The Power of Curiosity: How Simon Lovestone Helped Advance Alzheimer’s Research

Twenty-five years ago, the only way to get an accurate diagnosis of Alzheimer’s disease was to confirm the presence of disease-specific proteins, such as amyloid, in PET scans of the brain or in spinal fluids. These approaches, both just becoming available at the time, could be invasive, costly and not readily available. A blood test would be transformative.

When Alzheimer’s Research UK , a nonprofit organization, called for grant applications to identify blood biomarkers for AD, I was skeptical. I thought that it was unlikely that something occurring in the brain could be detected through blood because the blood-brain barrier tightly regulates the movement of molecules into and out of the brain.

Nevertheless, I applied, and I got the grant. Using proteomics, the large-scale study of proteins I set out to prove that the blood proteins in people with AD would be no different from those in people without the disease. What our research revealed was quite the opposite.

That work triggered an explosion of research in blood-based biomarkers for AD. Today, there are astonishingly accurate blood tests for AD approved and in development.

The ability to accurately detect disease-related biomarkers through a simple blood test is helping make clinical trials more efficient by identifying the right participants, reducing costs and enabling more precise measurement of treatment effects.

Tau is a protein found in brain cells, where it helps neurons stay stable and move the nutrients and signals they need to function inside the cells. In AD, tau is altered by a process called phosphorylation, causing it to detach and clump into tangles inside neurons. What makes tau so scientifically compelling is that those tangles follow a predictable path: they begin in the regions of the brain responsible for forming new memories and spread outward from there. As they spread, the abilities those regions support are progressively lost, which is why tau burden is so closely associated with the severity of a person’s cognitive decline. That process is what I spent my early career trying to understand. My doctorate asked a narrow question: which enzyme is actually responsible for phosphorylating tau? I spent the years that followed at King’s College London and then Oxford, with tau at the center of my work.

J&J’s tau work was part of what drew me into industry. They were the only company pursuing antibodies against specific sites on phosphorylated tau — I had spent my career asking how tau becomes phosphorylated; they had built a therapeutic strategy around what happens once it does. It was a bold bet, grounded in strong science.

As part of that strategy, J&J developed an antibody that recognizes the tau p-217 site, which is common in AD but rare in healthy brains. J&J then developed a blood-based pTau217 assay to support its clinical trials. By publishing what we learned and working with researchers outside J&J, we helped build a broader body of evidence around pTau217 and its potential as an AD biomarker.

I consider my current role as a Distinguished Fellow the pinnacle of my career because it enables me to pursue truly transformative science. Today, my work focuses on the intersection of artificial intelligence, large-scale data, drug discovery and biomarker research. It brings together teams across J&J and external partners to explore new approaches to some of the most difficult challenges in neuroscience.

The idea grew out of a conversation with a friend at Gates Ventures who leads AD research efforts there. We were frustrated that large, costly studies could generate valuable data and biological samples that were often used to answer only a single research question. We wondered whether combining data and samples from multiple studies could accelerate biomarker discovery and deepen our understanding of neurodegenerative diseases.

These diseases are highly complex and heterogeneous. People can experience the same disease very differently, and understanding what drives those differences is critical. J&J and Gates Ventures co-founded the Global Neurodegeneration Proteomics Consortium (GNPC) to do what no individual institute could achieve alone: create datasets large enough to identify and better understand patient subgroups.

To date, the GNPC has brought together more than 40,000 blood and cerebrospinal fluid samples, generating nearly 300 million protein measurements from healthy individuals and people living with neurodegenerative diseases. The public release of the first dataset was accompanied by a coordinated series of publications in Nature , demonstrating the scientific value of bringing these data together and making them widely available.

Researchers around the world are using these datasets to generate extraordinary scientific insights. The resource continues to grow in depth and value, creating opportunities to better understand disease biology, identify patient subgroups and uncover new therapeutic targets.

AI will be one of the biggest forces shaping the field. It needs massive, high-quality datasets to be useful, which is why collaborations like the GNPC are so important. I think it’s safe to say that AI will dramatically accelerate drug discovery and development and help us uncover treatments for diseases that are currently untreatable in a timeframe we could not have imagined.

But even as the tools change, the qualities that make us good scientists do not. AI has no curiosity. Diligence and perseverance remain vital, as do good instincts. These very human qualities will become even more important in the future.

It has been thrilling to be part of what we have achieved so far. I only wish I could see where the field will be 50 years from now.