MSACL 2022 Abstract

INTRODUCTION:
The tumor suppressor PTEN is the main negative regulator of PI3K/AKT/mTOR-signaling and is commonly found to be downregulated in many cancers, including breast cancer (BC), due to transcriptional and post-transcriptional mechanisms. Conflicting immunohistochemistry (IHC) and western blot data have sparked a controversy about PTEN's role as a prognostic and predictive biomarker in these cancers.

OBJECTIVES:
Because the existing data lacks the precision required to correlate minor PTEN-expression changes in tumors with clinical data, our goal was to develop and validate a fully-standardized, highly sensitive, robust anti-peptide immuno-multiple reaction monitoring (iMRM) assay for PTEN quantification.

METHODS:
We selected the PTEN peptide NNIDDVVR as the ideal target after querying databases, enforcing specific sequence and peptide criteria, and analyzing recombinant PTEN by DDA. The LC-MRM method for NNIDDVVR was developed on an Agilent 6495A triple quadrupole mass spectrometer. The LC conditions and collision energies for individual MRM transitions were optimized, and the resulting 11-min LC-MRM method was validated using CPTAC guidelines. Because direct quantitation of PTEN from cell or tissue lysate was not feasible, and because PTEN protein levels are expected to inversely correlate with disease severity, we generated an anti-peptide antibody to enrich NNIDDVVR prior to LC-MRM. The resulting immuno-MRM assay was used to quantify PTEN in cell lines, fresh frozen-, and formalin-fixed paraffin-embedded (FFPE) cancer tissues.

RESULTS:
The average recovery of the anti-NNIDDVVR immuno-enrichment was 90%, and the average accuracy of the complete iMRM assay was 87%. Our iMRM assay enabled precise quantitation of PTEN concentrations in cell lines, fresh frozen- and FFPE tissues, down to 0.1 fmol/10 µg (of extracted protein), with high inter-and intra-day precision (CV 6.3%). iMRM PTEN concentrations in BC-derived patient-derived xenografts (PDX) were consistent (i) across biological replicates, e.g. 5.7±0.1 fmol/10 µg (PTEN-IHC-high) and 0.7±0.0 fmol/10 µg (PTEN-IHC-low); (ii) across technical replicates with an average %CV of 24% for three cores analyzed from the same FFPE block; (iii) generally showed the same trend as the IHC classification. The accuracy of iMRM assays may be affected by the selection of the tissue locus, protein extraction yields, digestion efficiency, and peptide binding capacity, for example, but our results for technical and biological replicates show a low variation in PTEN levels. Although iMRM required substantially less sample-input, minor differences in PTEN-concentrations could be determined even between samples that were deemed to be PTEN-negative based on WB/IHC. Interestingly, the IHC- and iMRM-derived PTEN levels did not agree well with PTEN copy number variations (r 2=0.15), confirming the concept behind proteogenomics, i.e. that genome-only analyses may often be insufficient to capture the phenotype of a tumor and for making optimal treatment decisions. Finally, we applied our PTEN iMRM method to PDXs derived from primary and metastatic triple negative BC samples. The PDXs were treated with paclitaxel and the regression in tumor size was evaluated over 21 days. All the metastatic samples showed a very good correlation (r2=0.86) between PTEN iMRM concentration and treatment-response. More samples will be analyzed to confirm these initial results and to validate PTEN as a clinically relevant biomarker of BC patient's prognosis and response to therapy.

CONCLUSION:
The anti-peptide immuno-multiple reaction monitoring (iMRM) assay which we developed includes an 11-min micro-flow LC-MRM/MS analysis on a triple-quadrupole MS. This assay provides a much-needed tool for the study of PTEN as a potential biomarker in many cancers.