New blood tests based upon genetic testing are starting to be marketed to clinicians to use during diagnosis, or to enhance screening programmes, which may have great potential to improve patient outcomes.

Tests like BCtect for Breast Cancer and ADtect for Alzheimer's, being marketed to clinicians in the UK at the moment, involve single blood samples and are good examples of where gene testing has the potential to improve diagnosis and care. However, we weren't really sure what the full story was, so we started digging into the tests a little further.

The two tests we've looked at are based upon analysis of blood samples for characteristic changes in the activity of selected genes, rather than mapping full sets of genes. Without getting hugely technical, the uniform chemical nature of RNA (a nucleic acid similar to DNA) makes it a good "bio marker" and therefore useful for diagnosis.

The tests extract RNA and then covert it into cDNA (a collection of complementary, or cloned, DNA synthesised from mature RNA) and you pretty much need to be a genetic scientist to fully get to grips with the conversion process. The process takes a few days, so the results take slightly longer than traditional blood testing.

The tests then assume that the over expression of mRNA of a certain type will be associated with an increased risk of disease, in these cases, Breast Cancer or Alzheimer's. The sensitivity (proportion of actual positives correctly identified) and specificity (proportion of negatives correctly identified) of the tests are relatively low though, at around 75% for each test (which implies one in four test results may not be wholly accurate).

One of the basic tenets of professional health screening is that any test should be effective in detecting a disease whose mortality rate can be significantly improved by early detection, thereby justifying the medical intervention required to conduct the tests. We're not confident that's the case with these particular tests, especially Alzheimer's, so we wouldn't currently view the tests as being appropriate for inclusion in a screening programme.

Which then leads us to the suitability of the tests in assisting diagnosis for symptomatic patients. For Breast Cancer, it's debatable whether the blood test taking up to a week to produce results is, on balance, better or worse than a fine needle biopsy, which is a standard and lightly invasive procedure yielding results instantly (but most commonly in a day or two).

The Alzheimer test may be more useful, enabling early intervention to reduce the impact of the disease. Early stage Alzheimer's is notoriously difficult to detect, which implies the test may be really helpful, although it's yet to be clearly established that early intervention positively effects Alzheimer's mortality rates, so the main benefit would potentially be in providing more time for patients' to organise affairs and come to terms with the condition.

The tests are a glimpse into the future direction of diagnosis, however, the key concern currently is the accuracy of the tests, compared with the best alternative approach to current 'gold standard' diagnostic techniques for Breast Cancer and Alzheimer's.

In summary, we're excited about the developments in these areas and the potential to improve diagnosis and care, however, we're going to keep watching until the accuracy of the tests improves.