In early May 2017, I interviewed Dr. Nancy Klimas, head of the Institute for Neuro Immune Medicine at Nova Southeastern University. I first learned of Dr. Klimas when reading the book Osler’s Web, which is a history of Chronic Fatigue Syndrome, its “discovery” in the 1980s and 90s, and the attempts at definition and research. Dr. Klimas has been doing CFS research and clinical treatment since 1987. (Currently, as of May 2017, Dr. Klimas’ group is doing a CFS study involving genetic testing.)
I came down with CFS symptoms two years ago, in 2015, at the age of 37. The reason I wanted to talk to Klimas is that, while I’ve read a lot about CFS, it’s been hard to separate the good information from the bad, and the new information from the outdated info, so I wanted to speak to someone respected, who was on the front lines, to get a sense of the current state of the research and opinions on it.
This talk has been edited slightly for clarity. Also, it’s possible I’ve made some errors in transcription due to my ignorance of medical terminology; those would be entirely my fault, not the fault of Dr. Klimas.
Zach Elwood: There are lots of theories about the overlap between CFS and ME, and CFS and fibromyalgia. I know some people think they’re all a manifestation of the same condition. Do you have any thoughts on them being discrete entities or just being different manifestations of a similar process?
Dr. Klimas: I think ME and CFS are iterations of the same thing. And very much case-definition-dependent. I know some people in England feel very strongly that ME and CFS are not the same. The reason why is because the CDC case definition for CFS does not require exercise-induced relapse; it didn’t have any requirements, you just had to have four basic symptoms. So it cast a wider net than the ME case definition. But there’s a half a dozen case definitions.
The person who’s researched this the most is Leonard Jason; he’s got a whole series of papers that look at caseness.If you go to his website, he’s at DePaul University, you’ll see that he’s spent a great deal of time trying to understand the iterations and case definition. He has a nice paper on the IOM’s report on case definition as it relates to the other case definition.
Zach Elwood: That was my understanding, too. It sounded like it was a matter of all these different case definitions and people going in different directions due to that. Do you have any thoughts on sudden onset versus gradual onset. I’ve seen some thoughts on there being potentially two different kinds. Do you have any thoughts on that?
Dr. Klimas: So you bring up the question of subgrouping. And one of the early requests from the earliest case definition, the CDC case definition, is that people try to describe very well in the papers we publish what definition we use, and if we use any kind of subgrouping strategy. And one of the earliest [inaudible] was acute versus chronic onset. So when you try to do these studies, they’re tricky, because it’s all about what you call acute or slow. If an illness evolves over a month, is that an acute onset or is that a slow onset?
So I would say that, first, I certainly wouldn’t separate them. I’m more of an umbrella person myself and not a lumper. Because my concern is that if you split—Let’s say there are a million people with CFS and we discover 40% of them are auto-immune. Is that a different illness? Or is that a subgroup of CFS that is auto-immune? I’d like to claim the latter so that we can count our successes as we go alone. Because in the end we’re going to beat this thing by subgrouping and having directed therapy by subgroup. But it’s not going to be by gradual versus sudden onset. It’s going to be by autoimmune or viral, or toxic exposure; it’ll be all these subgroups that are based more on their mediators, the things that set them off, what caused them to be sick.
Our research is more focused on why do you still have a chronic illness; why didn’t you recover?
Zach Elwood: It sounds like you believe it can have varying causes that lead people to this state. Would you say that’s accurate; that it could be many things that kick it off?
Dr. Klimas: That’s right.
Zach Elwood: I just finished the book Osler’s Web, which I know is old and has a lot of old thoughts in it, but one thing that struck me is how it seems like they explored a lot of viral ideas of the disease back then, in the 80s and 90s. I was wondering; is that still an active area of research or have people kind of ruled out viruses are a potential cause?
Dr. Klimas: Viruses are still in play. They’re absolutely in play. They’re not in play in the way you’re thinking.
There are two places where viruses are in play. One is: do you have a chronic [inaudible] infection; if so, why didn’t we figure that out yet? The answer to that would be: because it’s in a protected department, an area we don’t usually assay.
The person who’s doing that work is John Chia in Los Angeles, who’s an infectious disease doctor. He’s great. He’s been looking at enteroviruses, particularly the Coxsackie virus. He’s published a number of papers; four of them so far, where they biopsy the stomach or colon or whatever and they look for DNA fingerprints or look for viral staining in the cells. Very well-powered studies; couple hundred patients in each study.
Anyway, he’s after enteroviruses and he’s of the opinion that people like Ian Lipkin aren’t going to find them because they’re not looking in the right place. You’re not going to find the signature for that virus in the blood if it’s not circulating in the blood. So he’s equally convinced they won’t find them in the microbiome because they’re not looking for viruses in the microbiome, they’re just looking for bacteria. That’s one thought.
The other place is in the world of re-activated viruses. There you don’t really think the virus started the problem or was the ultimate cause, but that the virus was screwed up enough so that the virus re-activated and now are contributing to the degree of illness and the types of symptoms you have. So in that theory, you would wonder: the symptoms would again relate to where the virus is. So in John’s case, it would be in the gut; enteroviruses in the gut and muscle. In the HHV6, which is probably the one with the best data for being re-activated, it’s in the blood stream but it’s also in the brain, the brain compartment. Epstein Barr virus, which we all carry around, can reactivate.
Our group just published a paper with Ohio State, Dr. Marshall, that suggested that viral DNA or viral RNA in an activated cell codes for viral “stuff” and sometimes that viral stuff is bad for you. For instance, in the paper we published, we looked at EBV dUTPase, an enzyme that’s specific to EBV. But when it circulates, it causes all kinds of havoc in different kinds of systems [inaudible]. So high levels of dUTPase would interfere with the normal regulation of a number of metabolic pathways and systems, such as the endocrine system.
Our group is doing research looking within the cell, using a single-cell-biology way of looking, and we can look at basically what’s happening in the cytoplasm and mitochondria and the nucleus of the cell, and in those we’re looking at EBV peptides in the cell, which we can [inaudible]. So that’s interesting.
So viruses are still in play. Are they the be-all and end-all? No, they’re part of the big picture. And you have to think of this in terms of the big picture instead of a single button you can push to make it all go away.
Zach Elwood: It seems like in the 80s and 90s, would you say it’s fair to say people were looking for a quick way to find a cause for this and when they couldn’t find that, people were frustrated, which led to people not thinking it was real. Would you say that’s fair to say some of the dismissal was due to not finding a cause quickly?
Dr. Klimas: I think it’s way more complicated than that. That book you read [Osler’s Web] describes the political aspects of that. That work was being done at a time when psychiatry was being well-funded and biological sciences were much less so. So most of the people doing that work then were doing it in very small samples because we didn’t have enough money to publish the big number studies and they weren’t getting funded. That was the environment of the time.
I think right now, because there’s a new interest, particularly around the microbiome, they might discover that we’re allowed to go back there again and think this through again. And also the technology is way better. God Almighty, what the last five years have brought us. We can do whole genome analyses now and just probe for viral signatures in the DNA and RNA of the cell; it’s much, much better than trying to culture something out and probe with PCR; all the things we had that were all we had to rely on before.
Elwood: If you had to name a few of the CFS/ME researchers right now. You’ve named a few of them, and along with yourself, could you name a few more as far as top people, that would be interesting for people to look into what they’re doing? I realize that’s probably a hard question.
Dr. Klimas: There are a lot of them. It’s too broad a question. There are a lot of people doing good science right now. This last consortia application I was doing; I think there were nine centers that responded. But the big one is Stanford; they have a major donation from a private donor, it’s a very nice center.
Cornell, with Maurine Hanson; they also are establishing a center of excellence.
I told you about Ian Lipkin.
The Whittemore Peterson Institute in Nevada is still plugging away.
Jarred Younger, in Birmingham, Alabama doing very important work, and very clinically oriented.
The Open Medicine Foundation could be grouped with the Stanford group I mentioned.
John Chia I mentioned.
In Canada there are some good groups.
But yeah, there are a lot more…
Elwood: Getting back to the history of it, I read a lot about the immune system dysfunctions that were shown in CFS sufferers. In Osler’s Web, it talked about it a lot. Have those things been proved definitely to be the case? Are there verifiable immune disruptions in CFS sufferers?
Dr. Klimas: So the Institute of Medicine report was a very conservative look at this literature. It’s about 2 years old. Even this IoM report [https://www.nap.edu/html/19012/MECFScliniciansguide.pdf: see p.13 for some info on this] said that and they were really not willing to stretch very far. But they said that the evidence for natural killer (NK) cell dysfunction is very strong. In truth they would have put it in the case definition had the assay been widely available in a reliable way.
But anyway, it’s a very difficult assay to perform and it doesn’t travel well so that one is not considered a rule-out for diagnosis. But it’s certainly a very reasonable thing to understand why it happens. Cytokine papers, if you total them all up and do the meta-analysis, the preponderance of evidence would be that yes, there is inflammatory cytokine expression and yes, there’s an abnormality of the cytokines, the Type 1 and Type 2, TH1 and TH2, cytokines; that they’re abnormal. And I think that that’s very important work.
A paper that came out right after the IOM report; I think it might have swung it if it had been a couple months earlier, which was Mady Hornig’s paper; she had a large study where she said there were inflammatory cytokines but they were more evident in the earlier years of illness. So she suggested that there was a split at the 3-year mark for one sort of immune signature versus another immune signature.
Elwood: I’ve read also that the low sedimentation rate (ESR) is one key indicator. I’m not sure if that’s still a respected observation, but do you know much about that?
Dr. Klimas: I don’t know what I think about that. It doesn’t make a lot of sense because there are other papers that say C-reactive protein (CRP) is a little elevated, so you could argue the opposite direction. I don’t think it’s agreed upon. I don’t think the literature supports that finding. Although the elevated C-reative protein; there’s clearly a subgroup with elevated CRP. But that might be the other immune group; we don’t know. But we’ll find out. Time will tell.
Elwood: Is it surprising that it’s been so hard to get answers for this condition? Or is that understandable considering the complexities of it and the lack of funding, maybe in the early days?
Dr. Klimas: This is a hard nut to crack. Obviously funding matters, and funding is still poor, don’t get me wrong. It’s not good. The fact that they’re going to create these centers is a beginning. But I study another illness, Gulf War Illness, that has $40 million a year funding, and we’re much further along. We’ve had 22 clinical trials, we’ve had all this work done and it’s been fantastically aggressive and supported by a decent funding track and an innovative way of thinking about funding. It’s mostly DoD [Department of Defense] money, a little bit of VA, but mostly DoD, because there’s something called the Congressionally Directed Medical Research Program, CDMRP, and with that, a congressperson can sponsor an earmark to a specific illness. We in the CFS community; we haven’t learned how to use that correctly, and it’s certainly another avenue to get directed funding.
But we’ve stuck around the $3 million a year mark for a long time. But these new centers; these will bring us up to $9 million; it’s not bringing us up a lot. It’s nowhere near the point where we can move what we know through the clinical trials. If the end game is treating people based on new knowledge, we’re not there yet. We don’t have the funding avenue to get there; there’s no place to go to get that money. So that’s a problem.
We need to promote the science better. We should be using the Gulf War Illness research as a template. The CDMRP had a consortia for Gulf War Illness that resulted in around 21 new applications from people related to the consortia using their resources. That’s what you want with a consortium; you want it to be the epicenter, giving people access to content experts and clinical experts, and samples tied to really fine clinical work.
Politically, we’ve got a ways to go to get where we need to be.
Elwood: Is there a comparable condition that’s been as hard a nut to crack?
Dr. Klimas: I think the two best examples would be Multiple Sclerosis, another nerve inflammatory disorder, that presents in a multitude of different ways and has different courses, and that probably, underneath it all, has different pathogeneses. The worst MS is clearly auto-immune, but it’s not so clear that all the MSes are autoimmune; there’s still a viral theory going around. And of course, before there was a diagnostic test, the CAT Scan, MS was called hysterical paralysis; they thought it was psychiatric. So MS is a great example that follows the path that we’re following.
Gulf War Illness is a great example of an almost identical illness, where we clearly know the inciting event was toxic exposure. The difference between Gulf War Illness and CFS is by definition GWI had to have started in 1991, you had to have been in that Gulf War, and in that war there was—the stars and the moon aligned in such a way that they were exposed to a tremendous number of toxins in a very short window, and there were neurotoxins included. 1 in 3 veterans in that first Gulf War are sick today, 25 years later. So that was a powerful model of how knocking the brain around with an inflammatory process can cause a long-term chronic illness.
So a couple good examples. And that’s not to say I’m not optimistic about the chances of CFS. I’ve been working on this for a very long time. And we have some very exciting research that we think is trying to go for a cure and not just for effective treatment.
Elwood: You had mentioned some of your research earlier. Is that related to the genetic study you’re doing now, or is that a whole other study?
Dr. Klimas: We have a lot of studies going on here. It’s not just me; we have this amazing group. We have a genomic team, we have a computational biology team, we have a cell biology group, an immunology group, and a clinical care group. A lot of groupings. And we work in an integrated system so that the computational biology drives quite a bit of the exciting findings that we have. The gene study is a part; that’s looking at genetic risk, looking at if certain things that can put you in a subgroup
We integrate everything into a computational platform. We use supercomputers and we use something called dynamic modeling, which is where you force a small relapse and then map out what’s happening. We put people on a bicycle, because as you know exercise causes relapse, and we exercise right up to but not beyond their aerobic threshold. On average about 8 minutes. With some people it’s only 3 minutes. Depends on their fitness, their aerobic threshold. But we draw blood frequently; in the first 4 hours there are 8 blood draws, and there’s one more the next day. Then we measure all the pathways that are turned on or off genetically. We measure cytokines, we measure hormones, we measure a lot of stuff. It’s too much for any one mind to think about. So the computational biologists plug all of that data in to a supercomputer platform, and it shows what things are connected. Don’t ask me to explain how the program works, but it works. It’s not hypothesis-driven, it’s just mathematically-derived data.
And then we sort that all out and you put the horse before the cart; you look at what’s the first thing that happens that causes the other things. You look at the entire cascade of events that occur from that first moment when you just start to exert yourself. And we find very clear patterns. And so then we have a virtual model of the illness, so we have cleaner models, and then you can do virtual clinical trials and try to put it all right again. And that’s very exciting work, because it doesn’t require us to do the methodical kind of clinical trials work that we’re used to doing.
For example, we did an HIV clinical trial, we did a single drug for four years, then we did Drug A vs Drug B trials for a couple years, then we did Drug A plus B plus C vs Drug B plus C. Et cetera. That stuff took ten years before we were saving lives. It took ten years trying to pull apart the pieces and put them together. And in this kind of computational biology, you put it all together from the get-go, you say: Okay, the model doesn’t work with just one drug, there’s no one point I can push to make this better. I can make people feel a little better down that path, but I can’t make them better-better. But if I push two points and I know the timing of that and I time everything correctly, then theoretically I can reset that homeostasis, I can reset all the abnormal pathways that are causing illness to persist and try to push them back to a normal homeostatic pattern.
It sounds crazy and it’s very theoretical, but in Gulf War Illness we were very well-funded to do this, and we have an animal model and we were able to use those drugs in the animal model, and we reset the homeostatic network. And we’re moving to Phase 1 clinical trials this summer. All based on computational modeling of the illness.
So for CFS; we’ve been funded for some years to do some of the modeling, but we haven’t been funded yet to do the subgrouping, which is turning out to be important. We have some of it done, but not all of it. The gender difference is huge in both of these illnesses; it’s quite different if you’re a male or a female. Way more important than say acute onset versus slow onset. We’re wired differently and we’re talking about homeostasis, so not a huge surprise. But it is a surprise that, for example, in a woman it matters where they are in their menstrual cycle; you’d do one type of treatment if they were in their high-estrogen phase versus hypo-estrogen phase. We take all these things into account when we do our modeling. It’s very complicated but in the end we have really clean clinical trials.
The other thing we do is we only use repurposed drugs; we don’t use any pipeline drugs. Because that shaves many, many years off you guys someday having treatment. Because we didn’t have to wait for FDA approval and [inaudible] drug companies to get behind us and all that stuff.
The flip side, though, of repurposed drugs is that: who’s going to pay for that? You might have seen that the Gates Foundation funded this amazing study. They used a drug that cost $1 in women that were bleeding out post-partum, who were going to die from hemorrhagic blood loss, one of the most common causes of death of young women in the world. They found a drug that costs $1, and it reduced mortality by 30%. That’s like many thousands of women every year. They could not have gotten that funded anywhere else. No way, no how, could that study have been funded by anybody but a philanthropist. Because it’s too big for the NIH; it was a massive, multinational bazillion-dollar study, and it’s touting a drug that nobody’s going to make any money off of. How cool is that? That’s worth me not buying an Apple next time; let the Gates have my money.
Elwood: We need Bill Gates to get CFS.
Dr. Klimas: Yeah, hate to say it, but sometimes you get attention that way. Or a family member. But that was damn impressive. And that’s my point about repurposed drug; if you have a drug that’s repurposed, generic, then we can roll it out instantly. It’s approved, we know it’s safe, the FDA knows it, we can roll it through without headaches. But the downside is: I can’t go to pharma to help pay for it. I have to get it paid for by someone else.
Elwood: That seems to be a real weakness in our capitalist market.
Dr. Klimas: Not that capitalism isn’t great for new drug development. Not so great for that. My Gulf War study; I’m very realistic. I work at the VA, too, and work with these patients. I know for sure if I picked some pipeline drug or an expensive drug, there’s no way it would ever be available for these guys. So I deliberately pick drugs that are already on the formulary. And it was a royal pain for me because one of them was about to become generic and the company wouldn’t even give it to me for free. Usually you get free drugs to run the trial, but in this case I have to pay for every nickel of everything I’m spending in the trial.
Elwood: Well, it’s good work you’re doing. Speaking of, you had mentioned briefly, the metabolic work. Do you know much about the Ron Davis and Naviaux work on metabolic markers. Do you feel that’s pretty important and is it related to some of the work you’re doing.
Dr. Klimas: We published some papers 3 or 4 years ago when we were doing this dynamic modeling work that would have predicted what Naviaux’s work found, which is that this hypometabolic state would result in hypometabolic indicators in the metabolome. To me, there’s a lot of theses coalescing to confirm that. When we put people on the bike, what’s supposed to happen is their energy levels are supposed to increase; their mitochondria are supposed to generate a bunch of new energy. You’re supposed to have a little inflammatory response, but you’re supposed to counter it instantly with an anti-inflammatory response. And then a bunch of important things, mostly related to energy and anti-oxidant production; these things are what you’d expect in a healthy control.
In a CFS patient; what happens is instead of turning on energy pathways, you don’t have anything to draw from, you crank up your inflammatory response, you have this big spike in we’ll call it the “inflammazone.” Normally the anti-inflammatory stuff would kick in but instead it turns off and there’s nothing to counter this kick in inflammation. And then oxidated stress goes really, really high, and there’s not sufficient anti-oxidant response to that. All that sort of sounds what like Naviaux’s talking about with the metabolome.
Elwood: Would finding these metabolic traits of CFS sufferers give you insight into what causes the condition?
Dr. Klimas: Is it related to mediators, is what you’re saying. Yes. Everything is biology in the end, right? So there are some common symptoms, like body pain. What’s that about? Mediators could be systemic inflammation, like the cytokines we were talking about. So a different mediator might be the small nerve de-enervation, as was described by Dr. Oaklander in fibromyalgia. She found an [inaudible] to fibromyalgia that appears to be auto-immune and it was attacking the nerve endings. That would be a mediator that would be tremendously painful. Or it could be the central nervous system regulation of pain that most of the current fibromyalgia-type drugs work on.
So there are a lot of different spaces in there that can be considered biologic causes of a common symptom. It also says that there’s more than one way to get the same symptom. There’s more than one biologic cause that would result in, for example, fatigue. It makes it complicated; we have a heterogenous population, we’ve got men, we’ve got women, we’ve got some people that were clearly post-infection and not post-viral, we have persistence that affects things that would make it worse, like sleep.
And fitness, and that’s sometimes a bad word, but deconditioning is not your friend; it’s lousy for you. There are a lot of common elements to what happens to people who are chronically ill, for any reason; that illness stops them from being able to do normal activities.
So you have biologic consequences to pain, and you have biologic consequences to low energy level, and you also have mood causes. For instance, we’ve been denying this influence for 30 years, because the psychiatrists were pounding on it so much we had to push back, but the truth is, you know, 70% of people with renal failure have depression. Why wouldn’t you? This one, too. But there’s also sometimes biology to that, too. You have neuro-inflammation, you’re going to muck up your normal balance of normal peptides in your brain.
So there’s a lot going on, and a lot of it has to do with this multi-system balance, however it got that way. A while back I stopped trying to figure out how it got that way, because it felt like I was beating my head against the wall. It didn’t matter. What I have in front of me as a doctor is a patient who is that way. So now I have to figure out why do you stay that way. So I don’t really care that much anymore. I mean, it’d be great to know because that would inform my treatments. But truthfully I want to know even more about who you are today than what happened in the first 3 months of your illness. That’s my perspective.
And I’m in a different perspective than some people because I actually treat patients. That puts you in a different frame of mind. I’m in a hurry to get to effective therapy, as I suspect you are. And many people are.
Elwood: So you’re on both the clinical and research side.
Dr. Klimas: I’ve been treating CFS patients since 1987. I’m one of the early ones.
Elwood: I remember that now in Osler’s Web; it mentions you started a practice for that. That’s awesome.
Update 9/26/17: There were a couple questions I realized I had forgotten, that I wanted to ask Dr. Klimas and I thought I’d leave them here to spark further discussion, or perhaps if Dr. Klimas was able to answer. For one: the Osler’s Web book talked a lot about the brain scan abnormalities in CFS sufferers; I haven’t heard good recent information about whether those findings are still respected, or if most people have abandoned the brain scan abnormalities research.