Mar 05 2012
SNPwatch: The “Jist” on JAK2 and Myeloproliferative Neoplasms
Myeloproliferative neoplasms (MPN) have a complicated name but it’s an apt descriptor for the biology underlying this group of rare blood disorders. MPNs result from excessive production of “myeloid” blood cells that originate in the bone marrow, and come in different forms including essential thrombocythemia (ET), polycythemia vera (PV), myelofibrosis (MF), chronic myeloid leukemia (CML), and systemic mastocytosis. Most forms of MPN are cancerous and all types are rare, with each affecting fewer than 100,000 Americans.
In 1951 a physician named William Dameshek proposed the idea of myeloproliferative disorders, but it wasn’t until 2005 — when four articles were published back-to-back on the V617F mutation in the JAK2 gene — that researchers began to uncover clues about the genetic basis of these diseases. Discovered in blood cells of people with MPN, the V617F mutation is defined as a somatic or acquired mutation, which means that it arises during the course of person’s life as opposed to being a change in DNA that someone is born with. (See sidebar for more about the JAK2 V617F mutation and its clinical use.)
JAK2 in the Clinic
The JAK2 protein is a tyrosine kinase and plays important roles in the cell by directing the activity and movement of other proteins. Although more research is needed to clearly define the role of the JAK2 V617F mutation in myeloproliferative disease, there is strong evidence that cells bearing this mutation are less prone to dying and are very good at making multiple copies of themselves, a hallmark of cancer. The World Health Organization (WHO) has recommended that the JAK2 V617F mutation be used as a diagnostic marker of disease and numerous companies are developing drugs to inhibit the activity of JAK2, though it is too early to tell if they will be successful.
This figure shows that people who inherit predisposing JAK2 variants have higher odds of acquiring the JAK2 V617F mutation, which is thought to push the cell to make multiple copies of itself. But it’s not the only factor leading to MPN. Although most people with the polycythemia vera (PV) form of MPN have the V617F mutation, many people with other types of MPN do not.
In addition to the V617F mutation, a number of studies, largely performed in people with European ancestry, have shown that inherited genetic factors may also increase risk for MPN. A cluster of SNPs in the JAK2 gene, called the 46/1 or GGCC JAK2 haplotype, has been associated with MPN and seems to be a strong predisposing factor for acquiring the V617F somatic mutation. People with a G at rs12340895 (equivalent to rs12343867 in the study) had nearly four times higher odds of developing V617F-positive MPN compared to people without the G version. 23andMe recently replicated this association, though we see a smaller effect — in our database, people with a G at rs12340895 have about two times the odds of developing V617F-positive MPN compared to people without the G version.
A more recent study carried out with a Japanese cohort suggests that the association between relatively common variants in the JAK2 gene and MPN also applies to individuals with Asian ancestry. Junko Ohyashiki and colleagues from Tokyo Medical University report that people with an A at rs3780374 in the JAK2 gene (equivalent to rs4495487 reported in the study and highly correlated with rs12343867) have about four times higher odds of developing V617F-positive MPN compared to individuals without the A version.
(23andMe customers can view their data for rs12340895 (applicable to people with European ancestry) and rs3780374 (applicable to people with Asian ancestry) in their MPN Preliminary Research report.)
When they looked more closely, they found even stronger evidence for the association and a larger effect in patients with the polycythemia vera (PV) form of MPN, which is consistent with the fact that over 95% of people with PV have the V617F mutation while only about half of people with essential thrombocythemia (ET) or myelofibrosis (MF) have V617F. The rs3780374 SNP did not appear to be linked to V617F-negative MPN.
Although these findings are still preliminary since they were collected from relatively small studies (Ohyashiki’s findings came from just 95 people with MPN), they are still exciting progress in research on these rare disorders. That relatively common genetic variation might predispose individuals to acquire other mutations during their lifetime is an intriguing concept that will require more research to understand.
SNPwatch gives you the latest news about research linking various traits and conditions to individual genetic variations. These studies are exciting because they offer a glimpse into how genetics may affect our bodies and health; but in most cases, more work is needed before this research can provide information of value to individuals. For that reason it is important to remember that like all information we provide, the studies we describe in SNPwatch are for research and educational purposes only. SNPwatch is not intended to be a substitute for professional medical advice; you should always seek the advice of your physician or other appropriate healthcare professional with any questions you may have regarding diagnosis, cure, treatment or prevention of any disease or other medical condition.
I have PV and participated in the 23andme study, and sure enough – I just looked and I do have “substantially increased odds of developing an MPN.” It’s really great to see progress in the research and hopefully finding a cure!
I have CML, and my genome shows:
JAK2, LOC100129248 5089677 rs3780374 A or G GG
JAK2 5066691 rs12340895 C or G CC
Will there be an analysis of family members of Jak2 positive mpn people? My father submitted his DNA sample and I wonder what his results look like. Which side of my family did this snip ome from? I have thought of having my children submit samples. If they do will they be able to learn if they too have this snip?
Hi Mary,
The V617F mutation associated with many MPNs is actually an “acquired” mutation, meaning that it’s a spontaneous change that happens in a person’s cells rather than a genetic variant that you’re born with. Someone with this mutation cannot pass it down to their children since their sperm or egg cells do not have the mutation.
The other SNPs mentioned in the studies, however, are associated with higher odds of acquiring the V617F mutation and these are inherited. You can compare different people’s results for these SNPs like you can with any other SNPs on our chip. If someone has, say, “AG” at a SNP and her mom is “AA” and her dad “GG”, then you know that she inherited the “A” from mom and the “G” from dad. Similarly, if her mom is “AG” and her dad “GG”, you again know that she inherited the “A” from mom and the “G” from dad. In some cases, though, it’s not definitive who passed down what — for this example, this would be the case if both parents are “AG”.
Hope this helps!
My mother has polycythemia vera rubra. She’s 83 and has lived with it for many years. It’s been managed primarily through a drug called hydroxyurea. Thank you 23andme for identifying the fact that I may also develop this disorder.
I have just been diagnosed with early primary MF, at age 61, after living with CLL with 16 years and doing well. I am happy to participate in any genetic studies if that will help.
My results show positive for both GG and AA. Is this possible? It doesn’t sound like it from the knowledgable sounding post from Shwu. I do have the JAK2 variant and both ET and SM.
I have both snps markers for myeloproliferative neoplasm that are linked to higher odds of devolping this illness. What should I do as prevention?
Hi Andrew,
We are sorry to hear of your diagnoses. If you haven’t already, please go to http://www.23andme.com/mpn and enroll in our MPN research initiative. If you have any questions, contact us at mpn-help@23andme.com. Thank you for your interest in contributing to research!
Hi Carolyn,
My previous reply was with regards to being able to tell which version of SNP someone might have inherited from which parent, apologies if it was unclear. It’s actually expected that if you’re GG at one of the SNPs that you’ll be AA at the other — these two SNPs are highly correlated but one association has been found in studies of people with European ancestry while the other association has been found in individuals with Asian ancestry. You should consider your results at the SNP applicable to your ancestry only.
@Franco,
As I wrote above to Carolyn, both of the SNPs we currently report on (rs12340895 and rs3780374) probably represent the same biological signal, however one of the SNPs has been shown to be associated with MPN in several studies in people with European ancestry while the other SNP is specific to people with Asian ancestry, so you should just look at whichever association is applicable to your ancestry. Keep in mind that MPN is still extremely rare, so having higher odds of developing JAK2-positive MPN is still a very low probability overall and there are likely other genetic factors that influence who develops this condition as well. Scientists are still learning more about this set of diseases and the different factors involved!
I have 2 kids on here with cutaneous mastocytosis (never had the bone marrow biopsy to find out if it is systemic). One child showed high risk of Jak2 mutation, and the other showed normal risk. My question is: is jak2 indicative of systemic mastocytosis? In masto, they typically talk about the c-kit mutation, not jak2. Can we also check for likelihood of c-kit mutation? If so, can someone tell me which gene to look at in the raw data.
Thanks,
Bonnie
Hello Bonnie,
We are very sorry to hear about your children’s diagnoses and wish all of you the best. Although systemic mastocytosis is classified as a myeloproliferative neoplasm (MPN), individuals with this form of MPN do not usually have the JAK2 V617F “acquired” mutation in their blood cells. (A study published in 2009 showed that only 4% of people with systemic mastocytosis had the JAK2 V617F mutation: http://bloodjournal.hematologylibrary.org/content/113/23/5727.long).
The results discussed in this blog post and in our MPN preliminary research report are for predisposing, inherited variations in DNA that seem to make people more susceptible to acquiring MPN and the JAK2 V617F mutation later in life. As you mentioned, systemic mastocytosis is often associated with KIT D816V, which just like JAK2 V617F, is a mutation that people “acquire” later in life as opposed to being one that they are born with. Since 23andMe only tests for inherited variations in DNA and not for “acquired” mutations, we do not test for KIT D816V. We are not aware of any inherited variants that predispose for acquiring the KIT D816V mutation. Hopefully scientists will learn more about the genetics underlying systemic mastocytosis in the near future! If you would like to learn more about our research efforts on MPN, please go to http://www.23andme.com/mpn. If you have any questions, contact us at mpn-help@23andme.com.
The report Prevelance of Polycythemia Vera and Esstentional Thrombocythemia lists the occurances of the disease per 100,000 people as 22 and 24.
What I would like to see is the total number of people listed on this site as rs12340895 GG or CG or AA and rs3780374 the same and broken down per GG,etc.. That way if there were only say a total of 100 rs12340895 and 22 get PV you could better see the rarity. Or say if there were 50,000 and only 22 get PV then even rarer.
These are some of the things that computer data mining can solve very quickly. the power of a data base can be immense or nothing, depending on how it is used.
The problem has been presented, I am graciously awaiting the solution. thanks.
Hello Tom,
You raise a very good point. Since all forms of MPN are very rare, it helps to know what percentage of people have the “riskier” version of the SNP. We know from data collected through the International HapMap Project (http://hapmap.ncbi.nlm.nih.gov/index.html.en) that nearly 50% of people with European ancestry have at least one G at rs12340895 and about 40% of people with Asian ancestry have at least one A at rs3780374. Beyond this, more research is needed to understand why only some people with the inherited variant go on to acquire the somatic mutation or develop MPN, but it’s a question we’re very interested in and have the potential to study given our unique database.
Thanks for the response BHromatka. So when shall we get to studying this problem as the sooner we do the sooner an answer can be obtained. And possibly how to prevent the somatic mutation. thanks again.
My son has been diagnosed previously with autism, PANDAS and underwent treatment for ALL — the course was complicated with frequent persistant infenctions, one of which was bartonell grahamii.
His results show the trifecta GG AA GG when you include rs10974944
This would make his MPN risk aprox 4096 normal.
Each parent is heterozygous = 64 fold risk.
My marrow results are pending.
Any advice on where to turn?
When you caculate 4 fold risk for each allele shouldn’t this pop up at the top of the browser?
Hello Randall,
We are very sorry to hear about your son’s diagnoses and hope that he is doing much better. We also wish you the best with your bone marrow tests.
Being GG at rs12340895 increases one’s odds of developing JAK2 V617F-positive MPN, but being GG at rs10974944 doesn’t further increase one’s odds. This is because these SNPs don’t individually confer risk and thus the effects are not additive. If a parent is heterozygous at each SNP then he/she still only has ~ 4 fold increased odds of developing MPN. Just for some more background, the SNPs rs12340895, rs10974944, and rs3780374 are almost always inherited together — for instance if you are GG at rs12340895 then you will most likely also be GG at rs10974944 and AA at rs3780374, too. As mentioned earlier, you should only consider the results that are applicable to your ethnicity.
There are a number of patient support groups for MPN including the MPN Research Foundation (http://www.mpnresearchfoundation.org/) and the Myeloproliferative Disorders Research Consortium (http://www.mpd-rc.org/home.php). You may also direct any questions to mpn-help@23andme.com.
I have both rs12340895 GG, rs3780374 AA, and rs10974944 GG. This is described as greatly/substantiallly increased risk for MNP. Wondering what the prevalence of these SNP variants in the normal population is, can only find for a single G or single A, respectively, which seems to be quite common (40-50%?).
Also wondering what the total/combined increase in risk of MNP is in my case with three of the “worst” variants. Also, how common is this triple variant combined in the population? Should I worry?
@ MPN SNP
As noted in previous comments to this post, you should consider your results at the SNP applicable to your ancestry only. Nearly 50% of people with European ancestry have at least one G at rs12340895 and about 40% of people with Asian ancestry have at least one A at rs3780374. However, only about 5% of people will be GG at rs12340895 (for Europeans) or AA at rs3780374 (for Asians). Although being GG at rs12340895 increases one’s odds of developing JAK2 V617F-positive MPN, being GG at rs10974944 doesn’t further increase one’s odds. In this particular case these SNPs don’t individually confer risk and thus the effects are not additive. See responses to earlier comments for more explanation on these points. I hope this helps!
I have the AG Asian variant. I think 23andme is really on to something here. My daughter was diagnosed with acute myelogenous leukemia, subtype acute promyelocytic leukemia when she was two. I was diagnosed with erythromelalgia just a few years ago. I just read that Polycythemia Vera can cause erythromelalgia. To date, aside from positive Lyme tests which I was told were false positive by some doctors, no doctor has been able to give me a definitive diagnosis for the constellation of symptoms I have. I went ahead with Lyme treatment, despite the disapproval from some doctors. Surprisingly, my erythro improved greatly. I still have the erythro, but it is greatly diminished. It does make me wonder if the Lyme bacteria can be one of the causes of the mutation. There are studies that show that the Lyme bacteria does enter the organs of unborn babies so that could explain a mutation with my daughter having the leukemia. I’m just brainstorming. Perhaps the Lyme is a false positive and we have a genetic predisposition. Either way, I find it too coincidental that I have erythromelalgia and my daughter had APL, both of which are rare and both which can fall under this genetic variant.
Forgot to mention that I have the CG variant at rs12340895. Both of the variants I carry increase my odds. So while these variants may be found in 50% of persons of European ancestry and 40% of Asians, it is not common to have those variants and already have one AML in my family and erythromelalgia. I think those diagnoses together with the variants definitely paint a little different picture in my case as far as risk factors go.
I also find it interesting that Randall’s son was found to have an infection, Bartonella, which can also be transmitted by ticks. Researchers are finding that ticks are now passing more than just one bacteria at a time. They are passing Lyme, Bartonella, Babesiosis, Rocky Mountain Spotted Fever, etc.
Perhaps these bacteria are the triggers that turn these oncogenes on. I would love to see a study with mice with these genetic variants, who are then given these types of bacteria. It would be interesting to see if they would serve as a trigger and the mice end up with MPN. I wish I had all the money in the world so I could fund these studies.
Hi Kristen,
We are sorry to hear about your diagnoses and hope that you and your daughter are doing better. You raise a good point about combining family health history with genetic results. If you haven’t checked it out already, please visit 23andMe’s Family Health History feature: https://www.23andme.com/family/tree/.
Hi All,
I’m new to 23andme, and still making sense of all the info. When flagged for MPN, is there a way to tell from the data available to us, if a mutation has already occurred in V617-F? Or does the data only reflect a predisposition to developing mutation?
Thanks for your help!
Anne
Hi Anne,
Welcome to 23andMe! Our MPN report tells you if you have higher odds of developing MPN and acquiring the JAK2 V617F somatic mutation. It does not, however, tell you if your blood cells have already acquired the V617F mutation. The JAK2 V617F mutation is defined as a somatic or acquired mutation, which means that it arises during the course of a person’s life as opposed to being a change in DNA that someone is born with. I hope this helps!
Hi BHromatka,
Yes, that is exactly what I wanted to know. Thank you for your speedy response!
# BHromatka
Thank you for your answer, and high quality of answers in general! You say that rs12340895 GG, rs3780374 AA, and rs10974944 GG combined does not pose an additive increase in risk for the JAK2 V617F somatic mutation. Do you have the publication where this information is taken from?
Hello MPN SNP!
In general, SNPs that are located close to one another (for instance in the same gene) often represent the same biological signal unless proven otherwise. In this case, it just so happened that one particular SNP in the JAK2 gene was the most statistically significant for Europeans and a different SNP in the JAK2 gene was the most statistically significant in Asians. This is more likely a reflection of differences in genetic linkage in the JAK2 region between the two populations rather than a reflection of two separate risk signals. Below are some of the primary papers for the association between SNPs in the JAK2 gene and developing MPN.
http://www.nature.com/ng/journal/v41/n4/full/ng.341.html
http://www.nature.com/ng/journal/v41/n4/full/ng.334.html
http://www.biomedcentral.com/1471-2350/13/6
I hope this helps and just to reiterate, you should only consider the results at the SNP applicable to your ancestry only!
I recently went to a conference on Bartonella — interesting pathogen with a propensity to hide in CD34 (marrow progenitor) cells — the most common form of ALL is CD34+. Intra erythrocytic location can make this a difficult pathogen to treat. There is fantastic work being done at NCU by a Dr. Ed Breidtschwerdt. Galaxy labs offers PCR testing. My son was also diagnosed with autism — which seemed to improve with antibiotics — perhaps it was neuroretinitis? Ticks can transmit bartonella — though typically it can be aquired simply through an abrasion — not an uncommon thing for a child. The most recent publication links bartonella and rheumatoid arthritis — a similar situation to post deployment ailments (infectious arthritis) seen after WW1. Also it is a noteworthy pathogen due to it’s interaction with H1N1 (Spanish Flu).
I suspect that the driving force behind some autoimmune ailments might just be chronic gram negative bacteremia. Broad PCR testing can help (ex spirostat), especially in the situation of chronic immunosuppression. We live in Burnett County WI — recently Mayo Clinic was through looking at Ehrlichia Muris in cancer patients.
I’m less inclined to blame Lyme — as we learn more about pathogens that clearly have a track record of persistance. If Lyme is there — you get around to treating that too, too much controversy there — and no retractions of any of the old literature.
Charles Kallick M.D. has a patent for treating aplastic anemia and leukemia with strictly antibiotics — also article in Medical Hypothesis that would support my ramblings here.
I don’t suppose you’ll ever see an affordable approach to cancer — but treating concomitant bacteremia is a no brainer.