Single Cell & Cancer Heterogeneity: CyTOF, CODEX & Ovarian Cancer

Brian McCloskey and Brad Power July 13, 2022 “We've had DNA and RNA seq. Those are all great proxies for proteins. I don't want to monkey around anymore. ” – Brian McCloskey “You need somebody that's really, really committed to understanding how new technologies can work.

Wendy Fantl, PhD, Assistant Professor, Urology, Stanford Medicine, led a discussion on "Single cell and cancer heterogeneity analysis". She gave a basic introduction to two single cell technologies she works with: "Cytometry by time of flight" ("CyTOF") and "CODEX" (Co- detection by indexing) multiplex imaging.

They provide important information at a single cell level about protein expression, using antibodies to identify cells based on the types of antigens or markers on the surface of the cells. Why are these single cell technologies important? Tumors are a mass of heterogeneous cell types.

Single cell proteomic technologies allow you to unravel tumor heterogeneity, to peer inside and identify key cell types in the tumor, their numbers, and how they are arranged spatially into neighborhoods. They can help characterize immune function and cancers, and understand cells' response to therapy.

You can identify minority cell populations that are the bad guys, that have key roles in tumor initiation, maintenance, cell migration, and drug resistance. For example, Wendy shared research that looked at blood collected from men with metastatic prostate cancer and found natural killer cells that didn't kill anymore. With CyTOF, you could measure populations of these distinctive natural killer cells to monitor disease progression.

Since CyTOF looks at cells in suspension (usually blood), you lose information about which cell is next to which cell. That's important, and that brings up seeing cellular neighborhoods using CODEX multiplex imaging. For example, you can see one neighborhood that has tumor cells, or another that has immune cells.

A ”hot tumor”, which is more likely to respond to immunotherapies, has immune cell infiltration into the tumor cells in a neighborhood, while a “cold tumor” does not. Brian McCloskey asked how he could have a conversation with his doctor about using these technologies in his treatment decision-making process. “We've had DNA and RNA seq. I don't want to monkey around anymore.

” Wendy responded: “You need somebody that's really, really committed to understanding how new technologies can work. You need really good physician scientists. They need to really get it. Sadly, a lot of physicians don't have time. I am a former patient. I had ovarian cancer. I fired , PhD) Wendy responded: “You need somebody that's really, really committed to understanding how new technologies can work. They need to really get it.

I am a former patient. I had ovarian cancer. I fired so many doctors you don't even want to know. I told them: ‘I am the patient from hell. I am going to ask you questions, I am the boss of me. ’ I did find somebody who was amazing. She's one of my best friends, and we are collaborating.

” The information and opinions expressed on this website or platform, or during discussions and presentations (both verbal and written) are not intended as health care recommendations or medical advice by Cancer Patient Lab/Prostate Cancer Lab, its principals, presenters, participants, or representatives for any medical treatment, product, or course of action.

You should always consult a doctor about your specific situation before pursuing any health care program, treatment, product or other course of action that might affect your health.

SUMMARY KEYWORDS cell, tumor, antibodies, proteins, ovarian cancer, question, samples, patient, technology, tumor cell, imaging, nk cells, tissue, important, papers, oligo, immune cells, cancer, neighborhood, assays SPEAKERS Wendy Fantl, Anonymous Caregiver, Dr. John Laird, MD, Brian McCloskey, Brad Power, Rick Stanton. Wendy Fantl 00:03 My lab is focused on applying single cell proteomic technologies to expedite patient benefit.

We're all about translation. The focus areas of my lab are ovarian cancer and kidney cancer. But all this work can be applied to any malignancy. There are two multiplex single cell proteomic technologies up and running in my lab: mass cytometry or cytometry by time of flight (“CyTOF”), and CODEX multiplex imaging.

(Mass cytometry labels antibodies with isotopes of rare-earth metals, resulting in a theoretical capacity to distinguish up to 100 unique characteristics of cells simultaneously. ) I’m also going to talk about the quality of clinical samples. Wendy Fantl 02:04 What is the motivation for developing multiplex single cell technologies? These multiplex single cell proteomic technologies allow you to deeply phenotype what kind of cell it is.

Is it a tumor cell? Is it a blood cell? What is it? They allow you to unravel tumor heterogeneity. And why is that important? If you took a tumor and you processed it, just mushed it up, you get sort of average values of transcripts and proteins. But because you're analyzing or measuring what's going on in single, intact cells, you can identify minority cell populations and those are the populations that you're interested in.

They have key roles in tumor initiation, maintenance, epithelial-mesenchymal transition (a process by which epithelial cells – cells that come from surfaces of your body, such as your skin, blood vessels, urinary tract, or organs – lose their cell polarity and cell–cell adhesion, and gain migratory and invasive properties to become mesenchymal stem cells – multipotent stem cells found in bone marrow that are important for making and repairing skeletal tissues, such as cartilage, bone and the fat found in bone marrow), and drug resistance.

Those are the bad guys. Those are the guys we need to understand. To address this need, about over a decade ago mass cytometry was developed. It characterizes single, intact cells in suspension, and it's based on protein co-expression patterns. I'm not going to talk about transcripts or DNA. This is all about proteins, which really are the functional convergence point from DNA, to RNA, and to protein.

You can include RNA in the assays, and you can look at protein-to-protein interactions. But CyTOF, or mass cytometry, depends on cells in suspension, and so therefore, you lose any structural information. That was the motivation for the development of multiplex imaging technologies, and several are available. Code detection by indexing, or CODEX, is the technology I have in my lab.

There are other technologies, such as Multiplexed Ion Beam Imaging (MIBI) and Cyclic Immunofluorescence (CycIF) imaging, a whole slew of them. But essentially, they're all doing the same thing. They're good at providing information about tissue architecture and allowing you to identify cellular neighborhoods. Samples of the highest quality are essential. There are many variables to consider when collecting samples, and the devil is in the details.

CyTOF cells in suspension means that blood is the perfect medium. People generate peripheral blood mononuclear cells from whole blood. But what I'm using is a buffer that allows you at the clinic, at the site, not to do any of the peripheral blood mononuclear side preparation, just add this buffer to incubate and freeze it. That's it.

And what's really cool about that is that with peripheral blood mononuclear cells, you don't have any granulocytes. Granulocytes are also important in cancer. Smart2 buffer allows you to preserve granulocytes. For other CyTOF analysis, blood is perfect, but it also works with solid tumors.

I think my lab was one of the first to look at solid image ovarian tumors and to work out in excruciating detail all the protocols and details of dissociation into high quality single cells. CODEX doesn't use cells in suspension, but it uses tissue slices. These can be fresh-frozen or formalin-fixed paraffin embedded (FFPE). And again, the devil is in the details.

Everybody is very sort of "hoo ha" about fresh-frozen paraffin embedded samples because they're available worldwide, so it shouldn't be a problem. But I can tell you, there is such variation in the preparation. Here are some of the papers and some of them are from my lab and others. They go into the details about producing high quality samples. I think this is very important for this group.

Maybe we could have a separate meeting about how to request samples and the quality control that we need to have working with the collaborators and the people providing the samples. Wendy Fantl 06:53 With flow cytometry, which is reused routinely, the technology is based on fluorescence and looking at fluorescence emission spectra. What you can see here from this diagram is that there's tremendous overlap.

That makes it very difficult to measure if you're a routine practitioner. There are six to eight experts in the field, but they're few and far between and they're swamped. Rick Stanton 07:28 Wendy, I don't think everyone knows how to associate these wavelengths with any problem, or what are these wavelengths? Rick Stanton 07:43 I mean, I know flow cytometry, but I don't think other people here do.

Does that clear it up? Rick Stanton 08:30 Fluorophores with these emission spectra, like the red and the green on the right, show there's overlap. If something was emitting, it would be between 600 and 100. You don't really know whether it was a red or a green, and that red and the green are bound selectively to different antibodies. Without some kind of deconvolution algorithm which is trying to optimize red from green, you don't know.

That's the meaning of this overlap - there's an ambiguity of whether I am looking at a fluorophore that's tagged to this antibody, or that antibody. That is going to make a difference in the cell type in your conclusions because you're querying with these antibodies tagged to these fluorophores. If you don't know where they're coming from, you lose your ability to interpret. Wendy Fantl 09:35 Absolutely.

But I would say the algorithms are actually very good at untangling that. In any case and to address the problem, enter Scott Turner. Scott Turner was a professor at the University of Toronto who is no longer there, but he reasoned that he would label antibodies with rare metals. Those metals are rare or absent in biology and they're at the bottom of the periodic table. Here they are - the lanthanides.

They're rare or absent in biology and what's cool is they are composed on the planet when they mine stable metal isotopes. There are ways to purify these stable metal isotopes. Now you can label antibodies with these metals, and you can get rid of the overlap that you've just so beautifully explained. The number of parameters per single cell is just dependent on what Earth provides us, but maybe we'll go to Mars and get more, but I don't know.

For example, you see that samarium, or Sm, has 6, a reasonable abundance. We label our antibodies with an isotopically-enriched lanthanide. Right now, we're up to about 60 parameters per single cell. We label ourselves with antibodies against proteins that delineate surface markers. If it was a T cell, the tumor cell, the cell type, and antibodies, we permeabilized the cells, and we could go inside the cell and look at what's inside the cell.

It's incredibly powerful. We've got the samples labeled, and this is a picture of their CyTOF instrument. So now you're introducing into the CyTOF hundreds of 1000s, if not millions of cells. For each cell, you get a readout of the amount of protein. This has really spawned the computational folks to develop all sorts of algorithms that take a lot of the intuition from RNA expression. These are just some of them.

You don't need to know the specifics. It's just the power of this that is just fantastic. Because this is publicly available, these are the panels that we made in my lab, and there are plenty, plenty more that we made ourselves – the conjugations. They're commercially available, and there are antibodies against immune cells. It'd be like the checkpoint proteins that you're familiar with: T cells, NK cells, natural killer cells, macrophages, etc.

against tumor cells. If you need anything else in the panel, and it's not available commercially, but there's a good antibody, we can conjugate it and include it in the CODEX panel. With prostate cancer, if there aren't any CyTOF antibodies metal conjugated, then they can certainly be made. Here are a few examples of some results from our publications. This is data from untreated ovarian tumors.

We identified a minority cell type with metastatic features. The more you have, the worse time to relapse. It's much, much shorter. We would have never found this any other way. And then another example, we were analyzing the intra-tumoral immune cells and found this cell type. This cell type turns out to be decidual-like. ) Why is that? Because this CD9 protein is a hallmark of decidual NK cells.

These NK cells, natural killer cells, are found in about 70% of lymphocytes in the first trimester of pregnancy. And what do they do? They prevent the mother rejecting her fetus. We found that in ovarian cancer, even though they have exhausted T-cells, ovarian cancer patients have had very poor response to checkpoint immune blockade.

Our working hypothesis is that these decidual-like NK cells could be overriding any effect from the immune checkpoint blockade.