00:00:00: You are listening to the Effective Statician Podcast.

00:00:04: The weekly podcast with Alexander Schacht and Benjamin Piske designed help you reach a potential, lead great science & serve patients while having a great work-life balance!

00:00:22: In addition our premium courses on the effective statistician academy.

00:00:28: we also have lots of free resources for you across all kind of different topics within Z-Academy.

00:00:38: Head over to theeffectivestatistician.com and find The Academy, And much

00:00:44: more For You To Become An

00:00:47: Effective Statistician.

00:00:49: I'm producing this podcast in association with PSI A community dedicated to leading and promoting students' statistics within the healthcare industry for the benefit

00:00:59: of patients.

00:01:01: Join PSI today to further develop your statistical capabilities with access to the ever-growing video on demand content library, pre-registration for all PSI webinars

00:01:12: and much more.

00:01:13: Head over

00:01:14: through the PSI website at psivab.org To learn more about PSI activities.

00:01:20: you can become a PSI

00:01:21: member too

00:01:22: great!

00:01:31: We don't just have one guest, we actually have three guests.

00:01:41: And not only that.

00:01:42: they're statistical methodologies which is a subject which is close to my heart as well and in the moment I'm going get them to introduce themselves.

00:01:51: but just say this podcast has been recorded due to publication of paper entitled Statistical Methodology Groups In The Pharmaceutical Industry Which Has Just Been Published In Statistics In Biopharmaceutical Research and a link to the paper will be included in the show notes.

00:02:10: So, I am going ask people introduce themselves.

00:02:32: My name is Jürgen Hummel, I'm a methodology statistician at Sightel in the Statistical Consulting Group.

00:02:38: I previously headed up a statistical methodology group at a CRO that was PPD now part of Thermo Fisher for seven years and also had a large pharma company Novo Nordisk for year-and-a-half.

00:02:50: And i've worked in industry for about thirty years overall originally from Germany but have spent my entire working life.

00:02:58: I'm Jenny Devonport, i am a statistician by training.

00:03:02: I worked in diverse settings for a long time public health device and drug development at medium-and large companies in individual contributor roles and leadership roles obviously most recently in methodology leadership really just dedicating my career to ensuring that appropriate methods are available to solve problems that matter.

00:03:22: perfect!

00:03:23: And as mentioned the introduction there's this paper which has been published And I wonder if you could just give the audience or give the listeners, not me.

00:03:34: Some top-line headlines of their paper but normally that is maybe some sort summary?

00:03:41: What we were trying to do with this manuscript which has been authored by the FSBI Statistical Methodology Leaders Group is to explore the remit and value of dedicated methodology groups, the organizational considerations that are necessary to address and maximize impact.

00:04:02: The characteristics that methodologists need to influence speed and efficiency in drug development—and our goal here was just in general to address what these groups do why they're needed and how they add value.

00:04:16: Juergen or David?

00:04:17: Do you have anything

00:04:19: Now?

00:04:19: I think that's a very good summary overall, Jenny.

00:04:22: And it was really good to have you were the main drive behind putting that forward and i think what we are trying do from our group is explore the similarities in how we're setting ourselves up also where maybe some of differences.

00:04:38: for example I work both on large pharma setup and CRO set-up.

00:04:43: there's lot of similarities.

00:04:44: like people in these groups have a sort of really deep methodological background in there but the focus areas they may be slightly different.

00:04:54: So, In a CRO setup it is typically focusing on client concerns to try and help them speed up maybe stages of development with innovative approaches.

00:05:05: .In a large pharma setting that's then targeting specific therapeutic areas more.

00:05:11: ,the area that the company difference is.

00:05:16: maybe that in a CRO setting, it's often bigger focus on implementation of published innovative approaches and tailoring them to the client needs whereas.

00:05:27: In large pharma settings there may be sometimes an approach more working within collaboration with academia in order to develop new methodology.

00:05:37: so that is our long-term approach as many less common in CROs.

00:05:43: David any thoughts from your side?

00:05:45: Yeah, maybe just a little bit.

00:05:47: Thank you.

00:05:47: on the longer term aspect of what you were talking about.

00:05:51: so yeah that is important and can be tricky.

00:05:54: part of our work Is getting buy-in from senior leaders to have sort of three year roadmap or five year road map towards success in adoption of new approach.

00:06:06: The paper goes into details how we need to work across all the way through implementation.

00:06:14: Of course, as you were alluding to Jürgen there are many different ways of doing that in big pharma or CROs but In Big Pharma we might be working a little bit more from the left hand side of that equation if you like.

00:06:26: But We still also have to tailor what's developed for specific project needs which is in effect very similar To What You're Doing in CRO.

00:06:35: So there's lots of similarities.

00:06:37: Maybe I could touch a little bit on the training aspect, which is great opportunity that we have in Big Pharma because we have such big audience.

00:06:45: but obviously it also creates difficulties and how you can educate people or inform people across various different sites across the whole world then drive adoption so he gets not standardisation necessarily But use these methods in quite broad fashion.

00:07:04: Fantastic, and that's some great aspects of just a general remit.

00:07:10: Of what you guys do in both big pharma and CROs.

00:07:15: I wonder if he could give me or the audience?

00:07:18: I should say again to give the audience so idea of What does it look like to be a great methodology statistician And what would have typical day look like.

00:07:29: So here maybe i can start by saying I think the one thing that attracted me most to this area is... ...I wouldn't be able say exactly what a great day would look like, because each day will be different.

00:07:41: So there's an amazing variety of what we'll be working on.

00:07:44: so i think that's one of the things to stress which i think it's fantastic!

00:07:48: One great day might be where somebody calls you up or meets at your coffee point if you're in the office and have short consultation with someone and immediately solve their problem.

00:07:59: or you might be spending a whole morning with the study team where your addressing very difficult problem that you've been working within for maybe months of several weeks on.

00:08:10: So, it's another part through day and then may have other days when people don't contact to you and they're new methods writing conference presentation and maybe making some corrections to a paper that you've written.

00:08:26: So there's all of these different aspects that you might be working on, And I think the only difficulty may be that some folks find is not having that concrete knowledge of what they will be working out because feel like it's a particular strength for working in this method or the T-group?

00:08:40: I mean

00:08:40: would you mention there David about variety as certainly very true nuts What i have experienced from years and ultimately what I found really exciting being able to solve often complex problems.

00:08:54: I've always been excited about adaptive designs and it's really become a buzzword over twenty years ago, so what can be great there is when you have a discussion with somebody who comes with the particular problem then during their discussions as they drill into them get an understanding of that maybe a solution around solving those problems or maybe a particular type of adaptive design And the ability and then to see there's a path forward, to explore that further with wider project team with simulations.

00:09:26: With more details... ...and then to figure out if There is solution and ultimately when you see it come to fruition.. ..and help the project team implementation of that.

00:09:37: I think That Is what i found most rewarding.

00:09:41: Maybe jump in here too.

00:09:42: I agree with both David and Jurgen that consultation can be extremely rewarding, the ability to collaborate closely with teams they're working on projects and help them solve new challenges that always seem to arise in drug development is important but something that i also enjoy.

00:10:03: It's going to need scaling and it could be even more impactful on the entire portfolio if everybody can do.

00:10:12: And so that process of developing education plans, creating subject matter expert networks that aren't just methodologists, but also project statisticians who can really help advocate and roll out the new methodology within a company so that everyone benefits.

00:10:32: And everyone feels a sense of ownership and responsibility for implementation—that to me is really exciting!

00:10:39: I think it's as special feature than dedicated methodology group

00:10:47: And maybe I can build on that, Jenny.

00:10:49: Maybe we're going to be a bit controversial here and hopefully not too controversial though?

00:10:53: There may be senior leaders out there who are listening to this saying get the methodology groups.

00:11:00: fantastic!

00:11:00: I'll get the idea but they were nice to have.

00:11:03: How would you respond those people?

00:11:05: So...maybe i make us start.

00:11:07: This is heavily addressed in introduction of our paper.

00:11:12: Our industry has an innovation problem and it's related to efficiency, And It is Related To The High Failure Rates In Drug Development.

00:11:22: The costs of drug development are driven very extensively by the different stages of trial investment... ...and your biggest investments occur in late stage.

00:11:31: But the latest statistics suggest that only sixty percent roughly of phase three trials are successful Which means there a lot room for improvement.

00:11:40: So we have this innovation problem.

00:11:42: This where our methodologists Statistical thinking in general can help is to help drive decision-making.

00:11:48: And so if you think that it's a nice to have, then maybe your not willing to address this forty percent but would be really good for drug development.

00:11:55: we could

00:11:56: If I add too much of that... So i've spent most my working life in the CRO setting and its probably fair say that Not many CROs do have a statistical methodology group.

00:12:08: APD & PsyTel are probably exceptions there.

00:12:13: many CROs focus more on the operational aspects of delivering study.

00:12:18: However, I think that misses an aspect a lot.

00:12:22: clients want to have efficient way running their studies and if we can help them in that then provides head start for actual operational delivery.

00:12:39: Absolutely not a must-have in the CRO setting, but I think it does provide competitive advantage by having that ability to do.

00:12:48: Yeah just add to there... In Big Pharma one of the advantages with our methodology group is they can span multiple different parts of their business.

00:12:58: Statistics are fairly agnostic to particularly therapy area where people who work in particular therapy areas thinks that the methodologies doesn't get used anywhere else, but so we're able to make those links and sometimes that can lead to advances in different areas or just awareness.

00:13:17: We can translate one problem in oncology into one in cardiovascular for example or met different areas And so we can provide that link as Jenny was talking about.

00:13:28: when deciding which problems should we scale obviously would like on.

00:13:33: let's go to be broadly applicable across a whole range of different areas in the business.

00:13:39: And then I think that offers very clear way for CD management to see the value methodology groups.

00:13:45: I'm going to mention an additional point brought up by our American Statistical Association colleagues, they and other discussants were advocating Acknowledgement of the fact that a lot different areas of drug development could benefit from statistical thinking.

00:14:11: So, decision-making which I already talked about but also operations which Jurgen mentioned... There are lots of places where we can gain some optimization or efficiencies and that would benefit from Statistical Thinking And Different Organizations choose to operationalize that differently.

00:14:31: They may have their own groups who do that.

00:14:35: but they may still benefit from the statistical knowledge of a methodologist,

00:15:04: statisticians who are young in their career, or they may be slightly more experienced.

00:15:10: They might be sitting there thinking how do I become one of these fantastic methodologists?

00:15:15: So actually that makes a great methodology statistician.

00:15:18: There's obviously the technical knowledge first and foremost.

00:15:22: In order to be good methodology statisticians you need to have an deep knowledge of statistics.

00:15:28: And what are the underlying assumptions Of different approaches?

00:15:31: What is the framework ?

00:15:32: What are the limitations?

00:15:34: but then also the ability to adjust an approach as and when needed rather than just applying it in a very specific way.

00:15:43: So that technical skill is a great starting point, But I think It's only a starting point because there are lots of other aspects too.

00:15:53: That.

00:15:54: so Jenny do you want add on some others?

00:15:57: Technical skills or this starting point what?

00:16:00: having a deep understanding of drug development So that you know what questions to ask and, You Know What Might Be Practical

00:16:09: is

00:16:10: also important.

00:16:11: Having the ability And The Social Skills To be able to interact with lots of different kind Of non-statistician personnel Is really Important.

00:16:20: having a sense of curiosity about the basic science About the medicine about the different players involved the intrinsic motivation to pursue problems and to pursue relationships, of course an extraordinary ability to collaborate.

00:16:36: But David I think you are ready to speak as well!

00:16:39: Thank you Jenny.

00:16:40: yeah.

00:16:41: so listening this sounds like we're looking for a unicorn or sort-of impossible superhuman?

00:16:47: And then sir i'd just like to reassure listeners that obviously we understand people start from different places different backgrounds and to have different paths too, how they've joined a methodology group.

00:17:01: So sometimes their academics who then join industry other times the people that worked in projects for awhile or several years but also want to go back a little bit through their methodology routes, if you like.

00:17:18: And so then they get the best of both worlds when they come back into a methodology group.

00:17:23: So just to emphasize that everyone in the team is different and obviously having a diverse team is very important.

00:17:29: It's very unusual or impossible for the whole team have all these skills we're talking about at the moment.

00:17:35: So it'll be a team exercise to do that... ...but he only wants them.

00:17:39: coming from that point

00:17:40: I wanted to say yes, we know that the environment is changing and there are a lot of additional types of expertise coming into the quantitative array in pharmaceutical development.

00:17:55: One example is software development The open source revolution in the pharmaceutical industry towards using more open-source software as opposed to licensed software to address innovative methods and get them implemented sooner.

00:18:12: This is something that has come into play, so some organizations have software development within the methodology groups—some of them somewhere else.

00:18:20: And then there's also artificial intelligence which finds its way in different parts of organisations.

00:18:28: In some organizations, they exist with the Methodologists in an innovation group.

00:18:34: And sometimes they sit in a parallel group.

00:18:38: but certainly perhaps there is benefit to those groups being synergistic both in terms of validation and compliance... ...and just realizing all value that possibly can.

00:18:52: Thank you for their expansion on what I was talking about which i fully agree.

00:18:57: Key skills that is relevant now as the world changes in an ever rapid fashion, obviously you would imagine methodologists will be very curious about new methods.

00:19:08: That's a key part of their role.

00:19:10: but also things come from AI side.

00:19:13: I've been delighted to see several people just lean-in immediately To understand how these new technologies are working and then provide expert guidance various senior leaders about the value some things bring and be honest, there's less value that other things bring which is a part of our role to be a trusted partner in those conversations.

00:19:35: And I actually had Monday's question around AI which was quite interesting... ...and i'm wondering if we could look into the future.

00:19:43: where would you see the methodology groups an AI working together?

00:19:47: Maybe it's okay.

00:19:48: so whether the methodology group have to evolve

00:19:52: It's a good question.

00:19:53: And maybe I make a slow start?

00:19:55: One of the things that current models are very good at is automation, so when you already know what to do and how it should be done better or faster... That is where todays' model can do an excellent job of accelerating work where there aren't as many questions about how we're doing.

00:20:21: Obviously that's not innovation, right?

00:20:23: That's implementation.

00:20:25: That scaling.

00:20:26: and so it's this innovation piece... ...that we have to see how the models of the future perform.. ..and how they can support methodologists & other scientists in their work To accelerate invention ,to accelerate adoption.

00:20:45: I certainly see The potential of these new technologies to help with scaling and automation, an adoption.

00:20:52: But I think the jury is still out on innovation.

00:20:55: but i'm sure that I can be corrected on this.

00:20:58: David?

00:20:59: Thank you.

00:21:00: I'm not sure if we're going correct you by just saying it's important thing in one two three five years time will be driving aspect organizations.

00:21:09: so who was driving?

00:21:10: Essentially, you might say no one is because these automatic cars exist already.

00:21:15: But I think the crucial component is that... ...I think it should be people like ourselves and our teams that are driving the implementations right.

00:21:24: so by asking the right questions And then doing research in the right methodologies and seeing how they can fit into new technology rather than other way round Because otherwise we run a risk of looking at those technologies sometimes failing for reasons that we would readily be able to have foreseen if we were consulted.

00:21:45: But, We shouldn't sit back and just wait for it happen because... ...we need big and bold and go through this.

00:21:52: then the chance of leading those conversations.

00:21:55: And If you look at some aspects of machine learning as well as artificial intelligence there are approaches becoming more popular in modern a varied adjustment that are machine learning based and do have the ability to make our statistical analysis more efficient.

00:22:16: Some of the questions we hear from regulatory authorities is ensuring that we still have good understanding on what drives treatment effect under certain circumstances in specific populations or even subpopulations, so I think this is where we as methodology statisticians can really help is try and get their understanding from the drug development perspective rather than simply the implementation of those approaches.

00:22:46: And I'm going to add on top that AI issue, you mentioned adaptive designs earlier on.

00:22:52: my way thinking as a human it's much better thing designing clinical trial particularly innovative clinical trial because an AIs only gonna be working for what he knows in database Whereas you might be thinking, actually that's not going to work.

00:23:07: And maybe this is gonna work?

00:23:08: Maybe this more innovative and a better way of looking at the design.

00:23:13: So for me we've got to embrace it.

00:23:15: I don't need the genies out in the bottle We can put them back But as a case of working with it there are human things That we could do And i certainly think designing an innovative clinical trial Is one those human things.

00:23:29: I'm fully with ya there.

00:23:31: Since your mention about the design There is no such thing as one type of adaptive design.

00:23:36: The ICH E-twenty guideline is out, there's a draft in it already mentions many different areas but they are additional types of adaptive designs that are not even mentioned in a lot of detail.

00:23:47: because ultimately what you're trying to do Is try and build the study design That addresses the main area of uncertainty You have at the beginning or your studies so then can take corrective action if needed during the study rather than just seeing the results at the end and finding out that, oh this is not what I needed or would have wanted.

00:24:09: In those situations you're absolutely right Alan it's the ability to try and understand where is our main area of uncertainty?

00:24:17: And how can i choose one out of many different types... ...to try and address

00:24:22: it?".

00:24:23: Yeah!

00:24:23: Just to add to that it's a remiss me not to mention the work we do and collaboration with regulators as an ex-regulator.

00:24:31: And, as Jurgen pointed out even if people within the company think that they might be able to adopt various things other people will be asking questions of a technical nature which we're the best groups to advise on.

00:24:43: I think thats very important thing.

00:24:45: flag up is well.

00:24:47: i think there's lot of scope their.

00:24:49: obviously everyone will learning about new approaches.

00:24:52: it'll be sort or roller coaster riders.

00:24:54: people learn more about them anyway.

00:24:56: I like roller coasters, so that's fine.

00:24:58: we look forward to the ride

00:25:00: and maybe i'm going to move away a little bit here.

00:25:01: and just your three methodology leaders.

00:25:05: statistical methodology leaders?

00:25:06: Just to guess what percentage of companies.

00:25:10: this is really difficult because it's obviously last small companies.

00:25:13: if you were to guess What percentage have statistical methodology groups?

00:25:18: can't give you an exact number and statisticians hate guessing and gambling, but I would say that they are fairly common in large pharmaceutical companies.

00:25:28: And maybe less common in medium-and smaller biotech companies.

00:25:33: I would completely agree.

00:25:35: and if we then look further, i already said earlier that they're not very common in a CRO environments even amongst the sort of medium to large size CROs.

00:25:45: They are not necessarily that common so...I think their biggest proportion is you right?

00:25:51: In the large pharma setting.

00:25:54: And it makes sense because larger pharmaceutical company Is going to likely have more diverse portfolio.

00:26:02: They may be pursuing more complex targets, and so they may have more demands to streamline and centralize.

00:26:12: I'm going say this has been a fantastic discussion And open.

00:26:16: the listeners will love this.

00:26:17: but i want to go as we're starting To wrap this up an know if touched on it already.

00:26:22: for what is The true goal of a methodology departments?

00:26:27: I think ultimately what unites us is that we all want to make drug development more efficient and i think that's been the driving force of F-SPY stats methods leaders group being established in order to help us jointly achieve.

00:26:43: That because if we are able to join force, maybe have helped with the spreading off new methodology.

00:26:54: that is going to be a big step forward for all of us.

00:26:58: So, me?

00:26:59: That's the biggest aspect!

00:27:01: For me ever since I started in statistics... ...I've always been looking to apply something as it were.

00:27:07: so if it was just theoretical and wasn't applied although might beautiful or very clever.. ..I would always be deflated some way they hadn't being used.

00:27:16: And so i think the biggest challenge we face at the pharmaceutical industry is slow.

00:27:20: adoption are very beautiful things.

00:27:23: And so that's the most important thing, and goal is to speed up that adoption of these family's new approaches being used in the right situations.

00:27:31: Because as we've said they shouldn't be used everywhere just finding the rights situations to use them.

00:27:38: That's the beauty of The Game.

00:27:40: has it where there were in?

00:27:42: We need get even better at that because I have been talking about recently some things are incredibly slow which for various reasons.

00:27:50: So we can make that, those timelines much sharper and then we can innovate more quickly.

00:27:56: And David if I could probe on that.

00:27:58: but how are you going to do that?

00:28:00: Better connections between various groups is one aspect in which were all working as part of the F-SPY.

00:28:07: Stance Methods Leaders Group was at an F-STANCE leaders meeting earlier this week.

00:28:13: so they're thinking about ways doing it so that it's partly those connections.

00:28:19: It is also about better communication of methods, and we can clearly articulate them to non-statisticians in a better way.

00:28:28: So I think if we do that We've got the potential to vastly speed up the adoption as new approaches.

00:28:36: The traditional academic route is you invent a new method.

00:28:41: You write a paper, you present it at the conference and then a miracle happens right?

00:28:47: Or everyone ignores it.

00:28:49: but I think that goal of methods in drug development is deployment not invention.

00:28:55: Right.

00:28:55: so... A concept i have found really helpful.

00:29:00: It's not mine!

00:29:01: Its ours.

00:29:03: its' not my method that I developed, it's our method that we're using together.

00:29:07: And so where you see that play out in groups like the FBI Statistical Methodology Leaders or American Statistical Association counterparts is the collaboration on specific topics.

00:29:22: there are special interest groups dedicated to kind of further development and implementation of specific methodologies, this collaboration that we don't have to do things alone.

00:29:34: In fact will be faster and more thorough if we do it together.

00:29:39: so all the steps in the innovation cycle benefit from collaboration whether its internally with project statisticians at your company which is huge or also externally with academics other statistician appropriate context regulators.

00:29:55: You can really take things further faster because inventions don't have value until they're adopted at scale.

00:30:03: And so you mentioned the F-SPY leaders in the S-Py status, if listeners are listening to this and I say how could i connect with F-Spy?

00:30:11: How do you connect with the SPY leader than the SPI statistical readers?

00:30:15: The F-spy has a website that details there about stats methods leaders group With contact names of people who lead those groups.

00:30:24: That's the best way, have a look there and contact to people that are mentioned.

00:30:30: And Jörg and I will put in the show notes... ...and how about you three?

00:30:34: What is your best way of connecting with youths?

00:30:36: LinkedIn!

00:30:37: Yes it works!

00:30:39: I'll

00:30:40: hear nods from everybody.

00:30:41: So let me take one moment.

00:30:44: really thank you for coming along this edition.... ...and like i said very special edition of The Effective Statistician.

00:30:51: I know the listeners will be delighted with this episode, and actually what i'm encouraging the listeners to do is if they like this episode put some comments to say what you liked about it.

00:31:02: If there's any questions that you'd like to ask Jenny, Jürgen or David then maybe put those in as well so we can answer them on a future video or chat form.

00:31:12: So thank-you very much!

00:31:21: This show was created in association with PSI.

00:31:25: Thanks to Rain and her team at VVS, we are position on the background and thank you for listening!

00:31:31: Reach your potential, read great science and serve patients.

00:31:34: just be an effective statistician.