARPis and IRN administered in the low-dose, protracted schedule optimized for pediatric patients should be considered for clinical development. A reduced dose of TMZ may be incorporated to provide full potentiation of PARP inhibition in EWS. While veliparib was not as active as olaparib and BMN-673 in the preclinical phase III study, a higher dose of veliparib was well tolerated in preclinical phase I and had efficacy comparable to olaparib and BMN-673 in a preclinical phase II with IRN and TMZ.

DNA-Damaging Agents and PARPis in EWS

EWS cells express high levels of SLFN11 and PARP1 compared to other cancers, providing a rationale for combining DNA-damaging agents and PARPis. TMZ has been favored in the PARPi EWS clinical trials developed to date because in adult, camptothecins (topotecan or IRN) caused dose-limiting myelosuppression and diarrhea when combined with PARPis (Kummar et al., 2011; Samol et al., 2012). However, the low-dose, protracted schedule and the sensitivity of EWS cells to IRN make this camptothecin an attractive agent to combine with PARPis. Our PK, PD, and in vitro drug-combination studies showed that concentrations of PARPis and IRN can be achieved in vivo at levels sufficient to potentiate PARP-mediated cytotoxicity in vivo.

We did not test ionizing radiation in our studies, but this is another therapy that should be considered in combination with PARPis and DNA-damaging agents. EWS tumors are sensitive to IR, and similar potentiation might be achieved with IR in vivo (Lee et al., 2013). The majority of EWS cell lines have TP53 mutations (Tirode et al., 2014) so it is difficult to compare the sensitivity of wild type and TP53-deficient EWS cell lines to PARPis and DNA-damaging agents. However, TP53-deficient osteosarcoma cell lines are insensitive to those drug combinations, so p53 status alone cannot explain the sensitivity to PARPis and DNA-damaging agents in our study. Indeed, Oplustilova and colleagues found that PARPis can sensitize cells to camptothecin or ionizing radiation independent of p53 status in colorectal carcinoma cells (Oplustilova et al., 2012).

Our results suggest that STAG2 status does not correlate with sensitivity to DNA-damaging agents or PARPis in EWS, and this has important clinical consequences. In a separate whole-genome sequencing study, 17% of EWS tumors had inactivating somatic mutations in STAG2, a component of the cohesin complex (Tirode et al., 2014). Those data are consistent with previously published data from Solomon et al. showing that STAG2 is frequently lost in EWS (Solomon et al., 2011). While STAG2-deficient glioblastoma cells appear to be more sensitive to PARPis (Bailey et al., 2014), this does not appear to be the case in EWS. It is encouraging that the TP53;STAG2-deficient EWS lines remain sensitive to the drug combinations because those patients have the worst overall survival (Tirode et al., 2014).

PARP Trapping

While it is unlikely that high PARP1 levels alone are sufficient for conferring sensitivity to PARPis—a DNA-repair defect must also be present—we did observe that the in vitro sensitivity of PARPis in EWS correlated with PARP trapping potential (Murai et al., 2012; Murai et al., 2014). The apparent differences in PARP trapping may be the result of differences in drug retention times. Specifically, BMN-673 may be more active as a single agent because it remains bound to PARP1 longer than does olaparib or veliparib, resulting in more persistent PARP-DNA adducts. This may also contribute to its reduced tolerability. The implications of these data are related to dosing of each PARPi: efficacy may be improved if veliparib is dosed at a higher intensity to overcome these differences in tumor clearance and ultimately retention of PARP-bound drug in the tumor cells. Indeed, our preclinical phase II study with high dose veliparib+IRN+TMZ showed efficacy similar to that of BMN-673 and olaparib. Similarly, once-daily dosing of BMN-673 may be better tolerated than twice-daily dosing without affecting PARP inhibition and efficacy. As PARPis are combined with DNA-damaging agents to treat patients, it will be essential to balance the dosing and schedule of the PARPis with DNA-damaging agents to achieve maximum efficacy and tolerability.

Preclinical Testing for Pediatric Solid Tumors

Survival of patients with recurrent or metastatic EWS is among the worst of pediatric cancers. More importantly, outcome has not been significantly improved in 20 years. This explains the enthusiasm for treating recurrent EWS with PARPis, once the original discovery showed that EWS cell lines are sensitive to olaparib (Garnett et al., 2012). Indeed, the first clinical trial (NCT01583543) opened within a month of that publication. However, several unanswered questions from that first study had important implications for the clinical trial: Could the levels of olaparib required to kill EWS cell lines in vitro be achieved in a patient’s tumor in vivo? Could the PARPi be combined with a DNA-damaging agent? Which DNA-damaging agent should be used? What is the best dose and schedule for the PARPi and the DNA-damaging agent?

In this study, we developed and optimized a comprehensive preclinical-testing paradigm to answer these questions in EWS. The advantage of the program presented here is that it directly relates in vivo PK and PD to cytotoxicity in vitro and then uses MEDs and schedules to test those predictions in an array of systems (i.e., cell culture to relevant orthotopic EWS xenografts). Also, our preclinical phase I, II, and III trials mirror the approach used in clinical trials and provide an efficient system to advance new therapies. However, the key of any preclinical-testing effort is the predictive power of the preclinical results. The data presented here predict that single-agent olaparib will not be an effective treatment of EWS, and the efficacy of olaparib+TMZ may be limited by tolerability. As the ongoing trials with BMN-673 accrue patients, we will have additional opportunities to determine the predictive power of our preclinical-testing efforts.

The drug combination results from this study, specifically the utility of individual PARPi, should not be generalized to non-EWS tumors in part because of the sensitivity of EWS cell lines to IRN (Barretina et al., 2012). Also, in vivo drug efficacy is the product of a multitude of effects beyond intrinsic cell sensitivity, such as influence of the tumor niche, immune system, and pharmacokinetics. Future efforts should focus on primary human orthotopic xenografts to complement the EWS cell line data presented here.

EXPERIMENTAL PROCEDURES

Comet Assay

50,000 ES-1, ES- 6, ES-8, or U2OS cells or 100,000 mesenchymal stem cells were plated onto 12-well dishes per well. Cells were incubated for 24 h and then treated with Parp inhibitor: 10 uM BMN-673, 1 uM BMN-673, 10 uM Veliparib, 1 uM Veliparib, 10 uM Oliparib, 1 uM Oliparib, or DMSO. Cells were incubated with the inhibitor for 14 h and then treated with 10 Gy IR or were left untreated. Cells were incubated an additional 10 h and then trypsinized and collected. For un-repaired control, cells were treated 10 Gy IR and immediately harvested. 50,000 cells were suspended in 30 uL PBS. 10 uL of this suspension was mixed with 200 μL of 0.5% low melting point agarose (Sigma) and layered on CometSlides (Trevigen). Alkaline single cell gel electrophoresis was performed as described previously (Benavente et al., 2013).

Animals

CD-1 nude immunodeficient mice were purchased from Charles River (strain code 087, heterozygous). Esterase-deficient SCID mice were bred and obtained from Philip Potter (St. Jude Children’s Research Hospital). This study was carried out in strict accordance with the recommendations in the Guide to Care and Use of Laboratory Animals of the National Institute of Health. The protocol was approved by the Institutional Animal Care and Use Committee at St. Jude Children’s Research Hospital. All efforts were made to minimize suffering. All mice were bred and housed in accordance with approved IACUC protocols. Animals were housed on a 12-12 light cycle (light on 6am off 6pm) and provided food and water ad libitum.

Drugs used for in vitro studies

Veliparib was purchased from Selleck (S1004, CAS 912444-00-9), Olaparib was purchased from LC Labs (O-9201, CAS 763113-22-0), BMN-673 was purchased from Abmole (BMN673, 1207456-01-6), Temozolomide was purchased from Combi-Blocks (OR-2567, CAS 85622-93-1), and SN-38 was purchased from (CAS 86639-52-3). The purity of all compounds was confirmed to be ≥95% using LC/MS coupled with UVTWC/ELSD detection, and concentration was verified using CLND if nitrogen was present in the compound.

Gene Expression Array Analysis

Microarray assays of Ewing sarcoma tumor samples (GSE37371) were compared to osteosarcoma arrays (HGU133v2 Affymetrix arrays, SJ unpublished). The data were RMA normalized, evaluated for quality by PCA and, after outlier removal, statistically compared using the unequal variance t test in Partek Genomics Suite 6.6. Select data known to be associated with DNA repair were then evaluated.

QPCR Analysis

Real-time RT-PCR experiments were performed using the Applied Biosystems 7900HT Fast Real-Time PCR system and custom TaqMan Array Micro Fluidic Cards (Life Technologies). RNA was prepared using Trizol following manufacturer instructions (Life Technologies, 15596018). cDNA was synthesized using the High Capacity cDNA to RNA kit (Life Technologies, 4387406) per user instructions. Samples were analyzed in triplicate and normalized to ACTB2 expression levels.

Histology and Immunohistochemistry

Paraffin-embedded formalin-fixed EWS tumors were serially sectioned for routine hematoxylin and eosin staining. Immunostaining for histochemical analysis was done with Ki67 (ThermoShandon, CA, USA, cat. RM-9106) (dilution 1 200) using hematoxylin as a counterstain (1:10 dilution). Histology images were obtained using Aperio® ImageScope (Leica Biosystems).

PARP1 Knockdown

Gene-specific siRNAs (mix of 4 sequences) for PARP1 (ThermoScientific, L-006656-03-0005) were transfected into ES-8 cells using Libofectamine RNAiMAX Reagent (Invitrogen, 13778). Cells were harvested 48 h post transfection and lysed for Western analysis to confirm knockdown of PARP1. Also, cells were harvested by trypsinization, seeded onto 96 well plates at a density of 5000 cells per well in 40 uL of media. 72 h post transfection, cells were treated with 0, 1, 2.5, 5, 10, 20, 50, or 100 uM Olaparib, Veliparib, or BMN-673 in triplicate for each condition. 72 h post addition of PARPi, cells were analyzed for cellular activity using Cell Titer Glo (Promega, G7570).

Supplementary Material

01

Figure S1 related to Figure 1. QPCR analysis of DNA repair genes and response to DNA damage. A) Histograms showing normalized relative fold for QPCR using Taqman probes for the indicated DNA repair genes. Each bar represents the mean of duplicate PCR reactions for ES1, ES8, EW8 and RDES Ewing’s Sarcoma Cell Lines. As controls, human mesenchymal stem cells (MSCs), neuroblastoma cells (NB), osteoblasts (OB) and osteosarcoma (OS) cells were used. All data were normalized to beta-actin expression and plotted relative to osteosarcoma (1.0). B) Micrographs of cell lines stained for gamma-H2AX (red fluorescence) before and after exposure to 5 Gy IR. Histograms are shown of scoring of 100 cells in duplicate per sample. The number of foci per nucleus (blue immunofluorescence) was scoreed and the mean and standard deviation are plotted in the histogram on the bottom. The distribution of foci are indicated for the untreated cells and the treated cells are shaded in gray. C) Micrographs of ES-8 cells treated with 5 Gy IR in the presence of 10 micromolar of each PARPi. Gamma-H2AX is shown by red fluorescence and nuclear DNA is indicated by blue fluorescence. D) Histogram of the proportion of gamma-H2AX immunopositive cells (>20 foci per cell) for each treatment group. 100 cells were scored in duplicate and the mean and standard deviation are shown.

Figure S2 related to Figure 2. Sensitivity of EWS cell lines to PARPi+SN-38 and temozolomide and mechanism of action. A) Dose response curves for each PARPi alone (red curve) or in combination with fixed concentrations of SN-38 (blue curve). Each data point is the mean and standard deviation of triplicate wells in each experiment. The gray dashed line indicates the level of cytotoxicity of the SN-38 alone. B) Dose response curves for each PARPi alone (red curve) or in combination with fixed concentrations of temozolomide (blue curve). Each data point is the mean and standard deviation of triplicate wells in each experiment. The gray dashed line indicates the level of cytotoxicity of the temozolomide alone. C,D) Immunoblot for the nuclear soluble and chromatin bound proteins in cells treated with10 micromolar PARPi+ 0.01%MMS for 2 hours. A second group was treated with 10 micromolar PARPi for 2 hours, washed and then exposed to 0.01% MMS for 2 hours. A third group was treated with 10 micromolar PARPi, washed and then exposed to 0.01% MMS for 2 hours and then washed again before harvesting for analysis. Topoisomerase I is used as a positive control for the nuclear soluble fraction and histone H3 is used as a positive control for the chromatin bound fraction. E) Histogram showing the quantitation of the bands in the blots in A,B above. The normalized relative fold level of PARP1 in each fraction is plotted.

Figure S3 related to Figure 3. PK and PD of PARPis in EWS. (A-C) Concentration versus time plot for BMN-673, olaparib, and veliparib in plasma after a single oral dose as indicated. (D-F) Concentration versus time plot for BMN-673, olaparib, and veliparib in the orthotopic EWS tumor after a single oral dose as indicated. The gray box indicates the minimum level of drug concentration maintained for 12 hours in the orthotopic tumor. (G-L) Response surface model fits for PARPi combined with TMZ (G-I) or SN-38 (J-L). ‘Relative survival’ is the reduction in CellTiter-Glo signal relative to the negative control. Kappa values reflect the sign and magnitude of the drug-drug interaction: κ <0 indicates antagonism, κ =0 indicates Loewe additivity, and κ >0 indicates synergy. A90 values quantify the proportion of the area of the response surface defined by concentrations from 0 to Cmax for each compound with efficacy >90% (10-fold reduction in survival) The maximum value of A90 is 1.0. (M-O) In vivo PD of PARP inhibition in an orthotopic EWS tumor at different timepoints after a single oral dose of BMN-673, olaparib, or veliparib. Each point represents the mean ± SD of triplicate tumors from different mice.

Figure S4 related to Figure 4. Histological analysis of treated and untreated orthotopic xenografts. A-E) Micrographs of H&E staining of placebo (A), IRN+TMZ (B), veliparib+IRN+TMZ (C), olaparib+IRN+TMZ (D), BMN-673+IRN+TMZ (E) tumors. F-J) High magnification micrographs of the tumors shown in A-E. K-O) Immunohistochemical analysis of Ki67 (brown) for each of the tumors shown in A-E. Scale bars: 50 microns.

Figure S5 related to Figure 5. Pharmacokinetics and preclinical phase II efficacy in esterase deficient mice. A,B) Plot of tumor burden (xenogen signal) for individual mice from a preclinical phase II study of ES6 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of olaparib+IRN+TMZ(50%) and BMN-673(80%)+IRN+TMZ(30%). C,D) Plot of tumor burden for individual mice from a preclinical phase II study of EW8 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of olaparib+IRN+TMZ(50%) and BMN-673(80%)+IRN+TMZ(30%). E) Concentration versus time plot of plasma levels of SN-38 in wild type (red) and esterase deficient mice (blue) following an IP injection of irinotecan. F) Xenogen signal for orthotopic ES-8 tumors in immunocompromised esterase deficient mice treated with olaparib alone (black line) or olaparib+IRN (red line). G) Xenogen signal for orthotopic ES-8 tumors in immunocompromised esterase deficient mice treated with BMN-673 alone (black line) or BMN-673+IRN (red line). H) Plot of tumor burden (xenogen signal) for individual mice from a preclinical phase II study of ES8 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of veiliparib+IRN+TMZ(50%) at a high dose of veliparib (62.5 mg/kg BID x5 x2).

Table S1 related to Figure 1. Expression of DNA repair pathway genes.

Provided as a separate file.

Table S2 related to Figure 2. Drug sensitivity data.

Provided as a separate file.

Table S3 related to Figure 4. CBC-D data for preclinical testing.

Provided as a separate file.

Table S4 related to Figure 5. Preclinical Phase III data.

Provided as a separate file.

ACKNOWLEDGMENTS

We thank Jieun Kim for assistance with PET-CT and MRI, Angela McArthur for editing the manuscript, Fred Krafcik for help with cell screening and David Finkelstein for assistance with bioinformatics analysis. This work was supported, in part, by Cancer Center Support (CA21765) from the NCI, grants to M.A.D from the NIH (EY014867 and EY018599 and CA168875), and the American Lebanese Syrian Associated Charities (ALSAC). M.A.D. was also supported by a grant from Alex’s Lemonade Stand Foundation for Childhood Cancer and HHMI. Finally, we thank John and Andra Tully and the Tully Family Foundation for generous support of Pediatric Solid Tumor Research at St. Jude Children’s Research Hospital.

REFERENCES

  1. Ashworth A. A synthetic lethal therapeutic approach: poly(ADP) ribose polymerase inhibitors for the treatment of cancers deficient in DNA double-strand break repair. J Clin Oncol. 2008;26:3785–3790. doi: 10.1200/JCO.2008.16.0812. [DOI] [PubMed] [Google Scholar]
  2. Bagatell R, London WB, Wagner LM, Voss SD, Stewart CF, Maris JM, Kretschmar C, Cohn SL. Phase II study of irinotecan and temozolomide in children with relapsed or refractory neuroblastoma: a Children’s Oncology Group study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2011;29:208–213. doi: 10.1200/JCO.2010.31.7107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S, Wilson CJ, Lehar J, Kryukov GV, Sonkin D, et al. The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 2012;483:603–607. doi: 10.1038/nature11003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Benavente CA, McEvoy JD, Finkelstein D, Wei L, Kang G, Wang YD, Neale G, Ragsdale S, Valentine V, Bahrami A, et al. Cross-species genomic and epigenomic landscape of retinoblastoma. Oncotarget. 2013;4:844–859. doi: 10.18632/oncotarget.1051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brenner JC, Feng FY, Han S, Patel S, Goyal SV, Bou-Maroun LM, Liu M, Lonigro R, Prensner JR, Tomlins SA, Chinnaiyan AM. PARP-1 inhibition as a targeted strategy to treat Ewing’s sarcoma. Cancer research. 2012;72:1608–1613. doi: 10.1158/0008-5472.CAN-11-3648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E, Kyle S, Meuth M, Curtin NJ, Helleday T. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature. 2005;434:913–917. doi: 10.1038/nature03443. [DOI] [PubMed] [Google Scholar]
  7. Casey DA, Wexler LH, Merchant MS, Chou AJ, Merola PR, Price AP, Meyers PA. Irinotecan and temozolomide for Ewing sarcoma: the Memorial Sloan-Kettering experience. Pediatric blood & cancer. 2009;53:1029–1034. doi: 10.1002/pbc.22206. [DOI] [PubMed] [Google Scholar]
  8. Delattre O, Zucman J, Plougastel B, Desmaze C, Melot T, Peter M, Kovar H, Joubert I, de Jong P, Rouleau G, et al. Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours. Nature. 1992;359:162–165. doi: 10.1038/359162a0. [DOI] [PubMed] [Google Scholar]
  9. Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB, Santarosa M, Dillon KJ, Hickson I, Knights C, et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature. 2005;434:917–921. doi: 10.1038/nature03445. [DOI] [PubMed] [Google Scholar]
  10. Fong PC, Boss DS, Yap TA, Tutt A, Wu P, Mergui-Roelvink M, Mortimer P, Swaisland H, Lau A, O’Connor MJ, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;361:123–134. doi: 10.1056/NEJMoa0900212. [DOI] [PubMed] [Google Scholar]
  11. Furman WL, Stewart CF, Poquette CA, Pratt CB, Santana VM, Zamboni WC, Bowman LC, Ma MK, Hoffer FA, Meyer WH, et al. Direct translation of a protracted irinotecan schedule from a xenograft model to a phase I trial in children. J Clin Oncol. 1999;17:1815–1824. doi: 10.1200/JCO.1999.17.6.1815. [DOI] [PubMed] [Google Scholar]
  12. Garnett MJ, Edelman EJ, Heidorn SJ, Greenman CD, Dastur A, Lau KW, Greninger P, Thompson IR, Luo X, Soares J, et al. Systematic identification of genomic markers of drug sensitivity in cancer cells. Nature. 2012;483:570–575. doi: 10.1038/nature11005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Granowetter L, Womer R, Devidas M, Krailo M, Wang C, Bernstein M, Marina N, Leavey P, Gebhardt M, Healey J, et al. Dose-intensified compared with standard chemotherapy for nonmetastatic Ewing sarcoma family of tumors: a Children’s Oncology Group Study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2009;27:2536–2541. doi: 10.1200/JCO.2008.19.1478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Horton TM, Thompson PA, Berg SL, Adamson PC, Ingle AM, Dolan ME, Delaney SM, Hedge M, Weiss HL, Wu MF, Blaney SM. Phase I pharmacokinetic and pharmacodynamic study of temozolomide in pediatric patients with refractory or recurrent leukemia: a Children’s Oncology Group Study. J Clin Oncol. 2007;25:4922–4928. doi: 10.1200/JCO.2007.12.0667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Howlader N, Noone AM, Krapcho M, Garshell J, Neyman N, Altekruse SF, Kosary CL, Yu M, Ruhl J, Tatalovich Z, et al. SEER Cancer Statistics Review, 1975-2010, NCI. 2013 [Google Scholar]
  16. Hsiang YH, Lihou MG, Liu LF. Arrest of replication forks by drug-stabilized topoisomerase I-DNA cleavable complexes as a mechanism of cell killing by camptothecin. Cancer Res. 1989;49:5077–5082. [PubMed] [Google Scholar]
  17. Kummar S, Chen A, Ji J, Zhang Y, Reid JM, Ames M, Jia L, Weil M, Speranza G, Murgo AJ, et al. Phase I study of PARP inhibitor ABT-888 in combination with topotecan in adults with refractory solid tumors and lymphomas. Cancer Res. 2011;71:5626–5634. doi: 10.1158/0008-5472.CAN-11-1227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lee HJ, Yoon C, Schmidt B, Park do J, Zhang AY, Erkizan HV, Toretsky JA, Kirsch DG, Yoon SS. Combining PARP-1 inhibition and radiation in ewing sarcoma results in lethal DNA damage. Mol Cancer Ther. 2013;12:2591–2600. doi: 10.1158/1535-7163.MCT-13-0338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lessnick SL, Ladanyi M. Molecular pathogenesis of Ewing sarcoma: new therapeutic and transcriptional targets. Annu Rev Pathol. 2012;7:145–159. doi: 10.1146/annurev-pathol-011110-130237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Morton CL, Iacono L, Hyatt JL, Taylor KR, Cheshire PJ, Houghton PJ, Danks MK, Stewart CF, Potter PM. Activation and antitumor activity of CPT-11 in plasma esterase-deficient mice. Cancer Chemother Pharmacol. 2005;56:629–636. doi: 10.1007/s00280-005-1027-y. [DOI] [PubMed] [Google Scholar]
  21. Morton CL, Wierdl M, Oliver L, Ma MK, Danks MK, Stewart CF, Eiseman JL, Potter PM. Activation of CPT-11 in mice: identification and analysis of a highly effective plasma esterase. Cancer Res. 2000;60:4206–4210. [PubMed] [Google Scholar]
  22. Murai J, Huang SY, Das BB, Renaud A, Zhang Y, Doroshow JH, Ji J, Takeda S, Pommier Y. Trapping of PARP1 and PARP2 by Clinical PARP Inhibitors. Cancer Res. 2012;72:5588–5599. doi: 10.1158/0008-5472.CAN-12-2753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Murai J, Huang SY, Renaud A, Zhang Y, Ji J, Takeda S, Morris J, Teicher B, Doroshow JH, Pommier Y. Stereospecific PARP trapping by BMN 673 and comparison with olaparib and rucaparib. Mol Cancer Ther. 2014;13:433–443. doi: 10.1158/1535-7163.MCT-13-0803. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Oplustilova L, Wolanin K, Mistrik M, Korinkova G, Simkova D, Bouchal J, Lenobel R, Bartkova J, Lau A, O’Connor MJ, et al. Evaluation of candidate biomarkers to predict cancer cell sensitivity or resistance to PARP-1 inhibitor treatment. Cell Cycle. 2012;11:3837–3850. doi: 10.4161/cc.22026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Quiros S, Roos WP, Kaina B. Processing of O6-methylguanine into DNA double-strand breaks requires two rounds of replication whereas apoptosis is also induced in subsequent cell cycles. Cell Cycle. 2010;9:168–178. doi: 10.4161/cc.9.1.10363. [DOI] [PubMed] [Google Scholar]
  26. Samol J, Ranson M, Scott E, Macpherson E, Carmichael J, Thomas A, Cassidy J. Safety and tolerability of the poly(ADP-ribose) polymerase (PARP) inhibitor, olaparib (AZD2281) in combination with topotecan for the treatment of patients with advanced solid tumors: a phase I study. Invest New Drugs. 2012;30:1493–1500. doi: 10.1007/s10637-011-9682-9. [DOI] [PubMed] [Google Scholar]
  27. Solomon DA, Kim T, Diaz-Martinez LA, Fair J, Elkahloun AG, Harris BT, Toretsky JA, Rosenberg SA, Shukla N, Ladanyi M, et al. Mutational inactivation of STAG2 causes aneuploidy in human cancer. Science. 2011;333:1039–1043. doi: 10.1126/science.1203619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Stahl M, Ranft A, Paulussen M, Bolling T, Vieth V, Bielack S, Gortitz I, Braun-Munzinger G, Hardes J, Jurgens H, Dirksen U. Risk of recurrence and survival after relapse in patients with Ewing sarcoma. Pediatr Blood Cancer. 2011;57:549–553. doi: 10.1002/pbc.23040. [DOI] [PubMed] [Google Scholar]
  29. Su JM, Thompson P, Adesina A, Li XN, Kilburn L, Onar-Thomas A, Kocak M, Chyla B, McKeegan E, Warren KE, et al. A phase I trial of veliparib (ABT-888) and temozolomide in children with recurrent CNS tumors: a Pediatric Brain Tumor Consortium reportdagger. Neuro Oncol. 2014 doi: 10.1093/neuonc/nou103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Tirode F, Surdez D, Ma X, Parker M, Le Deley MC, Bahrami A, Zhang Z, Lapouble E, Grossetete-Lamali S, Rusch M, et al. Co-association between STAG2 and TP53 mutations defines an aggressive subgroup of Ewing Sarcoma. Cancer Discovery. 2014 doi: 10.1158/2159-8290.CD-14-0622. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tutt AN, Lord CJ, McCabe N, Farmer H, Turner N, Martin NM, Jackson SP, Smith GC, Ashworth A. Exploiting the DNA repair defect in BRCA mutant cells in the design of new therapeutic strategies for cancer. Cold Spring Harb Symp Quant Biol. 2005;70:139–148. doi: 10.1101/sqb.2005.70.012. [DOI] [PubMed] [Google Scholar]
  32. Wagner LM, Villablanca JG, Stewart CF, Crews KR, Groshen S, Reynolds CP, Park JR, Maris JM, Hawkins RA, Daldrup-Link HE, et al. Phase I trial of oral irinotecan and temozolomide for children with relapsed high-risk neuroblastoma: a new approach to neuroblastoma therapy consortium study. J Clin Oncol. 2009;27:1290–1296. doi: 10.1200/JCO.2008.18.5918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Zoppoli G, Regairaz M, Leo E, Reinhold WC, Varma S, Ballestrero A, Doroshow JH, Pommier Y. Putative DNA/RNA helicase Schlafen-11 (SLFN11) sensitizes cancer cells to DNA-damaging agents. Proc Natl Acad Sci U S A. 2012;109:15030–15035. doi: 10.1073/pnas.1205943109. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01

Figure S1 related to Figure 1. QPCR analysis of DNA repair genes and response to DNA damage. A) Histograms showing normalized relative fold for QPCR using Taqman probes for the indicated DNA repair genes. Each bar represents the mean of duplicate PCR reactions for ES1, ES8, EW8 and RDES Ewing’s Sarcoma Cell Lines. As controls, human mesenchymal stem cells (MSCs), neuroblastoma cells (NB), osteoblasts (OB) and osteosarcoma (OS) cells were used. All data were normalized to beta-actin expression and plotted relative to osteosarcoma (1.0). B) Micrographs of cell lines stained for gamma-H2AX (red fluorescence) before and after exposure to 5 Gy IR. Histograms are shown of scoring of 100 cells in duplicate per sample. The number of foci per nucleus (blue immunofluorescence) was scoreed and the mean and standard deviation are plotted in the histogram on the bottom. The distribution of foci are indicated for the untreated cells and the treated cells are shaded in gray. C) Micrographs of ES-8 cells treated with 5 Gy IR in the presence of 10 micromolar of each PARPi. Gamma-H2AX is shown by red fluorescence and nuclear DNA is indicated by blue fluorescence. D) Histogram of the proportion of gamma-H2AX immunopositive cells (>20 foci per cell) for each treatment group. 100 cells were scored in duplicate and the mean and standard deviation are shown.

Figure S2 related to Figure 2. Sensitivity of EWS cell lines to PARPi+SN-38 and temozolomide and mechanism of action. A) Dose response curves for each PARPi alone (red curve) or in combination with fixed concentrations of SN-38 (blue curve). Each data point is the mean and standard deviation of triplicate wells in each experiment. The gray dashed line indicates the level of cytotoxicity of the SN-38 alone. B) Dose response curves for each PARPi alone (red curve) or in combination with fixed concentrations of temozolomide (blue curve). Each data point is the mean and standard deviation of triplicate wells in each experiment. The gray dashed line indicates the level of cytotoxicity of the temozolomide alone. C,D) Immunoblot for the nuclear soluble and chromatin bound proteins in cells treated with10 micromolar PARPi+ 0.01%MMS for 2 hours. A second group was treated with 10 micromolar PARPi for 2 hours, washed and then exposed to 0.01% MMS for 2 hours. A third group was treated with 10 micromolar PARPi, washed and then exposed to 0.01% MMS for 2 hours and then washed again before harvesting for analysis. Topoisomerase I is used as a positive control for the nuclear soluble fraction and histone H3 is used as a positive control for the chromatin bound fraction. E) Histogram showing the quantitation of the bands in the blots in A,B above. The normalized relative fold level of PARP1 in each fraction is plotted.

Figure S3 related to Figure 3. PK and PD of PARPis in EWS. (A-C) Concentration versus time plot for BMN-673, olaparib, and veliparib in plasma after a single oral dose as indicated. (D-F) Concentration versus time plot for BMN-673, olaparib, and veliparib in the orthotopic EWS tumor after a single oral dose as indicated. The gray box indicates the minimum level of drug concentration maintained for 12 hours in the orthotopic tumor. (G-L) Response surface model fits for PARPi combined with TMZ (G-I) or SN-38 (J-L). ‘Relative survival’ is the reduction in CellTiter-Glo signal relative to the negative control. Kappa values reflect the sign and magnitude of the drug-drug interaction: κ <0 indicates antagonism, κ =0 indicates Loewe additivity, and κ >0 indicates synergy. A90 values quantify the proportion of the area of the response surface defined by concentrations from 0 to Cmax for each compound with efficacy >90% (10-fold reduction in survival) The maximum value of A90 is 1.0. (M-O) In vivo PD of PARP inhibition in an orthotopic EWS tumor at different timepoints after a single oral dose of BMN-673, olaparib, or veliparib. Each point represents the mean ± SD of triplicate tumors from different mice.

Figure S4 related to Figure 4. Histological analysis of treated and untreated orthotopic xenografts. A-E) Micrographs of H&E staining of placebo (A), IRN+TMZ (B), veliparib+IRN+TMZ (C), olaparib+IRN+TMZ (D), BMN-673+IRN+TMZ (E) tumors. F-J) High magnification micrographs of the tumors shown in A-E. K-O) Immunohistochemical analysis of Ki67 (brown) for each of the tumors shown in A-E. Scale bars: 50 microns.

Figure S5 related to Figure 5. Pharmacokinetics and preclinical phase II efficacy in esterase deficient mice. A,B) Plot of tumor burden (xenogen signal) for individual mice from a preclinical phase II study of ES6 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of olaparib+IRN+TMZ(50%) and BMN-673(80%)+IRN+TMZ(30%). C,D) Plot of tumor burden for individual mice from a preclinical phase II study of EW8 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of olaparib+IRN+TMZ(50%) and BMN-673(80%)+IRN+TMZ(30%). E) Concentration versus time plot of plasma levels of SN-38 in wild type (red) and esterase deficient mice (blue) following an IP injection of irinotecan. F) Xenogen signal for orthotopic ES-8 tumors in immunocompromised esterase deficient mice treated with olaparib alone (black line) or olaparib+IRN (red line). G) Xenogen signal for orthotopic ES-8 tumors in immunocompromised esterase deficient mice treated with BMN-673 alone (black line) or BMN-673+IRN (red line). H) Plot of tumor burden (xenogen signal) for individual mice from a preclinical phase II study of ES8 cells with standard of care (IRN+TMZ(50%)) versus triple drug combinations of veiliparib+IRN+TMZ(50%) at a high dose of veliparib (62.5 mg/kg BID x5 x2).

Table S1 related to Figure 1. Expression of DNA repair pathway genes.

Provided as a separate file.

Table S2 related to Figure 2. Drug sensitivity data.

Provided as a separate file.

Table S3 related to Figure 4. CBC-D data for preclinical testing.

Provided as a separate file.

Table S4 related to Figure 5. Preclinical Phase III data.

Provided as a separate file.

NIHMS667197-supplement-01.pdf (20.3MB, pdf)

RESOURCES