Two different headlines:
"Fifteen months after he was initially diagnosed with brain cancer, Kennedy succumbed to the disease at age 77."
ABC News, Aug. 26, 2009
"The Remarkable Cancer Treatment That Helped Jimmy Carter Combat Brain Tumor."
ABC News, Mar 7, 2016
Did Jimmy Carter combat* brain cancer? Absolutely not. He never had brain cancer to begin with.
You may have heard of lymph node metastases. Cancers, specifically carcinomas, will usually spread first into the nearby, draining, lymph nodes**. But sometimes after this first stop a tumor will spread to other tissues and organs. These may be in an area remote from the original tumor - how Jimmy Carter's melanoma spread to his brain.
An example of a lymph node metastasis almost replacing the entire node. On the right is a dark blue crescent shaped area, the residual lymph node. The larger purple deposit is the tumor.
There is no universal cancer. After cancer leaves its original site it persists as itself. Its identity is highly specific to where it started - in an organ or even specific to a body site. Molecular information is now telling us that each person's cancer is specific to even him or herself.
When cancer spreads, (we say it metastasizes), it is unusual for it to lose its original markers***. This consistency is good for those of us who have to diagnose these tumors, because they leave lots of clues. No matter where they go, we can usually tell the tumor's original address and identify the tumor cells. This kind of sleuthing makes pathology interesting.
First off, we look at where the metastasis is located. Of course all cancers can and do spread everywhere, but somethings are predictable. For example, colon cancer metastasizes to the large umbrella organ that drains everything in the abdomen - the liver. Lung cancer spreads to the brain. Renal cancer spreads, well, everywhere, including odd places like the sinuses. Melanoma is the master of metastasizing, in part because it came from a lineage of cells made to travel. As part of normal development embryonic melanocytes migrate out from near the central neural tube (the embryonic spinal cord) and disperse widely throughout nearly every organ. An inherent part of what makes a melanocyte a melanocyte is this ability to spread. We shouldn't be surprised then that malignant melanoma easily metastasizes.
Sometimes patients don’t know they had a cancer until they have a metastasis. We say they "present" with a metastatic tumor. We, the medical team, then go back and fill in the first part of the story, much like a writing a prequel.
Here are some cells found within a blood vessel, in transit to their remote destination. The tumor cells are remarkably larger than any other cell in this photo.
Based upon the site of metastatic spread, radiologists and oncologists make their best guess as to what kind of tumor it is. But it takes a sample of the tissue, a biopsy read by a pathologist, to confirm the identity.
I get asked this clinical question in the form of a slide. I make a point to not review the age of the patient, the site of the biopsy, nor the clinical history before scrutinizing the slide. That way, the images form my first impression.
The first thing I do when confronted with a tumor that may be a metastasis is to evaluate how the cells are arranged - the pattern. Does it form glands? Is it in single cells? Are the cells arranged around a central area of dead tissue? All of these items point me towards a set of possible answers what we call our 'differential diagnosis". The differential diagnosis is a list of possible diseases to consider. It's good to have a broad differential, so that you don't overlook something. But, you don't want to be too broad, and waste time and money reviewing stains to exclude possibilities that are unlikely.
Here is an example from my practice this past month. The clinical suspicion was a cyst.
I saw this tumor deep in the skin. Even on low magnification, this pattern alerted me to the idea that this was likely a metastatic tumor - a large deposit deep in the dermis (the deep part of the skin, between the surface and the fat). There are not too many skin tumors that look like that.
On higher power examination (at a closer look) the cells look large and different from one another. I found a mitotic figure. These features are the signs of a cancer. My impression favored a metastatic breast cancer, but a lung cancer was also in my list of possibilities.
Then, I read the patient's history. She was a middle aged woman with no known cancer elsewhere. If she had a history of another cancer, I might have been able to diagnose the metastasis right then, as: Metastatic carcinoma, consistent with known primary cancer from X organ. However, with no history, my responsibility to identify this was greater. Whatever this tumor was, it was going to be startling news for everyone.
To help clarify what this was, I ordered a panel of stains. Immunohistochemical stains are an important tool for pathologists. This is a technique that uses antibodies to attach to markers on cells. Each cell in our body is coated with thousands of markers. In the lab we have hundreds of relevant antibodies that can search out markers, ideally markers that are unique to specific cell lines. These antibodies are labeled with a chemical that will turn brown (usually) when they find the matching marker.
Stains have several purposes: they give information about what is on the cell to that could respond to specific drugs, they can tell us how rapidly a tumor is growing, and, as in this case, they can identify the cells of a tumor type.
I chose stains for markers found on breast cancer cells, markers on lung cancer cells, and markers that might be found if this was a rare skin tumor of sweat glands. I considered this my 'first round' of stains - if nothing turned up with these choices, I would have to revisit my differential diagnosis list, and likely ask a colleague for their opinion.
This panel shows the positive (in this context positive means reactive) stains. Clockwise, the stains are Mammaglobin, Progesterone, Estrogen and GATA3. Mammaglobin antibodies highlight the outside of the cell, directed towards cell membrane markers. The other markers highlight the nucleus, because they are directed towards nuclear receptors.
About 24 hours later, I got my answer. These cells stained positively for all the breast cancer markers I used and nothing else. These markers, used together and with negative stains to rule out a lung tumor, meant that with near 100% confidence we have identified the original location of these cells. With this certain identification, we could name this tumor: Metastatic carcinoma, consistent with a breast primary tumor. I called this patient's dermatologist who performed the biopsy and let her know that this was not a cyst and that the patient should be referred to a breast surgeon. The patient went on to see an oncologist and surgeon, where her breast cancer could be properly diagnosed and where she could be screened for other potential metastases.
She did not have a skin tumor, even though she presented with a tumor in her skin. Her breast cancer was growing, and had metastasized to her skin without being detected in the breast tissue. Just like President Carter, she did not have separate cancers of each organ. Her doctors will treat her for one cancer, not separate cancers.
If it is cancer when it metastasizes, what about when it never leaves it's original spot? It's a pathology zen koan – if something stays in one place, is it cancer? Out of control local growth is technically considered cancer, but we might also categorize these tumors as benign. The contradiction is that you could have a benign tumor that could be deadly. Like a meningioma, a benign tumor can be a damaging, and will require local treatment. On the bright side, such tumors will likely never spread.
There are unanswered questions and some questions that have answers but as of yet no corresponding use in medicine. Why do tumor cells metastasize to one organ and not another? Do all tumors shed a few cells into the blood, but only in the 'right' conditions are these cells able to take root in a new location? As we screen people more intensively, we find more examples of subtle problems - what is the meaning of a single cell in a lymph node? What is the meaning of circulating tumor cells? Pathologists are the ones asking and finding the answers to these questions.
This kind of pattern recognition is what I love about pathology. The thrill of correctly naming a tumor, especially when it is acting as an imposter, is real. I was an exceptional player of the childhood memory match game called Husker Du. I always remembered where the matching image was. This recognition skill is at the root of a lot of pathology. I won't say that my job is like a game, but it is a great feeling to know your skills at matching and identifying images can help save someone's life.
*Let's leave the battle terminology for another post - I don't like thinking of cancer as a war and think this analogy doesn't serve patients or physicians well.
** The very first lymph nodes to receive a metastatic tumor are called Sentinel lymph nodes; standing guard like sentinels they drain a particular organ or body area.
**Rarely a tumor might lose it's original markers. So with these cases, our panels of immunohistochemical studies and inspection of the cells and pattern do not lead us to one specific origin. When we don't know where the tumor comes from, we'll call it so - cancer of unknown origin.