Abstract C004: Endpoint readouts in a zebrafish larval xenograft platform for in vivo chemosensitivity testing in neuroblastoma

Abstract

Neuroblastoma is the most common extracranial solid tumor of childhood and accounts for ∼15% of pediatric cancer mortality, with pronounced cellular heterogeneity driving variable treatment response. Because of this, there is a need for tractable in vivo platforms for individualized drug testing. Here, a zebrafish larvae pipeline was evaluated as a rapid, low-cost xenograft model to quantify neuroblastoma response to doxorubicin, comparing the sensitivity of imaging- versus single-cell–based readouts. Far-Red–labeled SH-SY5Y neuroblastoma cells were characterized (CD44+, CD133 weak, CD15–, CD70–; doubling time ∼24 h, equivalent at 33°C and 37°C) and microinjected (∼100 cells) into the yolk sac of 72-hpf larvae. Doxorubicin concentrations (0.1 and 1 mg/L) were selected from in vitro cytotoxicity and larval toxicity assays; ≥10 mg/L was excluded for reducing survival below 30% at 24 h. After baseline light-sheet fluorescence microscopy (LSFM) imaging at 48 hpi, larvae were immersed for 48 h, and tumor response was quantified by relative tumor-area change by LSFM with a custom Python segmentation pipeline, and residual tumor-cell burden by enzymatic dissociation and flow cytometry at 96 hpi (n=7 per group). In vitro, doxorubicin reduced viability at 1 and 10 mg/L (P=0.005). Wild-type zebrafish larvae survived treatment with 0.1 and 1 mg/L (survival ≥94% at 36 h). Engrafted cells proliferated in vivo, completing three and five division cycles at 72 and 96 hpi, evidenced by Far-Red dilution, and xenografts persisted through 144 hpi. Tumor-area reduction did not differ among groups (median 63.3%, 39.1%, and 63.6% for 1 mg/L, 0.1 mg/L, and control; Kruskal–Wallis H=3.33, P=0.19). In contrast, cell-number cytometry readouts revealed a lower residual tumor burden in treated larvae, with averages of 12.4, 5.4 and 5.9 cells per larva in controls, 1 mg/L and 0.1 mg/L, respectively, a difference significant for both absolute count (H=6.35, P=0.04) and proportion (H=6.86, P=0.03). These findings show that zebrafish larvae provide a feasible individualized xenograft platform for neuroblastoma drug testing, with the single-cell flow cytometry readout more sensitive than tumor-area imaging in detecting doxorubicin response, likely because area-based imaging misses cytostatic effects and is confounded by xenograft compaction and dye dilution.