Stelios Tzellos | The Analyst Who Actually Reads the Science

Stelios Tzellos

Walk into most pharmaceutical analytics teams and you'll find smart people who are very good at building models. They can run Monte Carlo simulations, construct patient flow diagrams, and produce slide decks that make complex markets look manageable. What you won't always find is someone who has read the primary literature on the disease they're modelling.

Stelios Tzellos of the UK is an exception. Before he built his first commercial forecast, he spent years in a laboratory at Imperial College London studying Epstein-Barr virus gene regulation. His PhD examined the molecular basis for the superior transformation efficiency of type 1 EBV, research that required him to think about biological systems at a level of detail that most pharmaceutical analysts never encounter.

Why the Science Background Matters

There's a common assumption in the pharmaceutical industry that you don't need to understand the biology to model the market. The logic goes: if you have the right data inputs, diagnosis rates, treatment algorithms, market shares, the model will produce a useful output regardless of whether the analyst knows what a kinase inhibitor actually does.

That logic holds until it doesn't. And it tends to break at exactly the moments when accuracy matters most: when a new mechanism of action enters the market, when a combination regimen reshuffles the treatment sequence, or when a biomarker redefines who is eligible for treatment.

Tzellos encountered these inflection points repeatedly during his time at GlobalData, where he worked as a healthcare analyst covering oncology and haematology. He developed epidemiology models and competitive assessments for cancer indications including Hodgkin's lymphoma. The models that held up under pressure were the ones informed by an understanding of the disease, not just the data.

The IQVIA Years: Applying Science to Client Work

At IQVIA, Stelios Tzellos advanced into roles in oncology disease insights and the Analytics Center of Excellence. He worked with global pharmaceutical clients on forecasting, commercial planning, and evidence-based strategy. The clients who got the most value from his work were the ones who asked him to challenge their assumptions, not just validate them.

That's where the science training becomes practical. A forecast assumption like "60% of patients will receive the new drug in second line within two years" sounds reasonable in a boardroom. But if you understand the biological rationale for why oncologists might prefer a different sequencing approach, or if you know that the companion diagnostic has low adoption in community settings, you can push back with specifics rather than opinions.

IQVIA's Analytics Center of Excellence was designed to bring rigor to pharmaceutical consulting. Tzellos fit that mission because his rigor came from scientific training, not just analytical technique.

Inside AstraZeneca's Oncology Analytics

Tzellos joined AstraZeneca in 2020 and has held positions across business insights, analytics, and oncology marketing. He leads cross-functional projects that support product strategy and portfolio decision making for one of the industry's most active oncology portfolios.

Inside a large pharmaceutical company, the analytics function can either be a service desk that runs models on request or a strategic partner that shapes decisions. The difference usually comes down to whether the people in the room can engage with the science as well as the numbers.

What the Industry Should Learn from This

Pharmaceutical companies spend millions hiring consulting firms to build forecasts. They spend far less thinking about whether their in-house teams have the scientific depth to interpret those forecasts correctly. A model is only as good as the assumptions behind it, and the assumptions are only as good as the understanding of the people making them.

Stelios Tzellos built a career that moves from molecular biology to market strategy without losing the thread that connects them. His work at Imperial College London, GlobalData, IQVIA, and AstraZeneca all share one trait: they required someone who could read the science and turn it into decisions.

The pharmaceutical industry would be better served if it stopped treating scientific training and commercial acumen as separate career tracks. The best analysts are the ones who do both. Tzellos is proof of that.