Scientists have created a new test for identifying people at risk of developing acute myeloid leukemia and related cancers, years before they do. The new platform, “MN-predict,” will allow doctors and scientists to identify those at risk and to design new treatments to prevent them from developing these potentially lethal cancers.
The study is published in Nature Genetics.
Researchers at the Wellcome-MRC Cambridge Stem Cell Institute (CSCI), the University of Cambridge’s Department of Haematology, and Instituto de Investigación Sanitaria del Principado de Asturias (ISPA) analyzed data from more than 400,000 individuals participating in the United Kingdom Biobank.
Using this data, the scientists have created “MN-predict,” a platform for predicting the risk of developing blood cancers such as acute myeloid leukemia, myelodysplastic syndromes and myeloproliferative neoplasms over a 10–15-year period.
This test, now available in NHS clinics, requires patients to provide a blood sample from which DNA is extracted for limited sequencing, alongside basic blood cell counts. With this information, MN-predict identifies those at high risk of any of these cancers and can be used in specialist clinics for leukemia prevention.
Professor George Vassiliou, senior author of the study said, “We all know that prevention is better than cure, but it is not easy to prevent diseases like leukemia without knowing who is at risk. MN-predict makes it possible to identify at-risk individuals, and we hope it can become an essential part of future leukemia prevention programs.”
The myeloid neoplasms are a group of related cancers encompassing acute myeloid leukemia, myelodysplastic syndromes and myeloproliferative neoplasms. Treatments for these cancers have improved in the last few years, but most cases remain incurable.
In the last few years, scientists discovered that these cancers develop over decades through the accumulation of DNA mutations in blood stem cells, the cells responsible for normal blood formation. These mutations encourage these stem cells to grow faster than normal and, as more mutations accumulate, they can progress towards leukemia.
Thankfully, while mutations that promote cell growth are common, leukemia develops only in a small minority of cases. Identifying these cases early on helps efforts to prevent the cancers from developing.
Dr. Muxin Gu, first author of the paper, said, “We hope that MN-predict will help clinicians to identify people at risk of myeloid cancers and use novel treatment to prevent the cancers from developing.”
Dr. Pedro M. Quiros, joint senior author of the study, said, “Despite some recent advances in their treatment, these cancers remain lethal to many sufferers. We hope that our efforts will help advance prevention in favor of treating the full-blown disease.”
Although patients with myeloproliferative neoplasms (MPN) often experience many symptoms either related to the disease or from treatment itself, nurses can help patients navigate symptom management and help seek relief, one expert said.
Oncology Nursing News® spoke with Tetyana Furmanets, CRNP, MSN, an oncology nurse practitioner at Penn Medicine Abramson Cancer Center in Philadelphia, to learn more about how nurses can advise patients with MPN on symptom relief and tools available for nurses to gauge treatment responses.
Oncology Nursing News: What are some of the symptoms associated with MPN and what are some ways nurses can help patients manage them?
Furmanets: MPN comes with a lot of symptoms, the most prominent one being probably fatigue. A lot of patients report debilitating, generalized fatigue. That is probably one of the hardest ones to manage as well because there’s no specific targeted agent for that. I recommend [that] our patients continue to exercise as much as possible while listening to their body, going on daily walks while taking time to rest at home. Certain medications that patients are taking for MPN might help with the symptoms of fatigue.
Some of the other symptoms that we see with myeloproliferative neoplasms are itching. That’s one of the big ones. Specifically, patients report severe itching after they take a shower. Our recommendation is either lowering the temperature of the water before taking a shower or using topicals. There is one lotion—which is over the counter—that we use a lot, Sarna cream, which is very helpful for our patients. We recommend applying that after taking a shower while their skin is still wet.
There are some side effects of the myeloproliferative neoplasms that are very tricky to deal with. Some of them may be fevers, which you can take Tylenol, but there comes a point of the disease process where Tylenol is just not helping with it. So promote fluids, hydration. Sometimes that can be very helpful with symptoms of fevers as well as bone pain, which we see a lot with this patient population as well.
Some of the more vague symptoms that we see is difficulty with concentration, which is a little hard to get out of the patients to talk more about, but when you ask them about it, they’re like, ‘I definitely started noticing I’m having more issues with that.’ This one is a little harder to treat. But I feel like going for those walks and trying to like breaks, take rest and listen to your body and don’t push it too hard, have been definitely helpful.
The other big one we see with myeloproliferative neoplasms is getting full after a few bites of food. A lot of patients are not able to finish full meals because of their spleen size. They have some discomfort associated with their spleen. That comes hand in hand along with fatigue and is probably one of the biggest symptoms we see in this patient population. Again, some of the treatments help with reducing the spleen size. When patients do experience that, they’re so grateful and they feel amazing. They’re like, ‘I could finally finish a full plate and I’m able to sleep on that side.’ So that’s very encouraging.
Unfortunately, sometimes patients don’t respond that well to treatment, so they’ll experience some of that left-sided abdominal pain. We work with nutritionists a lot for those patients; we encourage them to [try] some small, frequent meals that are high-protein, high-calorie content, so that even though they’re not getting a lot of food in at one time, they are still getting their adequate nutrition and their caloric amount during the day.
We work a lot with our palliative care team to help with the pain management aspects when we get to severe cases of myelofibrosis. Pain medication might help with that, as well [as] avoiding sleeping on that side, avoiding certain types of activity or exercise to avoid more trauma to the spleen.
Are there tools that nurses can use to educate their patients about the side effects?
I utilize NCCN guidelines a lot during the treatment phase. We use an MPN treatment symptom assessment during our visits. It’s a questionnaire; patients score [their symptoms] on a scale from zero to 10, zero being no symptoms at all and 10 being the worst imaginable. It lists all of the most common symptoms, fatigue, pain, itching, abdominal pain. It is very helpful as far as determining where the patients are on the scale of the severity.
It might be beneficial if the nurses utilize it and give it to the provider, something to compare it to because a lot of times when you ask the patients, how are you feeling and they’re telling you they feel fine. And when you give them the questionnaire it’s like everything’s like nine or 10 out of 10, so you have to dig a little deeper with those questions.
It is a very tricky disease to manage because everybody’s so different as far as that goes. But we have been utilizing that symptom assessment form a lot and have been helpful to determine if we’re moving in the right direction or making any progress, if we are addressing those symptoms at all.
What’s the most important thing for nurses to keep in mind when caring for patients with MPN who are experiencing symptoms?
Unfortunately, a lot of treatments don’t work overnight. It takes weeks to a month to fully kick in. It can be very frustrating for our patients. We have a lot of patients who are coming in and reporting that they just started this medication, they’re still not feeling too great, and they get a little discouraged. Reinforce that it might take some time for the medication to kick in.
Cincinnati Children’s experts show, in mice, that targeting DUSP1 eradicates JAK2 mutated MPN
Myeloproliferative neoplasms (MPNs) are malignant bone marrow diseases that cause dangerous overproduction of red blood cells, white blood cells, and/or platelets. These conditions mostly strike adults around age 60 but can occur at any age. Some of these patients ultimately develop acute myeloid leukemia (AML).
Based on successes achieved in treating chronic myeloid leukemia (CML) with a class of drugs called ABL tyrosine kinase inhibitors (TKI), cancer researchers had high hopes that a similar class of drugs called JAK2 inhibitors would be a breakthrough for treating MPNs. However, clinical studies have found that JAK2 inhibitors are ineffective.
Now, a study recently published in the journal Leukemia reports achieving curative response in mice when they selectively knock-out a negative regulator of MAPK signaling: DUSP1. This highly complex study took a team of scientists at three institutions seven years to complete. The work was led by senior author Mohammad Azam, PhD, Divisions of Cancer Pathology and Experimental Hematology and Cancer Biology.
“This study, for the first time, provides mechanistic understanding why JAK2 inhibitors are ineffective in vivo and how JAK2V617F signaling suppresses P53 function required for MPN transformation and progression,” Azam says. “Selective targeting of DUSP1 opens up a completely novel therapeutic approach and a potentially curative treatment outcome in MPNs.”
OVERCOMING DEAD ENDS
Inspired by the clinical efficacy of TKI therapy for treating CML, a race began to identify similar molecular drivers in MPN that could be targeted for intervention.
Scientists initially found mutations of interest within three genes JAK2, MPL, and CALR. Further study of mouse genetic models revealed that all the mutations produced a common outcome: elevated and persistent JAK-STAT and MAPK signaling. This provided a strong rationale for developing small molecule inhibitors to target JAK2 kinase activity.
Numerous JAK inhibitors have been assessed in MPN and myelofibrosis (MF), another rare, chronic blood cancer. So far, three JAK2 inhibitors are approved by the FDA to treat MPN and MF while almost a dozen JAK inhibitors are currently undergoing pre-clinical and clinical assessment for potentially treating conditions such as arthritis, psoriasis, inflammation, graft-versus-host disease (GVHD), and autoimmune disorders.
However–unlike the success of TKI therapy in CML–JAK2 inhibitors do not induce remission. Instead, they simply slow cell division. Similarly, inhibitors targeting the MAPK pathway by blocking MEK1/2 or ERK1/2 either alone or in combination with JAK2 inhibitors failed to induce remission.
“Even the most potent kinase inhibitors failed to kill MPN cells in vivo,” Azam says.
It became clear that other mechanisms must be involved in preventing the effectiveness of JAK2 inhibitors. In prior studies, Azam’s lab had explored another mouse model of MPN that involved a different cancer cell growth factor called BCR-ABL kinase. That work revealed that growth-factor signaling in the context of oncogenic signaling induces the expression of c-FOS and DUSP1 that causes resistance to TKI treatment.
Inflammatory cytokine signaling is one of the cardinal features of MPN, with about 60 different cytokines induced in the context of these conditions. Azam reasoned that inflammatory cytokine signaling drives TKI persistence in JAK2 targeted MPNs.
ZEROING IN ON DUSP1
In addition to Azam, the research team on this project included first author Meenu Kesarwani, PhD, Division of Pathology; H. Leighton Grimes, PhD, director of the Cancer Pathology Program; and six other members of the pathology division at Cincinnati Children’s. Experts from the Medical College of Wisconsin and the Memorial Sloan-Kettering Cancer Center also contributed.
The team worked for seven years to conduct numerous experiments to tease apart the reasons for the persistent cellular resistance to JAK2 inhibitors in MPN. The co-authors conducted an extensive set of genetic analyses that revealed deregulation of 19 genes in TKI resistant cells. Ultimately, the team focused on DUSP1 because this gene appears to dampen the MAPK signaling that suppresses the P53 apoptotic pathway.
NOVEL APPROACH FOR MPN THERAPY
Importantly, their work revealed that mice lacking DUSP1 exhibit normal growth and reproduction, thus supporting the notion that a treatment targeting this gene’s function would have minimal side effects. When mice without the DUSP1 gene were further tested, the team gained crucial insight into the cell signaling mechanisms that help MPNs resist JAK2 inhibitors.
“In essence, inflammatory cytokine and JAK2V617F signaling converge to induce the expression of DUSP1, which prevents the function of P53.” (See figure)
P53 is often referred to as the “guardian of the genome” due to its role in regulating diverse external or internal stresses, such as DNA damage, activation of oncogenes, nutrient deprivation, and hypoxia. Importantly, it plays a critical role in deciding the cell fate, cell death or division arrest for DNA repair. Consequently, it plays a significant role in treatment outcomes to chemotherapy as most resistant patients harbor P53 inactivating mutations.
NEXT STEPS
While the genes and signaling pathways involved in the mouse research also appear to exist in humans, much more research is needed to determine whether a selective eradication of DUSP1 can be achieved to cure JAK2-induced MPN, Azam says.
Meanwhile, co-authors say the new discoveries about how to control growth factor signaling in MPN cells may also lead to improved treatments for other forms of cancer.
Blood Cancer UK and Cancer Research UK recognise the need and impact of artificial intelligence approaches developed by Professor Daniel Royston and Professor Jens Rittscher for early detection and assessment of these blood cancers.
Cancer Research UK and Blood Cancer UK have awarded funding to a multidisciplinary team from the University of Oxford. These awards will help to advance the AI-based methods and predict the progression of myeloproliferative neoplasms (MPNs) more accurately.
MPNs are a group of closely related disorders of the bone marrow affecting around 5000 people every year in the UK. Patients with MPNs are at higher risk of developing leukaemia, especially those with a subtype called myelofibrosis (the most severe) where this develops in >10% of patients.
Because the treatment strategy varies depending on the MPN subtype, accurate assessment of MPN type at diagnosis is crucial for optimal treatment selection. In addition to mutational and blood count analysis, morphological analysis of a bone marrow biopsy is a key component for classification. Unfortunately, this is highly subjective, reliant on qualitative observations and there is great variability even when it is done by expert haematopathologists.
There is unmet clinical need for a more accurate method for diagnosing MPN from a bone marrow biopsy. The team, led by Professor Daniel Royston (Radcliffe Department of Medicine and Oxford University Hospitals NHS Foundation Trust) and Professor Jens Rittscher (Institute of Biomedical Engineering and Big Data Institute), have already developed artificial intelligence approaches to help pathologists extract quantitative data from scanned images of bone marrow biopsies. These algorithms will enable more accurate and reliable classification of MPN type.
With the new funding, the team now wish to refine and validate these methods with the aim of integrating them into existing NHS pathology workflows to bring about earlier diagnosis of MPNs in the clinic. Importantly, this work will include input from patient representatives from the Oxford Blood Group. They will give feedback on the visualisation tools designed to help patients better understand what’s happening in their bone marrow and the progress of their disease.
Better diagnostics and management of MPN disease are key priorities for our patients. Receiving this funding from Cancer Research UK and Blood Cancer UK will allow us to make significant progress towards our aim of applying our AI-based tool for more accurately diagnosing MPN type in the clinic so that patients can benefit. – Professor Daniel Royston (Radcliffe Department of Medicine and Oxford University Hospitals NHS Foundation Trust), research lead.
Individualized myelofibrosis treatment begins with correctly identifying a patient’s disease subtype and considering their symptoms, from which accurate decisions regarding the use of JAK inhibitors vs radiation vs hypomethylating agents (HMAs) can lead to spleen and symptom burden reductions, according to Raajit K. Rampal, MD, PhD.
During an OncLive® State of the Science Summit™ on hematologic malignancies, Rampal and colleagues highlighted the role of JAK inhibitors in myelofibrosis; considerations for CAR T-cell therapy in follicular lymphoma (FL); efficacy and safety findings with asciminib (Scemblix) in chronic myeloid leukemia (CML); the future of BTK inhibitors in mantle cell lymphoma (MCL) and chronic lymphocytic leukemia (CLL); unmet needs in diffuse large B-cell lymphoma (DLBCL); and research on the horizon in lower-risk myelodysplastic syndrome (MDS).
Rampal, who chaired the event, is the director of the Myeloproliferative Neoplasms Program and an associate attending physician at Memorial Sloan Kettering Cancer Center in New York, New York.
Rampal was joined by his colleagues:
Alexander P. Boardman, MD, assistant attending physician, Memorial Sloan Kettering Cancer Center
Michael J. Mauro, MD, leader, Myeloproliferative Neoplasm Program, Leukemia Service, Memorial Sloan Kettering Cancer Center
Prioty Islam, MD, MSc, assistant attending physician, Memorial Sloan Kettering Cancer Center
Jennifer K. Lue, MD, clinical director, Lymphoma Service, Memorial Sloan Kettering Cancer Center
Jan Philipp Bewersdorf, MD, hematology/oncology fellow, Memorial Sloan Kettering Cancer Center
Below, Rampal, Boardman, Mauro, Islam, Lue, and Bewersdorf summarize the main messages from their presentations.
Current and Emerging Treatments in Myelofibrosis
Rampal: [Myelofibrosis] treatment depends on the issue. This is 1 of the major principles in treating [patients with] myelofibrosis. It’s not a monolithic entity. This disease has different manifestations, and we need to treat the manifestation that is causing the patient the major issue. [When] some patients [present with myelofibrosis], anemia is the major [symptom] they’re [experiencing], not spleen [issues]––nothing else, just anemia. For those patients, a JAK inhibitor may not necessarily be the right choice, but [treatments such as] erythropoiesis-stimulating agents [ESAs], danazol, corticosteroids, or even immunomodulatory agents may be an appropriate first-line choice.
However, for patients with symptomatic splenomegaly or constitutional symptoms, JAK inhibitors have made their mark. Compared with [treatments such as] hydroxyurea, JAK inhibitors have superior efficacy, reducing both spleen size and symptom burden. We have 3 FDA-approved [JAK inhibitors] currently.
For other manifestations of this disease, there are other [treatments] we can use. For patients with extramedullary hematopoiesis in the lungs or bones, radiation is appropriate. For patients who are early in their disease, sometimes pegylated interferon can be useful; some data support that. When patients progress to accelerated or blast-phase disease, HMAs are the backbone of therapy, usually in combination with other agents.
CAR T-Cell Therapy in FL
Boardman: CAR T-cell therapy is clearly active in relapsed/refractory FL, with high response rates, and is effective in patients with high-risk disease. Current data suggest that these remissions are durable but given [that FL is] an indolent lymphoma, we need more time to [confirm these data]. Adverse effects [AEs] [including] cytokine release syndrome and immune effector cell–associated neurotoxicity syndrome are common and must be considered when weighing treatment options for patients, especially those who may be frailer. Lastly, the optimal timing of CAR T-cell therapy vs bispecific antibodies and other targeted agents will require further study.
Current and Novel TKIs in CML
Mauro: We might think of ponatinib [Iclusig] being favored [over asciminib] in patients with primary resistance [and] high transcript levels. For patients with compound mutations and pan cytopenias, neither drug may be successful. These can be challenges. Patients with a mixture of intolerance and resistance who have perhaps more preserved response or at least more residual response from prior therapy exposure may have a better AE experience and better long-term outcomes with asciminib, at least speculatively.
The T315I [mutation] is up for grabs. [Either ponatinib or asciminib may effectively target this mutation] because [both drugs seem to have activity there], and we don’t have concerns about the differences in dosing [between] asciminib [and ponatinib].
Looking into the future, other drugs are under further study. Olverembatinib, which is approved In China, is the drug that is closest to ponatinib and is in trials in the United States [US] now. [Regarding] other agents, we have some trials at Memorial Sloan Kettering Cancer Center. The [phase 1] ELVN-001 trial [NCT05304377] is open, [investigating an] ATP-competitive inhibitor that’s probably [similar to] ponatinib in its activity but may be much safer. Some other second-line allosteric inhibitors, such as TERN-701, [will also be investigated in clinical trials].
[The evolution of TKIs in CML is] a good story. A growing number of patients with CML are surviving CML, so survivorship is another effort and project at Memorial Sloan Kettering Cancer Center. The number of patients living with CML in the US will probably be 10 times what it used to be by the middle of the century. Asciminib will be a big help, [because] it offers better safety [than other TKIs]. We’ll see how other trials look.
BTK Inhibitors in CLL and MCL
Islam: BTK inhibition with small-molecule–targeted drugs has transformed the way we treat [patients with] B-cell malignancies over the past decade, ever since the advent of ibrutinib [Imbruvica] in the early 2010s. Newer-generation BTK inhibitors continue to improve safety and efficacy and are now even trying to overcome the resistance mechanisms we’ve seen with covalent BTK inhibitors. These drugs are being studied as monotherapies and have promising efficacy as single agents. [They] are also being combined with already-approved and emerging therapies. This has been a paradigm shift in both CLL and MCL, away from combination chemoimmunotherapy and more intensive therapies like autologous stem cell transplantation, potentially. There’s much to come, and many exciting data will be published in the next couple of years.
Updates in DLBCL Management
Lue: POLA-R-CHP [rituximab (Rituxan), cyclophosphamide, doxorubicin, polatuzumab vedotin-piiq (Polivy), and prednisone], has become the standard of care [(SOC) for patients with DLBCL with an] International Prognostic Index [score of] 2 and activated B-cell biology. We still believe dose-adjusted R-EPOCH [etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin, and rituximab] or more intensive therapies for high-grade B-cell lymphoma or double-hit or triple-hit lymphoma should be the SOC. The role of central nervous system prophylaxis in DLBCL is controversial. [We may] need to develop ways to identify truly high-risk patients, [such as through] novel assays like cell-free DNA to find subclone populations. Long-term follow-up for CD19-targeting CAR T-cell therapies [demonstrated] an overall survival benefit in refractory patients, and bispecific antibodies are having significant efficacy in the relapsed/refractory setting, although patients who relapse after bispecific antibodies and CD19-targeting agents [have] an unmet need.
Updates in Lower-Risk MDS Management
Bewersdorf: The treatment landscape for anemia in lower-risk MDS is finally moving. The [phase 3] COMMANDS trial [NCT03682536] showed luspatercept-aamt [Reblozyl] to be superior to ESA in patients with lower-risk MDS. In the second-line setting, imetelstat seems to be an effective option in ESA-refractory patients, independent of their [disease’s] molecular subtype.
At this point, [the role of] roxadustat is unclear. We’ll see what the final presentation of the [phase 3] MATTERHORN trial [NCT03263091] yields. There are still many open questions in the field. How do we sequence luspatercept and imetelstat? What do intriguing data [regarding] allele fraction reduction [show about roxadustat] as a disease-modifying effect? There’s more work to be done, but finally [we’re seeing] some progress.
Patients with hematologic myeloproliferative neoplasms (MPNs)—a group of rare blood diseases that include primary myelofibrosis, essential thrombocythemia (ET), and polycythemia vera (PV)—should be more active in their treatment plan, according to experts in oncology pharmacy who participated in a Pharmacy Times clinical forum in Chicago, Illinois, in June 2023. “Our role as pharmacists is to give [patients] as much information as we possibly can, then encourage them to move forward with advocating for themselves,” said Krystal Preston, PharmD, BCPS, a senior clinical oncology pharmacist at CVS Health and a clinical pharmacist at the University of Chicago Medicine.
Patients who are serious about taking an active role in their treatment could inspire health care providers to collaborate more, both with them and with other members of the care team, according to discussion leader Zahra Mahmoudjafari, PharmD, MBA, BCOP, FHOPA, clinical pharmacy manager of hematology, blood and marrow transplant, and cellular therapy at the University of Kansas Health System in Mission.
Subtypes of MPNs, ET, and PV typically transform into myelofibrosis, which can subsequently turn into acute myelocytic leukemia (AML). At least 20% of MPNs may transform into AML; therefore, the goal for treatment is to prevent this from occurring, Mahmoudjafari explained.
With more than 90% of patients with PV having a JAK2 mutation, “it is probably, by far, the mutation that we have the most actionable ability to do something about,” Mahmoudjafari said. She noted that there are 3 FDA-approved Janus kinase (JAK) inhibitors for MPN—ruxolitinib (Opzelura; Incyte), fedratinib (Inrebic; Bristol Myers Squibb), and pacritinib (Vonjo; CTI BioPharma Corp)—which were approved based on results from the COMFORT-I (NCT00952289), JAKARTA (NCT01437787), and PERSIST-2 (NCT02055781) pivotal trials, respectively.
Ruxolitinib and fedratinib are primarily for patients with intermediate- or high-risk myelofibrosis, including intermediate-2 risk and primary and post-PV/ET myelofibrosis, Mahmoudjafari explained. Pacritinib is indicated for patients with a platelet count below 50,000; all 3 JAK inhibitors have expected adverse event (AE) profiles, which include thrombocytopenia, anemia, bruising, dizziness, headache, and diarrhea.
Although the only true potentially curative treatment for myelofibrosis is transplant, there is a 30% mortality risk associated with it, Mahmoudjafari said. Further, patient adherence remains a predominant issue in patient care, according to Connor Roth, PharmD, BCOP, hematology/oncology pharmacy specialist with Franciscan Alliance, Inc in Chicago, Illinois. Whether due to dosing schedule, toxicities, cost, or all these reasons, people remain forgetful, Roth said. It is much harder to contact patients with a reminder via phone call because “nobody picks up [a call from] a phone number they don’t know,” Roth added.
Tammy McClellan, PharmD, a clinical oncology pharmacist at Riverside Healthcare in Kankakee, Illinois, said one of the greatest unmet needs she is seeing is timely access to medications. The faster a patient can get on a proper treatment regimen, the better they can prevent a blood-clotting event.
Insurance is another barrier to access; however, pharmacists understand how to work within the system and are best positioned to advocate for patients, according to Roth. Location is equally important for access to medications because patients living close to a city can access treatment centers and pharmacies more easily than those in a rural setting. Patients in cities also have better access to clinical trials, Preston said.
McClellan noted that an unmet patient need is effective communication with care providers. She said patients frequently mention that their provider does not listen to their input often enough.
McClellan said a solution may be individualized patient care. Further, Latha Radhakrishnan, PharmD, BCOP, BCPS, a clinical oncology pharmacist and an assistant professor in the College of Pharmacy at the University of Illinois at Chicago, noted that pharmacists and providers can foster improved individualized care through better organized collaboration with the patient and care team. This can make it easier to manage AEs and drug-drug interactions because treatment can be exceedingly difficult, according to Mahmoudjafari. Therefore, improving AE management can improve patient quality of life. “[Symptoms can be] enough to drive these patients absolutely insane,” McClellan added.
Additionally, financial burden is a significant issue for many patients. For this reason, some clinics have financial navigators who work with pharmacists and patients to coordinate benefits, co-pays, and prior authorization. Other institutions may assign these tasks to specialty pharmacists, who typically have experience with patient assistance programs that help older adults or individuals with limited resources to access affordable medications via grants, foundational support, or other means. Ideally, insurance or patient assistance would be connected to the patient’s electronic medical record, according to Roth. The panelists also emphasized patient education. “I really try to explain to [patients], in layman’s terms, what’s going on and just listen to what their issues are,” Preston said.
The panelists said that a best practice is to provide as much information about the disease state and treatment as possible. Because many patients do not understand their disease state, improving their understanding can provide the patient with more control, which can lead them to feeling better able to express concerns and be their own advocate.
“You can’t make the assumption that the patient already knows [everything],” Mahmoudjafari said. This is especially important because oncologists or other providers may be struggling to keep up with a complicated, changing treatment and guidelines landscape.
“Guidelines are dividing, and there’s so many things to know [about the drugs],” Roth said. “Pharmacists can be the ones to extend the hands of the physicians and be a patient advocate when [the patient] doesn’t always have one.”
Reference
American Society for Clinical Oncology. Pharmacy Times Clinical forum. 2023 American Society of Clinical Oncology Annual Meeting; June 2-6, 2023; Chicago, IL. Accessed July 13, 2023. https://conferences.asco.org/am/attend
Alessandra Carabbio, Alessandro Maria Vannucchi, Elisa Rumi, Valerio De Stefano, Alessandro Rambaldi, Giuseppe Carli, Heinz Gisslinger, Francesco Passamonti, Juergen Thiele, Naseema Gangat, and Tiziano Barbui
Ample evidence has been provided that accurate discrimination between essential thrombocythemia (ET) and early prefibrotic primary myelofibrosis (pre-PMF) has an impact not only on presenting laboratory data but also on complications, like thrombosis, progression to overt myelofibrosis (MF), transformation to blast phase (BP), and overall survival [1,2,3,4,5,6,7,8]. However, studies estimating the epidemiology of these critical events in the two entities have mainly focused on one isolated outcome at a time, without considering the entire spectrum of multiple intermediate disease states, possibly affecting probabilities and risk factors of the outcome of interest. This situation calls for a multistate model approach, a technique that allows a more in-depth insight into intermediate factors likely influencing the progressive transitioning from one status to another.
The aim of the present investigation was to estimate the probabilities that intermediate-state passages, including thrombosis, overt MF, and BP, impact the final absorbing state (death) in ET versus pre-PMF. To this purpose, retrospective data from two multicenter and well-documented studies [1, 9] were used: (i) ET patients (n = 791) from a multicenter international study of 891 cases, selected for the availability of complete disease history [1] and (ii) pre-PMF patients (n = 382) from four different Italian centers [9]. Both studies were approved by all institutional review boards or ethical committees of participating centers.
At the time of diagnosis, treatment-naïve ET and pre-PMF patients revealed different hematologic and clinical characteristics (Table S1). A parametric Markov multistate model [10] was applied to analyze data on survival considering intermediate states that are part of the natural history of ET and pre-PMF. The model included five states with ten possible transitions (Fig. S1): all patients begin in the initial state of diagnosis (ET, panel A, n = 791 or pre-PMF, panel B, n = 382) and then they could transit through the occurrence of an incident thrombotic event (Table S2) and/or the evolution to overt MF and/or BP (transient states) before death (absorbing state).
In ET, transition-1 from diagnosis to thrombosis included 101/791 patients (12.7%), but this status was transient in 21/101 patients that moved to death (21%) after a median time of 4.0 years (IQR: 1.6–6.4), 3/101 (6%) and in 1/101 (2%) to MF and BP, after a median time of 4.7 and 5.2 years, respectively. Remarkable was that in pre-PMF, the direct transition to thrombosis was found in 13.9%, a figure not different from ET (i.e., 12.7%). Conversely, pre-PMF substantially differed from ET for a higher rate of direct transition to overt MF or BP, that was 13 and 4% vs. 4 and 0.6%, respectively.
After 10 years, the state occupation probability of being event-free was 70 and 50% in ET and pre-PMF, respectively, and progressively decreased, particularly in pre-PMF (Fig. S2), due to earlier mortality, particularly for a greater probability of hematological evolutions. This trend was even more evident for death; regardless of the pathways through hematological evolutions, deaths were double in pre-PMF than ET, reaching 30, 60, and 80% vs. 15, 30, and 60% at 5, 10, and 20 years, respectively.
Probabilities to direct transition to thrombosis (n = 101 in ET and n = 53 in pre-PMF) and overt MF (n = 29 in ET and n = 51 in pre-PMF) are compared in Fig. 1. The trend of experiencing thrombosis directly after ET diagnosis showed to increase in the first 10 years (10%) and to decline subsequently (less than 5% at 30 years). On the contrary, in the first decade after diagnosis (<5%), the same probability grew slowly in pre-PMF while subsequently rose up to crossing the ET trend (8% after 30 years). Instead, the direct transitions from pre-PMF to overt-MF had an opposite trend: in the first 10 years, it reached a peak of 11%, while in ET, the trend was less pronounced, reaching a probability not exceeding 2.3% in the same period post-diagnosis.
Fig. 1: Direct transition probabilities to thrombosis and evolution in overt MF.
Direct transition probabilities over time from diagnosis of ET or pre-PMF to thrombosis (A) and overt MF (B). Transition probabilities are defined as the probability of going from a given state to the next state in a Markov process. Direct transitions refer to all the 791 and 382 ET and pre-PMF patients, respectively, initially at risk; thus, they represent the probability that a patient can first experience thrombosis or evolve into overt MF.
The performance of the IPSET-thrombosis score [11] was tested in both ET and pre-PMF for the direct transition to thrombosis. In ET, considering the low-risk group as a reference, the intermediate and high-risk groups determined by IPSET-thrombosis were confirmed to predict the thrombotic risk (HR = 2.08, 95% CI = 1.28–3.37, p = 0.003 and HR = 3.13, 95% CI = 1.82–5.40, p < 0.001, respectively). In pre-PMF, the same model was unpowered to reach statistical significance in the intermediate-risk group (HR = 2.50, 95% CI = 0.87–7.21, p = 0.089), while it was in the high-risk category (HR = 3.93, 95% CI = 1.52–10.11, p = 0.005).
Concerning survival, most of the deaths in ET and pre-PMF occurred directly from diagnosis (Fig. 2). The intermediate events that most influenced death were thrombosis (25.3%) in ET and BP (23.8%) in pre-PMF. In comparison with ET, the probability of direct transition from diagnosis to death in patients with pre-PMF increased linearly over time (Fig. 2) and was twofold higher, reaching values of 15, 30, and 60% at 5, 10, and 20 years, respectively. Of note, the probability of death in ET patients with an intermediate thrombosis state maintained a fourfold higher value over time than the ones without thrombosis. As expected, the probabilities of death in MF or BP status were higher and occurred faster, and not substantially different in ET or pre-PMF.
Fig. 2: Transition probabilities to death in ET and pre-PMF.
Comparison of the direct and indirect (via thrombosis, evolution in MF or BP) transition probabilities to death (absorbing state) over time from diagnosis of ET (dash lines) or pre-PMF (solid lines).
We confirmed the good performance of the IPSET-survival score [12] to differentiate the risk of direct mortality in ET (intermediate: HR = 4.38, 95% CI = 1.63–11.73, p = 0.003 and high-risk: HR = 20.17, 95% CI = 7.71–52.78, p < 0.001, compared to low-risk). The IPSET-survival score was equally well performing in pre-PMF (HR = 3.52, 95% CI = 1.22–10.12, p = 0.019 and HR = 13.37, 95% CI = 4.63–38.57, p < 0.001, for intermediate and high-risk groups, respectively, compared to low-risk). However, the discriminatory power of the IPSET-survival in ET was lower when the multistate model evaluated the mortality mediated by the thrombosis state; only high-risk patients were discriminated (HR = 19.27, 95% CI = 2.46–51.01, p = 0.005), whereas the intermediate-risk group was not significantly different from the low-risk one (HR = 3.84, 95% CI = 0.55–26.12, p = 0.194). Thus, in addition to the IPSET-survival risk factors (i.e., age ≥60 years, previous thrombosis, and white blood cells count ≥11 × 109/L) we found that platelets count ≥1000 × 109/L (HR = 5.74, 95% CI = 1.79–18.40, p = 0.003) and arterial vs. venous thrombosis in the follow-up (HR = 4.43, 95% CI = 1.04–18.91, p = 0.044) were independent predictors. The low number of deaths after thrombosis (12/105, 11%) in pre-PMF did not allow us to analyze the IPSET-survival performance in this transition.
The present multistate analysis, provides new insights for a better understanding of ET and pre-PMF disease processes. For example, in pre-PMF, the probability of thrombosis in the first decade was lower (<5%) due to a strong competitor represented by the evolution in MF (up to 11%). Consequently, occupation of thrombosis state in the first decade was lower in pre-PMF than in ET patients but became comparable in the last decades of observation (13 and 14% in ET and pre-PMF, respectively), supporting our previous cumulative estimates obtained with conventional methodology [1]. This notion might have practical implications to differentiate treatments during the course of the two entities by preferring antithrombotic prophylaxis according to IPSET thrombosis in ET that kept its discriminatory power also in this multiple competing adjustment analysis. In pre-PMF, therapy of first choice might be directed to prevent myelofibrosis evolution, provided agents able to do that are positively evaluated in appropriate clinical trials.
Regarding BP evolution, we highlight that the direct transition from the diagnosis was predominant in ET (n = 6/7, 86%), and it was modestly influenced by the pathway through thrombosis (n = 1/7, 14%). Unfortunately, we could not provide sufficient information on the role of cytoreductive therapy in these transitions due to the unreliable timing of drug administration.
Mortality prediction in ET was the topic addressed in a previous study [12]. On the basis of the hazard ratio estimates from Cox regression models, the IPSET-survival model was constructed, and 867 ET patients were allocated into three risk categories with significantly different survival [12]. In the present analysis, in a selected group of patients (n = 791) from the same database, we re-evaluated the risk factors of death considering the possible influence of the intermediate states that occurred before death, and confirmed the performance of the IPSET-survival scoring system for the prediction of direct mortality. However, we also found that the effect on mortality exerted by the intermediate thrombosis state was not negligible (accounting for 25% of deaths) and fourfold higher than in patients without incident thrombosis. Whether the reduction of vascular complications may impact survival remains to be demonstrated in appropriate prospective studies. Furthermore, we found two additional independent predictors of mortality in thrombosis-mediated transition, namely platelet count >1000 × 109/L (HR = 5.74, 95% CI = 1.79–18.40, p = 0.003), in line with a previous observation [13], and the incident arterial vs. venous thrombosis (HR = 4.43, 95% CI = 1.04–18.91, p = 0.044).
Limitations of this study concern its retrospective design and a possible bias related to the reporting accuracy of events, in terms of completeness and timing. In addition, since in these databases, the administration times of the cytoreductive drugs (hydroxyurea in absolute prevalence) were not well specified, we could not reliably evaluate the influence of the pharmacological cytoreduction on the post-diagnosis events. Furthermore, given that current results were obtained in the same ET database used for IPSET scores, a possible “self” confirmation bias could not be excluded. However, our aim was not to confirm the overall performance of the two scores, but to evaluate whether the transition from one state to another could have affected the overall survival or the cumulative incidence of thrombosis in a different way.
Strengths of the study are the relatively large number of patients for rare diseases such as ET and pre-PMF and the clinical and hematological diagnostic accuracy of the two entities and outcomes.
In conclusion, this multistate analysis provides novel information on the temporal probability of intermediate critical events occurring in ET and pre-PMF, and their impact on mortality. This knowledge might inform clinical practice and could also make more feasible the design of clinical trials.
Helna Pettersson, Jenni Adamsson, Peter Johansson, Staffan Nilsson, Lars Palmqvist, Bjorn Andreasson, Julia Asp
Introduction: Myeloproliferative neoplasm (MPN) is a heterogenous group of hematological malignancies including polycythemia vera (PV), essential thrombocythemia (ET) and primary myelofibrosis (PMF). JAK2V617F is the most frequent driver mutation in all three entities, but in PMF and ET mutations in CALR and MPL are also frequent. Mutations seen in additional genes are also often the same regardless of subtype of MPN. The aim of this study was to analyze a population based MPN cohort for genetic variants with prognostic value that can guide clinical decisions.
Methods: MPN patients from Western Sweden diagnosed between 2008-2013 (n=248) were screened for mutations in 54 genes associated with myeloid malignancy.
Results: Mutations in the genes SRSF2 and U2AF1 correlated significantly with impaired overall survival but did not correlate to increased risk for vascular events, neither before nor after diagnosis. Rather, mutations in these genes showed an association with disease transformation. Several recurrent gene variants with allele frequency close to 50% were confirmed to be germline. However, none of these variants was found to have an earlier onset of MPN.
Discussion: In conclusion, we identified gene mutations to be independent markers of impaired survival in MPN. This indicates the need for more individualized assessment and treatment of MPN patients and a wider gene mutation screening already at diagnosis. This could ensure the identification of patients with high-risk mutations early on. In addition, several genetic variants were also identified as germline in this study but gave no obvious clinical relevance. To avoid conclusions from non-informative genetic variants, a simultaneous analysis of normal cell DNA from patients at diagnosis should be considered.
Introduction
Polycythemia Vera (PV), Essential Thrombocythemia (ET) and Primary Myelofibrosis (PMF) all belong to the Philadelphia chromosome negative myeloproliferative neoplasm (MPN) category. These three entities share the same characteristics of causing proliferation of bone marrow cells, resulting in an increase of blood cells of myeloid lineage in the bone marrow and in peripheral blood. Advanced stages of PMF, on the other hand, is characterized by increase of reticulin fibers leading to decreased blood cells (1–3). The complications of these three entities are also similar regarding vascular events, i.e., thrombosis and bleeding. Furthermore, all three entities can transform into acute leukemia and have an impact on survival, however with large differences in frequency. Despite the common driver mutations in JAK2, CALR and MPL (4), the clinical presentation, risk and frequencies of complications and survival differ wildly between individual patients. Prognostic tools are therefore desired in clinical practice for follow-up and treatment decisions already at diagnosis. Modern sequencing techniques have given the opportunity to simultaneously analyze several mutations in blood malignancies. It has become widely used in both research and clinical practice (5). We and other groups have published data on risk mutations using this approach, but several studies analyze data on separate MPN entities or driver mutation groups. Since both driver mutations and several of the additional mutations found are shared between the subtypes of MPN, and since the occurrence of some additional mutations are rather rare, we hypothesize that analysis of mutations in MPN as one group has a potential to extend the prognostic value of genetic markers. Furthermore, there is a growing number of hereditary gene variants that have been linked to predisposition for development of hematological disorders, including MPN (6–11). This also represents a challenge when analyzing large amount of sequencing data, especially if comparison with normal non-malignant cells is not available. We also need to get more information regarding gene variants of unknown significance, to avoid overestimation of their importance but also to identify variants that might influence disease development and prognosis. In this study, we analyzed a well-defined, population based MPN cohort, regardless of subtype, for genetic variants. The aim was to search for additional prognostic markers that can be used to guide clinical decisions, as well as to investigate the potential impact of germline variants detected in the sequence data.
Materials and methods
Patients
All patients diagnosed with PV, ET or PMF according to the 2008 WHO diagnostic criteria (3) in Western Sweden at the Sahlgrenska University Hospital or NU Hospital Group between 2008 and 2013 and reported to the Swedish national blood cancer registry were identified. Basically, all patients in our health care region with suspected MPN are referred to and treated at these two hospitals. Thus, this cohort cover patients in the geographic area without any selection. Of this cohort, 248 patients were included in the study based on informed consent, availability of DNA sample from the time of diagnosis, and review of diagnosis. Data from a subset of this patient cohort have been published previously (12, 13). Details are outlined in Supplementary Table 1. Clinical characteristics, clinical course, vitality status, vascular complications, disease transformation and co-existing cancers were collected from the medical records of all patients. Each of the patient’s hospital records were searched after emergency care consultation and hospital admission records that is related to bleeding or thrombotic complications. Follow-up was done from diagnosis until June 2021. The study was performed in accordance with the Declaration of Helsinki after ethical approval.
Screening for myeloid mutations
Genomic DNA from whole blood from the same sample that was analyzed at diagnosis for the presence of JAK2, CALR or MPL mutation was screened for gene variants in 54 genes or mutational hot spots associated with myeloid malignancies. The TruSight Myeloid Sequencing panel (Illumina FC-130-1010) which also was used for diagnostics in the clinical laboratory at the time of the study, was used, and sequencing was performed on a MiSeq instrument (Illumina) according to manufacturer’s instructions. Secondary analysis was performed with MiSeq Reporter, v.2.4.60.8, using Burrows-Wheeler Aligner mapper and somatic variant caller (Illumina). Data were filtered and mapped to the human genome reference hg19 using Variant studio v3.0 (Illumina), where global filtering was set to >3% and coverage had a minimum of 500 reads. Variants causing missense, frameshift, an altered stop/initiation codon, in-frame insertion/deletion or variants affecting splice site were regarded as mutations. Variants with quality >Q30 and allele frequencies of at least 5% were considered positive for mutation. Known sequencing artefacts and variants previously found in normal controls were excluded from further analysis according to filter strategies used in the clinical laboratory. BAM files from secondary analysis were used to analyze selected variants by Integrative Genomics Viewer (www.broadinstitute.org). Variants in areas with difficult reads were excluded. Previously analyzed data was reanalyzed according to updated bioinformatic settings to make the results comparable regardless of time for sequencing.
Confirmation of germline variants
Blood sample was taken from patients with variants in CDKN2A (rs3731249), ETV6 (rs145477191), NOTCH1 (rs61751489) or MPL (rs41269541), with a variant allele frequency close to 50%. Blood was enriched for CD3+ cells, using MACS® cell separation kit StraightFrom™ Whole Blood CD3 MicroBeads (Miltenyi Biotech) and Whole Blood Column Kit (Miltenyi Biotech), according to the manufacturer’s protocol. Genomic DNA from CD3+ enriched cells were extracted using QIAamp DNA Blood Mini Kit (Qiagen) and 10 ng of DNA were genotyped using TaqMan SNP genotyping assay (Applied biosciences, Life technologies) according to manufacturer’s protocol. The following assays were used: CDKN2A (assay ID: C_25611114_10), ETV6 (assay ID: C_162058060_10), NOTCH1 (assay ID: C_90123839_10) and MPL (assay ID: ANFV4EK). All samples were analyzed in triplicates using the QuantStudio 3 Real-Time PCR system (ThermoFisher Scientific). Genotypes were determined automatically based on dye component fluorescent emission data depicted in the X–Y scatter plot using Taqman genotyper software v.1.6.0. The gnomAD database v2.1.1 (https://gnomad.broadinstitute.org/) was used to compare frequencies in the MPN cohort with a normal Swedish population.
Statistical analysis
Fisher’s Exact Test was used to compare differences in frequencies between groups. To estimate overall survival (OS), defined as time from diagnosis to last follow up or death from any cause, the Kaplan Meier Log-rank test was used initially. For multivariable analysis, logistic regression and Cox Regression was used. P-values <0.05 were considered statistically significant. The statistical software used were Analyze-it v.6.15.4 (Microsoft Excel), GraphPad Prism v.9.4.0 and SPSS v29.0.0.0.
Results
A population-based cohort
Between 2008 and 2013, 300 patients were diagnosed with MPN at Sahlgrenska University Hospital and NU Hospital Group in Western Sweden. Of these, 83% (n=248; PV n=84, ET n=123, PMF n=41) were included in the study. All included patients fulfilled the 2008 WHO diagnostic criteria. Age, gender, and blood counts from the time of diagnosis, for the whole MPN group and for the sub entities, are presented in Table 1. The distribution of driver mutations found at diagnosis was consistent with expected findings in the different subgroups of MPN (Figure 1). One patient with PMF was found to harbor both JAK2 V617F as well as mutation in the CALR gene. Patients not included in the study either declined to participate, had another diagnosis when their medical records were reviewed, or diagnostic material was missing. The median age of these patients was slightly lower (66 years vs. 69 years) but there was no other significant difference between these and the included patients.
Table 1
Table 1 Age, gender and laboratory findings at diagnosis in 248 patients with MPN.
Figure 1
Figure 1 Patients by diagnosis and driver mutations. TN, triple negative.
Mutations and survival
A sequencing panel including 54 genes associated with myeloid malignancies was used to screen for genetic variants that could be used as prognostic markers. Variants regarded as mutations other than the diagnostic driver mutations (JAK2, CALR or MPL) were found in 37 genes in at least one patient and in 27 genes in at least three patients (Figure 2A; Supplementary Table 1). During analysis of the gene data, several recurrent gene variants with allele frequency close to 50% were noted, which implied a hereditary variant. Therefore, analysis of the most common variants CDKN2A (NM_001195132.1:c.442G>A), NOTCH1 (NM_017617.3:c.6853G>A) and ETV6 (NM_001987.4:c.602T>C) as well as a variant close to a splice site in the MPL gene (NM_005373.2:c.1565 + 5C>T) were also analyzed in separated T-cells from new blood samples. This confirmed the variants to be germline. Therefore, these variants were excluded from analyses of prognostic impact. Sixty-three percent of the MPN cases had other mutations in addition to the diagnostic driver mutation (Figure 2B). Presence of at least one additional mutation was found to be associated with inferior survival (Figure 2C). To investigate if it was mutations in general or mutations in particular genes that had impact on survival, all genes with mutations detected in at least three patients were correlated to survival with the Kaplan Meier Log-rank test. For the whole MPN group, only mutations in five genes correlated significantly with inferior overall survival, ASXL1 (P=0.0005), SRSF2 (P<0.0001), U2AF1 (P<0.0001), CBL (P=0.01) and SF3B1 (P<0.0001) (Figure 3). These were further tested with multivariable analysis using Cox regression where also age and type of diagnosis were taken into consideration. When the five genes were grouped together, they still correlated to OS (P=0.002) with a hazard ratio 3.248, and it was not dependent on type of diagnosis (interaction 0.592). Also, age at diagnosis correlated to OS (P<0.001) as expected. However, it should be noted that all cases with CBL mutation also harbored mutation in another of the four genes (Supplementary Table 1). Therefore, the genes were also tested separately. When these were adjusted for both age and type of diagnosis, only mutations in SRSF2 and U2AF1 correlated significantly to OS (Table 2).
Figure 2
Figure 2(A). Frequency of additional mutations. (B). Distribution of patients with driver mutation only or addition mutation(s) when confirmed germline variants have been excluded. (C). OS in patients with driver mutation only or with addition mutation(s) (confirmed germline variants excluded).
Figure 3
Figure 3 OS in patients with mutations in ASXL1, CBL, SF3B1, SRSF2 and U2AF1 respectively according to Kaplan Meier Log-rank test.
Table 2
Table 2 Mutation impact on OS in univariate or adjusted (including age and type of diagnosis) analysis.
Mutations and vascular events
Vascular events in MPN are potentially life-threatening. The vascular complications are either thrombosis or bleeding where the co-existence of MPN is a contributing factor. The most common incidences are those that are discovered at the time of diagnosis and the most common thrombotic events were myocardial infarction (n=27), cerebrovascular infarction (n=24), pulmonary embolism (n=19), transient ischemic attack (n=10) and deep vein thrombosis (n=8). The most frequent hemorrhagic complications were gastro-intestinal (n=11) and cerebral bleedings (n=6). Fisher’s Exact Test and logistic regression were used to analyze if SRSF2 or U2AF1 which correlated with shorter OS also correlated with occurrence of vascular events before or after diagnosis or in total. However, no such correlation was seen, neither when only the mutated genes were tested, nor when they were combined with age and type of diagnosis.
Mutations and disease transformation
All MPNs have a risk of transformation into secondary acute myeloid leukemia (AML). In our cohort, 17 cases had transformed to AML. Both mutated genes with correlation to OS were tested with logistic regression. This showed that mutations in SRSF2 correlated with AML transformation (P=0.002), but this was not the case for U2AF1 (P=0.236). The analysis was extended to find genes correlated to fibrotic transformation and co-existence with other myeloid hematological malignancies. There were 18 patients that had secondary myelofibrosis transformation from PV and ET. Other myeloid hematological malignancies that co-existed with MPNs included two chronic myelomonocytic leukemia and one myelodysplastic syndrome. Logistic regression showed that mutations in both SRSF2 and U2AF1 correlated with co-existing myeloid hematological malignancies (SRSF2 P=0.05 and U2AF1 P=0.014).
Gene germline variants
Identified germ line variants indicated a possible hereditary predisposition of MPN. Comparison of the frequency in our MPN cohort to a normal Swedish population cohort in the gnomAD variant database was performed. Only the variant found in ETV6 was more frequent in the MPN group (0.0282 vs. 0.00975 in allele frequency). This difference was statistically significant (Fischer’s exact test, p=0.0006). However, there was no correlation between any of the variants and occurrence of early onset MPN. We further used logistic regression to test if any of the variants correlated with occurrence of other cancers (both solid tumors and hematological malignancies outside the MPN group). These cancers occurred about the time and after diagnosis of MPN and were noted upon reviewing the patient´s hospital records. In total, 19 patients with non-hematologic cancers were found. The most common types were colon cancer (n=6) and pancreatic cancer (n=4). However, no significant correlation was seen.
Discussion
The serious risks all MPNs impose, although at various frequencies, are vascular complications, transformation to more severe hematologic malignancies and ultimately negative impact on OS. It is thus a priority to identify high risk patients in clinical practice. Age at diagnosis as well as occurrence of vascular complications have been reported as risk factors (14, 15). Access to an abundance of genetic data allows genetic profiling to further broaden prognostic information. Mutational status has progressively taken a big role in clinical practice. Occurrence of mutations have also been used to create scoring systems for MPN (16, 17). The initial focus on gene mutations in MPN was on the driver mutations’ importance on disease development. These mutations are found in the genes JAK2, CALR and MPL which all are involved in JAK-STAT signaling (18). Notably, the same JAK2 mutation is found in both PV, ET and PMF, and mutations in CALR and MPL are seen in both ET and PMF. Thus, the mutation itself does not seem to determine the MPN phenotype, instead, allele burden has been reported as one factor behind the phenotypic differences (19). The order of acquisition of the driver mutation in relation to additional mutations may also have influence (20, 21). If the JAK2 mutation precedes mutation in DNMT3A or TET2, the phenotypic picture would likely be PV. If mutations instead occur in reverse order, the MPN phenotype would likely be ET (22). Host factors also contribute to the development of disease (22, 23). Several predisposing gene variants have been identified that may influence not only the risk of developing disease but also the course of the disease (17, 24–26). It is therefore reasonable to investigate the whole MPN cohort as a group independent of diagnosis when analyzing it from a genetic point of view.
Aside from the driver mutations, several additional mutations have also been reported in MPN. These are subclassified to gene families: epigenetic regulators (ASXL1, EZH2, TET2, IDH1/2, DNMT3A), spliceosome (SRSF2, SF3B1, U2AF1, ZRSR2), transcriptional regulators (TP53, RUNX1, IKZF1), general cell signaling genes (KRAS, PTPN11) as well as specific negative regulators of JAK/STAT signaling (CBL) (27). Occurrence of additional mutations correlated significantly with inferior OS (Figure 2C). Genetic profiling has raised the question if the number of mutations in a particular case is more important for the occurrence of complications or OS rather than in what genes or type of gene the mutations are present. Our results suggest it is not the number of mutations but rather the presence of certain gene mutations that are more informative for prognostic guidance. Mutations in SRSF2 and U2AF1 correlated significantly with worse OS in our patient cohort. Although they were more frequent in PMF, which is well known to have an impaired survival compared to patients with PV and ET, mutations in these two genes correlated to worse OS regardless of MPN subtype.
Previous studies, including a subpopulation of our analyzed MPN cohort, have shown that triple negative MPN without JAK2, CALR or MPL mutation have worse prognosis (12, 28). It has also been shown that the presence of several other mutations in addition to a driver mutation correlate with survival (2, 5, 24, 26, 28–32). In this study we initially identified mutations in five non-driver genes (ASXL1, SRSF2, U2AF1, SF3B1 and CBL) to be significantly correlated to impaired OS. Mutations in both ASXL1 and SRSF2 have previously been classified as high-risk mutations in both PMF and PV (26, 33). Moreover, mutations in U2AF1 and SF3B1 have been identified as genetic risk factors in ET (17). When age at diagnosis as well as type of diagnosis was taken into consideration only mutations in SRSF2 and U2AF1 remained associated with shorter OS. Mutations in CBL were only found in those patients who harbored mutations in one or more of the four other genes, suggesting that mutated CBL might just be a passenger rather than a disease driver. Mutations in ASXL1 is commonly seen in clonal hematopoiesis, which increases with age (34). This could be an explanation why the presence of ASXL1 mutation no longer significantly correlated with OS when age was taken into consideration.
Since vascular complications are associated with impaired survival we wanted to investigate if the detected mutations correlated also to vascular events in our MPN cohort. However, neither mutations in SRSF2 nor U2AF1 correlated to vascular events before or after diagnosis or in total. Another complication with MPN is transformation to myelofibrosis for PV and ET or to secondary AML for all three MPN. In the present cohort, a significant correlation between mutations in SRSF2 and transformation to AML was found. Furthermore, mutations in both SRSF2 and U2AF1 correlated with transformation from PV and ET to myelofibrosis and development of other hematological malignancies. This is in line with previous findings were mutations in SRSF2 and U2AF1 have been reported to serve as prognostic markers for rapid blastic progression in newly diagnosed MPN (35). Moreover, mutations in SRSF2, U2AF1 and SF3B1 detected at presentation of disease have been associated with rapid fibrotic progression in PMF. This was not demonstrated for mutations in ASXL1, DNMT3A or TET2 (36).
Aside from the acquired mutations in our MPN cohort, several specific variants were identified which turned out to be germline. A five- to sevenfold higher risk of MPN among first-degree relatives to MPN patients have previously been reported in Sweden, which suggest a genetic predisposition (37). Also in other myeloid malignancies, the question for germline variants involved in disease have come into focus (38–40). Four variants in our study were more closely investigated, their allele frequency was close to 50%, which could imply a hereditary variant. These genes were CDKN2A (NM_001195132.1:c.442G>A), NOTCH1 (NM_017617.3:c.6853G>A) and ETV6 (NM_001987.4:c.602T>C) as well as a variant close to a splice site in the MPL gene (NM_005373.2:c.1565 + 5C>T). The most frequent occurring CDKN2A mutation leading to a p.A148T substitution has been reported as an inherited coding variant associated with leukemic transformation of hematopoietic progenitor cells (41). Comparison of the frequency to a normal Swedish population cohort, however, only revealed the ETV6 variant to be more common in the MPN patient cohort. This variant did not correlate to earlier onset of disease, which could be expected for an inherited predisposition. On the other hand, in the Landgren study the mean age at diagnosis did not differ between affected relatives and controls (37). The ethical approval of the current study did not include testing of relatives, but it would of course be of interest to see if any of these variants are associated with an increased incidence of hematological or non-hematological malignancies within these families. Nevertheless, it is important to correctly identify germline gene variants to avoid drawing conclusions from non-informative genetic variants but also to provide genetic counseling when called for. In this study we used CD3+ selection of T-cells from collected blood samples to get constitutive DNA, which turned out to be easiest for both referring doctors and gave acceptable DNA yield for the laboratory but might of course misdiagnose somatic variants that are also present in lymphoid cells. Another alternative is a skin biopsy but this may be considered too much of an intervention for some patients.
In conclusion, our study on a population based MPN cohort strengthens previous reports about prognostic value of genetic data in MPN. Thus, a wider gene profiling at diagnosis is of value. In addition, several genetic variants were also identified as germline in this study but gave no obvious clinical relevance. To avoid conclusions from non-informative genetic variants, simultaneous analysis of normal cell DNA from patients at diagnosis should be considered.
Myelofibrosis is a rare type of bone marrow cancer. In this condition, extensive scarring (fibrosis) occurs in the bone marrow, which keeps the bone marrow from producing the right number of blood cells.1
Healthy functioning bone marrow produces infection-fighting white blood cells (WBCs), oxygen-carrying red blood cells (RBCs), and blood-clotting platelets in the correct numbers the body needs to function normally.
When myelofibrosis occurs, some people may not have any symptoms, while others have severe symptoms that require immediate treatment. This article will explain the symptoms of myelofibrosis, how it is diagnosed, and how it is treated.
How Myelofibrosis Affects the Body
Bone marrow is present inside the middle of the bones. It is normally a soft, spongy texture. It produces WBCs, RBCs, and platelets.
In myelofibrosis, one of the cells of the bone marrow begins to grow abnormally, multiply, and continue to produce more abnormal cells. Eventually, abnormal cells are present in high enough numbers to crowd out the healthy cells.
These abnormal cells cause fibrosis, which prevents the bone marrow from producing the correct number of blood cells the body needs to function normally.2 Over time, there is an increased risk of developing acute myeloid leukemia (AML), another form of blood cancer.1
Myelofibrosis Types
The initial cause of abnormal bone marrow development determines the type of myelofibrosis. The two types are primary and secondary.
Primary Myelofibrosis
With primary myelofibrosis, the change in the bone marrow cells happens spontaneously. It doesn’t occur due to a previous bone marrow condition.1
Secondary Myelofibrosis
With secondary myelofibrosis, the fibrosis occurs due to another bone marrow disorder, specifically polycythemia vera or essential thrombocythemia.3
Polycythemia vera is a blood disorder in which the bone marrow produces too many blood cells, most often red blood cells, but can also include white blood cells and platelets.2 Essential thrombocythemia is a disorder in which the bone marrow makes too many platelets.4
Myelofibrosis Symptoms
As too few blood cells are made, symptoms will start to develop. The rate at which symptoms develop and how severe they become can vary from person to person, and may take years to be experienced. Symptoms associated with myelofibrosis include:2
Feeling tired
Shortness of breath
Pale skin
Headaches
Fever
Night sweats
Enlarged spleen
Enlarged liver
Frequent infections
Easy bleeding or bruising
Abdominal pain
Joint pain
Bone pain
Tumors may develop in the lungs, skin, liver, or gastrointestinal tract and cause further symptoms.2
Causes of Myelofibrosis
For those living with primary myelofibrosis, the exact cause of the disease may never be known.
However, in about half of the cases of primary myelofibrosis, a mutation in the JAK2gene is found.5 The JAK2 mutation is also frequently found in those with polycythemia vera and essential thrombocythemia.
This mutation isn’t inherited. Instead, it develops spontaneously in a bone marrow cell. It produces a protein that causes the bone marrow to overproduce platelet precursor cells called megakaryocytes. These cells stimulate other cells to produce too much collagen, a protein that then builds up and produces scarring in the bone marrow.
Other gene mutations that may play a role in developing myelofibrosis include the CALR and MPL genes.1
Risk factors that may play a role in developing primary myelofibrosis include:2
Increasing age
History of exposure to chemicals including benzene, fluoride, or phosphorus
Risk factors for developing secondary myelofibrosis include:6
Having another cancer that has spread into the bone marrow
Having polycythemia vera or essential thrombocythemia
Diagnosis of Myelofibrosis
The diagnosis of myelofibrosis often starts when someone presents to their healthcare provider for evaluation of a symptom that they are experiencing. During this evaluation, a healthcare provider may start with a detailed history and physical examination. Blood work may be taken which can start the process of finding a diagnosis.1
A complete blood count (CBC) and peripheral blood smear measure the number of WBCs, RBCs, and platelets, as well as their shape and size. Abnormal findings in the CBC may lead to further testing, which may include a bone marrow biopsy.1
During a bone marrow biopsy, a small sample of bone marrow is taken, often from the hip. This allows the pathologist (physician specializing in analyzing body fluids and tissues in a lab setting) the ability to evaluate for any changes or abnormalities in the bone marrow. This test can result in a diagnosis of myelofibrosis.
In addition, a physical exam or imaging study such as a computed tomography (CT) scan may reveal an enlarged spleen.
Other blood testing may include:1
Complete metabolic panel (CMP) to evaluate kidney and liver function
Coagulation studies
Iron levels
Lactate dehydrogenase (LDH) to assess inflammation and tissue damage
Testing for the JAK2, CALR, and MPL gene mutations
Once a diagnosis of myelofibrosis is made, it is further classified into different risk categories, which helps determine how likely the disease is to turn into AML and can help determine treatment options.
This score is determined by the person’s age, symptoms, hemoglobin level, platelet count, leukocyte (a white blood cell) and leukoblast (a developing white blood cell) count, and certain genetic changes. The higher the score, the more high-risk their myelofibrosis is.
Myelofibrosis Treatment
Some people diagnosed with myelofibrosis, especially those without many symptoms or who have low-risk disease, may not receive any treatment until they become symptomatic. Called a watchful waiting approach, this policy of taking no immediate action regarding treatment includes routine blood tests and visits with their healthcare provider to determine when treatment will be needed.2
If someone is experiencing symptomatic anemia (low RBCs) because of myelofibrosis, they may receive periodic RBC transfusions. There are additional medications that may be given to help the bone marrow make red blood cells.
This may not completely resolve anemia but can keep the red blood cells up at a tolerable level. If someone also has iron deficiency anemia, iron supplements may help improve red blood cell levels.2
To reduce high levels of WBCs and platelets, medications to suppress the bone marrow may be given. An example of one of these medications is hydroxyurea.
An enlarged spleen may need to be treated if it is contributing to symptoms, especially pain or severely low platelets. This can be done through radiation to the spleen or by surgical removal of the spleen.
A medication called Jakafi (ruxolitinib) can be prescribed to those with either primary or secondary myelofibrosis who fall in the moderate- or high-risk category. This medication interferes with the JAK2 pathway that the cells use to grow. Another medication, Inrebic (fedratinib) can also be used to treat intermediate or high-risk primary or secondary myelofibrosis.2
Can Myelofibrosis Be Cured?
The majority of cases of myelofibrosis are treated with the goal of decreasing symptoms of the disease. An attempt can be made to cure the disease through a stem cell transplant.
This approach requires large doses of chemotherapy to kill all of the cancer cells. Stem cells are collected before the procedure to be infused back in after chemotherapy has worked. These stem cells can then begin to resume making normal WBCs, RBCs, and platelets. This procedure is not recommended for everyone with myelofibrosis, as it can lead to severe complications.7
Complications Associated With Myelofibrosis
Complications associated with myelofibrosis are related to the severe decrease in the number of normal WBCs, RBCs, and platelets. As the disease progresses and the blood counts continue to decrease, complications may arise.
With the decrease of white blood cells comes a higher risk of developing infection. Infections can occur anywhere in the body, though most often in the lungs. The infection can be due to bacteria, viruses, or fungi. With infection may come fever, increased weakness, and cough.8
Low red blood cells can result in severe anemia, which prevents enough oxygen-rich blood from getting to the tissues in the body. With the decreased amount of available oxygen comes complications such as heart failure, in which the heart has to work too hard to try to keep up with the increased demand for oxygen.8
Not having enough normal platelets can lead to severe bleeding or hemorrhaging. The bleeding can occur following an injury or can occur spontaneously. The bleeding can become life-threatening if severe and not stopped quickly.8
Blood clotting, the opposite of bleeding, could also occur. If blood clots inappropriately, it can lead to clots moving around the body and getting stuck in areas they are not supposed to be. This can lead to damage to the brain, heart, lungs, and extremities.
Transformation to acute myeloid leukemia occurs in 5% to 10% of those diagnosed with myelofibrosis. This is most common in primary myelofibrosis and is a significant complication since the prognosis is poor with transformation into AML.9
Myelofibrosis Prognosis
The prognosis of myelofibrosis can vary from person to person. It depends upon the type and risk category of their disease. Prognosis can differ slightly based on which scale is used at the time of diagnosis. The table below references the prognosis scale MIPSS70, which is used for those 70 years old or younger and is based on risk group severity.9
Risk Group
10-Year Survival
Very high
Less than 5%
High
13%
Intermediate
37%
Low
56%
Very Low
92%
When to Contact a Healthcare Provider
See a healthcare provider if you are having symptoms associated with myelofibrosis. Many of these are also associated with other conditions. A workup and diagnosis can ensure you are getting the appropriate treatment.
If you have been diagnosed with myelofibrosis, contact your healthcare provider anytime you’re having concerns about the symptoms you’re experiencing, especially if they continue for some time without getting better.
Your provider may want to run additional tests or start treatment if your symptoms continue. If severe symptoms develop, notify your healthcare provider immediately or seek emergency care.
Summary
Myelofibrosis is a type of blood cancer in which abnormal cells cause the bone marrow to become extensively scarred (fibrosis). The fibrosis doesn’t allow the bone marrow to make blood cells properly, which leads to low blood counts and other complications.
Once formally diagnosed by a bone marrow biopsy, the results will be used by your healthcare provider to develop a treatment plan. Treatments are individualized, ranging from watchful waiting to stem cell transplant.
Oncology nurses must know what to look for — and what questions to ask — when treating patients with myeloproliferative neoplasms (MPNs), as side effects and other patient characteristics can play a role in determining the best treatment regimen, according to Julie Huynh-Lu, PA-C, a physician assistant from The University of Texas MD Anderson Cancer Center.
MPN-Related Symptoms
“Whenever a patient comes to see us, they also fill out the MPN10 questionnaire, [which is] a list of all the 10 symptoms that frequently occur in our patients,” Huynh-Lu said in an interview with Oncology Nursing News. “Ideally, this should be occurring at every visit. On top of them filling that out, I obviously will ask them specific pointed questions as well just to tease out some more information. But this should occur at every visit.”
Symptoms can vary based on the subtype of MPN a patient has. Huynh-Lu said that patients with polycythemia vera and essential thrombocytosis are more likely to experience headaches, confusion or difficulty focusing, or pain and tingling in the fingertips. Meanwhile, common symptoms for patients with myelofibrosis include anemia and thrombocytopenia; shortness of breath and fatigue; bleeding; and complications from spleen enlargement, such as having a poor appetite.
Knowing about these symptoms is key, as they could indicate a physical issue that warrants a change in treatment, Huynh-Lu said. For example, if a patient is not experiencing splenomegaly (enlarged spleen), there may not need to be prescribed a JAK inhibitor. However, if the patient starts to experience a decreased appetite or feel full after eating only a small amount of food, that could indicate that their spleen is becoming enlarged, and that patient may benefit from being put on a JAK inhibitor.
“It can also change the trajectory on whether or not talking about splenectomy is an option. It’s not really our go-to [treatment] in our department at MD Anderson, but that could certainly lend to a conversation into if surgery is an option,” Huynh-Lu said.
Sometimes symptoms can lead to a change in treatment, while other times there may be an easy fix to manage the issue.
If a patient is currently taking a JAK inhibitor, nurses should be sure to ask them about worsening itching, diarrhea, and frequent infections (such as urinary tract infections or pneumonia). Secondary skin cancers can also occur, said Huynh-Lu, “so we always recommend that they get dermatology checks every 6 months.”
“If their [blood] counts are starting to drop, or if their spleen is starting to grow, well, maybe the medication they’re on right now, the dosage needs to be altered. But if we alter the dose to a higher medication dose, and the side effects are worse, maybe then it’s time to switch to a different class of drugs completely, or same class of drugs, just a different type of drug. There’s also, of course, clinical trials that are available, so that could be an option as well,” Huynh-Lu said.
Patient Characteristics and Comorbidities
Regarding patient characteristics and comorbidities, clinicians should know if patients have a history of cardiac, renal, or hepatic complications, as certain medications can affect these organs.
Additionally, interferons are commonly used to treat patients with polycythemia vera. However, according to the National Institutes of Health, these drugs can impact the synthesis of serotonin, dopamine, epinephrine, and norepinephrine, thereby increasing a patient’s risk for depression. That said, clinicians should know if patients have a history of depression or an autoimmune disease before they prescribe an interferon to a patient, Huynh-Lu said.
It also may be beneficial for oncology nurses to ask patients if they are experiencing financial struggles due to their cancer care.
“I know these drugs can be quite expensive. Financially, this can be a burden for some … Sometimes the local oncologists aren’t completely aware of financial assistance available for them. So maybe just ask and say, ‘Hey, I know this drug cost this much. Do you know of any financial assistance that you guys can provide for me?’ Because I know sometimes that’s not a question that gets asked,” Huynh-Lu said.
