A safe drug for increasing insulin sensitivity, improving lipid profi les, and reducing weight would be in instant demand for treating a host of patients, and many researchers believe the hormone fi broblast growth factor 21 (FGF21) could do all these things and more. What’s more, FGF21’s unusual binding properties could enable it to work in a targeted fashion that could minimize side eff ects. Several drug companies are exploring the preclinical possibilities, with impressive results in rodents and primates.
“FGF21 burst onto the scene in 2005 with this paper by Eli Lilly describing its really remarkable eff ects in obese rodents,” says Steven Kliewer, PhD, professor of molecular biology at the University of Texas Southwestern Medical Center.
In a landmark study in Th e Journal of Clinical Investigation, a team led by Alexei Kharitonenkov, PhD, group leader of diabetes research at Eli Lilly and Co., showed that, among its benefi ts, FGF21 could lower plasma glucose and triglycerides to near normal levels in diabetic mice. Th e eff ects lasted for 24 hours or more following the cessation of FGF21 administration, with no mitogenicity, hypoglycemia, or weight gain at any dose in diabetic, healthy, or transgenic mice.
Since that paper, a plethora of studies have found similar benefi cial eff ects in nonhuman primates, and the race has been on to explore its therapeutic potential to treat diabetes, obesity, and metabolic syndrome.
Longevity and Potency Issues
Th e research identifi ed FGF21 as an anomalous member of the FGF family. Most FGFs are true to the name—heparin-binding proteins that act locally in an autocrine or paracrine manner in growth and diff erentiation. Produced by the pancreas, liver, and adipose tissue, FGF21 is part of a three-member subfamily that lacks the heparin-binding domain, which allows them to diff use into the circulation. Rather than promoting growth, they act as endocrine hormones that regulate metabolism.
“FGF21 shows a strong preference for one of the four FGF receptors, FGFR1c,” Kliewer says. “It requires a second protein to function, called βKlotho, another membranespanning cell-surface protein.” FGF21 needs to bind to this receptor βKlotho complex to exert its actions. “This is probably why FGF21 doesn’t have the growth-promoting actions of other FGF family members,” Kliewer says. And it is probably why FGF21 is free of the proliferative and tumorigenic eff ects of most members of the FGF family.
FGF21 is not a natural when it comes to use as a pharmaceutical, however, because of issues with its potency and longevity. Although its eff ects can last for 24 hours in mice and nonhuman primates, it has a half-life of about 30 minutes in rodents and two hours in cynomolgus monkeys, and remains in human circulation for less than fi ve hours. Just the same, the group at Lilly, many collaborators, and other researchers have published abundant data using FGF21 and longer-lived analogs. “There are several ways you can modify the biological and biopharmaceutical properties of the protein FGF21,” Kharitonenkov told Endocrine News.
Monoclonal Antibody Solution
Other researchers have looked to sidestep these problems through a relatively new technology by developing a monoclonal antibody that mimics FGF21 action.
More than 30 monoclonal antibodies are currently used to treat cancer, as well as autoimmune, infl ammatory, and infectious diseases, but most are antagonists or inhibitory in action. The FGF21 mimetics are among the fi rst to use an antibody as an agonist to stimulate FGF21’s eff ects by activating its receptor complex.
“Antibodies may have several advantages,” says Yang Li, PhD, scientifi c director at the pharmaceutical company Amgen. “One potential advantage is their extended pharmacokinetic properties in plasma. Usually as a class, they can circulate for days in the plasma, so you can design a therapy that perhaps does not require frequent injections. Another advantage is that antibodies are quite specifi c for their targets, and this can usually lead to reduced potential side-eff ect concerns with your therapeutic molecule. And manufacturing processes for antibodies continue to improve as well.”
In December 2011, researchers at Genentech, led by molecular biologist Junichiro Sonoda, PhD, published fi ndings on a monoclonal antibody they named R1MAbs designed to mimic FGF21 by being an agonist of the FGFR1. “A single injection of R1MAb into obese diabetic mice induced acute and sustained amelioration of hyperglycemia, along with marked improvement in hyperinsulinemia, hyperlipidemia, and hepatosteatosis,” they wrote in Science Translational Medicine. Blood glucose concentrations in the treated mice normalized within a week, and remained lower than levels in placebo-treated mice for a month. The antibody acted on FGFR1 without involvement of βKlotho, and some of the results indicated that R1MAb was not as targeted as FGF21 itself.
A year later, Li’s group at Amgen upped the ante by developing mimAb1, an antibody that works with the FGFR1c/βKlotho complex, and testing it in monkeys. “In obese cynomolgus monkeys, injection of mimAb1 led to FGF21-like metabolic eff ects, including decreases in body weight, plasma insulin, triglycerides, and glucose during tolerance testing,” the researchers wrote in Science Translational Medicine. “Because mimAb1 depends on βKlotho to activate FGFR1c, it is not expected to induce side eff ects caused by activating FGFR1c alone.”
The eff ects of mimAb1 were closer than those of R1MAb to FGF21’s, but still not identical. And the eff ects were also very long lasting. Many of the benefi ts lasted at least fi ve weeks, and after a second injection, the reduction in body weight lasted for nine weeks.
Amgen’s Li credited the long-lasting eff ects primarily to the extended time the antibody remains in circulation, and Genentech’s Sonoda suggested that an additional contribution may come from the ability of the antibody to modify the underlying disease itself and not simply treat symptoms. For example, by removing excess lipids, it leaves the body in a healthier state.
Researchers will be busy for some time sorting out FGF21’s various roles and mechanisms of action. FGF21 acts on both liver and adipose tissue, but the antibody forms appear to bypass the liver and act directly on adipose tissues in achieving some of their antidiabetic eff ects.
FGF21 levels are higher in patients who are obese or have type 2 diabetes, and yet it plays a role in the starvation response. Th e liver increases FGF21 production in response to fasting, and FGF21 aff ects starvation adaptations such as suppressing the activity of growth hormone and reducing the production of insulin-like growth factor.
Kliewer and colleagues recently published in eLife a study that found that FGF21 overexpression in transgenic mice can lead to the same kind of lifetime extension that can be induced by severe caloric restriction. Th e study turned up some side eff ects, however, including infertility in female mice and smaller size and bone density loss in both sexes.
A drug that could work better than the out-of-favor thiazolidinediones—without dangerous side eff ects and with weight loss instead of weight gain—would certainly be remarkably valuable for treating diabetes and metabolic syndrome. But the history of the thiazolidinediones illustrates that any such drug will face a very high bar in proving more eff ective than current treatments without major side eff ects. And although FGF21 researchers are excited about its promise, with the published data currently in the preclinical stage, any new therapy remains years away.