Coronavirus Vaccine: Lean Development Principles in Action

A Covid-19 vaccine has sped to public distribution in record time, four times faster than the fastest vaccine development to-date.  How did it happen?

Not without the efforts of thousands of researchers, engineers, technologists, factory workers, delivery drivers, and many others to be sure.  But the strategies deployed in the US and in Europe have been equally pivotal to the dramatically shortened timeline.  Those strategies will not come as a surprise to those familiar with lean product and process development (LPPD).   In this post, I discuss the striking similarities between the strategies used in the Covid-19 vaccine race against time and LPPD principles, and what it potentially means for companies and public-private ventures going forward. Most of this post is based on the US effort, but interestingly the core strategies have also been adopted by the European Medicines Agency in their oversight of Covid-19 vaccines development. 

How fast is fast

Before getting into the strategies, it will be helpful to have some background on vaccine development generally, and how Covid-19 vaccine development compares.  According to the US Department of Defense website on Operation Warp Speed, the typical timeline to get a pharmaceutical distributed to millions of people in the US is about 5 years.  This timeline consists of:

• R&D and pre-clinical trials
• Phase I, Phase II, and Phase III Clinical Trials
• Manufacturing
• Distribution

Each of the clinical trials requires FDA approval based on evidence provided by the manufacturer on the safety and efficacy of the drug before the trial can begin. FDA approval is also required to sell the drug. Because of the uncertainty of FDA approval, companies typically wait to begin large-scale manufacturing planning and distribution until approval is received. This results in each clinical trial taking many months, even years; and taking a year or more for distribution to reach the population after FDA authorization.

Vaccines for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) responsible for Covid-19 were developed and received emergency use authorization from the FDA in less than a year, with complete distribution planned within 14 months of development start.  That would represent an 80% reduction in development lead-time! 

Just as remarkable is that the effort involves multiple government agencies and many private firms.  It’s one thing to achieve a dramatic improvement in lead-time within one organization, but quite another to coordinate it across multiple public and private entities!

So how did it happen?  Let’s take a look….

Build Upon Existing Knowledge

Clearly a major reason for the rapid time to development was the ability of teams to leverage existing knowledge.  Companies have researched and produced vaccines for many years.  One of the newest vaccine technologies using messenger RNA to elicit a person’s immune response, was in development since SARS-1 and MERS appeared.  All the front-running companies were able to use knowledge generated in their prior research to rapidly design and test candidate vaccines.  In fact, Moderna had a vaccine design within two days of the release of the SARS-CoV-2 virus genome!  (See David Wallace-Wells’ article, “We Had the Vaccine the Whole Time”)

Equipment manufacturers equally leveraged existing vaccine processes and equipment platforms to develop the processes and equipment needed for large scale production.  Existing pharmaceutical supply chains could be used to get consumables where they are needed at volume, and existing distribution channels could be used to get vaccines to the places where they can be administered.

So, one key to the rapid development of this vaccine was that companies were not starting from scratch.  In fact, a great deal of knowledge based on prior research, production, and supply chain experience created a tremendous springboard for rapid development of a novel vaccine.

A Set-based Approach

In an article that appeared in Science in May 2020, four leading authorities (Drs. Corey, Mascola, Fauci and Collins) outlined a strategy for Covid-19 vaccine R&D.  One of the key elements of that strategy is to encourage swift vaccine development through a “multi-pronged endeavor” in which substantial investment is made in multiple vaccine platforms.  Operation Warp Speed in the US took that advice to heart by investigating 14 vaccine development programs in depth, then narrowing that field to seven programs to target for funding, spread across four vaccine platforms as data emerged showing promise.  Criteria for selection included: 1) robust pre- or early-stage clinical trial data, 2) ability to go to large-scale trial quickly, 3) potential for manufacturing scalability, yields and consistency, and 4) likeliness of being safe and effective.  Substantial funding for six companies so far has been approved over the past year:

• In March, $456M approved for J&J vaccine
• In April, $483M approved for Moderna vaccine
• In May, $1.2B approved for AstrZeneca vaccine (partnered with Oxford)
• In July, $1.6B approved for Novavax vaccine; $1.9B approved for Pfizer vaccine (mfg & distn); $2B approved for Sanofie & GlaxoSmithKine vaccine; and an additional $472M approved for late-stage clinical development of Moderna vaccine
• August: $1B approved for J&J vaccine mfg & delivery; and $1.5B for Moderna mfg & delivery

As of the writing of this post, two of those programs have received FDA emergency use authorization, so the bets are paying off.

It is interesting that when speed is important, the infectious disease experts essentially said that a set-based approach is wiser than trying to pick the winning technology or design early.  At that time, the several technologies showed promise, but had uncertainties and trade-offs associated with efficacy, safety, durability of immunity, ability to scale and cost of manufacturing, vaccine stability and cold chain requirements. So rather than betting all of one’s money on one company with one technology, the government opted to spread its bets across several technologies and companies.

Parallel Development

Funding alone would not be enough.  The powers-at-be realized that the FDA approval cycle also needed to be streamlined.  Thus, a third key to accelerating development was to move sharply away from a highly sequential process, where one phase needs to be completed before the next phase can began, to one where phases overlap.  This apparently happened in a couple of ways.  First, the FDA allowed clinical trial planning to proceed while the previous phase was still in progress as long as certain benchmarks were met.  Clinical trial execution had to wait, but the planning could move forward such that if FDA approval was granted, the trial could begin almost immediately.  Second, as seen in the timing of funding to different companies, development of manufacturing and supply chain capacity and planning of distribution began while clinical trials were in progress. In fact, the US government provided funding to produce vaccines “at risk”, meaning that if the vaccine is not ultimately approved, the manufactured product would need to be destroyed and the US government is out the money.  Thus, manufacturers not only stood up production facilities, they started vaccine production prior to approval which necessarily included supplies such as vials, bottle caps, packing and labeling materials, etc.

Creating a Value Stream, not Just a Vaccine

It seems clear, then, that the approach taken by Operation Warp Speed is to not focus the vaccine development effort on just a vaccine, i.e., on the product.  Rather, the effort had to focus on creating a new value stream for Covid-19 vaccination for a substantial portion of the nation’s population. In fact, the mission was formulated as follows: “Operation Warp Speed will accelerate the testing, supply, development and distribution of safe and effective vaccines, therapeutics and diagnostics to counter COVID-19 by January 2021.” Had efforts focused on just the vaccine, supply chain issues would not have been worked on, nor scaled up manufacturing processes, nor distribution networks. Having a vaccine without the ability to make it, package it and administer it to the population would do little good.  Similarly, without simultaneously working on these elements of the value stream, the timeline would not have been accelerated to the same extent.

Information Flows Versus Batches

The ability to overlap phases depends heavily on the flow of information, specifically the frequency and bandwidth of communications and data transfers.  In the traditional drug development process, information moves between companies and government agencies in batches.  Research, small-scale production and testing results are submitted as a batch to the government along with an application to move to the next phase.  The government agency then reviews and approves the application for the next phase, which of course takes time. Meanwhile progress on vaccine development slows or even halts while the company awaits the government’s findings.  By making its rules and processes more flexible to allow a rolling release of information from company to government agency, companies provided data to the FDA as it became available rather than at the end. Thus, data could be processed as it came in and preliminary findings communicated back to the company more quickly (and frequently) which reduced the companies’ uncertainty about continuing to move forward.  This rolling release approach results in fewer pauses and slow downs in development and testing.

The People Side

A final piece of the puzzle, and easily the most important, is the dedicated effort, knowledge and creativity of the people involved with the Covid-19 value stream development effort. In both the private sector and government agencies, teams of experts were dedicated full-time to it.  Not only does this eliminate waste associated with putting a job aside, switching to a different task, then coming back up to speed on it once it’s time to pick that job back up again, individuals become experts in the minute details and subtleties of the effort making them more efficient in their jobs.  Further, Covid-19 vaccine work need not wait in the queue behind other work these individuals could have been assigned, reducing wait times and other waste.

Although this effort was coordinated by the federal government, private companies and other organizations can take stock of the importance of each of the principles when it comes to developing new products and services:

1. Build knowledge through research, development and experience that can be leveraged in the future. That knowledge cannot reside solely in the heads of scientists, engineers and managers.  It must be documented so as to be accessible and usable on future endeavors.
2. Focus on creating a value stream, not just the product/service.  The product/service, of course, will be the centerpiece but it will not deliver value (or at least as much value) to stakeholders without an intentional focus on standing up the value stream end-to-end.
3. Invest in multiple alternatives to mitigate risk and crash time-to-market.
4. Parallelize and synchronize development activities through steady information flow. Eliminate batching.
5. Build teams who have the necessary expertise, who focus intently on the task and who increase their expertise in completing their work.

A major lingering question, though, is the role of government funding.  $10 billion has been spent by the US government, not to mention by companies and other governments across the globe.  Potentially hundreds of millions of dollars could be spent on efforts that ultimately do not pan out. So, while it appears to be working at getting a vaccine delivered faster than ever before, it also appears to be very expensive.  Of course, when we’re talking a global pandemic with millions of lives and livelihoods at stake, it is totally worth it.  But how critical is this enormous level of funding, and would a private company be willing or even able to afford such an approach?  Look for some thoughts on this question in a future blog post.