Can you tell me a bit about yourself and what brings you to AACR?

My name is Sizun Jiang, and I'm an Assistant Professor at Harvard and Beth Israel Deaconess Medical Center. I was invited to the conference by Cathy Wu, who's a leader in the field of immuno-oncology and cancer immunology, to give an educational session teaching non-immunologists a bit about tumor immunology. More specifically, the tissue and the tumor microenvironment, generating data in a meaningful way so that we can compare across different sites and new technologies that allow us to do some of that. And then ultimately, how do we use that to understand more about cancer. These types of sessions provide the best environment for that.

What projects or disease areas are you working on at the moment?

My lab is primarily split into three main areas of research: cancer, viral-related diseases, and immune dysregulation. It's interesting because you can see there are certain things that encompass all three of these pillars: in viral-driven cancers, there's intrinsically immune dysregulation, a viral component, and a cancer component. In the lab, we have people working on either one, two, or three areas at the same time, and we're hoping to teach ourselves, and maybe others too, that concepts we take from one can be applied to another.

One example is in the cancer field. There's this concept of the hot and cold tumor microenvironment –  it has been around for decades. It's a chronic disease in which the immune system is incapable of mounting an effective response. Similarly, if you think about other diseases like HIV, a chronic viral disease that people live with, the immune system is impaired and not able to fully respond. We've been identifying similar features, and I hope that by learning more about one, we can apply that knowledge to a wider range of diseases.

You’ve published some high-impact publications recently – what are they focused on?

I'm very fortunate to work with people who are smarter, more hard-working than me. Both the trainees in the lab were amazing, but also other faculties that I collaborate with.

In one paper, we are looking at tissue remodeling in this chronic autoimmune disease called chronic rhinosinusitis. It's incredibly interesting – patients have trouble breathing because there’s inflammation of the nasal tract. There are different versions of it, one of which causes these nasal polyps: grape-like structures in the nasal tract that block the airflow. We had an interesting hypothesis: when people remove it surgically, it almost always comes back, which means that there are tissue cues that recruit and remodel the tissue at that site. We've discovered some of the molecular cues and cell mechanisms that underpin that, so hopefully that will be published soon.

We have another paper where we're developing a new computational method, collaborating with researchers specializing in spatial transcriptomics. Traditional tools have been really beneficial; however, these tools were originally developed for single-cell analysis, so they often lose some of the complexity of the image. Our method incorporates information from the images themselves, and we’ve found that this approach is really powerful in teaching us not just about the cell types, but also their locations and arrangements. This spatial context helps us understand the cell states and how they are situated within the tissues.

What's getting you excited in spatial?

We're very fortunate to work with people who are amazing at what they do, folks such as one of my good friends, Faisal Mahmood from Brigham and Women's Hospital. AI is a massive buzzword right now – but we're essentially building these AI foundation models and tools that will enable us to accelerate our understanding. This isn't just about computation – it speeds up everything we do. Generating a high-quality data set can take three to six months, and analyzing it might take years. I feel we're often struggling to get this information out as quickly as we need to. Not just for the journals and the scientists – the major stakeholders here are the people living with the diseases. We want to be able to improve their lives faster, and AI-accelerated tools are definitely changing our capabilities in this area.

What current challenges are there in spatial biology or multi-omics fields, and what are your thoughts or hopes for addressing them in the future?

One of the biggest challenges is how we harmonize across different technologies and platforms, such that we can start comparing studies to each other. If a study were done here in Chicago and another in the UK using the same platforms and antibodies, there would likely be significant variability in results due to multiple factors. Naturally, technical variability plays a role, highlighting the need for an international effort to standardize and assess methodologies. But the major variable is the patient tissue itself. Each institution will have slightly different protocols for tissue fixation and processing, which can significantly impact downstream detection of protein epitopes and RNA. These variations can introduce batch effects that must be accounted for to prevent them from obscuring key insights. Fortunately, more people are starting to recognize that, but substantial support is needed and I don't currently know where that's going to come from. We're just bootstrapping everyone's own initiatives, but I’d really like to see more of an international effort come together.

How do you hope the field develops in the next 5-10 years?

I want to be able to get to a point where we’re not just generating data, but we’re doing it in a very meaningful, well-designed way, because the bottleneck is always the analysis. If we're able to spend 20-30% more effort in ensuring robust data generation, I think that's going to accelerate things downstream – there are efforts to try to set up international standards consortia to do that. It's difficult because there are many different stakeholders and strong opinions; everyone is coming from a place of real intention, but this makes it difficult to reconcile different priorities. I don’t know how it will be resolved, but I’m certainly hoping that we’ll engage in that together.

Are there any emerging proteomics-based technologies we should be keeping our eye on, and why?

I can tell you a few things I’m excited about. People always want more – 30, 40, even 50-antibody imaging plex in the same tissue is routine and accessible to many. But there’s a difference between what people want and what they actually need. What they need is a robust set of 10 to 20 markers. What they want is everything, everywhere, all at once. But that comes with trade-offs.

We have to be deliberate in how we approach this. For example, if we’re studying TCR signaling or CAR T cells, we need to design our panels strategically to target the key markers and signaling molecules. And beyond that, some of the emerging developments in this space are incredibly exciting. There are assays related to proximity ligation, allowing us to test whether two proteins are interacting, and that's a physical test. There are companies like Navinci and others who are working on this, and I think it's really interesting to think about it eventually as a clinically meaningful asset.

There's also a group that came out with 120 antibody, single cell, imaging recently – Bernd Bondenmiller’s group in Zurich – using mass spec imaging, and that's really exciting too, because they're able to couple that potentially with other things like glycans and metabolites. And then I think more and more, we're starting to go towards these actual mass spec readouts, where we see the protein using an antibody, we see the RNA. But ultimately, what about all the other proteins that we can measure and for that spatially, at least, there's a really exciting method called Deep visual proteomics coming out from Matthias Mann's lab.

There have been extensions of that by various other groups. One of them is a commercial company called Syncell, they have a method called Microscoop. This technique enables precise cross-linking and mass spectrometry of specific regions within tissues or cells, allowing for a detailed understanding of their protein compositions. I find these advancements incredibly exciting.

Selfishly, we have internal methods we're developing where we're trying to go towards multi omics methods such as getting protein and RNA from same tissue sections and this is with great friends from MIT, Laura Kiessling and Alex Shalek, are looking at lectins and these sugar modifications at the single cell level. I think that's the next frontier.

Think about the textbook definition of a cell – you draw this phospholipid membrane, some protein shell, you see a nucleus, you see the Golgi, your various different organelles inside. But I think what's very underappreciated is our cells are not just protein shells. They are choked full of these sugars or glycans on the surface, and any kind of recognition starts at that level. So that's again, underappreciated. We're just starting to see new tools starting to get developed for that, and I'm really excited to see how that will just form our understanding of cancer, but also cell-cell recognition and basic mechanisms of immune responses.