My view is that we are experiencing a pause in the marketplace, if not quite a correction, and we will start to see increased movement again quite soon. I think the pause occurred for a couple of reasons: one) performance did not quite measure up to the claims of some providers, and two) the worldwide financial meltdown took an awful lot of business with it in the form of reigned in development programs and planned studies that never got started to save precious cash. Let's look at the first reason, since this is not an economics blog.
In the first half of the 'aughts' or the decade just ended, I attended a lot of informational meetings and conferences on emerging markets and heard many vendors explain how they had so harmonized their processes, so locked in the regulatory pathways in each country that sponsors could entrust their studies to these vendors safe in the knowledge that the conduct and resulting data would be of the highest quality and the entire experience would be as painless as if the study were being run in the US or EU. Which is many things but painless is not one of them. And which inevitably turns out not to be true anyway. Sponsors experienced some significant buyer's remorse when they discovered that their project manager was not the person who came to do the bid defense; that the data, ethics, and medical practice standards could be very different in actual countries of conduct and required an experienced hand in the design phase of the protocol, not just the conduct and reporting phases; that their data were being collected many time zones away by people who the sponsor had never clapped eyes on. Instead of finding the experience painless, many sponsors felt unease with the entire process even when the data and trial results proved to be of high quality - they didn't quite trust what they saw because they didn't know who was running the trial - there was no relationship. Vendors who were paying attention have now retooled their presentations to highlight the relationship side of the business.
Even the FDA has got on the bandwagon of mistrust. A recent Pink Sheet ($) dated October 4, 2010 published findings by CDER's Office of Biostatistics that showed in 13 of 16 large multinational cardiovascular outcome trials, the measured drug efficacy was lower in the US than outside. Four drugs studied in these trials were not approved specifically because of this 'regional heterogeneity' and nine of the drugs were deemed approvable but required more data. Anecdotally I have seen this type of regional difference myself and have always interpreted it as proof of the quality of the conduct, the high acceptability of the data. FDA says not so fast. Some regional differences in treatment effect may be expected, but too much can arouse concern. The Pink Sheet report cites examples such as the extended release version of metaprolol (AstraZeneca) whose 'effect on mortality was virtually all outside the US"; AZ's ticagralor which demonstrated no effect in US patients, only exUS; and whether anti-epilepsy drugs bring an increased risk of suicidality, where studies showed that outside the US, suicidality was higher. Bridging studies, even PK/PD or surrogate studies may be indicated to explain regional differences that are not entirely due to chance.
photo credit: J. Mardell, 2010
2 comments:
Interesting post, Jacquie.
What do you think is accounting for the regionally different outcomes--any clear genetic differences (e.g. in metabolism)? Were there other outcomes that were more likely attributable to regional differences in the conduct of the trials, rather than in population differences?
Judy Stone, MD
Author, "Conducting Clinical Research: A Practical Guide for Physicians, Nurses, Study Coordinators, and Investigators"
www.conductingclinicalresearch.com
Thanks for reading and commenting, Dr Stone. You raise a good question and the answer is...hard to say. People have been observing these differences for a long time. Years ago, maybe in the 80's I remember conducting a study in East Berlin whose results showed something like a 90+% success rate for the drug with very few AEs, far surpassing any results we were getting in the US. We were at a loss to explain it and joked that our investigators were moonlighting for the Stasi. Because the results were so differentiated and we had not directly monitored the study - it was done by a third party with relatively little direct supervision (those were different times) - we ended up not being able to use the data, which is a shame.
Genetic and/or regional differences such as diet are very real and need to be addressed during the protocol development stage. Sponsors should not be surprised by these kinds of variances or assume that a US/EU protocol can be force fit onto another region. Once these kinds of variances have been controlled for or ruled out, and the sponsor has been diligent about supervising the monitoring and can verify the quality of the data, we should be able to safely assume that the results we get are valid.
Post a Comment