Expires July 2024
Myeloproliferative neoplasm (MPN) is a malignancy of hematopoietic stem cells that is associated with hematopoietic proliferation resulting in peripheral blood cytosis. MPN is subclassfied on the basis of the dominant cell line involved and each subtype has a different prognosis. The diagnosis of MPN must include an integrated approach and combine the clinical findings with laboratory results. Dr. He discusses the subclassification of MPNs and the use of JAK2, CALR, and MPL mutational analysis in diagnosis and prognosis.
Presenter and Credentials:
Rong He, M.D., Consultant in the Division of Hematopathology and Assistant Professor of Laboratory Medicine and Pathology in the College of Medicine at Mayo Clinic in Rochester, Minn.
PresenterWelcome to Mayo Medical Laboratories Hot Topics. These presentations provide short discussion of current topics and may be helpful to you in your practice. Today our topic is Myeloproliferative Neoplasm: Morphology, Molecular Updates and Cost-Effective Test Utilization
Our speaker for this program is Dr. Rong He, a Consultant in the Division of Hematopathology at Mayo Clinic in Rochester, Minnesota. Dr. He is also Assistant Professor of Laboratory Medicine and Pathology in the College of Medicine.
Dr. He, thank you for presenting today.
Thank you Cara, I have no disclosures.
As you view this presentation, consider the following important points regarding Myeloproliferative Neoplasm Testing.
The focus today will be the bcr-abl-negative classical MPNs, namely, polycythemia vera, PV; essential thrombocythemia, ET; and primary myelofibrosis, PMF.
Myeloproliferative neoplasm is a neoplasm of the hematopoietic stem cells. It is associated with effective hematopoietic proliferation, and the bone marrow cells successfully make it into the peripheral blood, resulting in peripheral blood cytosis. MPN patients usually present with a hypercellular bone marrow and peripheral blood cytosis. Organomegaly is present or absent. Myelofibrosis is common, whereas dysplasia is usually not evident. MPN is further subclassfied on the basis of the dominant cell line involved.
Different subtypes of MPN have different prognosis. ET and PV are faring better than PMF
But both PV and ET have the potential of evolving into myelofibrosis, more so with PV, and all 3 of them can develop into acute myeloid leukemia, most commonly seen in PMF. The choices of medical intervention are different with different subtypes and stages of MPN.
Therefore for diagnosis of MPN, we need to take an integrated approach and combine the clinical findings with laboratory results, such as CBC and differential, serum EPO, LDH, iron study results and peripheral blood and bone marrow morphology, together with genetic findings, such as chromosome analysis, molecular testing of JAK2, CALR and MPL mutational analysis, and other appropriate prognostic mutational analysis.
The goals for morphologic evaluation in MPN are:
When sub classifying MPN by morphology, we need to take a practical approach and acknowledge that we cannot accurately subclassify every single MPN with absolute confidence. So our first step is to rule out or rule in PV. By morphology, PV can be a big masquerader, however the lab test results might be of help here as PV is the only disease in the MPN category that gives a primary erythrocytosis with a secondarily decreased serum EPO level. Additionally, higher than 95% of PV cases are positive for JAK2V617F mutation and the rest are essentially all positive for JAK2 exon 12 mutations. Once PV is ruled out, the next step is to establish the diagnosis of cases that are straight forward ET or straight forward PMF. Everything else is MPN, NOS.
Here is a flow chart showing the same thing. First step, rule in or rule out PV. Once PV is ruled out, next diagnose clear cut ET and PMF on the 2 ends of the spectrum of MPN, anything in between is MPN, NOS.
With years of research, distinctive genetic abnormalities have been identified in the myeloproliferative neoplasms, and the molecular genetic landscape has evolved significantly during the past decade.
Here is the major molecular landscape of the 3 bcr-abl-negative classical MPNs, PV, ET and PMF. Three major driver mutations have been identified in these disorders and importantly, they are essentially mutually exclusive of each other. These are JAK2, CALR, and MPL mutations. The donut charts here illustrate their respective prevalences in the 3 MPN subtypes. In PV, higher than 95% of cases carry the JAK2V617F mutation, demonstrated here in red; the rest are essentially all positive for JAK2 exon 12 mutations. In ET and PMF, about 50 to 60% are positive for JAK2V617F mutation, again in red. After that the second most common mutation in ET and PMF is CALR mutation, shown here in blue. It was first described by 2 independent European groups at the end of year 2013 at the ASH meeting. CALR mutation accounts for about 20 to 30% of cases in ET and PMF. The third well-defined driver mutation in ET and PMF is MPL mutation, shown in green, accounts for about 3 to 5% of ET cases and 5 to 10% of PMF cases.
Note that in both ET and PMF, there are still about 5 to 10% of cases negative for all 3 markers, and those are labeled as “triple-negative” cases. And the triple-negative molecular signature is associated with unfavorable prognosis in PMF. JAK2 and MPL mutations activate the JAK-STAT pathway and were adequate to induce a MPN phenotype in animal models. CALR mutation has also shown to converge on the JAK-STAT pathway. Recently, a French group reported that the CALR mutations were sufficient to induce a MPN phenotype mimicking ET with later progression into myelofibrosis in a retroviral mouse model.
Listed here are the 2008 WHO diagnostic criteria for MPNs. Note that in each entity, molecular marker is being included. In the next WHO update, CALR mutation will be added in for ET and PMF, according to the recent presentation by Dr. Orazzi at the 2015 USCAP meeting.
Going into the specifics of each gene mutation testing, for the most common JAK2V617F mutation, a positive result supports the presence of a myeloid neoplasm, and strongly favors MPN over MDS (however, it can be seen in less than 5% of MDS and AML case, and pretty frequently, at approximately 60%, in RARS-T). A positive result does not help to distinguish between different subtypes of MPN as shown in the previous slides. A negative result, in general, is diagnostically not helpful, but it does argue against a diagnosis of PV or post-PV myelofibrosis given the high prevalence of this mutation in PV. The assay sensitivity is determined by the methodology being used. If it is Sanger sequencing, the sensitivity is about 20%, whereas the more sensitive allele specific PCR generally gives a sensitivity of at least 0.1%. The sensitivity of the Mayo quantitative allele specific PCR has a sensitivity of 0.01%. However, a word of caution here is that we should be very careful with the borderline positive values, as with any quantitative assays. Low level JAK2V617F mutation has been reported in normal individuals and could reflect the nature of the assay at the threshold level or a bona fide early clonal process.
It is not uncommon to see JAK2V617F testing ordered on both peripheral blood and bone marrow samples. But is this necessary? As we know that MPN is a disease with effective hematopoiesis, i.e., the cells in the bone marrow can successfully make it to the peripheral blood. So we did a study and looked at JAK2V617F test results on Mayo clinic patients, from year 2006 to 2009. We identified 267 patients with concurrent peripheral blood and bone marrow studies performed and unsurprisingly, they showed a very high concordant rate of 98.5%. There were 2 patients with positive results in bone marrow and negative results in peripheral blood, and 2 other patients positive in peripheral blood and negative in bone marrow. When we delved into the quantitative levels of the positive results, all of them were borderline values close to the lower limit of detection. Bone marrow diagnosis of these 4 cases included 1 overlapping MDS/MPN, unclassifiable, and 3 bone marrow without morphologic features of a MPN. In none of the 4 cases, the discrepant results had any impact on diagnosis, management and patient outcome. So the final conclusion was to cancel the duplicate tests and only perform it on 1 type of specimen.
Next, CALR mutations: CALR is a multifunctional protein, and the described pathogenic mutations are somatic insertions and deletions involving CALR exon 9, the last exon of CALR. These are always frameshift indels resulting in a specific alternative reading frame in the C-terminus of the mutant CALR protein. Whereas the inframe mutations are of doubtful clinical significance and most of them are germline alterations. CALR mutations are mutually exclusive of JAK2 and MPL mutations. They occur again in approximately 20 to 30% of PMF and ET cases, and about 50 to 80% of JAK2- and MPL-negative PMF and ET case. CALR insertion/deletion mutations can be detected using different methodologies, if a Sanger sequencing is used, the sensitivity is about 20%, whereas fragment analysis gives a sensitivity of approximately 5%.
MPN patients with CALR mutation exhibit distinctive clinical features. The favorable overall survival had been reported by multiple groups in PMF patients, and they usually present with higher platelet count and are less likely to have anemia or leukocytosis. In ET patients, CALR impact on overall survival is less clear due to the better overall survival in ET. But CALR -positive ET patients have a lower risk of thrombosis, higher platelet count, and a lower hemoglobin and white blood cell count.
The third driver mutation, MPL mutation: MPL gene encodes for thrombopoietin receptor, and MPL mutations are present in 5 to 10% PMF and 3 to 5% ET cases. They are commonly detected by Sanger sequencing with a 20% sensitivity.
Detection of the 3 driver mutations is not only helpful for diagnosis, but also carries prognostic value, especially in PMF: Where CALR mutation shows a favorable impact on survival, over JAK2-positive and triple-negative cases, independent of the current risk stratification system. And triple negative mutational status has been proven to be a high-risk molecular signature.
Besides the 3 major driver mutations, there are also other mutations emerging as prognostic markers in PMF. Such as ASXL1, SRSF2, IDH1/2, and EZH2 mutations. They were identified as “prognostically detrimental” and associated with shortened survival in PMF. However, after multivariate analysis, only ASXL1 mutation remained significant, independent of the current risk stratification scoring system. The high risk molecular signature identified are: Higher than at least 2 mutations in the 5 genes, CALR-negative, ASXL1-positive; or Triple-negative status. And the median survival in this group is pretty dismal at about 2 to 3 years. Again, CALR mutations favorably impact survival, independent of the number of mutations, ASXL1 mutation, and the current IPSS/DIPSS scoring system. Additionally, NRAS have also been reported to be associated with the highest DIPSS score in MPN. There is also evidence that combination of TP53 somatic mutation with loss of heterozygosity is associated with leukemic transformation.
Lastly, JAK2 exon 12 mutations, these are seen solely in PV, not in PMF or ET, and account for approximately 2 to 5% of PV cases. As its incidence is markedly lower than that of the JAK2V617F mutation, which accounts for at least 95% of cases, JAK2 exon 12 mutation testing should be reserved for suspected PV cases negative for JAK2V617F, to be cost-effective. Regarding the methodology used, JAK2 exon 12 mutations is commonly detected by Sanger sequencing with a sensitivity of 20%. It should be noted that JAK2 exon 12 mutation assay is not an appropriate JAK2 V617F screening assay, first of all, V617 is located on exon 14 instead of 12; and even in assays that do cover exon 14, the Sanger sequencing technology predicts a 20% sensitivity in contrast to a much more sensitive allele specific-PCR technology, whose sensitivity is usually at least 0.1%.
Going back to the main molecular landscape of the MPNs, we know that the molecular signature provides diagnostic and prognostic values, especially in PMF. Importantly, the 3 driver mutations are mutually exclusive of each other with different prevalences. Therefore we should use these information as a roadmap to guide our test utilization during workup of the classical BCR-ABL-negative MPN cases. We should test for the most common mutation first, if that is negative, then reflex to the next common one.
This is our updated diagnostic algorithm of MPN workup in a bone marrow evaluation, incorporating the JAK2, CALR, and MPL molecular testing in a value-driven, cost-effective approach.
And here is the link to the algorithm on the Mayo Medical Laboratory website.
From the algorithm, we start the initial testing of MPN with
The bottom line for the ensuing algorithm is basically divided into 2 arms: PV or none-PV. If PV is suspected, the first molecular test to order is JAK2V617F given its extremely high prevalence in PV.
If it is positive, then no further molecular testing is indicated. If it is negative, then JAK2 exon 12 testing should be considered. If it is positive, you get the supporting evidence for the PV diagnosis. If it is again negative, PV is essentially ruled out. The definitive diagnosis of the case depend on the morphology and clinical and laboratory findings. On the other hand, if ET or PMF is suspected, the first test to consider would be JAK2V617F given its highest prevalence in this group. If it is positive, no further molecular testing is indicated on the molecular aspect. If it is negative, then the next test to order is the 2nd most common mutation in this group, the CALR mutation. If CALR mutation is identified, then you are done. If CALR is negative, then the third mutation that should be evaluated is the MPL mutation.
We currently offer the MPN reflex panel through Mayo Medical Laboratories which follows a reflexive algorithm and sequentially tests for JAK2V617F, CALR and MPL mutations until a mutation is identified. In a cost-effective approach, this convenient panel provides value-based, clinically-relevant information in MPN, and takes out the burden of complex molecular test ordering of the hands of the busy clinicians and pathologists. It is recommended for cases suspicious for ET, PMF, and myeloid neoplasm whose differential diagnosis includes ET or PMF.
Here is our next-generation sequencing panel of 35 genes mainly targeting myeloid neoplasms. It will be available soon, so keep tuned.
I‘d like to thank Dr. Hanson, Dr. Kurtin, and Dr. Viswanatha for their lead and contribution to the Mayo Clinic test utilization drive.
And, thank you for listening.