But why has AstraZeneca created this focus on personalized medicine? It's the right thing for the patients, Miller explained. It's also a pragmatic approach, he added--once it's possible to show that a drug works in a patient group, this demonstrates its value and helps it penetrate the market.
But what about the patients with disease variants where personalized drugs and tests aren't available? Will they get ignored in the rush to personalize treatment? Miller provided reassurance: "We just don't know enough yet for these patients, but we are learning more about different diseases--with many, what we once thought was one disease is actually a spectrum."
As an example, around 70 years ago or so, leukemia was thought of as one disease and it was a death sentence. Fifty years ago, researchers split it into leukemia and lymphoma. Twenty years ago these were subdivided even further, and now leukemias can be divided into 70 or more different diseases, and most children are cured. Researchers need to look for biomarkers for the diseases within the spectrum, Miller explained. "It's not the haves and have nots, just the 'not yets.'"
Previously, some drugs have not made it to market because they are effective only within a subtype, and without the biomarkers or the technology for tests, the drug's efficacy could not be elucidated. Biomarkers can be used to design clinical trials, driving a smaller and more efficient study to produce data earlier, leading to lower-cost approval. This could allow drugs to market that might not otherwise have been approved, helping patients who otherwise may not have been treated. As an upside for the company, faster approval also leads to a longer patent life. But could personalized medicine segment the market too much? Miller accepts that this is a challenge, and that it could be addressed by using partnerships, which will make the markets more viable.
"If there is only one drug for an indication, there is no point in having a test, as no test is 100% accurate, and in this situation, the physician is likely to try the drug anyway. However, if there are many drugs for an indication, then physicians do need an answer, and even if the answer is 'not this drug,' the doctor can then move on to another treatment. So, the patient wins, the payor wins because they won't be wasting money on a drug that doesn't work, and the pharmaceutical companies win because they have a market for their drug," said Miller.
Even once biomarkers are discovered, there are still challenges for personalized medicine. Some personalized medicine can require highly sophisticated testing, and these can be carried out in high-end hospitals worldwide quite easily. But, at the next level down, access to testing is more of a challenge, particularly in the timeframe needed for emergency medicine, or for infectious disease identification. Miller suggested that to solve this, testing needs to be brought to the patient, in the form of point of care and near point of care devices, or centralized testing. Physicians will also need to be educated about the kinds of tests that are needed; while they don't always need to know how the tests and personalized medicines work, they need to be confident that it works, the evidence for which will need to be supported by the laboratory testing industry, the medical device industry, and the pharma and biotech industries, too.
At AstraZeneca, 75% of the pipeline is now based around personalized medicine, Miller told FierceBiomarkers. "We believe that this is a better way of developing drugs." So watch this space for AstraZeneca as it drives forward with markers and medicine. -- Suzanne Elvidge (email)