Cancer could be caught before it develops

An article published in the journal BMC Medical Informatics and Decision Making describes the creation of the first comprehensive listing and classification of precancers, drawn from the medical literature. Using this classification, the precancers have been organized into groups that share similar biologic profiles and, hopefully, similar treatments.

Precancers precede invasive cancers. They are localized changes in tissue – lesions – identifiable by their morphologic structure. During carcinogenesis, when normal cells are transformed into cancerous cells, it is possible to identify precancers. Treating or removing precancerous cells at this early stage could prevent the prolonged, painful treatment and deaths of cancer sufferers. According to the authors of the article,

”Premalignant lesions are arguably the most important disease entities of modern man. In theory, the successful treatment of precancers would result in the eradication of most human cancers.”

Despite their importance, until now there has been no attempt to produce a list of precancers, or to classify them according to their biological properties. Jules Berman of the National Cancer Institute at the NIH, and Donald Henson of George Washington University, Washington D.C., have produced a first draft of such a classification, which can be downloaded from http://65.222.228.150/jjb/presum.tar.gz.

Berman and Henson worked from the National Library of Medicine’s Unified Medical Language System (UMLS), extracting all of the terms relating to precancers from the ‘Metathesaurus’, and adding an additional 10% from their own knowledge. They identified 568 distinct precancer ‘concepts’, which have been described by over 4700 terms. Often one precancer is described in many different ways, and can have a number of terms that refer to it. The database includes terms from seven languages in addition to English terms.

Precancer concepts and their associated terms were then grouped into six classes, defined by the authors, according to their biological properties. Berman and Henson provide, in the journal article, an in-depth description of precancer, along with the typical characteristics of each class of precancer.

The classification is designed to be fully searchable and linked to other databases. Written in XML (eXtensible Mark-up Language), it was converted into a HTML file so that it can be viewed in a standard Internet browser.

The authors recognize that this first draft is by no means definitive, so have posted the classification publicly and invited review by the medical community