Subscribe to our podcast, The Life Science Rundown, if you haven’t already.
We’ve been wanting to have a conversation about the benefits of bringing Quality in earlier for a long time.
While we see some teams make progress here, it’s still the case that by the time QA is consulted, many technical decisions have already been made. The capital is committed, the headcount is set, and investor expectations are locked. So when a problem surfaces, the question isn’t whether to fix it — it’s how expensive the fix has to be.
That’s the argument Dr. Andrea Bell makes, and her framing is sharper than the usual case for involving Quality sooner. She argues (and we agree) that late QA involvement doesn’t just add cost. It removes the option to remediate cheaply, which is the whole premise behind right-the-first-time. We see teams find this out the hard way more often than we'd like.
Nick Capman sat down with Andrea to talk about moving QA from late-stage gatekeeper into an early, risk-appropriate technical partner.
Andrea is a biotech executive with more than 25 years of experience leading Quality and CMC organizations across biologics, RNA therapeutics, vaccines, and small molecules. She currently serves as Vice President of Global Quality at BioNTech, where she leads quality strategy for the company’s international mRNA BioNTainer manufacturing network.
Throughout her career at BioNTech, Vir Biotechnology, Alnylam Pharmaceuticals, Biogen, and Amgen, she’s been recognized for building high-performing teams, developing fit-for-purpose organizations and quality systems, and driving complex programs from early development through commercialization, helping bring innovative therapies to patients worldwide.
She holds a doctorate in Global Health Sciences, a master’s degree in Quality Assurance and Regulatory Affairs, an MBA in International Business, and is passionate about advancing global access to transformative medicines.
She was also candid about why the topic matters to her. She’s worked on both sides of the divide, in organizations where Quality wasn’t consulted early or empowered to act, and in companies where Quality was built in as a partner from the start. So rather than the typical theory, she speaks to what she’s actually seen work (and fail).
Apple Podcasts | Spotify | YouTube | Web + Others
Andrea’s key insights and practical takeaways
If you’re short on time, here are the most important lessons from the discussion.
Late QA involvement removes your cheap options. This is the core of it! When QA sits organizationally downstream and isn’t consulted until after technical decisions are made, the costs land after the company has already committed capital, headcount, and investor expectations. At that point, you can still fix the problem. You just can’t fix it inexpensively.
Most of the cost of poor quality sits below the waterline. Andrea referenced the FDA’s iceberg model. Above the line are the direct costs everyone counts: recalls, lost revenue, legal and regulatory expenses. Below it lies the larger, less visible aggregate, including scrap rates, cycle-time delays, market share loss, costly investigations, and reputational damage that can take years to repair or may never fully recover. Those eventually show up in stock performance!
A single 483 carries a real remediation cost before you count the opportunity cost. Andrea cited a range of roughly $250,000 to $5 million in direct remediation for one 483 and noted that the figure excludes delayed clinical timelines and stalled BLAs. Both of those affect patient access.
Early means early in the conversation, not just early in the lifecycle. This distinction is easy to miss. Andrea’s point is that “early” also applies to the decision forums that occur at every phase: design-of-experiments reviews, tech transfer meetings, and the discussions where the technical calls are actually made. Being invited to those is different from being involved early in a general sense.
ICH Q8 grounds the argument. Quality target product profiles and critical quality attributes should be defined before process development begins. Having QA at the table when those are defined, rather than reviewing them afterward, is where the value shows up.
What this looks like across the lifecycle. Andrea walked through it phase by phase. In R&D and discovery, through first GMP, QA joins cell line, cell bank, and asset qualification discussions as they happen, with systems staying fit for purpose and a focus on traceability and data integrity.
In Phases 1 and 2, QA reviews the design of experiments alongside process scientists so critical process parameters get identified upfront rather than reverse-engineered after a deviation.
In late Phase 3 and pivotal work, QA shifts toward process lockdown support, and the CPPs, CQAs, and control strategy get finalized there rather than under filing pressure or in response to agency questions.
At the NDA and BLA stages, validation is already complete, so PPQ confirms a known relationship rather than discovering one, and QA moves from inspection to inspection readiness by stress-testing systems before the PAI.
Post-launch, QA stays in the continuous process verification loop with change control and real-world manufacturing knowledge feeding back into the program.
Risk-based doesn’t mean less rigorous. It means rigor pointed where the risk is. ICH Q9 frames this around two principles: risk evaluation grounded in scientific knowledge and tied to patient protection, and effort, formality, and documentation commensurate with risk rather than maximal by default. Andrea’s read is that applying commercial-level scrutiny at Phase 1 doesn’t make you safer. It misallocates resources away from where the actual risk sits.
The rigor doesn’t loosen; it strengthens! You still identify hazards, analyze and evaluate risk, and put controls in place, using tools like FMEA systematically. What you gain is a defensible reason behind every decision, which is what agencies are looking for. A uniform checklist applied across the board is a weaker position, and agencies will challenge it.
Three questions focus the risk analysis. Is there an impact on the patient, the product, or the process? Andrea uses that frame during Phase 1 and 2 technical evaluations to keep attention where it belongs.
The barriers are cultural, not procedural. This is the part most organizations get wrong. The obstacles to embedding QA earlier are usually a lack of a quality culture and a reporting structure, along with a genuine misunderstanding of QA’s responsibilities. A QA function only involved in late development can often measure nothing but deviation closure rates, which pushes it into a reactive mode no matter how capable the people are.
Three concrete moves leaders can make. Give quality standing membership in the technical forums. Run a formal maturity assessment to establish a baseline worth tracking against, using ISPE’s Advancing Pharmaceutical Quality program or PDA’s quality culture assessment tool. And commit at the leadership level, which ICH Q10 identifies as foundational, meaning resources are committed and quality is genuinely empowered.
Culture follows what leadership measures and rewards. Andrea’s observation from culture survey data. If leadership doesn’t empower Quality and doesn’t reward behaviors that support patient safety, the organization won’t produce it, regardless of what the procedures say.
The Q8, Q9, Q10 trifecta is the operating framework. Q8 defines quality through QTPPs and CQAs. Q9 supplies risk prioritization methods and the tools for determining critical process parameters and controls. Q10 provides the pharmaceutical quality system lifecycle that carries those decisions forward. Newer capabilities, such as real-time release testing and process analytical technology, build on that foundation by generating data during the process rather than only at batch record review.
Data makes QA proactive only if people are empowered to act on it. Both halves are required. Andrea framed the quality management review as where this gets tested, and a functioning one does three things: it surfaces metrics that can be acted on rather than burying them in a report nobody reads until the next audit, it escalates when a threshold is breached instead of waiting for the next scheduled review, and it carries real authority.
Recurring unresolved issues in the QMR cycle are visible to agencies. Andrea was pointed here. Where health authorities have issued 483s or provided feedback, a pattern of issues recurring quarter after quarter without resolution signals either that they aren’t receiving attention or that leadership isn’t empowered. Raising the issue isn’t enough. Showing the data isn’t enough! State of control means acting in real time rather than batching problems downstream, where they compound and sometimes become unsolvable.
QA earns credibility by changing what it brings, not just when it shows up. The gatekeeper posture is arriving to say no. The partner posture is to sit with the technical team, ask what could go wrong and how bad it would be, then work on the problem together. Andrea’s framing is that QA becomes a co-author of the argument you’ll eventually make to the regulator, not the signature at the end.
Regulations are open to interpretation, and agencies want to see your logic. This is the nuance underneath the credibility question. Any regulation means something slightly different in the context of a specific company and program. Regulators want to see a consistent rationale and sound scientific logic behind a decision, rather than evidence that a box was checked. QA’s combination of regulatory knowledge and technical understanding is exactly what produces that.
The three-batch example shows where credibility actually comes from. Three consecutive validation batches became standard industry practice, and Andrea’s question is where that number came from. FDA guidance has moved away from prescribing it. Now you justify your number based on risk and process understanding. Put three down, and you should expect to be asked why three. Three may be too few, or a statistically justified alternative may be more defensible for your situation. Helping build that logic early is where QA earns its seat, not by enforcing a rule.
What to develop if you want to do this work. Andrea’s list starts with genuine curiosity, an open mind, and the soft skills that show up as “how” rather than “no.” Then scientific fluency, real operational knowledge of what you’re overseeing, statistical and risk analysis literacy, working knowledge of the ICH guidelines and tools like FMEA, and data and digital fluency, including how AI-driven tools can be leveraged, how they need to be validated, and what proper quality oversight of them looks like. Together, those build the trust that gets QA invited to the table.
Design governance so early involvement is the path of least resistance. ICH Q10 already lays out the pharmaceutical quality system lifecycle. Instead of a single late-stage gate, build phase-appropriate risk and analytical checkpoints into the process as programs mature. Practically, that means standing membership in technical discussions rather than only change review boards, explicit decision rights, and scoping the risk level up front. The goal is not an approval step bolted on top. It’s making the earliest possible involvement the easiest thing to do.
Three steps to start tomorrow.
Run a baseline quality culture maturity assessment using ISPE’s APQ or PDA’s tool, so you understand where quality actually sits in your development timeline based on evidence rather than anecdote.
Pilot the technical partnership model on a single active program, putting QA in the room for QTPP and CQA definition and DOE planning, and running FMEA jointly with the technical team.
Then establish QMR metrics with real leadership commitment, using FDA’s voluntary Quality Management Maturity program, and supplement deviation closure counts with leading indicators like right first time, cost of poor quality, cycle time, and root cause closure.
One thing to bring back to your team
Pick one active program and look at when QA first appeared in it. Not when QA was formally assigned.
When someone from quality was actually in the room for a technical decision. If the answer is a change review board or a batch record signature, that program is running the model Andrea describes, and the cost of that shows up later, at the point where fixing anything is expensive.
Then ask:
Were your QTPPs and CQAs defined with Quality present, or reviewed by Quality afterward?
Can your QA function measure anything other than deviation closure rates? If not, it has been structured into a reactive posture regardless of how good the people are.
When the same issue appears in your QMR for three quarters running, what happens? Because an agency reading that pattern draws a conclusion about whether your leadership has empowered quality at all.
Andrea E. Bell, DHS, MS, MBA, is Vice President, Global Head of Quality Assurance for BioNTainer at BioNTech SE, focusing on harmonization, phase-appropriate quality systems, and quality risk management. She has spent more than 25 years in biotechnology building quality and CMC organizations across RNA, biologics, and small-molecule programs, and has led complex programs from early development through commercialization. Before BioNTech, she was Senior Director and CMC Leader at Vir Biotechnology and held director-level product quality and QA roles at Alnylam Pharmaceuticals, following nearly nine years at Biogen in analytical project management and product quality roles spanning clinical and commercial biologics, small molecule programs, and combination products. She holds a doctorate in global public health with a vaccines focus from Massachusetts College of Pharmacy and Health Sciences, a master’s in quality assurance and regulatory affairs from Temple University, an MBA in international business, and a bachelor’s in biology.
Connect with Andrea on LinkedIn here.
Who is The FDA Group?
The FDA Group helps life science organizations rapidly access the industry's best consultants, contractors, and candidates. Our resources assist in every stage of the product lifecycle, from clinical development to commercialization, with a focus on staff augmentation, auditing, remediation, QMS, and other specialized project work in Quality Assurance, Regulatory Affairs, and Clinical Operations.
With thousands of resources worldwide, hundreds of whom are former FDA, we meet your precise resourcing needs through a fast, convenient talent selection process supported by a Total Quality Guarantee.
Here’s why 17 of the top 20 life science firms access their consulting and contractor talent through us:
Resources in 75 countries and 48 states.
26 hours average time to present a consultant or candidate.
Exclusive life science focus and expertise.
Dedicated account management team.
Right resource, first time (95% success).
97% client satisfaction rating.
Talk to us when you're ready for a better talent resourcing experience and the peace of mind that comes with a partner whose commitment to quality and integrity reflects your own.
Subscribe to The Life Science Rundown:
Apple | Spotify | YouTube | Web + Others
Check out our newly launched AI-powered QMS audit tool, AICA (the Audit Intelligence Compliance Assistant).
