In a disturbing post at ScienceInsider, Jon Cohen and Martin Enserink explain why the swine flu vaccine is running so late. Or at least they try to explain why it’s so late. For while all the suppliers are running into problems, we’re not allowed to know what they are.
The delays are substantial and critical. They leave us naked as the flu spreads through the country. The flu has now killed 1000 people, over 100 of them children. Even as this happens, the delivery dates keep moving back and the delivery amounts keep shrinking. As recently as a month ago, the CDC was telling us that we’d have 40 million doses by the end of this month. Last week, they were saying 30 million doses. Now they’re saying 16 million doses and not making any promises beyond that.
The U.S. contracted with five companies to supply vaccine. Only one, MedImmune — the company supplying the adjuvanted FluMist product — has come close to its promised delivery schedule. But because we’re so leery of adjuvanted product, we ordered only 12.8 million doses from MedImmune. The other four companies, meanwhile, are all running badly behind. Why? When ScienceInsider pressed Nicole Lurie, the assistant secretary for preparedness and response at HHS, for answers, she essentially she couldn’t say.
For proprietary reasons, Lurie would not name the delay issues at each manufacturer, but said “one company was just really overoptimistic” in its delivery predictions to HHS. Initially, all of the makers of inactivated vaccine had trouble with the “potency assay” used to ensure that 15 micrograms of influenza antigens are present in each dose. Once that was corrected, said Lurie, this company lost 20% to 35% of its antigen during the “recalibration.” Unfortunately, said Lurie, the company did not notify HHS of the problem. “They thought they were going to compensate by getting their yield up but they never let us know,” she said. “When we learned of problem, it was over and there was nothing we could do, which is too bad because we could have helped.”
After Columbus Day on 12 October, HHS learned about delays at two of the other companies. One had a new production line to fill doses of vaccine into individual syringes. “Getting up the new production lines with prefilled syringes was really slow,” Lurie said. The third company had “major production problems” that Lurie said she could not describe in detail for proprietary reasons. Basically they had difficulty making “acceptable” vaccine that did not contain an immune booster called an adjuvant, which is used in several countries but not the United States.
The final company wanted to supply its own country with vaccine before shipping product to the United States, which Lurie said would begin next week. She did not identify the company, but Australia’s CSL earlier said it would supply its country before exporting the product.
Given the drug industry’s recent problems with what one might generously call overpromising (and with what Lurie here even more generously calls “overoptimism”), this is troubling. We contracted with four companies, and each of the four was apparently either overoptimistic or misleading in its delivery predictions. The CDC apparently found them convincing — and was apparently left blind to the delays until the companies revealed them at a very late hour. The CDC now tells us simultaneously that no one could have foreseen the problems, for they came (at four different companies) from a perfect storm, or four perfect storms — and that these perfect storms are all to be expected, because this is vaccines, and vaccines are slow and glitchy.
If they have so much experience making the product [asked ScienceInsider], why so many snafus with the pandemic vaccine? “Almost every flu season, something goes wrong,” said Lurie, emphasizing that she was referring to manufacturing, not safety.
I don’t think the CDC is conniving here. But it appears it has let itself be misled by an industry that has a remarkably well-known record of misleading. It remains unexplained how this industry enjoys so strong a presumption of goodwill and integrity that even in a supply issue so directly vital to public health, we’re not allowed to know more about what slowing things down. I like industrial secrets as much as the next person. Doubtless these companies must keep certain hard-won information under wraps. But it would seem that when tens of millions of doses of vaccine are weeks late, leaving a country exposed to a fairly dangerous virus, we might get something more specific than that one company was overoptimistic and another had trouble filling syringes.
In a way, this is to be expected, for most people in virology and public health recognize that our conventional mode of making vaccines is outmoded and is no match for pandemic viruses. A six-month production schedule — six months when everything goes well — is no match for a virus that can go global in weeks.
So one big question is whether this shortfall — which would be catastrophic if this flu did its ugliest job on even, say, 5 percent more of the infected — will inspire a serious move toward other strategies. Nonadjuvanted vaccines, for instance, can get to more people sort of quickly, and other vaccines in development, such as those built by via cell culture or modeling or with virus-like particles, can be made much more quickly. But how vigorously will we pursue these? There’s a risk that simply because this flu is so far not killing vast numbers of people, the existing complacency about the threat of pandemics will couple with the antipathy or apathy about vaccines to leave the current inadequate system in place.. “Well, that wasn’t so bad, even though the vaccines were late. The strategy worked.” Combine that with the veil of proprietary secrecy over the production delays, and you get a formula not for better vaccines, but for simply trying the same strategy next time.
In short, if we care so little about the delays, and about vaccines in general, to accept that these delays were simply a bad-luck “perfect storm” (even while being to-be-expected), then we’re less likely to put in place something much better for next time. The conventional egg-based vaccine approach isn’t working for flu. It’s too slow, too glitchy — and apparently not terribly accountable. We’ve spent a lot of money and invested a lot of hope in vaccines that are getting here too late. If this holds up and is not seen as a major failure needing correction, we’re missing what at this point appears to be a learning opportunity that is vital but relatively inexpensive — if, that is, you can call thousands of deaths inexpensive.