Bringing global trials teams together

pharmafile | February 23, 2009 | Feature | Research and Development |  clinical trials, global 

The world is getting bigger and smaller at the same time. When the internet began to really catch on about a decade ago, I was immediately struck by how it was a global community. But I was wrong – I was only thinking about the developed countries that had the technology.

Now all that has changed, as countries which we previously thought of as remote are now well connected. Singapore is an example of one country which got itself properly wired up faster than we did in the UK.

Despite the revolution which the internet has brought about in so many walks of life, it hasn't yet transformed clinical trials.

It still takes around 10 years to get a new chemical entity out of the laboratory and onto the market – the same as it did 10 years ago. Clinical trials are the single most expensive stage, and costs continue to rise while time scales fail to improve. These are among the drivers towards globalisation.

Why trials are more commonly run in several countries these days is pretty clear. Prominent among the reasons is the rising pressure from regulators for more patients.  Recruitment into trials in Asia can be substantially higher, and no project planner should ignore the potential.

But a caveat is that you may only obtain regulatory approval to run your trial there on condition you don't disproportionately recruit Indian patients compared with other countries.

Of course, there may easily be other reasons why you might want some degree of balance between recruitment in different countries.

Quite apart from recruitment issues, marketing ones are likely to arise when considering the territories within your plan. Of course, it would be very nice if we could run exactly the same protocol in every country, but I have never been able to get away with that!

Despite the drive towards harmonisation, regulators still vary hugely around the world. In any phase III programme you will need pivotal efficacy data, but what comparator will you use? You definitely don't have a free choice. For example there is the trend in some therapeutic areas for the US to accept placebo-controlled trials and for Europe to require active comparators.

Apart from triggering different protocols for the two territories, there will be a major impact on study drug manufacture and supply chain. It will add a further layer of complexity with regard to expiry dates, which need to be tracked for both active drugs. So the message is to gather all the information on regulatory requirements as early as you can.

How do you intend to present your efficacy data to prescribers after launch? One reason for needing an active comparator is that regulators typically want to see how your new drug stacks up against the current gold standard. That should fit in well with your marketing plans, as – if you are confident that you have an advantage- you should be able to compete well against the current class leader.

But I was faced with a curious conundrum a few years ago. We were constrained to include a particular comparator in the study because it was the gold standard, but that was going to come off patent before we could get to the market. The effect of that was that its price would drop dramatically, so we would find ourselves with the better drug but at a much higher price. The outcome was that it wasn't worth developing the drug for that market, even though it had uniquely desirable properties, but in a non-serious therapeutic area. This meant that patients lost out in the countries affected.

These examples show just how difficult it can to design an international phase III programme. There is a complex web of scientific, regulatory, commercial, and logistical factors which can be extremely challenging to resolve.

How teams function

I am particularly interested in how people work together in project teams, and quite naturally how they operate becomes even more critical in clinical trials conducted in locations around the globe.

In recent years I have been both frustrated and elated by how such teams function. Interestingly I have found that as important as cultural differences are, interactions between the various functional groups and sub-teams are as least as important. Don't assume that because everyone is from the same country they will get on famously. This problem is thrown into prominence when a complex clinical trial programme is mostly or fully contracted out.

Despite the claims of some contract research organisations (CROs) to offer full services, I usually find myself managing several specialist service providers. At times I wonder if the trick is to find ways of stopping them from fighting each other. Actually its nowhere near as bad as that, and I have been hugely impressed on several occasions with the way people from different organisations can work to common goals, and achieve them.

So how best to set up and manage such multi-cultural and cross-functional teams? There are two main options. Either we set up separate contracts with all the vendors, or we can appoint one as lead contractor who will then manage the others as sub-contractors. For an international study, you will certainly want a CRO with solid presence in all the countries you propose to include. That's a minimum requirement. But if you want to appoint them as lead contractor, you need to be confident they can manage the specialist providers, whoever and wherever they may be.

But how can you have confidence in what they claim? I make it a rule never to engage any contractor without asking for references from previous clients. Quite naturally, you are only going to get the good ones, but if you are specific about what you ask for, you should at least know whether some client somewhere found the CRO to be competent at what you want. In this case, we want to know how effective the CRO was at managing multiple specialist vendors across geographical boundaries.

If we want to dig deeper, we can ask the CRO about their processes and methods for this. Ideally we want evidence that the CRO has the people and processes to handle this management structure, and evidence that it has worked in practice. Beware of the CRO that offers global reach, reeling off addresses of offices in many countries. Check that these are not just associates with a loose agreement to work together, or recently acquired subsidiaries not yet fully integrated.

The other management option is to manage all the specialist vendors yourself. This will naturally be the fall back position if you feel you can't trust any lead contractor to manage them. My first thought is that, if you can't trust them to do that, should you be engaging them at all? There is a cornucopia of CROs out there today, so maybe you didn't look hard enough.  

If you still go for this option, be prepared for a huge management overhead. Sponsors typically underestimate this. The mantra "We are a virtual company, we contract everything out" only works if you have full confidence in delegating all management to a contractor.

I have never heard of anyone doing this, and the multiple-vendor approach hangs on to most of the management and thus needs to be resourced. In particular it has a major impact on project management. If you do delegate management to a lead contractor, you will expect them to work up a detailed project plan including all their subcontractor activities. If you don't, you need to do the planning yourself, or at least have access to all vendor plans (if they have them) and link them into your top-level plan. You can imagine the amount of preparation and ongoing control effort that's required with this approach.

So given the choice, I would always go for the competent lead contractor approach. Ideally I would say to several such contractors at the outset:

"Here is the list of deliverables I require, and the dates I want them. Now go away and work up a plan for delivering them, or if you can't meet my dates tell me the dates you can meet.

If you need to subcontract to specialists that's fine, but I expect to see their tasks and deliverables in your plan. Please provide your rationale for placing the study in the countries you choose, and ensure that you model local factors into your plan. I will choose whichever one of you gets closest to what I want, cost-effectively."

In reality such a request for proposal is of course a lot more detailed, but you should be able to see the principle. There is of course another option if the sponsor is big enough – do the whole thing yourself. But I rather think that the constraints remain quite similar. Large multinational companies are not huge, amorphous and totally flat organisations. To make them work at all, they are broken up into functional departments, geographical units, and vertical hierarchies.

Cutting across all these are projects, necessarily temporary as they are time-limited. Projects assemble people and resources drawn from a range of functions and units, and to make them work agreements are needed, even if they are internal to the company. I don't see a conceptual difference between organising a multinational clinical trial using the internal resources of a large company, and a range of contracted out resources. Indeed multinational companies commonly use an internal market approach in order to make the network of agreements transparent and effective.

Bringing multinational teams together

Of absolutely central importance is this question: how do we get people in different countries to work to a common goal? There seem to me to be two main areas – processes and culture.

To take the latter first, I will upset a lot of readers if I provide a list of difficult behaviours broken down by country! They are really only perceived as difficult because they are unfamiliar and take us by surprise. The truth is that people from anywhere on the globe have more in common than we think. Their different levels of development and social structure are dependent on environmental factors, not on intellectual capacity. People anywhere can cope with new stuff much better than we think. Worth thinking about the next time you are getting exasperated with someone on the other side of the world.

Language is probably the single cultural factor of which we are most conscious. As English is the global lingua franca native English speakers often assume non-native speakers understand it as well as we do. One problem is that in some languages, such as Japanese, certain grammatical concepts don't exist. Therefore, although they may understand the words we are using, the listener may completely misunderstand the message.

In clinical trials, it is not uncommon to issue protocols to all countries in English. If we ask investigators whether they can cope with English, they proudly say yes. But what is the level of understanding? I find it really is worth testing this, as it is not unknown to detect misunderstandings about selection criteria well into the study – a potentially disastrous situation.

Patient-facing materials will have to be translated, and here the cultural domain crosses into the process one. Translation can cause major delays because of error, and trying to cut corners here is highly dangerous. Certification of back translations is essential in my experience, and there are vendors who will carry out the whole process for you. High on the list of negative impacts is that all patient-facing documents need ethics approval, so if you get it wrong, you have to re-submit.

On the matter of approvals, describing the various ethics and regulatory processes around the world would fill a book. But let me condense this into a word of warning: these processes are changing over time as well as across national boundaries. It is vital that they are monitored in real time, and planning templates updated as soon as a change happens. I have been misled in the past by receiving planning modules (cycle times and dependency networks) that were found to be out of date when deployed in the countries concerned.

Communication tools

Our current key tools, email, teleconferencing, and websites, enable multinational teams to interact to an extent unimagined even ten years ago. The danger is one of over-use, with a major clinical trial generating over 100,000 emails easily. I am fighting a personal battle against protracted and unnecessary teleconferences  with a little success!

They work best when highly focused on goals. Routine periodical calls rarely include decisions that can't be made outside them. I am hugely impressed by the way people who have never met, from different countries and companies, can focus on an outcome to which they have all committed.

The key is quite simple – just define the deliverable and when we want it. If the goal is clear and achievable, people will aim for it, wherever they are working.

SOME KEY CRITERIA FOR MULTINATIONAL TRIALS

Deliverables and outcomes: Have all commercial, regulatory and logistical factors been considered when planning geographical distribution for the study? Are targets realistic?

Organisation: What model will you use? If contracted out, lead contractors and subcontractors, or multiple vendors? If in-house, decide who will regulate and oversee inter-departmental agreements

Planning: Who will run the master plan – sponsor or contractor?  Decide on a planning hierarchy – sub-projects at country level, functional sub-projects (eg biometrics) etc

Communications: Decide what circumstances require face to face meetings and which need only teleconferences. Will there be a dedicated project team website hosting all documents and progress reports?

International logistics: Create a standard procedure for the processes and constraints for deploying study materials on time. Ensure you have people on the ground who know these processes – eg import controls, availability of reliable couriers, storage, and accountability

People: Are all team members properly trained and experienced in the country in which they are working? Is there adequate line management and oversight?

Change control: How will changes to deliverables be tracked for impact on time and cost?

Study sites and patients: Has the project plan been adapted to how the local health care system works? Consider any special requirements for patient consent. Will the typical patient profile vary across countries?

Les Rose is a freelance clinical science consultant specialising in project management.