Page 137 - SRPSKO DRUŠTVO ISTRAŽIVAČA RAKA

SDIRSACR Oncology Insights

Since detection/quantification of low levels of disease in plasma is still challenging regarding analytical and clinical
validations, we assessed association of baseline ctDNA presence/level with the clinical course of melanoma.
Patients and Methods: The study included 61 patients with BRAF-positive melanoma in clinical stage III/IV. All patients
received ICI as adjuvant/neoadjuvant therapy. Cell-free DNA (cfDNA) was extracted from plasma samples using the
MagCore Plasma DNA Extraction Kit (1.2 mL). CtDNA was quantified via BRAF V600 mutation using the QIAcuity Digital
PCR System.
Results: Patients with detectable ctDNA more frequently experienced disease progression (p=0.06). The average
fraction of variant allele (VAF) was significantly higher in patients who experienced disease progression (0.15 vs. 0.06;
p=0.02). When VAF of 0.15 was used as cut off (also obtained by ROC analysis), superior duration of progression free
survival and overall survival were observed in patients with VAF < 0.15 (median PFS 13.91 vs. 1.64 months, p=0.02;
median OS 18.91 vs. 7.63 months, p=0.041).
Conclusions: The ctCNA presence/fraction at baseline may be considered as a marker of poor prognosis. In order to
confirm these findings, a larger cohort of patients is needed to allow broader statistical analyses.

Acknowledgments and Funding: Science Fund of Republic of Serbia, ReDiMEL project, grant No 6795

P44

The use of pleural effusion from patients with lung cancer as a source of extracellular vesicles – first
report from the EXPAND-EV project

Miodrag Vukovic1, Lidija Filipovic2, Nina Petrovic , Andrej Zecevic4, Maja Kosanovic5, Miljana Tanic1, Radmila
1,3
Jankovic1, Tatjana Stanojkovic1, Milica Popovic2, Aleksandra Korac6, Sanja Stevanovic7, Milena Cavic1

1Department of Experimental Oncology, Institute for Oncology and Radiology of Serbia, Belgrade, Serbia
2Department of Biochemistry, Faculty of Chemistry, University of Belgrade, Belgrade, Serbia
3Laboratory for Radiobiology and Molecular Genetics, Department of Health and Environment, "VINČA" Institute of Nuclear
Sciences-National Institute of the Republic of Serbia, University of Belgrade, Belgrade, Serbia
4Clinic for Pulmonology, University Clinical Center of Serbia, Belgrade, Serbia
5Institute for the Application of Nuclear Energy, INEP, University of Belgrade, Belgrade, Serbia
6Faculty of Biology, University of Belgrade, Belgrade, Serbia
7Institute of Chemistry, Technology and Metallurgy, Department of Electrochemistry, University of Belgrade, Belgrade, Serbia

Keywords: Carcinoma, Non-Small-Cell Lung, Extracellular Vesicles, Pleural Effusion

Background: Pleural effusion (PE) is a frequent complication of lung cancer (LC), occurring in up to one-quarter of
patients. Despite being routinely sampled for diagnostic cytology, its sensitivity is often insufficient, leaving a need
for complementary molecular approaches. Small extracellular vesicles (sEVs) from PE are enriched in tumor-related
proteins and RNAs and could serve as a form of liquid biopsy. Most studies have relied on ultracentrifugation (UC), but
this method yields low-purity isolates, often unsuitable for proteomic profiling. Improved results were obtained by
combining UC with size-exclusion chromatography (UC-SEC), while ExoProK introduced Proteinase K pretreatment to
reduce contaminants. Nevertheless, clinically feasible and standardized protocols for PE-derived EVs are still missing.
Within the EXPAND-EV EU project, we focused on EV isolation from PE of patients with advanced non-small cell lung
cancer (NSCLC), evaluating methods which emphasize minimal input volume, short turnaround time, and compatibility
with downstream omics.
Materials and Methods: sEVs from PE samples were isolated by three approaches: (1) an in-house spherical porous
methacrylate-based polymer functionalized with VHH antibodies (chromatography-based method, CH), (2) UC, and
(3) a commercial Norgen kit (CK). Vesicles were characterized using nanoparticle tracking analysis (NTA), transmission
electron microscopy (TEM), atomic force microscopy (AFM), and flow cytometry for CD9 expression. Methods were
compared for efficiency, purity, and clinical applicability.
Results: Both CH and CK enabled isolation from as little as 1 mL, while UC required ≥12 mL. CK was the fastest, followed
by CH, whereas UC was time-consuming. CH yielded the highest particle counts and cleanest profiles, with UC producing
the lowest yield and greatest protein contamination. TEM, AFM, and flow cytometry confirmed vesicular morphology
and integrity, with CH isolates showing the highest purity.
Conclusions: CH and CK approaches are practical alternatives to UC, enabling efficient EV isolation from small PE
volumes. CH isolates are suitable for proteomics, while CK combines speed with RNA recovery, supporting transcriptomic

122