Mass spectrometry or immunoassay: est modus in rebus

As shown in a search run in PubMed with “mass spectrometry” in the title or abstract, the number of papers in the literature on this technique has increased exponentially, from one article in 1945 to more than 15,000 in 2016 (https://www.ncbi.nlm.nih.gov/pubmed?term=mass%20spectrometry%5BTitle%2FAbstract%5D). Thanks to progress made, liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) has undergone a transformation, and this technique is no longer considered an esoteric technology used in highly specialized laboratories for academic and research purposes, but a suitable methodology for clinical practice. Due to its high analytical specificity and sensitivity, this approach obviates the main disadvantages of immunoassays, which have relatively poor analytical performances with, for example calibration bias, lower analytical sensitivity and specificity, and a number of potential interferences. Immunoassays suffer from cross-reactivity with endogenous or exogenous related forms of the measurand, leading to false positives or false negatives on the basis of the assay design (sandwich or competitive); moreover, there is a lack of concordance across platforms, since the influence of cross-reactivity is closely related to the antibody used. Another advantage of LC-MS/MS over immunoassays is the lack of interference from autoantibodies and heterophilic immunoglobulins whose well-documented interference in immunoassays continues to be a critical problem [1], [2]. Initially, the sensitivity of mass spectrometers allowed the measurement of molecules in the micro-millimolar range: now manufacturers have developed instruments for measuring low-molecular-weight molecules present in very low concentrations, such as serum-free testosterone, serum aldosterone, and plasma-free metanephrines [3], [4]. Another advantage of LC-MS/MS is that it enables the simultaneous (multiplexing) measurement of more than one analyte, thus making it possible to achieve a complete pattern, such as in steroid profiling [5], [6]. This characteristic is particularly convenient for therapeutic drug monitoring (TDM), allowing the simultaneous measurement of drug subsets and the cost-effective management of patients on multidrug therapy [7].

LC-MS/MS is currently considered the method of choice for inherited metabolic disorder [8] and steroid hormone assay, especially in cases requiring accurate measurement at very low concentrations [9]. In the face of the strengths of LC-MS/MS, there are also some weaknesses, including high initial instrumental costs (albeit with the potential for important cost-saving), serial-access operation, need for highly skilled personnel, limited CE-IVD approved reagents and kits, and limited sample throughput [3]. However, interesting solutions have been proposed to address the latter critical pre-analytical issue [10], [11], [12]. LC-MS/MS assays also call for accurate validation and appropriate standardization/harmonization: a mass spectrometric method does not necessarily constitute a reference method. At present, however, an effort should be made by scientific societies to provide guidelines on MS-based method validation [13], [14].

In view of the above, it would be unfeasible to shift all the tests carried out by immunoassays to LC-MS/MS. Yet LC-MS/MS will have increasingly wide diffusion in at least two settings. One area is the development and validation of reference methods with MS-based methods, which cannot be transferred to routine practice due to their complexity, and their relatively long execution time. These reference methods will aid the definition of target values and help clinical laboratories define the trueness of their own methods, be they immunoassays or LC-MS/MS. In this regard, considerable improvement has been achieved in the comparability between results of immunoassays for C-peptide and insulin after the re-calibration of the immunoassays with a MS-based reference method [15], [16]. The other area of diffusion is the replacement of immunoassays for particular measurands: those with known problems of specificity and sensitivity should be analyzed with LC-MS/MS, while also taking into account the balance between the benefits (high accuracy measurement) and caveats (sample throughput, costs, skilled personnel). Some examples of this, in particular certain steroid hormones (e.g. cortisol), are given in literature [9], [17]. In a recent comparison made between four immunoassays and one routine LC-MS/MS method with a candidate reference LC-MS/MS method (cRMP), Hawley et al. [18] reported results that demonstrated a high variability in routine immunoassays in cohorts with well known interferences (pregnant women, patients on metyrapone or prednisolone). Only the routine LC-MS/MS method aligned extremely well with cRMP. Another example is TDM of immunosuppressants for which the accuracy of immunoassays can be compromised by the presence of potential interferences (cross-reactivity with metabolites or other drugs, and heterophilic antibodies) [19].

In our clinical laboratory practice, LC-MS/MS was initially implemented for TDM of immunosuppressants and for drugs of abuse; then analysis of urinary and salivary cortisol was transferred to this technique also to meet endocrinologists’ need for laboratory data that truly reflected the clinical status of patients [11], [12], [20], [21], [22]. Now LC-MS/MS is taking hold for the TDM of antiepileptic, antidepressive, antiarrhythmic, and neuroleptic drugs as well as benzodiazepines.

Several papers in this issue of Clinical Chemistry and Laboratory Medicine emphasize the wide range of applications of MS based methods [23], [24], [25], [26], [27], [28], [29], [30]. The review by El-Najjar et al. [23] highlights the crucial role of TDM of antibiotics in cancer patients: immunoassays are not available for all the classes of these drugs due to the difficulty in β-lactams ring reactivity or degradation. Therefore, LC-MS/MS methods have become the method of choice. Wiesen et al. [24] propose a multi-analyte LC-MS/MS method for the quantification of four direct oral anticoagulants (DOAC) in plasma suitable for TDM purposes and pharmacokinetics analysis, together with the possibility of screening when it is not known whether or, if so which, DOAC is taken. The study by Pantano et al. [25] reported the validation, and the application to real cases, of an LC-MS/MS method for the determination of oxycodone and its major metabolites in plasma and urine, and emphasized its rapidity and suitability in patients under oxycodone treatment. Ullah et al. [26] describe the versatility of this technique in analyzing phosphadylethanols (PEth), in oral fluid. Improved and validated, the method is suitable for measuring PEth in this non-traditional matrix. Another two papers discuss the possibility of measuring traditional analytes in a non-traditional matrix, highlighting the feasibility of using this instrumentation in more than one matrix [27], [28] for hormone measurement. The study by Spanaus et al. [29] serves as an example of differences in immunoassay performances, thus stressing the importance of evaluating each specific assay: results obtained on comparing two automated immunoassays for the measurement of 1,25 dihydroxyvitamin D with LC-MS/MS demonstrated that one immunoassay is appropriate for clinical purposes in view of its high sensitivity, low imprecision, broad measurement range and agreement with LC-MS/MS. Torma et al. [30] developed a cRMP for 1-32 BNP with the potential to support external quality assessment schemes and to establish the bias of immunoassays, in order to enhance harmonization between results.

In conclusion, the papers published in the present issue of the journal highlight the increasingly important role played by LC-MS/MS in laboratory medicine. Certainly, this technology is still limited to some laboratories, and manufacturers should make greater efforts to improve sample throughput and user friendliness, also to produce CE-IVD approved reagents and kits. Moreover, both national scientific societies and their federations should implement strategies to achieve harmonization and standardization of MS-based methods [31].