- [Resource Hub](/)
- [Resource](/?resource-type=webinar#library)
- Mastering Multiplex Immunofluorescence: Easy Strategies for Reliable Spatial Biology Results

# Mastering Multiplex Immunofluorescence: Easy Strategies for Reliable Spatial Biology Results

RareCyte’s Tad George and the La Jolla Institute for Immunology’s Simon Goldstein share the sample-prep, antibody-validation, and panel-design habits behind reliable single-round multiplex immunofluorescence on the Orion™ platform.

Reliable multiplex immunofluorescence (IF) depends less on the imager than on the habits around it: fix the tissue well, validate every antibody clone before it enters a panel, and match each marker to a channel that suits its brightness. In this RareCyte webinar, Simon Goldstein of the La Jolla Institute for Immunology (LJI) shares the sample-preparation, antibody-conjugation, and panel-design strategies his microscopy and histology core uses to get clean, quantifiable spatial results on the Orion platform, while RareCyte’s Tad George frames where single-round, 20-channel imaging fits in translational and clinical spatial biology. Orion stains and images up to 18 biomarkers in one round; reaching higher plex is a separate, optional cycling step.

In this video:

- George frames the trade-off Orion targets: microenvironment detail needs many markers, cohorts need throughput, met by 20 channels imaged in one staining-and-imaging round.

- Goldstein on why LJI's core chose single-round over low-plex or cyclic multiplex IF: less autofluorescence and cross-reactivity, fewer channel limits, no cyclic tissue damage.

- Histology first: a high fixative-to-tissue ratio and prompt fixation, plus archival-block rescues: high-adhesion slides, a 37°C slide warmer, gentler antigen-retrieval buffer.

- Antibody work: source IHC-paraffin-validated clones, screen one concentration across exposures, then simple amine-based ArgoFluor™ conjugation failing on one of ~100 antibodies.

- Panel design: map autofluorescence first, then assign low-abundance markers (FOXP3) to the brightest and abundant ones (CD3) to the dimmest, unmixed by a flow-cytometry matrix.

- Q&A: same-cell overlap, 16-bit dynamic range, “smart cycling” to add plex, stain vs scan time, controls from pellets to knockouts, and picking markers back from the analysis.

Full transcript

Moderator: Welcome everybody to today’s webinar, “Mastering Multiplex Immunofluorescence: Easy Strategies for Reliable Spatial Biology Results.” Today’s speakers are Simon Goldstein from the La Jolla Institute for Immunology and Tad George from RareCyte. Tad.

Tad George (RareCyte): All right, thanks, Rob. So again, welcome. I’m going to give a brief overview of the Orion approach to multiplexing, and we’re really excited to have Simon talk about how they’ve used the system and how they’ve worked through the specific challenges of multiplex immunofluorescence at La Jolla. They’ve done a really nice job. From a big-picture standpoint, Orion is essentially a tool deployed across the discovery, translational, and clinical space in spatial biology, and it’s therapeutically oriented. Modern therapies are designed to alter microenvironments. At a high level, that means recruiting the right cells in the right state to improve patient outcomes. When you’re in that translational and clinical space, you want to measure these things with confidence, and that requires two things at once. The biomarker panels need to be large enough to resolve those microenvironments, but at the same time you need sample throughput to analyze cohorts for statistical power. That’s where Orion fits nicely, because before Orion, or if you don’t have Orion, you’re basically forced to choose one or the other.

If you think about biomarker panel size on the y-axis and study throughput on the x-axis, the traditional way of getting enough information on a microenvironment has been to run cyclic immunofluorescence, because of spectral overlap. Those are great tools that give you a tremendous amount of information, but they’re throughput-limited for the statistical significance you need. There are also plenty of high-throughput tools, but they typically carry insufficient information, like IHC or low-plex multiplex assays. What Orion does is fill that need in the clinical and translational zone. We like to say it gives you the essential information with statistically significant studies.

The major technological breakthrough is that we can measure 20 channels of fluorescence in a single staining and imaging round. That translates into a couple of things. The big one it unlocks is throughput: it’s fast, but, as Simon will talk about, it also has very low run cost and high data quality, which is critical. The tissue is preserved, and it’s whole-specimen, so you can image regions of interest or the whole specimen for complete spatial context. In all of these spaces, flexible panel design is really necessary. This year we’re also releasing a high-capacity system with a 30-slide loader, which I’ll touch on. It lets you operationalize large cohort studies and accommodate multi-user environments like La Jolla has. The system works for essentially any tissue and any indication, 20 channels in a single round.

I’d encourage you to visit our website, in particular the tissue atlas page. You can click on any of these examples; this one is a whole-slide mouse ileum, for instance, at very high resolution. The other nice thing about the single-round approach is that you can do a same-section H&E, so in this panel we didn’t really stain for muscle, but you can see the muscle and goblet cells nicely in the same-section H&E. It enables really nice pathology-oriented workflows.

In terms of building panels, we use antibodies directly conjugated to ArgoFluor dyes, and they’re all IHC-validated. I should probably update this slide, because we now have over 175 biomarkers available in the catalog. We also offer off-the-shelf panels, dye kits, and services for custom biomarkers and panels, plus a panel-design tool to configure custom panels. One nice thing is that even in the off-the-shelf panels, all the reagents are supplied in individual tubes, like flow cytometry. So if you’re interested in diabetes, for example, you can take a panel, and if you’re not interested in CD20 you can take it out, put in DC-LAMP, and submit an order.

Developing panels is very straightforward. If you’re starting from scratch, you select biomarkers with the panel designer; if we have them all in the catalog, the designer will assign the channels, build an order, and you verify panel performance through a single titration on the intended tissue type. For any customs you have, you validate the clone against IHC, label it with the ArgoFluor dye, and verify that the immunofluorescence pattern looks like the IHC. We do lots of these in house; it usually takes us a couple of weeks to validate up to four customs.

The testing workflow is designed for high-volume or multi-user settings, with the staining and imaging performed in parallel. That lets you stain samples as they come in while you’re scanning material from last week. The scanning automation delivers what we call a continuous stream of quantitative results, because while it’s imaging it’s also processing the images and doing quantitative analysis with templates, so you get results quickly. A few things to highlight: it’s whole-slide, over eight square centimeters of imaging area. You can stain dozens of slides per day. It’s really just a cocktail of direct conjugates, just like flow cytometry. What’s nice about these dyes is that they’re very photostable, so you can scan right away or bank slides for up to a year without any loss of signal, and schedule your staining and imaging whenever it’s convenient.

The system runs 20 channels in a single scan, with 24/7 unattended operation and up to 30-slide capacity. Once you’ve done a scan, you can also perform a same-slide H&E or an additional informed immunofluorescence round, so it’s fully compatible with cyclic staining if you want to go higher plex. The quantification is very simple: there’s a human-in-the-loop ROI annotation, where typically pathologists get involved looking at the bright-field image or the IF to identify regions of interest. You classify cell types and states using thresholds and Boolean logic, and most commonly people measure density in and around the lesion, then summarize across a whole cohort. We’re usually analyzing dozens of samples in a study, not one. So Orion is really an ecosystem for the translational and clinical part of spatial biology, with a single-round workflow, reagents validated for the system, low run costs, easy panel building, and strong onboarding and support. That’s a quick overview of Orion, and with that I want to hand off to Simon.

Simon Goldstein (La Jolla Institute for Immunology): Okay, thank you. I first want to introduce our core facility. I work at the La Jolla Institute for Immunology, and we are a joint microscopy and histology core facility. We have seven full-time employees with 102 years of combined experience, and we describe ourselves as an end-to-end core: we help with all stages of experiments, from experimental design all the way through sample prep, imaging, and analysis and interpretation. To give you a sense of our output, we assist with around 20 to 30 publications a year, and we’re co-authored on some of those. Last year we generated around 6,000 FFPE blocks and 10,000 slides. I also want to acknowledge our funding sources.

First, briefly, why we chose the Orion platform for our core. We had traditionally offered multiplex IF services through traditional multiplex IF, and we were facing multiple challenges. The first was autofluorescence: when you’re trying to max out these smaller, low-plex multiplex IF panels, autofluorescence becomes problematic because you already have such a limited number of channels. The second was cross-reactivity: we were using primaries from different hosts and then going in with secondaries, and it can be hard to find antibodies from different hosts that work well together. And third, as I said, the limited number of channels; the labs and companies we work with really just wanted more information. Orion helped us address all three. The autofluorescence is extracted from the channels it affects, which is nice. Like Tad mentioned, all Orion antibodies are conjugated, so there are no cross-reactivity issues, and you can go with antibodies from any host. And you can stain up to 18 biomarkers in one round, which we really liked.

Some other considerations: we help people with image analysis, so we wanted a system that generated really high-quality images, and specifically one that didn’t rely on cycling. There are a couple of reasons for that. One, you have a higher risk of tissue damage with cycling. Two, it can be complicated to decide the order in which you stain. Being able to stain all your markers in one go was really attractive to us. We also considered the time it takes technicians to do the staining and the overall cost. And we already had a long history of doing immunofluorescence at the core, so we didn’t want to reinvent the wheel; we wanted a system we could fold into the workflows we’d already established.

Here’s a quick look at how the protocol works. What’s really nice is that if I’m staining, say, 60 slides in a day, and half of them are a basic single-plex fluorescent stain and some are Orion slides, I can go through the initial process all together. If you’re familiar with traditional immunofluorescence, the workflow is pretty much identical. First you bake your slides, then go through deparaffinization and antigen retrieval. RareCyte has a nice autofluorescence-quenching protocol where you use LED panels, with the slides in a hydrogen peroxide buffer, plus a UV bleach. We then typically stain our primary antibodies overnight. It provides a better stain, and splitting the work into two days helps us from a core-facility perspective. The next day you go in with your secondaries and nuclear dye and coverslip.

One of our big considerations was low cost. I won’t get into all the details, but we use staining racks that let us cut down on reagent cost. What’s nice about Orion is that really the only cost is the antibodies themselves; there are no extra fancy parts that build up cost. Another advantage is the generous scanning area, with a lot of usable space to place tissue. We work with some very talented histologists, and to give you a sense of the room, we can comfortably fit around five mouse lungs on one slide. Being able to add multiple tissues from multiple blocks onto one slide lets us dramatically reduce the cost of the assay, which is very useful from a core perspective, because it lets labs actually afford these stains.

I also want to briefly highlight image quality. There’s a high dynamic range of signal, so you can see both dim cells and bright cells, and that becomes super important for analysis. This is an example from a human tonsil. Another attractive point is that the data output is very analysis-friendly: it generates an OME-TIFF, and you can use your preferred software. Our core happens to like using QuPath, and Orion output is very compatible with QuPath, which is nice. All the channels end up pre-labeled, so everything is easy to work with.

I want to step back and go into the basics of histology, because one of the pain points that comes up is working with difficult samples that were not properly handled. This is a general issue, unrelated to Orion; it’s a histology issue. You want your basics down if you want high-quality multiplex IF results. We’ve found the type of fixative isn’t super important; we haven’t noticed a huge difference across the most popular fixatives. Something that is important: don’t try to jam your tissue into cassettes if it doesn’t fit, or you can end up with tissue that looks squeezed, like a waffle. If it doesn’t fit, use a deeper cassette or trim your sample so it fits properly. You also want a high ratio of fixative to tissue; we recommend at least 20:1 fixative to tissue.

It’s also really important to get samples into fixative as soon as possible. If you place tissue in PBS or media and wait a couple of hours before fixing, you’re compromising the quality of the sample; we really don’t recommend that. You want to fix at minimum 24 hours and up to 72 hours at room temperature. Selecting the fixation container matters too. Please don’t fix tissue in small Eppendorf tubes, and avoid conical-bottom tubes, because the tissue sits at the bottom and the bottom of the tissue won’t fix as well as the top. We recommend urine cups or larger, wide-bottom containers, and rocking your tissue as it fixes so there’s good flow of fixative. If you notice the tissue is still pink on the inside or you see visible blood, it’s not fixed enough and you need to keep fixing. For very thick tissues, it can help to trim mid-fixation so all areas fix properly.

There are cases where you don’t control how tissue was fixed, say archival clinical samples or blocks a collaborator made for you; you have to make them work. We’ve learned some tips to rescue difficult samples that weren’t properly fixed. My first big recommendation is finding high-adhesion slides; we’ve had a lot of success with the TOMO brand. They hold onto tissue a little better, and for tissue that’s falling off the slide I highly recommend switching to a slide like that. Another basic but underappreciated technique: after your slide is fully dry, place it on a slide warmer at 37°C for 24 to 48 hours. The 37°C temperature helps the chemical reaction of the tissue binding to the slide, and the tissue stays on better.

Probably the most important thing we’ve noticed involves antigen retrieval. There are two classical antigen-retrieval buffers: a high-pH (pH 9) Tris-EDTA buffer, or a low-pH (pH 6) citrate buffer, and there’s a trade-off. Many people default to the high-pH Tris buffer because it typically gives a brighter stain; however, it’s much harsher on tissue. If you’re having sample-quality issues, I recommend switching to the low-pH citrate buffer. Just keep in mind that you need to redo your antibody optimizations and titrations for the buffer you’re using, because there are differences in signal intensity depending on the buffer. As an example, these are sequential slides: on the left, a normal slide with Tris-EDTA retrieval; on the right, a sequential slide using citrate. It makes a dramatic difference, and we were able to rescue this sample by changing the antigen-retrieval buffer. If you ever see a webby-looking effect, that’s not an Orion or staining artifact; it’s a sample-prep issue, and you want to troubleshoot ways to correct it. Here’s another example: this time citrate on the left and Tris on the right, in a mouse spleen.

Now I’ll talk about how I go about antibody validation when building these Orion panels. As Tad said, RareCyte has an extensive, great catalog of antibodies, so that’s really your first step, and chances are they’ll offer a lot of what you’re looking for. But inevitably there are markers they don’t offer, and in that case our core has moved to conjugating our own antibodies. One of the biggest and most important steps is finding good, IHC-validated clones to work with. A tool I’d recommend is the website CiteAb; it’s a great resource for finding IHC-validated clones that have been highly cited. It shows how many times they’ve been cited, and you can filter specifically for IHC-paraffin. That’s a common point of confusion: antibodies that work in frozen tissue often will not work in paraffin, so look specifically for clones that are highly cited and validated in IHC-paraffin. There are other resources too; a lot of papers test different clones. On the bottom right, it’s nice when you can find papers that show negative data, which is rare but great. In one project, in Rhesus macaque, we were going through our human antibody catalog trying to figure out which antibodies were cross-reactive and would work in the macaque tissue, and two papers provided negative data, listing clones that didn’t work, which saved us a lot of time and effort.

For the amine-based conjugation we do, it’s important that you look for antibodies that are carrier-free and don’t have BSA. A typical rule for us: if an antibody is recombinant, the company has spent time, effort, and resources on it, so it has a higher chance of working; a recombinant is a good green flag for us. And when we have a choice, we like to go with monoclonal over polyclonal antibodies, because it’s a little cleaner, you know exactly what you’re labeling, and there are fewer lot differences. We have used polyclonal antibodies as well, but if you have a choice, it’s better to opt for a monoclonal.

Here’s how I go about the validation process; we’ve developed a streamlined workflow. When we receive the antibody, I test it in both antigen-retrieval buffers to get a sense of how it works in each. We stain at a fixed concentration of 2 micrograms per milliliter with the primary antibody, then go in with a secondary. That concentration gives us a good yes-or-no on whether the antibody works for Orion, and we try not to squeeze further; if we don’t see signal at that concentration, it’s probably not a suitable antibody. We scan the image at three different exposure times, which lets us analyze the signal abundance, decide which channel to place the antibody in, and then proceed with the conjugation and titrate.

It’s really important, after you conjugate an antibody, to test that the conjugation was successful. I’ve conjugated probably over a hundred antibodies at this point, and I’ve only had one that didn’t like to be conjugated and failed, so it’s a very high success rate. But there are rare cases where, for whatever reason, the antibody doesn’t bind anymore after the amine conjugation, so don’t just assume it will work; test it and find an appropriate titration. This is also a stage where robust controls are extremely important. You don’t want to do all this work and find out you were looking at non-specific signal, so having good positive and negative control tissues is key. If you know one tissue has high expression of your protein and another has low expression, that’s ideal. If you don’t have such tissue, there are other options: you can overexpress your protein in a cell line, generate a cell pellet, add HistoGel, and create an FFPE block out of those cells. We’ve also started experimenting with protein gels, where we order the physical protein and incorporate it into gels and make blocks from that. Classic IgG controls are always useful, and knockout tissues are an extremely useful tool as well.

I want to stress that it’s very important to test and optimize your antibodies in your end tissue of interest, because certain targets are expressed at very different ranges depending on the tissue. Here’s an example: a multi-block with spleen, liver, and kidney, stained with Na/K-ATPase. In spleen it’s relatively dim, in liver it’s pretty bright, and in kidney it’s extremely bright. So if you’d done all your optimization in spleen and then applied it to kidney, it wouldn’t work out well, because you’d expect a much dimmer signal than kidney actually gives. If you end up with an antibody that’s oversaturated, it bleeds into neighboring channels and you get weird digital artifacts. If you ever see something like that, it means something went wrong at the optimization and titration stage, and you need to go back and retitrate correctly.

Now, panel design. It really is like a puzzle. My first recommendation is to look at the native autofluorescence in your tissue. There is typically always a high level of autofluorescence in the green range; however, there’s a lot more variation in what we call the secondary autofluorescence peak, depending on the tissue type, and you won’t know what you’re dealing with until you look. So the first thing I do is run a blank slide with no antibody through the Orion protocol, take a quick scan, and look at the tissue in RareCyte’s Artemis software. It lets you go channel by channel and see how the autofluorescence looks, and you can manually set your coefficients and see how the autofluorescence gets subtracted from each channel. I do this first because that secondary autofluorescence peak is variable, and depending on where it falls, you can create a secondary-autofluorescence channel to subtract that peak, which ends up taking the place of one of your antibodies. So you need to know where that secondary peak is before you assign markers to all your channels.

On the right here are all the ArgoFluor channels RareCyte has. There’s a rank column on a scale of 1 to 4 that correlates to how bright each fluor is, and this becomes very important when you decide which channel to conjugate your antibodies to. For your low-abundance antibodies, which aren’t as bright, you want to conjugate to your brightest fluors; on the flip side, if you have a highly abundant antibody signal, you can place it in the dimmer channels. Here’s a quick mock example: on the right are antibodies you’ve validated and assigned an abundance ranking, and on the left are your open channels, and you decide which fluors to conjugate each marker to. Look at FOXP3: that was our lowest-abundance antibody, so it’s very important to place it in the brightest channel; channel 4 is rated very bright, so I put FOXP3 there. On the flip side, CD3 is very highly abundant, so I can place it in the dimmest channel. It’s also important at this stage to keep the analysis in mind. Markers you expect to be co-expressed, like CD3 and CD4, are probably better not placed as direct neighbors; putting some distance between them can help, because otherwise you might not know whether you’re looking at spillover or the true signal of each antibody.

From there you conjugate all your antibodies. It’s a very simple amine-based conjugation; really any technician or student in your lab should be able to do it. RareCyte provides a simple protocol, and you can buy all the fluors directly from RareCyte. It’s a quick protocol, done in a couple of hours. Then you titrate in your tissue of interest, which is an important stage because it tells you the concentration to stain your final panel at. After that, you generate single-stain controls of each marker, scan them, and upload them to the Artemis software, which helps you build an extraction matrix used to process your scans. If you’re familiar with flow cytometry, this will look very familiar: you build this matrix, and each value shows the extraction coefficient, with your donor channels on the left and recipient on the right. Having accurate, representative single-stain controls is super important, because that gives you the best final, properly extracted image.

I want to highlight what’s possible when you work out all these methods, and what we’ve achieved at LJI using Orion. In the past year we stained with 96 unique biomarkers across four different model systems: human, mouse, pig, and Rhesus macaque. Maybe the most striking stat is that in the past year we stained 49 unique combinations of antibodies using the Orion system. And if you generate pretty images, you might be featured on RareCyte’s yearly calendar; this year we made the cover, which was very exciting. That was a project we did with Cecilia Becker from the University of Copenhagen, showing piglet jejunum; she came here to LJI and worked on this Orion project with us. That’s all I have. I think we can take some questions.

Tad George (RareCyte): Thanks, Simon. We did get several questions that span a range of technological and practical topics. I’ll answer the first one, because it relates to when you mentioned the placement of CD4 and CD3 given the spectral overlap on the system. The question is whether we’ve experienced any issues with markers present on the same cell that have closely overlapping spectra. It’s actually okay to have spectral overlap for spatially co-localizing biomarkers, because that extraction matrix Simon mentioned will subtract it. The main issue is that you erode a little on the high end of the dynamic range. The camera is 16-bit, and when you’re collecting data, if you have a lot of cross-talk between two biomarkers in the same pixel, spatially co-localized, you can saturate the detector faster than if they were spaced. The extraction matrix will subtract that pretty accurately, so if you’re able to space them, you do, but it’s not a problem at all. The same cell isn’t bad, because the resolution is fine enough that if you have, say, FOXP3 in the nucleus and CD3 on the surface, those are totally different pixels.

The next one is more for you, Simon, because you mentioned the importance of controls and experimenting with protein gels, which is new for a lot of people. Do you have a recommended gel that you find works well?

Simon Goldstein (La Jolla Institute for Immunology): Yeah. For the cell pellets, we use HistoGel; you spin the cells down into a pellet, and it has an agar-like consistency. The protein gels are relatively new, so if you have questions about that, I’d email histology@lji.org. Our core director, Z, is the one who’s been experimenting with that recently, and he can provide our protocol for how we’re doing it.

Tad George (RareCyte): Excellent. This next question I’ll combine, because two related questions came in about how many biomarkers you can run and how you enable multiple staining rounds. Simon, maybe you can talk a little about the average panel. It looks like you’ve done 49 panels. Are you finding people who want to cycle, and what’s the average panel size?

Simon Goldstein (La Jolla Institute for Immunology): On average, the most common would be 14 to 16 markers, though a lot of times we’re also maxing out at 18. We haven’t gotten into cycling beyond 18, but we are interested in getting there very soon.

Tad George (RareCyte): I’ll answer the other part, which was how you enable multiple staining rounds. In house, we’ve done a 51-plex across three rounds. A lot of customers are doing something we call smart cycling, where you stain with one round, put it in the freezer for up to a year, learn something new, and then pull it back out to do another round for a different cell type. As for the mechanics: the system has 20 channels, one dedicated to a DNA marker and one to dedicated autofluorescence, which leaves 18 ArgoFluor channels for biomarkers. In that first round, before imaging, you stain, put a coverslip on, and scan the slide on the Orion. If you want a subsequent round, you remove the coverslip. We screened a whole bunch of antibody-removal protocols and ended up finding a really nice product from Thermo that removes the signal without damaging the tissue, and then you stain for the subsequent round, coverslip, and image. We’re thinking of coining the term smart cycling, because usually we’re not doing it up front; it’s more of a reflex, once you’ve learned something from the sample or a reviewer asks you to look at something else, you do an informed second cycle, which is what most people are doing.

Another question, even though you answered it directly, is worth repeating: what is the success rate of amine conjugations for an antibody across fluors? You mentioned something like 99%.

Simon Goldstein (La Jolla Institute for Immunology): Yeah, it really is a high success rate. I’ve had one antibody that doesn’t like to be conjugated through amine conjugation.

Tad George (RareCyte): I can expand on that. When we were developing Orion, there were multiple reagent strategies we considered, and a lot of them involved amplification schemes or bulky side chains, but we ended up going with a direct conjugate at a low degree of labeling, because it had so much success over 50 years in flow cytometry. It did put stress on the instrument; we ended up having to put nine high-powered lasers into the instrument to get sensitivity to low-abundance biomarkers with antibodies conjugated to, on average, about two fluorochromes. One issue some people have with amine conjugation is thinking more is better, but you don’t want to overly perturb the antibody. Our degree of labeling is typically one to five, random across the ensemble of antibodies, so our success rate is about the same as what you’re seeing.

This last one is interesting given your diverse set of users. You mentioned technological challenges, but how do you actually get users to decide on a panel? I imagine there are some existential problems in just deciding which markers to use.

Simon Goldstein (La Jolla Institute for Immunology): We like to work backwards. The first question we address is how they actually want to use the data they’re generating, because that informs the panel and how they’ll do the analysis, like which cell types they’re trying to parse out. The worst thing that could happen is choosing a more random panel and then missing markers that would have been essential for the analysis. So you really want to think about how you’ll analyze your images. From there, RareCyte has a catalog, we have our own catalog of the antibodies we’ve generated, and we talk through what would work for them; if there are gaps, we move to finding those custom antibodies to conjugate, and then either add them to our catalog or, sometimes, the user owns the antibody. It just depends.

Tad George (RareCyte): We had a flurry of questions come in. The first is more sales-oriented, so I’ll take it: what if I want to see data from the Orion platform on my own samples before committing to a purchase, and do we provide services and support? Whenever someone’s evaluating the platform for purchase, we have proof-of-concept programs, and we connect interested people with existing users. Simon, I don’t know how many existing customers reached out to you at La Jolla just to get that confidence. There are two things you’re looking at: is it going to work technologically, and am I going to be able to master the technology? We have programs to help you evaluate that. We also do services, including custom-biomarker services if you don’t want to make your own reagents. Many of our biopharma partners own instruments, and also have instruments at a CRO; we have panel-transfer programs to CROs and customers, and some academic partners do that too. So it’s very flexible.

A different question is about throughput: how long does it take to do an 18-plex staining, and how long does it take to scan?

Simon Goldstein (La Jolla Institute for Immunology): From the staining aspect, it takes really the same amount of time as staining one antibody, which is what was so attractive to us; the only difference is physically pipetting the extra antibodies. It’s like flow cytometry: you’re just adding all the antibodies at once to your mix and staining. The scanning depends on the tissue size, but I’ve found that a slide fully maxed out on tissue, scanning all channels, takes around six hours.

Tad George (RareCyte): That’s what we found; it takes about an hour and 15 minutes per square centimeter, so if you’re doing four or five square centimeters, that makes sense. And I suppose you run the system unattended overnight when you can.

Simon Goldstein (La Jolla Institute for Immunology): Yeah, typically. Right now we just have the two-slide-holder system, so when I have a lot of slides to scan, I’m pretty much scanning every day. Typically I’ll start some scans in the morning that I know will finish by end of day, and then load two slides before I leave that will scan overnight. That’s how I get through my scanning most efficiently.

Tad George (RareCyte): Great. The last question is whether RareCyte makes its list of clones available, and how we work with customers who have good clones. For both that and the proof-of-concept evaluation, definitely reach out to us. We freely show what our clones are and what’s in the catalog. We also have what I’d call a reverse biomarker-transfer program: if Simon develops a really nice biomarker, he can share that information with us, and we can put it in the catalog ourselves, as long as we do the validation, because anything we sell needs a QC infrastructure for making lots. With the panel-design tool, you have access to the entire catalog and any custom biomarkers you make, and you can share those with the community, including with us if you want; you don’t have to, but it’s an easy way to build that community. We’re sitting around 175 biomarkers now, and some biomarkers are offered on multiple clones, so the number of clones is higher than 175. We’re on pace to add another 100 to 150 this year, because so many labs like yours are pushing the boundaries in different application areas. I think that was the last question. Thanks, Simon, for doing this; it’s wonderful. And thanks everybody for taking the time to be part of this webinar. Have a great day, everybody.

This transcript was generated automatically from the recorded webinar and edited for readability, speaker labels, and the correct spelling of names, products, and technical terms. Wording may differ from the spoken audio; quantities, product details, and study specifics are reproduced as stated by the speakers and may differ from formally published values.

[Back to all resources](/#library)
## Related resources
[Article](https://www.selectscience.net/article/advanced-spatial-biology-sheds-new-light-on-therapeutic-research) Orion SelectScience
### Advanced spatial biology sheds new light on therapeutic research
[Read Article](https://www.selectscience.net/article/advanced-spatial-biology-sheds-new-light-on-therapeutic-research) [Article](https://www.selectscience.net/editorial-articles/bringing-spatial-biology-to-the-clinic-a-new-lens-on-cancer-biology/?&artID=58147) Orion SelectScience
### Bringing spatial biology to the clinic: “A new lens on cancer biology”
[Read Article](https://www.selectscience.net/editorial-articles/bringing-spatial-biology-to-the-clinic-a-new-lens-on-cancer-biology/?&artID=58147) [Article](https://www.selectscience.net/article/how-to-unlock-the-complexity-of-tumor-microenvironments) Orion SelectScience
### How to unlock the complexity of tumor microenvironments
[Read Article](https://www.selectscience.net/article/how-to-unlock-the-complexity-of-tumor-microenvironments) [Brochure](https://go.pardot.com/l/455452/2020-07-15/8p7453/455452/1606951069teslfZQe/BR21_101_Orion_Brochure_201202.pdf) Orion
### Orion Spatial Proteomics Platform
[View Brochure](https://go.pardot.com/l/455452/2020-07-15/8p7453/455452/1606951069teslfZQe/BR21_101_Orion_Brochure_201202.pdf) [Spec Sheet](https://pardot.rarecyte.com/l/455452/2026-02-24/965v3w/455452/1771949067AeuR6syb/Orion_HT_spec_sheet.pdf) Orion
### Orion HT
[View Spec Sheet](https://pardot.rarecyte.com/l/455452/2026-02-24/965v3w/455452/1771949067AeuR6syb/Orion_HT_spec_sheet.pdf) [Spec Sheet](https://pardot.rarecyte.com/l/455452/2025-10-31/961nyb/455452/1761933957WEJWJyeN/2025.07.14___SS___Orion__LE_.pdf) Orion
### Orion LE
[View Spec Sheet](https://pardot.rarecyte.com/l/455452/2025-10-31/961nyb/455452/1761933957WEJWJyeN/2025.07.14___SS___Orion__LE_.pdf) [Spec Sheet](https://go.pardot.com/l/455452/2020-07-15/8p742r/455452/1607104572nfrlJouR/DF_24_101_201203_Orion_Datafile.pdf) Orion
### Orion
[View Spec Sheet](https://go.pardot.com/l/455452/2020-07-15/8p742r/455452/1607104572nfrlJouR/DF_24_101_201203_Orion_Datafile.pdf) [Research Spotlight](/resources/spatial-profiling-immune-correlates-mss-crc/) Orion 2026
### Spatial Profiling Uncovers Immune Correlates of Durable Response in Phase 1/2 MSS Colorectal Cancer Trial
[Read Spotlight](/resources/spatial-profiling-immune-correlates-mss-crc/) [Publication](/resources/copanlisib-nivolumab-microsatellite-stable-colorectal/) Orion Nature Communications 2026
### Copanlisib in combination with nivolumab for microsatellite stable colorectal cancer: a phase 1/2 trial
[Read More](/resources/copanlisib-nivolumab-microsatellite-stable-colorectal/) [Publication](https://doi.org/10.1158/2159-8290.CD-26-0171) Orion Cancer Discovery 2026
### Spatial Integration of Protein and Chromosomal States Reveals Early Copy-Number Changes and Genotype-Associated Immune Neighborhoods in Serous Ovarian Cancer Evolution
[View Publication](https://doi.org/10.1158/2159-8290.CD-26-0171) [Publication](https://www.nature.com/articles/s41467-026-74933-w) Orion Nature Communications 2026
### Multiple human transgenes prolong survival of triple-carbohydrate knockout porcine kidney xenografts in nonhuman primates
[View Publication](https://www.nature.com/articles/s41467-026-74933-w) [Video](https://www.youtube.com/watch?v=KOiToZIGN0g) Orion 2026
### Dr. Tanjina Kader on mapping spatial proteomics and chromosomal changes on the same tissue section
[Watch Video](https://www.youtube.com/watch?v=KOiToZIGN0g)