A biotech flu vaccine and the seduction of hype

(UPDATED: See below.)

protein-sciences-logo.gifVaccine developer Protein Sciences is all over the blogosphere today (see, for instance, items on Pharmalot and the WSJ health blog), thanks to a breathless Bloomberg piece on the little Meriden, Conn., biotech. According to Bloomberg reporter John Lauerman’s article, Protein Sciences is poised to vanquish establish flu vaccines from Big Pharma giants Novartis and GlaxoSmithKline, in an interesting case study of what happens when a quote-worthy biotech executive catches the ear of a reporter eager for a new story line.

From Bloomberg:

GlaxoSmithKline Plc and Novartis AG, two of the world’s biggest vaccine makers, may have bet on the wrong technology in the race to develop a better flu shot.

The drugmakers are building U.S. factories to grow influenza virus in animal cells as an advance over the decades- old technique of making flu shots using chicken eggs. Now a small, privately held biotechnology company may leapfrog ahead of them with a more advanced method using DNA.

That company is Protein Sciences, which plans to use genetic engineering — technically called “recombinant DNA” — to produce influenza surface proteins from genes inserted into insect cells. The company’s chief operating officer, Manon Cox, scoffed at the traditional vaccine makers, telling Bloomberg that even their newer animal-cell technology “is an amazingly stupid investment to make for the future…. It’s as if we’re still living 100 years ago and recombinant DNA was never developed.”

The background: Flu vaccines are traditionally time-consuming and laborious to produce, since the most widely-used vaccine consists mostly of proteins harvested from inactivated flu virus, requiring mass quantities of virus. Until recently, the most reliable way to culture the virus was to grow it in vast collections of sterile chicken eggs, a 60-year-old process that takes about six months from start to finish. In the past few years, big vaccine makers like GSK and Novartis have been moving toward growing the virus in mammalian cell cultures, a much faster process than the old chicken-egg technique, although it’s not yet in widespread use.

Generally speaking, injecting people with a mixture derived from the inactivated virus immunizes them against infection by the real, live virus. A more recent intranasal vaccine developed by MedImmune — soon to be part of AstraZeneca — uses attenuated live flu virus to create the same effect.

Enter Protein Sciences and Novavax, of Malvern, Pa., which produce immunizing proteins directly, without the intermediate step of growing cultures of live virus. Using recombinant DNA, Protein Sciences inserted the genes for influenza surface proteins called hemagglutinins into insect cells, which then obligingly pump out zillions of hemagglutinin proteins when cultured. Ideally, injecting people with these proteins will also protect them against actual infection.

Recombinant production offers a number of advantages even over cell-cultured virus production. It’s fast and efficient, for one thing, and tinkering with the inserted genes can yield new vaccines relatively quickly. Both Protein Sciences and Novavax aim to produce vaccines against a possible avian flu pandemic, should that virus ever acquire the ability to pass between people.

On the other hand, this sort of vaccine — technically known as a “subunit vaccine” because it consists only of viral pieces — has some drawbacks as well. Although researchers have been working on subunit vaccines for well over a decade, only two have so far proven themselves, for typhoid and hepatitis B. In general, subunit vaccines tend to produce weaker immune responses than those derived from whole virus, which may reflect the possibility that recombinant proteins don’t always retain the same shape as their natural counterparts.

In the case of Protein Sciences, its recombinant flu vaccine consists solely of hemagglutinin proteins, whereas traditional vaccines also incorporate a second flu protein called neuraminidase — not to mention whatever other viral remnants survive the chemical centrifuging process that harvests the surface proteins in the first place. (See, for instance, the production process described in this PDF link to the package insert for Novartis’ Fluvirin.)

All of which is to say that it’s a bit soon to declare recombinant subunit vaccines victorious over the traditional version. A working recombinant flu vaccine would unquestionably be a great step forward for public health, but the evidence really isn’t there yet. Protein Sciences, for instance, conducted a large study of its vaccine — which goes by the contrived name FluBl0k (yes, that’s a zero in place of the letter “O”) — in 460 volunteers during the 2004-2005 flu season and says a high dose of FluBl0k “showed 100% protective efficacy against laboratory confirmed influenza,” but hasn’t yet published those results. An earlier study showed that FluBl0k was more effective in raising certain antibody levels than traditional flu vaccine, but didn’t test its efficacy against actual infection. (Read that study in PDF form here.)

UPDATED: Turns out there’s another reason to at least be cautious about Protein Sciences’ technology and its claims. In the early 1990s, the company — then known by the name MicroGeneSys — was developing an HIV subunit vaccine called VaxSyn that it wanted to test in human trials. When the NIH rejected its application, MicroGeneSys hired a lobbyist and got Congress to earmark $20 million for the trials in the Pentagon budget — a hugely controversial move at the time. In 1996, the resulting Army-run trial of VaxSyn showed “no clinical improvement that could be attributed to the vaccine.”

To be fair, it appears that the company’s management has turned over completely since those days. On the other hand, this saying is always worth bearing in mind.