Ovarian cancer (OC) is an umbrella term for cancerous malignancies affecting the ovaries, yet treatment options for all subtypes are predominantly derived from high-grade serous ovarian cancer, the largest subgroup. The concept of "functional precision medicine" involves gaining personalized insights on therapy choice, based on direct exposure of patient tissues to drugs. This especially holds promise for rare subtypes like low-grade serous ovarian cancer (LGSOC). This study aims to establish an in vivo model for LGSOC using zebrafish embryos, comparing treatment responses previously observed in mouse PDX models, cell lines and 3D tumor models. To address this goal, a well-characterized patient-derived LGSOC cell line with the KRAS mutation c.35 G>T (p.(Gly12Val)) was used. Fluorescently labeled tumor cells were injected into the perivitelline space of 2 days’ post-fertilization zebrafish embryos. At 1 day post-injection, xenografts were assessed for tumor size, followed by random allocation into treatment groups with trametinib, luminespib and trametinib + luminespib. Subsequently, xenografts were euthanized and analyzed for apoptosis and proliferation by confocal microscopy. Tumor cells formed compact tumor masses (n = 84) in vivo, with clear Ki67 staining, indicating proliferation. Zebrafish xenografts exhibited sensitivity to trametinib and luminespib, individually or combined, within a two-week period, establishing them as a rapid and complementary tool to existing in vitro and in vivo models for evaluating targeted therapies in LGSOC.

Zebrafish serve as a valuable model organism for studying human genetic diseases. While generating knockout lines is relatively straightforward, introducing precise disease-specific genetic variants by knock-in (KI) remains challenging. KI lines, however, enable more accurate studies of molecular and physiological consequences of genetic diseases. Their generation is often hampered by low editing efficiencies (EE) and potential off-target effects. In this study, we optimized conventional CRISPR/Cas9-mediated homology-directed repair (HDR) strategies for precise KI of genetic variants in zebrafish and compared their efficacy with prime editing (PE), a recently developed technique that is not yet commonly used. Using next-generation sequencing (NGS), we determined KI EE by HDR for six unique base-pair substitutions in three different zebrafish genes. We assessed the effect of 1) varying Cas9 amounts, 2) HDR templates with chemical modifications to improve integration efficiency, 3) different micro-injection procedures, and 4) synonymous guide-blocking variants in the protospacer sequence. Increasing Cas9 amounts augmented KI EE, with optimal injected amounts of Cas9 between 200 and 800 pg. The use of Alt-R™ HDR templates (IDT) further increased KI EE, while guide-blocking modifications did not. Injecting components directly into the cell was not superior to injections into the yolk. PE, however, increased EE up to fourfold and expanded the F0 founder pool for four targets when compared to conventional HDR editing, with fewer off-target effects. Therefore, PE is a very promising methodology for improving the creation of precise genomic edits in zebrafish, facilitating the modeling of human diseases.