Sir

Your News Feature “Running before we can walk?” (Nature 429, 496–497; 2004) highlights concerns about a new proteomic test for ovarian cancer known as OvaCheck. This test measures breakdown products of proteins in blood serum and looks for patterns that may indicate disease. The appeal of the method, now under scrutiny, derives from its simplicity and potential power; the total number of breakdown products could eventually exceed the number of gene transcripts and proteins, and may therefore provide a subtler correlation between biological events and disorders than regular expression genomics and proteomics. This approach should not be dismissed before it has been fully investigated with the best possible technological means and expertise, and under rigorously standardized conditions.

Experts quoted in the story cite concerns about mass spectrometry, biostatistics, and also environmental, dietary and psychological factors that could produce artefacts in the data. In our experience, serum preparation, handling and storage could be the largest source of artefact in mass-spectrometry readings, especially for archived samples. We are currently investigating the effects of, for instance, different blood-collection tubes and the number of freeze/thaw cycles.

The whole concept is now in danger of being declared null and void, both by some who previously embraced it and by others who lack the adequate experimental experience to participate in the debate. Although unbiased, your News Feature is likely to embolden the critics and reinforce the bleak view of this particular type of proteomic analysis, and perhaps of proteomic profiling in general.

No final assessment can be made at this point of the concept underlying the OvaCheck test. We hope that leaders from the clinical and analytical-chemistry communities will come together and make a concerted effort to refine the procedures before testing them in carefully designed clinical studies. Only then can we establish whether this approach can disclose hidden patterns of disease or whether it is the molecular equivalent of the Emperor's new clothes.