A lack of evidence that links diagnostic tests to health outcomes is hindering the growth of personalised medicines in the US, according to a new study.

This shortage of evidence is leading payers in the US to be sceptical about the clinical usefulness of those tests and is negatively affecting the potential of personalised medicines, says the study, which has been published by the Tufts Center for the Study of Drug Development (CSDD).

Companion diagnostics are tests linked to a therapeutic drug that stratify populations into responders and non-responders and indicate the likelihood of adverse events in particular patients. And while the number of personalised medicines and companion diagnostics in use in the US has increased gradually, from a handful in 2001 to several dozen in 2011, the lack of evidence for the clinical usefulness of many current companion diagnostics is a major factor limiting the potential of personalised medicine, says the report. 

"Scientifically, the process of biomarker discovery and validation in general, and parallel development of drugs and companion diagnostics in particular, has been slow,” says Joshua Cohen, senior research fellow at Tufts CSDD and author of the study.

"Additionally, regulatory and reimbursement issues have limited uptake in clinical practice, particularly with respect to companion diagnostics, but also for drugs lacking effective diagnostics," he adds.

Without clinically-useful diagnostics, development of personalised medicines is likely to continue at a relatively slow pace, Dr Cohen forecasts.

The report also finds that the problem with companion diagnostics has led payers in the US to deny or restrict reimbursement of tests, while a minority of payers require documentation to show that a diagnostic test has been conducted prior to prescribing personalised drugs, even when the diagnostic is included on the label.

The study, which is published in the July/August edition of the Tufts CSDD Impact Report, also quotes pharmacogenomic experts as forecasting moderate growth over the next five years in post hoc development of companion diagnostics to personalise already-approved medicines, and in the co-development of companion diagnostics and personalised drugs.