ON the outside it is an ordinary brick building. Inside, scientists pad around in white suits and blue booties amid ranks of stainless steel incubators. This is the assembly line of tomorrow. In an atmosphere that is 250 times cleaner than the air outside, its operators are tending to a product that has never before been mass-produced: human skin. Later this year California-based biotech company Advanced Tissue Sciences hopes to get the go-ahead for full production, growing translucent sheets measuring four by six inches from the cells of foreskins.
This will be the first commercial venture for the young, yet potentially lucrative technology of 鈥渢issue engineering鈥�. It won鈥檛 be the last. Also in the pipeline are products based on laboratory-grown cartilage and breast tissue. Some companies hope to equip surgeons with ready-made spare parts to replace damaged or worn tissue. Others want to customise replacement tissue for individual patients. 鈥淚n the next 25 years, we will replace more and more sick tissue with new tissue, instead of trying to treat it therapeutically,鈥� predicts Michele Barzach, a former French Minister of Health who now advises companies like ATS on how to introduce new medical technologies into the French market.
Optimistic words. Yet there are substantial scientific and medical obstacles to be overcome before off-the-shelf tissues can become a mainstay of the surgeon鈥檚 toolbox. For a start, growing 鈥� and especially mass-producing -tissues is far from routine, even though ersatz human skin has been available for small numbers of patients since 1988. Engineered tissues are only useful if they can be integrated with healthy tissue in a patient鈥檚 body. This means ensuring that implants have a blood supply and can react normally to signals from genes and hormones. Another problem is how to prevent cells from tissue implants migrating to other parts of the body where they may cause diseases such as cancer. And how will tissue engineers stop the rejection that commonly occurs when foreign tissues are transplanted into a patient?
Advertisement
So far, researchers have been able to side-step these difficulties by carefully selecting the tissues they cultivate. In future, they will need to face these problems head-on if the technology is to produce the goods.
Even so, progress has been remarkable. Ten years ago, creating human tissues in the laboratory would have been unthinkable. There were just too many obstacles, not least the complexity of the signals that pass between cells as they build a tissue. The breakthrough came when researchers realised that cells could solve these problems for themselves; that they need little more than an inert scaffold to grow on. It was this that brought together Joseph Vacanti, a surgeon at Harvard Medical School in Boston, and Robert Langer, a chemical engineer across the river at the Massachusetts Institute of Technology in Cambridge.
Vacanti, a transplant specialist, had been frustrated by the chronic scarcity of tissue for transplant. He had experimented with building makeshift replacement organs from tissues taken from other parts of a patient鈥檚 body. 鈥淭hat method works in principle,鈥� he says. But all natural tissues have evolved to perform particular functions and the stress that transplantation to another part of the body places upon them can lead to complications. However, artificial implants such as knee joints or heart valves also have their drawbacks. Because they cannot grow they are of little use in children. Even in adults they rarely last more than 15 years. Tired of having to improvise, Vacanti decided in 1984 to try to design replacement tissues in the laboratory. His insight was to collaborate with Langer, a biomaterials expert with over a decade of experience developing synthetic polymers to use in the human body, for example to deliver measured doses of drugs.
While biologists had been growing cells in culture for years, they had done it in only two dimensions, with most cells squatting on flat surfaces or floating around a nutrient broth in tiny amorphous blobs. Vacanti and Langer realised that if they were to engineer human tissues they must help the cells conquer the third dimension. So they devised synthetic compounds made of polylactic acid and polyglycolic acid 鈥� biodegradable polymers used as surgical sutures. The researchers processed hard pellets of these plastics into felt-like, porous textiles containing about 97 per cent air, then moulded these to give scaffolds through which cells could grow into tissue. The beauty of this approach is that once implanted, the scaffold simply dissolves into carbon dioxide and water.
One of the first tissues they attempted to recreate was skin. Although flat, skin is a highly complex organ made up of two layers with very specialised functions. Several groups of tissue engineers have attempted to recreate skin over the past decade. The incentive is financial. Companies testing products for patients with stubborn bedsores or diabetic ulcers estimate, perhaps optimistically, that for these two applications alone the world market approaches $3 to 5 billion annually. The other potential market is burns victims. This is smaller, but it could still reach $300 million.
The tissue engineers soon discovered that they could minimise their problems and still produce functional tissue by growing the two layers of human skin separately and allowing the body to generate the other after transplant. The dermis, the inner of the two layers, is the most popular choice, because it creates fewer problems of rejection. This is because the fibroblast cells that make up that layer lack surface molecules called MHC class 2 antigens. Most of our cells have these marker molecules, which are highly specific to each person and so allow our immune systems to recognise and attack foreign cells that are not similarly labelled.
Working under licence from a biotechnology company called Neomorphics, bought in 1992 by ATS, Vacanti and Langer began experimenting with their new polymer scaffolds. They seeded them with healthy young fibroblast cells from the foreskins of circumcised infants. Within weeks the cells had grown through the polyglycolic acid mesh to produce sheets of dermis. The next step was mass production. Last year ATS completed construction of its 鈥渟kin factory鈥� at La Jolla. Already it has stockpiled enough fibroblasts for years of manufacture. 鈥淭hat wasn鈥檛 hard to do鈥� says ATS vice president Gail Naughton. 鈥淎 single foreskin yields six football fields鈥� worth of dermis.鈥�
But ATS has still to seek approval for its engineered skin from the Food and Drug Administration. It is hoping to get the go-ahead for treatment of diabetic ulcers some time this year, following successful trials on 400 patients at 20 clinics in the US and 16 in France.
Its study, reported last August at the annual meeting of the European Tissue Repair Society in Oxford, showed that the cultured dermis stimulated closure of deep foot ulcers in half of the patients compared with 8 per cent in the control group. They received standard ulcer treatment consisting of cleaning the wound with saline and dressing it with gauze. When placed onto a wound, the opaque sheets, resembling cheesecloth, secrete hormones that spur healing. 鈥淏lood vessels from underneath the dermis grow in, and over a year鈥檚 time, the dermis completely remodels itself with the patient鈥檚 own cells,鈥� says Naughton. Also, cells from the outer layer of healthy skin around the wound migrate in and lay down a fresh layer over the top of the transplanted dermis.
Benefits for burns
Ultimately, severe burns victims could also benefit. At the moment they receive transplants of their own skin. When there is not enough to go round doctors can send a postage stamp sized sample of epidermis to Genzyme Tissue Repair (GTR), formerly BioSurface Technology, in Cambridge, Massachusetts. The company has been providing this service since 1988. The keratinocyte cells that make up this skin layer are cultured into sheets of replacement epidermis and then shipped back. It is an expensive treatment: a piece 5 centimetres by 5 centimetres costs $350 and, because the skin often has to be applied several times, the average bill per patient is $62 000.
While the technique saves lives there is a second major drawback. It takes GTR a month to grow replacement skin from a burn victim鈥檚 own cells. Meanwhile, doctors have to cover the patient鈥檚 wounds to ward off deadly infections. For this they use skin from a corpse, an emergency measure that is growing increasingly unpopular because the dead skin is scarce, gets rejected by the patient and carries its own risk of contamination.
鈥淪urgeons need an off-the-shelf skin that they can put on immediately,鈥� says Naughton. Here, products like ATS鈥檚 dermis would come into their own, buying time for the patient鈥檚 own skin to be grown. Frozen sheets of dermis could be kept alive for at least a year in hospitals, ready for use in emergencies. Clinics are already testing the product.
John Hansborough, a surgeon who directs the burn centre at the University of California in San Diego, is among those doing clinical trials. So far he has treated nine patients including a 15-year-old boy who had third degree burns from a firebomb covering 60 per cent of his body. 鈥淚 was very pleased with the new dermis. It stayed on for 45 days, during which time human cadaver skin next to it [was] rejected several times,鈥� says Hansborough.
This meant that he was able to do a more gradual job of harvesting the boy鈥檚 remaining skin for permanent transplantation, Hansborough says, thereby reducing the long-term problems of scarring and disfigurement.
Now GTR is testing its own engineered skin replacement and plans to apply for US regulatory approval next year, says David Castaldi, a consultant for GTR who was chief executive of BioSurface Technology before it merged with Genzyme in December. GTR has chosen to produce patches of the outer skin layer, the epidermis. The advantage of this approach is that no polymer scaffold is required 鈥� when placed in sealed flasks, the cells crawl on top of one another forming sheets about five cells thick.
GTR hopes that the laboratory-grown tissue will encourage a patient鈥檚 own epidermal cells to move in from the wound鈥檚 edges. And as the outer cells of skin naturally slough off, the patient鈥檚 epidermis will replace the foreign covering. But one big question remains unanswered. Will the patient鈥檚 immune system attack transplanted epidermis? Susan Schaeffer, a scientist with GTR, says that rejection has not been a problem in preliminary trials, but it remains a possibility, especially since scientists do not fully understand what is happening inside a wound.
Rebuilding the breast
Whatever difficulties remain in building skin, at least it has the virtue of being flat. An ambitious project just getting under way aims to build a decidedly more three-dimensional part of the body 鈥� the female breast. David Mooney, a chemical engineer at the University of Michigan in Ann Arbor is collaborating with researchers at the Carolina Medical Center in Charlotte in trying to reconstitute breast tissue lost to cancer. Their work might one day offer women an alternative to silicone and saline implants, both of which have been linked to longterm complications such as autoimmune responses and rampant, painful growth of connective tissue around implants.
Like most research in tissue engineering, this work is being supported by a biotechnology company in return for a licence to make money out of the technology. Jim McNab, chief executive officer of Reprogenesis in Dallas, hopes to reap huge rewards, given that 250 000 mastectomies are performed every year worldwide. But problems abound, says Mooney, who has been working as part of Langer鈥檚 team for the past seven years.
To begin with, the sheer size of a female breast poses a challenge. 鈥淭he tissues grown in the past are only up to a centimetre thick. Even for a small breast of, say, 10 centimetres, you cannot simply scale up the procedures,鈥� says Mooney. The main obstacle is in providing the inner cells of a transplanted tissue with oxygen and nutrients. To do this the tissue needs a blood supply but scientists cannot yet grow blood vessels as strong as those found in the body. 鈥淭he blood vessels we make look histologically just perfect, but when you pass blood through them they blow apart,鈥� says Chris Breuer, who works with Vacanti.
In an attempt to overcome this, Mooney is developing a synthetic 鈥渂reast鈥�, a fine-meshed honeycomb that doctors would implant and then populate with successive injections of the women鈥檚 own cells reared in culture. The first groups of cells, tucked in near the outer margins of the implant, could survive on nutrients diffusing in from the surrounding tissue until blood vessels start to grow into the honeycomb. Once this occurs doctors would then inject more cells into pockets adjacent to the new vessels and deeper inside the breast.
To speed up the process Mooney and his team plan to mix tiny polymer pellets with the implanted tissue. The pellets will lure blood vessels into the new tissue by oozing growth factors 鈥� chemicals found naturally in the body which spur growth. For example, a protein called fibroblast growth factor encourages the division and growth of endothelial cells that line blood capillaries.
Building up the breast a bit at a time brings problems, however. It is tricky to anticipate how the implant will respond to the pressures created by surrounding connective tissue and skin. Further distortion could be caused by tissue growing in from the host, explains Mooney. 鈥淵ou could end up with a shape that would be very inappropriate.鈥� The scaffold he is designing aims to minimise such effects by combining sturdy polymer braces on the outside to give mechanical stability with a fluffy, highly porous polymer centre through which the cells can grow.
Normal female breasts are made largely from fatty tissue. But to date there has been little reason for scientists to study fat cells and understanding of how to grow them in culture is poor. Mooney and his team aim to get round this problem by growing breasts made from smooth muscle cells. More usually found lining intestines, blood vessels and gall bladder, these cells have been the subject of much research. Smooth muscle tissue has an elastic, springy feel that makes it a leading candidate for replacement breasts, says Mooney.
But who wants a breast that could flex like a muscle? Mooney鈥檚 team intend to ensure that won鈥檛 happen. The slow, rhythmic undulations of smooth muscle occur because the cells align in groups that contract together. 鈥淲e hope to manipulate the material such that the cells grow in a more chaotic pattern, he says. 鈥淪o you might have individual contractions of cells, but no large-scale, noticeable motion.鈥� It is easier said than done. Mooney estimates that it will be four years before he can begin human trials.
Meanwhile a third engineered tissue is already being tested. Cartilage, the tough slippery shock absorber that lines the ends of bones and prevents them from grinding together, was the first tissue that Vacanti and Langer attempted to recreate. It has two major advantages for tissue engineers. First, it has far fewer blood vessels growing within it than most human tissues, so the problems of integrating implants with a patient鈥檚 own tissue are minimal. Moreover, cartilage seemed an easy target because it consists of only one cell type.
Another enticement comes from Wall Street. In the US alone, surgeons perform around 628 000 operations a year involving damaged joint cartilage, though at present they can do little more than smooth frayed edges and lubricate the joint to alleviate pain. About 13 per cent of these operations are on people with wounds like potholes, where after years of arthritic decay the only option is to replace the entire joint. ATS hopes to patch up the damage using ready-made cartilage 鈥減lugs鈥� based on a technology developed by Langer and Vacanti.
ATS is currently testing its plugs in rabbit knees. They excise damaged cartilage, cut out a slightly larger piece of homemade tissue and simply pop the tough but malleable blob into the hole like a piece in a jigsaw puzzle. 鈥淭he cartilage integrates into the surrounding host cartilage and gives a nice, smooth surface,鈥� claims Naughton.
Assembly line
Human trials should begin next year Meanwhile, the company has designed an assembly line for the mass production of cartilage plugs. Each is grown in its own bioreactor, a plastic capsule containing a polymer scaffold and fluids designed to simulate physiological conditions. The capsules containing polyglycolic acid meshes around 5 millimetres in diameter are strung together in their hundreds so that the process can be computerised.
First, healthy cartilage cells 鈥� the leftovers from standard surgical treatment 鈥� are sprinkied onto the scaffolds. A nutrient brew containing carbohydrates, amino acids, and vitamins, is then added. The computer keeps these concentrations constant throughout the four-week growth period, while recirculating growth factors secreted by the cells and removing wastes such as urea. Once the plugs are ready they are frozen, allowing them to be shipped and stored in hospitals.
It sounds simple, but problems remain. Langer鈥檚 co-worker Lisa Freed, who studies the growth of cartilage in the test-tube, says that after 3 to 4 weeks in the bioreactor, the tissue contains only about a third of the diverse proteins and sugars that give cartilage its essential properties of toughness and lubrication. A more natural product can be made given more time, but such tissue does not integrate with the patient鈥檚 cartilage so easily. Unless the researchers can get the cartilage plug to integrate seamlessly with surrounding tissue, lack of nutrients and growth factors will impair its function, says Freed. They do not know how this can be achieved other than by improving the surgical procedure.
Already, other tissue engineers with a less futuristic method of culturing cartilage have improved the lot of several dozen people with severe knee problems (Technology, 29 October 1994). Rather than trying to build generic, three-dimensional cartilage outside the body, a group of Swedish researchers led by Lars Peterson of the University of G枚teborg took samples of healthy cartilage from patients, grew the cells in culture and injected them back into the damaged knee.
Now GTR has licensed the technique, and plans larger trials next year. Because the product is based on a patient鈥檚 own cells, no regulatory approval is required. Treatment involves two surgical procedures: the first to get cells to culture, the second to remove a piece of periosteum 鈥� the skin-like tissue that covers bones 鈥� and then sew it over the implanted cells to keep them in place. Castaldi estimates that about 85 000 patients in Western Europe could be treated in this way 鈥� the same 13 per cent with pothole-like damage targeted by ATS. So how do the competing technologies compare?
ATS is confident that it can produce a cheaper repair than GTR, largely because its method is much less labour-intensive. Neither company is willing or able to give precise costings, but the market cost of a plug insertion could be around $1000. The advantages of the Swedish method, according to Castaldi are that the implanted cells integrate more effectively with patients鈥� healthy tissue.
And because their own cells are used, there is no chance of rejection. By contrast, ATS鈥檚 plugs might be attacked by the patient鈥檚 immune system. However, Vacanti anticipates that rejection might be less of a problem with cartilage implants than it is with other tissues. This is because the cells responsible for immune attack are found in the blood, but cartilage contains almost no blood vessels, making it difficult for immune cells to enter a transplant.
So far, none of ATS鈥檚 cartilage transplants has been rejected. But there is no guarantee that this record will hold once the plugs are tested in humans. Indeed, graft rejection poses such huge hurdles for tissue engineers that in general they are leaning toward custom-tailoring tissues using a patient鈥檚 own cells rather than creating off-the-shelf products. Other scientists are using genetic engineering to try to rid cells of their immune recognition molecules 鈥� the MHC class 2 antigens 鈥� in the hope of making them immunologically neutral, but this work is still in its early stages.
Vacanti and Langer are not the only innovators in this field. William Otto, a researcher at the Histopathology Unit of the Imperial Cancer Research Fund in London, has recently tested another approach that initially looked promising. But when his engineered skin was transplanted it quickly became pulpy as the patients鈥� enzymes attacked the animal collagen cells on which the skin was grown. 鈥淭his approach is probably a non-starter,鈥� he admits. 鈥淲e are back to the drawing board.鈥� Otto is now convinced that polymer scaffolds that can be implanted without fear of rejection will corner the market.