New, comprehensive tumor classification combines molecular biology and classic pathology

Information about the genetic make-up of tumors should, in the long term, help clinicians decide on the most effective course of treatment for patients with cancer. To be most helpful these molecular data must be incorporated into a tumor classification that includes morphological and clinical information. Jules Berman describes his ideas for a new comprehensive tumor classification in BMC Cancer this week.

Berman, the Program Director for Pathology Informatics within the National Cancer Institute’s Cancer Diagnosis Program, USA, writes: “This classification is simple (reducing the complexity of information received from the molecular analysis of tumors), comprehensive (providing a place for every tumor of man), and consistent with recent attempts to characterize tumors by cytogenetic and molecular features.”

“Traditionally, tumors have been classified by their morphologic appearances,” he continues. “Unfortunately, tumors with similar histologic features often follow different clinical courses or respond differently to chemotherapy.” The alternative, a classification based solely on the molecular characteristics of tumors, would sacrifice the clinical/pathological experiences that inform virtually every clinical decision related to the diagnosis and treatment of patients with cancer.

Both the type of cells from which a tumor is derived and the molecular characteristics of those cells determine how a tumor will behave – information of great importance to clinicians. Berman has organized his classification using these two features.

At the highest level of the classification, tumors are grouped according to the component cells’ developmental history – their ’histogenesis’. For example leukemia would be classified as a mesenchyme-derived tumor. Patterns identified from the gene expression and proteomic data from tumor samples can then be integrated as new groups within this classification.

Berman writes: “Some of these groupings will prove to have a specific biologic feature, for example an increased tendency to metastasize, a higher response to a chemotherapeutic agent or lengthened survival.”

As each tumor is derived from a cell with an individual lineage, it will have a unique position within the classification. The commonly used tumor classifications list tumors by the body site of origin. However, every organ contains connective tissue, vessels, and lymphoid tissue. When you start to list the tumors that can occur in an organ, you find that every organ can develop many of the same tumors. Classifications that are organized by anatomic site become massively redundant, and this becomes a major problem when dealing with many thousands of tumor terms that link to other biologic datasets.

“A good classification system should help drive down the complexity of enormous databases and help us discover relationships among different data elements by assembling data under sensible group hierarchies,” writes Berman. “The most important value of this classification is the disengagement of tumor type and tumor place of origin […] This permits tumors with similar molecular profiles to be classified according to biological attributes rather than anatomical location.”

The classification outline (in XML) and the latest version of the classification (with 55,000 entries) are available with the article as supplemental Open Access documents that can be used or criticized freely by the biomedical community.

This press release is based on the following article:

Tumor classification: molecular analysis meets Aristotle Jules J Berman
This will appear in BMC Cancer, volume 4 on Wednesday 17 March