By studying the cerebellum (the structure in the brain largely responsible for coordinating motor activities) Drs. Alvin Kho and Isaac Kohane at Children's Hospital Boston, Dr. David Rowitch at Dana-Farber Cancer Institute, and their research colleagues in Boston have developed a novel method for comparing the genetic changes associated with normal development in mice with that of the most common malignancy of the pediatric nervous system, medulloblastoma.

"With information derived from the Human Genome Project we now have the ability to easily compare and identify meaningful patterns of gene expression between species such as mouse and human," said Kho, a postdoctoral fellow and the paper's lead author. Such cross-species comparison provides a powerful new tool for understanding the genetic changes associated with human tumor development.

In a developing organ, the pattern of gene expression changes as the individual cells commit to their own specialized functions. By analyzing the changing patterns of expression of more than 2,000 genes in the developing cerebellum in mice and comparing these to genes expressed in human medulloblastomas, the investigators were able to characterize the malignant cells from a developmental perspective.

''These findings have exploited our ability to analyze thousands of independently segregating genetic markers to confirm the classic theories by investigators such as Lobstein and Cohnheim in the 19th century that tumorigenesis recapitulates aspects of development,'' Rowitch said.

Kohane and Rowitch's research is important for two reasons. Firstly, this novel method provides a generalizable framework within which gene expression in development and tumorigenesis can be studied. This means the role of a particular gene in tumor progression can be better understood. Secondly, since this analysis is readily applicable to other tumor types it may be developed as a useful tool for both tumor diagnosis and prognosis.

Indeed, the investigators show that similar findings are obtained when a human squamous cell lung cancer is compared with the developing rodent lung.

''Much work remains to be done to determine whether the 'developmental perspective' can lead to clinically meaningful insights such as advances in diagnosis or our understanding of tumor behavior,'' Kohane said. ''This study provides a foundation to ask these questions and facilitate translation of a large basic science knowledge base into the clinical sphere.''