MSACL 2017 US Abstract

Molecular Markers of Serous Ovarian Cancer Aggressiveness and Surgical Outcome by Ambient Ionization Mass Spectrometry Imaging

Marta Sans (Presenter)
The University of Texas

Bio: Marta Sans from Barcelona, Spain. Graduate student at the University of Texas at Austin, working under Professor Livia S. Eberlin on ambient ionization MS for health related research.

Authorship: Marta Sans (1); Kshipra Gharpure (2); Li Liang (3), Jinsong Liu (3); Robert Tibshirani (4), Anil K. Sood (2)*; and Livia Eberlin(1)*
(1) Department of Chemistry, The University of Texas at Austin, Austin, TX, 78712 (2) Department of Gynecologic Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 77030 (3) Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, 77030 (4) Departments of Health Research and Policy and of Statistics, Stanford University, Stanford, CA 94305

Short Abstract

High-grade serous ovarian cancer (HGSOC) and borderline serous tumors (BOTs) present underlying differences regarding tumor invasion and prognosis. Surgical resection is a key component of HGSOC treatment but post-operative disease adversely impacts patient survival. Here, desorption electrospray ionization (DESI) mass spectrometry imaging (MSI) together with multivariate statistical analysis was used to diagnose HGSOC, and tumor overall with 96.4% and 96.2% agreements, respectively. Molecular markers of cancer aggressiveness were also identified and selected as significant contributors for HGSOC and BOT classification. Finally, metabolic profiles were used to predict HGSOC patients at high risk for residual disease after surgery.

Long Abstract

Introduction

High-grade serous ovarian cancer (HGSOC) is the most aggressive form of ovarian epithelial cancer(1,2). HGSOC develops rapidly due to genetic instability and presents unfavorable prognosis, while borderline serous ovarian tumors (BOT) are indolent non-invasive neoplasms that can slowly progress to malignant low-grade serous carcinomas (LGSOC) (3,4). Due to their distinct tumor invasion behaviors, characterization of the molecular differences between malignant and borderline serous ovarian tumors could provide valuable insights to unravel the biological mechanisms driving tumor aggressiveness (5). Moreover, development of robust methodologies for detection and characterization of disease could potentially improve disease management (6).

Here, we used DESI-MS imaging to investigate the molecular profiles of serous ovarian cancers and characterize lipids and metabolites that could potentially serve as diagnostic markers of aggressive disease. 2D molecular images allowed unambiguous correlation between molecular signatures and regions with specific histological features. Classification models were built using the least absolute shrinkage and selector operator (Lasso) technique (29,30) and tested to predict disease and tumor malignancy of SOCs. Furthermore, the relationship between metabolic profiles and residual disease was investigated to better understand its role in tumor invasion and consequently provide potential biomarkers associated with surgical outcome. Selected significant species attributed to the models were tentatively identified by high mass accuracy/high mass resolution and tandem mass spectrometry analysis as lipid and metabolite species of biological relevance. Our results strongly suggest the capabilities of DESI-MS for SOC diagnosis and for the identification of potential markers of disease aggressiveness and surgical outcome, which can aid in the investigation of biological pathways involved in disease development and serve as a clinical tool for rapid diagnosis of ovarian cancers.

Methods

A total of 116 frozen human tissue specimens were obtained from the Cooperative Human Tissue Network and MD Anderson Tissue Bank under approved IRB protocol. Samples were stored in a -80oC freezer until sectioned. Tissue samples were sectioned at 16 µm thick sections using a CryoStarTM NX50 cryostat (Thermo Scientific, San Jose, CA). After sectioning, the glass slides were stored in a -80oC freezer. Prior to MS imaging, the glass slides were dried for approximately 15 min. A 2D Omni Spray (Prosolia Inc., Indianapolis, IN) coupled to an LTQ-Orbitrap Elite mass spectrometer (Thermo Scientific, San Jose, CA) was used for tissue imaging. DESI-MSI was performed in the negative and positive ion mode from m/z 100-1500, using a hybrid LTQ-Orbitrap mass spectrometer of high mass accuracy (<5 ppm mass error) and high mass resolution (240,000 resolving power) measurements. Spatially accurate ion images were assembled using Biomap and MSiReader software (200 µm spatial resolution). The same tissue sections analyzed by DESI-MSI were subjected afterwards to standard H&E staining protocol. Regions of clear diagnosis of serous ovarian cancer were assigned in the glass slides and differentiated from normal ovary stromal regions. Clinical diagnosis of the samples as BOT and HGSOC was also confirmed by pathology. After pathological evaluation, the MS data corresponding to the areas of interest was extracted using MSiReader software. The extracted data was statistically evaluated using the Lasso technique.

Results

DESI-MS imaging was performed in the negative and positive ion modes for a total of 116 tissue samples, including 27 normal ovarian, 23 BOT and 66 HGSOC tissues. Characteristic metabolic profiles were observed in both polarities and presented a remarkable diversity of molecular species. Specific histologic features characteristic of HGSOC and serous BOT tissues were observed and annotated for all samples analyzed. The spatial resolution provided by DESI-MS imaging allowed for visualization of key features of tumor heterogeneity in ovarian cancers which were directly correlated and clearly outlined the histological details of the tissues samples. High relative abundances of glycerophospholipid (GP) species such as glycerophosphoinositol (PI) 18:0/20:4 or glycerophosphoglycerol (PG) 16:0/18:1 or glycerophosphocholine (PC) 36:3 (in the positive ion mode) were observed by DESI-MSI in regions with high density of tumor cells, which allowed visualization of regions with tumor clusters in HGSOC. For BOT samples, high relative abundances of ceramide species such as Cer 42:1, Cer 42:2, was highly specific to the discrete tumor regions in BOT samples. Conversely, other species such as ascorbic acid, were more prominent in normal tissue. The distinct molecular compositions associated to the three histological subtypes suggested lipid and metabolite species as potential biomarkers for cancer diagnosis and aggressiveness.

Due to the large amount of molecular and spatial information from DESI-MSI analysis, the Lasso method was employed to build classification models for normal, HGSOC, and BOT and to select the molecular features that contributed to each class. First, HGSOC was classified in comparison to healthy tissue, with outstanding 96.4% overall agreement, for both negative and positive ion mode data. Next, we evaluated the ability of our method to discriminate the three classes of ovarian tissues, normal, BOT and HGSOC, and thus predict diagnosis and aggressiveness of serous ovarian cancers, yielding overall 91.9% agreements. A subset of m/z values were selected by the Lasso as potential biomarkers of cancer aggressiveness for negative and positive ion mode data, which were identified as relevant molecular species of important biological roles. Finally, in order to investigate the predictive value of metabolic markers for surgical outcome, lipid and metabolite data was correlated to clinical data on surgical outcome (residual versus no-residual disease following cytoreduction surgery) for a total of 24 high-grade SOC patients. Negative ion mode data showed good sensitivity with 91.6% agreement to classify patients with residual disease, while 77.4% agreement predicting residual disease was achieved in positive ion mode.

Conclusions

Ambient ionization mass spectrometry imaging provided the needed spatial resolution to visualize key heterogeneities within the tissue samples, which allowed for careful characterization of the phenotypic subtypes. Overall agreements of 96.4% were achieved to classify HGSOC in comparison to healthy tissue which demonstrate the capability of DESI-MS-Lasso to diagnose the most aggressive form of serous ovarian cancers, which is highly relevant due to the high occurrence and poor prognosis of the disease (5). Molecular markers for cancer aggressiveness were selected by the Lasso, building molecular models to discriminating between normal, BOT and HGSOC samples with 91.9% overall accuracy. Moreover, clinical data on surgical outcome was related to significant alterations in metabolic profiles for HGSOC, showing potential trends for the selection of a group of patients at high risk for residual disease. These results show the powerful capabilities of DESI-MS as a diagnostic tool for SOC, that can be used by clinicians to design personalized surgical approaches and general course of action, which is of high demand within the surgical community to improve overall patient survival (10,11).


References & Acknowledgements:

References

1. Leong HS, Galletta L, Etemadmoghadam D, George J, Kobel M, Ramus SJ, et al. Efficient molecular subtype classification of high-grade serous ovarian cancer. J Pathol 2015;236(3):272-77.

2. Hayano T, Yokota Y, Hosomichi K, Nakaoka H, Yoshihara K, Adachi S, et al. Molecular Characterization of an Intact p53 Pathway Subtype in High-Grade Serous Ovarian Cancer. Plos One 2014;9(12).

3. Vang R, Shih IM, Kurman RJ. Ovarian Low-grade and High-grade Serous Carcinoma Pathogenesis, Clinicopathologic and Molecular Biologic Features, and Diagnostic Problems. Adv Anat Pathol 2009;16(5):267-82.

4. Fischerova D, Zikan M, Dundr P, Cibula D. Diagnosis, Treatment, and Follow-Up of Borderline Ovarian Tumors. Oncologist 2012;17(12):1515-33.

5. Saad AF, Hu W, Sood AK. Microenvironment and pathogenesis of epithelial ovarian cancer. Horm Cancer 2010;1(6):277-90.

6. Vaughan S, Coward JI, Bast RC, Berchuck A, Berek JS, Brenton JD, et al. Rethinking ovarian cancer: recommendations for improving outcomes. Nat Rev Cancer 2011;11(10):719-25.

7. Nick AM, Coleman RL, Ramirez PT, Sood AK. A framework for a personalized surgical approach to ovarian cancer. Nat Rev Clin Oncol 2015;12(4):239-U78.

8. Bowtell DD, Bohm S, Ahmed AA, Aspuria PJ, Bast RC, Beral V, et al. Rethinking ovarian cancer II: reducing mortality from high-grade serous ovarian cancer. Nat Rev Cancer 2015;15(11):668-79.

9. Tucker SL, Gharpure K, Herbrich SM, Unruh AK, Nick AM, Crane EK, et al. Molecular biomarkers of residual disease after surgical debulking of high-grade serous ovarian cancer. Clin Cancer Res 2014;20(12):3280-8.

10. Hotamisligil GS, Johnson RS, Distel RJ, Ellis R, Papaioannou VE, Spiegelman BM. Uncoupling of obesity from insulin resistance through a targeted mutation in aP2, the adipocyte fatty acid binding protein. Science 1996;274(5291):1377-79.

11. Furuhashi M, Hotamisligil GS. Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets. Nat Rev Drug Discov 2008;7(6):489-503.

12. Uysal KT, Scheja L, Wiesbrock SM, Bonner-Weir S, Hotamisligil GS. Improved glucose and lipid metabolism in genetically obese mice lacking aP2. Endocrinology 2000;141(9):3388-96.

13. Suhre K, Romisch-Margl W, de Angelis MH, Adamski J, Luippold G, Augustin R. Identification of a potential biomarker for FABP4 inhibition: the power of lipidomics in preclinical drug testing. J Biomol Screen 2011;16(5):467-75.

14. Soga T. Cancer metabolism: Key players in metabolic reprogramming. Cancer Sci 2013;104(3):275-81.


Financial Disclosure

DescriptionY/NSource
Grantsno
Salaryno
Board Memberno
Stockno
Expensesno

IP Royalty: no

Planning to mention or discuss specific products or technology of the company(ies) listed above:

no