Commercially Available Insulin Products Demonstrate Consistency With Product Labeling Throughout All Seasons in the U.S.

In 2017, the stability of insulin in the cold supply chain in the U.S. was questioned by a report from Carter and Heinemann (1). Specifically, the study stated that nearly all insulins purchased from local pharmacies in the U.S. failed to meet the required standard of 100 units/mL (±5 units/mL) set forward by the U.S. Pharmacopeia (USP). This was an especially surprising finding as the U.S. Food and Drug Administration and the European Medicines Agency require that insulin formulations be subjected to rigorous testing prior to release from pharmaceutical manufacturers. The results of that study were also subjected to widespread reporting in various forms of social media, generating a series of commentaries noting that insulin potency is of vital importance to diabetes management (2,3).

To provide a follow-up to that report (1), we designed a comprehensive study to assess nine different insulins, purchased from four pharmacies per region (representing both rural and urban areas) in five regions across the U.S. on four different occasions (winter, spring, summer, and fall; designated seasons 1–4 below) (4). The regions included 1) Seattle, WA (Northwest); 2) Houston, TX (Southwest); 3) Gainesville, FL (Southeast); 4) Boston, MA (Northeast); and 5) Ann Arbor, MI (Midwest). Insulins obtained at each site included Humalog, NovoLog, Humulin R, Humulin N, Novolin R, Novolin N, Humulin 70/30, Lantus, and Basaglar (rapid-acting, short-acting, intermediate-acting, long-acting, and premixed formulations produced by three major manufacturers, Eli Lilly, Novo Nordisk, and Sanofi). We noted that the methods used by Carter and Heinemann (1) were not optimal for the measurement of commercial insulin and subsequently demonstrated that using such methods generates lower numbers of international units (IUs) than expected due to incorrect assumption of extraction recovery. Importantly, we also demonstrated that two complementary methods (USP monographs and liquid chromatography–tandem mass spectrometry) indicated all insulins analyzed were stable across the U.S. without exception, indicating that insulin purchased from U.S. pharmacies was, in fact, consistent with product labeling. Given the immediate importance of these findings to the medical community, we elected to report results from the first seasonal sampling (winter) only (4), leaving the impact of seasonality unaddressed.

Unfortunately, our ability to address the issue of seasonality with respect to insulin stability was severely impacted by travel-related issues associated with the coronavirus disease 2019 pandemic in the spring of 2020. As a result, we were able to obtain the insulins across all five regions for only two seasons. It was feasible to collect insulin during all four seasons through the summer of 2021 only in the Southeast region. Here, we report new results related to the impact of seasonality as a function of region.

For season 2, results observed were similar to those of our initial study (4). All insulins (n = 174) tested across every region in season 2 were noted to be within USP guidelines for potency (100 ± 5 units/mL) using USP monographs. A key aspect of our study was the assessment of potential regional differences associated with shipment that could be due to temperature fluctuations, and our results indicate potency is intact. In addition, we confirmed that USP methods to measure potency are reliable.

In the Southeast region, where data for all four seasons were collected, we again observed that all insulins were stable across the four seasons (Table 1). Our results ranged from 96 to 104 IU/mL across all insulins and seasons, well within the potency specified by USP. These results also establish confidence in shipping of insulin during a pandemic, since our study was conducted before and during the coronavirus disease 2019 pandemic. Together, data from the two seasons across the U.S. and all four seasons in one region provide unequivocal evidence that no seasonality differences exist in potency associated with shipment of insulin to pharmacies. It is important to note, however, that these newer tests utilized only USP monographs, since they correlated with liquid chromatography–tandem mass spectrometry methods in our previous efforts (4).

These results provide strong evidence that insulins purchased in the U.S. are stable regardless of location or time of year. The issue of seasonality was considered important as multiple variables, including weather/temperature during insulin distribution, potential impacts of shipping duration, and local variation resulting from warehouse/wholesale distributors (particularly between urban and rural areas), could have an impact on insulin stability. These findings provide important information necessary for both patients and health care providers to have confidence in the stability of current insulin formulations and their labeling.