EP207: Dr. Kevin Ko: New Horizons in Oral and Dermatopathology

Dr. Kevin Ko on Biomarkers, Oral Dysplasia, and the Limits of H&E Diagnosis

Christine interviews Dr. Kevin Ko (DMD, MD), a pathologist at the BC Cancer Agency with training in oral and maxillofacial pathology, anatomic pathology, and dermatopathology. They discuss his ASDP 2025 lecture on using p53 in oral dysplasia as a potential new approach and the broader problem of diagnostic discordance and over-diagnosis when relying on H&E alone. Dr. Ko shares examples from practice, including recognizing oral porokeratosis (previously followed as dysplasia for years) and a chemotherapy-related lip lesion initially suspected to be severe dysplasia but supported by wild-type biomarker results and clinical history, resolving after stopping chemotherapy drugs. He emphasizes the need for reproducible biomarkers and possibly molecular-based classification to improve consistency and patient outcomes, while also describing the pressure to be near-perfect in pathology, the risk of burnout, and efforts to build sustainable systems (QA sessions, colleague consultation, protected time). The conversation closes with his approach to presentations as storytelling, interest in prospective multi-center research, and a final message about balancing perfectionism with rest while remaining open-minded to new diagnostic methods to improve patient care.

00:00 Welcome & Meet Dr. Kevin Ko (DMD/MD, Dermpath at BC Cancer)

01:00 The Controversial Idea: Using p53 Biomarkers in Oral Dysplasia

01:18 Oral vs Skin Pathology: Discovering Porokeratosis in the Mouth

02:07 Diagnostic Error & Overdiagnosis: Why Reproducible Biomarkers Matter

05:19 Case Study: “Severe Dysplasia” vs Toxic Erythema of Chemotherapy —Context Changes Everything

06:36 The Perfectionism Trap in Pathology (and Why 95% Isn’t Good Enough)

08:04 Burnout, QA Systems, and Building Sustainable Workflows

09:14 Work–Life Balance, Kids, and Choosing Priorities (Family vs Research)

11:14 How to Build a Great Talk: Storytelling, Cases, and Future Studies

11:38 Final Takeaways: Balance, Open-Mindedness, and Better Diagnostics

Transcript

Christine Ko: [00:00:00] Welcome back to The Girl Doc Survival Guide. Today I am very pleased to be with Dr. Kevin Ko. Dr. Kevin Ko, DMD and MD, graduated from UBC Dentistry and then from Columbia University for his MD degree, with two residencies, one in oral and maxillofacial pathology, and a second in anatomic pathology at Vancouver General Hospital and Stanford University, followed by a fellowship in dermatopathology at Memorial Sloan Kettering. He currently works at the BC Cancer Agency.

Welcome to Kevin. 

Kevin Ko: Thank you so much for having me, Christine. It is an honor and a great opportunity to participate in this. I'm a big fan. 

Christine Ko: Oh, thanks I really loved your lecture at the American Society for Dermatopathology meeting, this past November, in 2025. I really appreciated how you presented in a manner in which I felt was very honest, high level, and thoughtful.

Kevin Ko: [00:01:00] Thank you. It might be a little controversial. It's basically using p53 in oral dysplasia. It's a very new concept. My hope is that more and more people will try this in their practice, and once we gather more data, maybe there's a chance that this may become the future standard. 

Christine Ko: Yes. Before we move on to anything else, can you share a personal anecdote? 

Kevin Ko: Definitely. I did my oral pathology residency first. Pathology from a view of a dentist. Then I did med school and then did the anatomic pathology and dermatopathology, seeing oral lesions from the view of dermatopathologist. I'm starting to find diseases that we didn't know exist in the mouth. For example, porokeratosis. We started seeing cases in the tongue. Clinically, it looks perfect for porokeratosis, H and E is perfect for porokeratosis. Some of them were actually diagnosed as dysplasia and followed as dysplasia for like 20 years, and we realized, oh, this is just all porokeratosis. So we're trying to [00:02:00] publish this soon. I realized oral and dermatology really just one entity in a way.

Christine Ko: Yes. So given your training, which you just mentioned, as a dentist and then oral pathology and also skin pathology, you touched on this a little bit right now, like diagnostic error, by saying that we weren't recognizing porokeratosis in the oral mucosa. It was called something else. And one of your patients, you mentioned, was actually thought to have something sort of like a precursor to cancer. Even though it's not. And I think we see that in the skin too, that sometimes the cells in porokeratosis do look a little bit abnormal or atypical, but we don't call that a cancer or a pre-cancer. It's just porokeratosis on sun damaged skin. Do you have an opinion, a feeling about this, about diagnostic error or diagnostic discordance or even over-diagnosis? 'cause I guess that [00:03:00] applies to the patient who is carried as having some sort of pre-cancer or dysplasia when actually they had something benign, which is porokeratosis.

Kevin Ko: Yeah, definitely. This is one of the big reasons why I really want to come up with biomarkers that can be reproducible among pathologists. When I was in training, I was taught that oral dysplasia diagnosis is an art. Every person might interpret differently, and there might be no consensus. But then this diagnosis really have consequences, right? The moment you call a patient with dysplasia, they might be followed for life. Some surgeons might be very aggressive. They might actually excise the entire lesion. So we have to be very accurate. Be very precise. But it might not be possible with only H and E. To separate out the true dysplasia versus actually is not dysplasia is to me very important. In the world of severe dysplasia, the agreement is higher.

My hope is with new biomarkers, with molecular evidence [00:04:00] that we can maybe use a molecular based classification system as a way to move forward and be more reproducible.

Christine Ko: It's interesting that you said that early on, your teachers and mentors said that, oh, you know, diagnosing is an art. That it is somewhat subjective rather than based on something objective like math, like two plus two is four. It's interesting you say that because I don't know that I understood that until much later. I think I always thought that there is one right answer or maybe one best answer. And I think I have shifted more recently to think that by only looking at a microscopic slide, the standard staining, the hematoxylin and eosin staining, that sometimes we can't know.

And so like you said, using a biomarker, whether it's a immunohistochemical stain or some molecular test or something, might be [00:05:00] necessary. And so I've been wondering, maybe we're expecting too much of ourselves? When I spoke to Dr. Wolfgang Weyers, he said a similar thing, like the perspective matters, and sometimes we're expecting too much from just one perspective, meaning just looking at one slide.

Kevin Ko: Definitely, I'm a hundred percent in agreement with this. You know, there are examples where on histology we came to one diagnosis and everyone is in agreement. And then, we received a clinical photo later, and our diagnosis completely changed. A recent example is a patient with a lip biopsy that the pathologist who got the biopsy was quite worried about severe dysplasia, which means the lip usually will come out. Then it was shown to me. I was also thinking, yeah, I think it's likely severe dysplasia, but it's so inflamed, something's weird about it. And we tried our biomarkers and they were actually wild type, and they made us pause, say, Hmm, you know, most severe dysplasia should be mutant pattern in our markers, but this is wild type. [00:06:00] So then we get more history and clinical photos. And turns out patient actually had chemotherapy, and then we're like, oh, this could be toxic erythema of chemo in a mucosal lip, another condition very known to the pathologist, but not that well known among oral pathologists. We ask them to stop the chemo drugs and then lesion resolved a few weeks later. It is one great example where, you know, knowing the history, seeing a clinical photo, using a biomarker might actually be the way to go.

But the problem with that is, you know, we have a lot of cases. Are we able to do that in all cases? And then there's a, a, a part of me who is trying to be perfect. I'm not a perfectionist. At least I don't think I am according to my wife. But in this job I try to be perfect. Because I know there's a patient behind every slide, and I want to do the best I can. So what happens is I find myself staying very late and then trying to figure everything out. [00:07:00] There's part of me who wants to get every case right. In dental school, in medical school, the exams, right? If you get like 95% great, amazing score. But that's not how it works in pathology, 'cause if I'm only 95% correct in my diagnosis, that means 5% error. That's way too high for pathologist. If I'm only 95% accurate and correct, that's a big problem. 'Cause we might see 5,000, you know, 10,000 cases a year. 5% of that is way too many. And then, so the urge is to get like 99.99% accurate. But the pursuit of perfection is quite difficult and maybe perhaps unhealthy. And this is something I'm struggling with right now. 

Christine Ko: I agree with you on the perfectionism thing. You're totally right. Even 99% correct, one outta a hundred times you're wrong. That's still so high. I don't wanna make those errors on patients. But it's unavoidable, right? We're not gonna be able to be [00:08:00] perfect. That's just a fact. So it's a hard thing. So you are saying that you have been dealing with this a little bit by staying later at work and working harder.

Kevin Ko: It's so unhealthy. 

Christine Ko: Probably not so sustainable long term though, in terms of thriving. 

Kevin Ko: No, it's not. It's it's getting to the point where I have to stop and pause and think. I think burnout is a real concern. I find myself coming to work on Saturday and Sunday, probably worse than my surgery rotation in med school.

To the point, you know, I was talking to my family the other day, they were like, listen, you can't continue to do this, you know that. And so one thing I'm trying to do is to have a system that allow me to rest where I feel comfortable and then know when to stop.

I'm still trying to resolve this. It's not easy. But the things I try to do is say, have more QA sessions where case I'm unsure about, talking cases with senior colleagues [00:09:00] and have like days where I can have a protected time off to catch up with my cases and work on research, the type of thing.

These are all the things I'm trying to implement, and hopefully I will have a more balanced work life balance. 

Christine Ko: Yeah. Work life balance is hard. It sounds like you have young kids. 

Kevin Ko: Yeah, I have a 6-year-old and then, you know, because I'm getting older, I realize the most precious asset I have might be time. Something I didn't cherish before. Now I'm getting older, I realize no one's around forever. We have to know what's important in our lives. And I feel like family is my number one, being around my kid when she's growing up. So I'm trying to spend more time with her now. 

Christine Ko: It's hard because, for a lot of careers and definitely for medicine, you know, our kids are young, right? When we do have to put a lot of time into our careers and establish ourselves in terms of job and research and other things. It's [00:10:00] definitely not easy to have the right balance. So think about your priorities. 

Kevin Ko: That being said, there's also a part of me who really wants to do more research that could possibly change how we diagnose things better. It's tough knowing research is interesting and important, but family is first. And how to find the right balance is something I'm still trying to work. 

Christine Ko: Yeah. I think it's really hard when you have more than one passion, right?

Imagine it would be so much easier if I was like, oh, you know, family's number one, and that's the only thing that I care that much about. Then I could just focus on family, like one thing. But as you're saying, if you're like, oh yeah, family's my number one, but also a close second, is this research that I can do and maybe really change patient's lives. Your story before where the lip lesion and the patient would've been given a diagnosis [00:11:00] of basically cancer and have their lip removed versus getting a diagnosis of, oh, this is probably related to the chemotherapy you're already on for a different cancer. So different for that patient. You're doing good work.

Kevin Ko: Oh, thank you.

Christine Ko: Going back to your ASDP presentation, it was wonderful. Do you have a method for creating a great presentation?

Kevin Ko: I guess what I'm trying to do is to tell a story, a journey, right? Then the next part is to find cases. The future direction I'm hoping to do is to work with other centers in a prospective study.

Christine Ko: That's great. I love the idea of, you know, really create a story. Do you have any final thoughts? 

Kevin Ko: I think we are all trying to do the best we can as pathologists and clinicians. There's a drive to do everything perfectly. We have to understand that sometimes it's important to stop, pause, and then think about balance a little bit. And also be open-minded with new ideas, new techniques, [00:12:00] technologies. If there's a chance that we can do better, be open-minded and try a new method, maybe we can actually improve our diagnostic accuracy and improve patient care.

Christine Ko: I like it. Thank you. Thank you so much for your time. 

Kevin Ko: Thank you so much for your time.