Abstract Introduction and objectives: Malignant effusions (MEs), including pleural, ascitic and pericardiac fluids, occur in 33% of patients with solids tumors, either at diagnosis and/or at progression to therapy. These fluids are very valuable when tumor tissue samples are inaccessible, as circulating free DNA/RNA (cfDNA/RNA) obtained from MEs enable accurate molecular diagnosis. Primary cultures can be established from tumor cells present in MEs, allowing for the in vitro testing of antitumor drugs and being of help in treatment selection due to their similarity with the patient’s tumor. In this study, we aimed to incorporate MEs analysis into the routine clinical practice to determine clinically relevant alterations and to establish 2D and 3D primary culture models to be used in drug testing. Materials and methods: MEs from patients with solid tumors were prospectively collected, with volumes ranging from 5-5,000 mL. cfDNA/RNA was extracted from all fluid samples and subjected to genotyping using a 30-gene DNA next generation sequencing (NGS) panel and a commercial mRNA nCounter panel containing 770 mRNA hybridization probes. In parallel, tumor cells contained in the MEs were isolated by centrifugation of the entire volume, cultured in complete culture media to establish 2D and 3D in vitro models, and molecularly characterized by NGS and nCounter. In successfully established pure tumor primary cultures (PTPC), cell viability assays were performed with chemotherapeutic agents and/or targeted therapies based on the detected alterations. Clinical follow-up data were collected to evaluate prognostic associations. Results We collected 382 MEs from 314 solid tumor patients. Of them, 49.7% (n=190) were collected at therapy progression, 28.3% (n=108) at baseline and 8.4% (n=32) while the patient was still on treatment. In 13.6% (n=52) cases, no data regarding collection time was available. Genotyping of cfDNA/RNA from 334 samples identified clinically relevant alterations in 77.3% (n=258), including mutations, amplifications, and/or fusions at variant allele frequencies (VAF) ranging from 0.33 to 93%. Among these, 35.3% (n=91) were considered targetable. Information on paired cytology samples was available in 11.6% (n=93) cases, showing a concordance rate of 80% with cfDNA/RNA genotyping. PTPCs could be successfully established from 61 MEs, mainly from pleural effusions (70.5%, n=43) and lung cancer cases (54.1%, n=33). Drug sensitivity assays were performed in 47 PTPCs (77%), including chemotherapy and targeted therapies. In 14 cases (29.7%) correlation between treatment received and sensitivity results could be analyzed, showing a 93% agreement. Conclusions: MEs can be prospective collected in the clinical setting and used for a dual purpose. cfDNA and cfRNA can be isolated from the fluid fraction and employed for genotyping, while cells can be cultured and antitumor drugs tested. Citation Format: Elba Marin, Silvia Garcia-Roman, Cristina Aguado, Marta Vives, Sonia Rodríguez, Susana Muñoz, Noelia Armiger, Alejandro Martínez-Bueno, Maria Gonzalez-Cao, Ana Velasco, Serafin Morales, Reyes Roxana, Ekaterina Meshoulam, Andrés Aguilar, Clara Mayo de las Casas, Miguel Ángel Molina-Vila. Malignant effusions for cfDNA/RNA genotyping and drug testing: Turning malignant fluids into personalized cancer models abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1132.
Marín et al. (Fri,) studied this question.