From picking out cancerous tissue areas during diagnosis, to uncovering novel biomarkers of early detection, mass spectrometry (MS) has become an indispensable analytical technique in cancer research. In the past few years, we have seen significant advances in MS instrumentation that allow for more sensitive, reproducible, and rapid analysis.
These developments have led scientists to couple mass spectrometry to progressive surgical processes in the cancer clinic. Recently, devices like the iKnife and MasSpec Pen have created quite a buzz with their ability to provide quick, real-time analysis of tissue sections during surgery.
Knives and pens transforming cancer diagnosisThey say the pen is mightier than the sword, but modern research has shown that both “knives” and “pens” have proven equally useful when it comes to cancer diagnosis and surgical margin evaluation. The iKnife, a hand-held sampling device developed by scientists at Imperial College in London, uses electrospray ionization (ESI) and mass spectrometry detection to distinguish in real time which tissues are cancerous, and which are not.1
The iKnife works by capturing the “smoke” released during tissue dissection and analyzes the different molecules in each puff by their masses. (Figure 1) This biologically-rich smoke is generated because modern surgery uses “energy devices” that immediately seal tissues by electrocauterization (and hence are favored over blades). The vapor from the electrocauterization is aspirated into a mass spectrometer and analyzed using a technology called rapid evaporative ionization mass spectrometry (REIMS). Initially invented by Prof. Zoltan Takáts of Imperial College, London, and published in Science Translational Medicine,1 the REIMS Research System does not require sample preparation or chromatographic separation, as is often the case with complex biological samples.
At the data analysis stage, a software called Progenesis QI is used to mark similarities and differences between samples, and group them in an objective and unbiased manner. The molecules associated with cancer are identified by applying machine learning techniques to rapidly search large databases of previously validated spectra. The speed of the MS characterization and deep computing tools enable surgeons to know within a matter of seconds if the tissue they are cutting open is cancerous.
Figure 1: How the iKnife works. Credit: Cancer Research UK
iKnife now in clinical trialsThe iKnife-REIMS analysis reveals a lot about genetic and molecular functions of specific cancers. As such, the system could be used to guide medical treatment, in addition to surgical decisions. So far, the technology has only been used in clinical trials such as the current ongoing trial in breast cancer at Charing Cross Hospital.
In a recent study of ovarian cancer tissue, published last year2, samples collected during surgery were tested with the iKnife and compared with pathology reports. Results showed that the iKnife was able to distinguish ovarian cancer from normal, non-cancerous tissue with 100 % accuracy.2 Armed with these results, Imperial College researchers are now looking to run a clinical trial aimed at ovarian cancer detection with the iKnife in a surgical setting.
MasSpec Pen keeps pushing forward
The MasSpec Pen is another hand-held device that is being developed by researchers at the University of Texas - Austin, USA, to rapidly distinguish tumors from healthy tissue during surgery. In the case of the “pen”, the molecular profile of a tissue is obtained from a small droplet of water containing molecules extracted from tissue.3 By depositing a single water droplet onto a tissue sample for a determined amount of time, biomolecules are efficiently extracted, while tissue integrity is preserved. Following several seconds of contact with the tissue surface, the MasSpec Pen collects the water droplet and sends it straight to a mass spectrometer which is wheeled into the surgery room. Then, ambient ionization mass spectrometry is used to characterize diagnostic molecules like proteins, lipids and other metabolites. A diagnosis of the health of the tissue is then generated using machine learning algorithms.
Similar to the iKnife, the MasSpec Pen is currently being studied to understand the sensitivity, specificity, and overall accuracy of the device on various cancer types (e.g. breast, lung, thyroid, and ovarian cancers). Just last month, the research team from UT Austin published data on the utility of the MasSpec Pen in rapid diagnosis of ovarian cancer.3 A total of 192 tissue samples were tested using a linear ion trap mass spectrometer. They were then compared with pathologist evaluations. Results showed that the MasSpec Pen was capable of detecting and differentiating ovarian cancer from normal healthy tissue with very high sensitivity and specificity when using the ion trap mass spectrometer.4
Targeted mass spec approaches are leading the way in biomarker discoveryThroughout the United States (US), dozens of research labs investigating cancer biomarkers and early detection mechanisms are participating in a national effort to accelerate the understanding of the molecular basis of cancer through proteogenomics. Two of the National Cancer Institute-funded research programs in the US heavily involved in proteogenomic research are the Clinical Proteomic Tumor Analysis Consortium (CPTAC), and the Early Detection Research Network (EDRN).
According to Dr. Amanda Paulovich, Aven Foundation Endowed Chair at the Fred Hutchinson Cancer Center in Seattle, Washington, USA, CPTAC has led efforts to develop and distribute targeted mass spectrometry-based assays and to build a database of proteogenomic analyses of a variety of tumor types, with the aim of enhancing diagnosis and treatment of cancer. “The teams in the EDRN have benefited from the work done by CPTAC,” remarked Paulovich, whose is a member of both the CPTAC and EDRN networks.
The consortium’s online assay portal and antibody portal allow researchers access to highly-characterized assays, along with SOPs, reagents, and assay characterization/ validation data. The methods on the portal are rigorously validated according to CPTAC assay characterization guidelines to ensure reliability and reproducibility between experiments and even labs.
“In the last few years, targeted mass spectrometric analyses have really ramped up cancer biomarker research,” said Prof. Michael Lewis of the Lester and Sue Smith Breast Center, Baylor College of Medicine, Houston, TX, USA. Lewis’ lab is part of a National Cancer Institute-designated Biomarker Development initiative within the EDRN network, together with Paulovich’s lab at the Fred Hutchinson Cancer Center. “We’ve been using quantitative MS-based targeted assays because they work better in various aspects to conventional protein detection systems, or are highly complementary. For example, these targeted assays allow for robust multiplexing and provide high specificity when detecting lower abundance proteins. One can, not only evaluate the quantity of an analyte, but can also find out the precise identity of the protein being evaluated.”
Typically, proteomics-based biomarker development consists of two main phases – the discovery phase and the validation phase. “In the discovery phase, we try to identify almost all the peptides or proteins in a sample that may be involved in the disease,” Lewis noted. “Then this initial list of candidate biomarkers must be validated, and that’s where targeted MS assays play a huge role.”
In the validation phase, the initial biomarker panel is tested across a large clinical cohort. At this point, multiplexed analysis becomes vital and hence targeted MS provides a great solution. “Our use of mass spec approaches runs the spectrum from label-free methods of protein profiling in samples to tandem mass tagging for discovery proteomics to accurate inclusion mass screening (AIMS) and multiple reaction monitoring (MRM; also known as selected reaction monitoring or SRM) for evaluation of selected proteins,” said Lewis. “Choice of method depends mainly on whether we are trying to identify as many different proteins as possible in a given sample, or whether we are trying to identify and quantify a subset of proteins.”
MRM is normally performed with a triple-quadrupole mass spectrometer, where certain precursor peptide ions can be picked out as candidates to represent their respective protein. The first quadrupole mass filters these precursor ions, and the second quadrupole fragments them into product ions, which are then guided through the third quadrupole to the ion detector. The precursor-product ion combination is known as a “transition pair”, and researchers can program the mass spectrometer to give a signal only when certain transition pairs are present.
In 2011, Paulovich and her colleagues published a key study that investigated the analytical performance of targeted mass spectrometry in prioritizing and verifying candidate biomarkers from a large pool of previously identified candidates.4 Her team used a data-dependent triage process to prioritize plasma biomarkers that were among some 1000 candidates detected in a mouse model of breast cancer. “It’s easy to find candidate biomarkers using high-throughput omics technologies – but a high rate of discovery doesn’t mean a high rate of validation,” said Paulovich. “Developing immunoassays for all these candidates is a very expensive and time-consuming task. So, we wanted to find a scientific approach that could be used downstream to validate hundreds of candidates at the same time. Targeted modes of mass spectrometry allowed us to do this, enabling us to weed through the false positives that came from discovery experiments and retain biomarker candidates of value.”
Paulovich’s team developed, multiplexed, and evaluated eighty-eight novel quantitative assays based on MRM. As a result of these efforts, thirty-six proteins were verified as being elevated in the plasma of tumor-bearing mice.5
“The goal of this proof-of-concept study was not to find new breast cancer biomarkers, but to test the utility of the mass spec platform for biomarker triage and validation,” explained Paulovich. “So, we picked a well-characterized, highly-controlled animal model without much biological variation to see if the platform was good enough to find and verify true biomarkers.”
The success of this feasibility study helped Paulovich and Lewis obtain a grant from the National Cancer Institute to continue using the technology to find biomarkers of human breast cancer.
It is clear that mass spec-based studies in cancer biomarker discovery and development have headed in the direction of targeted assays. According to Lewis, there are at least two places where significant advances could be made in this regard: “One is the up-front high and medium abundance protein depletion and fractionation of samples to allow detection of proteins expressed at lower levels, particularly in plasma. The second is in the sensitivity of detection in the mass spec itself, which is already progressing at a very rapid pace such that instruments are eclipsed by newer more sensitive instruments in very short order.”
- Balog, J., Sasi-Szabó, L., Kinross, J., Takáts, Z. et al. Intraoperative Tissue Identification Using Rapid Evaporative Ionization Mass Spectrometry, Sci. Transl. Med. 2013, 5 pp.194ra193.
- Phelps, D. L., Balog, J., Gildea, L. F., Takáts, Z. et al. The surgical intelligent knife distinguishes normal, borderline and malignant gynaecological tissues using rapid evaporative ionisation mass spectrometry (REIMS), Br. J. Cancer. 2018, 118, pp. 1349-1358.
- Zhang, J., Rector, J., Lin, J. Q., Young, J. H. et al. Nondestructive tissue analysis for ex vivo and in vivo cancer diagnosis using a handheld mass spectrometry system. Sci. Transl. Med. 2017, 9, 406, pp. 3968.
- Sans, M., Zhang, J., Lin, J. Q., L. S. et al. Performance of the MasSpec Pen for Rapid Diagnosis of Ovarian Cancer. Clin. Chem., 2019, clinchem.2018.299289.
- Whiteaker, J. R., Lin, C., Kennedy, J., Paulovich, A. G. et al. A targeted proteomics–based pipeline for verification of biomarkers in plasma, Nat. Biotechnol., 2011, 29, pp. 625-634.