ABSTRACT
References ....................................................................................................534
During the past century, the clinical behavior of human cancer has been predicted using its microscopic appearance. This has been a useful approach because cancer has hundreds of ‘‘faces’’ under the microscope, but unfortunately ‘‘looks’’ alone can only predict general categories of biological behavior [1]. Tumors with a similar histological appearance can follow significantly different clinical courses and show different responses to therapy [2]. For example, prostate carcinoma arising in two patients may look virtually identical under the microscope, but each patient may have a different clinical outcome. Given the wide diversity of tumors, even for those derived from the same tissue, additional methods of classification are urgently required.
