Volume 36, Issue 2 pp. 268-273

Prognostic significance of Ki-67 (MIB-1) proliferation index in childhood primitive neuroectodermal tumors of the central nervous system

Michael A. Grotzer MD

Michael A. Grotzer MD

Division of Neurology and Oncology, The Children's Hospital of Philadelphia, Pennsylvania

Search for more papers by this author
Birgit Geoerger MD

Birgit Geoerger MD

Division of Neurology and Oncology, The Children's Hospital of Philadelphia, Pennsylvania

Search for more papers by this author
Anna J. Janss MD, PhD

Anna J. Janss MD, PhD

Division of Neurology and Oncology, The Children's Hospital of Philadelphia, Pennsylvania

Search for more papers by this author
Huaqing Zhao MA

Huaqing Zhao MA

Division of Biostatistics, The Children's Hospital of Philadelphia, Pennsylvania

Search for more papers by this author
Lucy B. Rorke MD

Lucy B. Rorke MD

Department of Pathology, The Children's Hospital of Philadelphia, Pennsylvania

Search for more papers by this author
Peter C. Phillips MD

Corresponding Author

Peter C. Phillips MD

Division of Neurology and Oncology, The Children's Hospital of Philadelphia, Pennsylvania

The Neuro-Oncology Laboratory, Abramson Research Building 515, 3400 Civic Center Boulevard, The Children's Hospital of Philadelphia, PA 19104Search for more papers by this author

Abstract

Background

Primitive neuroectodermal tumors (PNET) of the central nervous system, including medulloblastomas, are the most common malignant brain tumors of childhood. Whereas some patients experience prolonged disease control after surgery and adjuvant therapy, others with tumors that appear comparable will relapse and eventually die from progressive disease.

Procedure

Because proliferative activity may provide a potential correlate of biologic aggressiveness, PNETs of 78 well-characterized patients were evaluated by Ki-67 (MIB-1) immunohistochemistry. Proliferation indices (PI) were determined by counting Ki-67 (MIB-1) positive tumor cells either in the highest staining region (hot spot PI), or in at least 15 randomly chosen fields (random PI).

Results

Twenty-five of 78 PNETs showed amore than twofold higher value of hot spot PI(median 9.3%; range 0.6–56%), compared to random PI (median 5.6%; range 0.2–41.3%). Univariate Cox regression analysis revealed that PNETs with a high hot spot PI had a significantly greater risk of progression and death than PNETs with a low hot spot PI (hazard ratio 1.58, P = 0.04). The hazard ratio remained significant after adjusting for M-stage in multivariate analysis. In contrast to hot spot PI, random PI proved not to be a significant prognostic predictor.

Conclusions

Hot spot PI is a significant and independent prognostic factor in PNETs. Its assessment is uncomplicated, reliable, and may supplement routine histologic examination as a means for improving the accuracy of predicting the biologic behavior of childhood PNETs. Med. Pediatr. Oncol. 36:268–273, 2001. © 2001 Wiley-Liss, Inc.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.