Yesterday we reviewed the TEDxDelMar talks on immunology and living with diabetes; today we approach the second half of the fabulous program. Don’t take my word on any of this, though—I highly recommend you watch the videos for yourself as soon as they’re available!

Beta-Cell Replacement

Michael German: Where Do Beta Cells Come From?

Dr. Michael German is both a professor and an active clinician at the University of California, San Francisco.  His research focuses on the development of pancreatic beta cells, and all the genes and triggers that guide a stem cell to an insulin-producing fate.

Dr. German was the first of a series of beta cell scientists, and he gave an overview of the life of a beta cell, from embryo to pancreas:

The beta cell is a unique cell in that it is the only cell in the human body that makes insulin.

On average, the human pancreas has about one million islets, each of which has about a thousand beta cells, for a total of about a billion beta cells per pancreas. That may sound like a lot, but it sums to a precious few cells, relatively speaking, and they would all fit on the tip of your finger.

The beta cell is key in both type 1 and type 2 diabetes, and replacing the lost beta cells will be essential for any cure for diabetes. In order to better replace beta cells, though, we need first to understand how they develop.

It all starts in the embryo. The gut endoderm is a little tube of tissue that, during embryonic development, branches off to become the liver, lungs, and the pancreas.

The pancreas itself starts as a little cluster of what are called progenitor cells on the side of the endoderm. These cells are a form of stem cell, capable of becoming any of the numerous cells contained in the pancreas.

So, the endoderm cells differentiate into pancreatic progenitor cells, which in turn differentiate again into secondary pancreatic progenitors, which are capable of becoming any one of the endocrine cells in the pancreas, including the insulin-producing beta cells and the glucagon-producing alpha cells. These progenitors don’t actually produce any endocrine hormones yet, though, as the blood sugar of the embryo is controlled by the mother still.

As the embryo nears birth, the secondary pancreatic progenitors differentiate further, becoming mature beta  and alpha cells. This birth of new cells from stem cell parents is called neogenesis.

How can we use this knowledge of the embryonic development of beta cells to our advantage? Well, since there exists a pathway under normal circumstances by which an embryonic stem cell can differentiate a bunch of times and eventually end up as a beta cell, if we can figure out how this pathway progresses, we can recreate it in the lab and induce embryonic or other stem cells to become beta cells.

How does a stem cell progress along this differentiation pathway? At each stage, a certain set of proteins need to exist within the cell; if scientists can figure out which genes for those proteins need to be expressed at each stage, we can walk a stem cell along the path to insulin production, telling it at each stage by expressing certain genes what it should differentiate into.

So, how far have we come?

• German and others have identified many of the genes along this pathway, and perhaps the majority of genes, according to German.
• A company in San Diego, ViaCyte, was the first to put theory into effect by differentiating embryonic stem cells into endoderm stem cells, and subsequently they and others have taken us all the way to the secondary pancreatic progenitors.
• However, that last mile is proving very hard—we can make cells that produce insulin, but they behave more similarly to the progenitor cells than mature beta cells.

The clues for this final step, and for figuring out how to make existing beta-cells proliferate, will come from looking deeper at beta cell development under normal circumstances. As we clarify the crucial pathways and signals that govern processes like the expansion of beta cell mass during pregnancy, we will get closer to being able to regenerate beta cells at will.

Peter Butler: Preserving Your Beta-Cells

Dr. Peter Butler is the Division Chief of Endocrinology and the Director of the Larry Hillblom Islet Research Center at the University of California, Los Angeles. His research focuses on the nature of beta cell death and methods of beta cell regeneration.

Whereas Dr. German spoke about the embryonic development of beta cells, Dr. Butler talked about what happens to beta cells later in life, and how this can help build our understanding of beta cell regeneration:

Butler works primarily with human islets and pancreases, but that proves a difficult task; the pancreas in an inconvenient organ to access, and the islets themselves make up only about 1% of the cells in the pancreas.

When we are born, we have about 100 milligrams of beta cells, or about 100 million beta cells.

In early childhood, we see rapid expansion of this beta cell pool, and by five years of age, we have about 600 milligrams of beta cells.

However, at that point, the growth plateaus, and beta cell proliferation slows down dramatically. (Butler noted that perhaps this is why many diabetics are diagnosed in their teenage years, as the metabolic demand on the body increases rapidly, but the number of beta cells is already established.)

Rats and mice, we know, have the capacity to generate new beta-cells into old age; the question Butler’s lab has asked is, “What is the capacity, if any, to make new beta cells after the age of five?”

We know that it’s not impossible to increase beta cell capacity because we see that there is an increase in the number of beta cells in the pancreas during normal pregnancy. Stains of pancreas tissue show that these new cells do not show up in existing islets; rather, new islets are formed with new cells, meaning adult females at least are able to create new islets and new beta cells.

Butler then asks the obvious question: “If that is the case, why don’t people with type 1 diabetes make new beta cells?”

• The answer, perhaps, is that they do. Butler and his team stained pancreases from people who had type 1 diabetes for more than thirty years for insulin. These stains show the presence of a few islets of beta cells.
• Butler found beta cells even in the pancreas of an eighty-year-old type 1 diabetic.
• This implies that type 1 diabetics are constantly trying to generate new beta cells, but likely these cells are being killed off by the attacking immune cells.

What does this mean for diabetics? There does seem to be some natural capacity to regenerate beta cells, by some combination of neogenesis from other pancreatic cells and subsequent proliferation. If we can harness this capacity, either in plates in a lab, or by stimulating the body’s natural capabilities, this can form an essential part of the ultimate cure for diabetes.

Maike Sander: From Stem Cells to Beta-Cells

Dr. Maike Sander is a professor of cellular and molecular medicine at the University of California, San Diego. Her research focuses on how to turn stem cells into beta cells as a source for beta cell replacement therapies.

Dr. Sander has been working on that journey Dr.German spoke about—making functional, mature beta cells from embryonic stem cells:

As hard as we might try, diabetics cannot accomplish what a normal human pancreas can, and as a result many people are trying to perfect the art of islet cell transplantation.

Islet cell transplantation is currently possible; beta cells are isolated from the pancreases of cadavers, and the cells are infused into the liver. However, this process has two major downsides: it requires suppression of the immune system to protect the infused cells, and beta cells from organ donors are very hard to come by.

Many researchers, Sander included, have therefore set out to find a source of replacement beta cells.

• Thus far, scientists have tried using pig islets, and expanding them in culture, but this hasn’t worked well.
• Currently, some groups are trying to coax cells from other organs like the liver to become pancreatic, insulin-producing cells.
• Another approach, and the one taken by Sander, is to create beta cells from stem cells by instructing the cells how to differentiate.

Sander’s lab works with stem cells.

• Scientists have been able to isolate and maintain embryonic stem cells in culture for many years now, and embryonic stem cells that result from in vitro fertilization can be separated off and grown or frozen.
• In addition to embryonic stem cells, recent developments allow us to create induced pluripotent stem cells, or iPSCs, by taking normal skin cells and expressing certain key genes that transform the cells back into stem-cell-like cells.
• These cells can give rise to any cell type in the body; in the past ten years, we have gotten better and better at telling cells what to differentiate into by forcing the expression of certain genes at certain times during the development of the cell.

How successful have we been at turning these stem cells into beta cells?

• The goal is to imitate certain aspects of human development in a culture dish, and this, as it turns out, is a hard task. It’s not a single step, but requires the orchestration of many factors introduced into the cells at the correct time.
• Still, as Dr. German discussed, we’ve come a long way, and Sander is making progress on that last mile, figuring out what is necessary to make fully functional beta cells.
• An additional challenge will be how to protect the cells from the immune system in the body. ViaCyte and others are working with scientists like Sander to create little envelopes, mini-bags, to hold the cells such that insulin can get out, and nutrients can get in, but immune cells can’t.

How long will all of this actually take? Sander says that to give a date would be disingenuous: “It wouldn’t be science if we exactly knew how to do it.” Much like passing through a maze, science requires taking an educated guess, checking one’s surroundings, and trying again at each roadblock.

Peter Stock: Islet Cell Transplant Update

Dr. Peter Stock is a professor of surgery at the University of California, San Francisco. He specializes in multi-organ transplant, and in finding ways to block immune and autoimmune rejection of the transplanted organs.

Dr. Stock shared some promising news about how far we’ve come in both whole-organ and islet cell transplant for diabetics:

Usually, whole-organ pancreas transplant is only attempted for patients who have progressed to end-stage renal failure and will therefore need kidney transplants anyways.

• At that point, the patient will already need immunosuppressants to protect the new kidneys, and the transplant of a pancreas at that point is not much of an added burden.
• Moving the pancreas is no easy task. It is a risky surgery. As Stock said, “We all learn in medical school never to touch the pancreas, let alone transplant it.” With a few precious islets “lodged in a sea of digestive enzymes,” many surgeons prefer the only-if-absolutely-necessary approach to pancreas transplant.

More recently, as the technique has improved, pancreases are also being transplanted without kidneys, for patients with “life-threatening diabetes,” who have little hope without a new start.

There are about 12,000 pancreas transplants per year, with the number reaching a peak a few years ago and then beginning to taper off. This tapering is likely due to a combination of patients achieving better glycemic control and more islet cell transplants rather than whole-pancreas transplants.

One of the most difficult aspects of organ transplant is that we need to prevent immune rejection of the foreign organ and cells. This immune reaction against foreign tissue is called “alloimmunity.”

• In type 1 diabetes, this task is especially difficult, as doctors must fight against autoimmune attacks against the beta cells as well.
• Immunosuppressants—drugs that limit the effectiveness of the immune system body-wide—are given to patients to prevent rejection.
• In the past, steroids were also required, but the immunosuppressants have improved to the point that long-term steroids are no longer necessary. This is a great gain for transplant patients.
• With these techniques, immune rejection rates of transplanted pancreases are down to 10 – 15%.

Is it worth all this difficulty? Does transplanting pancreases work?

• Data shows that retinopathy stabilizes in 70% of patients, and neuropathy improves over one to two years. Damage already done cannot be reversed by the transplant, but a lot of further damage is prevented.
• Anecdotally, Stock reports that every pancreas transplant recipient he has spoken to would do it again if required.

Whole-pancreas transplant, though, is difficult and requires too many whole pancreases. Is there a quicker and easier solution?

• Islet cell transplantation is becoming more and more popular as an alternative to whole-pancreas transplant.
• In a single, same-day procedure, isolated islets are infused into the liver, where they take up residence and begin to produce insulin.

However, islet cell transplant has shortcomings as well.

• As with whole-pancreas transplant, alloimmune and autoimmune reactions are still a problem.
• The islets themselves are very sensitive to the immunosuppressants that are administered, and finding an immunosuppressant regimen that is protective but not toxic can be tricky.
• We have seen a decay in islet function over time after infusion, though it is not entirely clear why. Perhaps the lack of the pancreatic milieu, with whatever precursor and support cells it contains, leads to slow loss of the islets.

We’re getting better, though. Recent studies have shown that better immunosuppressants delivered earlier keep islets functioning for longer, and islet cell transplant is beginning to match whole-pancreas transplant in terms of longevity of function.

A JDRF trial recently showed that islet cell transplants led to insulin-independence in eight out of the ten recipients for four years now. The patients showed normal HbA1cs and no liver toxicity.

For now, then, Stock recommends that patients with kidney failure or life-threatening diabetes receive islet cell transplants, unless the patient has a “large body habitus,” in which case he should receive a whole-pancreas transplant, as that will have greater capacity to produce insulin.

Daniel Anderson: New Materials for Medicine

Dr. Daniel Anderson is a molecular geneticist and materials scientist at the Massachusetts Institute of Technology. He studies new materials and nanoparticles for a variety of medical applications, including islet cell transplant and glucose-sensitive drug delivery.

Dr. Anderson talked about some of the exciting new ways people are looking at beta cell replacement using new materials and tissue engineering:

Tissue engineering is the science of using living cells and natural or man-made materials to build tissues and even whole organs that can then be incorporated into living organisms.

The classic example is the Vacanti mouse, which has a functional human ear built from cartilage on a biodegradable polymer scaffold.

How can we use tissue engineering to build a pancreas? There are three major design criteria for the engineered pancreas:

• It must be able to eat and breathe—that is, it will need access to blood vessels that supply nutrients.
• It must do its job, which is to stay alive, sense glucose, and secrete hormones like insulin.
• It must be protected from the immune system and any alloimmune or autoimmune attacks.

Many attempts have been made to address these three criteria. The first is particularly difficult, as vascularization of tissue is an unsolved problem in tissue engineering right now.

• Any organs that are too thick don’t get properly vascularized, and therefore don’t get enough food and oxygen.
• One option is to create many tiny, little pancreases instead of one big pancreas.

Another challenge is protecting the tissue from the immune system. One approach that has seen some success is to encase the cells in alginate, a long sugar made from seaweed with a consistency similar to jello.

As early as 1980, these two ideas were put together, with scientists successfully implanting alginate microspheres filled with beta cells into diabetic rats.

• The rats were able to produce insulin and achieve normoglycemia for a few weeks, but then the microspheres stopped working.
• The problem was that the immune system, though unable to get through the alginate to attack the cells, recognized the foreign microspheres and encased the entire alginate ball with fibrous tissue, much like scar tissue. This blocking off of the microspheres, called fibrosis, starved the cells contained inside.

Where do we go from here? Scientists are experimenting with new tissues that can interact with the body in new ways. One promising example is that of memory materials, which can take certain pre-defined shapes when introduced to certain conditions, like body-temperature.

Bernhard Hering: Is There a Pig in Your Future?

Dr. Bernhard Hering is a professor of surgery at the University of Minnesota, and the Scientific Director at the Schulze Diabetes Institute there. His research focuses on islet cell transplantation, and on finding new sources of islets for patients.

Dr. Hering spoke about another option for beta cell replacement—xenotransplantation from pigs:

Hering proceeded with a fair sense of humor, beginning with a Winston Churchill quote: “I am fond of pigs. Dogs look up to us. Cats look down on us. Pigs treat us as equals.”

Hering does not limit the possibility of transplantation to extreme cases. People who should actively care about transplantation include those who have had type 1 diabetes for more than five years, who experience acute complications like hypoglycemia and insulin reactions, and who are beginning to have long-term complications.

Human islet transplantation, Hering notes, is a mature technology, with some patients with transplanted pancreases still functioning more than ten years later.

However, there is a limit to the number of human islets available, and stem cells haven’t gotten us all the way there yet.

The other option: pig islets.

• Pre-clinical studies in monkeys have shown promising results. Diabetic monkeys were given transplanted pig islets and a clinically relevant immunosuppression protocol.
• Five different groups have now shown long-term diabetes reversal in monkeys.

The issue is not just lack of human islets, though; as Hering puts it, “Pig islets are not just the ugly cousin of human islets,” as pig islets offer some unique benefits:

• Instead of relying on organ donors of questionable health and age, medical-grade pig islets come from young, healthy pigs with no comorbidities.
• Pig tissues have been used in humans for years, and have an established track record in the rebuilding of heart valves, hernia repairs, and other operations.
• We could raise an unlimited pool of pigs, meaning there is no wait time for new islets.
• Human islets are known to form amyloid plaques, which are tangles of protein fragments like those that appear in the brain in Alzheimer’s Disease. Pig islets do not form amyloid plaques.
• Pig islets are less susceptible to autoimmune attack, as the cells don’t appear to the immune system exactly as human beta cells do.
• Pigs can be genetically engineered to have islets that produce both insulin and immunosuppressants, meaning the islets could release immunosuppressants locally and potentially not require whole-body immunosuppression.
• Recipients could be vaccinated against pig antigens prior to transplantation, conferring additional protection from immune rejection.

With so much in favor of pig islets, Hering looks forward to the progress being made by the World Health Organization and others on the regulation front.

Hering notes, as a final illustration, that the nearest bridge to a cure is pig islets, and that bridge is almost completed.

Pamela Itkin-Ansari: Islet Cell Transplants—A New Approach

Dr. Pamela Itkin-Ansari is a professor at the Sanford Burnham Medical Research Institute in San Diego. She focuses on the signaling systems that regulate pancreatic development and pathogenesis, and on developing therapies that don’t require immunosuppression based on these systems.

Dr. Itkin-Ansari addressed the other side of the equation that Dr. Sander and others had mentioned—how do we protect transplanted islet cells, derived from stem cells or otherwise, from the immune system?

Itkin-Ansari points out that despite the success of whole-organ and islet cell transplants previously presented, we don’t do transplants in children. Why? The immunosuppressant regimen is dangerous and burdensome. Clearly, we need a better way.

A better way is indeed possible, though, because of two amazing things about islets:

• As many of the earlier speakers mentioned, the total mass of the islet cells in a normal pancreas is very small. We need less than an ounce of materials, which makes the prospect of transplant more feasible.
• Additionally, islet cells don’t need the whole pancreas to work and survive, further limiting the burden of transplantation.

Given that we’re talking about a few precious cells, would it be possible to transplant the cells in such a way as to be simply hidden from the immune system? Possibly—that’s where encapsulation technology comes in.

Any system that enclosed the islets would have to do several things: let glucose and nutrients in; let hormones out; keep immune cells at bay; and do this all quickly enough to allow for physiological insulin response.

Itkin-Ansari, in conjunction with ViaCyte, has been working to develop and test an encapsulation device made from a material similar to Gore-Tex and Teflon called polytetrafluoroethylene.

• These devices have been tested in a number of animal models, with promising results.
• Thus far, it appears that the device is not subject to alloimmune rejection or much autoimmune rejection—stains of the cells inside after removal show that the cells are healthy and producing insulin.

One big question is what kind of cells to put inside the envelopes.

• Mature beta cells are fragile cells, and trying to isolate them and put them in the device leads to lots of loss.
• Making beta cells from stem cells, as Dr. Sander noted, is an ongoing process. Nonetheless, Itkin-Ansari and her team have shown that putting the beta cell progenitors into the encapsulation device is a promising option; studies are showing that these progenitors mature and begin to produce insulin within the device once implanted. Luciferase labeling of the cells shows that once inside the recipient, insulin production increases, but the number of cells inside doesn’t change to match, implying the cells are differentiating and learning to make the necessary insulin.

The plan now is to design encapsulation devices to hold enough cells to supply a human with insulin, while still allowing sufficient blood flow to the cells. Then the cells could be implanted in any of a variety of locations in the body.

Itkin-Ansari closed by affirming the continuing promise of these technologies: “These things are all possible. They are not the stuff of fantasy; they are the stuff of funding.”

Bruce Buckingham: The Bionic Diabetic

Dr. Bruce Buckingham is a much-loved professor and clinical endocrinologist at the Stanford School of Medicine. In addition to running a number of key studies in the development of the artificial pancreas, I saw Dr. Buckingham—completely of his own volition—start to help clean up tables after the conference. In a suit and everything!

Dr. Buckingham gave an overview of diabetes technology over the last century, bringing us up to the current advances in creating an artificial pancreas.

The first records we have of managing diabetes come from about 1700 BC, when Egyptian doctors tasted the urine of patients to see if it was sweet.

In the early 1900s, urine strips were developed and used for diagnosis of diabetes.

It wasn’t until 1924, with the isolation and manufacture of insulin, that diagnosis led to a chance of survival.

Insulin, however, was not a cure. “Insulin turned out to have a very narrow therapeutic margin,” Buckingham noted.

So, better and better ways of deciding on and dosing insulin had to be developed. These included the first insulin pumps, behemoths, and have since progressed to today’s relatively svelte insulin pump options.

Even more recently, continuous glucose monitors (CGMs) have been developed and popularized.

The trick now is to tie the glucose measurements in with the insulin pumps to create a glucose-responsive insulin dosing system that imitates the pancreas. Many teams are working on algorithms that would control this, but there are a number of complications:

• Modern sensors have a 10 – 15 minute delay as compared to direct blood glucose measurements.
• Modern insulins also have at least a 15 minute delay.
• Glucose sensors are sometimes inaccurate.
• Insulin absorption rates vary according to person, activity, time of day, and so on.

Even so, we are making progress. The first iteration of a thinking insulin pump is designed to prevent severe hypoglycemia, especially at night.

• Even with CGMs, many diabetics become hypoglycemic at night without waking up. 71% of nocturnal CGM alarms go unnoticed by sleeping wearers.
• This tragically can lead to what is called “Dead in Bed Syndrome,” in which a diabetic goes very low at night but does not know or respond, ultimately dying in his or her sleep.
• One preventative measure would be to have the insulin pump itself respond to the ignored CGM alarm, turning off insulin delivery until the CGM shows the user’s blood sugar recovering. This technology, called “low-glucose suspend,” is available right now in Europe in the Minimed Veo pump. However, the FDA has not yet approved this functionality for use in the US.

The second iteration will be to allow the pump to increase and decrease insulin delivery to help keep patients in range.

• Such devices have been tested in clinical trial settings with recent-onset diabetics. Patients were kept on the self-adjusting insulin pump for up to four days, and showed good glucose control during that time.
• Even two years later, patients still showed improved glucose control on normal pumps at home. In theory, the initial period of better control helps protect and maintain beta-cells for longer.

There are many remaining challenges, though, including how to make the device usable in a home setting.

• Patients want a single device, not a hospital-room worth of gadgets.
• Integration between the various inputs will need to be coordinated, ideally through something like a smartphone, that many patients already have.

David Gough: Implantable, Long-Term Glucose Sensors

Dr. David Gough is a professor of bioengineering at the University of California, San Diego, and the co-fouder of GlySens, Inc. His research focuses on biosensors, especially for glucose sensing, and he is a self-proclaimed “card-carrying engineer,” which wins him points in my book.

Dr. Gough addressed the development and engineering of one version of that key component in the artificial pancreas Dr. Buckingham spoke about—the continuous glucose monitor.

GlySens, the company Gough co-founded, aims to develop and bring to market a long-term, implantable glucose sensor.

The goal of having an implantable glucose-sensor is by no means a new one; Gough showed images he was given in the early 1970s with illustrations of how such a sensor should function.

However, such glucose sensors have not been successfully produced yet. Instead, we have a set of other types of glucose sensors that Gough classes as:

• Those that work, but are non-optimal. These include the subcutaneously injected, short-term sensors like the DexCom. These have to be calibrated frequently, vary in accuracy, and have to be changed often.
• Those that don’t work, such as optical-based sensors. Many have been tried, but thus far, none have been sufficiently accurate or durable to be usable.
• Those that measure external fluids like tears or urine. These generally work, but provide a more qualitative measure, and have not achieved the specificity of milligrams per deciliter required for regular use.
• Sensors that measure secondary effects of blood glucose variation, like other chemicals, heart rate, and so on. These, like the optical sensors, suffer from too much noise in the signal, and have not been accurate enough to date.

Gough and his team developed a long-term, implantable glucose sensor that measures glucose in relation to oxygen levels in tissue.

These sensors have been extensively tested in pigs.

• Why pigs? Pigs have tissue similar to humans, in which the skin is tightly coupled to the tissues underneath, whereas mice, dogs, and the like have skin that is much more loosely attached.
• Humorously echoing Dr. Hering’s praise of pigs, Gough noted that pigs are very lovable and affectionate animals.

In the animal trials, the sensors worked for up to eighteen months, tracking glucose accurately in both non-diabetic and diabetic pigs.

• The implantation procedure did not cause an tissue damage.
• On average, there was about a 9 minute delay in the glucose reading as compared to blood glucose measurements.
• The device now has a collective total of 31 device years in all the pigs they tested.

The device holds lots of promise for consumer use:

• The device is inserted under the skin via a simple outpatient procedure.
• The battery lasts for at least two years.
• The device does not require frequent calibration.

Small human trials are currently underway, with GlySens hoping to have the device ready in several years time.

Alberto Hayek: Closing Remarks

Dr. Alberto Hayek is a professor and pediatric endocrinologist at the University of California, San Diego, and also the co-director of the Pediatric Diabetes Research Center at the university. His research is focused on creating beta cells from stem cells in culture.

Dr. Hayek reviewed the day’s talks and closed the conference, with endearing enthusiasm and camaraderie. As one of the key organizers of the conference, he put together the list of speakers, and expressed much gratitude to all of the speakers for coming, and even paying their own ways.

Dr. Hayek reflected that the talks showcased the duality of this disease, characterized by both the immune system and the endocrine system, as well as the duality of the speakers, focused on both the coming cures and what we can do today.

Dr. Hayek’s research focuses on turning stem cells to beta cells, and he recognized the technical difficulties still being faced by his field. He compared the struggles with stem cells to trying to talk to his nine-month-old grandson; his grandson babbles and babbles, trying to tell Dr. Hayek something, but all Dr. Hayek can do is babble back until he understands more. Similarly, Dr. Hayek knows his stem cells are trying to tell him something, and so he responds in kind, trying to understand more.

Ultimately, Dr. Hayek noted, “the message is that it’s going to take time and patience,” but that we are progressing, and learning a lot along the way.

Karmel Allison is ASweetLife’s science editor and a regular contributor.  She writes the blog Where is My Robot Pancreas?