The Slide and the Sequence
For a comprehensive education in pathology, we can’t afford to overlook either one
The practice of pathology has changed a great deal over the last two or three years. It’s no longer enough to provide a glass-based “analog” diagnosis – a histological description of a cancer for the oncologist to decipher before making a clinical decision. Today, you have to focus much more on molecular and genomic findings that can inform the oncologist’s decisions. Pathologists are slowly getting on board with the additional responsibility – but it can be hard to convince those who are already happy with their workflow to change it. The answer? We need to make a concerted effort to integrate genomic pathology into training right from the start. It’s crucial; if we don’t, other specialties (medical or non-medical) will step in to take over. And the way I see it, genomics is – and should remain – an integral part of pathology. Many alterations to the genome drive the appearance or behavior of disease, and that’s what “pathology” means to me: the description and identification of disease mechanisms. I want the next generation of pathologists to learn how to integrate genomic pathology into the clinical workflow, and how to use it to obtain a more integrated and comprehensive diagnosis.
Transitional teaching
Pathology education is falling behind. Either there isn’t enough of it or it isn’t focused on the right areas anymore – so how do we get the most out of it?
In my opinion, we start by updated the training to reflect the modern landscape of pathology. We’ve been teaching our students that it’s enough to look at glass slides and memorize them as if they’re curating a database of picture memories. But they’ll need much more. The images our students are memorizing are driven by underlying genomic and epigenomic changes, and I don’t think we currently place enough emphasis on that aspect. Clinical skills are clearly vital, but we give them so much of our attention that medical schools (some more than others) are failing to integrate genomic medicine into the curriculum at all. Our postgraduate trainees are underserved by this lack of attention, and it’s my mission to convince every school to change. Ten years from now, people will come into the doctor’s office for real-time sequence analysis from a fingerprick or cheek swab, and the doctor will need to figure out what that genetic data means – so it’s vital to start educating at the ground level.
Most medical students have studied undergraduate biology, so they know about DNA and RNA. But there are so many clinical skills to learn in medical school that they forget the basic science – and that’s what we need to reignite in pathology! Our number-one focus should be to ensure that our trainees not only have a solid grasp of those fundamentals, but know how to actually apply them to medicine. There are already science questions on the pathology board exams, but that’s not good enough; residents study for their exams, but don’t bother to retain the knowledge into the future, even though they need to understand its practicality in their day-to-day work.
Fortunately, the number-two concern is easier to change right now. Genomics plays a role in many diseases but, right now, most of the research is in cancer – a big part of clinical pathology – which gives us a great opportunity to integrate molecular teaching into our regular curriculum. For example, a resident who sees a lot of lung cancer should be taught to understand the molecular changes that take place in lung cancer, and how some of them may serve as biomarkers. Target each resident’s subspecialty and you’re guaranteed to keep their interest!
At the University of British Columbia, our first-year residents go to a week-long “molecular boot camp” where they review the basics of molecular biology. What’s an intron? What are the different types of mutations? How do they translate to diseases? Next, we say, “As pathologists, it’s important for us to spot these changes to inform our diagnoses.” How do you optimize a tissue sample to test for mutations? What are the pros and cons of a particular molecular assay? In 2017, for the first time, we’re also incorporating four hours of molecular genomics into the orientation – both the basics and its practical application to cancer diagnosis and treatment.
A curriculum in competencies
There is a strong emphasis on molecular pathology in hospitals devoted to its practice, training, and implementation in clinical care. Every Monday morning at the Massachusetts General Hospital, there are molecular pathology rounds to discuss the latest discoveries in the field. It’s pretty impressive that, even in a clinically oriented specialty, residents are encouraged to do that! There is also a requirement for residents to complete four weeks of molecular rotations – things like inherited disease or cancer genetics – so that, even before entering daily practice, they have a well-rounded exposure to molecular pathology.
I think many teaching programs realize the importance of molecular biology, but I also think that American hospitals are ahead of Canadian ones in that respect. In the United States, molecular genetics is a specialty certified by the American Board of Medical Genetics and Genomics – but in Canada, we have no similar program. We’re making strides now; I’m working with the Ontario Molecular Pathology Research Network to develop a teaching curriculum for genomic pathology. Initially, we want to implement it in all of Ontario’s medical training programs – ultimately, we’d like to submit the goals and objectives to the Royal College so that it can be applied to pathology training programs throughout Canada.
Hopefully, our efforts will tie in with the larger changes happening in the medical education system – namely, the transition to “competency by design” (see “Skill Switch”). What kinds of competencies would we want to see in genomic pathology? A first-year resident might be expected to understand the fundamental science and be able to process tissue optimally for genomic testing. The next year might involve more tissue-specific goals; for example, identifying the necessary tests for different types of diseases and responding appropriately to the results. With increased training and experience, the goals and objectives become more challenging – just like in any other branch of medical training.
How can trainees prepare?
Every resident is different. Some want to work in hospitals, others in private practice, still others in research. But no matter what your aims, genomics will be a big part of your work. A decade ago, when a clinician asked, “What does this diagnosis mean?” the response might be: “This is a carcinoid tumor and here’s what each of these features indicates.” Ten years from now, as well as giving the histological diagnosis, you’ll probably have to say, “The sequencing has revealed a mutation, and here’s the implication for the patient.” So every pathologist – no matter what career path he or she takes – will need to understand genomic medicine, use it safely and effectively alongside glass-based histological description, and relay it proficiently to a clinician.
At the same time, aspiring pathologists must not overlook the ongoing importance of glass slides. Sometimes I get asked, “If you can sequence a tumor, why don’t you just sequence everything and forget about glass altogether?” But that takes time and resources. An experienced pathologist can look through a microscope and – in five seconds – make a very tight differential or even a diagnosis based on the tumor’s appearance. Not to mention the fact that glass slides are transferable between hospitals; not every site has the capacity for genomic testing, but everyone has a microscope. Glass slides remain the common currency among pathologists, so this is not about replacing them; it’s about making the most out of tissue- and glass-based pathology by integrating additional information. At the moment, that might be imaging or immunohistochemistry, but in the next five to 10 years, it will be biomarkers and other genomic findings.
And let’s not forget that we will still need to make decisions about which samples need to be sequenced. Some might argue, if sequencing is becoming so much cheaper, why not just sequence everything and develop a huge database for each case? I don’t think that’s practical in a clinical pathology lab; the scale of data storage, computation and interpretation required is still beyond what we can efficiently handle. So we need to triage our samples. It’s something we already do – for instance, we use immunohistochemistry (along with a host of other factors) as a triage tool for additional FISH testing, because the test is too costly to perform on every sample. Glass will continue to be extremely important for the foreseeable future, because we’re already familiar with the visual features of tumors and we’re now linking that knowledge to our understanding of the underlying genomic aberrations.
In short, trainees shouldn’t toss their microscopes just yet. Instead, they should focus on connecting the disparate pieces of information they gain from both slides and sequences.
What’s coming next?
Usually, when we talk about investing in new technologies, it’s about buying the equipment, hiring staff and changing laboratory infrastructure. People often overlook the fact that we need to invest in educating the next generation of pathologists. We need to teach our residents to take ownership of genomics and its integration into the study of disease as a whole. We need to prepare residents for the role of the future pathologist.
There is a sort of dichotomy: educators need to move as quickly as possible to include genomics in their teaching curricula – but to go slowly when they’re actually teaching it. Trainees must have the fundamentals in place before studying more complex aspects of genomics, or else they’ll just end up with an overdose of information with no background knowledge to process it. It also takes time to instill what I would argue is an essential passion for genomic pathology. Young pathologists need to see that genomic pathology is the future of their field – then they’ll realize how exciting it really is and how affective it is. Pathology is really pushing the forefront of medicine. And, in my opinion, there has never been a more thrilling time to be a pathologist!
Stephen Yip is Assistant Professor of Pathology and Laboratory Medicine at the University of British Columbia, Neuropathologist at Vancouver General Hospital, and Associate Member at Canada’s Michael Smith Genome Sciences Centre.
Stephen Yip is Assistant Professor of Pathology and Laboratory Medicine at the University of British Columbia, Neuropathologist at Vancouver General Hospital, and Associate Member at Canada’s Michael Smith Genome Sciences Centre.
