I believe you have done a great disservice to both clinicians and patients by allowing the conclusion in the recent paper by de Roda Husman and colleagues [1] to be published.

The conclusion that "The addition of CCR5 genotype to currently available laboratory markers may allow better estimation of the clinical course of HIV-1 infection" is not supported by the paper. Although the authors are able to suggest again that heterozygosity for the 32-nucleotide deletion in the C-C chemokine receptor 5 gene (CCR5 Delta32) may be associated with prolonged survival, they give no reason why the test should be ordered. They make no claims of indication for use, clinical benefit, or even clinical importance. Indeed, they note that 12 of their 23 long-term survivors were not CCR5 Delta32 carriers.

Too often, insupportable claims of clinical relevance are thrown off by authors reviewing one or another correlate of clinical disease. This then becomes the trendy-and extremely expensive-laboratory test du jour, which adds nothing but cost to patient care. A similar case occurred with the HIV-1 genotype resistance assays. No clinically interpretable information is available from these tests, yet hundreds of thousands of dollars of patient funds are being spent on the tests each year and possibly ill-advised therapeutic changes are being made solely on the basis of these widely hyped tests.

Even more interesting is the review by Price and Ridker in the same issue [2]. The authors review the published data and go through an exhaustive discussion about when or if screening for the factor V Leiden mutation should be done. Their review is very useful; the paper and recommendation by de Roda Husman and colleagues are not.

The rapid development of new clinical laboratory tests is very important. However, patients and clinicians are ill served when information on the specific use of the test and interpretation of the result is not available. By allowing de Roda Husman and colleagues to publish a conclusion about the use of CCR5 Delta32 in HIV-1 infection without specifically outlining who should have the test and how the result should be interpreted, Annals has done a significant disservice to patients and clinicians alike. Currently, neither CCR5 nor HIV-1 genotyping offers useful clinical information. They should be left in the research realm, where appropriate guidelines for their use can be determined through population studies and where their developers should have the responsibility to define data-derived use and interpretation measures before the tests are commercially available.