Candid Cancer: Know thy pathologist
Still in the operating room, I asked the surgeon to show me what he had removed. Before my sleepy eyes, he placed a vial containing what looked like a bloody pea, a small piece of me that was about to seal my fate.
If I was lucky, the tumors that a recent CT scan had revealed would
be benign. Unlucky, and I was facing cancer. And somewhere in a remote
area of the hospital, a pathologist I would never meet
— who couldn't possibly know how much I wanted to live — would
decide whether I was lucky or unlucky. Panicked, I wanted to bolt off
the operating table, hand deliver the bloody pea and beg the pathologist
to treat it as his or her own. That, of course, was impossible.
We patients never meet these doctors who judge whether our cells are malignant or not, and if they are, just how aggressively they are misbehaving. Routinely, we place our destinies in their hands, but what do we really know about them? To find out, I turned to Dr. Kojo Elenitoba-Johnson, professor of pathology at the University of Michigan where he is the director of translational pathology and the director of the Molecular Diagnostics Laboratory.A tall, gracious man with a warm, friendly smile, Dr. Elenitoba-Johnson first takes me on a tour of his lab where he introduces me to colleagues and shows me where he's growing and studying cancer cells. Nine years after my own diagnosis of non-Hodgkin lymphoma, I get goose bumps standing in the very spot where other bloody peas are being analyzed, instantly empathizing with those who await their fates.
Back in his office, Dr. Elenitoba-Johnson tells me why accurate pathology is the foundation on which the best possible outcome rests. "Pathology not only establishes the correct diagnosis of the condition but implicit with the diagnosis is the prognosis and the selection of the appropriate treatment. So you can interpret that to mean: right diagnosis, right treatment; wrong diagnosis, wrong treatment."
It can't get any clearer than that, and I think about a family member who'd received less than conclusive results when tested for cancer at a small hospital where its two pathologists are charged with what seems like the impossible responsibility of interpreting everything from pap smears to skin lesions to cancer cells. Knowing the answer, I still ask if larger hospitals have specialists who are better equipped to make more definitive diagnoses.
Dr. Elenitoba-Johnson tells me that for every oncology sub-specialty, there is also a pathology sub-specialty. "That means that a particular condition is interrogated with expertise that is both broad and deep," he explains. Hematopathologists, for example, deal primarily with conditions of the blood, bone marrow, lymph nodes and spleen, and Michigan has nine board-certified hematopathologists. These pathology sub-specialties require additional training of one to three years after residency.
He adds that larger institutions have more sophisticated technology and are able to recruit top specialists. Of course. The cost of state-of-the-art diagnostic equipment is too prohibitive for every hospital to have it, but top specialists are drawn to places that do. And for patients, state-of-the-art equipment means more diagnostic accuracy and potential earlier detection when cancers are easier to treat.
Dr. Elenitoba-Johnson also points out that it's important for pathologists to be familiar with new developments, and I think back to my own diagnosis in 2002 when the World Health Organization recognized 30 sub-classifications of non-Hodgkin lymphoma. In 2008, it recognized 56 — nearly double in six years! Who but a specialist — a hematopathologist — has time to keep up with all that?
I suggest that because pathologists see tissue samples, they should be credited for identifying new sub-classifications. Dr. Elenitoba-Johnson demurs, "This isn't done in a vacuum. It's collaborative. Every now and then you'll offer a diagnosis to a clinician who says, "This isn't behaving like 'x' kind of lymphoma we've seen before," and then we go back and forth and study it to find out what's different about this than the other ones. The process of establishing distinct diagnostic categories is an evolving process, and this is often refined and informed by the application of new sophisticated tools to identify homogeneous clinicopathologic entities. That's how new categories get born."
It occurs to me that as treatment options are not only increasing but also improving, the significance of identifying sub-classifications can't be overstated. Using the multiple types of lymphoma as an example, Dr. Elenitoba-Johnson elaborates, "There are molecular abnormalities that unite the diseases that otherwise look the same under the microscope. Based on the molecular abnormalities, the disease behaves biologically differently, and the survival characteristics are often determined by the sub-type. That's the case with lymphoma as well as many other types of cancer."
I hear the words "survival characteristics" and interpret them to mean that correctly identifying the sub-classification means identifying the therapy that will give us the best potential outcome.
I ask if some types of cancer are more difficult to diagnose than others and therefore more prone to diagnostic errors. "Yes," Dr. Elenitoba-Johnson answers, and I ask why. "I think the most pervasive reason is accessibility to lesional material." Expounding, he tells me that it is difficult to extract adequate material from some tumors because they are located in an inaccessible location, such as the retroperitoneum where the adrenal glands, kidneys, bladder, esophagus, rectum, colon and part of the pancreas lie. With too little tissue to work with, it can be difficult for pathologists to say without question that cells are malignant or benign.
Dr. Elenitoba-Johnson adds, "So being able to interpret whether you have a malignant condition depends on what we receive. And sometimes the material is not appropriately processed and that handicaps the pathologist who reviews the material microscopically."
I then ask if malignant cells can sometimes appear to be benign and vice versa. Again, the answer is "yes."
Oh dear, I think. If pathologists have trouble identifying the disease, how then can patients have confidence that the best treatment will be chosen? Dr. Elenitoba-Johnson assures me that, at least at Michigan, several sub-specialists — several experts — review bloody peas to arrive at a consensus before a final diagnosis is confirmed.
I feel better, but I ask if pathology is as much art as science. Without hesitation, Dr. Elenitoba-Johnson replies, "I would take a contrary position on that. Pathology training is very rigorous and there are strict criteria to trigger any kind of diagnosis. All the pathologist needs to do is ask whether those criteria are present."
To show me an example, he pulls the latest edition of the "World Health Organization International Classification of Diseases" from his shelf and opens it to a random page, which happens to be for B cell lymphomas.
I feel like a very lucky student being taught by a master as he points out the necessary criteria for making a diagnosis. "See, here's the definition, and the morphologic description of what it should be if you're going to make the diagnosis of a particular category, and this is what the immunophenotype should be. You can see a table distinguishing category A from category B and these are the characteristics, so if you know these, you ought to be able to make the diagnosis. Then it talks about what the genetic profiling should be if it's going to be in one category or another."
Smiling broadly as he closes the book, Dr. Elenitoba-Johnson adds, "So I can say with reasonably good conviction that in most cases, we've moved towards a criteria-driven diagnostic categorization paradigm in pathology."
Still, I wonder why pathology reports sometimes vary. "This may come," says Dr. Elenitoba-Johnson, "from one pathologist's understanding of a particular disease presentation profile versus another pathologist's understanding." All the more reason, I think, to make sure that people who are looking at our bloody peas see our particular type of cancer day in and day out. And have every piece of technology at their disposal to perform the most sophisticated tests.
I ask Dr. Elenitoba-Johnson what he would say to a patient who has received conflicting pathology reports. Empathetically, his eyes reflect the vulnerability that patients feel in this situation. "This is a difficult position to be in, and from a patient's position, it's very confusing." Expounding, he explains certain highly sophisticated tests that can identify molecular characteristics to support one position or another and thus break the tie. He also reiterates that sometimes the material pathologists receive is inadequate, and they may request an additional biopsy if a lesion is accessible.
It's easy to see that Dr. Elenitoba-Johnson knows, crystal clearly, that each small sliver he sees represents a whole human being whose life matters. "At the end of the day," he says, "all that we do is offer better patient care. The reason of our existence is for patient care to be offered at the best possible level. If we're coming up with any diagnosis that has the kind of important implications like cancer, it had better be correct."
I couldn't agree more. So know thy pathologist. He or she is the most important doctor you'll never meet, but you have every right to ask for his or her credentials.
As I walked to my car after our interview, I prayed that I will never need another biopsy, but if I do, I know, with certainty, that the next bloody pea will be analyzed by expert specialists who will indeed treat it as their own. And I'm comforted that my future will be in such good hands.
The University of Michigan Pathology offers consults. Learn more by calling 800-862-7284 or visit the website.