Neuroblastoma screening data. An epidemiologic analysis
S. N. Goodman
Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, Md.
Neuroblastoma is one of the most common malignant neoplasms in children
under age 5 years. Little progress has been made in the prognosis of
advanced stage disease in the past three decades. Since the development in
the 1960s of a simple urine test to detect neuroblastoma metabolites, there
has been hope that mortality from this disease could be reduced by early
detection via mass screening of young infants. Encouraging reports from
Japan on mass screening programs instituted in the 1970s have been
appearing in the medical literature since 1982, resulting in widespread
interest in screening. This article applies standard epidemiologic criteria
for screening evaluations to the Japanese reports. We find that the data
needed to definitively assess the value of screening were not a part of
those reports and that the benefits claimed from the reported data could be
due to overdiagnosis. In addition, the most recent Japanese data, combined
with recent advances in biologic understanding of neuroblastoma, show that
screening at age 6 months may not detect tumors with poor prognoses. Even
if it could, it is uncertain whether those outcomes could be substantially
altered by earlier diagnosis. Although a final verdict on the value of
neuroblastoma screening is not yet possible, these neuroblastoma studies
are an excellent example of how screening results must be viewed with
extreme caution in the absence of age-specific, population-based incidence
and mortality rates.
