COVID vaccine development was astonishing, but it's not over

I am an ever-optimistic believer in the power of life science to solve the problems that matter the most to us all. People die of diseases now that will in the future be cured by life scientists; we will make more, more nutritious food, with more efficiency; and we can shift the manufacture of much that we need in our lives to bio-based, sustainable production.

Even with this belief in the power of biotechnology, it’s been astonishing how rapidly vaccines have been developed and rolled out. COVID is not beaten, but vaccines have already saved huge numbers of lives [1]. We simply wouldn’t be in the position we are in today without the blisteringly fast development timelines that were achieved by multiple groups all over the world. At the time of writing, 3.76 billion doses have already been administered worldwide [2]. It has been a scientific, engineering, and logistical endeavor of Apollo-sized ambition and triumph [3].

Have we really arrived at a new normal?

This lightning-fast development has led some to extrapolate the implications for the therapeutics industry as a whole [4]. A recent McKinsey article leads with “The standard view in the pharmaceuticals industry has been that it takes many years to develop a new drug. And, until recently, that has generally been true” [5], implying that the world’s efforts with COVID vaccines will lead to transformed drug development timelines measured in months. This, and many other articles [6] [7] [8], focus on the accelerated clinical timelines, unprecedented global collaboration, and at-risk investment of vast amounts of funds, and the lessons from these that could be learned for future therapeutic development.

But here’s the rub – they pay no attention to the nature of the biological challenge that was solved in this unprecedented timeframe, in comparison to the complexities inherent in most therapeutic development. With the COVID vaccines, most of the problem was solved ahead of time. In the wake of MERS and SARS and in the knowledge that a global pandemic was just a matter of time [9], foresighted research groups spent years unpicking the biology of coronaviruses and developing platforms designed specifically to enable extremely rapid vaccine design [4]. It’s striking that the actual design of vaccines took only a couple of days for Moderna [10] and Oxford-AstraZeneca [11], and a few hours for Pfizer [12]. They had already solved the fundamental biological problems before they started.

There is also the fact that (whisper it) it was a relatively easy biological problem to solve [13]. Elucidation that spike proteins are good targets for an immune response is a much simpler proposition than working out the complex web of biological interactions that underlie the development of diseases like cancer or Alzheimer’s. The understanding of the biology of these diseases is still significantly incomplete after decades of research and untold billions spent. The NIH alone reports spending almost $20B on cancer research just in the six years between 2013-2018 [14]. The process of target identification and validation, followed by lead identification, validation, and refinement takes many years for most therapeutics, and this won’t change overnight, simply because of the immense complexity of the biology that must be understood. The life sciences community’s response to COVID is similar to more literal moonshots, in that it is hugely inspiring, but simultaneously highlights how far there is still to go.

The future we need

The impact of COVID is massive and ongoing, but future pandemics may very plausibly be dramatically worse and even harder to deal with than this one is [15]. And there are all the other diseases that will one day be cured, where every avoidable delay in the development of their therapies leads to people dying unnecessarily. How can we reduce the years needed for the discovery of new therapeutic candidates?

To transform time to market and reduce attrition of therapeutic candidates in the pipeline, we need effective ways to understand and harness biological complexity. This will inevitably involve AI [16]: human brains are ill-equipped to grapple with the high-dimensional intricacies of biology. But we should also think about what experiments look like in this new world of digitalized, AI-augmented life science. We need scientists to be able to collaboratively envision, design, and refine their experiments anywhere, with cloud-based platforms that then talk directly to automation in the lab. That automation can carry out a scale and sophistication of experiments that would be impossible by hand, which produce comprehensive and robust datasets that can be analyzed, shared, and collaborated on in the cloud.

The good news is that although the speed of COVID vaccine development isn’t an indication that everything in therapeutic development is solved, it is a huge inspiration and a call for action [16]. In this way, it is accelerating the wave of cloud and automation-enabled transformation that will truly unlock the power inherent in biology to solve our biggest problems.