Never bet against occam, p.4
Never Bet Against Occam, page 4
This paper was followed less than a year later by another report of a similar discovery by a team led by the pre-eminent mast cell researchers Cem Akin at Harvard and Dean Metcalfe at the U.S. National Institutes of Health (NIH), who often collaborate with Dr. Valent. Their report, too, described a small group of patients suffering anaphylaxis of unknown cause (“idiopathic anaphylaxis”) whose marrows were found to harbor only small populations of abnormal mast cells which did not qualify them for being diagnosed with systemic mastocytosis.
Oddly, though, the particular mast cell abnormalities found in the patients described in these two reports were largely the same as those found in patients with systemic mastocytosis, thus raising a new question: if the patients in these two reports and patients with systemic mastocytosis had largely the same abnormalities, then what else was different between the two groups that produced far more mast cells than normal in those with systemic mastocytosis but only a few (if any) more mast cells than normal in those with MMCAS and idiopathic anaphylaxis?
The likely answer appeared in yet another paper that was published, around the same time (spring 2007) as the Harvard/NIH report, by a team led by pharmacologist and geneticist Gerhard Molderings at the University of Bonn. Dr. Molderings and his team had examined 17 patients with a form of systemic mast cell activation disorder in which GI symptoms were pronounced, but the testing they did on their patients was different in an important way from the testing done by the other two teams. This different testing focused on genetic mutations in the abnormal mast cells. For almost a decade by that point, it had been known that one key mutation (called D816V) was present in a key protein called KIT in abnormal mast cells in the great majority of patients with systemic mastocytosis. KIT is thought to be the dominant regulator of the mast cell’s functions. In their 2007 reports, both the Vienna team and the Harvard/NIH team did the standard (“PCR,” for polymerase chain reaction) testing looking specifically for the KIT-D816V mutation and found it in some (not all) of their patients, but they didn’t do any other testing looking for any other mutations (in KIT or any other mast cell regulatory gene).
Dr. Molderings and his group, though, did far more extensive testing for mutations in KIT in mast cells they extracted from their patients’ blood. The Bonn team determined the full sequence of KIT messenger RNA (mRNA). (When cells want to make a particular protein out of the genetic blueprint for that protein contained in the corresponding gene, they first translate the DNA sequence in the gene in question into a complementary string of nucleic acid molecules called mRNA, and then that mRNA string is transcribed (by the cell’s protein making units called ribosomes) into the corresponding string of amino acids which is the desired protein.)
Lo and behold, when the Bonn team fully sequenced mast cell KIT mRNA in their patients with mast cell activation syndrome (looking for any and all mutations instead of just the D816V mutation sought by the Harvard/NIH team), they found roughly 35 different mutations – interestingly, none of them D816V – in the mast cell KIT mRNA sequences in their 17 patients. That’s right: 100% of their patients not only had mutated KIT but in fact had multiple mutations in KIT, but it seemed that virtually every patient had a unique set, or pattern, of KIT mutations (just not the D816V mutation almost always found in mastocytosis). (The Bonn team later (2010) published another paper showing even more mutations across the larger group of patients they had found by that time, and again virtually every patient had a unique set of multiple mutations in KIT – including, in this report, D816V, though just in a single patient. Importantly, in this study they also sequenced KIT in a set of seemingly healthy control subjects, finding only a few mutations in a few of them. Perhaps the mutations found in the healthy controls were biologically insignificant, or perhaps these “healthy” controls would eventually develop MCAS symptoms, but regardless of that matter, the bottom line was that there was a clear difference between the few mutational findings in the healthy controls vs. the copious mutational findings in the MCAS patients.)
At this point things were starting to click together for me. It appeared that abnormal mast cell activation was associated not only with the KIT-D816V mutation classically associated with systemic mastocytosis (and might it be the case that actually there were usually a lot of mutations in systemic mastocytosis beyond KIT-D816V? later research would indeed find that to be the case) but also with a wide variety of other mutations which evidently aren’t associated with the significant excess accumulation of mast cells seen in mastocytosis. It certainly wasn’t clear yet whether the mutations were the actual cause(s) of the abnormal activation, but I nevertheless found the association intriguing and provocative.
An important puzzle, though, remained for me. Like most physicians, I had been taught that, under normal circumstances, mast cells – present in all tissues, albeit in scarce numbers and sparse distributions – produce and release small amounts of very potent biochemical signals, generically called mediators, which interacted with other cells and tissue to cause adjustments needed to maintain a state of health. The problems in mastocytosis, I was also taught, not only stemmed from there being too many mast cells but also that they were releasing their mediators inappropriately, causing other cells and tissues to make adjustments which were unnecessary, counterproductive, and causative of illness. However, like most physicians, I was taught that the list of mediators produced by mast cells is pretty short. In my textbook readings, the lists I saw contained roughly 10-20 different mediators. This was a dilemma for me. How could a disease capable of presenting with so many different symptoms be due to the inappropriate release of only a couple handfuls of different mediators? It just didn’t make sense…
…until my searching for more information about mast cell mediators led me to discover the fabulous “COPE With Cytokines” online encyclopedia of information about “cytokines,” some of the potent biochemical signals used by cells to influence the behavior of other cells. (The array of mediators produced by the mast cell includes cytokines as well as other molecular signals that belong to other molecular classes assigned names other than cytokines.) COPE, as I learned, is the remarkable result of the almost obsessive (meant kindly!) work over two decades by Dr. Horst Ibelgaufts, who trained as a molecular biologist in Germany and has come to teach the subject at a medical school in, of all places, the Philippines, where he has continued to work on COPE, surely a labor of love since it seems that, unfortunately, few of COPE’s users contribute to support the site. (He deserves better; support him if you can. I suspect that, for the most part, a little bit of money goes a long way in the Philippines, but at the same time, certain resources that are critical to him, like Internet bandwidth, are probably a lot more expensive than in first-world countries.)
COPE’s listings, as with most encyclopedias, are arranged alphabetically, so I clicked on M, and down amongst the long list of entries beginning with M, I found an entry for mast cells. A click on the link took me to a page that initially looked innocent enough – a few paragraphs of well-referenced, dense text generally describing mast cell function, below which was the beginning of the (similarly well-referenced) list of cytokines known to be produced by the mast cell.
I reviewed the introductory text and then started scrolling down through the list of mast cell cytokines – and scrolled, and scrolled, and scrolled, and kept scrolling, my eyes getting steadily wider. Clearly, the world of mast cell function I had previously been taught was but the tiniest portion of the true, and far more complex, world of mast cell function. Now I could see how it might be that different patterns of mast cell mutations (in KIT and perhaps other mast cell regulatory genes, too) could lead to different patterns of aberrant release of a large spectrum of mast cell mediators, in turn leading to not only a large spectrum of clinical presentations but also a large spectrum of responses to therapies targeted at mast cells and assorted mediators.
Out of curiosity, since each of the cytokines listed in that encyclopedia page was hyperlinked to another page in the encyclopedia chock full of information about that specific cytokine, I started clicking on each link, starting with the first, and continuing my reading to learn more details about the function of each cytokine.
You can imagine my surprise as I read the page of information about merely the second cytokine on the list, activin A (which I had never heard of before), learning that one of its activities is promoting red blood cell growth. One of the things about Shelly I could not understand is why she had become polycythemic (i.e., produced too many red blood cells) despite an unremarkable epo level. Here, now, merely at item #2 in a list of more than 200 entries, was a cytokine product of the mast cell that had potential for causing Shelly’s polycythemia. There was no need to find a cause for her polycythemia that was distinct from causes of her other problems. Instead, it might take just one particular mutation, or set of mutations, in her mast cells to overproduce activin A (or any of the several other mediators made by the mast cell which also directly or indirectly promote red blood cell growth), among a spectrum of mediators, to simultaneously cause her polycythemia together with her other symptoms. (Interestingly, at the time of this writing there now is a pharmaceutical company that is trying to develop an activin-A-like product into an epo-like drug that will boost hemoglobin and red blood cells in anemic patients.)
On a hunch, I then searched the literature to see if norepinephrine, too, was a known mast cell product. Indeed, I soon found not only a paper demonstrating exactly that but also another paper describing how mast cells produce and release a hormone called renin which, through a chain of reactions, then leads to increased norepinephrine in the blood. Just as with activin A and polycythemia, there was a way to connect mast cell activity with increased norepinephrine. This didn’t prove, of course, that it was inappropriate mast cell activity causing increased norepinephrine in Shelly herself, but at least I wasn’t just imagining/hoping that there would be such a connection. Instead, such a connection was already known to exist in at least some patients. And while it remained possible that something other than inappropriate mast cell activity was causing Shelly’s norepinephrine level to rise, the odds – Occam’s Razor, about which I’ll have much more to say later – said it was more likely that the elevated norepinephrine was due to the same root explanation causing her other problems than due to some other explanation.
A systemic mast cell activation syndrome – a “new” disease – was now really beginning to look like the cause of Shelly’s “weird” illness with JAK2-normal polycythemia and other features, but the number of times in a doctor’s career when he discovers a “new” disease in one of his patients is usually zero, so I knew I was still far from proving the case in her. More evidence had to be found.
The Seventh Visit
It was now the second week in July. On Shelly’s seventh visit to my center, she saw not me but one of my colleagues in the dermatology department. She told the dermatologist the same story she had told me and also noted that these rashes were “extremely” itchy (“pruritic”) when they came on. She felt she had some of these rashes scattered about her legs at the time of the visit, and the dermatologist obtained a biopsy and sent it to the pathologist with instructions to do the special processing needed to look for mast cells (since normal processing of skin biopsies by pathologists usually won’t reveal mast cells even if they’re present).
The Eighth Visit
Shelly returned next to see me a week later. She told me that after enjoying that brief period of feeling quite well, her symptoms had returned with a vengeance beginning a week before her visit to the dermatologist, finally simmering down somewhat the day after the visit to the dermatologist. She said the symptoms this time included “waves of nausea 24 x 7” which were not relieved with scopolamine (an anti-nausea drug commonly used for seasickness) or famotidine (a “histamine H2 receptor blocker” used mostly to reduce stomach acid), episodes of assorted spots of intense itching on exposure to heat (whether sun or bathing), insomnia, fatigue, intermittent dizziness and lightheadedness, intermittent fast pulse (“tachycardia”) upon minimal exertion, soaking sweats at night, decreased exercise endurance, diffuse swelling (more so in the legs than her arms), and a couple days of great difficulty initiating her urine stream. The day after the visit to the dermatologist, all of these symptoms suddenly vanished, but then, about four days later, she developed shortness of breath and tachycardia with minimal exertion, and on attempting to exercise that day, she found a blood pressure during exercise of 180/104 compared to her normal 140/76 with exercise. By that evening, though, she once again felt completely fine and had remained so since.
The physical exam at that visit showed only the slightest traces of the leg rash that had been biopsied. The biopsy wound seemed to me to be healing well, though Shelly felt it was taking much longer to heal than she had typically experienced with small wounds in the past.
Her hemoglobin that day had increased yet a bit further, to 11.6, almost normal, though frankly, with her wide extent of inflammatory symptoms, I would have expected her to be more anemic, as inflammation has long been recognized to cause anemia, and she certainly was markedly, chronically inflamed. The serum tryptase level was in the middle of the normal range (certainly not typical for systemic mastocytosis), and so was the urinary N-methylhistamine level – though I again noted in her chart at that time that she had been feeling well when the specimens were collected, and my reading on systemic mastocytosis at that visit had told me that these markers would be expected to rise and fall in accordance with how she was feeling. The pathologist reported the biopsy of her rash was showing “dense infiltrate urticaria” (exactly what one would get with a skin reaction to a noxious chemical or an insect bite), but there was no increase in mast cells. However, I saw that he had not used the tryptase and toluidine blue stains that I had requested. I spoke with him, and he said those stains were not readily available to him and that he had felt the stains he had used, Giemsa and chloroacetate esterase, would yield equivalent results. That was OK by me; there was no reason for him to not know his business.
I noted another bit of strangeness, too, in the labs that day. Her iron stores were not improving. It had been more than four months since her last phlebotomy, and although her hemoglobin was returning to normal, there was no sign whatsoever that her iron stores were improving. To be sure, I wouldn’t have expected her to rapidly develop marked improvement in the iron deficiency she had shown at her first visit because she was using iron to make the increasing amount of hemoglobin shown in her blood – but there nevertheless should have been some improvement in iron stores, and yet there in fact had been none.
As there was nothing about her history to suggest an ongoing bleeding issue of any sort, in my chart entry that day I wondered whether intestinal mastocytosis might be interfering with her ability to absorb her dietary iron. I decided to check an oral iron absorption test that day. It’s a simple test in which a baseline iron level in the blood is determined and then the patient drinks a certain quantity of liquid iron, after which the iron level in the blood is re-tested at certain intervals. If an increase in the blood iron level of at least a certain degree isn’t seen within a certain amount of time, then the patient clearly is not properly absorbing iron, and the question then becomes “Why?”
I also pondered that day about checking a brain MRI for any signs of a pituitary gland issue that might explain her malaise, insomnia, nausea, and swelling. I decided, too, to check levels of markers for a class of cancers called germ cell tumors, which sometimes can cause odd symptoms despite having only very small sizes. I also began thinking about doing a bone marrow biopsy, a key test for diagnosing the “systemic” form of mastocytosis in which significantly increased numbers of mast cells are found in the marrow.
The Ninth Visit
Shelly returned for the ninth time a month later, in mid-August. She told me she had just gotten over another “spell” of her symptoms that had started about four days earlier and lasted about two days. Symptoms had included fidgetiness, nervousness, inability to sit still, exertional shortness of breath (“dyspnea”) and tachycardia, nausea, indigestion, disrupted sleep (for example, she would sleep for three hours and then suddenly wake for a 3-4 hour period afflicted with pruritus, tachycardia, and indigestion), unsteady balance, and tachycardia and hypertension. She also noticed onset of a very pruritic rash near her belly button; she brought out-of-focus pictures suggesting a 2-3 cm wide lesion of a central clear blister surrounded by a red base. The day of the visit was one of her better days. She had smartly taken advantage of the just-passed spell to collect another 24-hour urine for N-methylhistamine and had submitted it to the lab. Thanks to some additional reading I had done by that point on mastocytosis markers, I asked the lab to also test the specimen for a prostaglandin D2 level. However, I found out that even though Shelly had been very careful to keep the specimen chilled throughout collection and transport, the lab personnel had left the specimen on their desk, unchilled, for about three hours before beginning to check it in, so I wasn’t sure I was going to get an accurate level of PGD2, a notoriously short-lived, heat-sensitive metabolite.
