Industry Insights: Let’s Automate the Future

What would you say has been the most significant advancement in molecular diagnostics during your career?

I feel fortunate to have witnessed the beginning of the evolution of molecular diagnostics. I was born in the early 1980s, when PCR was just being established as a technology. Since then, I’ve lived throughout what I would call a revolution in molecular diagnostics – to the point where next-generation sequencing can now happen almost anywhere in the world.

I like to draw a parallel to something we're all familiar with: the smartphone. I remember the very first mobile phones being released, and now we're in the smartphone era. That's exactly what has happened in the world of molecular pathology – we've gone from analog to fully digitalized. In a nutshell, I've witnessed a quantum leap in our ability to generate molecular profiles of patients.

What do you think is still missing from the molecular diagnostic toolkit?

In the last couple of years, we've seen strong advancement in the automation of laboratory systems. But customers are still asking us to push even further in that direction – making sure that workflows within the laboratory are fully automated. Let me add an important point here: they don't want a black box. They want to know what's happening at every step, but they don't want to be doing it manually themselves. 

But I would also add that we're now starting to understand that clinical data means nothing unless we're able to decipher it. The combination of automated data generation with minimal hands-on work – combined with the emerging capabilities of artificial intelligence for data interpretation – is like squeezing every single drop of juice from your lemon. And I think we've left a lot of juice in the lemon that we've squeezed so far. There's still a lot of understanding to be distilled from the data we've accumulated.

What are the most pressing barriers to equitable access to molecular diagnostics?

I think anything that's extremely complex is, by default, not deployable everywhere. Industry – and I think Thermo Fisher is really at the cutting edge of this – is investing in making solutions accessible and easy to use. In previous years, people tended to confuse "easy to use" with "simplistic," which is not the case. We still provide extremely complex, in-depth molecular diagnostic tools and solutions, but we're just making them much more straightforward.

This is an extremely important point because anything that requires an engineer to come to your lab for maintenance is not practical. So molecular assays have to be scalable, fully automated, and easy to use. Ideally, they shouldn't necessarily require highly trained personnel, because we can’t always guarantee an abundance of them. Some of the technologies are already out there, including some of ours, and I think we're taking great steps forward toward equitable access.

What is industry doing to bridge the gaps?

Delivering a fantastic, fully automated instrument isn't sufficient. Industry has understood that you can't just place an instrument and run away. You need to ensure that training collaborations with medical associations are in place – as we do at Thermo Fisher. I think we're moving away from the classical model where only academia provides rigid education. Educational partnerships – conducted in a very transparent and compliant manner – are on the rise, and I see this as a positive development. We need to bridge the gap, and the way to do it is through education.

What is your vision for the future of precision medicine?

I hope that before the end of my career, precision medicine will become a reality everywhere in the world. Today, in all honesty, it is a reality – but only in a very selective number of places. What we need to work toward is a situation where – no matter where in the world patients are – they have access to the best possible treatments available.

If I move this a little closer to what we're doing, I think the power of precision medicine can’t be fully unleashed unless it's combined with robust and reliable diagnostics. In order to connect the right patient with the right therapy at the right time, diagnostics is the glue that sticks the parts together. 

We've made substantial investment in therapeutics and precision medicine, and we'll continue to do so, but this needs to be matched with investment in diagnostics. Otherwise, precision medicine will remain just a buzzword amongst underserved communities.

What are the emerging trends or innovations that you are most excited about?

Of course, I have a lot of expectations for AI. Because I worked as a molecular pathologist and laboratory team leader myself, I can appreciate the bottlenecks that manual processes – including data annotation – cause. I see how complicated molecular diagnostics can be. I think AI can really open up technologies to the next level.

One specific thing I believe will happen within the next five years is the combination of AI capabilities with computational pathology. It's very important to mention – and as an employer of pathologists, I wouldn't say anything different – we don't need to substitute pathologists; we need to enhance their abilities. We need super-pathologists enabled by AI tools.

But at another level, what excites me is the ability of AI to help us understand data from omics in a much deeper way – and therefore extract clinically valuable information. Not just sequencing of DNA or RNA, but the overlaying of different omics levels and the combination of these data. Proteomics, for instance, is the next big thing, and that's where most of the value will come from.

What specific improvements to the molecular oncology workflow do you think would be particularly impactful?

We want to make sure that automation will really allow pathology labs to process even the most limited amount of specimen – whether it's tissue or, hopefully even more in the future, a liquid biopsy. By embracing automation, molecular pathologists will be able to extract meaningful data that can inform physicians to make the most appropriate treatment decisions. The breakthrough is happening, but it needs to happen faster.

We also need to make sure that laboratory professionals are able to manage the increased complexity of molecular diagnostics by reducing the complexity of their workflows. That is the best way to deliver results as fast as possible for the benefit of patients.

What else would help with molecular diagnostics implementation?

I think it’s important to acknowledge that the amount of pressure on molecular pathology and medical genetics laboratories is increasing. Why? Because to enable precision medicine, you need molecularly informed diagnoses. That's where I think companies like Thermo Fisher and others will be transformative in this field – by enabling laboratories to satisfy this increased demand coming from clinical work.

The connection between laboratories and clinicians will get closer and closer, and the full integration of different capabilities – from molecular biology professionals to clinical bioinformaticians, all the way to pathologists, molecular pathologists, treating physicians, and so on – they all need to come together in an ecosystem that's better structured. I think that's exactly what's happening today. We want to enable this ecosystem to be more efficient for the benefit of healthcare systems and for the benefit of patients. I feel honored to stand at the heart of this unfolding journey.

Newsletters

Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.

I have read and understand the Privacy Notice*

Subscribe