The Route Is the Drug

The gene is called GULO, and it codes for the last enzyme in the assembly line that turns glucose into ascorbic acid — vitamin C. Nearly every mammal still runs that line. A goat manufactures its own, quietly, all day, and cranks production up when it's sick or stressed. It has never thought about oranges. It does not need to.

You carry the broken copy — a burnt-out pseudogene sitting in your DNA like a shuttered factory on the edge of town. Every molecule of vitamin C in your body arrived through your mouth. Which was fine, evolutionarily, because our ancestors lived in fruit; the gene broke and nobody noticed, and then several million years later we invented long sea voyages and it turned out to be one of the more consequential typos in the genome.

Hold that, because a fifty-year scientific argument turns on it.

In the 1970s, Linus Pauling — a man with two unshared Nobel Prizes, which is a sentence that should probably be illegal — worked with the Scottish physician Ewan Cameron on high-dose vitamin C for patients with advanced cancer. Their early patients got it intravenously, and Cameron reported survival results startling enough that Pauling spent the rest of his life on it. Everything that came after moved to tablets, because tablets are easier.

The Mayo Clinic then ran two randomized, double-blind trials to check the claim. They used tablets. They found nothing — no survival benefit whatsoever. Pauling became a cautionary tale, the brilliant man who wandered off the map, and vitamin C sat in the credulous-people file for four decades.

Here is what nobody adequately weighted at the time: your gut is not a funnel. It is a bouncer.

The transporters that move vitamin C from your intestine into your blood saturate. Swallow more and the extra simply doesn't get in; past a point your plasma concentration flattens and stays flat no matter how many grams you take, and the surplus leaves in your urine at roughly the speed it arrived. Put the same molecule into a vein and you route around the doorman entirely — blood concentrations go tens, even hundreds, of times higher than any pill can reach.

And at those concentrations, vitamin C stops behaving like vitamin C. The thing everyone knows about ascorbic acid is that it's an antioxidant, which is true at the doses a diet supplies. At infusion levels it flips role and turns pro-oxidant, generating hydrogen peroxide in the tissue around it. Cancer cells appear especially poor at handling this — they're already running under heavy oxidative stress, with less headroom — while healthy cells, defenses intact, mostly shrug it off. That's a completely different pharmacology from "eat more citrus."

So the Mayo Clinic did not disprove Pauling's finding. The Mayo Clinic ran a rigorous test of a different intervention that happened to share a name with his. Same molecule, same nominal dose on the label, different physics on arrival. The route was the drug.

Which is not the same as saying Cameron was right, and I want to be exact about the size of that gap. Cameron's study wasn't randomized either. He compared his hundred terminal patients against a thousand case records pulled retrospectively from the same hospital and matched on paper — and his patients tended to be labeled untreatable earlier in the course of their disease than the controls had been, which manufactures a survival difference out of bookkeeping alone. Route explains why the Mayo null doesn't reach an intravenous infusion. It does nothing at all to make Cameron's numbers trustworthy. Absence of refutation is not vindication. It is an open question with the door left unlocked.

And route wasn't the only variable loose in the setup — it was just the one nobody chased. Pauling's own objection to the 1979 trial was that most of its patients had already been through chemotherapy, which he argued had wrecked the immune function his hypothesis leaned on. Mayo took that seriously. The second trial, in 1985, enrolled patients who had never received chemotherapy at all, and found the same nothing. The field was perfectly willing to control a variable. It controlled the wrong one, twice, and then stopped.

Careful here, though — this is exactly the kind of story that gets flattened into a miracle by the third retelling, and I refuse to be link two in that chain. Intravenous vitamin C is not a cancer cure. The human trials are small — ovarian, pancreatic, brain, mostly — and the survival results are genuinely mixed. What shows up consistently is softer: patients getting infusions alongside chemotherapy report less fatigue, less pain, less nausea. A real thing to give someone in the worst year of their life. Not the thing Pauling claimed.

What actually got vindicated wasn't a cure. It was a distinction.

For forty years, two sets of competent scientists produced results that disagreed, and the field settled it by declaring a winner rather than asking the boring question of what differed between the setups. The answer was sitting in first-year physiology the whole time. Everyone involved knew intestinal absorption saturates. Nobody set it beside the trial design, because by then the argument had stopped being about chemistry and started being about whether you were the sort of person who took Linus Pauling seriously.

That's the failure mode worth naming, and it isn't unique to vitamin C. When two honest measurements contradict each other, the contradiction is information — some variable you thought was held constant wasn't. Picking a side throws that away and calls the discarding consensus.

But tribalism is the easy answer, and it stops one floor above the machinery. Ascorbic acid cannot be patented. Nobody can own the answer to this question. A trial large enough to settle intravenous vitamin C in even one cancer would run to tens of millions of dollars and return its funder nothing exclusive — no molecule, no license, no protected years on the far end. Which means the study has no natural sponsor, and never has. The route hypothesis was cheap to think of and expensive to test, and there is very little in the architecture of clinical research that pays for expensive tests of unownable claims.

So the contradiction didn't sit unread for four decades because scientists were stubborn. Or not only. It sat unread because nobody was structurally positioned to read it — the reading was nobody's job, and nobody's budget. The signal doesn't vanish when it goes unfunded; it just waits. And there is no reason on Earth to think this is the only one waiting. Somewhere in the same lot sit a hundred other contradictions with no owner, perfectly legible, parked for exactly as long as looking stays free and confirming stays expensive.

Fifty years of argument about a molecule your body used to make for free. The goat, still making its own, has no opinion on any of this. The broken gene is still in you, doing nothing, patient as a fossil.

And the drip, when it works, isn't really supplementation at all. It's a prosthesis for something we lost so long ago we forgot it was ever ours.