Failure of iron chelators to reduce tumor growth in human neuroblastoma xenografts.

Neuroblastoma (NB) is a high risk tumor of childhood, and raised serum ferritin is an adverse prognostic factor. The hypothesis that iron chelation therapy impacts tumor status and patient prognosis through changes in iron metabolism has been systematically evaluated here in a xenograft model of human NB. One of two iron chelators was given in seven different regimens to nude mice xenografted s.c. with either IMR-32, an established cell line, or JBN-1, heterotransplanted directly from a patient. Nude mice (a total of 160 in 24 cohorts) were given: desferrioxamine (DFO) by s.c. bolus or reservoir; 1,2-dimethyl-3-hydroxypyridin-4-one (L1), i.p. or orally; or saline. Measurements of mean Hb and liver iron levels were compared with corresponding saline cohorts per regimen as well as for pooled cohorts per agent for both cell lines. For IMR-32 xenografts, significant differences in Hb were achieved with L1 (10.9 g/dl pooled versus 13.7 g/dl controls) and in liver iron with DFO and L1 (235 microg/g and 306 microg/g, respectively, versus 520 microg/g). For JBN-1, the pattern was similar. With L1, H6 was 10.2 g/dl and controls were 11.7 g/dl (individual DFO cohorts were also significant); liver iron with DFO was 303 microg/g, liver iron with L1 was 270 microg/g, and controls were 387 microg/g. Additional therapy prior to tumor injection (67 mice and 10 cohorts) did not increase the depletion. Despite documentation of iron depletion, no reductions in tumor engraftment, latency, or tumor size at end point were achieved in the chelator-treated mice, compared with controls populations. Accordingly, inclusion of these iron chelators in clinical trials for NB appears unwarranted.

Ferritin production and desferrioxamine cytotoxicity in human neuroblastoma cell lines.

In patients with neuroblastoma, elevated serum ferritin is correlated with adverse outcome. In this investigation, three human neuroblastoma cell lines have been characterised in terms of their levels of both intracellular and secreted ferritin and their response to the iron-chelating agent, desferrioxamine (deferoxamine). The cell lines differed markedly in respect of ferritin production as determined by radioimmunoassay. Intracellular and secreted ferritin concentrations for SH-SY5Y and BE(2)-C cells were several fold higher than that determined for IMR-32 cells. IMR-32 cells were most sensitive to desferrioxamine, BE(2)-C intermediate and SH-SY5Y the most resistant to the drug in terms of the respective ID50 values. Combining the differentiating agent retinoic acid with desferrioxamine did not enhance cytotoxicity in the neuroblastoma cells. The present data suggest that neuroblastoma cells secreting a relatively low levels of ferritin may be most responsive to iron chelating agents.