Inhaled Insulin for Diabetes: Is Afrezza Worth the Risks?

Afrezza, an insulin product administered with a special inhaler, rather than requiring injection, has just been approved by the United States Food and Drug Administration for pre-meal administration to control blood glucose in people with both type 1 and type 2 diabetes. There has been interest in this approach to insulin administration for several decades, with clinical trials and new approaches being developed over the past ten years.

Inhaled insulin is clearly effective in the treatment of both type 1 and type 2 diabetes in a fashion similar to that of the rapid-acting subcutaneously injected insulin analogs, with some evidence that its peak effect occurs at or before 20 minutes after administration, as opposed to the peak action of rapidly acting insulin analogs after 30-40 minutes, and of human regular insulin at 60 to 120 minutes. An advantage, then, is that inhaled insulin may act even more quickly than insulin analogs, which could have the advantage of reducing the immediate post-meal spikes in blood glucose that occur after meals with a relatively high carbohydrate content. This might lead to less late post-meal hypoglycemia, and there is a suggestion that inhaled insulin led to less weight gain than injected insulin before meals in the studies that have been reported. The studies are a bit inconclusive, however, because relatively few participants were treated to achieve good control of their blood sugar. Nearly half of participants had HbA1c levels over 8% at the end of the study, 49% in the type 1 and 43% in the type 2 diabetes trials. Seven-point glucose profiles in the type 1 studies (measuring the blood glucose before and after each meal and at bedtime), showed consistently higher glucose levels with Afrezza before and after lunch and dinner and at bedtime, so that less hypoglycemia and less weight gain might simply be a function of having taken what amounted to a less effective insulin.

While speed of action may be a benefit of Afrezza, we await clinical trials which show this actually to occur. In addition, there are some concerns about safety of inhaled insulin. In clinical trials of Afrezza and in trials with several other inhaled insulins that have been studied, cough definitely occurred more often with the inhaled insulin preparations. There is also evidence of a reduction in pulmonary function with inhaled insulin. A pulmonary function test used in assessing asthma and other forms of bronchospasm, the forced expiratory volume in 1 second (FEV₁), decreased by about 0.04 liters in diabetic persons not using inhaled insulin and by about 0.08 liters by those using inhaled insulin during the first three months, with this difference in average decline persisting over two years.

Not only did FEV₁ decrease to a greater extent in those using inhaled insulin, but more severe decreases were at least twice as common with inhaled insulin, despite screening all participants in the trials for lung disease before enrollment. Even with careful screening, giving Afrezza only to volunteers who had normal lung function and who had never smoked or had quit smoking at least six months before starting the treatment, about one in twenty had a 15-20% reduction in the FEV₁ and about one in forty had a reduction by more than 20%.

Afrezza is not the first inhaled insulin to come to market. Between September 2006-October 2007, Pfizer marketed Exubera, an inhaled insulin that may have caused lung cancer. The risk of lung cancer was more than five times as high in diabetic patients who took this product, and this did not occur only in cigarette smokers, although a smoking history appeared to greatly increase the likelihood of lung cancer developing. It’s important to note that insulin acts in the body not only to lower blood glucose levels, but also as a growth-promoting hormone. Lung cancer cells appear to be more sensitive to such effects than normal lung cells, and there is no doubt that with inhaled insulin the lungs are exposed to much higher insulin concentrations than those seen in people not taking insulin injections, or those seen after insulin injections. Exubera also reduced the lung’s ability to exchange oxygen and carbon dioxide, the pulmonary diffusing capacity, to a greater extent. Exubera was withdrawn voluntarily from the market by Pfizer, its manufacturer, and we do not know whether the cancer risk was real, or whether, if it exists, it would apply to Afrezza.

At the FDA advisory committee presentation on Afrezza, it was estimated that a randomized clinical trial to detect risk of lung cancer doubling with this treatment would require more than 60,000 participants, followed for many years. So, it’s fair to say that we are a long way from really knowing about Afrezza and long term cancer risk

What’s the bottom line? Inhaled insulin works. Many people who need to take insulin might find it much more comfortable to take a “puff” than a “shot” before meals. This could lead to more use of insulin where it is needed, and to better control of diabetes. However, inhaled insulin causes cough, although this tends to improve over time. It causes some reduction in lung function – and we just do not know whether an adverse effect in less than one in ten people over short term studies will more commonly lead to such effects over years of clinical use. More worrisome, another similar product appears to have increased lung cancer.

Will I prescribe Afrezza for people with diabetes? Not a lot, and not for past or present smokers, and not for people with any breathing problems (or heart issues that might cause difficulty breathing), and not without getting pulmonary function testing and explaining all of what I see as potential risks, and not without close follow-up. But some people who need insulin just cannot bring themselves to inject it, and, for a few of these, recognizing the potential for harm, Afrezza might still be a good answer.