Screening for neuroblastoma in late infancy by use of EIA (enzyme-linked immunoassay) method: 115000 screened infants in Austria

Biochemical Diagnosis of Catecholamine-Producing Tumors of Childhood: Neuroblastoma, Pheochromocytoma and Paraganglioma

Catecholamine-producing tumors of childhood include most notably neuroblastoma, but also pheochromocytoma and paraganglioma (PPGL). Diagnosis of the former depends largely on biopsy-dependent histopathology, but this is contraindicated in PPGL where diagnosis depends crucially on biochemical tests of catecholamine excess. Such tests retain some importance in neuroblastoma though continue to largely rely on measurements of homovanillic acid (HVA) and vanillylmandelic acid (VMA), which are no longer recommended for PPGL. For PPGL, urinary or plasma metanephrines are the recommended most accurate tests. Addition of methoxytyramine to the plasma panel is particularly useful to identify dopamine-producing tumors and combined with normetanephrine also shows superior diagnostic performance over HVA and VMA for neuroblastoma. While use of metanephrines and methoxytyramine for diagnosis of PPGL in adults is established, there are numerous pitfalls for use of these tests in children. The establishment of pediatric reference intervals is particularly difficult and complicated by dynamic changes in metabolites during childhood, especially in infants for both plasma and urinary measurements, and extending to adolescence for urinary measurements. Interpretation of test results is further complicated in children by difficulties in following recommended preanalytical precautions. Due to this, the slow growing nature of PPGL and neglected consideration of the tumors in childhood the true pediatric prevalence of PPGL is likely underappreciated. Earlier identification of disease, as facilitated by surveillance programs, may uncover the true prevalence and improve therapeutic outcomes of childhood PPGL. For neuroblastoma there remain considerable obstacles in moving from entrenched to more accurate tests of catecholamine excess.

Diagnostic and prognostic impact of urinary catecholamines in neuroblastoma patients

BACKGROUND

Urinary catecholamine metabolites are well-known to be elevated in patients with neuroblastoma. Some investigators have described different patterns in favorable and unfavorable cases. However, extended studies have not been published.

PROCEDURE

We investigated urinary catecholamine patterns and their correlation to stage, biological features, and outcome in 114 consecutively clinically diagnosed neuroblastoma patients.

RESULTS

Sensitivity of vanillylmandelic acid (VMA), homovanillic acid (HVA), and dopamine (DA) was 80.7, 71.9, and 61.3%, respectively. In 91.2% of patients at least one parameter was above normal. High VMA levels were associated with favorable biological features, high DA levels were predominantly found in biologically unfavorable disease. Whereas patients with normal HVA levels had a significant better outcome, the other parameters showed no significant association with prognosis. For disseminated neuroblastoma of infancy, DA/VMA ratio proved to be helpful for the discrimination of stage 4 versus stage 4s.

CONCLUSION

Urinary catecholamines appear to be useful to give a first but important hint about the biological behavior and thus the prognosis of the underlying disease. Particularly DA/VMA ratio may serve as a tool for "biological grading"-especially in disseminated disease of infancy. In addition, it may be speculated that HVA negativity and low DA/VMA ratio may be helpful for the decision of a "wait and see" strategy in selected neuroblastoma patients with localized disease.

Neuroblastoma Mass Screening in Late Infancy: Insights Into the Biology of Neuroblastic Tumors

Purpose: Neuroblastoma screening in early infancy has detected predominantly “favorable” tumors. We postponed screening to an age between 7 and 12 months to test whether this shift of screening age might influence the detection rate of genetically/clinically unfavorable tumors.

Patients and Methods: In a 10-year period, 313,860 infants were screened by analysis of urine catecholamines. When a neuroblastoma was diagnosed, at least two different areas from every tumor were analyzed for genetic features (MYCN amplification, 1p status, ploidy). Furthermore, neuroblastoma incidence and mortality of the screened group and the cohort of 572,483 children not participating in the screening program were compared.

Results: Forty-six neuroblastomas were detected by mass screening. In 17 tumors (37%) at least one of the biologic features was “unfavorable.” In 10 of 17 patients, one or more of these alterations were only focally present (tumor heterogeneity). In the screened cohort, neuroblastoma incidence was significantly higher when compared with unscreened children (18.2 v 11.2/100,000 births), while there was a trend towards lower incidence of stage 4 over 1 year (2.2 v 3.8). Mortality was not significantly different (0.96 v 1.57).

Conclusion: In contrast to other neuroblastoma screening programs, more than one-third of patients were found with unfavorable genetic markers in our study. The high proportion of focal alterations suggests that biologically young neuroblastomas may consist of genetically favorable and unfavorable parts/areas/clones. We conclude that at least one-third of neuroblastomas detected by screening in late infancy are anticipated cases. This, however, does not result in significantly reduced mortality.