High Throughput Screening Lessons Learnt: How Biobanking Can Benefit?

Mark Divers is the head of the Karolinska Institute’s biobank project in Sweden. He talks us through some of the key concerns before his seminar at this year’s conference. To listen to the podcast now go to: Converting Problems into Opportunities When Building Up a Biobank

Pharma IQ:  Mark, thanks for your time. Welcome to the show. Could I start off by asking you to give us an overview of the Karolinska Institute’s Biobank? Essentially, how long it’s been in operation, what its aims are, and how it’s being managed…in as much of a summary as it’s possible to ask for!

Divers:  Yes, certainly. The Karolinska Institute Biobank has been running since 2004. In 2004 it had its operational start and it started off by collecting samples from, what I think is, the well known Swedish Twin Registry, and it has built up over the years since then. It’s the central core facility for the Karolinska Institute, the Medical Research University in Stockholm, in Sweden, and it provides a biobanking service to the whole of the Karolinska Institute and some of the local research hospitals in the Stockholm region.

Its themes have always been to provide research infrastructure for research on human samples, and, as such, it’s been building up a profile and operation that supports epidemiological research amongst other types of research. And now we’re in a phase when we’re going to be scaling up the operation, so we’ve set the aims even further now to be able to support the really large epidemiological studies that are becoming more and more possible. In Sweden we have a very large study that’s about to kick off later this year, looking to map half a million citizens of Sweden, and then we have a number of other large epidemiological studies with large cohorts of either patients or volunteers who are prepared to enter into these research studies.

Pharma IQ:  Excellent. I know that up-scaling will be the focus point of your seminar, and you will be looking at converting problems into opportunities when building up a biobank. So, can you tell us where you use the major areas of risk throughout this process, and could you provide us with an idea as to how we should be strategising to lessen the likelihood of coming up against those problems?

Divers:  Yes, certainly. Of course, everything I tell you now is my personal view and my opinion as I see it today. What I should also say right up front is, I’m new to the biobanking arena. I joined Karolinska Institute’s Biobank as Executive Director at the end of last year. I’ve got my background in the Pharma industry where, amongst other things, I set up a compound bank and a high throughput screening operation.

So it’s with a newcomer’s eyes that I’m going to answer your question. For me, where I see the risks, are really twofold. The first area is technical, and it’s just simply whether we can… whether we’re able to put in place the kind of technical solutions to, not just to collect the samples and store them in a safe and high quality way, but also to access them, and access them at a high enough speed to support the research needs.

[inlinead]

There’s a whole area that goes under that title of technical risk there, and it’s a lot to do with the kind of automation infrastructure that we can put in place that can operate under the storage conditions. We want to store samples for long term quality. It’s typical that most biobanks want to store samples that are either -80° in freezers, or even in liquid nitrogen, and to get lab automation to operate under those conditions is really tough. We’re really at the cutting edge there with the technology.  And when I say we, I mean the industry as a whole, and the suppliers of the industry, the automation suppliers. So that’s one area of risk.

The other one is probably a more wide-reaching risk, and that is the ethical situation – being able to make sure that we comply with existing law, but also ethical expectations of, not just the users of the biobank but the individuals who provide their samples to go in there. Are we able to handle the data in an ethically and sustainable and defendable way, and ensure that people who have volunteered to take part in these studies have their data and rights respected? That’s an area that’s still not fully developed, especially when we look at the big international perspective.

Pharma IQ:  Absolutely. I’m sure it’ll be one of the big topics at the discussion as well. Apart from that, a lot of the discussion at this time is concerned with this potential collaboration between various national and international biobanks. Now, Mark, if more partnerships like this are enacted, what do you feel – obviously, just from your perspective – would need to be in place from the initial outset to ensure such collaboration could continue smoothly?

Divers:  First and foremost is the ethical and legal situation. We need to make sure when we collaborate across the borders that the partners involved are working within the legal frameworks of the companies they exist in, and that a mutuality [exists] that enables samples to be handled in a way that fulfils the ethics for the citizens of both those partner countries. It’s not to be underestimated, that. But also, it shouldn’t be seen as an insurmountable problem, but I think it’s one that’s still developing. Some countries have established law to protect individuals and citizens’ rights with regard to their samples; other countries don’t. So there’s still a way to go there, to sort some of those problems out. Then the other area is more technical, and that is ensuring that quality standards are compatible so that if we are collaborating – where the collaboration involves transfer or exchange of samples – we’re able to treat them in the same way as protocols that are compatible.

Pharma IQ:  Okay, and on this issue of protocol, and actually standardising protocol for quality control when it comes to data collection, how important is this, given the difficulties such an arrangement could present in terms of pooling data from different studies, particularly from large data sets? Can you suggest any areas you feel are particularly in need of standard protocols, or can you suggest any that possibly exist at Karolinska that you feel could be a good avenue towards a universal approach?

Divers:  This question of universality is always a tricky one because wherever you go people have different views on how to do things, and, of course, even though science is built on a bedrock of reproducibility and protocols that are clearly described and defined, we still have many different ways we want to address different problems. So it’s often hard to get people to agree on a completely identical process or a standard for a process, although the outcome is often identical.

I don’t think we at the Karolinska Institute’s Biobank have any…what should I say?...any “clear leadership” in standardised processes. We have some processes that we believe are right for the samples that we’re collecting and storing, and standard operating procedures that suit the research studies that we support, but I know that there’s variety in what the studies want between them and we often have to discuss and compromise with them and find enough of a common ground to be able to make progress at the same time as allowing different research studies to address their samples in the ways that suit them. So I think it’s a matter of finding the common ground where we can, but then recognising that we can’t delay progress by trying to over-define things that will be really hard to reach a very common standard on. I’m afraid the answer I’m giving you is rather abstract…

Pharma IQ:  No, it’s a very good point.

Divers:  My background in compound management and in high throughput screening taught me to be a pragmatist, and I’ve seen this kind of area often hindered by ground attempts to standardise beyond what is needed. I think the key is to find a minimal core of standardisation around which you can have a degree of coordination that allows a variation at the edges, if you like.

Another thing that’s really critical is in the information management and having metadata standards that are agreed internationally, and what I mean by that are the data categories that we agree are relevant for the samples that we’ve got. What kind of phenotypic descriptions do we use to classify the analyses we do on the samples? There I believe there’s a wide variation and a long way to go, whereas if we look at the genotypic data that, almost by definition, is a lot more standardised…but maybe there are some general lessons we can take there as we move forward with phenotypic descriptions.

Pharma IQ:  I’m sure there is, thank you. Now, moving on, it’s been theorised that a shortage of trained personnel in this field is slowing down effective analysis. What should the industry be focusing on, do you think, or investing in to tackle this issue, both now and in the future?

Divers:  In one sense, yes, I agree with the premise there that it’s rare to find people with the experience of, let’s say the combination of the technical expertise and the experience of the legal or ethical side of things. But, that said, biobanking…it reminds me very much of the high throughput screening and compound banking world. It’s really one of these…it’s not a pure discipline. It’s built on a number of key scientific and technical areas, stretching from bio-analytic through to automation and information management, and, in fact, it’s rare to find people who have covered every base. But you can also rapidly make progress in this field with expertise from any one of those areas, and you can gain the experience you want by joining part of an established team.

As I said earlier, I, myself, am new to the biobanking world. I’m fortunate even being able to rely on some of my close colleagues who have been working in the area longer than me, but also the experience I bring from a different area I see can help in making progress in some of these tricky questions that we were talking about before. I think it’s important to collaborate in this area and to build teamwork as much as possible. That was one of my lessons from the high throughput screening world – collaboration is everything. This is not the kind of area where you make big progresses as rugged individuals. It’s definitely a teamwork area.

Pharma IQ:  Okay, and speaking of collaboration and it being as important as you’ve said it is, do you think there’s a distinct lack of harmonisation at the moment between industry academia, health services and governments? And where can we perhaps begin making progress in this area to offset that?

Divers:  Yes, as a newcomer, I have to say, my early impressions are that it’s an area that could do with a lot more collaboration in those areas. I’m not sure whether I’d call it harmonisation, but certainly between academia and the health service organisations we need to be clearer in what we think we can achieve together. What I see is a landscape of many, many different divergent biobanks in a country, whether it’s this country or any of the others that I’ve had a glimpse into. And maybe a certain isolationism between biobanks – that’s probably putting it too strongly – but I think the different parties involved could establish a more effective dialogue with one another.

Again, this is an area where it’s hard to avoid some of the abstract generalisations, but we definitely need to be more collaborative between the different kinds of biobanks involved, and also recognise that some of them have different purposes.

Please note that we do all we can to ensure accuracy within the translation to word of audio interviews but that errors may still understandably occur in some cases. If you believe that a serious inaccuracy has been made within the text, please contact +44 (0) 207 368 9334 or email richard.desilva@iqpc.co.uk .