No pharmaceutical company can fund everything in its pipeline. Resources are finite, and oncology pipelines have grown so large that even the biggest companies have to choose. Which programmes get full investment, which get held back, and which get cut entirely. These decisions shape a company’s future more than almost anything else it does, and they are made with incomplete information under real uncertainty.
Stelios Tzellos of the UK works at the centre of these decisions. As a professional in business insights, analytics, and oncology marketing at AstraZeneca, with earlier roles at GlobalData and IQVIA, he helps build the analysis that informs which assets a company prioritises.
The Problem With Picking Winners
Every programme in a pipeline has a champion who believes in it. The scientific team sees the mechanism’s promise. The commercial team sees the market. Left unchecked, prioritisation becomes a contest of advocacy, where the loudest or most senior voice wins rather than the strongest case. Good portfolio analysis exists to replace advocacy with comparison.
That comparison is harder than it sounds. The programmes differ in disease area, stage of development, probability of success, and time to market. Comparing a late-stage asset in a crowded indication against an early-stage asset in an untapped one is not a simple calculation. It requires translating very different programmes into terms that can be weighed against each other.
What Goes Into the Decision
A sound prioritisation weighs several factors at once. The size and accessibility of the patient population. The strength of the clinical rationale. The competitive field the drug would enter. The probability that development succeeds. The time and cost required to reach the market. Each factor carries uncertainty, and the analysis has to hold all of them together without pretending any single number settles the question.
Tzellos developed this kind of multi-factor analysis at GlobalData and IQVIA, building forecasts and competitive assessments that clients used for investment decisions. The forecasts that helped most were the ones that made their assumptions visible, so the people deciding could see what they were actually betting on.
Why Science Literacy Changes the Analysis
Probability of success is where scientific understanding earns its place in the room. A programme’s odds depend heavily on the biology: how well understood the target is, how clean the mechanism looks, how much the early data supports the hypothesis. An analyst who can only see the financial model will take the probability estimate as given. An analyst who understands the science can question it.
Tzellos studied molecular biology at Imperial College London, where his doctoral research examined gene regulation in Epstein-Barr virus. That training lets him engage with the scientific case for a programme rather than treating it as a black box.
Honest Analysis Over Tidy Answers
The temptation in portfolio work is to produce a clean ranking that makes the decision feel automatic. That is the wrong goal. A ranking that hides its uncertainty gives leadership false confidence and invites bad decisions. The better output presents the trade-offs clearly and lets the people accountable for the choice see what they are accepting and what they are giving up.
At AstraZeneca, Tzellos leads cross-functional projects that support portfolio decision making across one of the industry’s most active oncology pipelines. Deciding which drugs to back is among the hardest work a pharmaceutical company does. It cannot be made certain, but it can be made honest.