Share
Printing vaccines at the pharmacy or at home will be the way of the future

Printing vaccines at the pharmacy or at home will be the way of the future

Enlarge / Artist’s impression of a vaccine printer.Getty / Aurich Lawson reader comments 3 with 3 posters participating Share this story Share on Facebook Share on Twitter Share on Reddit We’re running a series of companion posts this week to accompany our special edition Ars Lunch Break podcast. This is the third of three guest…


Artist's impression of a vaccine printer.
Enlarge/Artist’s impression of a vaccine printer.
Getty / Aurich Lawson

reader comments

3 with 3 posters participating

We’re running a series of companion posts this week to accompany our special edition Ars Lunch Break podcast. This is the third of three guest posts centered around Rob Reid’s TED talk from Tuesday. Today, microbiologist Andrew Hessel weighs in with his opinions and recommendations about the future of biomanufacturing.

The US government doesn’t skimp on bio-preparedness. Vaccines and other countermeasures are carefully developed in anticipation of disease outbreaks or bioterrorist attacks. The Strategic National Stockpile maintains a hefty inventory of medicines, supplies, and equipment, which can be shipped almost anywhere within 12 hours. In situations ranging from the 2001 anthrax attacks to 2016’s Zika scare, Americans have been lucky to have strong biodefenses.

But as anti-vaccine hysteria allows measles to regain long-lost beachheads, we’re reminded that human folly is a dynamic element of the disease landscape. Meanwhile, the number of human actors and actions in a position to stir the pot is set to explode. Tremendous improvements in core bioengineering technologies are tearing down the technical and economic barriers that once prevented the development of “designer” viruses and bacteria. Those entrusted with our defense will inevitably face an even more chaotic battlefield than exists today.

Currently, our vaccine inventory is designed to defend against a very short list of well-known diseases. Vaccine fragility calls for refrigeration and expiration dates, as well as regular testing and replenishment. Deployment requires transportation, communication, and person-to-person networks to be functioning.

If infections can arise from engineered organisms with no natural precedents, agility in response is paramount. If we develop an agile threat response system, it can handle engineered and emerging diseases, as well as the old threats from familiar pathogens, which would make our existing National Stockpile obsolete.

Be flexible

A model for truly effective biosecurity lies in the dynamism of our own immune systems. Human immunity is astoundingly sensitive and nimble, capable of sensing and responding to almost any invader. The technology to build a global pathogen detection network that sniffs out threats in a way similar to our bodies’ immune systems is within reach. Technology drove exponential curves that cut the cost of genomic sequencing by a factor of three million. As similar approaches expand to other areas of biotechnology, highly acute sensors could become inexpensive enough to follow the path that smoke detectors took to the ceilings of every home and business within a decade or two. Instead of changing batteries once a year, you might change a pack of solutions every few months.

But detection has limited value if the system can’t respond to carefully identified threats. And key elements of our responsiveness are atrophying.

Fewer vaccines and antibiotics are being made as companies focus on higher-margin medicines. Pharmaceutical manufacturing requires specialized facilities that are not widely distributed. Drug development invariably takes years. Should a truly novel pathogen appear—due to bioterror, bio error, or a natural run of bad luck in a world that can produce things like Ebola by chance—a defense centered on stockpiles could be swiftly outmaneuvered.

It’s time to rearchitect our defenses by leveraging the very force of proliferation that threatens to destabilize the system. Rather than warehouses of refrigerated cures for static diseases, we need a highly distributed agile system for producing vaccines and medicines. We need biomanufacturing at the edge—not just the hub.

But don’t envision edge biomanufacturing as giant factories and smokestacks. Instead, think of bio-printers that resemble inkjets, flexible enough to print a wide array of medicines.

Print on demand

This isn’t as far-fetched or as far-off as some might imagine. Medical printer prototypes are already out there. For instance, biotechnologist Craig Venter’s team unveiled its “digital-to-biological converter” in 2017. About the size of a chest freezer, the “converter” is a DNA printer mated to a liquid handling robot. It produces genetic “programs” for the downstream production of biologics, including proteins, vaccines, and viruses. These capabilities are now being miniaturized for the desktop.

While traditional vaccines involve producing proteins or even entire organisms on a massive scale, tests have shown that it’s possible to vaccinate an animal by injecting some of its cells with DNA that encodes one of a pathogen’s proteins. So a miniaturized DNA printer may be all we need to protect ourselves from many diseases.

Imagine versatile self-upgrading bioprinters extending into every pharmacy and medical office—each with a vast FDA-sanctioned repertoire of templates. This would be a game changer in public health and emergency response.

Think of bioprinters that resemble inkjets, flexible enough to print a wide array of medicines.

Prepping the nation for flu season? No need to guess which flu variant will be spreading months in advance and then bet it all on a massive centralized production run. Just print the precise vaccine required at thousands of locations across the country, adjusting the design to account for genetic drift. And in a worst-case bioweapon nightmare, antidotes made in every neighborhood will get to where they’re needed, unlike ones made in a lone urban center with fast-unraveling distribution networks.

Low food miles are fashionable. Low vaccine miles could be lifesaving.

Biochips and beyond

We can and should push our production nodes to be ever smaller, cheaper, and more widespread. This is already happening with key enabling technologies. We’ve moved beyond microfluidics to nanofluidics and molecular electronics. DNA sequencing is already done with “biochips” that are close kin to electronic chips; DNA and protein synthesizers may soon be chipsets, too. These would control the synthesis of molecules in a way that gets rid of the need for complicated chemical reactions and extensive infrastructure to supply large volumes of the right precursors. Entire biotech and pharmaceutical companies could be reduced to a few square millimeters, ready to be installed in “smart” syringes, inhalers, patches, and implants.

“Biomanufacturing at the edge” could start at the pharmacy, then move to our homes, then to our pockets, and finally on or under our skin.

This last step is feasible because of something I mentioned above: human cells are tiny manufacturing plants, which continuously make thousands of proteins and other compounds based on blueprints stored in DNA. If we give them the right DNA, they can make vaccines for us. Looking past the near term and into the murky future, our ultimate agile defense layer could be built around tiny biochips, which print nucleic acids and then deliver them into our bodies. Imagine a dermal patch that integrates these chips with nanoneedles, which inject their output into the skin’s epithelial cells. These cells could then churn out protein-based therapeutics targeting the latest biohazard.

Some well-targeted R&D muscle could make this sort of system attainable within a decade or two. Paired with a biosensor network of the sort that George Church described on Ars yesterday, it could supercharge global public health efforts, improve national biosecurity, and put the biopharmaceutical industry on a technical foundation better suited to address the demands and threats of the 21st century.

Andrew is the CEO of Humane Genomics, a seed stage company developing synthetic viruses targeting cancer. He is also a co-founder of the Genome Project-write, an international scientist-led effort to advance whole genome engineering.

Read More