Approved On Wednesday. The Same Molecule Was Refused A Trial In 2009.
The pancreatic agent licensed this week is a patented salt of a compound that has been in the catalogues since 1998. What changed was not the chemistry. It was who could hold the patent.
A months-long reporting project. Documents cited below are held in The Vault and available to readers.
SILVER SPRING, United States — On Wednesday the Food and Drug Administration licensed cetralate meglumine for metastatic pancreatic adenocarcinoma, on a phase III of 811 patients across 74 sites that reported a median overall survival gain of 4.1 months.
The active part of that molecule is cetralic acid. It has been in the commercial chemical catalogues since 1998. Twenty-five grams costs $3.10 a gram and a laboratory can order it this afternoon.
It first shrank a human pancreatic tumour in a study reported in 2009.
What the 2009 study was, in its author's own words
Sixty-one patients, one hospital, no control arm, and a comparator group assembled from the same hospital's own records. Median overall survival was 11.4 months against a matched historical 6.2.
Professor Adaeze Marchetti-Sole, who ran it, made us write that paragraph before she would answer anything else.
"It was small and it was uncontrolled and anybody who tells your readers otherwise is selling something," she said. "It was also the kind of result that is supposed to start something. It started three telephone calls."
The three refusals we cannot show you
Between 2009 and 2011 three sponsors were approached. All three declined. We have not seen any of the three letters.
Two people describe them and their accounts agree. One of them, Ines Barraclough-Otu, was a development director at a company that received one, read it, and does not have it.
What we can put in front of a reader is a letter of the same kind, written to this newspaper in June about an entirely different compound, filed and published as PX-2065. It is four sentences long.
We are citing that as an example of the form. It is not evidence about cetralic acid and we are not going to pretend it is.
What changed in 2016
Not the biology.
In 2016 a small development company patented the meglumine salt of the acid. The salt raised oral bioavailability from four per cent to sixty-one. The exclusivity runs to 2036. The phase III opened in 2022 and cost $58m — the same figure that was described to Marchetti-Sole in 2010 as unrecoverable.
Six years from patent to trial. Seventeen from her result to Wednesday.
Why we print three numbers instead of one
We asked what the interval cost, and we can give three answers depending on an assumption nobody has ever tested.
If the free acid at an achievable dose delivers what the salt delivers, the registry arithmetic puts the interval at somewhere between 41,000 and 96,000 lives in the four countries we can count.
If it delivers half, the range roughly halves.
If the acid is effectively inert taken orally and only the salt was ever going to work — the sponsor's position, and the one our own commissioned pharmacologist thinks is most likely — then almost nothing was lost to the refusals, and what was lost is the six years between a patent and a protocol.
We do not know which is right. We are publishing the extract and the code so that a reader can get a different answer and tell us where we went wrong.
What this desk takes from it
Six years is the smallest possible version of this story, and six years is the number produced by the assumption most favourable to everybody in it.
Nobody suppressed cetralic acid. There was no meeting, no memorandum, no hand on a shoulder. There was a molecule, a price, and a rule that a company may only spend money it has a mechanism to recover — a rule that worked precisely as designed, for seventeen years, on a disease that kills nine people in ten within five.
That is the thing this newspaper wants held next to Wednesday's good news, and it is good news. The machine started the moment somebody found a coin that fitted.
It had been standing there since 1998. </content> </invoke>
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We took Wednesday's approval notice, followed the pivotal trial's citation chain backwards until the citations stopped, and then went looking for the paperwork in the gap between the last human study and the first patent. The registry arithmetic is our own; the extract, the code and the three model specifications are filed in The Vault with this story so that a reader can get a different answer and show us where.
Who we spoke to
- Prof. Adaeze Marchetti-Sole, Medical oncologist; ran the 2007–09 study. Interviewed by video call three times; supplied her original protocol, her ethics submission and her funding correspondence July–August 2026 Asked us to print that her study was small and uncontrolled before we printed anything else about it.
- Dr. Hedda Kalfayan-Brooke, Clinical pharmacologist; has never worked on this compound. Commissioned to model achievable exposure from the free acid using only published data; her working is published with this story August 2026 Her conclusion is the most damaging thing in this piece to our own framing, and it is printed at length.
- Ines Barraclough-Otu, Development director at the sponsor, 2006–2013. Interviewed in person twice; shown this article in draft and given four days on it 12 and 19 August 2026 Read one of the three 2009 refusals. Will not supply it and says she does not have it.
- A national cancer registry, Incidence, stage at diagnosis and survival, 2009–2025. Standard data request, £0, granted in eleven days; our arithmetic and code filed in The Vault August 2026
- A pharmaceutical sponsor, Reply to this newspaper's funding enquiry on a different compound. Received in writing, filed and published as PX-2065 June 2026 Four sentences. Cited here as an example of the form, not as evidence about cetralic acid.
Documents
- PX-2065 — Reply to this newspaper's funding enquiry, from a pharmaceutical sponsor, declining to fund a phase III accepted
What we could not confirm
- The three 2009 refusals themselves. Two people describe them consistently and one of them read one; we have never held a copy and no company will give us one. What we publish is a letter of the same kind, about a different compound, written to us this June.
- Whether the free acid, given at a dose a patient could tolerate, would have delivered what the salt delivered. It has never been tested against the salt and now never will be. This is the sponsor's strongest objection to this story and we cannot answer it.
- What the interval cost in lives. We publish three estimates and they do not agree with one another. We are not going to choose between them in a headline for the sake of having one number.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Continental WireRegulator Approves New Pancreatic Cancer Drug
Ran the agency's notice and the trial's primary endpoint, accurately, in 340 words.
Had that we did not
The approval, the survival figure, the patient numbers, the sites.
Left out
That the active part of the molecule is twenty-eight years old. The wire item has no field for that and nobody asked for one.
-
Signal & LedgerCetralate Launch Seen Adding $2.1bn By 2029
Priced the approval for subscribers within ninety minutes of the notice.
Had that we did not
The salt, the patent term, the exclusivity, the launch curve — all of it, correctly, and first.
Left out
Nothing, in its own terms. The trade press understood this story better than the news pages did. It reported it as good news, which for its readers it is.
-
They Buriedthis newspaperApproved On Wednesday. The Same Molecule Was Refused A Trial In 2009.
Worked backwards along the pivotal trial's citation chain until the citations stopped, then looked for the paperwork in the gap.
Had that we did not
The 2009 protocol entire, the registry extract, our code, and all three of the estimates we could not choose between.
Left out
We published a story about this molecule's cousin on 9 August and stopped at the sentence 'the reason is not a conspiracy, it is arithmetic'. It was arithmetic. It was also seventeen years, and the arithmetic was ours to do that day. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Dr. Cassian Wrede-Machado, chief development officer of the phase III sponsor Sent the full draft, the registry extract, our code and all three estimates; given nine days. Answered in writing and asked that it run whole. Printed unedited.
I am going to answer this properly, because the version of it that will circulate by Friday is not the version you have written, and I would rather argue with yours.
First, the correction. In 2009 there was no drug. There was a molecule with four per cent oral bioavailability, which means that ninety-six per cent of what a patient swallowed went to the sewer, and the four per cent that did not was so variable between patients that no regulator on earth would have accepted the exposure data. Professor Marchetti-Sole's patients did well and I have read her paper twice and I am glad she ran it. Sixty-one people, no control arm, and a historical comparator she chose herself. That is a hypothesis. My industry is asked to spend fifty-eight million dollars on hypotheses roughly two hundred times a year and gets it right about one time in nine.
Second, the salt. Your headline calls it 'the same molecule'. It is not the same molecule and the difference is the entire drug. Somebody sat with that compound for nineteen months and found a counter-ion that took it from four per cent to sixty-one. That is not a legal trick. It is the reason a patient can take it at home instead of not taking it at all. Your own pharmacologist says so in your own sources list and I notice you have printed her, which is more than I expected.
Third, and this is where I concede something, so please do not cut it.
Nobody in my industry was ever going to do that nineteen months of formulation work on a compound they could not hold. Not out of malice. Because a development committee that approves it gets replaced by one that does not. The salt work happened in 2016 because a small company worked out that the salt was ownable, and the moment it was ownable everything else followed in ten years, which for this disease is fast.
So you are right that the obstacle was ownership. You are wrong that the obstacle was ownership of the acid. It was the absence of anything ownable at all, and the day something became ownable the system moved. That is not a conspiracy and I do not think you have called it one. It is a machine that only starts when you put a coin in, and for seventeen years nobody had a coin that fitted.
I would rather you printed that sentence than the number you cannot choose between.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.