Fax: (323) 667-1123
TrkA expression in peripheral neuroblastic tumors
Prognostic significance and biological relevance
Article first published online: 31 AUG 2004
Copyright © 2004 American Cancer Society
Volume 101, Issue 8, pages 1873–1881, 15 October 2004
How to Cite
Shimada, H., Nakagawa, A., Peters, J., Wang, H., Wakamatsu, P. K., Lukens, J. N., Matthay, K. K., Siegel, S. E. and Seeger, R. C. (2004), TrkA expression in peripheral neuroblastic tumors. Cancer, 101: 1873–1881. doi: 10.1002/cncr.20557
- Issue published online: 1 OCT 2004
- Article first published online: 31 AUG 2004
- Manuscript Accepted: 1 JUL 2004
- Manuscript Revised: 29 JUN 2004
- Manuscript Received: 24 MAY 2004
- National Cancer Institute, National Institutes of Health, Department of Health and Human Services. Grant Numbers: CA 13539, CA60104, CA22794
- The T. J. Martell Foundation for Leukemia, Cancer, and AIDS Research
- peripheral neuroblastic tumors;
- quantitative polymerase chain reaction;
- clinical stage;
- International Neuroblastoma Pathology Classification;
This study was conducted to investigate the prognostic significance and biologic relevance of trkA expression levels in peripheral neuroblastic tumors (pNTs) (i.e., neuroblastoma, ganglioneuroblastoma, and ganglioneuroma).
Levels of trkA expression from a total of 265 pNTs were determined by quantitative polymerase chain reaction analysis with Genescan software. The results were analyzed according to histopathology (favorable histology [FH] vs. unfavorable histology [UH] according to the International Neuroblastoma Pathology Classification) and MYCN tumor status (amplified vs. nonamplified) along with clinical stage and outcomes of the patients.
The levels of trkA expression differed significantly between the group of patients who were alive and well (n = 170 patients) and the group that had progressed or died (n = 95 patients) and between the group that was alive (n = 188 patients) and the group that died (n = 77 patients). However, the trkA expression levels were not independent predictors of clinical outcome when the proportional hazards model contained the known prognostic variables of clinical stage, histopathology, and MYCN status (all tests were done in 196 patients). In the neuroblastoma category (n = 173 tumors), tumors in the FH/nonamplified MYCN subset (n = 112 tumors) expressed higher levels of trkA and showed an age-dependent neuroblastic differentiation: They were classified into either a poorly differentiated subtype (n = 91 tumors; all patients age < 1.5 years at diagnosis) or a differentiating subtype (n = 21 tumors; 57% of patients ages 1.5–5.0 years). Tumors in the UH/amplified MYCN subset (n = 30 tumors) expressed significantly lower levels of trkA and showed very limited neuroblastic differentiation. Tumors in the FH/amplified MYCN subset were very rare (n = 3 tumors) and expressed higher levels of trkA. Tumors in the UH/nonamplified MYCN subset (n = 28 tumors) had trkA levels in a wide range and showed limited neuroblastic differentiation.
For patients with pNTs, levels of trkA expression did not add significant information to prognostic grouping, as defined by the combination of clinical stage, histopathology, and MYCN status. There was a biologically relevant correlation between molecular properties (trkA expression and MYCN status) and histopathologic features of the tumors in the neuroblastoma category. Cancer 2004. © 2004 American Cancer Society.