For decades, treatments for and clinical trials involving type 1 diabetes have been based on the understanding that the disease’s autoimmune process rapidly curtails pancreatic β-cell function, as measured by C-peptide levels. However, research involving a new, ultrasensitive C-peptide assay is providing fresh insight into the course of diabetes. The study findings are covered in this issue of Strategies.
The conventional understanding of the pathophysiology of type 1 diabetes is that pancreatic β-cell function drops off dramatically and soon—within a few years—after diagnosis. This well-described process can be tracked through declines in C-peptide values over the same period, C-peptide being the protein clipped when A- and B-chain proteins of proinsulin synthesized by pancreatic β-cells form insulin. Rapid β-cell death and the associated drop in insulin production have led to what researcher Denise Faustman, MD, PhD, has called therapeutic nihilism. “The thought has been that the pancreas is not functioning after about one-to-two years, so our treatment approach has been kind of fatalistic. ‘Oh, your pancreas is dead, we don’t have anything to save the remaining pancreas’,” she explained. Faustman is director of immunobiology at Massachusetts General Hospital and an associate professor of medicine at Harvard Medical School in Boston.
However, diabetologists also have recognized that some residual β-cell function remains after diagnosis and that early, aggressive treatment with insulin or immunosuppressive drugs can slow the autoimmune destruction of β cells, enabling patients to go a little longer before they need full doses of insulin routinely. Equally well-known is that some newly diagnosed diabetics run quickly into complications of the disease, while others fare much better clinically for years. Often this has been attributed to the latter managing their disease better, but new evidence is casting a different light on this. Faustman and her colleagues recently used a new ultrasensitive C-peptide immunoassay to measure serum C-peptide levels in long-term type 1 diabetics (Diabetes Care 2012;35:465-70). To their surprise, the subjects not only had detectible C-peptide levels—including 10% of those who had had diabetes 3-4 decades—but also functioning β-cells. “We found using a new ultrasensitive assay that the time course of the disease is totally altered. Even at 30 years out, 20 percent still have a pancreas making some insulin, as assayed by C-peptide,” said Faustman. “It’s a whole new way to view the disease as it relates to the pancreatic function two years out after diagnosis. It’s all come about because of this ultrasensitive assay, which has altered understanding of the kinetics of this disease in a good way.”
The study involved two components, including 182 type 1 diabetics, who were a median of 39 years, had had the disease a median of 15 years, and were a median of age 13 at disease onset. C-peptide values for all these subjects were measured first with an ultrasensitive C-peptide assay, and only those with values higher than the detection range of the ultrasensitive assay also were measured using a standard assay. The second part of the study involved four diabetics who provided serum samples weekly for up to 20 consecutive weeks. C-peptide values in these samples were measured using a different standard C-peptide assay but with the same ultrasensitive assay that had been used to measure samples from the 182 patients. The standard C-peptide assays have lower limits of detection of 15 pmol/L and 33.1 pmol/L, respectively, whereas the ultrasensitive assay has a lower limit of detection of 1.5 pmol/L with inter- and intra-assay coefficients of variation of 5.5 and 3.8% at 37 pmol/L.
Researchers categorized the 182 samples into six groups based on the patients’ years of disease duration and found declining levels of C-peptide detection across the groups; however, the decline was gradual over decades, rather than a few years. Nearly 80% of samples in the 0-5 years’ duration of disease group had C-peptide levels above the ultrasensitive assay’s limit of detection, 230 pmol/L, versus 10% from patients who had had diabetes for 31-40 years. C-peptide was not detected in subjects who had had diabetes 31- 40 years. In the group of four subjects who provided a total of 54 serial samples, the standard C-peptide assay did not detect C-peptide in any of the samples, whereas the ultrasensitive assay detected C-peptide in 63%.
The researchers also assessed subjects’ β-cell functional capacity by evaluating fasting and/or non-fasting glucose levels in both the group of four patients who provided serial samples and in samples from the 182 patients. In both groups, samples from hyperglycemic subjects with glucose values >150 mg/dL had significantly higher C-peptide levels than those from normoglycemic subjects.
These findings have important clinical ramifications, according to Daniel Stein, MD, who was not involved in the study. “The implications of this study are two-fold. One is that people even 30 years out from the diagnosis of diabetes may have minimal C-peptide levels that are not detectable by the normal assays but when an ultrasensitive C-peptide assay is brought to bear they have measurable levels,” he observed. “Perhaps just as important, is the finding that C-peptide responds in a biologically appropriate way to stimulation of the beta cells to secrete insulin. These long-standing diabetics don’t have a lot of beta cells. They have a few of them. But that’s a very important finding because the clinical trial data suggests both in terms of intervention studies of immunosuppressants or in conventional therapy with insulin, patients maintain better glycemic control if they have some residual beta cell function. These patients also have a lower risk of complications.” Stein is professor of medicine and scientific director of the Einstein/Montefiore Institute for Clinical and Translational Research at Albert Einstein College of Medicine in New York City.
Faustman agreed that the findings could reshape diabetes-related research. “The standard C-peptide assay drove every clinical trial protocol in this field for the past 25 years because everybody thought that if were you’re going to do an immune intervention trial in type 1 diabetics you had to get the kids within a week, or a month of diagnosis or do very complicated things and try to identify them in populations before they even get a high blood sugar. That’s because the drugs being developed weren’t stopping the disease, they were slowing the disease. So that meant you had to have C-peptide present to slow the destruction of the pancreas,” she explained. “This new assay will open up the field tremendously to say people who have had the disease two to even 40 years shouldn’t be rejected from these kinds of trials and we should be doing more clinically for these people.”
Faustman and Stein both predicted that clinicians would be eager to use the ultrasensitive assay in clinical practice, which is available in the U.S. as a kit from the manufacturer, Mercodia. “There’s value in knowing whether a patient has C-peptide positivity even at low levels detected by an ultrasensitive assay,” said Stein. “These patients may have a small amount of residual beta cell function, may respond better, and potentially have a better prognosis in terms of their ability to maintain good glycemic control without an insulin pump. They might do better on intensive insulin therapy, such as four times per day insulin injections or an insulin pump, compared to someone who truly has no detectible C-peptide.”
Faustman urged laboratorians to stay abreast of developments involving the ultrasensitive assay. A spokesperson for Mercodia indicated that the company may seek additional Food and Drug Administration clearance for the assay as a marker of β-cell decay, based on findings from this study. Stein’s lab also is developing a mass spectrometry-based method to measure very low levels of the analyte.
Faustman predicted ultrasensitive C-peptide measurement would fit a particular niche in diabetology. “For an endocrinologist evaluating a patient with a suspected tumor that’s making too much insulin from the tumor, the standard assays with range of detection from 40 to 660 pmol/L are totally appropriate,” she said. “However, I would think clinicians following people with type 1 diabetes would want to pair results from this ultrasensitive assay with HbA1c measurements to discern whether a patient is struggling with compliance or if their pancreatic function is shutting down.”
Join us for a discussion of this article, and interact with your peers, on LinkedIn