A note on the decision nobody had to defend

Britain discovered monoclonal antibodies and never patented the method. The usual telling blames the scientists. The record does not support that, and what it does support is less comfortable.

Köhler and Milstein's paper appeared in Nature on 7 August 1975. Under British practice then, publishing before filing closed the door. The only window that mattered was the weeks before that date.

The referral was made inside it. In July 1975 Milstein presented the work at an internal Medical Research Council meeting. An MRC official, Tony Vickers, asked him for the proofs and alerted the National Research Development Corporation, the body that handled patenting for MRC inventions. Vickers moved before the article went to print, because he thought the work had commercial implications.

The NRDC replied in October 1976, after several months and some prompting from the MRC. It acknowledged the possible value and said it could not "identify any immediate practical applications which could be pursued as a commercial venture." Nothing was ever filed. The window had been shut for over a year.

Who was exposed

The NRDC held a monopoly on the decision. Until 1985 it had right of first refusal over publicly funded inventions in Britain. Declining cost it nothing, and no reporting system records a decision not to file.

The scientists had no exposure either. Researchers in British government-funded biomedical laboratories were not entitled to royalties on patents from their work at the time. MRC royalty sharing arrived in the 1980s.

So the decision sat with a body that could not lose, and the people closest to the work had no claim on the outcome.

What happened to the cells

In September 1976, a month before that reply, Milstein sent his X63 myeloma line to Hilary Koprowski at the Wistar Institute on the ordinary terms of the day: acknowledge the source, ask before passing the cells on, do not patent products made from them. Nothing was signed, so none of it was enforceable.

Wistar applied in June 1977. Two US patents followed, on tumor antibodies in October 1979 and on viral antibodies in April 1980. I read both. Neither claims the hybridoma method itself, which stayed unpatented.

The verdict

A joint working party reported to government in 1980. It found "a lack of appreciation of the NRDC, particularly by young scientists" and concluded that protection had not been sought early enough, reducing British advantage.

The blame went to the bench. The referral had been made by an administrator, before publication, and the delay belonged to the office.

What I take from it

Give one body a monopoly on a decision and no exposure to the outcome, and it will reliably decide to do nothing. Nothing is always defensible. It generates no file, no cost line, no entry in any annual report.

This is the fourth time these notes have arrived in the same place. We count disclosures and licenses because they are events with dates. A decision to decline is not an event, so nobody answers for it.

The question for anyone designing a transfer function is not who is authorized to file. It is who loses if the filing decision is wrong. If the answer is nobody, the function will produce the NRDC's answer, and produce it in good order.

What I could not establish

Whether the first refusal rested on statute or on administrative arrangement. Two ministers put the end of the monopoly at 1985 and neither named the instrument behind it. I found no such right in the 1967 Act's provisions on the corporation, which is not the same as establishing there was none.

The interval between referral and reply. One account has it going across at once, another has thirteen months elapsing. The October 1976 date is firm and the window had closed well before it, so the dispute does not change the outcome.

The printed 1980 report, which I quote as the literature carries it rather than from the report itself.

One request

In your office, who can decline a filing, and what happens to them if the decision proves wrong? If the answer is nothing, I would like to hear from anyone who has changed it.

---

Sources

• G. Köhler and C. Milstein, Nature 256 (1975), 495-497, issue of 7 August 1975.
• Lara Marks, The Patent saga, whatisbiotechnology.org, and "Overlooked scientific innovation reaps rewards 40 years on", History & Policy, for the meeting, the referral, the 1976 reply and its wording, the X63 supply and the 1980 report. Checked 22 October 2026.
• US 4,172,124 (tumor antibodies, 23 October 1979) and US 4,196,265 (viral antibodies, 1 April 1980), Wistar Institute. Claim 1 of each is limited to a class of antibody. Read 22 October 2026.
• On 1985: John Redwood, HC Deb 12 February 1991; Lord Reay, HL Deb 3 June 1991.
• E. M. Tansey et al., eds., Technology Transfer in Britain: The Case of Monoclonal Antibodies, Wellcome Witnesses to Twentieth Century Medicine, vol. 1 (1997). Not read directly.