Companion diagnostics to personalise medicine

@HealthMed What common link is there between confusion, dementia, dyspepsia, heart failure, hypertension, liver failure and stroke (apart of course from excess alcohol in some)? They are all syndromes - clusters of symptoms and physical signs with many possible causes, risk factors and co-morbidities.
Earlier approaches to treating these health-related conditions were largely based on trial on error until the most effective treatment(s) were identified. While pragmatic, in many that may lead to a delay in achieving effective disease control, while exposing some patients to unnecessary risk of adverse drug reactions.
The relatively new term 'companion diagnostics' describes the concept that coupling careful selection of biomarkers of disease and risk factor phenotype, of therapeutic response and early warning of risk of adverse reactions, will provide a more rapid route to safe and effective drug selection and monitoring.
Major current challenges include the relative immaturity of research evidence on cost and clinical effectiveness of combining companion diagnostics with specific therapeutic strategies. A major potential driver to improve systematic assessment of companion diagnostics will come from the extension of the activities of NICE (the UK's National Institute for Health and Clinical Excellence) from pharmaceutical and other treatments into assessment of diagnostic technologies (Diagnostics Assessment Programme), including companion diagnostics. This will be complemented by the role of NICE in commissioning new research where significant gaps are identified. 
A further important challenge is the fragmentation of diagnostic services within clinical health services. There are practical reasons for having multiple local diagnostic capacity for serious acute illnesses for which very rapid access to sophisticated diagnostics is potentially life-saving or critical to minimise preventable complications, from early selection of effective treatment. Where there is the luxury of more than a few hours delay without significant risk to the patient, e.g. while using 'holding' empirical treatment, there needs to be more research on relative effectiveness of local compared with remote lab diagnostic strategies. There are also practical issues to be resolved among biotechnology companion diagnostic developers where joint licensing agreements are needed when multiple diagnostics are indicated from different source providers. And standardisation of testing is needed across national health service and private laboratories to ensure that results of these new diagnostic tests are validated.
And of course when there is a single test, implementation is more straightforward. However where multiple tests are needed within a companion diagnostics portfolio, there will need to be a regular programme of induction and refresher education for prescribers on how best to use and interpret results of this new testing strategy. This will be particularly important for interpreting the currently less familiar genotype based tests.