| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
Nagoya City University Medical School, Nagoya, Japan
Shionogi Research Laboratory, Osaka, Japan
To the Editor-in-Chief:
Using comparative genomic hybridization (CGH) and microsatellite analysis, Inoue et al1 have characterized some of the common genetic abnormalities found in thymomas.2–3 The most frequent genetic abnormality detected was loss of genetic material or LOH on the long arm of chromosome 6.
The advent of high-density oligonucleotide microarray technology, with its capacity to simultaneously monitor thousands of genes, also provides a unique opportunity for high-throughput genetic analysis of a tumor. We have examined and reported differential gene expression in patients with invasive/non-invasive thymoma by means of the Affymetrix Hum95000 array (Santa Clara, CA) Biochip (microarray) method.4
Although we have used the D-chip analysis method in the previous paper cited above,4
we have now changed the analysis method to Gene Spring analysis (Silicon Genetics Co., Redwood City, CA) and found that several genes at chromosome 6 overexpressed in invasive thymoma (Table 1)
. In search of genes involved in the progression of thymoma, we compared gene expression between advanced thymoma (two stage IVa B3 cases) and early thymoma (one stage I A and one stage II B3 case) samples.4
We should mention that the comparative differential gene expression analysis of advanced stage thymoma versus early stage thymoma revealed that four genes had significantly altered levels of expression by twofold or greater at 6q21–24 lesions.
|
In our cDNA microarray analysis, we identified several differentially expressed genes at chromosome 6, of which the potential roles in tumor progression have been described previously. However, we could not say whether those overexpressed genes were mutated or not from our analysis. Inoue et al1 determined that chromosome 6 is a target of frequent chromosomal aberrations in thymoma and suggested the presence of several putative tumor suppressor genes on chromosome 6 that might contribute to the pathogenesis of thymoma. Further MSI or mutation search for these genes in thymoma are warranted to determine whether the relation with tumor progression of thymoma.
We believe that this in vivo functional genomic approach not only provides an evolving opportunity to rapidly and directly monitor in vivo gene expression in human thymoma, but also promises to provide novel insights into fundamental cancer biology. Furthermore, the application of this approach to clinical thymoma specimens may provide a key step to rapid advances in thymoma prevention, detection, diagnosis, and therapeutics.
References
Roswell Park Cancer Institute, Buffalo, New York
Author’s Reply:
Sasaki et al1 have undertaken another step in elucidating which of the plethora of genetic aberrations occurring in thymoma are important in the progression of this disease from early to advanced stages. They examined gene expression patterns of several early and advanced thymomas looking for differences between those two groups. They came out with a list of genes showing different expression levels. However, their results are speculative at best.
The number of cases investigated (as referred to in the above letter) is completely insufficient. To draw conclusions based on the results obtained on four (moreover, heterogeneous) cases does not allow any meaningful statistical analysis. The low number of cases actually precludes any use of statistics. These results, based on the analysis of four cases, seems to belong to the realm of random error. A somewhat different picture emerges looking at their recent publication.1 Here, they focused on glycosylphosphatidyl-inositol (GPI)-anchored glycoprotein (GPI-80) and analyzed its levels in the tumor, thymoma, and in peripheral blood. While the GPI-80 mRNA results for thymoma show huge variation, GPI-80 protein serum levels are more consistent. However, I have doubts about the relevance of the data for the clinician in the real life (the test would have a terrible specificity) given the considerable overlap in values not only between different thymoma stages but also between patients with thymoma of any stage, myasthenia gravis, or normal controls.
The above study shows how important it is to use proper statistical methods when analyzing microarray results. Do not pick a reason to prove retrospectively a favorite hypothesis. A much better way how to find meaningful differences between early and late stage thymomas is to look at differences between signaling pathway activation patterns. Only then it will be possible to elucidate the pathway of thymoma development, the succession of the individual aberrations, and their contribution to pathogenesis. That is what we owe to our patients.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
