Sir

Two conclusions in your Business story 'Missing the mark' (Nature 449, 770–771; 2007) about the usefulness of cancer biomarkers should evoke a response from translational researchers and clinicians.

First, you say that the overall impact of early bladder-cancer detection on patient survival rates may be relatively small because surgery remains the treatment of choice. This may be why survival from bladder cancer has barely changed during the past two decades: more accurate early-detection tools, such as biomarker tests, are needed. Since the 1980s, the American Cancer Society has issued guidelines and recommended several early-detection tests because of their clinical benefits. These have led to an increased likelihood of complete tumour removal and therefore of a better outcome and reduced costs. Public–private partnerships are therefore expediting the development of biomarker diagnostic tools.

Second, you say that, in the absence of new drug therapies, “there isn't a huge incentive for doctors and the health-care insurers that pay for most medical services in the United States to buy the tests”. But this is contrary to established fact. For example, DNA-based testing for human papilloma virus to augment or replace pap smears is widely accepted and reimbursed. Her-2/neu gene testing is necessary for breast-cancer patients undergoing treatment with Herceptin. The US Food and Drug Administration's approval in 2004 of EGFR-pharmDx (which detects colorectal cancers expressing epidermal growth-factor receptors) for treatment with the cancer drug Erbitux led to rapid uptake, with reimbursement by Medicare and private insurers in the United States.

The problem lies not with physician uptake and reimbursement, but with the slow development and validation of accurate tests providing useful information for early detection and treatment.