Despite a vast amount of research into improving therapies and developing new ones, our ability to treat cancer has not significantly improved over the past few decades. Detecting cancer early — before it has had a chance to metastasize — remains one of the best strategies for reducing cancer deaths, but are we any better at detection than treatment? This month's Focus on early detection addresses the promise and pitfalls of this strategy.

The case for early detection is clearly described by Leland Hartwell and colleagues on page 243. They focus on the steps required to produce a useful population screening test. Michael Pollak and William Foulkes introduce a note of caution, however, in their opinion article on page 297. They acknowledge that screening is a worthwhile endeavour, but highlight the complex biological issues that can affect its success.

The initial transformation events in a target cell can be genetic or epigenetic and can alter gene and protein expression patterns. The ability to detect these changes would facilitate cancer diagnosis at an early stage. Peter Laird discusses ways of detecting epigenetic changes on page 253, and Emanuel Petricoin and colleagues discuss proteomic approaches to biomarker identification on page 268.

These technologies are still far from being translated into the clinic, so what is the current status of early detection? Screening procedures for breast, cervical, prostate and colorectal cancer exist at present, and recent articles — from highlights to reviews to primary research — from the Nature Publishing Group can be found on these and other important issues in the accompanying Web Focus (http://www.nature.com/nrc/focus/earlydetection). This should prove a useful resource for those interested in early detection, and all content will remain free throughout April.