Stelios Tzellos | The Real Reason Pharma Launch Strategies Fail in Oncology

Stelios Tzellos

Three months before a major oncology drug launch, the commercial team realized they had been planning for the wrong patient population. Their market research pointed to second-line treatment. The clinical data supported first-line use. By the time the strategy caught up with the science, they had lost six months of potential uptake to a competitor that read the data correctly.

Stelios Tzellos of the UK has watched versions of this play out across the pharmaceutical industry. As a professional working in business insights, analytics, and oncology marketing at AstraZeneca, Tzellos understands that launch failures in oncology are rarely about the drug. They are about the gap between what the data says and what the commercial plan assumes.

The Data Is There. The Translation Isn't.

Pharmaceutical companies generate enormous amounts of data before a launch. Clinical trials, real-world evidence, market research, competitive intelligence, physician surveys, payer analyses. The problem is never a lack of information. The problem is that the people making launch decisions and the people generating insights often operate on different timelines with different priorities.

A clinical development team thinks in endpoints and p-values. A commercial team thinks in patient starts and net revenue. When these two perspectives don't align, the launch plan reflects the loudest voice in the room, not the most accurate read of the market.

Tzellos has spent his career at the intersection of these perspectives. He studied Biochemistry and Molecular Biology at Imperial College London, then moved into healthcare analytics at GlobalData and IQVIA before joining AstraZeneca. That background means he can sit in a launch planning meeting and spot the disconnect between the scientific evidence and the commercial assumptions.

What Gets Missed in Pre-Launch Planning

Most pre-launch models assume that physician awareness translates directly to prescribing behavior. It doesn't. An oncologist might know about a new drug and its mechanism of action but still default to the regimen they have used for the past five years. Changing prescribing behavior requires more than data. It requires changing the treatment narrative, and that takes time.

Stelios Tzellos worked on disease insights and competitive assessments during his time at IQVIA, where he saw this pattern repeatedly. The drugs that launched successfully were the ones backed by commercial teams that understood the clinical context, not just the market opportunity.

Another common gap: payer access timelines. A drug can have the best clinical profile in its class and still underperform if the commercial team didn't account for how long formulary decisions take in key markets. In oncology, where treatment windows are narrow and patients move through lines of therapy quickly, a three-month delay in access can mean a permanently smaller market.

Cross-Functional Work That Actually Works

At AstraZeneca, Tzellos leads cross-functional projects that bring together medical, commercial, and market access teams early in the planning process. This sounds straightforward, but most companies only achieve it on paper. In practice, these teams still work in sequence rather than in parallel.

The difference between a strong launch and a struggling one often comes down to whether the analytics function had enough weight in the room during planning. When business insights teams are treated as a support function that runs models on request, the strategy suffers. When they're part of the strategic conversation from the beginning, the plans are grounded in reality.

Why This Keeps Happening

The pharmaceutical industry is structured around functions, not problems. Medical does medical work. Commercial does commercial work. Insights sits in between and serves both. That structure makes it easy to stay organized and difficult to make good decisions.

Stelios Tzellos built a career that crosses those functional lines. From molecular biology research at Imperial College London to consulting at GlobalData and IQVIA, to an industry role at AstraZeneca, his path reflects what the industry needs more of: people who understand the science, the analytics, and the commercial reality well enough to connect them.

Oncology launches will keep failing as long as companies treat science and strategy as separate departments. The fix isn't structural. It's about who is in the room and whether they have the range to bridge what the data shows with what the market needs.