← Drug Development Suite Scope & Resource · Vitalis Therapeutics Inc.

Roles & Responsibilities

Download Word
Contents
  1. Why This Document Exists
  2. The Shape of the Organization
  3. Governance — Who Decides
  4. The Functions, One at a Time
  5. The Bodies the Program Cannot Instruct
  6. The External Organizations
  7. Who Decides What
  8. Translating From Roles You Already Know

1. Why This Document Exists

Every other artifact in this suite names functions and headcounts. The organization chart shows eleven boxes. The resource plan shows 109 people and 191,420 hours. The work breakdown structure assigns eleven control accounts to functions that own them. None of that explains what any of these people do on a Tuesday.

This document does. It is written for a reader who has run programs in another industry — software, insurance, financial services, government — and who will recognise the governance, the stage gates, the change control and the earned value, and will recognise almost none of the disciplines doing the work.

The one idea worth taking from this page. A drug development program is not a generalist organization with domain knowledge added. It is an assembly of disciplines that own different kinds of truth, deliberately kept separate. The medical monitor decides what is safe. The biostatistician decides what is proven. Quality decides whether either can be believed. The regulator decides what may be claimed. A program manager who merges those judgements to move faster has not simplified the program — they have removed the checks that make its conclusions worth anything.

That is also why the org chart looks strange to an outside reader. The Program Director is accountable for $243,040,000 and line-manages four people. Everything else is matrix, contract, or deliberately outside the program's reach.

2. The Shape of the Organization

109 people sit inside the program: 102 in ten working functions that carry deliverables, and 7 on the Development Committee, which decides rather than delivers.

FunctionLed byPeopleWhat it owns
Clinical OperationsDr. R. Molyneux
Senior Director, Clinical Operations
24Sites, monitoring, enrolment, clinical supply
Technical Operations / CMCDr. K. Oyelaran
VP, Technical Operations (CMC)
16Process, analytical methods, scale-up, validation
Clinical DevelopmentDr. S. Aldridge
VP, Clinical Development
12Protocols, medical monitoring, clinical study reports
Biostatistics & Data ManagementDr. F. Achterberg
Head of Biostatistics
12SAP, EDC, database lock, analysis datasets
Regulatory AffairsG. Petrossian
Director, Regulatory Operations
10Strategy, submissions, agency interaction, labeling
Pharmacovigilance & Drug SafetyDr. N. Halloran
Head of Pharmacovigilance
8Case processing, signal detection, DSUR
Quality Assurance (GxP)Dr. I. Solberg
Chief Quality Officer
7Audits, CAPA, GxP compliance, inspection readiness
Nonclinical Safety & PharmacologyDr. M. Sørensen
VP, Nonclinical Safety
5Toxicology, pharmacology, study oversight
Program Management OfficeT. Nakashima
VP, Program Delivery
4Governance, planning, risk, reporting
Market Access & CommercialL. Whitcombe
Director, Market Access
4Payer evidence, pricing, launch readiness
109 is the sponsor's team, not the number of people working on this drug. Behind it sit a full-service CRO running the trials, a contract manufacturer making the drug, a central laboratory, outside counsel, and 260 investigational sites each with its own investigator, coordinator and pharmacy. The real population working on VitaFlow is in the thousands. The 109 are the people accountable for it, which is a different and smaller thing. Reading the roster as the whole effort is the most common mistake an outside reader makes.

3. Governance — Who Decides

The Development Committee is the gate body. It does not manage the program; it decides whether the program continues, on what conditions, and with how much money released.

MemberSeatVote
C. TyrrellProgram Director — Chair, seated from the PMONo vote
Dr. A. OkoyeChief Medical Officer — Executive SponsorVoting
Dr. H. VasquezChief Scientific OfficerVoting
Dr. P. RaghunathanSVP, Regulatory StrategyVoting
M. ThornburyChief Financial OfficerVoting
J. BarringtonSVP, Market Access & CommercialVoting
Dr. I. SolbergChief Quality OfficerObserver, no vote

Two features are deliberate and worth noticing. The chair (C. Tyrrell) has no vote: the person who prepares the decision does not get to make it. And the Chief Quality Officer attends as an observer: quality must be able to say what it sees without owning the outcome it is commenting on.

4. The Functions, One at a Time

Clinical Operations — 24 people

Dr. R. Molyneux — Senior Director, Clinical Operations

The largest function, and the one that physically makes the trial happen. Clinical Operations identifies and selects investigational sites, negotiates site contracts and budgets, ships and accounts for investigational product, tracks enrolment against the recruitment curve, and monitors the sites — visiting them, checking that what the case report form says matches what the patient's medical record says. In this program most of that execution is contracted to a CRO, so the sponsor team's real job is oversight of the CRO: agreeing the plan, auditing the work, and owning the outcome when it slips. Note what that means for scaling — this function grows with the number of sites, not with the value of the program.

Clinical Development — 12 people

Dr. S. Aldridge — VP, Clinical Development

Small headcount, high cost per head, because these are largely physicians. Clinical Development owns the scientific design of the trial: what question it asks, which endpoints answer it, who is eligible, and what the comparator is. During the trial they provide medical monitoring — reviewing serious adverse events, ruling on eligibility questions, deciding whether a given event is plausibly caused by the drug. Afterwards they interpret the result and author the clinical study report. When people say a program is "medically led," this is the function they mean.

Biostatistics & Data Management — 12 people

Dr. F. Achterberg — Head of Biostatistics

Two disciplines under one roof. Biostatistics writes the statistical analysis plan before anyone is unblinded, sets the sample size, produces the randomisation schedule, defines the rules for any interim analysis, and ultimately performs the analysis that says whether the endpoint was met. Data Management designs the case report form, builds the electronic data capture system, writes the edit checks that catch impossible values, chases queries with sites, and drives the database to lock. Database lock is the hinge of the whole program: after it, the data cannot be changed, which is precisely why unblinding is only permitted afterwards. Everything a statistician will later claim depends on the analysis plan having been fixed before anyone could see which way the results fell.

Technical Operations / CMC — 16 people

Dr. K. Oyelaran — VP, Technical Operations (CMC)

CMC stands for Chemistry, Manufacturing and Controls, and it is the half of drug development that outside readers consistently underestimate. This function develops the manufacturing process for both the drug substance and the finished product, develops and validates the analytical methods that prove each batch is what it claims to be, scales the process from laboratory to commercial volume, transfers the technology to the manufacturing site, makes the registration batches, runs process validation, and conducts the stability studies that establish shelf life. CMC forms Module 3 of the submission and it can delay a filing entirely on its own, with a perfect clinical package sitting ready beside it.

Regulatory Affairs — 10 people

G. Petrossian — Director, Regulatory Operations

The program's interface to the agency, and its interpreter. Regulatory Affairs sets the strategy — which pathway, which meetings to request, what the label should eventually say — and runs the formal interactions: the pre-IND meeting, the end-of-Phase-2 meeting, the pre-NDA meeting. They assemble and publish the submission itself, handle information requests during review, and negotiate the final wording of the label. A useful way to understand the function: everyone else establishes what is true, and Regulatory establishes what may be said, which is a narrower set.

Pharmacovigilance & Drug Safety — 8 people

Dr. N. Halloran — Head of Pharmacovigilance

Processes every adverse event report, maintains the safety database, reports suspected unexpected serious adverse reactions to authorities on a statutory clock measured in days, and performs signal detection — looking across all events for a pattern no individual case would reveal. It issues periodic safety reports throughout development. Two things make this function unlike anything in a commercial program: its deadlines are set by law rather than by the plan, and its obligations continue for the entire commercial life of the product, long after this program has closed.

Quality Assurance (GxP) — 7 people

Dr. I. Solberg — Chief Quality Officer

The most commonly misread function on this chart. Quality Assurance does not test the product and does not check the statistics. It audits whether work was performed according to the written procedure and whether the documentary evidence would withstand a regulatory inspection. It audits clinical sites and vendors, manages corrective and preventive actions, governs the standard operating procedures, and maintains inspection readiness. "GxP" is the family of regulated practices — Good Clinical, Good Laboratory and Good Manufacturing Practice. The function's independence is the point: it reports outside the program, so it can report badly on the program.

Nonclinical Safety & Pharmacology — 5 people

Dr. M. Sørensen — VP, Nonclinical Safety

The smallest working function and the earliest. Toxicology and safety pharmacology studies establish that it is defensible to give the compound to a human being at all, and at what starting dose. The work continues alongside the clinical program — longer-duration toxicity, carcinogenicity and reproductive toxicity studies are needed to support the eventual label, not just first-in-human dosing. Much of this function's output is invisible until a regulator asks a question that only an animal study can answer.

Program Management Office — 4 people

T. Nakashima — VP, Program Delivery

Four people, including the Program Finance Manager. The PMO owns the integrated schedule, the budget, the risk register, the governance calendar, gate preparation and reporting. It owns no technical content whatsoever. Its product is not the drug; its product is that the program is decidable — that when the committee meets, the evidence in front of it is complete, current and reconciled. This is the function a reader from another industry will recognise immediately, and it is deliberately small.

Market Access & Commercial — 4 people

L. Whitcombe — Director, Market Access

Four people, and their presence at all is a point about scope. They build the payer evidence strategy, which is not the same as the regulatory evidence strategy: a regulator asks whether the drug works and is safe, while a payer asks whether it is worth paying for relative to what already exists. They run health-economic analysis, pricing work and the value dossier, and they drive launch readiness. What they do not do is sell anything. Sales, brand marketing, distribution and lifecycle management are explicitly outside this program — they belong to the business, and the handover to them is a closure deliverable.

5. The Bodies the Program Cannot Instruct

Three groups make binding judgements about this program and take no direction from it. This is a feature.

The Data Monitoring Committee. An independent group of clinicians and an independent statistician who periodically review unblinded safety data while the trial is still running. The sponsor cannot see what they see. They may recommend that the trial continue, change, or stop. A program that could overrule its DMC would not have one in any meaningful sense.

Institutional Review Boards and Ethics Committees. Every site's protocol and consent form is approved by an IRB or ethics committee whose duty is to the participants, not to the sponsor. They are the reason a protocol amendment is expensive: the change must be re-approved site by site, and participants may need to be re-consented.

The regulator. Reviews the submission on its own clock and decides what the label says. No amount of program management changes the review period.

The most consequential lines on the org chart are the ones that leave it. Quality, the DMC and the independent statistician are drawn outside the program on purpose. A program that quietly redraws those lines has not gained control — it has removed its own ability to be told bad news, and it will find out at inspection or at filing rather than now.

6. The External Organizations

Most of the physical work of this program is done by organizations that do not appear on the payroll.

OrganizationWhat they areWhat the program keeps
Meridian Clinical ResearchPhase 2 and Phase 3 execution (D-04)Accountability. Work can be contracted out; responsibility cannot.
Aldergate Biologicsdrug substance and drug productAccountability. Work can be contracted out; responsibility cannot.
Kestrel LaboratoriesCentral laboratory and bioanalytical servicesAccountability. Work can be contracted out; responsibility cannot.
Thorne & Vale LLPOutside regulatory counselAccountability. Work can be contracted out; responsibility cannot.

The single most important sentence about all of them: work can be contracted out; accountability cannot. When a CRO misses an enrolment curve, the sponsor misses the enrolment curve. The contract determines who absorbs the cost, never who owns the outcome.

7. Who Decides What

The single most useful table on this page for a reader learning the domain. The right-hand column matters as much as the middle one.

DecisionWho actually decidesWho cannot
Whether the program continues past a gateDevelopment Committee, by voteThe Program Director chairs it and has no vote
Whether a participant may be dosedThe site investigator, within the protocolNobody in the sponsor organization
Whether an event is drug-relatedMedical Monitor in Clinical Development (Dr. S. Aldridge)The program, on schedule grounds
Whether the trial should stop early for safetyThe independent Data Monitoring CommitteeThe sponsor — it cannot see the unblinded data the DMC sees
Whether an endpoint was metBiostatistics (Dr. F. Achterberg), against the pre-registered analysis planAnyone, after seeing the results
Whether a batch may be releasedQuality Assurance (Dr. I. Solberg)Technical Operations, who made it
What the label saysThe regulator, negotiated by Regulatory AffairsThe sponsor, unilaterally

8. Translating From Roles You Already Know

If your experience is in software, insurance or financial services, these are the mappings that will and will not hold.

If you know this roleThe nearest equivalent hereWhere it differs
Business analystClinical Data Manager, and to a lesser extent the Statistical ProgrammerThe requirement is a Case Report Form and its edit checks. It is specified once, frozen, and thereafter changed only by a documented amendment — because the data it collects is evidence in a regulatory submission.
Project managerProgram or project manager inside the Program Management OfficeSubstantially the same discipline. The difference is that the schedule is gated by biology and by a regulator's queue, neither of which can be crashed with more people.
QA / test engineerThere is no single equivalent, and this is the most common misreadingQuality Assurance (GxP) does not test anything. It audits whether work was done to the written procedure and whether the evidence would survive an inspection. The nearest thing to software testing is Data Management's edit checks and the validation work inside CMC.
Subject matter expertMedical Monitor, Biostatistician, CMC scientist, Regulatory strategistThese are not advisers consulted by the program. They hold decision rights the program cannot overrule — a medical monitor's safety call is not a recommendation.
UAT / user acceptanceDatabase lock, followed by unblindingThere is no user to accept anything. The equivalent moment is the point after which the data can no longer be changed, which is why it is irreversible and heavily ceremonied.
Change control boardTwo separate things: the change control process, and protocol amendmentA protocol amendment is the heavyweight form. It may require re-approval by every ethics committee at every site and re-consent of participants already enrolled, so its true cost is rarely the work itself.
Production supportPharmacovigilanceRuns for the life of the product, not the life of the program, and reports to the regulator on a statutory clock rather than to a service-level agreement.
The mapping that matters most is the one that fails. In most industries "QA" means the people who check the work before it ships. Here, Quality Assurance checks whether the work was done the way the procedure said, and something entirely different — Data Management's edit checks, CMC's method validation, the statistician's pre-registered plan — does the job you would call testing. Reading GxP Quality as a test team is the fastest way to misunderstand how this program is controlled.