Analysis of insulin and insulin analogues by mass spectrometry

We read with interest the recent study by Dayaldasani et al., ¹ in which three insulin immunoassays were evaluated for cross-reactivity against insulin analogues. Certainly, the study is a valuable addition to the literature and will be a useful reference for those laboratories currently using the assays described. However, without doubt the introduction and widespread prescription of structurally modified analogues poses a significant analytical dilemma when insulin analysis is carried out using immunometric methods. This is especially true in cases where endogenous insulin and exogenous insulin(s) may be present in the same sample.

It was perhaps surprising then that no mention was made in this study of recent developments in the use of mass spectrometry-based detection methods for the unambiguous identification and simultaneous analysis of insulin and insulin analogues. Amongst others who have developed similar approaches, ² in our laboratory, we have found that mass spectrometry is a powerful tool that allows one to differentiate insulin analogues by their differing mass-to-charge (m/z) ratios. Coupled with immuno-affinity sample preparation (mass spectrometric immunoassay, MSIA™, Thermo Scientific) and isotopically labelled insulin as an internal standard (Peptides International, Louisville, KY), we have been able to achieve fully quantitative insulin analysis, with very low limits of detection (1 mIU/L for insulin, 500 µL serum) and are able to identify, even within the same sample, structural analogues (lispro, aspart, glargine, detemir, glulisine, degludec) and non-human insulins (bovine, porcine). ³ Chromatographic separation of the isobaric species insulin and lispro has also been achieved. ⁴

Our recent studies using high-resolution mass spectrometry, which allows retrospective interrogation of data, also suggest that further studies are warranted to investigate not just cross-reactivity of insulin analogues but also of their metabolites and/or degradation products. Studies where insulin analogues are ‘spiked’ into serum or serum pools, whilst useful in their own right, do not address this issue. For example, in vivo, it has previously been reported that insulin glargine is extensively metabolized and is excreted in the urine as two metabolites or degradation products. ⁵ Cross-reactivity data of the glargine metabolites in samples from patients prescribed glargine may be more useful than the cross-reactivity of glargine itself. Aside from insulin glargine, and applying this principle more widely to other insulins, no studies exist demonstrating the cross-reactivity (or lack of cross-reactivity) of insulin metabolites and/or degradation products with the range of commercially available immunoassays. Further, in complex situations such as those when multiple insulins/degradation products are present in a single sample, identification (let alone quantitation) of insulin species is an almost impossible task using only immunometric methods. For such cases, mass spectrometry-based analysis must be considered.

Insulin analogues as well as their respective degradation products should be considered as targets for further development of mass spectrometry-based methods for insulin analysis. As such, it would be useful for providers of reference materials to consider the production of insulin analogues and their degradation products as reference standards as well as, ideally, production of isotopically labelled versions for inclusion into quantitative workflows such as in MSIA.