Antibodies to TPO, receptor, platelets cause ITP

Only 40-60% of cases of ITP test positive for autoantibodies against platelet glycoproteins. What accounts for the remaining cases?

This study, published online December 1, 2016, tested for the presence of antibodies to thrombopoietin (the natural hormone that stimulates platelet production), antibodies to the receptors for thrombopoietin, as well as platelet glycoproteins GPIIb/IIIa and GPIb/IX. Out of 43 active ITP patients, everyone had antibodies to at least one of the three antigen categories tested, and many had multiple antibodies. Antibodies to platelet glycoproteins were the LEAST prevalent.

www.bloodjournal.org/content/128/22/2548
Autoantibodies to Thrombopoietin and the Thrombopoietin Receptor in Patients with Immune Thrombocytopenia
Ishac Nazy, Jane C. Moore, Rumi Clare, James W. Smith, Nikola Ivetic, Vanessa D'Souza, John G. Kelton and Donald M. Arnold
Blood 2016 128:2548;
Abstract
Background:
... Several studies have exploited the possibilities of other immune-mediated mechanisms to account for low platelet counts in the absence of detectable antibodies. Thus, we hypothesized that other autoantibodies that target antigens involved in the platelet lifecycle can be important in ITP, such as thrombopoietin (TPO) and the thrombopoietin receptor (cMpl). The objective of this study was to evaluate the frequency and clinical significance of autoantibodies against TPO and cMpl in patients with ITP compared to patients with other thrombocytopenic disorders and healthy controls.

Results: Among patients with active ITP, 36/42 (86%) had antibodies to TPO or c-Mpl: 4/42 (10%) had anti-TPO autoantibodies only and 5/42 (12%) had anti-cMpl autoantibodies only and 15/42 (36%) had both... Platelet bound antibodies to GPIIb/IIIa, GPIb/IX or both were detected in 18/42 (43%) active ITP samples; ... ITP patients with anti-TPO or anti-cMpl antibodies required fewer ITP therapies before remission was achieved compared with patients who had anti-GP autoantibodies.

Conclusions: Testing the entire panel of autoantibodies that included anti-TPO, anti-cMpl and anti-GP, we were able to identify all patients with active ITP...

Great find, Rob. Completely explains the lack of antibodies in some people....they weren't tested for all of them! Simple....

This article has made me rethink my understanding of ITP, and my head is still spinning from it:

Conventional wisdom is that ITP is mostly a platelet destruction problem. That is mostly wrong.

I had always thought production/destruction was a binary either/or choice. It turns out that in most cases the answer is "both".

I had thought that thrombopoietin receptor agonists (TPO-RAs) like Promacta work by causing the body to crank out enough platelets to overwhelm the rate of platelet destruction. Now it seems logical that in many cases (at least at lower doses) the TPO-RAs may actually be replacing the natural thrombopoietin that the immune system is destroying.

When TPO-RAs do not work, the reason may be that the receptors are being destroyed too quickly, rather than the platelets.

Patients in remission continued to show antibodies to TPO and TPO receptors, while healthy controls did not have these antibodies. Thus it seems that long term remission does not mean that the autoimmunity is gone, but rather that it has somehow become regulated (by regulatory T-cells?)

"ITP patients with anti-TPO or anti-cMpl antibodies required fewer ITP therapies before remission was achieved compared with patients who had anti-GP autoantibodies." I wonder why?

" all patients with non-immune thrombocytopenia had circulating autoantibodies to TPO or c-Mpl" (mostly to both)

Which antibodies are present during acute ITP? Childhood ITP? Are the numbers any different?

Can the presence of different types of antibodies predict treatment outcomes? or help choose the optimum treatment?

Interesting find Rob..will be printing it off and handing to my consultant on Monday .
Anne

I've always believed that most people were affected by both production and destruction. Most of the articles that I've read seemed clear about that. Everyone is different as far as how much they are affected by both, which explains why treatment success differs.

As far as I know, treatment outcomes cannot yet be predicted. There may be lab studies, but I don't know of any specifically tailored treatment methods just yet. It would be nice.

Sandi, You have been ahead of most researchers, to have long believed that ITP is caused by a combination of production and destruction issues. I have seen statements in the literature as recently as last year such as this one:

• The prevailing paradigm for the pathogenesis of ITP is that it is caused by IgG auto-antibodies against platelet receptors.

Occasionally, I have seen production mentioned, but there has always been the implication that platelets are being destroyed during the production process by the same antiplatelet antibodies attacking megakaryocytes.

In my experience, this research from McMaster University is something unlike what I have seen before, involving completely separate antibodies from those previously discussed.

Rob:

I wouldn't say that I'm ahead; I just remember what I read. Finding it again is another story. I usually bookmark articles but have lost many of the ones that I had over the years.

Another interesting theory: I have read articles that stated that people with ITP tend to have platelet fragments. I'm sure you know about this...platelet microparticles. All I can remember is that something causes them to shatter. Challenge - see if you can find out why. This is yet another 'cause' of low platelets but not sure if it's on the production or destruction end.

I guess I have had some theories over the years that ended up being proven eventually. Nothing earth-shattering though.

Oh my. This is a very provocative article. Perhaps even revolutionary in understanding of ITP. This should open up some doors. Apparently, thanks to some new ultra sensitive antibody testing apparatus.

Rob16 wrote: Patients in remission continued to show antibodies to TPO and TPO receptors, while healthy controls did not have these antibodies. Thus it seems that long term remission does not mean that the autoimmunity is gone, but rather that it has somehow become regulated (by regulatory T-cells?)

To me, this is the best part Rob. I want to expand on this idea but first I would like to point out what appears to be an error in one of the numbers. Please confirm. The numbers in this line don't add up right.
"
Results: Among patients with active ITP, 36/42 (86%) had antibodies to TPO or c-Mpl: 4/42 (10%) had anti-TPO autoantibodies only and 5/42 (12%) had anti-cMpl autoantibodies only and 15/42 (36%) had both.
"

I think the correct numbers are:
"
Results: Among patients with active ITP, 24/42 (57%) had antibodies to TPO or c-Mpl: 4/42 (10%) had anti-TPO autoantibodies only and 5/42 (12%) had anti-cMpl autoantibodies only and 15/42 (36%) had both.
"
This is because 4 + 5 + 15 = 24 and not 36.

Now also notice this statistic.
"
Platelet bound antibodies to GPIIb/IIIa, GPIb/IX or both were detected in 18/42 (43%) active ITP samples;
"
So here we see that 18 is correct. Why? Because 18 + the previous 24 = 42, which is the population size. Right?

Correct me if I've made an error, but doesn't this mean that there is no population overlap between the two groups. One group (TPO or c-Mpl) is 24 and the second group (GPIIb/IIIa) is 18. The sum is the total population (42). So for example, there is no population category where one has say a TPO problem and GPIIb/IIIa problem. The two groups are orthogonal, yes?

Ok, now to the point I want to make. Let me propose the following hypothetical. Then tell me if its 'toys in the attic' crazy.

For round numbers lets use 50% of folks have a GPIIb/IIIa problem, 25% have a TPO problem, 25% have a c-Mpl problem. This is roughly the proportions presented in the paper. Note that 50% have a TPO problem, a c-Mpl problem, or both.

Recall that the response to IVIG is about 75% of population. Now let me theorize that IVIG works on two groups - the GPIIb/IIIa group and the TPO group. Also that IVIG does not effectively treat the c-Mpl group. Ok, there you have it. All 'active' ITP is represented by the three groups and this explains why some 25% don't respond to IVIG. It doesn't work for the c-Mpl group.

In math, I think they call this a tautology ?? Each of the treatments (eg steroids, IVIG, etc) can be described in terms of their effects on the three ITP antibody groups presented in the paper.

This also suggests if a drug can be developed which targets the c-Mpl antibodies issue, there will no longer be a situation where some folks appear refractory to all drugs.

Folks, am I barking up a tree that doesn't exist ?

Hal,

Your point about an apparent discrepancy in the numbers which does not change the implications of the study,
so I hope readers who find my explanation of the numbers too esoteric will skip over it.

I, too, saw what looked like a discrepancy in the numbers, but it is possible that rather than making a mistake, the authors expressed themselves poorly. In fairness to them, it is hard to express what they were trying to say without using a lot more words and/or a Venn diagram.

Your "corrected" numbers do work if they are described as follows:

04/42 patients had TPO antibodies only and not Platelet antibodies

05/42 patients had c-Mpl antibodies only and not Platelet antibodies

15/42 patients had c-Mpl and TPO antibodies and not Platelet antibodies

24/42 patients had c-Mpl and/or TPO antibodies and not Platelet antibodies

18/42 patients had Platelet antibodies, with or without antibodies to c-Mpl or TPO

42/42 patients total, all of whom had at least one type of antibody

Assume all of the above, and assume that of the 18 patients with Platelet antibodies, 12 also had antibodies to TPO and/or c-Mpl, and 6 did not:

24/42 patients had c-Mpl and/or TPO antibodies and did not have Platelet antibodies

12/42 patients had c-Mpl and/or TPO antibodies and did have Platelet antibodies

36/42 patients had c-Mpl and/or TPO antibodies (86%)

06/42 patients had Platelet antibodies alone (14%)

42/42 patients total, all of whom had at least one type of antibody

Based on these numbers only 14% of ITP patients have ONLY a destruction problem!
Of course, this is a relatively small sample size, and the confidence intervals for the percentages would be fairly large. Nonetheless, they paint an interesting picture.

Hal, now for your main point:

I think that your tautology assumes away the tremendous overlap between the groups.
Based on my previous post, there is tremendous overlap between the three groups:

04/42 have antibodies only to TPO

05/42 have antibodies only to c-Mpl

06/42 have antibodies only to Platelets

15/42 have antibodies only to just one of the three antigens (36%)

27/42 have antibodies to some combination of two or three antigens (64%)

42/42

I agree with your conjecture that it is the c-Mpl antibody that is most difficult to treat. After all, the TPO receptor agonists (TPO-RAs) solve the problem of a TPO autoimmunity, but require there to be healthy TPO Receptors.

It is useful to make simplifying assumptions, because it is so hard to get one's head around this process.
However, the devil is often in the details. Some details we are leaving out:

• There are at least two antigens involved in platelet destruction, GPIIb/IIIa and GPIb/IX. This means that treatments might be effective against one, the other, or both of the related antibodies.
• There are more than one immune mechanisms at work for platelet destruction, and this may well apply to TPO and TPO receptor antibodies, as well.
• Immune mechanisms are not all-or-nothing. A patient may have varying degrees of autoimmunity for different mechanisms.
• Considering the large possible combinations of overlapping immune mechanisms, it may be necessary for some patients to receive multiple types of treatment.

Sandi wrote: Rob:
Another interesting theory: I have read articles that stated that people with ITP tend to have platelet fragments. I'm sure you know about this...platelet microparticles. All I can remember is that something causes them to shatter. Challenge - see if you can find out why. This is yet another 'cause' of low platelets but not sure if it's on the production or destruction end.

Platelet microparticles are caused by apoptosis, which is a form of programmed cell death. Certain antibodies cause platelet apoptosis by triggering changes in the cell wall membrane, causing platelets to fall apart into microparticles. I like your image of the platelets being caused to "shatter" better.

Here is an article that explains the process with more information than I can take in, but here it is:

www.ncbi.nlm.nih.gov/pmc/articles/PMC4975454/
Published online 2016 Aug 5. doi: 10.1371/journal.pone.0160563 PMCID: PMC4975454
Platelet Apoptosis in Adult Immune Thrombocytopenia: Insights into the Mechanism of Damage Triggered by Auto-Antibodies
Similarly to nucleated cells, platelet life span is controlled by an intrinsic apoptotic program, being major players in this process the anti-apoptotic protein BcL-xL and pro-apoptotic proteins Bak and Bax [11]. Pro- and anti-apoptotic protein unbalance triggers mitochondrial outer membrane permeabilization (MOMP) that is followed by mitochondrial inner membrane potential collapse (ΔΨm), efflux of cytochrome c into the cytoplasm, activation of caspase 3 and 9, phosphatidylserine (PS) externalization and microparticle shedding [12]. Since some of these events also take place during platelet activation, markers of platelet apoptosis should be carefully analyzed.
...
The higher incidence of apoptotic platelets in ITP patients carrying auto-antibodies against the major platelet glycoproteins suggests a causal role for these antibodies in triggering platelet apoptosis. Platelet apoptosis was evident in all five ITP patients with anti-GPIIb-IIIa auto-antibodies and the one with anti-GPIb auto-antibodies. The possible link between anti-platelet antibodies and platelet apoptosis has been previously suggested by Leytin et al [13], who described that injection of anti-GPIIb antibodies in a murine ITP model triggers thrombocytopenia that is associated with platelet apoptosis.
...

Good learning thread here. Another thing to remember is that antibodies can come and go. A person could be tested one year and in six months, have different antibodies.

While we're on the subject of microparticles, here's an interesting article that I found.

BACKGROUND:

Splenectomy is frequently employed for therapeutic and diagnostic purposes in various clinical disorders. However its long-term safety is not well elucidated. Although risk of infection by encapsulated organisms is widely recognized, less well-known are risks of thrombosis and cardiovascular disease.

METHODS:

We investigated levels of cell-derived microparticles (C-MP) in 23 splenectomized ITP (ITP-S) and 53 unsplenectomized ITP patients (ITP-nS). Assay of C-MP derived from platelets (PMP), leukocytes (LMP), red cells (RMP) and endothelial cells (EMP) were performed by flow cytometry. Coagulation parameters included PT, aPTT and activities of FVIII, IX and XI. Results of all measures were compared between the two groups, ITP-S vs ITP-nS.
RESULTS:

Levels of all C-MP were higher in ITP-S than ITP-nS but only RMP and LMP reached statistical significance (p = 0.0035 and p < 0.0001, respectively). The aPTT was significantly shorter in ITP-S (p = 0.029). Interestingly, correlation analysis revealed that RMP, but not other C-MP, were associated with shortening of aPTT (p = 0.024) as well as with increased activities of factors VIII (p = 0.023), IX (p = 0.021) and XI (p = 0.0089).
CONCLUSIONS:

RMP and LMP were significantly elevated in splenectomized compared to non-splenectomized ITP patients. This suggests that the spleen functions to clear procoagulant C-MP, and that elevation of C-MP might contribute to increased risk of thrombosis, progression of atherosclerosis and cardiovascular disease following splenectomy.

www.ncbi.nlm.nih.gov/pubmed/18334267

Ooooh, I went away on vacation and returned to such a wealth of interesting information!

Thanks for sharing Rob, I'm even more happy in my second line treatment choice (TPO-RAs) after reading. Lots of old school thinking about ITP is falling by the wayside. Good stuff.

Based on my previous post, there is tremendous overlap between the three groups:
04/42 have antibodies only to TPO
05/42 have antibodies only to c-Mpl
06/42 have antibodies only to Platelets
15/42 have antibodies only to just one of the three antigens (36%)
27/42 have antibodies to some combination of two or three antigens (64%)
42/42

>>>>>>>>>>>>>>>>>>>>>>>>>>
Helo Rob,
This was a good clarification.

Now my point is TPO related issues and platlet related issues are two different items.

I beleive they are coincidental but not causal to have GPM\I related and TPO related ones.

Root cause is the stress on the liver.

I know I was drinking alcohol tad more than is needed and combine that with couple of stress events my TPO was mis-firing causing platelet production issues.

I still don't believe in any other autoimmune issues but it is just me.

So you think that in your case, you don't have an autoimmune issue; the low platelets were caused by drinking too much?

Yes, type cirrhosis and thrombocytopenia in google and you will see too many people suffer from liver failure and low platlet count.

It is too much of a coincidence that liver is the center of all these issues.

TPO secretes in the liver, issue could no TPO or TPO which is compromised.


TPO receptors cannot process with no TPO or bad TPO.


Now nplate in my case is such a success because they are basically substituting TPO.

Alcohol or abstaining from it has made absolutely no difference to my platelet count or my liver function whatsoever.

Yes, liver problems can cause thrombocytopenia. Are your liver function tests abnormal or is there any evidence of liver disease?

All liver tests normal and they did extra tests to test for cirrhosis.

Still it is a unusual coincidence to get itp.

Hello mrsb,
How long did you abstain?
How many units weekly did you consume?

My issue was i was dieting/exercising and drinking 25 units a weeks.
Lost weight from 210 to 180 but same alcohol content.
Wife went to my inlaws for summer increased my intake to 35
Units, couple of stressful work incidents and voila got itp.

Jay:

Usually if liver disease is a cause of thrombocytopenia, it would show up on tests, or the spleen would be enlarged.

Background: Liver disease is frequently missed as the cause for a patient's thrombocytopenia.

Aim: To evaluate the role of liver disease in causing thrombocytopenia within a tertiary hospital.

Design: A hospital-based retrospective study.

Methods: Analysis of medical records of outpatients and inpatients with a platelet count <100 × 109/l seen at St Mary's Hospital, London in October 2011, was conducted. Cause for thrombocytopenia was determined in each case and patients with liver disease were analysed further looking at factors associated with their low platelet count.

Results: In total, 223 patients were included in this study, 109 of them were outpatients and 114 were inpatients. The mean age was 57.1 years (range 22–106), 64% male and 36% female. Liver disease was the cause for thrombocytopenia in 58% of outpatients. Overall, 92 patients with liver disease were identified; cirrhosis and/or splenomegaly were present in 78%, a further 8% were on interferon therapy. Thrombocytopenia was not explained by the extent of liver disease in 8%, significantly more in hepatitis C than other causes of liver disease (P < 0.05). Factors correlating with low platelet count in patients with liver disease were spleen size (P < 0.05) and serum bilirubin (P < 0.001). There were none, or mild abnormalities only in liver function tests in 19 patients with liver disease-associated thrombocytopenia.

Conclusions: Liver disease should be considered in all patients with an unknown cause of thrombocytopenia. Advanced liver disease does not have to be present for thrombocytopenia to develop. All patients with unexplained thrombocytopenia should be evaluated to see if liver disease is present, even when liver function tests are normal.

qjmed.oxfordjournals.org/content/early/2013/01/23/qjmed.hcs239

Hal9000 wrote: "Samples were tested for circulating antibodies against TPO or cMpl using newly developed enzyme immunoassays (EIAs) and for antibodies against platelet glycoproteins (GPIIb/IIIa and GPIb/IX) using the antigen capture assay."

pdsa.org/forum-sp-534/6-general-itp-discussion/29526-antibody-absorbent-materials-apheresis.html#57645
Hal, I missed the importance of the fact that there are now assays for all of what appear to be just four antibodies responsible for at least the vast majority of all ITP cases. Just by itself, that is huge news! When the new immunoassays get FDA approval I expect that testing will become routine, and treatment tailored to which antibodies are present.

I think some of the most important research to be done as soon as possible is to determine which existing treatments work best against each combination of antibodies.

Edit: You can even buy the assays online!
www.mybiosource.com/prods/ELISA-Kit/Human/Thrombopoietin-Receptor-Antibody/TPOR/datasheet.php?products_id=29620
www.mybiosource.com/prods/ELISA-Kit/Human/Thrombopoietin-antibody/TPOAb/datasheet.php?products_id=720418

Helli sandy,
Did you miss out the last sentence?

I have read the research And many more like this.

The statistical study never tells you why?

So novice patients like us will
Have to dig deeper?

Once a connection has been made them it is no longer itp but what is the cure.

Abstinence will not immediately work they it takes just 3 months to get the liver/bone marrow to correct itself.

I think eventually we will find smoking, alchohol, over weight and causes of cancer all cause thrombocytopenia in adults.
Then at least we will be treating things correctly than just doing a carpet bombing veryone with every possible medication.

Jay
You may be right that treatments will become more disease specific. However I think it is more a case of saying leads to remission rather than cures.

jayinchicago wrote:
Abstinence will not immediately work they it takes just 3 months to get the liver/bone marrow to correct itself.

I think eventually we will find smoking, alchohol, over weight and causes of cancer all cause thrombocytopenia in adults.
Then at least we will be treating things correctly than just doing a carpet bombing veryone with every possible medication.

I hope your theory works for you - abstainance from alcohol, smoking and losing weight are good ideas but didn't work for me and my Haemotologist told me it wouldn't and he was right!

I think you may be over simplifying what can be a spectacularly complicated condition to try and treat.

Rob
The tests are for 'research only'. I will have to look into how they can be used since it will not be covered by insurance. I am not sure a doctor will do the blood test. Hmmm.
Emily

jay, I have to wonder if one of us, you or I, have a basic misunderstanding of ITP. If there is no misunderstanding, please disregard. But, as I see it, once a bad/self destructive antibody is present in your immune system it is remembered perhaps forever.

As an example think of flu viruses. There are many mutations, most different from year to year. If your immune system sees a virus today that is the exact same virus as in years past, it will successfully ward off the virus with little effort. There are already circulating antibodies. Your body continues to produce antibodies for that strain for years. Each time that strain comes along again, there is a refresh which reinforces the memory for that antibody.

If a flu virus comes along that your immune system has never seen, it has to start mutating its antibodies until it matches the new strain. This can take time and causes extended illness.

Now consider a permutation during the period of antibody mutation for a new virus that ends up attacking platelets. Realize that old platelets are being consumed all the time and concurrent to flu infection. Some platelets are mixed in with the flu virus.

Once platelet destruction is learned platelets are attacked continuously - even if these antibodies never helped kill any virus. The platelet antibodies are just mistakes in trying to figure out how to kill a new invader. The memory for platelet antibodies is renewed constantly - no chance in ever forgetting how to kill platelets.

So, curing the trigger to ITP never causes remission. Your immune system must gain the ability to recognize the attacked platelet (or TPO or TPOR) as 'self', as opposed to 'foe', to stop your immune system from attacking.

With that, if ones liver was wacky one day in the past, that would be the trigger - on that day, and learned forever or until your immune system can figure out that platelets are 'self'.

I can only speculate that 'self' is realized after the immune system has been stopped from continuously producing bad antibodies for some period of time.

Yea, apparently its an 'enzyme' immunoassay. Does that mean TPOA and TPOR are enzymes (instead of proteins) themselves, or, does it mean they use enzymes to detect TPOA and TPOR?

Which existing treatments work best against each combination of antibodies? LOL, that's what I was thinking too. It should be the easiest thing to do next. Do tell Rob, what is your background? Engineer or medical research or ? A little management/team leader mixed in?

What's with this 'mybiosource.com'? Part of your work in the past? Looks like they call it TPOA and TPOR tests. A for antibody and R for receptor.

Hal: "Do tell Rob, what is your background?"
Former life, pit bull. This life: Psychology; engineering; management; self-trained medical hack; but most of all, google-fu master.

Google [enzyme immunoassay thrombopoietin antibody] and mybiosource.com pops up first.

A lot of antibody tests use ELISA (Enzyme Linked Immunosorbent Assay) techniques. I think the assay uses enzymes.

Hello Hal thanks for the clarification.

It is only me having a huge misunderstanding of the whole immune system life cycle.
Sorry Iam a computer engineer with no background in biology.

>>>>>>>>>>>>>>>
So, curing the trigger to ITP never causes remission.
>>>>>>>>>>>>>

How sure are about this statement.
Dengue fever causes low platlets but with no dengue patients recover why?

A patient had some wacky heart valve, one fine day he started getting low platlets.
When they fixed the valve, the platlets came back.


What about situations where TPO production is slowed down considerably and thus mackrocytes just die because TPO is not there.

It is still a good discussion to have, because of my lack of background in biology, I will have to question every assertion which does not make sense and try to rationalize it in my mind.