Skip to Content

Science News from Vanderbilt-Ingram Cancer Center

April 16, 2010

Here are the latest science news nuggets from the Vanderbilt-Ingram Cancer Center, first reported in the April 16 issue of the VUMC Reporter‘s Aliquots.

FISHing out lung cancer

Diagnosing lung cancer from bronchoscopy specimens can be difficult using standard microscopic examination of cells (cytology) – in part due to the small amount of specimen that can be collected.

In bronchial brushing specimens from a clinical population, Otis Rickman, D.O., and colleagues evaluated the use of fluorescence in situ hybridization (FISH) – which tags cells with chromosomal abnormalities often seen in cancer – as an adjunct to routine cytology.

In the March 1 issue of the American Journal of Respiratory and Critical Care Medicine, they report that FISH was able to detect an additional 32 percent of lung cancers that were missed by routine cytology. FISH was especially useful for peripheral nodules, detecting an additional 28 percent (tumors greater than 2 cm) to 44 percent (tumors less than 2 cm) of lung cancers. The results suggest that, used in conjunction with routine cytology, FISH testing – which can be performed on specimens with few tumor cells present – can improve lung cancer detection, especially in peripheral lung nodules.

Melissa Marino

Tool helps mine for cancer variations

Understanding how cancer develops and progresses is becoming increasingly dependent on making sense out of the mountains of genomic and proteomic data being generated and compiled in public databases. Several genomic and proteomic databases exist, but differ in variation and disease types, and in scope (genome-wide vs. single gene).

Bing Zhang, Ph.D., Jing Li, Ph.D., and colleagues developed an integrated database – called the Cancer Proteome Variation Database (CanProVar) – by compiling known protein sequence variations (with a focus on cancer-related variations) from these various public sources. As reported in the March issue of Human Mutation, CanProVar contains 8,570 cancer-related variations in 2,921 proteins – and 41,541 non-cancer specific variations in 30,322 proteins. They also show that the database can help reveal functional characteristics of cancer-related variations and proteins bearing these variations.

The authors suggest that CanProVar can serve “as a bridge between genomic data and proteomic studies” for identifying cancer-causative mutations or cancer biomarkers. The database can be accessed at:

Melissa Marino