Introduction: Cancer cell lines (CLs) and patient-derived xenografts (PDXs) are essential for biological and preclinical therapeutic studies. CLs and PDXs can be established from post-mortem neuroblastoma (NB) samples. We assessed the activity of a chemotherapy regimen commonly used for re-induction chemotherapy in 72 NB CLs and 9 xenograft models which were established from either pre-treatment (DX) or post-mortem (PM) samples.
Methods: STR-validated, tyrosine hydroxylase-positive, EBV-negative continuous NB CLs (47 DX, 25 PM) were profiled for 4-hydroperoxy-cyclophosphamide (4-HC, 0-30µM) and topotecan (TOPO, 0-1µM) cytotoxicity with the DIMSCAN assay. Response of 9 SQ xenograft models (7 PDXs, 2 CL-xenografts (CLX)) to cyclophosphamide (CYCLO) + topotecan was assessed in nu/nu mice. Cytotoxicity in vitro of 4-HC+TOPO was assessed for the PDX models and their matching cell lines.
Results: To date, 25 validated, continuous CLs and 10 PDXs have been established from 40 NB PM samples: 65% blood (50-200mL), 12.5% bone marrow, and 22.5% tumor; 25 of 40 placed in culture (63%) generated a CL, 10 of 16 xenografted (63%) generated a PDX. Success with PM samples was higher than observed for 1838 pre-treatment (CLs=6.6%, PDXs=9.7%) and 246 progressive disease (CLs=17.1%, PDXs=2.9%) samples. Both DX and PM models demonstrated heterogeneity in response to 4-HC+TOPO in vitro and CYCLO+TOPO in vivo; some were highly resistant while others highly sensitive. Complete tumor responses after 1-2 cycles were observed in 1 of 2 DX and 2 of 4 PM PDXs. PDXs and CLs established from the same samples demonstrated comparable responses in vitro and in vivo.
Conclusion: CLs and PDXs are readily established post-mortem from heavily pre-treated patients and demonstrated marked heterogeneity in response to CYCLO (4-HC) + TOPO in vitro and in vivo. PM NB models will enable delineation of molecular mechanisms of drug resistance, provide ideal models for testing new agents, and are available at www.COGcell.org.