The complexities of manufacturing and open source technology

Contract out facilities for mRNA technology and open source everything else

First, a quick note about Clover. Today, Hindenburg Research (which probably has a short position in Clover) published a cutting tweet thread about how Clover and/or its SPAC partners failed to disclose several major problems with the business. Namely, Clover is the subject of a previously undisclosed DOJ investigation, doctors hate its software and only use it because Clover pays them, and other sundry things that scream Not Good™️. I covered my qualms with Clover back in October but I… did not see a DOJ investigation coming!

So now: vaccines. I thought this week I’d continue talking about pharma more broadly, but vaccines keep dragging me back in. It might be because I’m perpetually stuck in my house waiting for a vaccine that never seems any closer.

Anyway, here we go again.

There’s an ongoing debate about how to ensure greater production of the COVID-19 vaccine.

I think there’s a missing point here about the complexities of the manufacturing process, which is important to nail down now that we have two FDA-approved candidates and likely a few more on the way. (I should also note that this is entirely my opinion, not necessarily that of my institution!)

The two vaccines currently approved, Moderna and Pfizer, both use mRNA. This is a brand-new, first-of-its-kind application of this kind of technology; before this, Moderna had been trying for years to develop a successful candidate using mRNA (Pfizer, which closed its infectious disease unit years ago, partnered with a smaller biotech firm that had developed similar tech as Moderna).

It’s exciting! It opens a lot of possibilities! Moderna has at least 11 additional vaccine candidates for diseases other than COVID-19 in their pipeline, and several other mRNA drugs lined up! When I first read about the marvel of Moderna having this breakthrough, I’m being completely serious, I cried.

But now here we are almost two months since those first trucks rolled away from the Pfizer plant in Kalamazoo, and while there have been some wins (Feb. 1 marked the first day that more Americans had received at least one dose of the vaccine than had been diagnosed with COVID-19 since it first hit U.S. shores), those of us stuck at home feel like the rollout will never, ever end.

So why not make the patents open source, spread around the IP, and possibly increase manufacturing speed? As I’ll argue here: I don’t think that’s the right move for Pfizer and Moderna’s mRNA vaccines—I think the U.S. could have more impact by direct-contracting manufacturing facilities and maybe microfluidics devices. Then, when vaccines using older technology are FDA-approved, we should strongly consider making that IP open source—the vaccine market prior to 2020 was far from being innovative or speedy, and it could use a shake-up.

mRNA technology

First, a little more about mRNA technology. mRNA is a nucleic acid produced by most human cells, an mechanism to transfer instructions for making proteins. In other words, it’s absolutely crucial. It’s also a great point of leverage if you want to cause human cells to make proteins that aren’t already encoded in DNA—for example: antibody proteins targeting COVID-19 spikes.

Sorry, this is purely because I love how insane stock photos are

Scientists know this, and they’ve been trying to adapt mRNA to produce desirable proteins for at least 30 years. But the human body is too good at recognizing foreign mRNA and destroying it; mRNA applications never got off the ground.

Then, after a series of key lab discoveries by scientists at the University of Pennsylvania, a few scientists in Boston realized they could monetize mRNA applications. Around 2010, they formed Moderna (ticker symbol: MRNA) with the goal of leveraging mRNA technology.

They didn’t do it alone. Not only was Moderna venture-capital funded, but the company also took more than a hundred million dollars from DARPA and BARDA, two government agencies that fund research projects for the good of national security.

Despite this funding, Moderna had a rocky start. Its early drug candidates were either too weak or too immunoreactive. In response, the company pivoted its technology to vaccine production (probably at least partially motivated by the available millions in DARPA and BARDA funding for vaccine research).

Its first success is with the COVID-19 vaccine.

mRNA technology, in other words, is still extremely new. It’s also fairly fragile. The Pfizer vaccine has to be kept at subarctic temperatures, and the Moderna one must be frozen. (Interestingly, it’s possible the Pfizer vaccine is actually stable at normal freezer temperatures—but the corporation decided that, in the interest of moving as fast as possible, it would just go with ultra-cold storage.) The precise mixing of mRNA within a medium of lipid nanoparticles probably requires bespoke microfluidics devices that take months to acquire at scale.

So there aren’t many facilities capable of producing mRNA vaccines, although that’s subject to change in the future. A better solution for mRNA COVID-19 vaccines is direct contracting of facilities.

(I talk more about the government incentivization of the Moderna and Pfizer successes in my American Prospect piece from January, if you want to see more on this.)


Our friends at PrEPforAll, along with a professor from Columbia, recently proposed direct contracting in the pages of the New York Times. They compare their proposal to the President’s Emergency Plan for AIDS Relief, an early-2000s initiative launched by the GW Bush government that provides prevention, treatment, and care for HIV/AIDS. PEPFAR is, in the words of the op-ed, “one of the most successful global health initiatives ever implemented.”

Just as PEPFAR was a large-scale, global initiative with an eye to preventing viral disease, the writers of the NYT op-ed propose a PEPVAR, or President’s Emergency Plan for Vaccine Access and Relief. This plan would be more manufacturing-heavy than PEPFAR; under it, President Biden would authorize government contracting of available facilities capable of producing an mRNA vaccine at scale and act as a liaison between facility and vaccine manufacturer (in this case, Pfizer and Moderna).

This method is particularly useful because of the previously mentioned technological constraints of the mRNA vaccine; there simply aren’t many vaccine facilities capable of producing mRNA vaccines right now. Pfizer is relying on a network of its own facilities in Kalamazoo and Puurs, Belgium; Moderna appears to be doing its microfluidics work in-house and relying on Lonza, a large corporation that prepared for months, to handle other steps of the process.

These companies are also already making agreements with other facilities to build up mRNA technology capabilities. BioNTech (the relatively small biotech firm that developed the technology used in the Pfizer vaccine) recently announced a contract with Sanofi, a huge multinational drug corporation, to produce more of BioNTech/Pfizer’s vaccine after Sanofi’s candidate failed. Some Wall Street analysts suspect that more partnerships are possible.

But these agreements are months in the future. There’s still room for the government to make a list of availabile facilities and microfluidics devices, along with other rate-limiting equipment or supplies, contract for it directly, and then demand that Pfizer and Moderna use those additional facilities as much as is possible.

And before I sound too sanguine about big drug manufacturers and their ability to pull together in a time of crisis: the level of cooperation we’re seeing required a restructuring of typical vaccine market incentives, a restructuring that, in the U.S., was achieved by Operation Warp Speed. Usually, would-be vaccine manufacturers wait until they’re sure an epidemic is becoming a pandemic. Their trials move slowly, they get caught in bottlenecks. Manufacturers that aren’t the first mover may hang back, waiting to see if the first mover is successful. But because OWS funded a portfolio of diverse vaccine technologies, provided billions in research funding, and guaranteed purchases ahead of time, all of these corporations are incentivized to produce as quickly as possible and, therefore, as collaboratively as possible.

So there’s still room for the government to contract out some facilities.

The power of open source

All of this being said, there is power in open source science and possibly a role for government-mandated open source IP for some of the older vaccine technologies, like the candidate produced by AstraZeneca.

For example, insulin (again, more forthcoming from me on this) hasn’t been the focus of real innovation for decades. Almost the entire U.S. supply is controlled by just three manufacturers that are probably colluding with one another, so there’s no incentive for new insulin manufacturers to step up, or for the existing manufacturers to innovate.

In response, grassroots, open source insulin groups have cropped up across the country. Literally, it is people making their own insulin. The insulin space could obviously benefit from patents and technologies being made open source so that other companies can freely make it, and insulin users won’t have to rely on their own homemade insulin.


Similarly, the older technology used in some COVID-19 candidates doesn’t really represent scientific breakthrough, and there probably isn’t too much room for further innovation with the tech. Above all else, people need these vaccines at scale—and there are far more facilities capable of producing these more familiar types of vaccines. Government funding has already supported a large portion of the vaccine research (for both mRNA vaccines and old technology vaccines, to be sure); governments should leverage that to either break open production everywhere that’s possible, or to outright make the patents part of the public domain.

And once the mRNA space calcifies, the government should closely police companies who try to patent thicket or otherwise monopolize the market; if and when that happens, there’s a strong case for making that tech more open source and begetting another era of scientific advance.


The first part of the COVID-19 vaccine—the innovation part—has been stunningly successful. The government flexed its power to restructure and incentivize a shoddy, calcified market and now we have more than one amazing, successful, cutting edge vaccine.

The distribution process has been much rockier, from the manufacturing bottlenecks to the struggles of hospitals to roll out the vaccine to individuals. Operation Warp Speed dropped the ball.

The manufacturing bottlenecks we’re seeing could handle some restructuring and assistance from competent government logistics experts. By using a targeted approach—direct contracting with facilities and pursuing a PEPVAR for advanced tech, considering open IP for older, less innovative tech—the administration stands a better chance of seeing production at the necessary scale.

Then, when it’s over, I think we need to apply this strategy to all sorts of other drug markets. The pharmaceutical industry showed how fast, accurately, and innovatively it could move. Think of the possibilities if we always held them to this standard when taxpayer dollars are involved!