Shortly after the launch of Solving Diabetes we received a number of penetrating questions from “J P Marat”. With his permission the resulting dialogue with Scott King has been edited for inclusion in Straight Talk. “J P Marat” is the pen name of a diabetes researcher in Canada. The dialogue has been formatted to enable further dialogue from readers.
Encapsulations will certainly never be cost-effective if they cost $100,000 each, as LCT estimates, if they have to be replaced every two years, and if they are not 100% effective even while they do work.
Is a price of $100,000 cost effective?
The Marat/King Dialogues
By J P Marat - July 9th, 2009
Cost of the device(s), cost and morbidity from surgery, and longevity of implant function form a fairly comprehensible set of takeoffs. A commercially viable product will need to work at least 18 months. We hope it will work a lot longer. If the market is improvement of conventional islet transplantation, and the device requires no chronic drugs, then the drugs cost saving are more than $100,000 annually so the price for the device can be comparable and still make economic sense. The price will be lower than $100,000 annually when production is scaled up I think.
The question of how long an islet can maintain its functionality in a capsule is important and not yet known. I think your suggestion “encapsulated beta cells will always have a life expectancy so short that they will simply not be cost-effective” is a rather bleak reading of the evidence. For instance we know that the encapsulation systems in clinical studies in Russia (and soon New Zealand) and Brussels seem to have their peak efficacy at six months. Our hope rests on sporadic reports in the literature of same capsules working for years, including Randy’s own microcapsule system (Transplantation Proceedings 27:6 (1995) pp3297-3301). We believe that the Islet Sheet will function at least eighteen months, and perhaps significantly longer.
Yes, the cost of $100,000 for LCT therapy might seem high, but it will come down over time. Remember LCT (and for that matter Cerco Medical) are funded by investors and they want a profit – and deserve one for making a risk investment. You are right that the cost of pig tissue is the biggest single item of direct costs for the LCT product (based on my own knowledge – I am not privy to LCT’s numbers). But if we end up selling a pig Islet Sheet it will probably be priced at a similar level. If the pig Islet Sheet needs to be replaced at 18-month intervals, then the total cost will be lower than that of conventional islet transplantation.
Over time the stem cell research will make safe and effective ‘islets’ relatively cheap. In principal these cells can be much less expensive than pig islets because culture methods can be scaled fairly easily. The cost most likely will be governed by the cost of materials needed for stem cell culture and differentiation. And regulatory requirements might jack up the expense.
At Cerco we are going to prove that the Islet Sheet works in humans using human islets, the sort already approved for clinical studies. Then we will tackle the supply issue. The volume of islets needed to treat every type 1 diabetic on earth is several tons per year. A huge investment will be required.
Continuing with the theme of the cost-effectiveness of LCT’s islet encapsulation treatment, I wonder what you think of its value in comparison with current stem cell treatments for type 1 and type 2 diabetics? From the results published so far, it appears that most patients with an islet cell encapsulation can expect to obtain only reductions in insulin requirements, although a few are totally free of exogenous insulin for a while. But compare islet encapsulation with stem cell treatment:
Expected duration of islet encapsulation’s insulin requirement reduction: 18 months
Present maximum duration of insulin independence by stem cell treatment: 4 years
Typical exogenous insulin reduction from islet encapsulation: ca. 40%
Exogenous insulin reduction from stem cell treatment: 13% — no effect; 43% — 25% to 50% reduction; 43% — 50% reduction or more.
Cost of islet encapsulation: $100,000
Cost of autologous stem cell treatment: $10,500 to $12,400
Earliest date islet encapsulation will be clinically available: 18 months (LCT’s estimate)
Earliest date stem cell treatment will be clinically available: Available in Europe since 2006
Possible complications of islet encapsulation: peritonitis, subclinical immune response leading to background inflammation and cytokine release, cancer
Possible complications of adult autologous stem cell treatment: Embryonic and allogenic stem cell implants are known to be carcinogenic, but there is only a theoretical risk that adult autologous stem cell implants are.
Given that all medical research estimates, especially those made by researchers concerning their own work, are overly optimistic, it may well be many years before islet encapsulation will be available, in contrast to the option of stem cell implants, which are available today.
Sources: LCT website; X-cell.com website (Duesseldorf and Cologne stem cell clinic); C. Court, et al, “C-Peptide Levels and Insulin Independence following Autologous Stem Cell Transplant,” Journal of the American Medical Association, 301(15) 1599 (2009); J. Voltarelli, et al, “Autologous Hemoatopoietic Stem Cell Transplantation for Type 1 Diabetes,” Annals of the New York Academy of Medicine, 1150 (2008); R. Elliott, et al, “Live Encapsulated Porcine Islets for a Type 1 Diabetic Patient: 9.5 Years after Xenotransplantation,” Xenotransplantation, 14(2) 157 (2007).
Couri (JAMA 2009) is for newly diagnosed diabetics (never having experienced ketoacidosis) and is essentially a way to extend the honeymoon; there is no evidence it will work on the vast majority of people with T1D with fully developed disease. Although I personally regard the side effects of this protocol to be unacceptable, I agree that there is a reasonable case for it (if you are lucky enough to be diagnosed before ketoacidosis).
We believe that encapsuled islets can eliminate the need for insulin injections. We are working now to prove it.
Given this history of encapsulated islet transplantation and the predictions made by practitioners, your point on likely delays is well taken.
My main concern about encapsulated porcine islet cells as a method to improve the medical fate of type 1 diabetics — and indeed about any intervention focused on blood glucose control — is that there is now increasing evidence that the complications of diabetes may be caused, at least in part, by autoimmune and/or genetic processes which would remain unaffected by even the most rigorous control of blood glucose levels.
Dr. Faustman at Harvard has demonstrated that diabetic autoimmune disease persists throughout the life of the patient, long after the initial attack on the pancreatic beta cells. Most autoimmune diseases, such as lupus and vasculitis, attack a whole range of targets, so autoimmune type 1 diabetes would be unusual if it damaged only the beta cells of the pancreas and nothing else. It seems reasonable to suspect that diabetic autoimmunity also causes damage beyond the pancreas, since type 1 diabetics have a higher than normal incidence of other diseases caused by autoimmunity, such as multiple sclerosis and thyroiditis. Intuitively, it seems strange that lifelong diabetic autoimmunity should cause all of its associated injury only as a side-effect its initial attack on the pancreatic beta cells — which manifests as non-physiologic blood sugar levels consequent on the lack of normal insulin output — but do no harm as a result of an autoimmune disease which was initally so lethal to beta cells.
Recently some empirical evidence has emerged to support this logical inference. Autoimmunity has now been implicated as a cause of diabetic neuropathy (V. Granberg, et al, “Autoantibodies to Autonomic Nerves Associated with Cardiac and Peripheral Autonomic Neuropathy,” Diabetes Care, 28(8) 1959 (2005)), diabetic nephropathy (K. Ichinose, et al, “Recent Advancement of Understanding of Pathogenesis of Type 1 Diabetes and its Relevance to Diabetic Nephropathy,” American Journal of Nephropathy, 27(6) 554 (2007)), and diabetic retinopathy (S. Kastellan, et al, “Could Diabetic Retinopathy be an Autoimmune Disease?” Medical Hypotheses, 68(5) 1016 (2007)). One opthalmological researcher is now so convinced that diabetic retinopathy is an autoimmune disease that he even recommends immunological intervention as the primary treatment for it (D. Adams, “Autoimmune Destruction of Pericytes as the Cause of Diabetic Retinopathy,” Clinical Ophthalmology, 2(2) 295 (2008)).
Similarly raising concerns for the theory that hyperglycemia is the essential cause of diabetic complications is the research indicating a role for genetic factors in the etiology of diabetic complications. Genetic factors have been suggested as the cause of diabetic nephropathy (S. Rich, “Genetics of Diabetes and Its Complications,” Journal of the American Society of Nephrology, 17, 353 (2006)), and the distinct similarities in the natural history of diabetic complications within the same family over the years suggest that genetics rather than blood sugar control explains these patterns (M. Monti, et al, “Familial Risk Factors for Microvascular Complications,” Journal of Clinical Endocrinology and Metabolism, 92(12) 4650 (2007)). Mild ‘diabetic’ neuropathy can even be found in the non-diabetic offspring of type 2 diabetics (C. Haverslev Foss, et al, “Autonomic Neuropathy in Nondiabetic Offspring of Type 2 Diabetic Subjects,” Diabetes, 50(3) 630 (2001)), while evidence of diabetic vascular disease can be found in the children of type 2 diabetics (C. Giannattasio, et al, “Increased Arterial Stiffness in Normoglycemic, Normotensive Offspring of Type 2 Diabetic Patients, Hypertension, 51(2) 182 (2008)), just as elevated levels of enzymes associated with diabetic nephropathy are found in the first-degree non-diabetic relatives of diabetics with renal disease (C. Bau and S. Twigg, “Fibrosis in Diabetes Complications,” Vascular Health Risk Management, 4(3) 515 (2008)).
When you think of Gary Pittenger’s impressive experiment showing that normoglycemic diabetic serum from diabetics is toxic to nerve cells (G. Pittenger, “The Toxic Effects of Serum from Patients with Type 1 Diabetes Mellitus,” Diabetes Medicine, 10(10) 925 (1993)), you have to worry that a porcine pancreatic implant providing the patient with normoglycemia might just not be enough — or not enough to be worth $100,000 every two years.
The case that islet transplantation prevents diabetic vascular disease is extremely strong. One should be careful in extrapolating clinical meaning from laboratory observations.
For instance the Pittenger/Vinik paper you cited in the last paragraph last might well be a laboratory artifact. “The possibility that humoral neurotoxic factors contribute as a cause of diabetic neuropathy was tested by application of serum from patients with Type 1 and Type 2 diabetes to mouse neuroblastoma cells…” When your result comes from a tumor cell from a foreign species, the result is interesting but proves almost nothing. It does justify a study to see if this effect can be demonstrated in humans.
On the other hand, the relationship between diabetic blood sugars and diabetic vascular disease is established by enormous clinical experience, most famously DCCT, and also the experience of diabetics pursuing tight control, including me. Further, clinical experience with islet transplantation strongly supports the concept that the improved blood sugar reduces vascular problems and can even partially reverse them.
Your basic point that autoimmune disease might not be limited to islet damage is sound and deserves scrutiny. However, I don’t think we should lose sight of the fact that the clinical evidence supports the concept that the great majority of diabetes morbidity is mediated by diabetic blood sugars.
I look forward to a new era of diabetes research, after islet transplantation is routine, when we are pursing the remaining, ‘direct’ autoimmune effects.
Prior to the DCCT, there was always speculation that even if a statistical correlation between hyperglycemia and diabetic complications could be securely demonstrated, the possibility would still remain that this correlation might only be the marker for some underlying cause distinct from hyperglycemia which was actually causing the complications, but which was itself measured by the degree of hyperglycemia. After the DCCT, this logical possibility was generally forgotten, although now Dr. Duncan Adams in New Zealand (in the article cited in my previous post) has gone so far as to recommend that tight glucose control be abandoned as the preferred route to prevent diabetic retinopathy, and that immunosuppressive interventions be adopted in its place. While the DCCT does show a general correlation between lower blood sugar levels and lower complication risk on a population basis, many individuals in the study had rapid development of severe complications despite excellent blood glucose levels, while others suffered little or no worsening of complications despite poor blood sugar control. This is consistent with other research, which shows that 40% of patients suffering from diabetic neuropathy have excellent blood glucose levels (M. Centofani, “Diabetes Complications: More Than Just Sugar?” Science News, 149 (26/27) 421 (1995)), and that type 1 diabetics surviving 50 years or more with the disease generally have very mild complications despite a huge burden of hyperglycemia-years, but also have, on average, an extremely high HbA1c of over 10% (G. Gill, et al, “Insulin-Dependent Diabetes of Over 50 Years’ Duration,” Practical Diabetes International, 10(2) 60 (2005)).
Also, an important limitation of the DCCT study was that patients who found strict blood sugar control too difficult to achieve dropped out of the group attempting it, so those remaining in the strict control group constituted a self-selecting set of patients with easier blood sugar control. Was this because they already had a less severe form of type 1 diabetes, perhaps with a less virulent autoimmunity and thus more surviving beta cells to provide a buffer against blood sugar swings from exogenous insulin? It is well-established that patients with more residual insulin function can achieve better blood glucose control than those who have none (C. Steele, et al, “Insulin Secretion in Type 1 Diabetes,” Diabetes, 53, 426 (2004)). So were these patients doing better because they had a milder case of diabetic autoimmune disease in the first place, and thus less autoimmune attack on other organs, or because they had better blood sugar control as a result of their autoimmunity having left them with more functioning beta cells? These patients would also have more c-peptide, which itself has been shown to reduce diabetic complications independently of blood sugar control, according to references I cited in an earlier post.
Historical epidemiology also shows some puzzling results for the theory that hyperglycemia is the overriding cause of diabetic complications. For example, although glucose control was difficult before home glucose meters became available in the 1980s, and therapeutic regimens did not even recommend tight glucose control before the 1960s, the difference between the diabetic complication outcomes obtained long ago and now — a generation after the initiation of strict blood sugar control regimens everywhere — is disappointingly small. Thus while 50% of cases of type 1 diabetes resulted in diabetic nephropathy before 1950, today the data show that about 30% of cases still have this result (J. Ekoe, et al, ‘The Epidemiology of Diabetes Mellitus,’ London: John Wiley, 2001, 341), even though ACE-inhibitor therapy, introduced for diabetic renal disease in the 1990s, is estimated to have reduced the risk of diabetic nephropathy by 60% and so is probably responsible for much more of this decline than better blood sugar control (G. Viberti, “A Glycemic Threshold for Diabetic Complications?” New England Journal of Medicine, 332(19) 1293 (1995)). Similarly disappointing results from improved blood sugar control efforts prior to the introdution of home glucose meters were found for retinopathy and nephropathy (Elliot Joslin, et al, ‘Joslin’s Diabetes Mellitus,’ Philadelphia: Lippincott, 2004, 798; G. Pambianco, et al, “The 30-Year Natural History of Type 1 Diabetes Complications,” Diabetes, 55(5) 1463 (2006)). One epidemiological study has concluded that there is in fact little relation between glycemic control and coronary disease and lower extremity arterial disease in diabetics (C. Lloyd, et al, “Incidence of Complications of IDDM,” American Journal of Epidemiology, 143(5) 431 (1996)).
What all of this suggests is that when performing a cost-benefit analysis of encapsulated porcine islet cells as a new treatment for type 1 diabetes, the estimated benefit has to be multiplied by some value, less than 1.0, to take account of the extent to which glycemic control may not prevent all of the complication-causing factors in type 1 diabetes. We all hope that this discounting factor will prove small, but for the moment we do not know its magnitude. Ironically, the answer may only become evident because of encapsulated islet therapy.
Are crimes caused by police?
JP Marat’s several posts on immunology and type 1 diabetes have led me to reflect on what we know and don’t know about the matter, and immunology generally. We all think we understand the brain, because to all think, but we really don’t understand how it works and mental illness remains mysterious. The immune system, which is as complicated as the nervous system, has an obscure function and vastly complicated structure and actions.
You can’t locate the immune system because it is a fluid organ present everywhere. Its name comes from its most elementary function: everyone knows that if you recover from an infectious disease (chicken pox) you will never suffer from it again: you are immune. Once it became possible to identify infectious agents it became possible to analyze how the immune system works. Over time other effects of the immune system appeared. Tissue implants (organs or blood) are usually rejected. Cancer may be prevented and reversed by the immune system. Tissue healing includes immune cells. And sometimes the immune system turns on the body in a sort of biological ‘friendly fire;’ such diseases (including type 1 diabetes) are called ‘autoimmune.’
When cells started to group themselves into organisms the problem of policing ‘self’ from ‘non-self’ appeared. The immune system is the organ assigned the task. A bewildering array of cells is involved, but the basic logic is simple. During fetal development the immune system memorizes the components of self. Except for identical twins, every human being is genetically unique and has a unique set of self components never before seen in this world. Each proto-immune system is born naive and learns, affected by nutrition, infections, trauma, etc.
When an invader appears, scavenger cells eat it and present certain components to be evaluated by the immune memory. If the component was present in the fetus it is identified as self and is ignored; if the component is novel, it is identified as non-self and is attacked and destroyed.
Beyond this general scheme lies vast complexity of detail. Most of the immune system cells are white, so are called leukocytes (Greek for white cell). They travel by crawling through tissues and through a sort of subterranean sewer system called the lymphatic ducts where they are called lymphocytes. In fact one of the most confusing aspects of immunology is that the same cell has different names when found in different places. And analogous structures in mice and men have different names. Becoming an immunologist is joining a priesthood: they know you by how you talk.
The great strength (and also weakness) of contemporary biology is molecular biology. The component that identifies self and non-self is most often a small protein or peptide. It is presented by a cell surface protein called the antigen presenting protein, and the memory function implemented by a protein called the T cell receptor (the T comes from the thymus gland in the neck, the boot camp of leukocytes). Ask an immunologist about the interactions of T cell receptor, antigen presenting protein and peptide and you will get an exquisitely detailed answer at the molecular level. It is quite beautiful and is a fruit of molecular and structural biology. But it does not tell us much we really want to know if we want to cure human disease.
If our goal is to prevent or reverse immune disease we need to understand exactly what happens in the affected tissues. This is very hard to observe and we don’t have the tools to understand much of what is happening. I’ll use type 1 diabetes as an example.
If you believe type 1 diabetes is an autoimmune disease you are saying that the beta cells are destroyed by leukocytes that have (mis)identified the beta cells as non-self. Certainly the presence of activated leukocytes that appear to be killing the beta cells in newly diagnosed diabetic supports this concept, as does the fact that diabetics tend to have certain genes for key immune proteins. On the other hand, the islets are present when the immune memory is formed and thus islets should be remembered as ‘self’. So either the memory is faulty or the immune system has been persuaded to attack the beta cells. How does this happen? I have read countless proposals in the literature in the past 25 years and I have never seen one that persuaded me. As far as I am concerned the cause of autoimmunity in Type 1 diabetes is a near total mystery.
In fact you might call autoimmunology The Department of Mysteries. I once asked a senior Berkeley immunologist what causes autoimmune disease. His answer was “an autoimmune disease is the name we give something we don’t understand.”
Leukocytes are present when the beta cell fails. So they done it. But police are present after a murder. So the police done it? Maybe not. The presence of leukocytes proves nothing about causality.
JP Marat believes that a major component of diabetes complications is not caused by bad blood sugars but rather directly by autoimmune disease. Given our poor understanding of immunology generally, and autoimmunity particularly, I am skeptical, but concede it as a logical possibility. Given the more direct clinical evidence for bad blood sugars leading to vascular problems in diabetics I think that Islet Sheet is likely to clear up the majority of the problems.
We agree that a metabolic normalization brought about by the Islet Sheet (without drugs) will help answer this question.
I agree completely with your concerns about the defective character of today’s immological science. Just as 18th century physics went into a period of stagnation before the concepts of oxidation-reduction reactions and electromagnetic fields became available, and instead resorted to explaining phenomena by a multiplicity of imaginary subtle fluids such as phlogiston and the luminiferous aether, so too modern immology is forced to rely on a bewildering array of special cell types to explain things in order to compensate for the lack of a more general theory. It thus remains more a kind of natural history than a strict science.
Research by Dr. Faustman raises the interesting possibility that even the explanation of the cause of type 1 by autoimmune processes may still not have gotten to the root cause of what is happening with the pancreatic beta cells. Her work suggests that pancreatic beta cells, and perhaps other tissues as well, may exhibit neurological abnormalities long before they are attacked by the immune system. It may then be the disordered signals sent out from the abnormal nerves in these tissues which alert the immune system to identify them as foreign and attack them, rather than diabetes resulting from a primary misdirection of the immune system per se. See: A. Lonyai, et al, “Fetal Hox11 Expression Patterns Predict Defective Target Organs: A Novel Link Between Developmental Biology and Autoimmunity,” Immunology and Cell Biology, vol. 86, no. 4, p. 301 (2008).