Background

Insulin is a polypeptide hormone primarily produced by beta cells in the islets of Langerhans of the pancreas. It plays a crucial role in carbohydrate metabolism, and individuals who cannot produce insulin are unable to regulate their blood glucose levels effectively [1]. Insulin is a life-saving medication for individuals with type 1 diabetes mellitus and is also essential for some patients with type 2 diabetes mellitus who require daily injections to maintain glycemic control [2].

To preserve insulin’s concentration and efficacy, manufacturers and regulatory agencies such as the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and World Health Organization (WHO) recommend that all insulin formulations be stored within a temperature range of 2°C to 8°C (36°F to 46°F) [3]. This temperature range is specified in pharmacopeial monographs, such as the United States Pharmacopeia (USP) and European Pharmacopeia (Ph. Eur.), which outline stability requirements and quality control parameters for insulin storage and distribution [4]. However, in many Low- and Middle-Income Countries (LMICs), including Tanzania, maintaining these optimal storage conditions remains a challenge due to unreliable electricity supply, inadequate refrigeration facilities, and economic constraints [5]. Inadequate storage conditions have been widely recognized as contributors to potential insulin degradation. Nonetheless, an often overlooked but equally important issue is the source and manufacturing quality of insulin products, especially in LMICs where regulatory oversight of imported pharmaceuticals may be limited. Defective products may originate from manufacturers with poor quality control standards, making it difficult to attribute reduced concentration solely to storage conditions.

Insulin stability is affected by temperature fluctuations, direct exposure to light, and mechanical stress due to agitation, all of which can accelerate degradation processes. Degradation mechanisms include chemical degradation (e.g., hydrolysis and deamidation) and physical instability (e.g., fibrillation and precipitation) [6,7]. Regulatory guidelines like USP monograph, European pharmacopeia (Ph. Eur.), and ICH Q5C-stability of biotechnological/Biological products highlight that insulin formulations degrade rapidly at pH values below 5 and above 8, necessitating stringent control over formulation conditions to ensure product stability [8].

Different insulin formulations exhibit varying degrees of stability due to differences in peptide structure, excipients, and manufacturing processes [9]. Regulatory authorities require stability testing for each insulin formulation to establish shelf-life specifications under both recommended and stressed conditions. The International Conference on Harmonisation (ICH) guideline Q1A (R2) provides a framework for evaluating stability and defining appropriate storage conditions [10]. However, limited data exist on the quality of insulin formulations available in sub-Saharan Africa, necessitating region-specific studies to assess potential degradation risks.

In many parts of sub-Saharan Africa, including Tanzania, weak supply chain infrastructure coupled with limited regulatory capacity contributes to the importation of substandard medicines. Studies have shown that temperature excursions beyond the recommended range during transportation and storage can lead to chemical and physical degradation, reducing insulin’s content and clinical effectiveness [6]. Moreover, regulatory oversight of manufacturing facilities that supply imported medicines to these regions remains limited, with insufficient enforcement of pharmacovigilance measures to ensure insulin quality throughout the supply chain.

Several studies across Africa have highlighted significant concerns regarding insulin quality defects, primarily due to improper storage conditions and handling practices. For instance, research conducted in Ghana and Nigeria found that insulin vials stored in health facilities often exceeded recommended temperature limits, raising concerns about compromised potency [11,12]. In LMICs, evidence regarding insulin quality is scarce. One study conducted in Kenya found no significant change in insulin content after exposure to simulated laboratory conditions [6]. To address this gap and ensure insulin quality assurance in Tanzania, our study evaluated the quality of insulin obtained from community and hospital pharmacies in Mwanza, Tanzania.

Methods

Results

System suitability

The percentage Relative Standard Deviation (%RSD) values for peak area and retention time were within 2% as shown in Table 1 below, indicating the system’s suitability. The efficiency of the column separation, expressed by a resolution of NLT 2.0, and the peak symmetry, expressed by a tailing factor of 1.10 ± 1.1% (mean ± % RSD), also confirmed the system’s suitability.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. System suitability.

https://doi.org/10.1371/journal.pone.0325366.t001

Fifteen different insulin samples were used to assess the quality of human insulin as shown in Table 2. All brands were registered

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Commercial product information and insulin concentration in a different insulin brands.

https://doi.org/10.1371/journal.pone.0325366.t002

Identification of insulin in products

The retention time of the major peak of the sample ranges from 12.8 minutes to 14.0 minutes, with a mean retention time of 13.1 minutes and a percentage relative standard deviation (%RSD) of 1.4%. This corresponds to the retention time of the reference standard solution, which is 13.7 minutes with an RSD of 1.4%. Fig 1 confirms the presence of human insulin in the injected samples, as evidenced by peaks corresponding to the insulin standard. In contrast, Fig 2 shows no detectable human insulin, suggesting either degradation or absence of the compound in those samples. Each set of figures includes a chromatographic peak for the reference standard (Figs 1a and 2b), which serves to confirm the identity of insulin in the samples. The corresponding sample peaks are used to both verify the presence of insulin and determine its concentration by comparison with the standard.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Chromatograms for the samples and reference standard solution of human insulin: “a” shows the reference standard solution of human insulin CRS, and “b,” “c,” and “d” show Insu 009 (batch no. BF23000685).

https://doi.org/10.1371/journal.pone.0325366.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Chromatograms for the samples: “a” shows Insu 009 (batch no. BF23000685), “b” shows a standard solution of human insulin CRS, “c” shows 010 (batch no. BF21003364), and “d” is a blank with 0.01 N HCl.

https://doi.org/10.1371/journal.pone.0325366.g002

Discussion

The temperature storage conditions for the insulin samples during the study period were satisfactory. Therefore, the storage temperatures in the pharmacy premises do not appear to explain the out-of-specification results observed in two samples (13.3%).

In the current study we found human insulin samples matched the reference standard solution in terms of retention time (13 minutes), with tailing factors not exceeding 1.8 and peak resolution not less than 2.0. These results indicate that the method used for detection and assessing insulin concentration in the vials was suitable for separating human insulin and A-21 desamido insulin. These findings are consistent with another study that determined the identification of human insulin in various pharmaceutical products, where the retention time was found to range between 14 minutes and 20 minutes [15,16].

Of the 2 (13.3%) non-compliant or substandard human insulin samples as shown in Table 2. This discrepancy could be attributed to poor manufacturing processes or poor handling during transportation. Manufacturing of large molecules like proteins (insulin) is very complex with many parameters that are difficult to control. Manufacturers of large molecules face significant challenges due to batch-to-batch variability, which can lead to final products that do not meet minimum quality standards [17,18]. The World Health Organization, U.S., and EU have developed guidelines for the distribution of cold chain products like insulin, placing responsibility on manufacturers, regulatory authorities, and wholesalers to ensure proper transport conditions [3].

Out of 15 human insulin samples tested, two were substandard (13.3%) failed to meet the official USP monograph standards in the assay test. This non-compliance likely stems from cold chain distribution challenges common in Sub-Saharan Africa, including Tanzania, as well as potential manufacturing shortcomings. These findings underline the urgent need to strengthen quality control and distribution systems for insulin products.

The remaining human insulin samples comply with official USP standards, meeting the acceptance criteria of not less than (NLT) 27.5 mg and concentration range of 95%−105% [4]. Out of 15 human insulin samples, 13 (13.3%) comply with the official USP standards as shown in the Table 2 above, consistent with findings from other studies [6,19,20]. The chromatograms in Figs 1 and 2 indicate that the human insulin samples (Insu 009 & 010) were substandard, likely due to poor formulation during manufacturing process or environmental factors such as temperature, light, and agitation during transportation.

Insulin monomers are highly susceptible to mechanical agitation during manufacturing, transportation, and handling. Agitation induces partial unfolding of the insulin structure, exposing hydrophobic regions of the peptide that are usually buried within the native conformation [21]. These exposed hydrophobic regions facilitate the self-association of insulin monomers, leading to aggregation. Over time, these aggregates can evolve into insoluble amyloid fibrils through a nucleation-dependent process [21]. This fibril formation not only reduces the active concentration of insulin in the vials but also compromises its therapeutic efficacy [22]. Furthermore, injecting insulin containing aggregates or fibrils can contribute to injection-site complications, including lipohypertrophy, lipoatrophy, and insulin-derived amyloidosis, while also impairing glycemic control [22]. The main limitation of this study is the lack of assessment for insulin fibrils in the non-compliant human insulin vials. We recommend further research on non-compliant insulin samples [23].

Conclusion

The study reveals significant insights into the quality of insulin products, highlighting that while most samples meet the official USP standards, a few do not, reflecting potential issues in manufacturing and distribution. Specifically, two samples lacked active human insulin, which may be due to inadequate manufacturing processes or improper storage during transportation.

Acknowledgments

The authors extend their thanks to the Catholic University of Health and Allied Sciences (CUHAS) and the Higher Education and Economic Transformation (HEET) project for funding this study, as well as to the Tanzania Medicine and Medical Devices Authority (TMDA) for their support with laboratory work. We also appreciate the expert reviewers for their valuable feedback, which greatly improved the quality of this article.