Targeted proteomics has become the method of choice for biomarker validation in human biopsies due to its capacity to measure a set of peptides in multiple samples with high sensitivity, reproducibility, accuracy and precision. However, for targeted proteomics technologies to be transferred to clinical routine there is the need to reduce its complexity and make its procedures simpler, increase its throughput and improve its analytical performance. Biomarker peptide quantification with high accuracy and precision is one of the steps that still remains a challenge in clinical routine, mainly due to the difficulty to account for matrix effects (i.e. signal variation due to biological patient variability) and to establish a valid range of quantification for each analyte in each individual sample. Research-grade proteomics laboratories perform targeted peptide quantification with stable isotopically-labelled peptides (one per analyte), which are added to the samples and used as internal standards. This strategy, known as single-point calibration, enables to confidently assign the endogenous peptide and infer its quantity by direct comparison to the internal standard. This approach however assumes that both the endogenous and the internal standard peptides are within the linear range of quantification, which is not always granted as the linear dependency between peptide areas and concentrations only occurs in a limited sample-dependent range of concentrations. Although standard peptide abundances are adjusted to endogenous levels in low-throughput projects, this is not feasible when dealing with large cohorts of patients in clinical routine that exhibit a high variability of peptide abundances and different matrix effects. External multi-point standard curves in their different forms (calibrated, reverse) have been used to alleviate part of these limitations, but the requirement of a representative blank matrix and the fact that they do not account for matrix effects in individual samples nor establish a valid range of quantification for each individual sample limit their application. Here we present the Isotopologue Multiple-point Calibration (ImCal) quantification strategy, which uses a mix of isotopologue peptides to generate internal multiple-point calibration curves for each individual sample and to accurately quantify biomarker peptides in clinical applications without the need of expert supervision. ImCal relies on the use of five different isotopically-labelled peptides of different nominal mass―ranging from 10 to 39 Da mass shifts—mixed at different concentrations to be used as internal calibration curve for each endogenous peptide of interest

The performance of ImCal was compared with other quantification strategies, namely single point calibration, external calibrated curves, and reverse calibrated curves by selected reaction monitoring in a triple quadrupole mass spectrometer. For this purpose, two unique peptides for human ERBB2 receptor tyrosine-protein kinase 2 (ELVSEFSR and GLQSLPTHDPSPLQR; Protein HER2, P04626) were selected from the CPTAC assay portal. The use of isotopologue multiple-point internal calibration curves resulted in higher accuracy and precision for the quantification of the selected peptides. Finally, we used the ImCal method to quantify HER2 in formalin-fixed paraffin-embedded (FFPE) tissue biopsies from breast cancer patients.

Known amounts (5, 50 and 200 fmol) of each synthetic non-labeled peptide were analyzed in a background matrix consisting on a digested E. coli protein extract to assess the accuracy and precision of the ImCal method. In parallel, five isotopologues forms were synthesized for each selected peptide using different combinations of 13C615N4-Arg, 13C515N1-Val, 13C615N1-Leu, 13C915N1-Phe, 13C515N1-Pro and 13C515N1-Glu and mixed at different amounts (2, 10, 20, 100, 200 fmol/µL) to generate internal and external calibration curves covering three orders of magnitude. Clinical samples were obtained from formalin-fixed paraffin-embedded (FFPE) tissue biopsies from 10 HER2+ and 25 HER2- breast cancer patients