Abstract
Background/Aim: Metastatic Ewing sarcoma (EWS) causes high mortality. Precision medicine can help to mitigate dismal outcomes by targeted and more effective eradication of cancer cells. The aim of the study was to propose a targeted immunoliposome (IL) that has a potential of selectively binding to and destroying EWS cells. EWS expresses CD99 used for diagnostics. If CD99 is used for targeted IL, its payload must be more damaging to EWS than normal cells, which may also express CD99. Poly(ADP-ribose) polymerase (PARP) inhibitors are potential payload candidates, since they may enhance apoptosis of EWS-cells treated with DNA-damaging agents.
Materials and Methods: The R2 genomics platform was used to explore the differential expression of CD99 and PARP1 in seven different databases (two EWS and five various normal tissues). Kaplan-Meier analysis was conducted for prognostic significance of PARP1 expression. The side-effect profile of PARP inhibitors allowed selection of a potential payload for proposed IL.
Results: EWS samples (Savola/Surdez, n=196) had higher CD99 expression than 737 normal tissues. Endothelial compartment had twice-higher CD99 compared to other normal tissues but lower CD99 and PARP1 than EWS. Recurrent/metastatic EWS expressed more PARP1 than primary tumors (ANOVA p=0.02, Savola). Ten-year survival respectively for low versus high PARP1 expression was 36% vs. 14% for EFS (p=0.016), and 50% vs. 7% for OS (p<0.001). Normal hematopoietic/B-cell compartments had ≥2-times higher PARP1 than other tissues; therefore, niraparib (least lymphotoxic of PARP inhibitors), was favored as a payload for anti-EWS IL.
Conclusion: Immunoliposomes covered with anti-CD99 mAbs and loaded with small-molecule niraparib may be developed as an adjuvant therapy for advanced EWS. Projected 100 nm IL should provide vascular permeability and tumor tropism; however, extensive preclinical evaluations will also be required regarding hematopoietic and endothelial damage.
Introduction
Metastatic and recurrent Ewing sarcoma (EWS) poses significant mortality risk in children and adolescents (1). Advances in precision medicine can help to mitigate dismal outcomes by designing targeted delivery of drugs for more effective eradication of cancer cells (2, 3). This study aimed to identify a rationalized combination of antibody-covered immunoliposomes (IL) loaded with a small molecule against EWS. CD99 antigen is typically expressed on EWS cells and is not only used for diagnostics but was also proposed as a potential target for EWS cells (4, 5). However, CD99 is also expressed on normal tissues; therefore, if CD99 is used for targeting a nanoparticle, its payload must be more damaging to EWS than normal cells. FDA-approved PARP inhibitors may present such an opportunity, since EWS is treated with DNA-damaging agents and PARP inhibition may enhance EWS cell apoptosis (6, 7).
Figure 1 depicts the study flow chart detailing the aims as well as the expected results of this study. First, we aimed to confirm the differential expression of CD99 and PARP1 in EWS compared to normal tissues. Next, we examined PARP1 expression in primary versus recurrent and metastatic EWS. Lastly, we analyzed the potential effect of PARP1 expression on patients’ survival – all of which combinatorically rationalize validity of the proposed IL.
Study flowchart of aim to expected results for anti-EWS IL. EWS: Ewing sarcoma; IL: immunoliposome.
Materials and Methods
R2 genomics analysis platform (8) was used to explore the potential design rationalization of CD99-covered and PARP inhibitor-loaded IL. Seven different databases were used for differential expression of CD99 and PARP1: two EWS databases (Savola and Surdez) and five normal tissue databases (B cells, endothelial cells, hematopoietic, lymphocytes, and various normal tissues). Kaplan-Meier analysis was conducted for prognostic significance of PARP1 overexpression in one of the EWS datasets (Savola) containing survival data in R2 for patients with EWS. Lastly, comparative side effect profiling of variable PARP inhibitors (9) was analyzed and utilized for selection of a candidate payload molecule for the presumed IL.
Results
In two EWS datasets (Savola, n=117 and Surdez, n=79) CD99 expression was two to three times higher than in five sets of normal tissues (total n=737; Figure 2). Endothelial compartment had approximately two times higher CD99 expressions compared to other normal tissues (Figure 2), but lower CD99 and PARP1 than EWS (Figure 2 and Figure 3). Also, normal hematopoietic/B-cell compartments had about two times higher PARP1 than other datasets (Figure 3). Recurrent and metastatic EWS had higher PARP1 than primary disease as seen in Savola dataset (p=0.02, ANOVA; Figure 4).
Average CD99 expression for Ewing sarcoma and normal tissues.
Average PARP1 expression for Ewing sarcoma and normal tissues.
Average PARP1 expression for primary, recurrent, and metastatic Ewing Sarcoma in the Savola dataset.
Higher PARP1 expression was associated with worse event-free survival and worse overall survival, further validating potential role of PARP inhibition in EWS (Figure 5A and B). Ten-year survivals respectively for low versus high PARP1 expression were 36% and 14% for EFS (p=0.016), and 50% and 7% for OS (p<0.001).
Event-free survival probability (A) and overall survival probability (B) for high and low PARP1 expression for patients with Ewing sarcoma in the Savola Dataset.
Discussion
Overall results support potential feasibility of our proposed IL. Anti-CD99 Ab linkage and PARP inhibition load may provide sufficient tropism and relatively targeted killing of EWS cells. Although this IL may bind to endothelial cells more than other normal tissues (but less than EWS cells), it may not be detrimental since PARP1 expression is not high in endothelial compartment. In contrary, although with less CD99 expression, hematopoietic/B-cell compartments maybe more sensitive to PARP1 inhibition due to higher expression. Therefore, for IL payload it is crucial to select an inhibitor with less lymphotoxicity.
Small molecule niraparib (MW=320.4 g/mol) is an FDA-approved PARP1/2 inhibitor which causes less lymphopenia compared to other PARP inhibitors (9), thus it would be a preferred candidate for the suggested nanoparticle. The size of niraparib molecule and relatively hydrophilic structure should allow an ample amount to be packaged into a 100 nm IL. Niraparib is used for ovarian, fallopian tube and peritoneal cancers and its liposomal packaging is possible as preclinically tested for ovarian cancer (10). Niraparib can be loaded into IL followed by PEG-linked attachment of monoclonal antibodies (mAbs) against CD99 as demonstrated in Figure 6. This structure should have good vascular permeability and tumor tropism, based on the literature review of preclinical work for other but relevantly similar immunoliposomes, such as for neuroblastoma (11). Another work by Urey et al. describing the successful development and characterization of a MUC4-targeted immunoliposome showed targeted affinity and an anti-proliferative effect against pancreatic ductal adenocarcinoma (12).
Model of proposed niraparib packaging, anti-CD99 mAb cell-surface binding, and PARP inhibition in Ewing sarcoma cells.
Conclusion
ILs covered with PEG-linked anti-CD99 monoclonal antibodies and loaded with PARP inhibitor niraparib may be developed as an adjuvant therapy for metastatic and recurrent EWS. Extensive preclinical testing will be required to ensure acceptable hematopoietic and endothelial damage as seen with other targeted therapies (13).
Acknowledgements
Authors are thankful to the team of Dr. Koster from Amsterdam for the following website: ‘R2: Genomics Analysis and Visualization Platform (http://r2.amc.nl)’.
Footnotes
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
Authors’ Contributions
All Authors contributed to writing and editing the manuscript. The first author conducted the initial literature review, data extraction, analyses and drafted the manuscript with the second author. They received further detailed and critical mentorship from Dr. Lasky for re-writing the manuscript into its presented form.
Funding
No funding.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received June 19, 2025.
- Revision received July 13, 2025.
- Accepted July 25, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.