Chapter 06 / Understanding Trial Designs Through Real Project Work / Paid beta preview

Chapter 6: Interventional Trials from Phase I to Phase IV

How each development phase changes PM risk, readiness, evidence, and oversight. ## 6.1 A Phase Is Not a Calendar Step Lauren Brooks once brought me a development timeline that looked, at first glance, beautifully simple. Phase I. Phase II. Phase III. Phase IV. Four boxes. Four date ranges. Four neat arrows. It had the visual confidence of something that had not yet met a real study team. The program was DIAB-220, a fictional metabolic-disease development program for an investigational therapy intended for adults with type 2 diabetes and elevated cardiovascular risk. The molecule had completed nonclinical work, the clinical pharmacology team was preparing a first-in-human study, medical was already imagining a proof-of-concept trial, commercial wanted a line of sight to differentiation, and regulatory wanted the team to stop using the word fast as if it were a strategy. Lauren had been asked to build the integrated clinical plan. She did what many capable PMs do early in their careers: she made the sequence visible. Maggie Chen looked at the slide and said, "This is not wrong. It is just too flat." Lauren frowned. "Too flat?" "You made the phases look like train stations," Maggie said. "They are different operating environments." That sentence is the beginning of this chapter. In Chapter 5, we studied observational work and real-world evidence. The protocol did not assign treatment in REGISTRY-RA-10. The central question was whether routine-care data could honestly support a defined decision. Interventional trials are different. In an interventional study, the protocol prospectively assigns participants to an intervention, comparator, placebo, no-intervention or control strategy, procedure, diagnostic strategy, or other protocol-defined exposure to evaluate biomedical or health-related outcomes. The assignment may be randomized [ICH E9] or not randomized [ICH E9], blinded or open-label, controlled or single-arm, simple or complex. The common feature is that the study itself assigns the exposure or strategy being evaluated. That assignment creates obligations. It means the protocol is not only observing the world. It is changing a participant's experience under a research plan. That is why interventional clinical trials carry intense attention to participant protection, informed consent [FDA Informed Consent 2023], eligibility, dosing, safety monitoring, data reliability, investigational product control, protocol adherence, regulatory oversight, and documentation. Phase labels help organize that work, but they can also mislead if we treat them as administrative labels. Phase I is not merely "early." Phase II is not merely "middle." Phase III is not merely "large." Phase IV is not merely "after approval." Each phase asks a different kind of question and creates a different kind of project. For DIAB-220, the program team needed to move through several evidence questions: | Phase | Core Question | PM Translation | |---|---|---| | Phase I | Can humans receive the investigational product with an acceptable early safety, tolerability, pharmacokinetic, and pharmacodynamic profile? | Build a controlled, safety-intensive operating model where dosing, samples, escalation, review, and stopping rules are precise. | | Phase II | Is there enough evidence of biological or clinical activity, and what dose or regimen should move forward? | Manage uncertainty without letting enthusiasm outrun data quality, endpoint discipline, or decision criteria. | | Phase III | Can the product's benefit-risk profile be confirmed in the intended population with evidence strong enough for major regulatory and clinical decisions? | Protect scale, consistency, endpoints, safety oversight, data quality, supply, vendors, sites, and governance. | | Phase IV | What additional evidence is needed after approval about safety, effectiveness, use, comparative questions, or commitments? | Keep postmarketing work rigorous without confusing every real-world question with an observational registry. | That table is simple. The work underneath is not. Daniel Liang told Lauren to keep one sentence in her notebook: A phase is not a calendar step; it is a different kind of evidence machine. That became the operating principle for Chapter 6. When a PM understands that principle, phase planning changes. You stop asking only, "When does Phase II start?" You start asking, "What decision must Phase II support, what uncertainty remains from Phase I, what evidence must be protected, and what would make us stop?" That is project management. It is also humility. Not every development program follows a clean Phase I, II, III, IV path. Oncology studies may begin in patients rather than healthy volunteers. Rare disease programs may combine phases, use smaller populations, or rely on accelerated pathways when justified. Vaccine, device, diagnostic, pediatric, cell therapy, and digital therapeutic programs can use different development logic, regulatory frameworks, and evidence expectations. Some programs never reach Phase III. Some products are studied postapproval in both interventional and observational ways. So the point is not to memorize a rigid ladder. The point is to learn how the operating model changes when the evidence question changes. ## 6.2 Phase I: First Human Exposure and Dose Escalation Phase I is where the word first becomes serious. First human exposure. First dose level. First safety review. First sign that the assumptions from the laboratory, animal studies, modeling, and prior class knowledge may or may not survive contact with human biology. In many drug programs, Phase I studies

...