Evaluation of catecholamine metabolites, mIBG scan, and bone marrow cytology as response markers in stage 4 neuroblastoma.

BACKGROUND

The early biological response has been proved an excellent predictor in acute lymphoblastic leukemia and nephroblastoma. We asked whether catecholamine metabolites, mIBG scan, and bone marrow evaluation might be relevant response markers in disseminated neuroblastoma.

PROCEDURE

Three hundred sixty-seven unselected stage 4 neuroblastoma patients treated according the German cooperative trial NB90 were entered into the study. Catecholamine plasma and urine levels were centrally determined by gas chromatography/ mass spectrometry. Bone marrow cytology and mIBG scans were evaluated by local investigators.

RESULTS

At diagnosis, mIBG scan was positive in 306 patients (92%), borderline in seven patients (2%), and negative in 19 patients (6%). Bone marrow aspirates were cytologically positive in 292 patients (84%) and negative in 57 patients (16%). Plasma catecholamine levels were elevated in 79% (206 of 260 patients.), urinary levels in 91% (307 of 338 patients). The outcome of patients with normalized mIBG scan after four courses of chemotherapy [5 year EFS (event free survival) 0.22 +/- 0.07] was not superior to the outcome of patients with still abnormal uptake (5 year EFS 0.30 +/- 0.05). The event free survival of patients with still positive bone marrow aspirates after four courses (0.16 +/- 0.06) was inferior to the EFS of patients with negative bone marrow aspirates (0.26 +/- 0.04, P = 0.0054). Urinary catecholamine normalization after four cycles of chemotherapy (5 year EFS 0.35 +/- 0.06 versus 0.26 +/- 0.10) had no influence on outcome, whereas plasma catecholamine normalization after the first (5 year EFS 0.40 +/- 0.09 versus 0.14 +/- 0.07, P= 0.0364) or the fourth cycle (5 year EFS 0.35 +/- 0.06 versus 0.26 +/- 0.10, P = 0.0242) indicated a better outcome.

CONCLUSIONS

These data show that serial plasma catecholamine levels and bone marrow aspirates in the course of the disease are useful tools in predicting outcome.