In this QCast episode, co-hosts Jullia and Tom examine the dose expansion phase in phase one oncology trials — the crucial step that links dose finding to proof of concept. They unpack how expansion cohorts confirm safety, characterise tolerability, and explore early efficacy signals in targeted populations. The conversation highlights how clear objectives, adaptive design, and robust governance ensure reliable data and regulatory readiness. Along the way, they discuss practical approaches to cohort design, decision rules, and data integrity — helping sponsors translate early signals into confident, well-informed phase two programmes.
🎧 Listen to the Episode:
Key Takeaways
What the Dose Expansion Phase Is and Why It Matters
The dose expansion phase follows initial dose escalation in early oncology trials. Its purpose is to confirm safety, refine dose and schedule decisions, and explore early signs of efficacy in the intended patient population. By bridging exploratory dose finding and phase two design, it ensures that subsequent studies start on a sound scientific and operational footing.
Designing Strong Expansion Cohorts
Begin with a clear rationale for each cohort — why that dose, population, and endpoint are being tested. Define eligibility criteria, endpoints, and decision rules in advance, ensuring they align with the recommended phase two dose and biological rationale. Keep cohorts focused and sample sizes pragmatic to reduce uncertainty efficiently.
Maintaining Governance and Data Integrity
Regulators expect structured oversight. Manage adaptations through formal protocol amendments, version control, and clear documentation of rationale and decisions. Maintain consistent visit schedules, endpoint assessments, and data cleaning standards across all cohorts to protect traceability and audit readiness.
Making Sense of Early Activity Signals
Use objective and duration-based endpoints — such as response rate, time on treatment, and clinical benefit — to understand whether activity is meaningful and sustainable. Interpret cautiously, since expansion data are non-comparative, and use findings to shape future trial hypotheses and power assumptions.
Practical Tips and Common Pitfalls
Treat expansion as a disciplined learning phase, not a shortcut to proof of efficacy. Avoid uncontrolled cohort proliferation and eligibility drift. Build in interim reviews for safety and futility, tag subjects by cohort and protocol version, and act on predefined stop–go rules. Consistent documentation and data integrity turn exploratory evidence into a solid foundation for phase two success.
Full Transcript
Jullia
Welcome to QCast, the show where biometric expertise meets data-driven dialogue. I’m Jullia.
Tom
I’m Tom, and in each episode, we dive into the methodologies, case studies, regulatory shifts, and industry trends shaping modern drug development.
Jullia
Whether you’re in biotech, pharma or life sciences, we’re here to bring you practical insights straight from a leading biometrics CRO. Let’s get started.
Tom
In early oncology studies, people hear “dose escalation” all the time, but today we’re focusing on what comes after that. What do we mean by the dose expansion phase, and how does it fit within a modern phase one oncology programme?
Jullia
So, dose expansion is the bridge between first-in-human dose finding and decision-grade evidence for the next phase. After escalation identifies a well-tolerated range, a maximum tolerated dose, or a recommended phase two dose, teams open one or more expansion cohorts. These cohorts enrol a targeted population to characterise safety in a setting closer to real-world use, explore preliminary antitumour activity, and refine the dose and schedule. In practice, that means confirming adverse event profiles, understanding pharmacokinetics and pharmacodynamics in the intended population, and checking that biomarker and response signals justify further investment. Expansion can be single-cohort or multi-cohort, often stratified by tumour type, biomarker status, or line of therapy. The aim is not to prove efficacy but to reduce uncertainty, so that the selected dose heading into phase two aligns with biology, safety, and practical feasibility.
Tom
Building on that, sponsors sometimes treat expansion as a quick way to chase signals. Where do regulators sit on this?
Jullia
Current regulatory expectations emphasise clarity and control. Expansion cohorts should have prespecified objectives, eligibility criteria, and decision rules that connect back to the escalation data. You need a clear rationale for each cohort, including why the selected dose, schedule, and population make sense biologically and operationally. Safety oversight must be explicit, with timely reporting, cumulative safety review, and role definitions for medical monitor, pharmacovigilance, and data review governance. If you open multiple cohorts or add amendments as signals emerge, document the rationale, manage multiplicity in your interpretation, and avoid over-claiming activity. Data integrity matters: consistent assessments, traceable endpoints, and audit-ready processes. Finally, consent language should reflect evolving risks as the dataset matures, and you should keep line-of-therapy labelling and prior-treatment rules aligned to the intended use case. All of this keeps the programme coherent and the transition to phase two defensible.
Tom
Suppose a team has identified a recommended phase two dose during escalation. How do they design expansion cohorts to learn the right things without overcomplicating the protocol?
Jullia
Start with the key uncertainties you need to resolve. If safety at the recommended dose is the priority, focus on enrolling the population that mirrors the planned phase two setting and ensure adequate follow-up for delayed toxicities. If biological plausibility is central, embed pharmacodynamic endpoints or tumour-agnostic biomarkers with predefined sampling windows. Where heterogeneity is likely, consider parallel cohorts by tumour type or biomarker status, each with its own stop-go criteria. Keep endpoints pragmatic: objective response and duration of response are common, but also track clinical benefit rate and time on treatment to understand tolerability and persistence. Operationally, plan screening logistics for biomarker-selected cohorts, ensure central review where needed, and design interim looks for safety and futility. Crucially, tie every assessment to a decision rule. For example, continue if adverse events remain within a threshold and a minimum number of responses is observed; otherwise, close the cohort and record the lessons for protocol amendments or future arms.
Tom
Many programmes evolve mid-flight. New evidence emerges, supply chains flex, and biomarkers are refined. How do you keep scientific agility without creating an inspection headache or muddying the dataset?
Jullia
Treat agility as structured adaptation. Build a change framework into the protocol from the start, describing which elements may adapt, who decides, and how you will document transitions. When you add a cohort or refine eligibility, version the protocol and associated materials together: statistical analysis plan, pharmacy manuals, imaging charters, and data management documents. Tag subjects by cohort and protocol version so you can analyse equally. Keep analysis sets clear; for example, define safety and activity populations per cohort and per dose level. Do not let eligibility creep. If you relax criteria, explain the clinical rationale and its impact on external validity. From a data perspective, maintain consistent visit schedules and assessment windows to protect interpretability. Finally, keep oversight cadence steady: regular safety reviews with medical monitoring, targeted data cleaning, and timely reconciliation of external data like imaging and labs. This preserves flexibility while maintaining a coherent record.
Tom
On endpoints, teams can be tempted to chase every signal in sight. What are the most useful measures in expansion, and how should they be interpreted given the non-comparative design?
Jullia
Expansion endpoints should be decision-oriented. On safety, you want a granular picture: incidence and severity of adverse events, dose interruptions, reductions, and discontinuations, with attention to immune-related or targeted-therapy-specific patterns. On activity, objective response rate can be informative, especially with independent review in settings prone to bias, but contextualise it by line of therapy and prior treatments. Duration-based summaries such as duration of response and time on treatment help assess whether observed activity is durable enough to warrant phase two. Pharmacokinetic and pharmacodynamic data can close the loop between exposure, target engagement, and clinical observations. If exposure is variable, explore whether a schedule adjustment might improve the therapeutic window. Always interpret activity cautiously: these are single-arm data without a control. Use them to prioritise hypotheses and to refine the phase two design, not to make comparative claims.
Tom
Let’s talk sample size and operating characteristics. People often ask, how many patients do we need in an expansion cohort to be confident we have seen enough safety and a credible signal?
Jullia
Size follows purpose. If the goal is characterising safety, you choose a cohort size that gives a reasonable chance to observe events of interest at clinically relevant rates, then ensure follow-up is long enough to capture delayed toxicities. For activity, you can use simple threshold-based decision rules to judge whether the observed responses are consistent with a minimally interesting rate, without inflating claims. Adaptive approaches can be helpful: start with a modest cohort, review interim data at predefined points, and expand or stop based on safety and activity signals. Whatever you choose, document the rationale and the operating logic in plain language. Avoid overprecision. Expansion is not a substitute for a comparative phase two, so aim for enough data to reduce uncertainty and inform design parameters like response rates, variability, and event timing for the next study.
Tom
Operational complexity can spike when you run several biomarker-defined cohorts. What are the common pitfalls you see, and what practices help sponsors stay ahead of them?
Jullia
Three pitfalls recur. First, screening friction: if biomarker testing sits on the critical path, slow turnaround times and discordant assays can stall accrual. Solve this with upfront assay validation, clear sample logistics, and parallel screening pathways where feasible. Second, endpoint inconsistency: mixed assessment schedules across cohorts lead to noise. Standardise timing and methods, and use central review for imaging where bias is likely. Third, unmanaged multiplicity: when many cohorts run in parallel, teams cherry-pick favourable results. Avoid that by prespecifying stop-go rules and reporting outcomes for all cohorts, not only the winners. Good practices include tight site training, early pharmacy and supply planning for new strengths or schedules, and regular cross-functional reviews that bring medical, statistics, programming, and data management together. These habits keep the programme aligned as it grows.
Tom
I’d like a moment for practical guidance. If you had to give sponsors a short checklist for a strong expansion phase, what would make the cut?
Jullia
Here are quick takeaways you can apply today. Define the question for each cohort in one sentence, then align endpoints and assessments to that question. Anchor your dose and schedule rationale in the escalation evidence and exposure-response thinking. Keep eligibility tight and relevant to the intended phase two setting, rather than casting a wide net. Plan interim looks for safety and for futility, and act on them. Maintain traceability with cohort tags, version control, and clean analysis sets. Design consent and patient information to evolve with emerging risks. Finally, write decision rules you can execute operationally; if an event threshold is crossed or a minimum response pattern fails, close the cohort and document the decision. These steps protect scientific value and operational clarity.
Tom
Before we wrap, let’s connect expansion to what follows. How do findings from expansion cohorts shape the design of a strong phase two, and what should sponsors carry forward to avoid rework?
Jullia
Expansion sets the foundations for phase two by refining the target population, dose, schedule, and endpoints. Carry forward a clear definition of the intended treatment line and biomarker criteria, along with any stratification factors you will need later. Use observed safety to power monitoring plans, dose-modification rules, and patient-reported outcomes where tolerability is central. Translate activity signals into realistic assumptions for response rates, event timing, and drop-out, then choose a design that fits the decision context, whether single-arm with a stronger threshold or a randomised comparison. Codify data flows that worked in expansion, including imaging, labs, and external feeds, and retire what created noise. Most importantly, keep the estimand question front and centre: who the question concerns, what variable you will summarise, how intercurrent events will be handled, and which summary measure supports decision-making. That alignment keeps your programme coherent as it scales.
Tom
Thanks, Jullia. Let’s close with a crisp recap. What are the two or three ideas you want listeners to take away about dose expansion in oncology?
Jullia
First, dose expansion reduces uncertainty. It is the structured test of whether a dose and schedule are workable and biologically sensible in the intended population. Second, clarity beats complexity. A small number of well-designed cohorts with clear rules will teach you more than a sprawling matrix of exploratory arms. Third, carry lessons forward. Use expansion to lock down population, endpoints, operational rhythms, and data quality so phase two launches on solid ground. Treat the phase as a disciplined bridge, not a shortcut, and your oncology programme will move faster with fewer surprises.
With that, we’ve come to the end of today’s episode on dose expansion phases in oncology trials. If you found this discussion useful, don’t forget to subscribe to QCast so you never miss an episode and share it with a colleague. And if you’d like to learn more about how Quanticate supports data-driven solutions in clinical trials, head to our website or get in touch.
Tom
Thanks for tuning in, and we’ll see you in the next episode.
About QCast
QCast by Quanticate is the podcast for biotech, pharma, and life science leaders looking to deepen their understanding of biometrics and modern drug development. Join co-hosts Tom and Jullia as they explore methodologies, case studies, regulatory shifts, and industry trends shaping the future of clinical research. Where biometric expertise meets data-driven dialogue, QCast delivers practical insights and thought leadership to inform your next breakthrough.
Subscribe to QCast on Apple Podcasts or Spotify to never miss an episode.
