Recently, I received an email from the EMA stating that they acknowledge the receipt of my request and notify me of the start of the procedure. First I didn’t remember requesting anything from the EMA and had no idea what the email was about. It took me a couple of reads of the email when it all came back to me.

On 19th of July 2021, exactly 2 years and 3 months before, I requested to receive information about an industry survey conducted by the EMA. Thinking now, I probably should have remembered the request because I have written two blog posts where I complain about not having received the response yet.

What was the request about? I had requested the results of an industry survey. The results have been used by the EMA as a reference not to include local literature sources to their medical literature monitoring (MLM) service.

Why do I want to see the results? Basically, to prove a point. The point being that the survey results do not indicate that the industry would want to have the local literature screening included in the MLM service, because the survey does not ask about it.

The bottom line is that the EMA is not looking to improve the MLM service or at least in practice they are doing close to nothing. There are a few important aspects to this matter. Read further to learn more about what those issues are.

Flawed industry standards

When asked about adding local literature screening to the MLM, EMA told us back in early 2021 that nobody in the industry wanted changes based on the previous MLM survey sent to the marketing authorization holders.

We did not swallow this explanation, because there’s practically no one in the industry who would not be willing to pay for a service or automation that takes care of local literature screening. Then why nobody had stated that in the survey?

Lack of Transparency by the Authorities

The first issue related to this process, is that the survey results have not been published online. I had to send a request to receive the survey results by contacting EMA with their request for access to unpublished document form.

Asking the Wrong Questions

Without seeing the survey results yet, I assume the survey did not give the opportunity to express desire for additional scope or improvements to the MLM. EMA has since performed at least one more survey of this kind and they have also had a call for tenders for MLM services.

Lack of Interest in the Process

Tepsivo tried to publicly rally industry stakeholders to use this survey to express their dissatisfaction with the service that leaves companies still performing duplicate searches and reporting duplicate adverse events to Eudravigilance. However, looking at the EMA statistics only would indicate that maybe reducing costs and improving processes is not as dear to most of the companies as it is to us, because only 143 of about 4,000 (3.5%) marketing authorization holders answered the questionnaire.

This can be hard to understand because the cost, paid by the marketing authorization holders in the form of pharmacovigilance fees, and potential cost-savings are not insignificant. Could there be a certain amount of cynicism related to the potential effect that answering a questionnaire could have in the activities of the medicines agency? Perhaps there is a disconnect in the organizations between the QPPVs who were responsible for answering the survey and the budget holders, who are responsible for paying the bills? Maybe the EU QPPVs just don’t have the time to answer extra questionnaires?

Timeliness

Based on the the survey result request, one obvious issue is time. It takes the European Medicines Agency very long to get anything done. If responding to a simple request to receive unpublished documents, which probably should have been published in the first place, one can imagine how long it takes with more complex things.

Now, taking a quite bold leap from my document request to larger and much more important matters such as medical literature monitoring service and signal detection. MLM has been fully operational for 8 years now, but there have been no improvements to the service. The signal detection pilot started more than 5 years ago and was supposed to be over by the end of 2021, but it has been extended once again (because the process is useless and the legislation needs to be amended) to continue until the end of 2024.

Is there a fix?

Providing MLM service is one of the key responsibilities assigned to EMA in the European medicines legislation. Instead of actually developing a truly useful and effective service, EMA has opted to tick the box and outsource the service to a commercial organization whose only motivation is to charge as much as possible for the service.

By itself, there’s nothing wrong with this kind of approach, but the customer who is paying for the service (marketing authorisation holders) should have a choice or at least a say in how the process works. And they do! The survey, which 96.5% of MAHs did not respond to.

How to improve literature monitoring in the Fantasy World

The obvious root cause solution to this issue would be to amend the European pharmacovigilance legislation and to remove, amend, or clarify the literature monitoring requirements, because the current legislation says both too little and too much how literature monitoring should be done.

A lot can be learned from scientific literature and looking at data produced by other entities, for example healthcare registries, databases and even data generated by AI models. The EU regulation on pharmacovigilance recognizes that EMA should develop the process and monitoring service to meet the standards of the modern world.

The legal text was written almost 15 years ago, so probably the authors had something different in mind than what I have in mind now when I think about modern technologies for sharing and analyzing information. The directive on pharmacovigilance basically only refers to EMA’s responsibility to set up MLM and that MAHs must monitor literature outside of the MLM. This is the main issue.

The legislation should say that EMA will be responsible for setting up a service to monitor literature for all authorized products centrally and no additional screening is required. The service would cover as many relevant sources as possible and the stored data would include everything relevant to the monitoring of the benefit-risk profile. This would mean that the service could be relied on for the local literature screening, periodic safety reports, and for signal detection, at least for the commonly used active pharmaceutical ingredients.

Now, even when MAHs pay the mandatory fees for MLM, they are still required to screen the same literature (and local literature) to meet the legal requirements and expectations of the authorities.

How to solve the issue in the Real World

At Tepsivo, we’re part of the pharmacovigilance service industry whose main business is to support pharmaceutical companies to meet the current legal requirements for pharmacovigilance, but we have publicly stated that it would be best to get rid of the legal requirement for literature screening and signal detection outside of the centralized services. The most sensible option is to change the legislation.

As we are more than slightly skeptical about any possible changes to the legislation, we have built Tepsivo Literature. Tepsivo Literature uses simple everyday automation and does the mundane retrieval of relevant articles from thousands of sources and performs keyword searches on behalf of the highly qualified local pharmacovigilance experts who are traditionally used for the manual local literature screening.

Since my previous blog post about the cost-efficiency of MLM, our team has developed a tool that could easily automate not only 80% of the current MLM service, but also cover all the authorized (and non-authorized) medicinal products, and medical devices, and perform the searches in ten times more sources (including local literature, of course) than the current MLM covers. And this can be done for a fraction of the cost that the current limited MLM costs to the industry.

Sounds too good to be true?

Unlike some, Tepsivo truly aims to be transparent related to our processes and pricing. We invite anyone to contact us for a demonstration of our literature screening capabilities. We can demonstrate the system’s capabilities by automating your global and local literature searches and you can compare the results of the automated search to your current manual process.

It should be noted that you can immediately go and check the price of the service on our website.

Why 80% automation and not 100%?

Unlike some, Tepsivo truly aims to be transparent related to our processes and pricing. We invite anyone to contact us for a demonstration of our literature screening capabilities. We can demonstrate the system’s capabilities by automating your global and local literature searches and you can compare the results of the automated search to your current manual process.

It should be noted that you can immediately go and check the price of the service on our website.

Conclusion

In 2021, EMA reviewed the scope of MLM and they had a public call for tenders on the service. The process included the aforementioned survey to the QPPVs of MAHs. We described the survey process in our previous blog explaining the situation with local literature screening and MLM. Based on our analysis of the publicly available information, it looks like EMA ended up keeping the service exactly the same and they contracted both companies who submitted their proposals and awarded both Syneos and IQVIA with €15,200,000 contracts over 6 years. This means that EMA will be spending over €30 million during 2022 – 2028 years or around €5 million per year on MLM and reporting ICSRs to EudraVigilance. In practice it looks like there is not going to be a change to the situation for another 4 years.

So in the meantime, what should you do? Check what kind of processes you have related to literature monitoring in your organization and think. What is the value of this process? Could this be outsourced or automated?

Going back to the beginning of the text and the request for unpublished data from the EMA. What is the point of it? To whom do I want to prove a point? I already think I know what the data will be and most likely it will not change anything in practice. It may seem insignificant but receiving the information shows that at least this process (request for unpublished data) works and we are all able to get information and answers from EMA, even if it might take forever and the answers are not to our liking. And maybe, if this process works, the other processes might work as well and next time around, maybe there will be steps towards modernization of the medical literature monitoring service.