Zhenya Senyak

Based on the 23andMe MPN Patient database, new work in molecular biology has helped lift the curtain on the complex processes involved in the creation of an MPN. This emerging view of the interplay of chance and heredity is shaping our understanding of MPN realities. Whether or not it can lead us out of the dead-end wasteland of JAK2 inhibition trials into gene therapy pathways remains to be seen.



What seemed like a blessing just six years ago — the FDA approval of Jakafi for myelofibrosis treatment — has shaped up to be a disaster for MPN patients.

Discovery of the JAK2 v617f mutation a dozen years ago looked like the beginning of the end for MPNs. By 2011, the first drug to reduce the effect of this mutation in myelofibrosis was approved by the FDA. And the race to a cure was on. Except it was more of a stumble than a race.

Jakafi, the first of the JAK inhibitors, made life more bearable for MF patients, for a time. JAKAFI BOTTLES A key feature is reduction of splenomegaly, permitting return of appetite, improved nutrition and mobility and higher quality of life. But Jakafi doesn’t work for all patients, often becomes ineffective for patients for whom it did work, and drives down platelet levels, and increases anemia and fatigue.

Jakafi also affects only the protein produced by the mutated JAK2 stem cell, the enzyme that triggers the blood production process through the JAK-STAT pathway. The response to the drug is short-lived, requiring continual dosage and Jakafi has no effect on the stem cell or the progress of myelofibrosis.

Even worse for the general MPN population, the billion dollar blockbuster success of Incyte’s JAK inhibitor goaded giant pharmas into a pointless race to produce a look-alike JAK inhibitor. Thousand of MF patients joined clinical trials. Hundreds of millions of dollars have been spent on drugs like SAR302503, CEP-701, Pacritinib, AZD-1480, Momelotinib. The human experiments to test a better JAK inhibitor have either turned deadly and aborted, been suspended due to toxicity or dragged on ineffectively for years.

These trials and delays proved fatal to some high risk MF patients who got steadily sicker and less able to survive the progressing disease or a potentially curative stem cell transplant.

Funds and brainpower that could have been invested in basic science to uncover the actual causes and pathways of myeloproliferative disease were invested in a gamble to capture part of the Jak2 inhibitor market. And all that waste of life, time, energy and money has been essentially unsuccessful…. and pointless. The JAK2 mutation is only a contributing cause to some MPNs and far from the whole story.

As a result of all the wasted investment of brains, time and capital, the unpredictably effective meds MPN patients had available six years ago, are the same handful of drugs available today.

The JAK2 V617F is an acquired somatic mutation. Due to chance error in cellular reproduction or environmental insult the Janus Kinase 2 on Chromosome 9 in the multipotent hematopoietic stem cell suffers a mutation in which phenylalanine (F) is substituted for valine (V) at position 617 of the gene. This leads to increased production in one or more blood cell lines.

However, nearly half of all MPN patients don’t even have the JAK2 V617F mutation

More significantly, people can carry the mutation and not develop MPNs. And others could not have the mutation — or any of the known driving mutations like CALR and MPL –and still suffer from an MPN. Clearly there is something else driving this disease.

The likely co-conspirator is an underlying inherited component, a germline predisposition that could drive the MPN phenotypes with and without that mutation. Supporting that theory, the closest relatives of MPN patients report exhibit a three to four-fold higher incidence of MPN disease. And since the JAK2 mutation is a somatic event it could no more be inherited than could circumcision, tattoos or plastic surgery.

Factors that give rise to the onset and progression of MPNs are likely to be inherited variants in several genes that contribute to increased susceptibility to these blood diseases.

Which genes?

Slowly the secrets of MPNs, deeply buried in the bone-shielded vault of marrow, are being revealed

In 2010, A mostly European group of researchers, including Nick Cross, Tony Green, Claire Harrison, Peter Campbell, Alessandro Vannucchi as well as Richard Silver published their findings in Blood. nick-cross-haplotype-46-1They had discovered a specific group of nucleotides in the JAK2 gene – a small stretch of DNA called a haplotype or haploid genotype since it came from one parent only – that led to a predisposition for a clonal expansion of blood to blossom into a full bore MPN. They called this haplotype 46/1 or GGCC.

Thus the mystery started to unravel. There were stretches of DNA up and downstream of the JAK2 gene that could enable or accelerate the process of moving simple clonal expansion of blood lines, in some cases leading to an MPN. The Hinds, Gotlib et al. study featured below, for example, notes roughly two in 1000 hematologically normal people over 60 years old carry the JAK2 V617F mutation while only 6 to 8 per 10,000 suffer JAK2 V617F MPNs.

Age Related Clonal Hematopoiesis (ARCH), or expansion of one or more blood lines due to mutation of the JAK2 or other gene within the hematopoietic stem cell, is fairly common. Progression of ARCH into one of the myeloproliferative neoplasms is rare. MPN is a rare disease indeed. Why does it arise at all?

The likely answer to these knotty questions is one of the first fruits of the Great American Spit Initiative.

The paper, suggests answers to a few key questions.

Questions like: Why me?

Why are some MPNs so aggressive and others so indolent?

And how come we can have an MPN without a JAK2 mutation? Or any obvious mutation for that matter?

The key finding: The impact of random events – accidents, chance, transcription errors, etc. – resulting in the acquisition of the JAK2 V617F mutation could be increased, softened or even squelched by the genetic environment in which it occurred.

Hinds, Gotlib and co-workers discovered inherited variants associated with well studied genes – small differences in spelling – that turbocharged clonal blood expansion and helped direct it into MPN pathways. TET TERT, SH2B3 ( LNK), TET2, ATM and others.

As a result, their findings went well beyond heritability into the etiology of MPN itself.

“Previous studies identified genetic polymorphism that are associated with an increased risk of JAK2 V617F,” wrote George S, Vassiliou, Wellcome Trust.Sanger, in a published Comment in Blood ”…(but this study) significantly expands the number and identities of genes operating in diverse cellular processes.” Examples cited include JAK/STAT signaling (SH2B3), DNA cytosine methylation (TET2), transcriptional regulation and cell cycle control.

Vassiliou argues that since ARCH and MPN are together associated with increased risk from those same variant genes, “it would support the premise that chance acquisition of JAK2 V617F is not necessarily the dominant factor in the development of MPN” and that inheritance of a “permissive genome” is equally important.

In the past two years, these new studies using Genome Wide Association Studies have blown open the closely held secrets of MPN acquisition and progression. Along the way some other major questions troubling both MPN patients and hematologists appear to be answered.

A good part of the answer was provided in 2015 by Tapper et al publishing in Nature Communications. (Green, Cross, Harrison, Vannucchi and Campbell participated in this research which, in part, was a follow on to the 2010 study.)

“MPNs arise as a consequence of clonal proliferation driven by at least one somatically acquired driver mutation. Of these, the most recurrent is JAK2V617F, which occurs in >95% of PV and 50–60% of ET and PMF cases. In patients without JAK2 mutations, somatic mutations in CALR are the most common, occurring in well over 50% of JAK2V617F-negative ET and PMF.

“Although it is accepted that acquisition of these mutations drives clonal proliferation in MPNs, less is known about what factors influence the development, phenotype and severity of disease. Evidence from epidemiological and familial studies strongly suggest that common, low penetrance factors present in the general population contribute to the risk of developing MPN…. We previously characterized a major predisposition locus at the JAK2 gene, which is associated with the acquisition of V617F mutations. This specific JAK2 haplotype, called ‘46/1′ or ‘GGCC’ strongly predisposes to JAK2V617F-positive disease.”

In this very large three stage study — 3,437 MPN cases and 10,000+ controls — investigators identified two SNPs* with genome-wide significance in JAK2 negative MPNs. One SNP is located within the JAK2 gene and the other within MECOM. Two additional SNPS, were found one of which is in within the TERT gene. (see the Hinds et. al study below). The MECOM gene codes for a protein involved in the proliferation and differentiation of hematopoietic stem cells and fibroblasts.

The 23andMe MPN patient database

MPN patients had direct input into the Hinds, Gotlib et al. study presented at ASH last month. More than 1400 of us, recruited via the Internet, contributed DNA to the 23andMe MPN initiative of 2011. That program, conceived by Ashley Gould, daughter of MPNforum columnist and MF patient Harvey Gould, aimed at collecting DNA samples and basic information on the nature of each patient’s MPN along with some demographic data. The idea was to stimulate research into the genetic causes of MPNs as a means of targeting the mutation(s) driving the disease. Despite the massive patient response, it has not been a wildly successful initiative.

In 2012, 23andMe published the results of the first study using this MPN data base, discovery of “an association between MPNs and a variant in the TERT gene. The current large-scale study, published in Blood last year, was spearheaded by Gotlib, joined by Hinds and others. While highly significant in exploring and confirming the inherited and somatic roots of MPNs, the hoped for genetic associations that would lead to improved treatment have not yet materialized.

Originally Jason Gotlib (Stanford), Ruben Mesa (Mayo Clinic), Ross Levine ( Sloan Kettering) and Jim Zehnder (Stanford) joined 23andMe as unpaid scientific advisors in a “collaborative effort to guide the research discovery process and future project directions.” (Ross Levine reported he has not accessed the data and has no plans to do so.)

The 23andMe initiative was launched in 2011. The objective was to recruit an on-line cohort of 1000 MPN who would submit DNA samples via saliva. This mass of data would be used to study genetic associations that lead to MPNs. According to 23andMe, the MPN Patient Community is now closed. The only peer-reviewed paper in which independent collaborators and 23andMe actually used the data appeared last year in Blood

Using the personal genome MPN database of 23andMe, Gotlib et al. extended the landscape of genetic predispositions of those who are negative for the JAK2 mutation, reveal incipient, low level JAK2 V617F hematopoiesis in the general population, and clearly demonstrate the involvement of both inherited genetic variants and somatic mutations is triggering myeloproliferative neoplasms.

Dr. Jason Gotlib, Stanford

Dr. Jason Gotlib, Stanford

What they did

Hinds, Gotlib et al. tested all contributing MPN DNA samples and selected 726 MPN patients. The control general population – over 250,000 from the 23andMe aggregated data base –were genetically surveyed with a custom DNA probe and 497 turned up positive for the JAK2 V617F mutation. gotlib-gwas-cohortThis combined population of MPN patients and carriers of the JAK2 mutation formed the full cohort of 1223.

Within this cohort, investigators confirmed the presence of the previously discovered JAK2 46/1 haplotype and uncovered significant associations between MPNs and variations associated with TERT, TET, LNK,and other genes.

The associations were nearly identical with both MPN patients and those in the general population who simply carried the mutation but were not diagnosed with an MPN. Conclusion: The same inherited genetic variants predisposed both to myeloproliferative neoplasms and JAK2 V617F clonal hematopoiesis. “a more common phenomenon that may foreshadow the development of an overt neoplasm.”

Some other significant findings emerged from this research.

Likelihood of acquiring an MPN. Carrying the inherited genetic variations associated with the TERT, TET2, LNK and CHEK2, PNT and GF 1B increase the likelihood of developing into an MPN.

“One can hypothesize, ” says Gotlib, “not yet proven by this study, that individuals who carry an increasing number of these inherited genetic variations may have some additive risk of acquiring an MPN. Also, it is not yet understood how these variations functionally affect the genes with which they are associated, e.g. how changes in the DNA code affect the expression or biologic function of these genes, and in turn, how this promotes a higher risk of MPN development.”

Since these genetic variations uncovered in the study are inherited, those who are first degree offspring of MPN patients have a much higher risk of developing an MPN. The Hinds,Gotlib paper cited three studies claiming first degree relatives of MPN patients have a five to seven times greater risk of developing an MPN than the general population.

Participating mutations and variations are all over the molecular map. The V617F mutation and Exon 12 in the JAK2 gene on Chromosome 9 as well as distant variant genes play a role in expansion of blood volumes and acquisition of MPNs. Among the players are the TERT gene on Chromosome 5 (this one codes for the telomerase protein), TET2 on Chromosome 4, and SH2B3, also known as a lymphocyte adapter protein (LNK) found on Chrome 12. The interplay of biologic forces necessary to stimulate clonal hematopoiesis and then convert that flood into an MPN is a complex blend of acquired mutations and inherited propensities.

Hinds, Gotlib et al. also found the inherited variations in these genes “predispose to both age-related JAK2 V617F clonal hematopoiesis (ARCH) in the general population as well as to MPN, independent of V617F status.” That helps explain why MPNs are usually a disease of older adults. And the finding suggests why we can suffer from an MPN while being triple negative for the driving mutations in JAK2, CALR and MPL.

Much work lies ahead.

There are problems with SNP-based* Genome-wide Association Studies (GWAS) used in these recent investigations. And questions about the integrity of data derived from saliva instead of blood in the 23andMe cohort. Genotyping arrays designed for GWAS rely on linked anomalies to cover the entire genome by genotyping a subset of variants. As a result, the reported associated variants are not likely to be the actual causal variants. “Associated regions can contain hundreds of variants spanning large regions and encompassing many different genes, making the biological interpretation of GWAS loci more difficult.”

Whatever their limitations, these Genome Wide Association Studies have laid to rest the simple idea of MPNs caused by a JAK2 v617F switch that turns blood lines on and off and set the stage to study the interplay of acquired and inherited factors that actually drive myeloproliferative neoplasms.



*SNP: A Single nucleotide polymorphism or SNP (pronounced snip) is a DNA sequence variation occurring when a single nucleotide – A, T, C, or G – in the genome (or other shared sequence) differs between members of a species (or between paired chromosomes in an individual).

+ Nature or Nature,  from the 2010 MPN Research  Foundation archives.