Testing of rapid evaporative mass spectrometry for histological tissue classification and molecular diagnostics in a multi-site study

Abstract

Background: While REIMS technology has successfully been demonstrated for the histological identification of ex-vivo breast tumor tissues, questions regarding the robustness of the approach and the possibility of tumor molecular diagnostics still remain unanswered. In the current study, we set out to determine whether it is possible to acquire cross-comparable REIMS datasets at multiple sites for the identification of breast tumors and subtypes.

Methods: A consortium of four sites with three of them having access to fresh surgical tissue samples performed tissue analysis using identical REIMS setups and protocols. Overall, 21 breast cancer specimens containing pathology-validated tumor and adipose tissues were analyzed and results were compared using uni- and multivariate statistics on normal, WT and PIK3CA mutant ductal carcinomas.

Results: Statistical analysis of data from standards showed significant differences between sites and individual users. However, the multivariate classification models created from breast cancer data elicited 97.1% and 98.6% correct classification for leave-one-site-out and leave-one-patient-out cross validation. Molecular subtypes represented by PIK3CA mutation gave consistent results across sites.

Conclusions: The results clearly demonstrate the feasibility of creating and using global classification models for a REIMS-based margin assessment tool, supporting the clinical translatability of the approach.

Fig. 1. Workflow for multicenter study of clinical samples.

Human breast tissue was obtained from patients who received surgical treatment for breast cancer at the following sites: Imperial College London (C1), Maastricht University (C3) and Queen’s University (C4). Tissue was analyzed by REIMS and validated by histopathologists. Quality control material including two batches of single-source pork liver and meat homogenate (National Institute of Standards and Technology (NIST), Standard reference material 1546a) was analyzed at all four centres including Waters Research Center (C2) to compare method performance.

Fig. 2. Multi-site characterization of quality control material.

Representative REIMS spectra from two quality control materials analyzed: a single-source pork liver and b NIST meat homogenate. PCA plots of pork liver quality control data acquired at all sites c previously-acquired pork liver sampling points analyzed at C3 by 7 analysts are included in comparison to all quality control data acquired across sites in (c) and in comparison to pork liver quality control data acquired only at C3 (d).

Fig. 3. Multi-site characterization of human breast tissue.

Representative REIMS spectra from either normal breast adipose a or invasive breast cancer b. The spectra shown are 1 s scans acquired using cut mode, and subjected to background subtraction and lockmass correction. The PCA c and PCA/LDA d plots compare the overall tissue-type variability between cut and coagulation modes. e–g Differential abundance of selected ions identified by peak picking, used to create a model with only 11 targeted ions.

Fig. 4. Fatty acid metabolism in breast tissue.

Scheme of ω6 fatty acid metabolism a indicating selected enzymes involved, and the role of PI3K signaling in the liberation of arachidonic acid from the cell membrane by cPLA2. Fatty acids measured in the current study, emphasize pathways that contribute to tissue pools of arachidonic acid. Boxplots in b from a subset of data stratified by PIK3CA genotype indicate relative concentrations of the sum of selected fatty acids in normal breast adipose (N, N = 82) compared to tumor tissue harboring wildtype (WT, N = 12) or mutant (Mut, N = 23) PIK3CA. There was a significant difference in the level of all fatty acids between N and WT/Mut with all p < 0.005. c Boxplots of ratio analysis showing the change in the overall fatty acid metabolism.