Matthew Olson (Presenter)
Johns Hopkins University School of Medicine
Bio: Matthew T. Olson, M.D., is a practicing cytopathologist at the Johns Hopkins Hospital and an assistant professor of pathology in the Johns Hopkins University School of Medicine. As an active researcher with a longstanding interest and experience in mass spectrometry, I am excited about the development of mass spectrometric based ancillary tests to enhance the diagnostic process of cytological specimens. Developing these requires keen attention both to mass spectrometry fundamentals and to the workflow, strengths, and weaknesses of modern anatomic pathology operations, and my entire career thus far has been devoted to furthering these endeavors.
Authorship: Matthew T. Olson1*, Aparna Baxi1, Mariam ElNaggar2, Christopher Umbricht3, Alfred L. Yergey4, William Clarke1
Departments of 1Pathology and 3Surgery, The Johns Hopkins University School of Medicine Baltimore, MD 2Prosolia, Inc. Indianapolis, IN 4National Institute of Child Health and Human Development, NIH
| Short Abstract Cytological specimens are among the most abundant in anatomic pathology services worldwide because they are obtained through non- and minimally invasive methods. However, cytology challenges pathologists because limited sampling strategies that are so appealing to clinicians and patients are small. Thus, the cytologic diagnostic process is amenable to ancillary testing, especially when it does not interfere with classical morphology. Herein we present an ambient pressure ionization technique validated for both morphology and mass spectrometric analysis and compatible with routine cytological processing. We also demonstrate the need to choose a preservative that is amenable both to morphological and mass spectrometric analyses. |
Long Abstract
Background
Modern lipid analysis requires mass spectrometric techniques, and these have been directed primarily to histological specimens. Histological specimens are useful from the perspective of basic science research, biomarker discovery, and method development because the material in them is abundant and contains micro-anatomic features that facilitate morphological correlation for pathologists. However, the clinical usefulness of mass spectrometry on histological sections is less straightforward because the larger the section of tissue, the bigger the resection that was required to remove that tissue from the patient. As such, the predictive value of one of these sections is less than the predictive value that could be appreciated with a more limited sampling strategy. On the other hand, cytological specimens are limited in nature but so are often used for key clinical decisions – these include to treat or not to treat, to distinguish infection from neoplasm, to treat a neoplasm with surgery or chemotherapy, and so forth. Thus, it seems reasonable to focus the development of mass spectrometric techniques on cytological specimens as the potential for these techniques to add value is very high.
Methods
Cells, Cytological Preparation, and Preservation
All analyses were performed with a standardized breast cancer cell line MDA-MB-231 maintained in RPMI1640 10% FBS 1% penicillin-streptomycin at 37°C and 5% CO2 humidified incubator to near confluence. The harvested cells were re-suspended at quantifiable concentrations of 0.4 x 105, 0.6 x 105, 0.8 x 105, 1.0 x 105, and 2.0 x105 cells/ml in buffered saline and processed as a standard cytological specimen using Cytologic preparations were performed in duplicate for each cellular concentration. For each slide, a 400 µL aliquot was transferred to a Shandon™ EZ Single ™ cytofunnel with white filter card (Thermo Scientific, Grand Island, NY) and centrifuged at 1000 rpm for 5 minutes. After centrifugation and removal from the funnel device, one of the duplicate slides was allowed to air dry without any manipulation; the other slide was subjected to Flowprobe-MS (FMS) analysis, described in detail below. Afterwards, both slides were stained in Quik-dip and cover-slipped. To examine the effect of preservation on morphology and FMS analysis, numerous replicates of the cytospin preparations were prepared and stored in the following solutions: ethanol in buffered saline at concentrations of 0%, 10%, 25%, 50%, and 70%, ethanol in deionized water at the same concentrations, formalin, and Cytorich® clear. One slide from each preparation was subjected to FMS at 1 day and 1 week.
Mass Spectrometry
A Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Scientific) was used with a Prosolia (Indianapolis, IN) Flowprobe liquid microjunction surface sampling device. For all experiments detailed in this paper, a 2:1 solution of methanol and chloroform was used as the extraction solvent. The extractive junction was rastered in the same pattern across all the sample spots to allow full extraction of the entire spot. The probe height with respect to the surface of the glass slide was maintained at 80µm. Data acquisition in the mass spectrometer proceeded with a spray voltage of 3.5 kV, capillary temperature of 450°C, and a mass range of 70-1000 m/z. For higher energy collisional dissociation (HCD) analysis, the normalized collision energy (NCE) was set to 20-25% with an isolation window of 0.4 Daltons and mass range 100–2000 m/z.
Morphological validation
All slides were masked and reviewed by a panel of seven skilled cytomorphologists including board certified cytopathologists (2), licensed master-level cytotechnologists (3), and board certified anatomic pathology fellows (2). The panel was told that the cytospins were that of a ductal breast adenocarcinoma prepared in different ways. First they were asked to compare the morphology of two slides: one slide that had been and one slide that had not been subjected to FMS. The panel was then asked to rank the preserved material most against their chosen favorite of the FMS/ non-FMS slide pair.
Results
Lipid profiles of cytological specimens
The lipids expected from this particular cell line have been characterized using destructive techniques in a recent publication. In the present work, nearly all of the same parent ions have been detected with the non-destructive FMS. Additionally, FMS appeared to show less adduct so yielded more straightforward lipid assignments. The fragmentation spectra were unequivocal for approximately 20 of the most intense peaks, and collision energy conditions for the less intense peaks is currently in the process of optimization. Lipids identified included a mixture of phosphoglycerols, phosphoinositols, phosphocholines, and sphingomyelins.
Preservation and its effects
Spectra obtained from cells preserved under each of the 11 preservation conditions was compared to demonstrate several clearly superior preservatives for mass spectrometric purposes. These included 10 or 25% ethanol in normal buffered saline as well as formalin. There were also clearly inferior preservatives, and these included water-based preservatives, any preservative with an ethanol concentration >25%, and Cytorich clear.
Morphological validation
All panelists concluded that FMS analyzed specimens were at least as good as the non-FMS specimens, and all but one panelist chose the FMS analyzed specimen as the one with superior morphology. In a review of the preservatives, there was a plurality of support for low-ethanol in buffered normal saline. Other preservatives also ranked highly and can be discussed in further detail when space allows. Formalin universally received the lowest ranking – this preservative is known to be suboptimal for cytological specimens and is not routinely used in this context. Thus, there appears to be ideal preservative mixtures that suits both FMS and morphological analysis.
Conclusions
From these observations and others, to be described in greater detail as space permits elsewhere, FMS is fit for the purpose of analyzing cytology specimens in a way that does not hinder morphological analysis. To our knowledge, ours is the first comprehensive evaluation of specific cytomorphology-friendly techniques that are readily compatible with existing workflow in the conventional diagnostic workup of cytological specimens.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no | Prosolia ( |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |