Lombardi et al 2011 - cytokine signature in CFS

[Sasha]

Sasha, You must have seen it on the other forum:

http://www.mecfsforums.com/index.php/topic,7276.msg86605.html#msg86605

Hi Johnny - I don't visit the other forum - I saw it here on Phoenix Rising.

 

[jace]

Angela quoted my last post, but it made no sense without the quotes from the paper itself, so I'm doing it again because it seems some people haven't understood yet.

This one, IMHO, is really important. What it shows is a pattern in the immune system (expressions of cytokines) that predicts XMRV+ ME/CFS. There are also some clear pointers to help all those that fail to find XMRV. For instance:

Blood samples were drawn using green-capped vacutainer collection tubes with the anti-coagulant sodium heparin (Becton Dickinson, Franklin Lakes, NJ, USA) and processed within 6 hours of draw. Plasma samples were separated by centrifugation at room temperature, aliquoted and stored at 80 ?C until analysis. Samples were not subjected to more than 1 freeze-thaw cycle.

0/0 studies tend to handle their samples less rapidly, I have read that some freeze at -30C and also that their samples have been put through multiple freeze-thaw events.

I haven't got my head round the Random Forest Algorithm that was used, but I do know that they fed their raw data in, and the program then predicted with amazing accuracy which samples were ME/CFS XMRV+

we utilized the Random Forest data mining software package using patient status as the target variable and the cytokine and chemokine values as the predictive variables .... The final model accurately identified 128 out of the 138 controls (93% specificity) and accurately identified 113 out of 118 patients (96% sensitivity) (Table IV)

.
So in other words, this test randomizes the results and then picks out the sick people by analysing the cytokine profile.

The conclusion, in proper qualified scientific speak

Although the determination of XMRV infection is presently not a definitive stand-alone diagnostic tool for neuroimmune diseases or malignancies, a combination of XMRV and cytokine/chemokine analysis may prove to be a reliable diagnostic strategy and may assist in monitoring the success of treatment

In science, there is no "proof". The concept of certainty is anti-science. True scientists always allow for new information to change their view of reality. True science evolves.

The exciting thing is that this may be an objective tool to test how well treatments work. That is hugely important, and I am super excited about it.

 

[ahimsa]

IgA Deficiency

Hmmm, one of the few abnormal blood tests I've had is a almost nonexistent igA (Immunoglobulin A). igA is the antibody responsible for mucosal immunity. It exists in saliva, upper airway, GI tract etc. Coincidence? Anyone else test low igA?

Yes, one of my test results showed non-existent IgA. But a later test showed a normal level. Either the first test showing zero was a mistake or my IgA level changed. I'm not sure whether IgA can go up and down by that much, from zero to normal. I never got a straight answer from any of my doctors (to their credit, I did get a few "I don't knows"). I couldn't find a good answer online, either. I think there was one place that said certain infections can temporarily lower IgA. And since there is no known treatment for low IgA (unless that has changed lately?) I didn't bother pursuing this issue after the second test was normal.

On the other hand, IgA deficiency is known to run in families. My older sister has had consistently low IgA levels. She does not have ME/CFS or any of the other chronic illnesses (migraines, endometriosis, NMH/POTS) that I have. My sister fits the profile for low IgA more than I do because she gets lots of sinus infections and she has allergies. But generally she has a huge amount of energy and is always doing something - once told me she doesn't even like to sit still long enough to have her hair done.

 

[Cort]

I think there's some really interesting stuff here. Whether or not XMRV works out these are people with CFS.

It was great to see the clonal T-cell rearrangements back in the news so to speak and for them to find cytokine measures that appeared to be aligned with those findings. The cancer cluster, as I remember, was full of clonal T-cell rearrangements. Unfortunately, that study - which made such a big splash at the Reno conference - has, at least from what Dr. Peterson knew, never got completed. Perhaps it will be picked up again...if they can show a cancer subset in ME/CFS that would be big.

Consistent with a ?-? T-cell involvement, our analysis shows the up-regulation of MIP-1?, MIP-1?, TNF-? and IL-10, all of which are produced by ?-? T-cells. Recently, Gu et al. investigated the correlation between circulating cytokines and chemokines and the risk of developing B-cell non-Hodgkin lymphoma and reported that an increased level of IL-13 has a protective effect regarding the development of B-cell lymphoma, whereas patients with increased inflammatory cytokines and chemokines are at greater risk (41). This study supports the hypothesis that CFS patients are at greater risk for developing lymphoma as a consequence of their inflammatory condition.

It sound like it very well may be picked again - as it gets a prominent mention in this paper.

Then there's the HHV-6/herpesvirus -clonal T-cell arrangement association.

Furthermore, a proportion of this cohort has repeatedly tested positive for active herpes virus infections, including HHV-6A, CMV and EBV. A previous report by Lusso et al. has shown ?-? T-cells to be susceptible to HHV-6A infection, inducing the expression of CD4 de novo (35), and Vrsalovic et al. have shown an association between HHV-6 infection and ?-? T-cell clonality

I think this paper shows good avenues of research that are available no matter what happens with XMRV. These are all the 'WPI's goods'; they developed both the cancer and T-cell clonal re-arrangement findings; altho neither is as exciting as XMRV (nothing is) - they show real promise.

 

[Cort]

This is a stunning immune signature! I don't think anyone anywhere has shown such a dramatic difference in cytokine expression in CFS. These aren't little differences - they are consistently huge differences (almost too big!) compared to past studies.

You can look at this group and say "...these people are really sick".

What we really need, of course, are labs that validate each other's findings. Dr. Klimas has some interesting cytokine results which not everybody believes because prior studies usually haven't shown much. At the SOK workshop Dr. Natelson talked about his efforts and his failure to find cytokine differences.

On the other hand the tests may be getting better and/or it could be that Dr. Mikovits really narrowed in on a subset of patients and that allowed them to finally pop out like this.

Up-regulated IL-8 1045 (254) 13 (1.6) <0.0001 MIP-1? 1985 (556) 164 (41 <0.0001 MIP-1? 763 (216) 91 (19) <0.0001 TNF-? 148 (53) 13 (4.3) <0.0001 IL-6 336 (87) 29 (11) <0.0001 IL-2 113 (56) 28 (10) <0.0001 IP-10 98 (16) 32.8 (3.0) <0.0001 Eotaxin 271 (19) 95.8 (6.5) <0.0001 IL-12 289 (20) 211 (31) 0.0001 MCP-1 468 (42) 421 (41) 0.0003 Rantes 27107 (3400) 9564 (1012) 0.0018

Down-regulated IL-13 28.2 (3.6) 85.5 (6.5) <0.0001 IL-5 7.35 (0.66) 21.1 (4.9) <0.0001 IL-7 33 (11) 78 (6.9) <0.0001 MIG 48.2 (9.0) 80 (12) <0.0001 IFN-? 35 (5.9) 60 (4.3) <0.0001 IL-1RA 1010 (363) 1277 (429) <0.0001 GM-CSF 108 (23) 166 (28) <0.0001 IL-4 39.6 (3.9) 55 (9.3) 0.0003

This is pretty stunning as well

"... The final model accurately identified 128 out of the 138 controls (93% specificity) and accurately identified 113 out of 118 patients (96% sensitivity) (Table IV)

 

[alex3619]

Hi, the thing about machine learning algorithms is that they can not only find patterns, but can be analyzed to tell you what those patterns are. You can then go back and look for what is causing those patterns. I agree that even if XMRV does not hold up, this is very promising research. I suspect that what researchers should have done all along is ignore the mild cases, and go straight for severe patients. De Meirleir did that, WPI does that, who else does that? My guess is that only doctors/researchers who focus on severity or epidemics will get great results - mild sporadic cases might include a large percentage of patients who are misdiagnosed and dilute the results.

Bye
Alex

 

[oceanblue]

This is a stunning immune signature! I don't think anyone anywhere has shown such a dramatic difference in cytokine expression in CFS. These aren't little differences - they are consistently huge differences (almost too big!) compared to past studies.

You can look at this group and say "...these people are really sick".

What we really need, of course, are labs that validate each other's findings. Dr. Klimas has some interesting cytokine results which not everybody believes because prior studies usually haven't shown much. At the SOK workshop Dr. Natelson talked about his efforts and his failure to find cytokine differences.

On the other hand the tests may be getting better and/or it could be that Dr. Mikovits really narrowed in on a subset of patients and that allowed them to finally pop out like this.

This is pretty stunning as well

"... The final model accurately identified 128 out of the 138 controls (93% specificity) and accurately identified 113 out of 118 patients (96% sensitivity) (Table IV)

Yes the results are stunning though I also agree with the need for replication.

Also, I think there may be an issue with the approach used here (which equally applies to most other studies that cite specificity and sensitivity for identifying CFS): the same group of patients/controls appears to be used to derive the model and then for accuracy assessment. This will tend to maximise the results for this particular group of individuals. What's really needed is a second study taking a new group of patients and controls, and applying the algorithm derived from this first study. If the model still accurately distinguishes between patients and controls then this will look like a major breakthrough.

 

[alex3619]

Also, I think there may be an issue with the approach used here (which equally applies to most other studies that cite specificity and sensitivity for identifying CFS): the same group of patients/controls appears to be used to derive the model and then for accuracy assessment. This will tend to maximise the results for this particular group of individuals.

This problem is a big one for machine learning. It is called overfitting the data, iirc. It basically memorizes the data and doesn't generalize properly. Machine learning has its limits, but it depends on what you use it for. Any cytokine profile developed this way will require additional study to verify its effectiveness - but then that is true of just about every approach.

In order to use machine learning reliably, you typically use huge data sets, then test the results with a separate data set that were not used in development.

However, the other way to look at this is they have an hypothesis on cytokine markers. This hypothesis now has to be tested, and preferably with other approaches.

Bye
Alex

 

[jace]

Do you think this paper is worth an article on the Front Page, Cort?

 

[kurt]

This study in part seems to replicate an earlier study that they do not reference. Curious, until you consider that WPI has results that conflict with the earlier CFS cytokine study by Fletcher et al. However, good science says you survey the entire literature, and propose explanation for divergent findings. I wonder what WPI would say about this study? And I wonder what Dr Klimas would/will say, given it was her study that was omitted from reference.

J Transl Med. 2009 Nov 12;7:96.
Plasma cytokines in women with chronic fatigue syndrome.
Fletcher MA, Zeng XR, Barnes Z, Levis S, Klimas NG.
Source
Department of Medicine, University of Miami Miller School of Medicine, 1600 NW 10th Ave, Miami, FL, USA. mfletche@med.miami.edu
Abstract
BACKGROUND:
Chronic Fatigue Syndrome (CFS) studies from our laboratory and others have described cytokine abnormalities. Other studies reported no difference between CFS and controls. However, methodologies varied widely and few studies measured more than 4 or 5 cytokines. Multiplex technology permits the determination of cytokines for a large panel of cytokines simultaneously with high sensitivity and with only 30 ul of plasma per sample. No widely accepted laboratory test or marker is available for the diagnosis or prognosis of CFS. This study screened plasma factors to identify circulating biomarkers associated with CFS.

METHODS:
Cytokines were measured in plasma from female CFS cases and female healthy controls. Multiplex technology provided profiles of 16 plasma factors including the pro -inflammatory cytokines: tumor necrosis factor alpha (TNFalpha), lymphotoxin alpha (LTalpha), interleukin (IL) - IL-Ialpha, IL-1beta, IL-6; TH1 cytokines: interferon gamma (IFNgamma), IL-12p70, IL-2, IL-15; TH2: IL-4, IL-5; TH17 cytokines, IL-17 and IL-23; anti-inflammatory cytokines IL-10, IL-13; the inflammatory mediator and neutrophil attracting chemokine IL-8 (CXCL8). Analysis by receiver operating characteristic (ROC) curve assessed the biomarker potential of each cytokine.

RESULTS:
The following cytokines were elevated in CFS compared to controls: LTalpha, IL-1alpha, IL-1beta, IL-4, IL-5, IL-6 and IL-12. The following cytokines were decreased in CFS: IL-8, IL-13 and IL-15. The following cytokines were not different: TNFalpha, IFNgamma, IL-2, IL-10, IL-23 and IL-17. Applying (ROC) curve analyses, areas under the curves (AUC) for IL-5 (0. 84), LTalpha (0.77), IL-4 (0.77), IL-12 (0.76) indicated good biomarker potential. The AUC of IL-6 (0.73), IL-15 (0.73), IL-8 (0.69), IL-13 (0.68) IL-1alpha (0.62), IL-1beta (0.62) showed fair potential as biomarkers.

CONCLUSION:
Cytokine abnormalities are common in CFS. In this study, 10 of 16 cytokines examined showed good to fair promise as biomarkers. However, the cytokine changes observed are likely to more indicative of immune activation and inflammation, rather than specific for CFS. As such, they are targets for herapeutic strategies. Newer techniques allow evaluation of large panels of cytokines in a cost effective fashion.

Note that WPI found IL-8 up-regulated in CFS, while Fletcher et al found IL-8 down-regulated. Also, WPI found IL-5 down-regulated while Fletcher et al found IL-5 up-regulated. WPI found IL-1alpha normal while Fletcher et al found it up-regulated. However, both studies found IL-13 down-regulated and IL-6 up-regulated. So maybe those are stronger bio-markers. Those are two important cytokines, as Rrrr posted earlier:

IL-6: Stimulates chronic inflammation
IL-13: Inhibits inflammatory cytokine production

So if 6 is high and 13 is low, we have a major inflammation problem, so it appears. Anyway, I think we need further research before conclusions can be drawn.

 

[leaves]

I consider myself a fairly typical CCC Me patient (although i do have it since childhood/ no sudden onset) I have insanely elevated il-4 and il-1 b and il-12, as conform FLetcher, I know many (canadian criterian) pwme that have this. My il-8 is low-normal and my il-5 is (as conform WPI) very low, my il-1-a is also low, as conform to wpi. So I seem to be in between the two profiles. But my il-13 is normal, and my il-6 low. SIgh.. this is a mess.. Seriously I dont think this is going to be helpful other than for deciding on the individual treatment for the person with the cytokine profile. Too heterogeneous. Inflammation is an aspect, for sure, but it can present itself in many different ways, AND varies over time.
Moreover, all the immune modulators, high dose vitamins etc that we are taking will also likely affect the cytokine profile.
oh for completeness; tnf r11 and il 23 are also high

BTW I take high dose vitamin k2 and that can greatly lower il-6

 

[richvank]

Hi, all.

In my opinion, what is missing in the cytokine studies is a hypothesis to explain the results. Here's mine: The immune system is able to detect that there are pathogens present, and it produces cytokines in an attempt to mobilize a proper response to the threats. However, because of a combination of deficits that result from the vicious circle mechanism that is at the core of the pathophysiology of this disorder (described by the GD-MCB hypothesis), the immune system is not able to respond properly to certain types of threats (viruses, intracellular bacteria, and fungi, including yeasts) and consequently these are the types that are commonly found to be elevated in ME/CFS. Therefore, the cytokine levels remain high, continuing to sound the alarm and attempting to mobilize the defense, but there is only a partial response, and the infections continue to smolder on.

The deficits in the immune response include the following:

The NK cells and cytotoxic T cells cannot produce enough perforin, because of glutathione depletion. Glutathione is necessary to control the oxidation state of cysteine during protein synthesis, and perforin molecules are particularly vulnerable, because they have 20 cysteine residues.

The helper-T cells do not respond normally to signals, again because of low glutathione, which has been found to be essential to their function.

The T cells cannot undergo clonal expansion (proliferation) in response to a threat, because the folate draining from cells in the bone marrow as part of the vicious circle mechanism limits the rate of production of DNA and RNA to make new cells.

The RNase-L, which is intended to be a temporary stopgap mechanism until the T cell cavalry arrives, is forced to continue to be elevated. Because glutathione is depleted, calpain activates in the cells and cleaves the RNase-L molecules to form the unregulated low-molecular-weight version, and this causes the RNase-L mechanism to be poorly controlled.

The depletion of glutathione also causes poor control of oxidative stress that results from the production of reactive oxygen species by cells of the immune system as part of the inflammatory response to pathogens. This leads to damage of normal tissues.

Best regards,

Rich

 

[currer]

I understood the Lombardi study to be specific to ME/CFS and XMRV infection.

Would that not exclude the Klimas study from comparison, if it cannot be known whether her patients were XMRV+ve?

 

[leaves]

well I am xmrv+ (by culture) and I dont have the same profile as Lombardi et al, as I described in the post above. I know others who are positive and dont have it either. it is possible that that is because of the therapies that we take (I take immune modulators) but .. not so sure. ALso I found that my cytokines greatly varied over time.

 

[floydguy]

This study in part seems to replicate an earlier study that they do not reference. Curious, until you consider that WPI has results that conflict with the earlier CFS cytokine study by Fletcher et al. However, good science says you survey the entire literature, and propose explanation for divergent findings. I wonder what WPI would say about this study? And I wonder what Dr Klimas would/will say, given it was her study that was omitted from reference.



Note that WPI found IL-8 up-regulated in CFS, while Fletcher et al found IL-8 down-regulated. Also, WPI found IL-5 down-regulated while Fletcher et al found IL-5 up-regulated. WPI found IL-1alpha normal while Fletcher et al found it up-regulated. However, both studies found IL-13 down-regulated and IL-6 up-regulated. So maybe those are stronger bio-markers. Those are two important cytokines, as Rrrr posted earlier:

IL-6: Stimulates chronic inflammation
IL-13: Inhibits inflammatory cytokine production

So if 6 is high and 13 is low, we have a major inflammation problem, so it appears. Anyway, I think we need further research before conclusions can be drawn.

Yes, it is a mess. According to Klimas' labs I am very high IL-4, IL1B, TNFR11, high IL-6, IL-13, IL-5 and very low IL-23 and low IL-15, IL-2 and IL-8.

For the most part I guess I follow Klimas' profile except that I am rather high in IL-13.

 

[leaves]

hey floydguy; yes you and me both seem to have the classical th1/th2 shift

 

[kurt]

I understood the Lombardi study to be specific to ME/CFS and XMRV infection.
Would that not exclude the Klimas study from comparison, if it cannot be known whether her patients were XMRV+ve?

Lombardi does reference an earlier Klimas immune study as an example of cytokine/immune abnormalities in CFS. So not referencing the more recent study that shows some conflicting results is an odd omission. Anyway it doesn't really matter much what WPI chooses to reference, we certainly can read all the research in order to place their finding into perspective. I just found it very odd they would fail to mention such an important CFS cytokine study when that is their topic, XMRV or not, this is a CFS study also.

 

[Forebearance]

Thank you for that explanation, Rich.
It's interesting to compare your hypothesis of what's going on in the immune system to Dr. Shoemaker's. You both have part of the immune system sounding the alarm, and the calvary unable to come to the rescue, but for different reasons.

 

[5150]

Awesome, thanks for sharing. Looks like the WPI was finally able to publish something again?

This seems to establish CFS (or XMRV related CFS as they call it) as an inflammatory disease. Which makes sense as anti-inflammatory medicine has helped me a lot.

The bad news seems to be that we have a greater risk to get lymphoma...

But now we have a test?

hi jemal,
may i ask you which anti inflammatory medicine(s) has helped you?

i have a doctor's appt soon and would like to ask for something instead of Advil.

thank you.

 

[richvank]

Thank you for that explanation, Rich.
It's interesting to compare your hypothesis of what's going on in the immune system to Dr. Shoemaker's. You both have part of the immune system sounding the alarm, and the calvary unable to come to the rescue, but for different reasons.

Hi, Forebearance.

I think that's right. In the case of the biotoxin illnesses, which Dr. Shoemaker is dealing with, the threat is a biotoxin rather than a virus, intracellular bacteria, or a fungus itself (even though molds are fungi). In the case of the biotoxin illnesses, for genetic reasons, the people's cells are not able to present the biotoxins to the T cells so that they are recognized, so the T cells never get alerted. In this case, the innate immune system is left holding the bag, including the complement system. This isn't so effective, so the biotoxins build up and cause problems. So there are similarities and differences, but it is an interesting comparison.

Best regards,

Rich