Background: Clozapine is the most effective antipsychotic for treatment-resistant schizophrenia. Although serum clozapine levels can help guide treatment, they are underutilized owing to requirements for frequent venous blood draws and lack of immediate results.
Methods: Clozapine levels measured with a novel immunoassay technology (which enables point-of-care development) were compared with those measured by standard liquid chromatography/tandem mass spectrometry (LC-MS/MS). Frozen serum aliquots of 117 samples (N = 48 patients with schizophrenia on clozapine; N = 24 patients with schizophrenia not on clozapine; N = 45 healthy controls) were sent to a national reference laboratory (NRL) for clozapine level determination by LC-MS/MS, and matching samples were subjected to novel immunoassay (3 runs). At a later date, another frozen aliquot from the same date was sent to the NRL for repeat testing.
Results: The NRL obtained 18 false-positive clozapine results (mean 42.39 ± 32.06, range 21-159 ng/mL) in participants not on clozapine (N = 3) and healthy controls (N = 15). The immunoassay showed no false-positive clozapine results. The clozapine levels were correlated between both assays (r = 0.84, P < 0.0001), despite 16% higher clozapine levels with immunoassay (482.08 ± 270.88 ng/mL immunoassay, 414.98 ± 186.29 ng/mL LC-MS/MS [P = 0.03]). Agreement analysis using concordance correlation coefficient (CCC) for LC-MS/MS of the 2 aliquots yielded CCC = 0.869; 95% confidence interval = 0.690-0.970, whereas higher agreement results were observed for the 3 runs of immunoassay (CCC = 0.99; 95% confidence interval = 0.979-0.997).
Conclusions: The lack of false positives observed with immunoassay, higher repeat performance agreement, and good correlation with LC-MS/MS may indicate the more robust performance of immunoassay than that of LC-MS/MS clozapine-level determination.