DMPK stands for Drug Metabolism and Pharmacokinetics. It is crucial to the assessment of the safety of a medicine, as it measures not only what the medicine will do in the body, but also what the body will do to the medicine.

This is a great summary:

The main areas to test are:

• How the drug is absorbed.

• Where it is distributed in the body.

• How the body transforms (metabolizes) the drug.

• How quickly/by what route drug/metabolites are eliminated from the body.

Readers wishing more depth on the topic can refer to Metabolism and Pharmacokinetic Studies on the FDA website.

Module 5 electronic Common Technical Document (Clinical Study Reports)

If safety testing is successful (typically takes 3 years), the next stage is to run clinical trials to study the drug in humans. In the US, the application is known as an Investigational New Drug (IND). The company submitting the application is known as a Clinical Trial Sponsor (CTS).

The CTS is legally responsible for all the data and information submitted in an IND application. The detailed requirements for an IND can be found on the FDA website.

The IND application must contain information in three broad areas:

• Animal Pharmacology and Toxicology Studies

• Clinical Protocols and Investigator Information

• Manufacturing Information [Supply Chain Information]. This is information pertaining to the composition, manufacturer, stability, and controls used for manufacturing the drug substance and the drug product. This information is assessed to ensure that the company can adequately produce and supply consistent batches of the drug.

The drug development expert’s skill sets cover toxicology, drug metabolism, pharmacokinetics, process chemistry, pharmaceutical technology and chemistry, manufacturing and controls (CMC—the entire supply chain).

I’ve purposely keep this short as there is a lot to ‘absorb’ here (excuse the pathetic pun!).

Please do ask any stupid question you can dream up. I’d bet it’s a lot less stupid than you thought - and others will be wishing they’d asked it :O)

Leave a comment

Stay safe,

Hedley

In the last post, I traced how Big Pharma’s mid-1980s decision to sell off manufacturing, distribution, and development research created the fragmented ecosystem we now take for granted — contractors, biotech, virtual pharma, generics, and eventually biologics. I ended with a question I didn’t answer: how does an industry built to make one-size-fits-all products, distribute them globally, and sell them through third-party networks cope with therapies designed for a single patient, delivered at a single hospital, sometimes even in a single home?

There’s no better place to answer that than with CAR-T cell therapy — because the industry has already been trying to solve it for the better part of a decade, and the results tell you almost everything you need to know about where the model breaks.

What “vein-to-vein” actually means

CAR-T therapy starts and ends inside the same patient’s bloodstream, which is exactly the problem. The process runs roughly like this: a patient undergoes leukapheresis, where their T cells are extracted from their blood. Those cells are then cryopreserved and shipped — often thousands of miles — to a centralised manufacturing facility. There, the cells are genetically engineered to recognise and attack the patient’s cancer, expanded in culture, and put through quality control release testing. The finished product is then shipped back to the treatment centre, thawed, and infused into the same patient it came from (ATTC Network).

The industry has a name for the whole journey: vein-to-vein time. And the current median is around 27 days (MedPath), with some published ranges running from three to six weeks depending on product and process (Springer).

Twenty-seven days doesn’t sound catastrophic until you remember who’s waiting at the other end. These are patients with rapidly progressing, often relapsed or refractory blood cancers. Shorter vein-to-vein times are associated with meaningfully better complete-response rates and survival (MedPath); the current lag “can impact CAR-T eligibility for end-stage patients,” as McKinsey puts it — some patients simply deteriorate past the point of being treatable before their own cells come back to them (McKinsey).

There’s a fragility here too, one that echoes exactly the “process is the product” theme from the last post. Fresh leukapheresis material stays viable for only 24 to 48 hours before it must reach the manufacturing site (AllCells). Miss that window and the patient may have to go through apheresis all over again — not a paperwork problem, a re-run of an invasive medical procedure on someone who is already seriously ill.

This is what I meant when I said biologics could be lost “in the blink of an eye.” With CAR-T, it isn’t a batch that’s lost. It’s a person’s only chance.

Leave a comment

Why this breaks the model pharma actually has

Big Pharma spent forty years building a supply chain optimised for anonymity and scale: make millions of identical units, ship them through wholesalers and distributors, and let the product arrive at a pharmacy counter with no meaningful contact between the company that made it and the patient who takes it. CAR-T inverts every one of those assumptions, and the seams show in three places.

There’s no shared nervous system between hospital and manufacturer. Apheresis centres and manufacturing sites typically aren’t electronically integrated, so tracking a specific patient’s cells through the entire journey — what the industry calls chain of identity and chain of custody — often falls back on manual labelling and administrative procedures at the human touchpoints (Insights.bio). There is also no globally harmonised privacy framework governing how a patient’s identity should be protected as their cells move between institutions and across borders (Insights.bio).

Scheduling is a hard, unforgiving constraint. A patient’s apheresis appointment has to be booked against an available slot at the manufacturing facility, sometimes reserved well in advance. A change in the patient’s health, a courier delay, or bad weather can mean forfeiting that slot entirely — pushing treatment back by weeks (Clinical Supply Leader). High-performing centres now schedule apheresis a full 24 hours ahead of the manufacturing start date just to build in a buffer for exactly this kind of disruption.

The cold chain is unforgiving at both ends. Cryopreserved shipments need cryogenic shippers pre-conditioned with liquid nitrogen down to roughly -150°C, and those shippers have a limited working lifespan once conditioned — they have to be coordinated precisely against manufacturing completion and delivery windows that are themselves often uncertain (Cytiva; KPMG). And the fragility isn’t theoretical: transit times exceeding roughly 36 hours between leukapheresis and cryopreservation have been shown to significantly reduce the number of viable cells that survive the process (PMC).

None of these are exotic failure modes. They’re the ordinary friction of two institutions — a hospital and a manufacturer — that were never designed to operate as a single continuous system, now being asked to do exactly that, for a product that dies if they get it wrong.

What getting it wrong actually costs

The economics make the stakes even sharper. CAR-T therapies typically cost somewhere in the region of €300,000-500,000 per patient in Europe and the US (Springer). KPMG has modelled pathways toward a theoretical $30,000 CAR-T therapy as a long-term industry aspiration — a reminder of just how far current costs are from anything resembling accessible, routine care (KPMG).

This is where the two posts connect directly. In the outsourcing era, a failed batch of a conventional drug was an expensive write-off, absorbed somewhere in the cost base of a company making thousands of identical units. Here, a single mishandled shipment is not a percentage loss on a production run — it’s the entire treatment for one specific, named, seriously ill person, and there is no simple “make another batch” recovery. The GMDP discipline and systems thinking I’ve argued for elsewhere in this newsletter aren’t abstract quality-management ideals in this context. They are the only thing standing between a missed handoff and a patient who doesn’t get a second chance.

What the industry is trying

To its credit, the field isn’t standing still. Three approaches are being pursued in parallel, each addressing a different point of failure.

Point-of-care manufacturing moves production to, or very near, the hospital itself — collapsing the two long cold-chain legs (hospital-to-factory, factory-to-hospital) into a short internal walk. Where this has been implemented, vein-to-vein times have dropped from around 27 days to roughly 7-14 days (MedPath).

Rapid manufacturing platforms are compressing the production step itself. Conventional CAR-T manufacturing takes 10-21 days; newer rapid-manufacturing approaches have pushed some processes down toward a matter of days — though quality-control release testing still adds meaningful time on top, since you can’t skip verifying the product is safe before it goes back into a patient (Springer).

Digital chain-of-identity systems are starting to bridge the integration gap, with IoT telemetry streaming temperature, location, and shock data for each shipment, tied to an auditable record binding it back to the individual patient (MedPath).

All three are real progress. None of them is a solved problem yet. Point-of-care manufacturing requires hospitals to build and staff cleanroom-grade production capability they’ve never needed before. Rapid manufacturing still has to clear the same regulatory bar for safety and consistency as the slower processes it’s replacing. And digital tracking systems are only as good as the weakest human handoff they’re layered on top of.

Answering the questions I left open

Last time, I asked how the industry would adapt from global, one-size-fits-all distribution to something built around individual, circular, “vein-to-vein” supply chains. Based on where CAR-T has got to, here’s a provisional scorecard:

• Serving much smaller patient populations? Yes, but at a per-patient infrastructure cost that conventional pharma distribution never had to carry.

• Meeting the needs of individual patients? Being solved, slowly, through decentralised and point-of-care manufacturing — but this remains the exception, not the norm.

• Delivery in home settings? Still largely unsolved. The overwhelming majority of CAR-T infusion remains hospital-based, and nothing in the current logistics model points toward home administration any time soon.

• Genuine relationships between manufacturers and hospitals, pharmacies, and patient care? This is the biggest structural gap of all. Pharmaceutical companies still don’t “touch” the patient the way this therapy model demands — and the manual, non-integrated handoffs described above are the direct consequence.

Where this leaves us

The industry that spent four decades perfecting how to make and move the same pill to a million anonymous patients is now being asked to make and move a unique, perishable, irreplaceable product to exactly one person — and its supply chain, quality systems, and regulatory frameworks were never built for that job. What we’re watching with CAR-T isn’t a logistics inconvenience to be engineered away with better couriers and colder freezers. It’s a live test of whether an entire industrial model, built on scale and distance from the patient, can be retrofitted for intimacy and speed.

The honest answer, right now, is: partially — and only where hospitals, manufacturers, and regulators have been willing to redesign the relationship between them from the ground up, not just move the same broken handoffs faster.

Which raises the question I want to take up next: is this really a logistics problem at all, or is it a governance and quality-systems problem wearing a logistics costume? If chain of identity, cold-chain integrity, and slot scheduling all depend on discipline at the human handoff points rather than the technology layered on top of them, then the fix isn’t a better freezer — it’s a different quality culture. That’s where I want to go in the next post.

Pathway to Prescriptions is reader-supported. If this raised questions you’d like to see tackled next, hit reply or leave a comment — and if you haven’t already, consider becoming a free or paid subscriber to get the next instalment as soon as it’s out.

Leave a comment

In the previous post, below, we investigated how Big Pharma was born in the mid-1980s, as it adopted a new strategy of outsourcing (selling) all it’s manufacturing, distribution, and development research activities, to focus on patenting molecules and sales & marketing activities:

Next, we look at the series of events that followed this ‘strategic realignment’ adopted by the then large, vertically integrated, pharmaceutical companies.

Leavers and funders join forces

This mass divestment of assets resulted in the formation of business models that were not feasible when the Big Pharma companies owned the assets. The management teams in the new contractor base were under enormous pressure to survive beyond the initial contract period that their former employers gave them, typically five years. The search was on for new business, and this coincided nicely with newly “released”

Big Pharma executives looking for opportunities and investors looking enviously at the returns that marketed drugs were providing for their owners. From this marriage made in heaven, biotech companies were conceived.

The idea behind biotech’s appeared attractive. By assembling an executive team and a core staff of scientific and technical experts, the companies could complete all the activities of drug development without actually owning any of the “hardware.” All a biotech needed was the purchase of the appropriate mix of clinical and nonclinical contractors to do the heavy lifting of drug development (File It). The cash to pay the contractors would come from the investors keen to make a return on their money.

This model was further extended into what became known as “virtual pharma,” where the team of core experts was whittled down to a bare minimum, many of them comprising little more than ten or twenty people, often reliant on contractors to educate them in the niceties of drug development and the regulations.

This was considered acceptable because their mission would be to move the process to the point where the compound could be sold to bigger companies, or the entire company purchased en bloc by a suitor with the necessary resources to move things on towards marketed products.

Generic companies pick up the scraps

The dropping of drugs when out of patent also resulted in new companies willing to work to much smaller margins by copying the original product that they had been prevented from doing under patent law.

The Big Pharma companies seemed more than happy to abandon that piece of the market. These newly christened generic models did not have to go through the extensive clinical trials of patented drugs and therefore did not have to recover the costs of failures. The extent of the clinical research required of them was to prove that their copy was bioequivalent. That is, they had to prove the effect of the drug on the patient was equivalent (within an acceptable band) to the original drug. The regulators established the rules for the clinical trials required.

These companies received a welcome boost when the US Congress passed the Hatch-Waxman Act in 1984, actively encouraging the entry of generics into the market.

The intention was to force down drug prices. Other countries, such as the United Kingdom, also made moves to ease the entry of generics, establishing requirements for generic substitution whenever possible. Generic companies supply the vast majority of prescription medicines sold around the world today.

Others pile in

As time has marched on, companies large, small, mini, and micro are now involved in developing and supplying medicines. As university spin-offs and start-ups have entered the fray, we now have an interesting mix of business models:

• Pharmaceutical innovator

• Biopharmaceutical innovator

• generics

• biosimilar

• biotech

• virtual

• speciality Pharma

• university spin-off

• start-up

The models deal in health conditions (indications) ranging from mildly irritating to life threatening and terminal. To add to the models, we have breakthroughs in treatments of diseases.

Biologics lead the way to new therapies

In recent years, new classes of compounds have begun to emerge, namely biologics (or biologicals) and advanced therapy medicinal products. Biologics are made from living things using processes far more variable and complex than the traditional small-molecule products produced using chemistry (chemical synthesis). Their manufacture is also an order of magnitude trickier than making drugs based on chemistry alone. Even seemingly minor alterations in the process can change the product, with potentially devastating effect. This has led to the mantra in biologic Pharmaceuticals that “the process is the product.”

This draws a distinction with small-molecule compounds, where a particular molecule can be reproduced reasonably accurately independent of the facility and equipment used to make it. In biologics, the molecules are so large and complex that it is often impossible to define their molecular structures by analysis. All that is known is that a particular process has produced something that has a particular biological effect on a patient. Other manufacturers may not be able to replicate that product and its effect, even if the process appears to be exactly the same.

That is not the end of it. The sensitivity of biologics to temperature variation and other factors in the environment mean they can be lost in the blink of an eye. A moment’s loss of concentration from an operator or material handler can mean months of work wasted. A temperature data logger not properly validated, activated, or downloaded can yield the same result: valuable product in the bin.

Even that is not the end of it. The potential for input materials to affect yield, potency, and quality of output can be dramatic, as the strength of each new supply of materials can vary widely depending on factors that are not always obvious to the receiving company. Getting to the truth with suppliers, especially when the upstream supply chain leads to seemingly anonymous donors, can be a nightmare and sometimes even impossible.

Even that is not the end of it. The costs of goods for biologics often make a promising compound commercially nonviable. The net result of all the factors is that it is more expensive to develop a biologic than it is a small-molecule drug.

Advanced Therapy Medicinal Products arrive on the scene

Now we have come to the end of it, at least for the time being. A new generation of therapies based on the body’s own biology—advanced therapy medicinal products—is on the rise. These products are similarly biologic in nature, with cell therapy, gene therapy, and tissue engineering. These use the body’s own healing mechanisms and often target conditions associated with a patient’s genetic makeup. The potential to cure disease is phenomenal, but it is still in its infancy. Almost all of the clinical trial work is in the very early stage and involves small numbers of patients in hospital settings.

There is also a subset of advanced therapy medicinal products—autologous cell therapy, which is specific to an individual patient. The patient’s own cells are extracted, modified in some curative way, and then reintroduced into the body.

The diagram below shows the circular, also known as vein-to-vein, supply chain for autologous products.

There are many unsolved supply chain issues in making these therapies available to patients. If we consider that this is an industry that is used to making one-size-fits-all products for global distribution and sale via third-party networks, the picture gets even more concerning.

Pharma product license-holders and manufacturers have little, if any, contact with hospitals, community pharmacies, and patient care at home. If the industry is to respond to therapies that involve much smaller patient populations with specific needs, down to individual patients, even in home settings, how is it going to work?

Maybe you would like to ponder the question while waiting for the next instalment?

What is Find It, File It, Flog It?

‘Find It, File It, Flog It’ is the name I gave to Big Pharma’s business model that gave birth to the blockbuster era. I wrote about it in 2015, in a book self-published with Amazon. This is an excerpt:

Rich pickings begin

Pickings began to get rich starting in the mid-1970s, mostly from the battle of the stomach ulcer drugs Tagamet (Smith Kline & French) and Zantac (Glaxo). Even though Tagamet was the first to market (1976), Zantac overtook Tagamet soon after its launch in 1981 with what was reported to be a superior marketing effort. This seems to have been the birth of the blockbuster era.

By the early 1980s, industry players learned that a patented compound—new molecular entity—with an important license to sell could use nimble marketing to make huge profits under the shelter of patent protection. The industry focused increasingly on patenting as many compounds as seemed reasonable, selected the most promising for development, and then marketed the bones out of them once approved.

Below is a tongue-in-cheek description of the approach. Figure 1 shows a gifted scientist who has found a compound from the patent library that is showing some promise in the test tube. He’s having an ‘eureka’ moment.

Figure 1. The reigning paradigm of drug development (courtesy Expert Witness Dr Graham Cox for the cartoons)

We see that our eager scientist phones his boss, who is under pressure from above to move compounds into development, and he is more than happy to hear the news that the finding has potential. The race is now on to get the magic powder into trials by making larger quantities to test on animals. Figure 2 shows the scientist’s baby hastily passed along the development conveyor belt as the patent clock ticks.

Figure 2. The patent clock is ticking

The scientist moves on to find further medical breakthroughs—or not, as the case may be. Other scientists take over the baton and head toward the finish line. The next step is proving that this compound is safe for clinical trials to begin and has some scientific rationale as to why it is going to work.

In the name of conserving remaining patent life, the industry puts things on fast-forward. The patent fairy (or is it a wicked witch?) is omnipresent, and every pause for thought must be met with a poke of her broom handle or a whack with the bristles.

To recap in the real world, this “lifestyle” approach involved finding a promising patented compound (Find It), placing it into a development pipeline intended for regulatory approval to market (File It), and then marketing the approved product with the utmost verve and vigor (Flog It). In mathematical terms, we have:

F1 + F2 + F3 = $$$.

Where:

F1 = Drug discovery (Find it)

F2 = Regulatory review and approval (File it)

F3 = Marketing (Flog it)

$$$ = Megabucks

This is the equation that has driven the industry ever since—with devastating outcomes. From here on, we will refer to this approach of drug development and commercialization as Triple F, as we continue our exploration of the pharmaceutical industry.

The race to the clinic begins

We now dig more deeply into the “File It” stage, where Pharma companies must prove their compounds are fit to be marketed. We pick up where “Find It” is completed, where a fresh team of scientists takes over from our discovery friends above. They will be conscious of the wicked witch’s presence from years of conditioning. A day lost in development is a day’s patent life and, more important, lots of money in lost sales if it’s a blockbuster. Minds are concentrated appropriately. It’s time to get cracking with preclinical testing.

The basics of drug development

Using the metaphor, we are able to shine a light on the basics of drug development. Drugs (new molecular entities) selected as candidates for market have to be proved safe to test in humans. This is what is known as preclinical testing. It involves proving that the drug is tolerated sufficiently well in animals to give a high level of assurance that it can be tested in healthy human volunteers without causing undue or irreversible harm. There is also an attempt to gather evidence that the drug could be effective in the particular disease state under investigation, but animals are much different from humans, so this is always sketchy at best and useless at worst. There is normally a requirement, though, for scientific rationale as to why the drug should work in theory.

The supply chain is a vital component of the whole system because it is physically what will be doing the magic work during clinical trials and when the drug hits the market. The drug to be fed to animals is less pure than subsequent manufacture of the active ingredient for humans. This is known in the jargon as a “dirty batch.” The logic is clear: as development proceeds, each subsequent batch becomes increasingly pure. The dirty batch is the “worst case,” and if the regulators are happy for a company to use it for initial testing, the process will be easier.

If the preclinical phase gets approval, the company receives a licence to run clinical trials in humans. This is known as a clinical trial application in the European Union and an investigational new drug in the United States.

Phase I is testing in healthy volunteers. All the data necessary are collected and the results assessed. Every clinical study must have an end point that determines whether the study achieves what it has set out to prove. If it meets its end point, then it can progress to the next stage.

Phase II is sometimes split into an A and a B run in patients with the condition. Information is collected and, again, the company judges, through statistical analysis of the results, whether the study has met its end point. If it has, Phase III testing begins with a mission to gain approval from the regulators to market. If the regulators give approval, then it is hats in the air, yachts on order, and succulent sausages all around!

This was how Tagamet, Zantac, and other blockbusters emerged on the market.

This created another mathematical formula for Big Pharma companies to adopt:

S1 + S2 + S3 = $$$

Where:

S1 = Safe Supply-chain (preclinical testing)

S2 = Sample Supply-chain (clinical trials)

S3 = Succulent Supply-chain (marketing approval)

$$$ = Megabucks

As the years rolled on, increasing numbers of Big Pharma companies had success applying this formula. They became known as R&D-based Pharma companies or Big Pharma for short.

The formula is still the basis for the drug development paradigm of today.

Next time, well will cover Pharma’s Crippling Addiction and How to Cure it

Dr Angeliki Kotsianti, physician-scientist and molecular pathologist. She held the final medical judgment on the medicines of a major pharmaceutical company, including decisions on reformulation, recall and supply interruption, and built its global strategy for medically necessary products. She now advises on medical governance and regulatory strategy with PharmaFlow Ltd.

Read Angeliki’s view regarding a A Clinical Perspective on Supply Decisions in The Unconflicted View section of the PharmaFlow website.

For those who prefer to rest their clicking finger, here it is reproduced below:

A Clinical Perspective on Supply Decisions

Author: Dr Angeliki Kotsianti, physician-scientist and molecular pathologist. She held the final medical judgment on the medicines of a major pharmaceutical company, including decisions on reformulation, recall and supply interruption, and built its global strategy for medically necessary products. She now advises on medical governance and regulatory strategy with PharmaFlow Ltd.

Every structural fault in the way medicines are developed and supplied arrives, eventually, as a decision that a doctor has to make about a named patient.

Supply interruption is usually discussed as a question of pedigree, procurement, and track and trace. At the point of care, it is none of those things. It is whether to begin a course that may not be completed, whether to move a stable patient onto an alternative agent, and whether the evidence behind that alternative is strong enough to justify moving them. Those are medical judgments. The clinician making them has usually had no part in the decisions that produced the constraint, and rarely has time to establish what those decisions were.

The track and trace measures introduced after the Heparin contamination would not prevent a recurrence, and the reason has to do with sequence. The harm in that episode was clinical before anyone recognised it as a supply failure. Patients were injured and died while the system still believed the material was genuine, and the clinicians treating them had no route by which what they were seeing could be raised as a question about supply. Reporting runs in one direction. Until that changes, failures of this kind will keep being found out the way this one was, which is late.

The penicillin story is usually told as evidence about manufacturing capacity, and it is. From the clinical side it is also a story about consent. Albert Alexander was started on a course the Oxford team knew might not be finished. Someone weighed an incomplete course against no course at all, for a patient whose name they knew, on evidence that did not yet exist. That was a medical judgment, taken in conditions no protocol covered. The production figures from those months are recorded in some detail. The reasoning behind the decision to treat him is not.

In most companies benefit-risk is a named function with a named owner, and supply is a named function with a named owner. What a supply constraint means for patients belongs to neither. It gets answered late, informally, and often by whoever happens to be nearest to it. In my own experience the judgments that reached me on reformulation, recall and supply interruption had in substance been made already, months or years earlier, when a formulation or a presentation or a dose was chosen and nobody asked what it would mean at the bedside. What arrived on my desk was a choice between poor options.

A class of therapies has now arrived that makes the problem literal. Where a medicine is manufactured from a patient and returned to them, the decisions include whether that patient is eligible at all given the distance to the apheresis site, whether their bridging therapy will hold through the manufacturing window, and whether an out-of-specification product should be given to someone who has no alternative. Manufacturing failure is not a rare event. The UK National CAR T Panel reported a failure rate of just under four per cent after the first manufacturing attempt in large B-cell lymphoma patients approved for commercial autologous therapy, with further patients affected by out-of-specification product or by delay. Each of those is a patient who has already been leukapheresed and bridged. Each is a medical judgment carrying regulatory accountability, and the operating model has nowhere to put it, so it is handled as logistics at one end of the chain or the other. The batch here is a single patient. Scale has become a question about one person at a time, and it is still nobody’s to answer.

Geography is itself a clinical outcome. An analysis of SEER-Medicare data covering patients treated for third-line or later diffuse large B-cell lymphoma found that those who received CAR-T were less likely to carry multiple comorbidities and more likely to live in higher income areas. The same study modelled the effect of distance. Reducing the average journey to an authorised treatment centre in poor-access states from 104 miles to 34 would raise the proportion receiving CAR-T from 6.6 to 9.1 per cent. The relative gain is large. The absolute position is that more than nine in ten of these patients would still not receive the therapy. That is not a logistics statistic. It describes patients accepting less effective treatment because of where they live and what they earn, and it is the same pattern as any therapy that needs infrastructure the local hospital does not have. Decentralisation is a governance problem before it is a manufacturing one.

Those figures are American, and a single-payer system removes the cost of the medicine itself. It does not remove the pattern. UK data show that referral for CAR-T is broadly even across the country, but that patients from the most deprived areas are infused at 73 per cent against 86 per cent for those from the least deprived. Once infused, response, progression-free survival and toxicity are similar. The therapy does not work less well for poorer patients. They are less likely to receive it. Modelling of the English network finds that concentrating advanced cell therapies in a small number of centres increases travel time in every scenario tested, with the burden falling hardest on lower socioeconomic groups and up to 40 per cent of patients left more than an hour from a treating centre. Centralisation is a reasonable answer to a manufacturing problem. It is also a decision about who will find it hard to be treated, and it is not currently made by anyone who has to answer for that.

Pharmacovigilance carries substantial governance because harm caused by a medicine is understood to need a named owner, defined thresholds and mandatory escalation. Harm caused by the absence of a medicine has nothing equivalent. A patient who deteriorates because treatment was interrupted has been harmed as surely as one who suffers an adverse reaction, but no system requires that this is recorded, investigated or reported. Companies supplying medically necessary products should be expected to hold a clinical continuity plan for each one, owned by a medical officer, with escalation thresholds and a documented position on therapeutic alternatives worked out before a shortage rather than during one.

It is well established that a substantial proportion of published findings cannot be reproduced. A newer problem now sits alongside it. Evidence generated or filtered by machine learning is entering development decisions, including decisions on formulation, stability and demand. I have worked on these methods and they can be sound. They can also produce confident output from data that would not survive inspection. Any reform of the medicines system should require that evidence used to justify a development or supply decision is traceable to its source and validated by someone competent to judge it clinically.

Reform of this kind will not be delivered by supply chain professionals alone, nor by clinicians alone. At present the two describe the same failure in vocabularies that do not meet. My proposal is a narrow one. Wherever a supply decision is to be taken earlier or differently, require also that a named medical owner is accountable for what that decision will mean for the patient who receives the medicine, or does not.

Written in 2026 as an appendix to Medicines for the 21st Century, the paper presented to the House of Commons Health and Social Care Committee in July 2019 by Friends of Medicines Modernisation with PharmaFlow Ltd.

The pharmaceutical industry rightly presents itself as science-led. Yet science alone does not deliver medicines: it produces hypotheses, molecules and possibilities. A medicine only becomes an innovation when engineering, development, manufacturing, regulation, supply chains and clinical practice convert that possibility into a reliable product that benefits patients.

That distinction matters because the industry’s record of attrition remains painfully poor. Companies may screen thousands of potential compounds for every medicine that ultimately reaches the market, while many development candidates enter formal programmes with fundamental weaknesses still undiscovered. The result is a long, expensive journey through the “valley of death,” where promising ideas fail under the demands of toxicology, manufacturability, clinical performance, regulation, reimbursement and real-world use.

Science Is Necessary, Not Sufficient

There is a tendency in pharma to treat medicine as uniquely unknowable because the human body is immensely complex. Of course, biology is complex. But so are other fields in which society has learned to create reliable products despite incomplete knowledge: aircraft fly, satellites operate and engines perform without humanity fully understanding every aspect of the systems around them.

The practical lesson is not that science is unimportant. It is that innovation cannot wait for perfect knowledge. It must identify uncertainty, test it intelligently and design around it.

In commercial settings, science becomes powerful when joined with engineering. Scientists may establish whether a molecule can affect a biological target. Engineers and technologists then ask a different question: can this be turned into a robust, usable, manufacturable and economically viable solution for the person who needs it?

That is the journey from an idea to a medicine in a patient’s hands.

The Wrong Question

Systems thinker Peter Checkland drew a useful distinction between scientific and technological thinking. Science places the highest value on advancing knowledge; engineering and technology place the highest value on efficiently achieving a defined purpose.

His example from Imperial Chemical Industries is revealing. A research scientist objected to a project intended to create synthetic leather because natural leather’s three-dimensional structure was too complex to describe accurately. From a purely scientific perspective, the conclusion was reasonable: if the original material could not be fully characterised, it could not be faithfully replicated.

But the technological question was different. It was not, “Can we copy leather exactly?” It was, “Can we create a material that performs satisfactorily where natural leather is currently used?”

That reframing changes everything. The task becomes solving an end-user problem rather than reproducing every characteristic of the existing product.

Pharma needs to ask itself the equivalent question. Are companies primarily studying molecules and molecular mechanisms, or are they developing solutions that work for patients, clinicians, health systems and payers?

Candidate Selection Is the Critical Gap

The greatest missed opportunity may come before a development programme properly begins: when a company selects a candidate to take forward.

The industry invests heavily in getting compounds into clinical trials, but it too often treats clinical entry as the prize rather than as the beginning of a far more demanding test. The candidate must survive the full journey:

• It must be safe enough to develop and use.

• It must show a meaningful effect in humans, not merely in models.

• It must be formulated, manufactured and scaled reliably.

• It must meet increasingly demanding regulatory expectations.

• It must deliver sufficient value for patients, providers and health systems.

• It must be supplied consistently throughout its lifecycle.

Too often, these questions are answered late, after major investment has already been committed. The candidate enters the valley of death exposed to risks that could have been investigated earlier.

This is where modern science should play a more decisive role. Predictive technologies—including computational modelling and simulation, human tissue-based methods, advanced in vitro systems and other ex vivo approaches—can help identify likely failures before a compound consumes years of development effort. They cannot provide certainty, but certainty is not the standard. The standard should be a sufficiently rigorous assessment of whether a candidate is robust enough for the journey ahead.

Innovation Means Commercialisation

Professor Daniel Steenstra, a medical innovator with an engineering perspective, makes the point starkly: pharmaceutical R&D may consume vast resources, yet evidence from attrition rates and development timelines suggests that it often remains ineffective at translating discovery into successful products.

The issue is not a shortage of scientific ideas. It is the lack of a complete innovation system.

Innovation is not creativity alone. It is not the discovery of a new biological mechanism, a promising antibody or a novel gene therapy platform. Those are the beginnings of innovation. Innovation is the successful conversion of an idea into a product or service that creates value—whether by improving health outcomes, improving quality of care or generating sustainable commercial returns.

Other industries have developed practical disciplines to support this conversion:

• Concurrent engineering, where development disciplines work in parallel rather than sequentially.

• Rapid prototyping, where assumptions are tested early and repeatedly.

• Lean methods, which reduce waste, delay and avoidable rework.

• End-user-led design, which starts with the needs of the people who will use, prescribe, pay for or supply the product.

Pharma has adopted elements of these approaches, but not consistently enough or early enough in development.

From Technology Push to Patient Pull

Pharmaceutical R&D often remains dominated by what might be called “uppercase R and lowercase d”: an intense focus on research and on the compound itself, with too little equal attention to development as a multidisciplinary route to patient value.

That can lead to technology push: a promising technology seeks an application because it exists. The stronger alternative is patient, clinician and health-system pull: an unmet need defines the problem, and science and technology are mobilised to solve it.

The difference is profound. A technology-push question asks:

“We have discovered this molecule—how do we develop it?”

A patient-pull question asks:

“What problem do patients and health systems need solved, and what is the most effective, practical and sustainable way to solve it?”

The latter does not diminish science. It gives science a clearer purpose.

The Lesson

Big Pharma does not need less science. It needs science applied more intelligently across the full development system.

That means using predictive methods earlier to challenge development-candidate choices; bringing CMC, manufacturing, supply chain, clinical, regulatory, market-access and patient perspectives into decisions from the beginning; and treating innovation as the successful delivery of a solution, not simply the creation of an invention.

The “valley of death” will never disappear entirely. Drug development involves genuine uncertainty, and some failures are unavoidable. But fewer compounds should be sent into that valley unprepared—like Tennyson’s Light Brigade, charging onward because “someone had blundered.”

The central challenge is therefore not simply to discover more molecules. It is to make better decisions about which ones deserve to become medicines, and to develop them as solutions for end users rather than as scientific projects in search of a market.

The Future Pharmaceutical Supply Chain Starts at the End

What would the pharmaceutical supply chain look like if we began not with today’s fragmented arrangements, but with the patient and the desired end state?

Systems thinker Russell Ackoff offered a useful principle: begin an intellectual journey at its end, then work backwards until you reach where you are today.

That is the approach explored in Chapter 14 of my book: defining a feasible future state for the pharmaceutical supply chain through industrial engineering, systems thinking, historical evidence and practical experience.

The central problem

Modern pharmaceutical supply chains can be highly fragmented. Drug substance, drug product, packaging and distribution may be conducted at separate locations, across different countries and by organisations with different ownership.

Outsourcing can bring valuable specialist capability, capacity and flexibility.

But it can also create disconnections between the product licence holder, the production system, distribution, healthcare professionals and ultimately the patient.

A simple question illustrates the issue:

If a medicine reaches a patient with damaged packaging or a quality defect, how easily can the product be returned, investigated and used to improve the system?

The question is not whether formal complaint, recall or pharmacovigilance processes exist. They do. The question is whether the end-to-end supply chain is designed to enable rapid feedback, investigation, learning and prevention.

A different approach: industrial systems practice

The chapter applies a heuristic methodology—an approach intended to aid learning, discovery and problem-solving where exhaustive analysis would be impractical.

It combines:

• Primary analysis based on industrial engineering and systems thinking

• Supporting evidence from pharmaceutical history and operational experience

• A focus on the fundamental method-study questions:

What should be done? Where should it be done? When should it be done? Who should do it? How should it be done?

These are not merely manufacturing questions. They are questions for the whole medicines-development and supply system.

Lessons from an integrated model

Early in my career at Miles Laboratories in Bridgend (Wales), then part of Bayer, I saw an operating model in which production and distribution were closely connected.

Raw materials arrived at the manufacturing site; finished products were made, packed and supplied to hospitals, pharmacies and, in some cases, patients. Customer links were direct. Quality systems, production records, change control, supplier relationships and product feedback existed within a connected organisational system.

This was not an argument that every modern pharmaceutical supply chain must return to a single company under one roof. It is evidence of an important principle:

Product knowledge, process knowledge, quality accountability and feedback from the point of use must remain connected—even when work is distributed across specialist partners

The future model should therefore retain the benefits of outsourcing while restoring two-way physical and electronic communication between the product licence holder, manufacturers, wholesalers, healthcare settings and patients.

Penicillin: discovery was only the beginning

The familiar penicillin story is often presented as an example of serendipity. But Fleming’s observation was only the first stage.

Penicillin became a medicine because discovery was followed by purification, nonclinical and clinical work, process development, scale-up, industrial manufacturing capability and coordinated supply.

The teams associated with Howard Florey, Norman Heatley and Andrew Moyer were crucial in turning a laboratory observation into a product that could be made in sufficient quantity for patients.

The lesson is straightforward:

A promising molecule does not become a medicine until it can be produced consistently, controlled properly and supplied safely.

Process development and supply-system design should therefore begin early in development—not be treated as a downstream activity after the scientific work is complete.

Cimetidine: product and process evolve together

The development of cimetidine, marketed as Tagamet, provides further confirmation. Rational drug design in the United Kingdom was complemented by major work in the United States to create an economic, scalable and practical manufacturing route.

The result was not simply a successful molecule. It was a medicine that could be manufactured at the scale required for global use.

The lesson remains highly relevant: scientific innovation, CMC development, manufacturing process design, quality assurance, regulatory strategy and patient access are not separate projects. They are connected elements of one system.

Why biologics change the equation

For biologics, the need for integration becomes even more pressing.

Temperature-sensitive materials, intermediates and finished products require a control strategy that extends from raw-material sourcing through manufacture, storage, release, transport, distribution and administration.

For biologics, a supply chain cannot be judged only by whether a shipment arrives. It must demonstrate that product quality, identity, condition and traceability have been maintained throughout the journey.

This requires a genuine two-way interface between production, the PLH, distributors and healthcare providers—supported by shared data, agreed responsibilities and fast decision-making when an excursion, delay or quality event occurs.

Patient-specific therapies challenge the old model

Autologous therapies, including CAR-T products, reveal the limits of a conventional one-way wholesaler model.

In these treatments, the patient is both the source of the starting material and the destination for the finished therapy. Cells are collected at the hospital, transported to a manufacturing facility, processed, tested and returned for administration to that same patient.

This is not a standard distribution transaction. It is a time-sensitive, circular and patient-specific supply chain in which chain of identity, chain of custody, temperature control and quality oversight are critical.

The question is therefore not whether conventional distribution is valuable. It is. The question is whether a system built principally for standardised, one-size-fits-all medicines can safely and efficiently deliver medicines that must be made, handled and finished around an individual patient pathway.

A role for hospitals

Hospitals already possess important elements of the future model:

• Patients and clinicians at the point of care

• Clinical decision-making and treatment capability

• Pharmacy functions and internal logistics

• Buildings, engineering services, equipment and storage

• Planning, governance and patient-record systems

However, most hospitals do not yet possess the full GMP, aseptic-processing, CMC, ATMP manufacturing, cryogenic-storage and quality-assurance capabilities required for advanced-therapy manufacture.

That gap should be viewed as a capability-building agenda—not a reason to dismiss the model.

My visit to the GMP cell and gene therapy manufacturing suite at Guy’s and St Thomas’ Hospital (London) showed what can be possible when controlled manufacture is placed close to clinical care. Properly designed hospital-associated facilities can connect cell collection, manufacturing, clinical oversight and patient administration within a more coherent system.

The transformation challenge

The future will not require every hospital to become a pharmaceutical manufacturer. Nor will it eliminate the need for specialist CDMOs, wholesale distributors or centralised production.

It will require something more fundamental: supply-chain architecture that matches the product, the patient pathway and the risks that must be controlled.

For small molecules, that may mean stronger end-to-end data, quality and feedback links.

For biologics, it means integrated cold-chain control, traceability and rapid cross-organisational decision-making.

For patient-specific therapies, it may mean authorised, care-adjacent manufacturing or finishing capability integrated with hospital pharmacy and clinical operations.

The question for leaders

If medicines are becoming more complex, more temperature-sensitive, more personalised and more closely linked to diagnosis and care delivery, can the current supply-chain model remain unchanged?

Or must we begin, as Ackoff advised, with the desired end state—and work backwards to build the production, quality, distribution and healthcare system that patients will need?

Read Chapter 14:Heuristic Methodology to Define the Transformation

Leave a comment

It is vitally important to consider the E2E supply chain as the overall system that is responsible for the safety, efficacy, and quality of pharmaceutical products. At any of the stages shown in the diagram 1 below, and during transfer between the stages, unanticipated events can have a damaging, if not catastrophic, impact on the product.

Diagram 1 Production supply chain, small molecule products.

By way of recap, E2E means end-to-end. In other words, the entire supply chain from beginning to end, or from raw materials to the point where the end user receives the product.

Note that the product license holder (PLH) is responsible for monitoring the safety of its medicines at every point in the supply chain, under pharmacovigilance regulations.

Production Supply Chain: Small Molecule Products (Originator and Generic)

Diagram 2 shows the stages in the production supply chain for a small molecule product.

Diagram 2 Stages in the production supply chain for a small molecule product.

Finished Product

The finished product stage is where the drug product (DP) is labeled and enclosed in a carton with a patient leaflet. This involves labeling and packing operations to produce presentations suitable for patient use and filing into shipping outers ready for transfer to the distribution channel. None of the packaging is in contact with the product or able to affect product integrity. All printed matter must be closely controlled to ensure that the labeling on the product is as approved in the license. This includes purchased materials and materials printed in-house. Materials are typically based on paper, board, corrugated packaging, and overwrapping films of different types.

The DP inside the carton is enclosed in primary packaging, so the threat to product integrity is reduced. This reduced risk is, however, countered by the need for the labeling and all associated text to conform strictly to the information that has been registered with the regulators; also, mislabeling can have a dire effect on patient safety if it leads to a product being identified incorrectly. The majority of product recalls are related to mislabeling or packaging issues, so there remains a substantial element of noncompliance risk.

In supply chain terms, the challenges of secondary packaging lie in the complexity of end-item variants (stock keeping units [SKUs]) and the lead time of origination of packaging artwork and the associated approvals. Particularly in EU countries, where virtually every country has its own language pack for each product presentation, the range of different SKUs passing through the secondary packaging lines can be extraordinary. In addition, with such a variety of packaging component items, each in different languages, the mix of components delivered to the site of production can be vast.

Drug Product

DP is the dosage form. To produce the dosage form, drug substance (DS) is mixed with nonactive ingredients. These are commonly known as excipients. Excipients help the manufacture of the dosage form and the performance of the drug in the body. This stage is known as a primary packaging operation, involving the encasement of the DP in suitable materials to protect product integrity. This means that the packaging must be capable of protecting against the impact of moisture, humidity, and light, for example, to the extent required by the nature of the product as determined through development. These materials would typically be a bottle and closure, sealed aluminum foil blisters, sachet, vial and stopper, and so on. Once contained in the primary package, the product must remain stable over the registered shelf life of the product.

The important points to know about small-molecule drugs to be produced are these:

They are relatively stable at room temperature (c. 20 °C). That means during storage, they degrade slowly over time. The shelf life is, therefore, long, typically in the region of 2–5 years. The manufacturing processes involved are defined by specifications, and testing must be carried out to ensure manufactured products meet specifications. That means different companies, using different plants and equipment, can produce nearly identical products so long as they use the same route of synthesis.

They must also comply with current good manufacturing practice (cGMP) and formal distribution standards such as good distribution practice (GDP). Transferring the manufacturing technology between production plants is a well-defined series of activities for small molecule products, known as “technology transfer.”

The stage prior to primary packaging is conversion of the active ingredient into the basic dosage form through processes such as blending, mixing, and forming. If the basic dosage form (eg tablets) is to be stored awaiting primary packaging, studies must be carried out to confirm that the storage container (such as a fiberboard drum) can maintain product integrity for a period equal to or greater than the length of time in storage.

These three stages can be carried out independently or in a connected series. If they are independent, there must be adequate and validated containment to cover the period they are held awaiting the next stage. For example, tablets held in bulk prior to being packaged into blisters must be kept in containers that prevent any adverse effect on stability. It is always preferable from a supply chain best practice viewpoint to join the processes so that the temptation to make larger quantities and hold them in store is avoided.

Drug Substance

It should be noted at this stage that the pharmaceutical supply chain is a hybrid of process and discrete industries [1]. In simple terms, finished product (FP) and DP can be counted as discrete units of production, sometimes known as widgets. Losses in production are measured as a percentage of widgets that are rejected against the starting number, such as 4% wastage. In process industries, losses are measured as a yield, such as 96% yield.

The companies in the business of pharmaceutical manufacture and supply must invest time and resources in meeting the standards of regulators for compliance with GMP, good laboratory practices (GLP), and GDP.

From a supply chain viewpoint, suppliers to branded pharma have historically been in a strong position by virtue of the degree of exclusivity afforded by their presence in the regulatory filing. Since these companies’ heritage is based in the competitive world of chemicals, they tend to be skilled negotiators and commercially aware. These are often testing supply markets for those engaged in procurement; also, to be borne in mind is that this sector is capital intensive, and utilization is a key driver. This can be an advantage in times of overcapacity, but of course, a cause of concern where capacity is limited. When this is the case, costs and lead times tend to go up and out, respectively.

Starting Materials

These are materials that will form a significant portion of the molecular compound when it is finished. From the SM stage onward, production, storage, transportation, and distribution must all comply with cGMP and formal distribution regulations.

Production Supply Chain: Biologic Products (Originator and Biosimilar)

Diagram 3 shows the stages in the biologics production supply chain.

Diagram 3 Stages in the biologics production supply chain.

Readers should note that upstream processing corresponds to the production of DS and downstream processing corresponds to the production of DP. The rationale for the difference is unclear to the author.

This extract from the Biopharma Cold Chain Sourcebook, 2011, sets the scene for the challenges in the biologics supply chain [2]:

“Biological products are essentially perishable because they consist of organic matter and are therefore influenced by extremes of temperature and the actinic effect of light. Carelessness in the handling and storage of smallpox vaccine, for example, often results in such injury to the product that it becomes inert, although physically it may appear to be in perfect condition and is of recent manufacture. Smallpox vaccine should be stored in a refrigerator or someplace where the temperature does not exceed 50 °F (converts to 10 °C) (Parke, Davis & Co. 1919).”

Note that it was as early as 1919 when the unique challenges of temperature sensitive materials within the life sciences supply chain were mooted, by Parke, Davies & Co (latterly to be acquired by Pfizer).

This was a warning bell to an industry previously only having used, stored, and handled chemically synthesized products and materials. The risk of degradation due to temperature excursions was low because of the inherent stability of the small molecule chemical structures involved. That was not the case for the biologic supply chain.

This supply chain is very different when compared to small molecule products, due to the nature of living organisms. Biologic production involves processes far more variable and complex than small molecule products. Even seemingly minor alterations in the process can change the product, with a potentially significant impact on the clinical performance of a drug. This has led to the mantra in biologics that “the process is the product.”

This means that unlike small molecule products, where a particular molecule can be reproduced independent of the facility and equipment used to make it. For biologic products, the molecules are so large and complex that it is often impossible to define their molecular structures or identify them by analysis. All that is known is that a particular process has produced something that has a certain biological effect on a patient. Other producers may not be able to replicate that product and its effect, even if the process appears to be the same. This will be covered later as we discuss biosimilars.

A second complication is the sensitivity of biologics to temperature variation and other factors in the environment. When biologics arrived on the scene, additional product temperature ranges were required to keep them alive escalated:

• Refrigerated: from 2 to 8 °C,

• Controlled room temperature: from 15 to 25 °C,

• Freezer: from −25 to −15 °C,

• From −60 to −40 °C,

• Ultra-low Freezer: from −80 to −60 °C,

• Cryogenic (LN2): from −190 to −150 °C.

Biologic products and materials can be lost in the blink of an eye. A moment’s failure in concentration, whether by an operator or material handler, can mean months of work wasted. A temperature data logger that is not properly validated, activated, or downloaded can yield the same result of valuable product in the trash.

Input materials can also be problematic. They can dramatically affect yield, potency, and quality of output, as the strength (titer) of each new supply of materials can vary widely, depending on factors that are not always obvious to the acquiring company. Acquiring accurate, good pedigree information from suppliers, especially when the upstream supply chain leads to seemingly anonymous donors, can be extremely challenging and sometimes even impossible.

Finally, excessive cost of goods sold (COGS) can often make a promising compound commercially nonviable and lead to catastrophic outcomes for the sponsoring company.

Growth of the Biologics Production Supply Chain

In the formative years, the biologic product portfolio was predominantly vaccines, insulin, and blood/blood products. Over time, that expanded into areas such as monoclonal antibodies, hormones, cytokines, and advanced therapies (somatic cell and gene therapy, tissue-engineered products), as well as a range of biologically derived diagnostic materials and tests.

As the success of biologic therapies spiraled, most notably monoclonal antibodies (mAbs), the number of clinical trial sponsors (CTSs) began to increase. Biologically derived materials circulating in the supply chain also grew, to include:

• Antigens

• Antibodies

• Sera

• Plasma

• Albumin/globulin

• Cell-lines

• Blood

• Human tissue

Management of the Cold Chain

These product temperatures must be maintained through storage or en route to the next destination. Temperature data loggers must be included in product packaging containers to continuously monitor real-time temperatures.

Gene therapies are, of course, classed as biologics.

More to come soon. Please do comment if you want to know something specific :)

Leave a comment

What follows describes the ‘pivotal event‘, taken from Chapter 5 of Transforming the Pharmaceutical Supply Chain:

“Large Pharmaceutical Companies Outsource Their Assets

In November 1976, Smith Kline & French (SK&F) launched Tagamet, a drug to combat stomach ulcers. It quickly took off and was dubbed the world’s first “blockbuster” drug (≥ $1 Billion annual sales)

Glaxo launched a competitor product, Zantac (ranitidine), based on a similar compound but produced by a declared cleaner manufacturing process in 1981. Glaxo reportedly detailed physicians on the side effects of Tagamet in relation to Zantac. By 1987, Zantac had become the world’s biggest-selling prescription drug, outselling Tagamet 3:1 at one point.

This was a clear example of the power of marketing messages in differentiating a pharmaceutical product from the competition. It stimulated phenomenal growth in the therapeutic area, and the profits were immense for both companies, on sales of tens of billions of dollars. This was the birth of the blockbuster era.

Competitor companies and their investors were impressed by what the Glaxo strategy had achieved, beginning to grow interest in the power of the patent and marketing expertise.

The diagram below explains further, based on the lifecycle of a compound following the award of a 20-year patent.”

Lifecycle of a patented compound.

Of the timelines concerned, market exploitation was of greatest attraction to investors, naturally.

It turned eyes toward the period between approval to sell and patent expiry, and then leftward to the patent award. This shone a spotlight on the remaining patent life—the time available to exploit the market (orange line).

Consciously or unconsciously, large pharmaceutical companies and investor heads turned toward the activity that was eating up most of the time to exploit the market—product development (green line).

Armed with this apparently powerful strategic model, Big Pharma resolved to increase sales and marketing resources, awaiting molecular compounds coming down the development pipe. Discovery research grew exponentially, as libraries of patented molecules were required to fill future pipelines. Expert statisticians and medics were hired to help marketers frame the messages to doctors, and regulatory affairs departments were expanded significantly.

Coincidentally, during the 80s, other sectors were outsourcing “non-core activities,” claiming significant benefits in risk reduction, plus lower costs. That seemed like the perfect solution. Discovery research and sales and marketing were considered core activities.

The huge increase in resources now allocated to the core activities raised concerns among financial analysts, the investing community, and executive boardrooms. They queried the wisdom of owning drug development, production, and distribution assets, along with the people they employed. The large pharmaceutical companies of the day decided to categorize these activities as noncore.

Designated noncore activities included safety testing, running clinical trials, manufacturing, analytical methods development, movement and storage of goods, and customer services.

The exact sequence of events isn’t easy to pin down, but the results were unmistakable. Production plants and drug development facilities were placed on the market, along with the employees working in them. Exiting senior executives joined together and bought up the facilities, funded by whatever means possible. Their mission was to offer services to Big Pharma companies in return for a fee, under contractual terms. These became known as contract development and manufacturing organizations (CDMOs) and contract research organizations (CROs).

Another cadre of executives, displaced and seeking pastures new, put the assets to good use by setting up to develop drugs themselves. They established small companies that became known as biotech, developing drugs to either sell to Big Pharma or try to get to market themselves.

The CEOs in biotech were making a persuasive case to be the engine house of drug development, citing less bureaucracy and shorter chains of command. Venture Capital investors were impressed and began to dip their toes in the biotech water.

Many of the rest of the redundant staff became consultants. Not the McKinsey kind, more former employees selling their skills back into the industry under contracts of varying lengths. I became one of them in 2005, when I left to set up PharmaFlow.

Also on the agenda was the tricky business of supplying hospitals and pharmacies. Handling customer complaints and dealing with ever more frequent deliveries were not deemed core and Big Pharma handed over all of its warehousing and distribution assets to gratefully receiving wholesalers.

Similarly, specialist third-party logistics providers (3PLs) grew their businesses helping with the burgeoning volumes of materials and products that needed to be stored and transported around the globe.

The final aim of the strategy was the practice of abandoning existing products once the patent expired, as they did not meet the return on investment (ROI) targets the branded versions had enjoyed.

The New Strategy Appeared to Work

This was the first example of a pharmaceutical company detailing physicians to highlight competitors’ weakness to capture market share. It stimulated phenomenal growth in the therapeutic area.”

ENDS

What should we take away from the pivotal event?

This was the genesis of the blockbuster era, where patenting molecular structures rather than manufacturing processes took hold.

The patent monopolies, given a tailwind by power sales and marketing, became the new way of working for Big Pharma…

…Covid vaccines were the end result, four decades later.

Leave a comment

Eli Lilly’s dispute with the US Food and Drug Administration

Eli Lilly’s dispute with the US Food and Drug Administration over retatrutide raises an uncomfortable question for the pharmaceutical sector: where has the industry got to when a major company has to sue its regulator over an issue that should, in principle, be capable of resolution through the formal meeting process?

The question is not whether companies should ever challenge regulators. Courts have a legitimate role when an agency acts outside its statutory authority, applies its rules inconsistently, or reaches a decision that is arbitrary. The question is why a technical classification disagreement—one with obvious implications for development planning, CMC strategy, regulatory pathway, and intellectual-property economics—appeared to become a federal lawsuit before an approval application was even submitted.

The Retatrutide dispute

Lilly asked FDA to classify retatrutide, its investigational obesity and diabetes candidate, as a biological product. FDA instead concluded that the product should be regulated as a conventional drug. Lilly then filed suit in September 2024, arguing that FDA’s decision conflicted with the applicable statute and the agency’s own regulations.

At first sight, the dispute appears technical. FDA’s framework defines a protein, for these purposes, as an alpha-amino-acid polymer containing more than 40 amino acids. Lilly says retatrutide has 41 amino acids and also argues that it should qualify as a product “analogous to a protein.” FDA’s position has been that the relevant molecular construction does not meet its interpretation of the protein definition.(biospace)

But this is not an academic disagreement. The classification determines whether Lilly must pursue a New Drug Application or a Biologics License Application. More consequentially, it affects the statutory exclusivity framework: five years for a new chemical entity under the conventional drug pathway versus 12 years of reference-product exclusivity for a biologic.(biospace)

That gap is commercially enormous. But its scale does not make the legal dispute less revealing. It makes the failure—or perceived failure—of regulatory engagement more significant.

The question the system should answer

A sophisticated sponsor should not have to discover a foundational regulatory-pathway disagreement through litigation.

There are established mechanisms intended to prevent precisely this outcome: formal FDA meetings; written advice; meeting minutes; product-classification processes; Requests for Designation; Type B and Type C meetings; and, where appropriate, escalation through senior review.

These processes exist because drug development requires decisions well before a dossier is finalised. Companies must decide how to design studies, build supply chains, validate analytical methods, scale manufacturing, prepare quality modules, plan comparability packages, allocate capital, and communicate credible timelines to investors and patients.

For an innovative medicine, uncertainty is normal. But uncertainty over the applicable legal route to market is of a different order. It turns regulatory strategy into a binary bet.

The retatrutide case therefore invites a more fundamental concern: did the engagement process fail to produce a clear, adequately reasoned, and reviewable decision—or did the parties reach a genuinely irreconcilable interpretation of the law? Those are very different diagnoses, and they require different remedies.

Litigation is not scientific advice

A court can decide whether FDA has acted lawfully. It can set aside an unlawful decision and remand it to the agency. It cannot replace the sustained scientific and regulatory dialogue needed to develop, manufacture, assess, and monitor a medicine.

Indeed, the court’s mixed ruling illustrates the limitation. The district court vacated FDA’s reasoning on whether retatrutide was “analogous to” a protein, but the issue was sent back to FDA for further consideration. In other words, litigation did not eliminate the need for regulatory engagement; it returned the question to it. (biospace)

That should give the industry pause. If the endpoint of a long legal process is renewed dialogue with the regulator, why was a sufficiently robust dialogue not possible at the outset?

The answer cannot simply be that the stakes were too high. High stakes are exactly when transparent, predictable, senior-level engagement matters most. The point of a mature regulatory system is not to avoid difficult decisions. It is to make them intelligible, timely, evidence-based, and capable of being resolved without forcing sponsors into litigation as a project-management tool.

A wider trust problem

This case arrives against a backdrop of increasingly adversarial relations around GLP-1 medicines, compounding, shortages, exclusivity, pricing, and accelerated market expectations. Lilly has also brought six actions against entities it alleges are selling unauthorised versions of retatrutide while the product remains in Phase 3 development. FDA has stated that unapproved retatrutide products cannot lawfully be compounded or sold to consumers.(reuters)

Those enforcement cases are substantively different. A manufacturer seeking to stop alleged unauthorised, unapproved sales is protecting patients, product integrity, and the legitimacy of the regulatory system. But the contrast is striking: in one set of cases, Lilly looks to law and FDA authority to suppress an unregulated market; in another, it turns to the courts to challenge FDA’s interpretation of its own classification framework.

That is not hypocrisy. It is, however, a sign of a system under strain.

Regulatory legitimacy depends on more than enforcement authority. It also depends on predictable interpretation, procedural fairness, proportionate decision-making, and the capacity to resolve disputes before they become legal warfare. Sponsors must be able to obtain clear advice. FDA must be able to preserve scientific, technical and supply chain independence and avoid being pressured into outcomes by commercial stakes. Both can be true.

What better engagement would look like

The lesson should not be that companies must accept regulatory decisions without challenge. Nor should it be that FDA should negotiate statutory interpretation away in meetings with well-resourced sponsors.

Instead, the system needs a clearer route for resolving novel classification disputes early:

• A defined, time-bound escalation process for high-impact classification questions.

• A written, technically detailed FDA rationale that identifies the evidence, interpretive principles, and precedents relied upon.

• A structured opportunity for the sponsor to respond to the agency’s reasoning before the decision becomes final.

• Independent senior review where the decision may determine the applicable approval pathway and statutory exclusivity regime.

• Publicly usable precedent, suitably redacted, so that the next sponsor does not need to litigate the same ambiguity.

• Clear separation between scientific questions, legal interpretation, and commercial consequences—while acknowledging that all three influence real development decisions.

There will still be cases in which judicial review is necessary. A regulator cannot be the sole and final interpreter of the limits of its own authority. But litigation should be the exceptional safeguard, not an implicit stage of regulatory development.

The real issue

Retatrutide may ultimately be regulated as a drug or as a biologic. The court process and FDA’s renewed consideration will determine that. But the larger concern is institutional.

When an experienced global sponsor and the world’s most influential medicines regulator cannot resolve a definitional issue through established channels—before marketing submission, before a product reaches patients, and before a federal judge is asked to intervene—it points to a process problem as much as a legal one.

The industry should not ask whether Lilly was entitled to sue. It should ask a harder question:

What has to change so that the next genuinely difficult regulatory dispute is resolved through transparent, rigorous and accountable engagement—rather than through years of litigation?

Hedley Rees is the author of Wiley’s Transforming the Pharmaceutical Supply Chain (2025) and Supply Chain Management in the Drug Industry: Delivering Patient Value for Pharmaceuticals and Biologics (2011)

Leave a comment

The Quiet Redesign: What FDA’s Systems-Based Turn Really Changes

For years, “systems-based inspection” has been treated as old news in drug manufacturing circles — FDA’s Compliance Program 7356.002 has organised CGMP inspections around six systems (Quality, Facilities and Equipment, Materials, Production, Packaging and Labeling, and Laboratory Control) since the early 2000s, well before most people currently running a quality unit started their careers (FDA Chapter 5 – Establishment Inspections; FDA CP 7356.002F). So when I saw the phrase circulating again in 2026, my first instinct was to ask what’s actually new — because “FDA discovers systems thinking” isn’t a story. What I found underneath it is.

The six systems were never the news

The systems framework for drugs is genuinely old. CP 7356.002 built inspections around evaluating whether an establishment’s quality system, facilities, materials handling, production, packaging, and lab controls function as an interconnected whole — with a Full Inspection covering the Quality System plus at least three others, and an Abbreviated Inspection covering Quality plus no more than two (FDA – Approaches to GMP Inspection). Biologics got their own version of the same logic under CP 7345.848, organised around six key systems and three critical elements (Scribd – Team Biologics Inspection Guide). This is decades-old architecture, not a 2026 announcement.

What’s actually happening this year is that FDA is rebuilding the rest of its inspectional enterprise — devices, food, and the agency’s internal governance — around that same logic simultaneously, while quietly raising what counts as evidence of a system actually working.

What changed on February 2, 2026

The clearest concrete shift is on the device side. When the Quality Management System Regulation (QMSR) took effect, FDA retired the Quality System Inspection Technique (QSIT) — the four-subsystem checklist that had structured device inspections for decades — and replaced it with Compliance Program 7382.850 (FDA – QMSR; FDA Law Blog). Four QSIT subsystems became six QMS areas — Management Oversight, Measurement/Analysis/Improvement, Design and Development, Change Control, Outsourcing and Purchasing, and Production and Service Provision — plus four “Other Applicable FDA Requirements” covering MDR, corrections and removals, tracking, and UDI (FDA Law Blog).

Two details matter more than the relabelling. First, investigators now start by reviewing the company’s own risk management file before they walk the floor, using it to decide where to press (FDA Law Blog). Second — and this is the one I’d flag hardest to anyone holding a device or drug license — management review records, internal audit records, and supplier audit reports, which were carved out from FDA review under the old §820.180(c) exemption, are now explicitly inspectable (FDA Law Blog; Exponent). Documents written for internal, self-critical use are now written knowing FDA can read them.

The enterprise rewiring behind the curtain

The device reset didn’t happen in isolation. In May 2026, FDA formalised the Inspectional Affairs Council (IAC) through a new Staff Manual Guide — a cross-agency governance body pulling together every FDA Center, the Human Foods Program, and the Office of Inspections and Investigations to coordinate inspection strategy at the enterprise level for the first time (FDA – Investing in FDA’s Inspectional Enterprise). FDA’s own language for the goal is “right-sizing” — explicitly defined as “aligning inspectional effort with risk, complexity, and public health need” (FDA – Investing in FDA’s Inspectional Enterprise). A facility with a clean, stable compliance history “may justify a lower inspectional priority”; one with novel technology or known concerns “may warrant greater inspectional attention” — the same sentence, doing two very different things to two different kinds of establishment (FDA – Investing in FDA’s Inspectional Enterprise).

The foreign-inspection numbers show how far this has already moved. Unannounced inspections in India rose from 47% of the total in FY 2025 to 87% year-to-date in FY 2026; in China, from 19% to 61% over the same window (FDA – Investing in FDA’s Inspectional Enterprise). FDA frames this as closing the domestic-foreign “parity” gap and calls it “a public health imperative,” not just an operational tweak (FDA – Investing in FDA’s Inspectional Enterprise).

A two-speed system is emerging

At the other end of the risk spectrum, FDA piloted one-day inspectional assessments starting in April 2026 — shorter, AI-assisted screening visits across food, biologics, medical products, and clinical research, with roughly 46 completed by late April, most closing as No Action Indicated (FDA – One-Day Inspectional Assessments; Inside Health Policy). FDA is explicit that this isn’t a loosening of standards for everyone — it doesn’t apply to higher-risk or more complex facilities — but it is generating facility-specific risk scores that feed directly back into how future oversight gets targeted (FDA – One-Day Inspectional Assessments). Layer on the parallel BRIDGE project pushing routine domestic food inspections toward state partners under a shared risk framework (FDA – The BRIDGE Project), and the direction is unmistakable: FDA is bifurcating its own oversight model, doing measurably less at facilities it trusts and measurably more — with access to documents it never used to see — at facilities it doesn’t.

The real shift is evidentiary, not structural

Here’s what I think gets lost in coverage that treats all of this as a modernisation story. “Systems-based” was never the innovation — CP 7356.002 has said that for over twenty years. What’s changing is what FDA will now accept as proof that a system works. Under the old model, an investigator largely confirmed a procedure existed and was followed on paper. Under the new one, the investigator can go straight to the company’s own internal audit findings and management review minutes and ask why a problem the company already knew about wasn’t fixed before FDA arrived (FDA Law Blog). That’s not a procedural update. It’s a much higher evidentiary bar, and it falls hardest on establishments that have been treating internal audits as a compliance ritual rather than a genuine control mechanism.

This is where the outsourcing model gets exposed. Most originator and generic companies no longer run their own plants — they rely on CDMOs, while the Product License Holder retains the NDA, BLA, or ANDA and the legal accountability that comes with it. If FDA can now read a CDMO’s internal audit reports and supplier audit trail as inspection evidence, the question every PLH should be asking isn’t “did we pass the last inspection” — it’s “have we ever actually seen what’s written in the audit documents our manufacturing partner is now required to hand over.” For a license holder whose name is on the application but whose operations sit inside someone else’s facility, that’s a live gap, not a hypothetical one.

What this means in practice

The firms that come out ahead in a right-sized, two-speed inspectional model are the ones with a genuinely functioning quality system and the paper trail to prove it — they get shorter, less frequent, lower-friction encounters with FDA. The firms that come out worse are the ones that have been managing deviations and CAPAs as isolated, closed-out line items rather than as a system that learns — because that’s precisely the kind of gap a risk-file-first, records-open inspection is now built to find. If you’re a PLH relying on a CDMO for manufacturing, the practical move isn’t to wait for the next EIR. It’s to get contractual and practical visibility into your manufacturing partner’s internal audit and management review records now, on your own terms, rather than finding out what they say at the same moment an FDA investigator does.

---

Sources: FDA – Chapter 5: Establishment Inspections · FDA – CP 7356.002F · FDA – Approaches to GMP Inspection · FDA – Quality Management System Regulation (QMSR) · FDA Law Blog – Understanding FDA’s Risk-Based Inspection Model Under QMSR · Exponent – FDA’s QMSR for Medical Devices · FDA – Investing in FDA’s Inspectional Enterprise · FDA – FDA Launches One-Day Inspectional Assessments · Inside Health Policy – FDA Launches One-Day Inspection Pilot · FDA – The BRIDGE Project

This is a bittersweet moment for me

What follows here was written by Dr Ray Perkins. To save on words, you can see him in this short video of a conference I hosted in Wales on the 8th of May, 2019:

MEDICINES FOR THE 21ST CENTURY: SAFE, BETTER, CHEAPER (he is the only American citizen in the conference)

Dr Perkins subsequently contributed an expert statement to my second book for Wiley, titled Transforming the Pharmaceutical Supply Chain.

Sadly, Ray passed away with a heart attack last year, so is no longer with us. My condolences and flowers were passed to his wife, Teri, who accompanied him to the conference, and his family.

When you read his contribution to the White Paper written to summarise the conference findings, I think you will agree that this is a catastrophic loss to the world of medicine.

Appendix C: Medicines in the 21st Century, Contribution by Dr Ray Perkins, 8 July, 2019

“Here at the end of the second decade of the 21st century, the word “broken” best describes our ability to diagnose and treat disease. Our ability to select therapies that work is broken. Our ability to diagnose, even define, disease is broken.

Our ability to perform research that is reproducible is broken. Our fundamental hypothesis of biology is broken. The list goes on. This is certainly a depressing and, to many, surprising state of affairs. Commercial and non-commercial media outlets are crowded with excited faces assuring us that something akin to immortality is just around the corner. Even old and once-respected science publications now are filled with only “positive” findings, with headlines adjusted to secure maximum attention.

Institutions and universities employ public relations offices whose influence on scientists’ careers equals that of their peers. This segment of the larger document touches on the nature and extent of brokenness, holding a mirror up for a public viewing. In doing so it is hoped that those in a position to foster change will recognize the analyses for what they truly are: opportunities.

Now that it is known how not to proceed, the 21st century can still become the century of genuine progress in medicine. Topics will be briefly addressed proceeding from the clinic to the lab to the offices of decision makers.

The application of medicine is broken. Specifically, the ability to select medicines for patients presenting with symptoms is worse than hit-or-miss. In 2018 in the United Kingdom, some one billion prescriptions were written. Of that number, the best that can be said is that 900 million of those did nothing, providing neither benefit nor harm. The reality is much worse. Nine of ten patients are exposed to side effects, some worse than the disease itself.

Misuse of antibiotics leads directly to drug resistant strains of bacteria. Of course, cost is an issue given that the cost of a successful treatment, on average, is ten times the cost of individual treatments. The ability of any given medicine to “work” is expressed in a simple metric: the Number Needed to Treat or NNT.

No medicine is universally effective, i.e. has an NNT of one. Indeed, NNTs range from a “miracle” treatment with an NNT of five, a “good” medicine that works for every tenth person, to common medicines with NNTs of 40-50. There are even NNTs of infinity for prescribed medicines that don’t work at all.

NNTs must become the lingua franca for patients, clinicians, scientists and institutions.

Biological research is broken. In 2005 a seminal publication appeared, “Why Most Published Research Findings Are False” by John P. A. Ioannidis. Viewed over three million times, this publication became the warning klaxon. It is now universally recognized – by scientists and institutions alike – that a minimum of 50% of published findings cannot be reproduced. Heavily publicized, this “Reproducibility Crisis” has not been effectively addressed.

With the crisis now dating back over decades, scientific literature is corrupted, given that which half is good vs. bad cannot be discerned. Within the context of this document, it must be acknowledged that the foundational “knowledge” for development of new medicines and diagnostics is likewise corrupt. The same judgment also applies to selection of therapies and is, therefore, a contributor to the high NNT values of the pharmacopeia.

The Genes-are-Destiny model is broken. Readily refuted by simple arguments, genes-are-destiny persists as the century-long basis of biology. Gene mutations, for example, have long defined “treatable targets” in drug and diagnostic discovery. As such, it must be seen as a primary contributor to their failures. Even the recent “breakthroughs” in cancer immunotherapy reinforce this conclusion, all exhibiting NNTs comparable to chemotherapy.

The picture is understandably bleak. Enormous amounts of money are wasted on therapies that don’t work, and development of new therapies and diagnostics are hamstrung. However, there are findings that point the way forward. The impact of microbiome findings has only scarcely been felt. Humans are composite creatures and must be approached as such. Further, the failure of genes-are-destiny must be seen in contrast to its inverse: biology must be approached at the functional level of complex, interacting networks of molecules, cells, tissues, organs, organisms and the environment. Within this new and bracing context, the activity of proteins becomes paramount, as well as the relationship between that activity and biological function.

“Functional Proteomics” must occupy centre stage in the eyes of researcher and decision maker alike…

…Else a pig is simply a ham sandwich.

Recommended Reading:

Dance to the Tune of Life: Biological Relativity, Denis Noble, Cambridge University Press, 2016

The Black Swan: The Impact of the Highly Improbable, Nassim Nicholas Taleb, Random House, 2007

Functional Proteomics, Chapter One, “Making the Case for Functional Proteomics,” Ray Perkins

Rest in Peace, Dr Perkins, your work must not be in vain.

This is one of his contributions as an expert witness in my book:

Subscribe now

Pharmaceutical Products

In terms of the different types of pharmaceutical products, the definition from the Encyclopedia Britannica fits our purpose.

A pharmaceutical [product] is a substance used in the diagnosis, treatment, or prevention of disease and for restoring, correcting, or modifying organic functions.

Readers should note that I have inserted the word product into the definition, as the substances themselves require further processing before meeting the definition. The Encyclopedia Britannica also lists the products that fall into the pharmaceutical category:

Vitamins and minerals (functional medicines).

Medicines.

Vaccines.

Diagnostics (in vitro): In vitro diagnostics (IVD) are tests done on samples such as blood or tissue that have been taken from the human body. They can detect diseases or other conditions and can be used to monitor a person’s overall health to help cure, treat, or prevent diseases).

Medical devices.

Blood components for transfusion.

What Is a Drug (Medicine)?

Second on the list are medicines, more commonly also known as drugs in the United States. A drug is defined as containing an accurate dose of a pharmaceutical active ingredient (API), also known as drug substance (DS), in a form that is easy to use and can be effectively administered. To attain the status of a drug, the following objectives must be met:

• The drug must be shown to be safe.

• It must be shown to be effective.

• A process for manufacturing must be developed (end-to-end supply chain), along with test methods and controls to assure its quality.

There are two broad categories of drugs (medicinal products):

• Small-molecule products—those manufactured using chemical synthesis (industrial chemistry). Aspirin is an example.

• Biologic products (biologics)—those manufactured using biological processing of living organisms. A monoclonal antibody (mAb) is an example.

In more recent years, a subclass of biologics has gained traction in the industry, known as advanced therapies in the United States, or advanced therapy medicinal products (ATMPs) in the European Union. ATMPs comprise somatic cell therapy, gene therapy, and tissue-engineered products.

These are divided into two further subcategories:

Autologous therapies—using cells specific to each individual patient. Chimeric antigen receptor T-cell (gene-modified cell therapy) is a recent example.

Allogeneic therapies—using cells to treat a broad spectrum of patients.

Industry Business Models and the Supply Chain

Big Pharma

Prior to the 1980s, there was a single business model for researching, developing, and marketing prescription drugs. These companies were highly vertically integrated across the prescription drug life cycle. Production facilities were wholly owned by the pharmaceutical companies and generally located in every country where the companies marketed their goods. In-house production activities covered the sourcing of raw materials, the manufacturing of all stages required to produce packaged goods, and distribution to hospital and community pharmacies. Supplies for test materials for preclinical and clinical trials would be produced in-house. The staff carrying out the analysis of safety and efficacy results, as the trials progressed, would also be employed in-house, as would all the activities required to submit license applications to the relevant regulatory authority.

From the early 1980s onward, the then large, vertically integrated pharmaceutical companies began to outsource to third parties the activities required to develop new drugs, produce them for clinical trial supplies and commercial sale, and distribute them. This was a major strategic initiative, apparently aimed at reducing the fixed-cost base and passing some of the risk of drug failures onto its suppliers.

The large pharmaceutical companies retained and expanded the existing discovery research and sales and marketing competencies, deeming them core to the business. The result of this strategic realignment was a fundamental shift in the industry structure and the business models therein.

The large pharmaceutical companies, such as Pfizer, Merck, and Glaxo, became known as “Big Pharma.” Their role was to discover molecular compounds and secure their intellectual property rights (IPR), in the event that a molecular compound was selected as a development candidate. The other retained in-house skill sets, sales and marketing, assumed the role of demand creator for products in the development pipeline, along with priming demand for existing marketed products. Table 1 shows the top 10 Big Pharma companies in 2022.

Table 1 Top 10 Big Pharma companies in 2022. (Author note: FY 2022 is chosen as the approximate mid-way of the SARS-CoV-2 outbreak)

Contract Organizations

Big Pharma’s vacuum in drug development, production, and distribution competencies was filled by displaced employees working in the divested facilities, along with new entrants to the industry. A third-party service provider model grew rapidly, working on a fee-for-service basis, and became known as Contract Research Organizations (CROs) and Contract Development and Manufacturing Organizations (CDMOs).

Diagram 2 shows the top CROs.

Table 2 Top CROs in 2022.

Table 3 shows the top CDMOs in 2022.

https://uk.practicallaw.thomsonreuters.com/9-543-2565?transitionType=Default&contextData=(sc.Default)&firstPage=true

Table 3 Top CDMOs in 2022.

Generic Drugs

For generic small-molecule products (copies of the originals), the U.S. Congress passed the Hatch–Waxman Act in 1984 to proactively encourage the entry of generics into the market.

The intention was to force down drug prices. Other countries, such as the United Kingdom, also made moves to ease the entry of generics, establishing requirements for generic substitution whenever possible.

Developing a supply chain for a small-molecule drug is comparatively straightforward. Small-molecule drugs are not overly sensitive to temperature variation and typically remain stable for between 2 and 5 years. The development of a copy of an originator drug (generic) requires the developing company to submit an abbreviated NDA and undertake bioequivalence studies. The aim of these studies is to prove that the drug produced by the generic supply chain has a clinical effect broadly equivalent to that of the originator drug. The FDA website has some useful information for non-specialists:

“Generic drugs are copies that one company makes of a brand-name drug that was developed by another company. Generally, generic drugs sell at lower prices, and it is in the public’s interest to get generic drugs to the market quickly. But, like any other scientific and regulatory process, approval of a generic drug takes time. It takes FDA time to review the complex information needed to demonstrate that a given generic drug can be substituted for the brand-name drug that it copies, and that time also depends on the complexity of the drug product and the completeness of the application. Here is why:

Prescription drugs have significant, sometimes life-saving, positive effects, but they also may present significant risks. FDA approves a drug only after review of extensive testing showing that a drug will provide the benefits described in its labeling, and that those outweigh its risks.”

The FDA Office of Generic Drugs follows a rigorous review process to ensure that, compared to the brand-name (or innovator) medications, the proposed generic medications:

• Contain the same active/key ingredient;

• Have the same strength;

• Use the same dosage form (for instance, a tablet, capsule, or liquid); and

• Use the same route of administration (for instance, oral, topical, or injectable).

FDA’s review process ensures that generic medications perform the same way in the human body and have the same intended use as the brand-name medication. The FDA review for licensing is under an Abbreviated New Drug Application (ANDA).

Healthcare professionals and consumers can be assured that FDA-approved generic drug products have met the same rigid standards as the innovator drug. All generic drugs approved by the FDA have the same high quality, strength, purity, and stability as brand-name drugs. In addition, FDA inspects facilities to ensure the generic manufacturing, packaging, and testing sites pass the same quality standards as those of brand-name drugs.

Table 4 shows the top 10 generic companies in 2022.

Table 4 Top 10 generic companies in 2022.

Biosimilar Products

In more recent years, copies of the original biologic product have emerged, and it has become more challenging to prove that they are sufficiently similar to the originator product. We will learn in later chapters that the proof of bioequivalence is an order of magnitude more involved than for generic small-molecule drugs.

Biotech Company (Sometimes Referred to as Virtual Pharma)

The biotech name for a business model was originally adopted from biotechnology, which was an emerging idea for developing new drugs in the 1970s. Baking bread is an early example of biotechnology, where a living organism (yeast) is used to produce the product (bread). Brewing is another example. The perceived potential of making drugs from living things led to some drug development companies naming themselves biotech companies. British Biotech is an example, which was initially named British Biotechnology.

A biotech company has the intention of developing a drug up to a predefined stage, to sell its IPR to Big Pharma, in exchange for compensation payments. These payments are normally based on milestone payments and an agreed share of royalties. Biotech companies raise the majority of their funds from venture capital, beginning with initial seed funding, followed by further rounds of funding. The funds are used to build sufficient infrastructure and employ the skills to begin early-stage development of a compound. This only became a possibility when the fee-for-service contract organizations were formed.

Specialty Third-party Logistics Provider (Specialty 3PL)

With the geographic spread of the supply chain, as the contract organizations consolidated through acquisition in different geographic locations, demand for logistics services spiralled. Initially, it was specialists, with competencies in handling drugs, that felt the benefit. World Courier and Marken (Hedley joined the Marken Advisory Board in June 2012) were the main players in the market. Subsequently, freight forwarders and integrators (3PLs that own aircraft) entered the market.

Hopefully, this is sufficient to digest for now.

What is the point of this?

The COVID era demonstrated that doctors, other healthcare professionals and the layperson had been suffering from a lack of understanding of how pharmaceutical products arrive on the hospital or community pharmacy shelf.

Even with the limited knowledge I have shared here to-date, you know enough not to be taken in by the ‘drugs can be developed in 12 months’ slogan.

That is a definite plus as far as I’m concerned.

The other factor is that doctors have been side-lined by the big pharma companies, as all those companies are focussed on patented molecular compounds.

We need to bring doctors back into drug development to collaborate with experts in drug development WHO ARE NOT CONFLICTED IN ANY WAY.

To explain the new way to develop drugs I have been advocating through my writing in professional journals and published educational books, this little explainer animation should give you the idea (only 2.5 minutes):

Developing drugs the penicillin way

Yes, it took around 15 years to get penicillin to market. The world’s educational systems has been getting it wrong all this time!!!

Be back soon with more of the same brain food,

Hedley

Leave a comment

In Part 2, we continue with the basics of drug development and regulation. The aim is to help subscribers appreciate the challenge and complexity of keeping a pharmaceutical supply chain safe, as a minimum, and effective, as far as possible.

If doctors and other healthcare professionals were familiar with these aspects of drug development, they may have asked more searching questions during the approval of the SARS-CoV-2 sterile injectables.

If you are feeling this is all a bit dry, please bear with it for a wee while longer. Also, if you have any questions, no matter how simple, SHOUT OUT! The comment button below is the place to do it:

Leave a comment

Electronic Common Technical Document

Companies wishing to apply for a license to sell a drug must submit an eCTD in support of an NDA/BLA/MAA application. The FDA website contains an further details on the eCTD.

eCTD comprises three modules:

• Module 3 – Chemistry, Manufacturing and Controls (CMC)

• Module 4 – Nonclinical Study Reports

• Module 5 – Clinical Study Reports

A fully completed eCTD must be submitted to the relevant regulatory authority and evaluated in detail before a new drug can be authorized for sale.

The time taken for evaluation and review of the eCTD can be anything between 12 and 18 months, sometimes longer.

During that time, there is a cycle of questioning carried out by the regulatory authority to leave no stone unturned in assuring patient safety and sufficient efficacy of the drug produced by the supply chain, as detailed in the eCTD. Authorization to sell is based upon the result of the evaluation and review process. Preapproval inspections are carried out, as a minimum, of all drug substance (DS) and drug product (DP) manufacturers (producers). An unsuccessful inspection is likely to result in failure of the application.

With approval of the drug, the clinical trial sponsor (CTS) becomes a product license holder in the United States and a marketing authorization holder (MAA) in the European Union. These licenses lay out the obligations that must be respected by the license holder.

Drug Development Programs

The program of development is dictated by regulatory rules and guidance. Details can be found on the FDA website, titled The Drug Development Process.

In countries outside of the United States, the regulatory authority website will have the details required.

In layperson terms, the two supply chain stages of development are:

Nonclinical Testing—testing in animal models to prove acceptable safety to allow the person(s) or organization(s) studying the compound to progress to studies in humans. It is important to remember that safety of a prospective drug must be evaluated on the product of the supply chain that produced it. This is a recurring theme that will run through the book.

The person(s) or organization(s) become a CTS once an application to run trials in humans has been approved by the relevant regulatory authority.

Clinical Testing—this involves testing in humans to prove that the compound is safe, effective, and can be produced to the required quality (supply chain). To be able to test a drug in humans, a license from the regulatory authority is required, known as an investigational new drug (IND) [14] in the United States, and a clinical trial application (CTA) in the European Union [15].

The IND/CTA application must contain information in three broad areas:

• Animal Pharmacology and Toxicology Studies

• Clinical Protocols and Investigator Information

• Manufacturing Information [Supply Chain Information]

Sufficient safety data that will subsequently be included in Modules 3 and 4 of the eCTD must be collected and analyzed by the prospective CTS. Typically, the data are collected from the CDMO(s) and CRO(s) carrying out production and testing.

If the prospective CTS considers the data can support an application to embark on trials in humans, it must apply to the regulatory authority using an IND application in the United States. In the European Union, it will be a clinical trial application (CTA). This is the stage where the least is known about the safety and manufacturability of the compound.

The detailed requirements for a CTA can be found on the FDA website.

The phases of clinical development are typically:

• Phase 1 studies in healthy volunteers

• Phase 2a dose ranging studies

• Phase 2b studies with the chosen dosage form

• Phase 3 larger studies in the chosen dose form

• Phase 3 pivotal studies in preparation for license application

Regulatory Good Practices (GxP)

GxP is the general term for regulations to ensure safety, efficacy, and quality of drugs. These are made up of:

• Good Laboratory Practice (GLP)

• Good Clinical Practice (GCP)

• Good Manufacturing Practice (GMP)

• Good Distribution Practice (GDP)

All four apply across the supply chain in certain areas. The predominant ones relating to the supply chain are good manufacturing practice (GMP) and good distribution practice (GDP in the European Union).

In the United States, GMP is termed current Good Manufacturing Practice (cGMP) and there are no specific regulations titled GDP. cGMP is included in the U.S. Code of Federal Regulations (CFR). FDA’s portion of the CFR is Title 21, which interprets the Federal Food, Drug, and Cosmetic Act and related statutes, including the Public Health Service Act.

The main elements of cGMP

21 CFR Part 210: cGMP in manufacturing processing, packing, or holding of drugs [19].

21 CFR Part 211: cGMP for finished pharmaceuticals.

21 CFR Part 600: cGMP biological products.

FDA is known to rely on the U.S. Pharmacopeial Convention (USP) document titled USP <1079> for distribution standards.

In the European Union, GMP and GDP have been combined in recent years to form GMDP. This reflects the growing importance of transport, storage, and distribution in the pharmaceutical industry.

One role of a regulatory authority is to inspect and license companies and organizations in the supply chain, to ensure compliance with the conditions laid down in the product licenses to sell. This is known as process licensing.

There is an amount of regulatory flexibility when it comes to standards in the supply chain for producing preclinical test material, given the early stage of development.

For trials in humans, however, production in the supply chain intended for use in humans must comply with GMDP in the European Union and cGMP/USP <1079> in the United States.

Before a compound can be administered to humans it must be cleared for use in humans, by both animal studies and in vitro (test tube) assessments. Initial toxicity studies are designed to determine a safe starting dose in humans.

The toxicity testing bar is raised as human clinical trials progress. This comes under the heading of nonclinical testing. Nonclinical means it includes safety testing preclinic and the manufacturing supply chain, which produces the compound for testing. Readers wishing for more depth on the safety can refer to Drug Safety Information—FDA’s Communication to the Public on the FDA website.

Key Aspects of GMP [23], GDP [24], and Distribution Standards [25]

The following are the main elements of GMP as applied to the pharmaceutical supply chain.

Organization

The regulations require that a quality control unit be established whose responsibility it is to approve or reject materials and check and approve relevant documentation.

There must be a clear organizational reporting relationship that allows the quality unit to be independent of the production function so that no undue influence can be brought to bear when difficult decisions about product disposition must be made.

For example, rejection of a batch of products can have dramatic cost implications in terms of both the material lost and the potential opportunity cost of sales revenue forgone. Commercial pressures can therefore be intense, and in the opinion of the regulators, organizational separation is the only sure way to ensure that patient safety comes first.

Other aspects of organization cover the need for personnel to be properly qualified, trained, equipped, and informed of their job requirements. Staff must have relevant knowledge and experience of the areas in which they work, and their resumes must demonstrate that and be available for inspection. Standard operating procedures (SOPs) should be written and approved only by suitably qualified staff. Those working for SOPs should be kept abreast of the latest versions and training given, where relevant, in any changes to working practices. Training files must be kept up to date to provide evidence of training and competence to perform duties to the required standards. There must also be a sufficient number of personnel available to carry out the tasks. This may seem irrelevant to regulatory compliance matters, but in practice, it is vital.

Quality Management System

The regulations require that SOPs be established for all aspects of GxP. These SOPs, together with policy and guidance documents, should make up a fit-for-purpose management system to ensure product quality. These must be carefully documented and controlled made available to all relevant staff, who must be trained and confirm that they are competent to work under the SOPs. In recent years, there has been a move from the regulators to broaden quality systems and utilize the principles of ISO 9001/2 and similar systems. ICH Q10 is intended guidance for companies that wish to develop their quality management systems along best practice lines.

Quality and/or Technical Agreement

The regulations require that a written contract is in place between companies where services are provided by a third-party undertaking activity that may affect GxP. In the regulations, this agreement is referred to explicitly. 21 CFR states: “There should be a written and approved contract or formal agreement between a company and its contractors that defines in detail, the GMP responsibilities, including the quality measure of each party.”

For example, if a company is sponsoring a clinical trial and wishes to have the API made by a third-party contractor (CDMO), they will be the contract giver, and the contract manufacturer will be the contract acceptor. Similarly, if the contract manufacturer outsources the delivery of its service provision, there must again be a contract in place. In the United States, this contract, commonly termed as a quality agreement, ensures that obligations for quality assurance activities such as reporting and acting on out-of-specification results are clearly defined. The supply agreement will then normally contain the specifications, process, and working methods.

In E.U. countries, the contract document, more commonly called a technical agreement, includes both the quality procedures and the technical information relating to specifications. This has led to a degree of confusion between the two areas. In practice, both work when the supply agreement and quality/technical agreement are taken together. Personally, I prefer to see the technical information and quality procedures in a single document to enable easier documentation of changes. This is because the supply agreement rarely changes once executed, whereas technical details and quality arrangements tend to change more frequently. This makes it better to have quality and technical agreement, which becomes a working document that is properly updated under changing control.

Process Validation

Validation aims to confirm and document that the output from a process or operation is as specified. In the case of press producing tablets, for example, the producer must prove that under defined and documented production conditions, the tablets manufactured will consistently meet specifications. There are a number of stages to validation that follow the development life cycle:

• Design qualification: confirms that the design meets requirements.

• Installation qualification: confirms that the installation meets requirements.

• Operational qualification: confirms that the operation meets requirements.

• Process qualification: confirms that the entire process meets requirements.

Validation must be carried out to a protocol, sometimes called a master validation plan. 21 CFR defines a validation protocol as: “a written plan stating how validation will be conducted and defining acceptance criteria. For example, the protocol for a manufacturing process identifies processing equipment, critical process parameters and/or operating ranges, product characteristics, sampling, test data to be collected, number of validation runs and acceptable test results.”

Although validation is vitally important in ensuring that the product is produced as intended, it can also be extremely prohibitive to change, since it is a resource-hungry activity. One possible solution is that instead of defining processes as a single set of parameters, the development stage actually explores a range of the most relevant parameters (critical quality attributes [CQAs]) to define a “design space.” Then validation would be required only if operations extended beyond the space.

Change Control

Changes to any aspect of the supply chain must be carefully controlled. This involves assessment of the impact of any proposed change on safety, efficacy, and quality of the product. Companies are required to have a standard operating procedure that defines all the responsibilities and activities involved. Many companies form a change control board, so that assessment can be made more easily with all the key personnel present. The main stages are as follows:

• Define the change clearly.

• Perform an impact or risk assessment.

• Make a cross-functional review.

• Identify needs for validation, verification, and other risk mitigations.

• Sign-off on the implementation plan.

• Implement change according to the plan and document accordingly.

Historically, pharma companies have been averse to changing anything that would require authorization by the regulators. The reasons and counters for this are something to be considered in the light of the opportunity for improvement.

Stability

Products tend to degrade over time under the ravages of temperature and humidity effects, for example. Stability is a measure of how quickly or slowly that happens.

Some compounds are as “stable as old boots,” as the term goes. Others may last only days or weeks. In general terms, biological compounds are far more sensitive than small molecules to temperature and humidity. This means that cold chain storage and transportation is a major consideration in biologics. This does not mean that it is never a concern in small molecules. In the early stages of development, where relatively less is known about the stability profile, the evidence may not exist to assume that a compound is not temperature or humidity sensitive. This means that material may need to be stored and transported under controlled temperature conditions. This is an important point for those responsible for the supply chain to bear in mind; it is a compliance risk that has grown in recent years. Clearly, it is important to know the profile of degradation for any product so that it is not consumed in an inferior state.

Investigations and CAPAs

Any out-of-specification result or unplanned deviation must be investigated to find the root cause, which must then be corrected through corrective and preventive actions (CAPAs). Pharmaceutical companies are required to document the investigation carefully and not close the file until a satisfactory solution to the quality issue has been identified, approved, and implemented. It is the responsibility of the trial sponsor or license holder to ensure that this happens. If the manufacturer is the sponsor or license holder, this will be covered under the quality system SOPs of that company.

If the manufacturer is under contract, this must be covered within the quality and/or technical agreement.

Customer Complaints

Companies must monitor records and respond to customer complaints relating to products. They should then have a process of correcting any deficiencies identified and reporting any matters relating to safety to the appropriate competent authority.

Traceability and Recall

Traceability and recall are classic supply chain concerns. As a product is made and distributed, it is possible that quality defects could be identified in one of two ways:

• A finished product in the marketplace could be found to cause adverse patient events or to exhibit some other unacceptable problem.

• A constituent material in the product could be identified (by a supplier of manufacturer) as being of suspect quality.

In either case, the supply chain must be halted while investigations take place. This requires identification of all the batches of products that could be involved and taking them out of the system by quarantining them. To do this, there needs to be forward (in case 1 above, where did the product go to) and reverse (in case 2 above, where did it come from) traceability.

As mentioned, please let me know if you want me to cover a specific query you have in relation to the subject matter!

Leave a comment

Building understand for Subscribers

Now that Pathway to Prescriptions is up and running, it is time for me to share the foundational knowledge required to gain a sound knowledge of how the pharmaceutical industry works are a more detail level.

Remember, Hedley has written two book for Wiley covering the Pharmaceutical Supply Chain, and all related matters, in great depth, namely:

Transforming the Pharmaceutical Supply Chain, 2025.

Supply Chain Management in the Drug Industry: Delivering Patient Value for Pharmaceuticals and Biologics, 2011.

I’ve split it up into PARTS so that it is easier to consume. This is PART 1.

Regulatory Authorities

Regulatory authorities hold delegated authority for the safety, efficacy, and quality of drugs in a specific country or region. The general term for such bodies is competent authority, which is a legal entity with regulatory powers delegated by the government. Nearly every country in the world has an authority with the same mission.

In the United States, the regulatory authority is the Food and Drug Administration. The regulations were originally laid down in the Federal Food, Drug, and Cosmetic Act (FD&C Act, 1938 and subsequently incorporated into the U.S. Code of Federal Regulations (CFR), under Title 21.

In the European Union, the regulatory authority is the European Medicines Agency (EMA). Regulations are laid down in E.U. Directives. The principles of regulation are the same, with some differences in methods. Most regulatory authorities across the world have websites that will expand on the details contained in this chapter; also, since regulations and guidance documents are subject to change, readers should seek current versions before committing to any actions. The aim of this chapter is to help readers with navigation aspects when seeking answers to their questions.

The two authorities above, along with the Pharmaceuticals and Medical Devices Agency (PDMA) in Japan, signed up to global harmonization of regulations under the facilitation of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use. ICH was established in Switzerland in 1990. Their mission, published on its website, is to achieve greater harmonization worldwide to ensure that safe, effective, and high-quality medicines are developed, registered, and maintained in the most resource-efficient manner while meeting high standards.

Note: Common FDA drug definitions can be found at Drugs@FDA.

Role of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH)

Read more

Subscribe now

On 8th May 2019, I hosted a conference at Techniquest, the oldest science museum in the UK, nestled in the beautiful Cardiff Bay.

The title is in the heading, and following the conference I compiled a White Paper summarising the dialogue over the day.

I then sent it to the Clerk of the House of Commons, Health and Social Care Committee for the attention of Chair Jeremy Hunt MP and Committee Members.

This a taken from the stated aim:

“Aim

This paper is presented in the context of rapidly increasing volume and intensity of calls for major reform of companies developing and supplying medicines into healthcare systems (pharmaceutical companies). The modus operandi of those companies appears to be working against the best interests of those using their products – patients and healthcare professionals.

On May 8th, 2019, a group of clinicians, patients, representatives of relevant charities, experts in product development, legal, regulatory and supply chain specialists gathered together in Wales, with the aim of examining that claim, based on facts and evidence.

The day was organised in conference format, titled “MEDICINES FOR THE 21st CENTURY: Safe, Better, Cheaper”. It involved in-depth dialogue and transfer of knowledge, over three panel sessions, between invited attendees and panel members, considering issues and opportunities in relation to safe medicines, better medicines and cheaper medicines. Proceedings over the day were recorded on video, and live polling was used to collect inputs from those in attendance.”

This is the headline video of the event:

The gentleman shown above is Dr Ray Perkins, a genius in drug discovery and development. Ray sadly passed away towards the end of last year, but left a legacy of his knowledge that he contributed to Transforming the Pharmaceutical Supply Chain.

I will be sharing his insights as we go through future newsletters.

One of the attendees at the Conference, Alan Kennedy, provided the video below to set a marker for the state of the industry at the time.

This was the reason why I hosted the conference - to find and share a better way to develop drugs.

Seven years later, it has got worse

This was just before COVID arrived - it provided even more urgent evidence of the need for massive change in the pharmaceutical business model based on #PatentsNotPatients.

I’m looking forward to sharing my knowledge with you in Pathway to Prescriptions, so that you can challenge those who are still to learn how bad this has become, based on facts and evidence.

Please, feel free to comment, that’s how we all exchange thoughts and opinions.

Leave a comment

Share

Last time, we asked whether GLP-1 drugs like Ozempic and Wegovy were themselves the risk, or whether the risk sat in the system built around them. The honest answer was the second one. But that raises an uncomfortable follow-up question: once a drug is approved, licensed, and out in the world being used by millions of people the clinical trials never saw — who is actually watching it?

Picture this. A medicine has been on the market for two years. It has passed every trial, been reviewed by regulators on both sides of the Atlantic, and is now prescribed to hundreds of thousands of patients a month. Then a pattern starts to emerge — not in a lab, but in real prescribing data, adverse event databases, and clinicians comparing notes. Nobody designed a trial to look for this. Nobody could have, because the population using the drug in real life is bigger, older, sicker, and more varied than any trial could ever be.

So who is legally on the hook to notice, investigate, and act on that pattern?

The answer has a name, and it is not “the regulator.” It is the Product Licence Holder — called the Marketing Authorisation Holder (MAH) in UK and EU law, and functionally equivalent to the NDA or BLA holder in the US. This is the company whose name sits on the licence, and it carries a legal duty that does not end the day approval is granted. It starts there.

The approval is not a finish line — it’s a transfer of custody

It is tempting to think of drug approval the way we think of passing a driving test: a one-off hurdle, cleared once, valid forever. That is not how licensing works, and it is not how it is supposed to work. When the MHRA, EMA, or FDA grants a licence, they are not certifying that a medicine is permanently and unconditionally safe. They are certifying that, based on the evidence available at that moment, the benefits outweigh the known risks — and they are handing the licence holder an ongoing, legally binding obligation to keep checking that this remains true for as long as the product is on the market (UK MHRA guidance on MAH responsibilities; EU Medicines Evaluation Board).

In other words: the licence is not a trophy. It is a standing contract with the public, renewed every single day the product stays on the shelf.

That contract has a name too — pharmacovigilance — and it is far less glamorous, far less funded, and far less scrutinised than the multi-year, multi-billion-pound process that got the drug approved in the first place.

A live case, unfolding right now

You don’t have to look far for an example. In June 2025, the World Health Organization issued an alert on semaglutide medicines — Ozempic, Rybelsus, and Wegovy — flagging a possible association with a rare form of vision loss, non-arteritic anterior ischaemic optic neuropathy (NAION). The European Medicines Agency subsequently recommended updating the product information to include NAION as a very rare side effect (WHO, June 2025). None of this showed up in the original registration trials. It surfaced through real-world observational data and case reports, gathered after millions of prescriptions had already been written.

That signal has kept moving. In July 2026, Australia’s Therapeutic Goods Administration issued its own safety advisory covering the whole GLP-1 class — semaglutide, tirzepatide, dulaglutide, and liraglutide (ABC News / TGA, July 2026). New Zealand’s Medsafe opened a formal monitoring period on the same issue that is scheduled to close on 26 July 2026 (Medsafe). Meta-analyses published this year put the excess risk in context — small in absolute terms, but real enough that regulators across four continents have felt obliged to act (Neurology, May 2026).

This is pharmacovigilance actually working — signal detected, evidence weighed, label changed, prescribers and patients informed. It is also a reminder of the uncomfortable lag baked into the system: years of exposure at scale before a rare-but-real risk becomes visible enough to act on. The question worth sitting with is not “did the system fail?” It didn’t, this time. The question is: how much of that outcome depended on the licence holder doing its job properly, versus depending on regulators, academics, and clinicians noticing something the holder’s own systems were supposed to catch first?

That’s where the free version of this newsletter has to stop being comfortable and start being precise about who is actually accountable, for what, and on what timetable. Paid subscribers get the rest below — including the specific legal machinery that is supposed to make this work, and the structural reasons it sometimes doesn’t.

For paid subscribers: the machinery behind the promise

Read more

On 29 January 2026, the MHRA quietly updated the safety information for every GLP-1 and dual GLP-1/GIP receptor agonist licensed in the UK — dulaglutide, exenatide, liraglutide, semaglutide, tirzepatide, all of it — to flag something starker than the usual list of side effects: rare but real reports of necrotising and fatal pancreatitis. Between 2007 and October 2025, the regulator logged 1,296 Yellow Card reports of pancreatitis linked to this drug class. Nineteen were fatal. Twenty-four were necrotising (MHRA Drug Safety Update, 29 January 2026).

That’s the headline most outlets will run with this week. It’s not, in my view, the real story.

Buried in the same document is a single line that tells you where the actual risk in this drug class now sits: “privately prescribed GLP-1s … may not appear on the patient’s medical history.” Read that twice. The regulator responsible for monitoring a medicine’s safety is explicitly warning clinicians that they may have no idea a patient is even taking it.

That’s not a drug safety problem. That’s a governance problem. And it’s the subject of this issue.

What the approved drugs actually carry as risk

Let’s deal with the known, labelled risks first, because they matter and they’re not nothing.

GLP-1 receptor agonists — sold as Ozempic, Wegovy, Mounjaro, Zepbound, Trulicity, Saxenda and others — work by mimicking a gut hormone that slows digestion and signals fullness to the brain. That mechanism is also where most of the trouble comes from:

• Gastrointestinal effects (nausea, vomiting, diarrhoea, constipation) are common and usually dose-related.

• Acute pancreatitis is uncommon but, per the MHRA’s fresh warning, capable of severe and fatal outcomes (MHRA).

• Pulmonary aspiration under anaesthesia: since November 2024, every drug in the class carries a label warning that delayed gastric emptying can leave residual stomach contents during sedation, with rare reports of aspiration during surgery (WebMD).

• Suicidal ideation: the FDA has been evaluating a possible signal since 2023. Its most recent review found no clear causal link in either FAERS reports or clinical trial data, but hasn’t ruled out a small risk and continues to monitor it (FDA Drug Safety Communication).

• Thyroid C-cell tumours: a standard boxed warning since launch, based on rodent data whose relevance to humans remains unresolved.

None of this is secret. It’s in the patient information leaflet. A GP prescribing an approved product, at an approved dose, with the patient’s full history in front of them, is operating inside a system built — imperfectly, but genuinely — to catch these signals and act on them.

That system is not where most of the new risk in this category is coming from.

This is where it gets structural — and where I want to spend the rest of this issue. Paid subscribers get the full breakdown below: where the real exposure sits, who’s supposed to be catching it and isn’t, and what the MHRA’s own words are quietly admitting about a two-sided blind spot in how these drugs are now reaching patients.

If you’re reading this as a free subscriber, thank you for being here — the explainers above stay free, always. What follows is the deeper governance analysis Pathway to Prescriptions now runs as a paid feature.

Upgrade to paid →

Subscribe now

Read more

In 2023, I self-published with Amazon KDP: COVID Supply Chains: Fact NOT Fiction

Even though I have published two books with the global educational publisher Wiley, in 2011 (451 pages) and again in 2025 (350 pages), and they cost in the region of $100 each, I have more fingers on my hands than copies of COVID Supply Chains: Fact NOT Fiction sold.

That said, I can still share some of the contents here for the benefit of my free and paid subscribers.

This is what I envisaged as the typical audience for my work

You will be curious, someone who likes to ask questions and get answers, based on facts and evidence. As well as being curious, you may well find the contents here useful in your chosen line of work or profession. That includes people working inside and outside the industry, where a deeper knowledge of how a drug travels from the laboratory bench to the prescription pad will help them carry out their work more effectively. You may have been prescribed a medicine that later proved unsafe, or you may simply wish to understand the system before you, or someone you love, has to rely on it.

There was a tragic event in 2008 where at least 81 patients in the US died and 785 were seriously injured after oversulfated chondroitin sulfate, a cheap counterfeit substance, was substituted for genuine raw heparin sourced from a Chinese supplier and processed into Baxter’s finished product¹.² The contamination entered the supply chain undetected and reached patients through entirely legitimate prescriptions.

Despite intense activity between Governments, Regulatory Authorities and industry, there is nothing to prevent it happening again. That, and the issues of counterfeiting, diversion, and the widening gaps in oversight as a drug passes from developer to distributor to pharmacist to patient, will be covered later.

Finally, you may want to get your questions about how a drug reaches your prescription answered. Questions such as:

“How many years does it really take to get from a molecule to a prescription?”

“Why do so many promising drugs fail in clinical trials, and what happens to the ones that don’t?”

“Who decides what counts as ‘good enough’ evidence before a drug is approved?”

“What safety net exists once a drug reaches your pharmacist — and who is watching it?”

¹2008 Chinese heparin adulteration, Wikipedia, https://en.wikipedia.org/wiki/2008_Chinese_heparin_adulteration

²Blossom, D.L. et al., “Outbreak of Adverse Reactions Associated with Contaminated Heparin,” New England Journal of Medicine, 2008, https://www.nejm.org/doi/full/10.1056/NEJMoa0806450

Then there is the COVID era

You may have wondered:

“Why did so many critical items go into short supply?”

“Why were over 90% of raw materials sourced from China?”

“What risk management plans were in place?”

Most tellingly “Who was in charge of it all?!”

What About the Pharmaceutical Supply Chain

What picture comes to mind when you think of pharmaceutical supply-chains?

· Not a single one?

· Lorries pulling up at the hospital unloading bay?

· Vans delivering into your local pharmacy?

If those are mainly the pictures that come to mind, you should get a lot out of this Substack…

…there is much more to it than that. That’s only the tip of the iceberg. The journey through the various production stages, beginning with raw materials, sees drugs and their components travel tens, if not hundreds of thousands of miles. They go through multiple airports, seaports, countries, and continents. They are acted upon, handed over, acted upon again, handed over again…

…and so it goes.

The typical length, from beginning to receipt of the product in your hand, is around three years. That is the cumulative lead-time. It means that the companies at the beginning (raw material producers) are producing materials for drugs that will be needed in three years’ time.

The quantities they produce depend on projections, estimates, and forecasts. These are handed down from the companies along the chain. The company developing or selling the drugs at the top of the chain must start the ball rolling, based on sales expectations in their business plans.

When a seismic change in demand occurs, as with COVID-19, it is going to severely challenge the best of supply chains. When the supply chains have been neglected by their owners for decades, we get what we got – chaos and confusion everywhere.

This is an extract from Supply Chain Management in the Drug Industry, 2011:

“Left unattended, supply chains lay around doing the human equivalent of lounging on the sofa, drinking pop (soda), eating sweets (candy), and watching TV. They behave like neglected children.

No other sector seems to have neglected its (supply chain) children to the degree that pharmaceuticals have. The parents are now paying the price for all those years of neglect. The big question is: How do they get the children up off the sofa to start to become productive members of society?”

What will you get from Pathway to Prescriptions?

In Pathway to Prescriptions, you will join me in getting under the skin of what has gone wrong in pharmaceutical supply chains and the industry that creates them. That will help you make informed decisions on the drugs you take from a position of knowledge and understanding, rather than ignorance.

The typical timespan, from first discovery to a prescription in your hand, is ten to fifteen years. It means the scientists and companies at the beginning are developing medicines for diseases as they are understood today, hoping the science and engineering, the regulations, and the need will still align by the time of approval.

The evidence that supports that prescription depends on trial design, patient recruitment, statistical assumptions, and judgement calls made years earlier by sponsors, investigators, and regulators. The company developing the drug must start the ball rolling, based on scientific promise and a business case, long before a single patient is ever prescribed the finished product.

When a gap opens up anywhere along that pathway, whether in nonclinical testing, trial oversight, manufacturing controls, or post-market surveillance, it is going to severely challenge the safety of what lands in the patient’s hand. When that pathway has been under-scrutinised for decades, we get what we got: contaminated raw materials reaching prescriptions, and adverse events discovered only after the damage is done.

So, we start next Monday, when I will share the first Pathway to Prescriptions newsletter in this new format - stay tuned!

Regards, Hedley :O)

Leave a comment

Here’s the honest version of why I made that change.

For years I’ve written two different things in two different places: plain-English explanations of how drugs actually get from a lab bench to a prescription pad, and sharper, angrier pieces about where the system governing that process quietly fails — sponsor-investigator blind spots, supply chains optimized for cost instead of resilience, regulatory gaps that only surface after something’s already gone wrong. I wrote the second kind for free, on the side of a full client load, because I couldn’t not write it.

That’s not sustainable, and it was never going to get better by staying free. The pieces that took the most digging — the ones that actually named names, traced root causes, sat with FDA guidance long enough to find what it wasn’t saying — were the ones I had the least protected time for. So I’m changing that.

Starting today, Pathway to Prescriptions becomes the one place I write, and it works like this:

Free, always: the explainers — how drug development actually works, decoded for anyone who wants to understand it without a pharmacology degree. That mission doesn’t change, and it isn’t going behind a paywall.

Paid: the deeper work. The governance failures, the supply-chain and CMC analysis, the case studies on why specific trials or approvals collapsed. The work that used to live on Inside Pharma, given the time it actually deserves.

If you’ve read me for the reform writing, this is where it lives now, and it’s about to get more rigorous, not less. If you’ve read me for the explainers, nothing changes for you today.

Upgrade to paid →

Subscribe now

Thank you for being here for this, whichever door you came through.

Warm regards, Hedley