Journal of Clinical Oncology, Vol 14, 2066-2072, Copyright © 1996 by American Society of Clinical Oncology
Correlation between polysialic-neural cell adhesion molecule levels in CSF and medulloblastoma outcomes
D Figarella-Branger, C Dubois, P Chauvin, B De Victor, JC Gentet and G Rougon
Laboratoire d'Anatomie Pathologique et de Neuropathologie, Hopital de la Timone, Marseille, France.
PURPOSE: To quantify CSF levels of polysialic-neural cell adhesion molecule
(PSA-NCAM) in patients with medulloblastoma (MB) metastasis, to assess the
correlation with other diagnostic techniques (imaging and cytology) and
clinical features, and to determine whether it is a suitable marker to
monitor response to treatment and subsequent follow- up data. PATIENTS AND
METHODS: PSA-NCAM levels were measured using a double-site enzyme-linked
immunoadsorbant assay (ELISA) in 145 samples from 14 controls and 29
patients with MB. Clinical status of patients, imaging, and cytologic data
were available at the time of each lumbar puncture. Medians and ranges for
the 131 pooled PSA-NCAM concentrations were calculated for the MB versus
the control groups, and for MB patients for normal versus abnormal groups
at cytology or imaging, and for four clinical subgroups, respectively. For
patients with MB, three PSA-NCAM measurements that corresponded to
punctures performed during three time periods following surgery were
selected. The kappa measure of agreement was calculated between normal and
abnormal groups at cytology or imaging, and between groups of patients in
remission and refractory, respectively. For the same phases, sensitivity
and specificity of PSA-NCAM and cytology tests and their 95% confidence
intervals (95% CIs) were computed. RESULTS: PSA-NCAM was never detected in
control CSF. PSA-NCAM concentration medians were higher in CSF with
metastatic cells or that corresponded to abnormal imaging than in the
corresponding normal groups (P < .05). The PSA-NCAM concentration median
was significantly higher (P < .05) in CSF from patients refractory to
treatment or who relapsed than from patients in remission. Agreements
between PSA-NCAM and clinical status and between PSA-NCAM and cytology were
excellent during and after treatment. The sensitivity of PSA-NCAM test was
always better than that of cytology, whereas its specificity was lower for
phases that corresponded to more than 1 month following surgery. However,
specificity was 100% for patients refractory to treatment or with relapse.
CONCLUSION: PSA-NCAM measurement appears to be a new biologic marker of
possible use in the management of patients with MB.