Make your own spare parts

The tiny body on the operating table is not breathing. All around him the medical staff at the C.S. Mott Children’s Hospital in Ann Arbor, Michigan, work with a purposeful, practised efficiency. You’d never know that the team is attempting a first in emergency surgery.

The surgeon, Glenn Green, and his colleague Richard Ohye, have already opened three-month-old Kaiba Gionfriddo’s chest and rerouted the pulmonary artery that had wrapped itself python style around the airway leading to his left lung. But the crushed bronchus, one of the tiny branches leading from his windpipe, is weak and flaccid. Air cannot get through. The lung is still not moving.

Green picks up a small, C-shaped tube of white plastic, roughly the size of a cigarette filter. He places it around the collapsed bronchus, and begins to sew it onto the exterior wall. The rigid plastic splint holds the bronchus open, and with the last stitch in place the lung suddenly begins to move up and down in a rhythm reassuringly reminiscent of a sleeping child.

It is 9 February 2012, and this is the first time that an implant created by a 3D printer is being used to save a human life. “It was scary,” Green recalls.

“We characterise it as a ‘Hail Mary pass’,” says Scott Hollister, the biomedical engineer at the University of Michigan who made the splint. “I was elated when I heard it’d gone fine.”

Kaiba’s surgery was reported in May 2013 in the New England Journal of Medicine. When it comes to body parts, the splint is a simple one. But the successful surgery serves as a clarion call. The ‘Star Trek’ technology that in a matter of hours turns a virtual object on a computer screen into a real one on the surgeon’s table has arrived. Splints are just the beginning. In a few years, 3D printed tissues will be reconstructing peoples’ faces and providing women with customised breast tissue.

And in laboratories around the world, researchers are already tackling the ultimate challenge: how to print, using a patient’s own cells as the “ink”, the intricate mosaic of a bespoke organ. Creating an entire organ might not even be necessary. Printed patches of tissue could be enough to restore sufficient function to get a patient back on their feet again.

The challenges are daunting but with the waiting list for organs now stretching beyond 100,000 in the US alone, 3D printing offers a bold new solution. Some are confident it’s just a matter of time. “In the next 30 years we will be printing at least partial kidneys, livers and possibly whole organs,” predicts Terry Wohlers, an industry analyst at Wohlers Associates of Fort Collins, Colorado.

Since the dawn of the Stone Age, people have made tools and other objects by chipping away at blocks of rock or other material until they were left with the item they needed. Many of today’s production methods still rely on the same “subtractive” approach. But over the past 30 years, “additive” manufacturing, as 3D printing is more properly known, has begun to offer an alternative. 3D printing uses computer-controlled machines that build up three-dimensional objects from scratch, layer upon layer. Intricate plastic or metal objects can be made this way, including extremely complicated shapes that could not be made using other methods.

“We realised fairly quickly that the two obvious applications for high-end additive manufacturing parts were aerospace and medical,” says David Bourell at the University of Texas, Austin, a 3D printing pioneer. But what really spurred the field, says Bourell, was the increasing availability of intuitive 3D computer design software. “There was no need for a cheap 2D printer until we got desktop computers,” he says. Likewise, once people could easily produce 3D shapes on their computers, it was inevitable they would want a machine to turn the simulations into real objects. The engineering industry quickly adopted the technology to build prototypes, and medicine was not far behind.

The first applications found their mark in perfectly individualised metal and plastic prosthetics. The shape, contour and colour of a person’s teeth, for instance, are as unique as they are. “In the future, more and more dental technicians will pick up a computer mouse instead of a wax knife and use an intraoral scanner instead of an impression tray,” says a recent press release from German company EOS. But these synthetics are low-hanging fruit. The eagerly awaited main harvest is implants made the same way nature makes them, a sculpted organic scaffold plastered with cells.

Think tissue transplants, and organs such as the kidney, liver or heart, come to mind. But it was a different body part that led Hollister to 3D printing. It was the one that defines us, the human face. An entwined mosaic of bone, sinew, flesh and blood, it is a dauntingly complex piece of work. And of course, each face is unique. For cancer and gunshot victims, the loss of facial features adds a whole extra layer to their agony. Today’s crude synthetic prostheses do a poor job of integrating with the scarred remains of a face. In the mid 1990s, Hollister set out to see if biomedical engineering could offer something better. Tagging along with a friend to a conference on 3D printing, he had a Eureka moment. “I realised there was no better way to make these types of structures.”

Scaffolds could be printed and seeded with the patient’s own cells. Once implanted, the structure would be a perfect fit, and the growing cells would form a seamless connection to the rest of the face. The question was: which material to use for the scaffold? By 2004 he was experimenting with a printer that used a laser to heat up fine particles of a biocompatible plastic called polycaprolactone until they softened just enough to stick together (a process known as sintering). “The nice thing about polycaprolactone is that it’s fairly ductile. It doesn’t crack, it squishes a bit,” says Hollister. It also slowly breaks down in the body over about three years making it especially useful for children who quickly outgrow their implants.

Another good thing about polycaprolactone was that it was already being used in the clinic to mend holes in the skull. In cases of severe head trauma, internal bleeding exerts pressure on the brain, which surgeons relieve by removing a chunk of the skull. The conventional way to close the so-called “bone window” is with a titanium mesh.  But not only is the mesh expensive, the bone never grows back beneath it, and young children usually need a second operation to swap in a bigger plate as they grow. In the early 2000s, Dietmar W. Hutmacher of Queensland University of Technology (QUT) in Australia developed the first and second generation of polycaprolactone-based scaffolds with colleagues at the National University in Singapore. They were not only cheaper but also allowed bone to regrow naturally through the holes in the scaffold as the plastic dissolved. Thousands have since been implanted.

Hollister was still developing his bespoke polycaprolactone facial implants when he met Glenn Green in 2011. An associate professor of paediatrics at the University of Michigan, Green was looking for a way to make implants to repair children’s airways. Hollister listened to Green’s list of requirements then 3D printed a series of prototypes, all ribbed cylinders with a slice missing. With their C-shaped cross-section they were able to flex and expand as the child grew. They also sported a series of minute holes along their length so they could be sewn on to the exterior of the tissue.

The splint’s structure might sound simple – something that you might carve by hand – but it is anything but.  As Hollister points out, 3D printers craft objects with a precision far beyond the steadiest scalpel, building in the structural features that allow the splint to expand as the child grows. The holes in the splint are designed to be just the right size and shape for a surgeon to sew through. And each printed splint is identical, so multiple copies can be tested over and over to see how much pressure it can take. This helps the surgeon to be sure that the splint  which is actually sewn into the child  will perform in exactly the same way as the tested one. Improving the splint with each iteration, it took Hollister and Green just a few months before they had a device that maintained its structure and stayed in position when sewed into place during tests on human and pig cadavers.

Finally, Green and Hollister decided they were ready to make preparations for a clinical trial.

It was in January 2012 that the story took a dramatic turn. Green got an email from a clinician at the Akron Children’s Hospital in Ohio describing a patient who “may be a good candidate for your splint”. It turned out to be an understatement. Kaiba Gionfriddo had the most severe case of a condition called tracheobronchomalacia that Green had ever heard of. The artery wrapped around Kaiba’s left bronchus had left the cartilage floppy and useless giving the child the symptoms of a constant, severe asthma attack. “Outside an ICU, he would have died in a day,” says Green. There was no way that Kaiba could wait for a clinical trial to start. But going ahead with an implant was risky. The team had not even tested the splint in a live animal model. Nevertheless, Green got permission from the hospital’s review board, the US Food and Drug Administration (FDA), and Kaiba’s parents.

Shortly after Kaiba arrived at the Ann Arbor children’s hospital on 1 February 2012, Hollister used CT scan data to make a 3D computer model of the child’s lung and airways, then with the help of drafting software designed the splint to fit snugly around the collapsed bronchus. Finally the program sliced the virtual splint like a loaf of bread. This produced a series of templates to guide a laser, which traced out layer after layer in a printer bin containing finely-milled polycaprolactone powder. The zapped particles stuck together, with loose powder remaining at the sides, holding the growing structure in place. After each layer was sintered, the printer swept a 0.1 millimetre thickness of powder across the build platform, ready for the next laser scan. Several hours later the bin was full to the top with powder. After digging it out “like a kid with a sandbox”, Hollister sent the splint to be sterilised before Green sewed it into Kaiba during the four-hour operation. Three weeks later, the infant no longer needed a ventilator.

Kaiba’s small body has been able to grow its own cells around the splint so that he will have an entirely natural bronchus in a few years time. But the body is not always capable of regenerating tissue in this way. Sometimes the damage is just too extensive and nature needs a little extra help. Hollister is on to it. He is now close to achieving his original goal of printing parts for facial reconstruction surgery. Using a 3D printed scaffold, he soaks it in a brew containing natural cell growth factors and then hand-seeds it with cells. Rather like preparing a lawn with fertilizer and then sowing it.

The concept of seeding cells on to an artificial scaffold is not new. Almost 20 years ago, images of a mouse bearing a seemingly human ear on its back hit the headlines. The ear had been fashioned from an artificial scaffold impregnated with cartilage-generating cells. But today a patient in need of an ear is unlikely to be offered anything like this. Most often, they’ll have to surrender a piece of their rib cartilage for the surgeon to manually sculpt (as best they can) into an ear shape.  Patients have also had jaw reconstructions built from chunks of bone taken from their legs. “Surgeons can do remarkable things but it does have disadvantages,” says Hollister. Not only does it involve major surgery to harvest the leg bone, it’s also difficult to carve it into a realistic jaw shape.

3D printing promises to be the disruptive technology that finally wins surgeons over to artificial scaffolds. The procedure looks promising in animal tests. Working with researchers at the University of Illinois, Hollister printed a section of a pig’s jaw, seeded it with cells and implanted it on the pig’s back. Six weeks later, the implant had begun to grow bone, tissue and blood vessels and earlier this year the whole thing was transferred to replace a missing section of the animal’s jaw. “Yesterday the pig started eating again!” says Hollister in amazement. He is also printing human ear and nose shapes using polycaprolactone, seeding them with cartilage-forming cells and implanting them on the backs of lab animals. Green admits, “It’s a little eerie seeing a human ear on a pig.”

People occasionally need new ears and noses, but a growing multitude of women need new breasts. Following a diagnosis of breast cancer or, as in Angelina Jolie’s case, a genetic test showing she carried a high-risk breast cancer gene, the end result is often a double mastectomy. A woman can opt to have synthetic implants provided she has been able to retain the skin sacs of the breasts. If not, she faces a painful reconstruction using tissue from her stomach or hips.

3D printed breasts offer a new hope and Hutmacher is trying to turn it into a reality. Using data from laser scans of the patient’s breast, his team can print a soft, flexible breast scaffold made of porous polycaprolactone and polyurethane. The pores are filled with a cocktail of the patient’s fat cells, harvested from liposuction and suspended in a jelly-like matrix called a hydrogel. Compared to current reconstructions that use only a patient’s fat cells, he says these scaffolds should do a better job of achieving the right shape and volume and there should be no shrinkage. So far, Hutmacher’s team has shown that the procedure works well in pigs. Over three years, the scaffold dissolves leaving natural-looking breasts that have incorporated the pig’s own cells.

As the famous “ear-on mouse-back” photos attest, animals have now been sporting human-like body parts for a long time. But so far there are very few clinical trials of 3D-printed organs. “To see a major translation to the clinic we need big companies to come in to cover the expense of clinical trials. But if the processes aren’t scalable, they won’t buy into it,” explains Hutmacher.   One of the limitations to scaling up so far has been the slow process of hand-seeding scaffolds with cells.    Another is that tissue engineers have not yet solved the problem of how to provide printed tissue with a blood supply.

A new generation of so-called “bioprinters” may be the answer. In the cartridges of a bioprinter, living cells are treated as if they were just another type of “ink”. The machines not only print the scaffold but at the same time fill it with living cells, layer by layer. They can also print a rough approximation of the plumbing – the blood vessels that will supply the cells with the nutrients they need.

Researchers have printed almost every cell type imaginable in this way, including muscle and brain. In 2011, Anthony Atala, Director of the Wake Forest Institute for Regenerative Medicine in Winston-Salem, North Carolina, unveiled a 3D-printed kidney to the awestruck audience of a TED talk. It was made of more than 20 different human cell types combined with a hydrogel matrix, and took about seven hours to print.

Hutmacher too has converted to bioprinters and is already working on his next version of breast tissue that will be printed with its plumbing system intact.

In this strategy, he will use a bioprinter to print a tube composed of so-called endothelial cells, the ones that normally line blood vessels. Surrounding them he will print the cells that coat the outer layer, so-called smooth muscle cells and fibroblasts. Experiments from Hutmacher’s lab and others have shown that when these three cell types are put together, they will self-assemble to make a blood vessel. Finally the printer will deliver breast-making cells, juvenile long-lived fat cells called preadipocytes, as a type of cladding around the blood vessel. In terms of a timeline to success, Hutmacher says, “It is the longest strategy but it will be the best.”

Indeed functional organs are still a long way away. For all its complexity, Atala’s kidney was far from functional. Today’s bioprinters simply can’t yet recreate the finer structural details of real tissues, and so the tissues they produce are no match for nature. Developmental biologist Ian Smyth at Monash University in Melbourne, Australia, provides a reality check by pointing to the difficulty of making skin grafts for burn victims. Arguably the simplest of tissues, bioengineers are still unable to regenerate a sweat gland, so the size of an artificially grown skin graft is limited, lest the patient overheat. “I suspect we’ll be printing houses before we’re printing organs,” cautions Smyth. Overall, the detail of how an organ develops is still largely a black box. Mystery surrounds the precise identities of cells, and how they interact to give rise to a functioning tissue. Hutmacher is under no illusions. “Bioprinting is really complex. You need materials scientists, hardware specialists, software people, cell and molecular biologists.” He is now helping to set up a masters program in bioprinting, a collaboration between Queensland University of Technology, Wollongong University, and two institutions in The Netherlands and Germany. But the perfect should never stand in the way of the possible. A printed functioning organ may still be far away, but a good facsimile of a functioning tissue is not, as the latest results from San Diego-based bioprinting company Organovo demonstrate.

The liver is the body’s chemical factory producing hundreds of compounds that are crucial to everything from energy metabolism to neutralising toxins. Hepatocytes are the cells that do all the work and they are famously regenerative, able to regrow an entire liver from only a remaining third. Yet grow them in a culture dish and they wither and lose their function. But Organovo recently celebrated the one-month birthday of a sliver of 3D-printed liver tissue. The secret? The cells had been printed to mimic their natural architecture: hepactocytes arranged along a border of the endothelial cells that line blood vessels. Slivers of tissue up to one millimetre wide can survive in culture thanks to a primitive blood vessel network that develops between the cells. The tissue was also performing critical liver functions such as synthesising cholesterol and detoxification.

Tissue slivers like this could be used by pharmaceutical companies as guinea pigs to screen new drugs for liver toxicity, one of the most common reasons for drugs to be pulled off the market. Sharon Presnell, Organova’s chief technology officer, believes the working slivers of tissue created in her lab also offer hope of being developed as transplants. “I really think simple patches are achievable in a decade.”  Those sorts of tissue Band-Aids might be enough to restore sufficient kidney function, agrees Atala.

3D printing has already created an astonishing variety of prototype body parts but getting them into the clinic requires demonstrating their unique merits as well as their safety and effectiveness. Dramatic cases like those of Kaiba’s life-saving splint put 3D implants on the radar, says Hutmacher. It may have been a plain scaffold, made of a material already used in other patients, but its success was never a foregone conclusion. “The ground-breaking thing,” he says, “is that they did it.”