What does your work look like at Duke?

SARA:
I am a physician at Duke. I’ve been on staff as faculty for, I think this is my 8th or 9th year — who’s counting? I do a number of things including clinical care, research and education. In terms of my clinical work, I do cytopathology, which is basically things like doing fine needle aspiration biopsies in our clinic, looking at fluids, looking at samples from all over the body, as well as doing surgical pathology of the head, neck and endocrine system. So, looking at biopsies and specimens that come from patients with things like head and neck cancer, thyroid cancers and other endocrine neoplasms.

My research focuses on two areas. The more traditional academic niche is looking at thyroid molecular testing for diagnostics and prognostics and performance of fine needle aspiration for pathologists. My less traditional niche is the use of Social Media for medical professionals, for education, for networking and for research collaborations. In terms of teaching, I am very busy teaching everyone from medical students, to allied health students, to residents, to fellows, to my colleagues in Pathology. I do a lot of teaching internationally as well. I think I covered most of the bases there.

What drew you to this field? How did you get into pathology?

Sara: So, I have always been interested in science, and I’ve always enjoyed biology. I was really interested in neuroscience in undergrad, and I actually worked at Columbia for a few years in New York looking at neuroscience and biological psychiatry research. When I came to medical school, I thought I was going to continue along that path — neurology, neuro-oncology, neurosurgery. I actually did a year of research in Duke’s Brain Tumor Center, and while I was there, I got the chance to interact with all of the neuropathologists and really fell in love with pathology. I fell in love with the department and the people. I thought it was a great fit for my interests in research. The scientific, visual, looking-at-the-disease processes face to face, being the closest to the tissue, performing the diagnosis aspect was very appealing to me. That’s how I got into pathology.

When I came into medical school I had, as I think a lot of people have, a lot of misconceptions about pathologists, you know “all we do is autopsies, we work with dead people.” While that is certainly a part of what I and some of my colleagues do, all of my current work is with the living. I see patients face-to-face, I do procedures, so I am able to really do a lot of the things that drew me to medicine as well as have a little bit more flexibility to do research projects. Especially in terms of the fact that a lot of the research we do is very tissue-based, I am closest to the cells themselves.

I think it’s so interesting that your work focuses on looking at things on a cellular level but you are also still interfacing with actual human patients.

SARA:
Yeah, big picture, little picture.

What does your workflow look like currently when you’re looking at specimens? It sounds like you have specimens coming to you from a lot of different sources, can you give me an idea of that range?

SARA:
‍I do a number of different things in my workflow. A typical day of some of the cytopathology services might be looking at samples from the day before that were either obtained in radiology, or the procedure suites in endoscopy or bronchoscopy, or in my own clinic. Then, I go run back and forth to my clinic to do biopsies on an as-needed basis.

The actual diagnostic work happens with a regular light microscope. In terms of actually making the diagnosis, it’s just like Rudolph Virchow did back in the 1800s in some ways, which is that I take a slide out of a slide flat and I look at it under a microscope and use the light microscopy findings to make the diagnosis. If I need to, I can do some ancillary studies on the tissue, immuno-hystochemical stains, we can sometimes do molecular studies on the samples as well, but it’s still very much based on the glass slide.

We do use digital pathology somewhat in our department but we haven’t gone digital-only for primary diagnosis. We do use digital pathology for certain applications such as telepathology. So, for instance, one of the other things I do is called “frozen sections,” where I am looking at the tissue in real time from a patient who is in the OR, so that I can give the surgeon information that helps their management. For instance, we do something called “margins,” to see if all of the tumor is out of the patient, and they do a “frozen” to see if something is benign or malignant, as that might change their operative approach and things like that.

If I need some specialty help from one of my colleagues, I can put the slide on the scanner, scan it and show it to them, and they can look at it from wherever they are. Another instance where we use digital pathology would be overnight. We do an overnight call, and if there is an organ that needs to be transplanted and someone needs to look at that slide, one of my residents or fellows will scan that slide and I will review the findings from the comfort of my home in the middle of the night — which is nice for me.

What are some challenges or differences you notice between digital vs a glass slides?

SARA:
Right now, one of the challenges is that in order to make a digital slide, you still have to make a glass slide first — so all of the work that’s going into making the glass slide is still there. So, if you’re adding digital pathology, that actually adds more time in terms of scanning, and depending on your software you’re using to view the slides it could be a bit slower than using a traditional microscope.

The other limitations are particularly in the area of cytopathology — which is one of my areas of sub-speciality. You’re getting actual cells and clusters of cells on the slide itself, and they have a 3-dimensional aspect to them. The standard scanners that are available on the market right now are getting better, but they still don’t handle that 3D information very well. The 3D bit, Z-stacking, takes more time and more storage space. Having some solutions that could better capture that 3-dimensional element and maintain the information and the 3-dimensionality while at the same time minimizing storage size and scanning time, that’s where some of the challenges lie in cytology.

What do you feel could improve your workflow, in terms of technological developments?

SARA:
‍I’m thinking about cytopathology in particular — I think for surgical pathology, solutions are already out there. If you can scan the slide, you can start doing all kinds of things. Whether you’re using it for primary diagnosis, using it for remote applications, such as if you’re trying to look at it in the middle of the night, or if I’m trying to look at slides here from a hospital located across town. Also things like archiving that tissue. If you need to archive a slide, you can have a nice record of that slide and then you can use the actual tissue for molecular diagnostics. You can actually take the cells that are on that smear, scrape them off, and use them for prognostic information to help guide therapy, but you need to be able to archive that tissue somehow — digital would be a great way to do that.

Then, thinking about machine learning and AI applications, there’s all kinds of tools out there that have been developed to help with diagnosis in the surgery-pathology arena. Whether it’s something like counting mitotic figures, even doing preliminary classifications of tumor types. There’s a lot of things that machines are really good at doing, such as counting widgets, that really aren’t a good use of human time. But, without the ability to get the cytology slides into a digital format, all of those downstream applications are not really available for the cytopathology smears that I’m lookin at.

When you examine patient biopsy cytopathology slides, are there any features that at first glance immediately catch your eye and indicate something may be abnormal?

SARA:
The first thing that we always learn when we’re doing medical training is “what is normal?” and being able to identify “normal,” because then you can look at the populations that are abnormal. For instance, if I’m looking at a specimen from a lung, some of the normal elements would be bronchial epithelial cells, which have a very distinct appearance — they have cilia, so they always look like they have little hairs on them. Because I’ve looked at so many of them, I can identify the normal elements pretty easily from low-power (i.e. low magnification). Then, from low power you can say “these don’t look normal, there’s some intruders, or second population.”

When I teach my residents and fellows, I always say “look for the alien population.” Identify your normal populations, identify your alien populations. From there, you can take a closer look to see features of malignancy — nuclear pleomorphism, really big nuclei, really funky, wrinkly nuclei, mitotic figures, or necrosis. Those would be the things we would look for to identify malignancy.

What runs through your head when you catch a malignant feature, or an alien population?

SARA:
One of the things I also teach my residents is, after you’ve been doing pathology for a while, a lot of the cases do become straightforward. You look at it and you go “oh, I know what that is, that’s a cancer.” The next step, I always say, is “prove yourself wrong.” If you aren’t aware of the pitfalls — and there are many in pathology, medicine, biology in general — it is really important to make sure you’re not missing something.

For instance, if someone has an infection in the lung, they can get what we call Reactive Atypia to the cells — so the cells look funky, they look big, they look a little weird, but they aren’t necessarily malignant. So, when I look at a cluster of cells or a slide and I think it looks malignant, I always think “is there any way this could be reactive? Is there any way this could be a mimic of some kind?” Always trying to prove yourself wrong is a little bit stressful, but I think when you’re diagnosing cancer, it’s worth taking that second to make sure you’re not missing something.

As a patient, that’s a very reassuring thing to know — that people are going through that process of triple-checking. Can you talk more about the importance of social media for the scientific community?

SARA:
I think there’s so many reasons — certainly during the pandemic — that the interaction and networking that happened via social media was so important. In healthcare, it’s always been stressful but it has been a particularly stressful time, and some of the connections you make via social media can really help you. In your career as a scientist you’re always trying to build your network and your reputation, and social media is a great way to do that. But also, it can help you feel connected with a community of individuals who are all moving forward and trying to take care of patients and advance the science. It’s a little bit touchy-feely but social media has been a powerful tool to help keep us connected and keep those friendships going even when we weren’t able to meet face-to-face.

The other element that I think is really important is that there’s a lot of teaching that goes on, it’ all anonymized and patient identifiers are removed, but we are so fortunate in this country to have access to all the latest journals, textbooks, findings — I have sub-specialty experts just down the hall if I need any help. But, not everybody in the world has access to these resources, so social media has been a way to share resources, educational tools, and flatten those hierarchies to allow people in under-resourced areas to also get access to teaching and expertise. People in some of these regions may not have access to a desktop computer, but people have smartphones. If you have a smartphone, you can go on Twitter, Instagram, Facebook. Obviously, there are a number of issues at play with all those commercially owned entities and people who may or may not have taken them over in recent months, but at the same time they’ve still been a great resource to be able to share that education with our colleagues from around the world.

I think that’s excellent and so true. I love what you said about “flattening hierarchies,” because I think that’s one thing that is exciting about digital imaging, especially with regard to science but also art and archeology — it’s making things available and leveling the playing field in so many ways academically. Do you have any advice for folks interested in exploring pathology within their medical training?

SARA:
‍Just reach out to your local pathology department or pathologist. You know, we may be busy but most of us really love our jobs and love sharing what we do. I think we’re born educators, so reach out. Or reach out to me, you can go on social media and make connections and find pathologists there as well, but try to connect and see if you can find those opportunities.

Even just going on to pathology Twitter, you’ll get such a good sense of what the field is like, because we have so many passionate individuals out there doing educational efforts. It’s a great way to look at the community, see what kinds of people are out there, what kind of cases we’re looking at. I’d say explore it, pathology may not be top of most people’s list when they go to medical school or think about training but it’s a wonderful field filled with wonderful people. You do care for living patients, you can have patient interactions if you want, you can be in procedures if you want. It’s a hidden gem, so I definitely encourage anyone to hop on pathology Twitter and reach out.

Where can people find out more about your work?

SARA:
You can definitely find me on twitter (@sara_jiang). I’m also on Facebook (Sara Jiang, MD) and Instagram (@sarajiangmd). That’s probably the best way to see what I’m doing, where I am, and what I’m thinking about at any given time. It’s also a good way to contact me if people have questions or want to connect.

‍