Hello, my name is Rusica Lozova, and today it's my pleasure to discuss the diagnostic utility of mass spectrometry imaging in melanocytic lesions. I have no conflicts of interest to disclose. Proteins have an extremely important role in our bodies. They are the main carriers of biologic activity and have fundamental effect on cellular function and dysfunction. The genome and genetic alterations play important role in carcinogenesis. Genes, however, may or may not be expressed, and gene expression does not always correlate with protein translation and does not account for post-translational modifications. Therefore, measurement of gene activity at the protein level is more informative. Mass spectrometry imaging is a method that is capable of measuring larger molecules, such as peptides, directly can analyze tissue sections for the relative abundance and spatial distribution of those peptides. Here is a schematic of how the method works. On the left-hand side, there is tissue, which can be a whole tissue section, 5 micron thick, or a micro-dissected tissue placed on a specially charged slide. Matrix is deposited on the tissue section to hold it to the slide. When the section is hit with a laser, which evaporates practically the molecules, they fly through a chamber and mass spectra are obtained. Here is the tissue named MOLDI target on this slide. MOLDI stands for Matrix Assisted Laser Dissorption Ionization, Time-of-Flight Mass Spectrometer. What happens? The laser hits the tissue. The molecules fly through a chamber, and depending on their masses and charges, they fly with a different speed, and at the end, mass spectra are obtained. Here again, another schematic of time-of-flight mass spectrometry. On the top, we see the chamber. Depending on the masses or sizes of the molecules, the lighter molecules travel faster through the chamber, and the heavier molecules travel slower. At the end, the mz ratio is collected, mz ratio is mass-to-charge ratio, but all charges are 1+, therefore these are pure masses. In other words, if a molecule is smaller and lighter, it travels faster through the chamber, and that mass is reflected at the end as a result of collection of the mass spectra. Mass spectrometry can also be used on formalin-fixed paraffin-embedded tissue samples. Here is how it works. In the left upper corner is a slide with a 5-micron-thick tissue sample placed on this charged slide for mass spectrometry. Next to it is an H&E stained section with annotated areas by the pathologist to be studied by mass spectrometry. The two images are merged using Photoshop, and the areas of annotations are transferred to the unstained section. Trypsin and matrix are applied, and mass spectra are obtained from the precise areas of interest. Here is a short video that shows how the mass spectrometer goes to all annotated areas on the slide and obtains mass spectra. It is amazing what the accuracy of the mass spectrometer is going to the precise areas of interest. On the left hand side we have two samples, a sample of spitzweed melanoma and one of spitznivas that are annotated for mass spectrometry. From these areas of interest, mass spectra are obtained, and on the right we have a schematic showing three different peaks that were differentially expressed in the two lesions. The peaks in red are much higher in the malignant spitzweed melanoma, whereas the peak in black is much more expressed in benign spitzniva. How can we use mass spectrometry in our everyday life when we try to make a diagnosis and differentiate between spitzniva and spitzweed melanoma? We use mass spectrometry to study spitzniva and spitzweed melanomas, and to determine whether this method can help in their differentiation. We published our study at the American Journal of Dermatopathology, titled Imaging Mass Spectrometry, a new and promising method to differentiate spitzniva from spitzweed melanomas. We were able to determine a molecular signature which consisted of five peaks with the mz was listed below. These five peaks were able to discriminate between spitzniva and spitzweed malignant melanomas because they were differentially expressed in both entities. We concluded with that study that spitzniva and spitzweed melanoma may be successfully distinguished using mass spectrometry imaging analysis based on detection of proteomic differences. Our molecular signature consisted of five peaks. Mass spectrometry imaging is a valuable adjunct to histopathologic evaluation of spitzweed melanocytic lesions. It is objective, fast, and affordable. It is very easy to diagnose spitzniva and spitzweed melanomas if they have the classic histopathologic features and by using the criteria that we use in our everyday practice. However, there are many, many lesions that do not follow these criteria and present with mixed histopathologic features. We call these lesions atypical spitzweed neoplasms because we cannot precisely classify them as benign or malignant. So our next step was to see whether mass spectrometry analysis can help in those difficult cases. We called for international collaboration and in this study we were able to include 12 countries from four continents. There were 12 centers from the U.S. alone that contributed cases to our study. Two hundred and fifty-two cases were submitted of which a hundred and two included. One very important criterion for inclusion in the study is that the cases of atypical spitzweed neoplasms had a clinical follow-up. We divided the cases into four groups. Group one with best clinical behavior and group four with the worst clinical behavior. At the end, we correlated the clinical outcome with the mass spectrometry imaging diagnosis and compared that to the histopathologic diagnosis. These are the results of our study in a table format. These 26 cases that we see on this slide were all atypical spitzweed neoplasms, but all of them were favored to represent spitzweed melanomas on histology. You can see the column titled histology DX, histology diagnosis, in red. They were all classified or favored to represent spitzweed melanomas. Ten of them were also diagnosed as spitzweed melanomas based on mass spectrometry studies, the column next to the histology DX. One of these cases represented a patient who was dead of disease with metastases throughout the body. Sixteen of these cases that were diagnosed as spitzweed melanoma on histology were diagnosed as spitznevas on mass spectrometry. One of them, number 17, had a 14-year follow-up. This patient had one positive sentinel lymph node biopsy, but was well and without evidence of disease 14 years after the diagnosis. Another patient had a positive sentinel lymph node with one cell in it. This is patient number 23, four-year follow-up, no evidence of disease. This is the group in which histology, based on histologic diagnosis, spitznevas or spitzweed melanoma could not be favored. In other words, the diagnosis stayed clearly in the middle as atypical spitzweed neoplasm. All these cases were diagnosed as spitzneva on mass spectrometry. Only one of these cases had a positive sentinel lymph node, but six years follow-up with no evidence of disease. And then additional 18 cases were favored to represent spitzneva on histology and were diagnosed as spitzneva on mass spectrometry. We concluded with this study that mass spectrometry imaging diagnosis of atypical spitzweed neoplasms showed stronger association with clinical outcome than did histopathologic diagnosis. The diagnosis of spitzweed malignant melanoma by mass spectrometry imaging was statistically strongly associated with aggressive clinical behavior. MSI can help in the diagnosis of ambiguous cases by histology. Mass spectrometry imaging analysis using a proteomic signature can provide reliable diagnosis as well as clinically useful and statistically significant risk assessment for atypical spitzweed neoplasms. Our study titled Imaging Mass Spectrometry Assists in the Classification of Diagnostically Challenging Atypical Spitzweed Neoplasms was published in the Journal of the American Academy of Dermatology in 2016. Let me give you an example of an interesting case in which mass spectrometry was very helpful. This is the biopsy of a 50-year-old woman from the right posterior thigh. This lesion was present for a few months, so new lesion. And here are the histologic images. Nests, but in the lower portion of the dermis, sheet-like proliferation of large and atypical melanocytes with vesicular nuclei, abundant pale cytoplasm mitotic figures, and very irregular base, no maturation of the melanocytes with their descent into the dermis, and occasional deep mitotic figures as the one represented here on the slide. So of course, the histopathologic diagnosis was spitzweed melanoma. The depth was 4.75 millimeters. There was one mitosis per square millimeter. One sentinel lymph node out of three was positive. There was no residual lesion on reexcision, and completion lymphodymnectomy had negative lymph nodes. We analyzed this case with mass spectrometry, and the diagnosis was spitznevis. There is five-year follow-up for this patient. She is alive with no evidence of disease. And don't forget, the depth was 4.75 millimeters. We analyzed this case for copy number alterations, part of another study. And it turned out that this case had an 11p amplification and an H-RAS mutation. So most likely, it was a spitznevis. Mass spectrometry imaging in the diagnosis of all melanocytic lesions. Why not? In this larger study, we included different types of melanomas, particularly superficial spreading, nodular spitzweed, acral lentiginous, lentigum malignant, metastatic melanoma, desmoplastic, and nevoid. In addition, we included different types of nevi, banal or conventional nevi, acral nevi, and spitznevi. In the training set, we had 100 cases of melanoma and 111 of nevi. We developed the algorithm, or the linear discriminant analysis classification algorithm, which was generated on the training set and then validated in the validation set, which included 257 melanomas and 288 nevi. The validation spectral accuracy was close to 100% for malignant melanoma and 99% for benign nevi. Here is one of the peaks that was differentiating benign from malignant. The MZ ratio is 1198.8. Again, these peaks appear to be differentially expressed in benign or malignant lesions. This is a visual representation of a core microarray. On the left is a microarray with core biopsies of nevi, and on the right, microarray of core biopsies of melanomas. We color-coded the mass spectrometry results. So if the mass spectrometry classification was benign, it appears as green. If it's malignant, it appears as red. As you can see on the left, in the microarray of nevi, all samples are classified correctly as benign, green nevi. On the right, it's a microarray of melanomas, and all samples are red, correctly classified by mass spectrometry as malignant. We developed a histogram of histopathology-guided mass spectrometry protein profile scores. In this histogram, on the left are the benign lesions in green, and on the right are the malignant lesions in red. If a case obtained a score between minus 10 and minus 0.371, it was classified as benign. If the score was between plus 0.371 to plus 10, it was classified as malignant. That's the red color. You can see a very narrow zone of indeterminate lesions, for which the score was between minus 0.37 and plus 0.37, in which mass spectrometry could not, with absolute certainty, define the lesion as benign or malignant. Our sensitivity for the method of classifying melanomas and nevi correctly was 98.76%, specificity 99.7%, and overall accuracy 99.2%. Our conclusions were that mass spectrometry imaging is a reliable method to differentiate benign nevi from malignant melanoma based on proteomic differences. Mass spectrometry imaging can help in the diagnosis of histologically challenging cases and can accurately predict clinical behavior. Of course, future work to expand on the gray area cases and validate peptide identifications is needed. Thank you very much for your attention.

mass spectrometry imaging

melanocytic lesions

proteins

genetic alterations

protein translation