Myeloproliferative Neoplasms in the Adolescent and Young Adult Population: A Comprehensive Review of the Literature

Abstract

Myeloproliferative neoplasms (MPN) are characterized by a clonal proliferation of myeloid lineage cells within the bone marrow. The classical BCR-ABL negative MPNs are comprised of polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF). Historically, the majority of MPNs are diagnosed in adults older than 60 years of age, however in recent years, there has been recognition of MPNs in the adolescent and young adult (AYA) population. AYAs with MPN, typically defined as between the ages of 15 to 39 years old, may comprise up to 20% of patients diagnosed with MPN. They demonstrate unique patterns of driver mutations and thrombotic events and remain at risk for progression to more aggressive disease states. Given the likely long length of time they will live with their disease, there is a significant unmet need in identifying well tolerated and effective treatment options for these patients, particularly with the advent of disease modification. In this review, we provide a comprehensive overview of the clinical features, disease course, and management of AYA patients with MPN, and in doing so, highlight key characteristics that distinguish them from their older counterparts.

Graphical Abstract

Adolescents and young adults (AYA) with myeloproliferative neoplasms (MPN) demonstrate notable differences compared to older patients with MPN. With the most common MPN in this group being essential thrombocythemia (ET), patients are predominantly female, have a higher incidence of CALR mutations, and demonstrate a predilection for splanchic vein thrombosis. While their overall survival is superior compared to older adults, they remain at risk for transformation to more aggressive disease states, and priorities in treatment should not only target symptom burden and thrombotic risk reduction, but also disease modification. Unique considerations in their care include pregnancy/fertility-related issues as well as navigating psychosocial burdens.

Introduction

Essential thrombocythemia (ET), polycythemia vera (PV) and primary myelofibrosis (PMF) comprise the Philadelphia-negative myeloproliferative neoplasms (MPN) (1). While ET is characterized by megakaryocyte hyperplasia, patients with PV demonstrate significant erythrocytosis (2). PMF is associated with bone marrow fibrosis, cytopenias and extramedullary hematopoiesis (3, 4). ET, PV, and PMF carry the risk of developing vascular complications and progressing to acute myeloid leukemia (AML); transformation of ET and PV to MF may also occur. Patients are typically diagnosed in their seventh decade, with median ages of 69, 67, and 65 years for PMF, PV, and ET, respectively (5, 6). However, in recent years, there has been recognition in adolescents and young adults (AYA) being diagnosed with MPN.

Although varying definitions exist, here we consider the AYA population to encompass the ages of 15 to 39 years per the National Comprehensive Cancer Network (NCCN) and Surveillance Epidemiology and End Results (SEER) guidelines (7, 8). There have been several cohort studies specifically focusing on MPN AYA (9, 10, 11, 12, 13, 14, 15, 16). However, we do note there are minor inconsistencies in age cutoffs compared to the definition of AYA per NCCN and SEER, whereby several of these cohorts included patients younger than 15 years of age (9, 13, 16), as well as adults through the age of 45 (11, 12, 14, 15, 16) Although the inclusion of studies outside of the defined AYA age may lead to constraints in the interpretability of data (such as contributions of pediatric-age patients in AYA cohorts), we have, however, still included them as there is limited data in this field, and their median age at MPN diagnosis remained in the AYA range. The specific age ranges for each included cohort, as well as other pertinent clinical characteristics as discussed next, may be found in Table 1.

Table 1.

Clinical Characteristics of MPN AYA Cohorts

First author	MPN, (n)	Age range, years (median)	Driver Mutations	Thrombotic Events After Diagnosis	Transformation to post-MPN MF	Transformation to post-MPN BP
Harris	ET (114), PV (45), MF (12)	0 - 39 (31)	ET: JAK2 55%, CALR 26%, MPL 1%, TN: 13% PV: JAK2 100% MF: JAK2 42%, CALR 42%, MPL: 8%	Entire cohort: venous 19.8%, arterial 13.5%	ET: 9% PV: 24%	ET: 0.9% PV: 4% MF: 0%
Boddu	ET (105), PV (43), MF (37)	16 - 39 (33)	ET: JAK2 53% PV: JAK2 92% MF: JAK2 35%	ET: 2% PV: 2% MF: 3%	ET: 0% PV: 0%	ET: 2% PV: 0% MF: 1%
Szuber	ET (219), PV (79), MF (63)	18 - 40 (32 - 37)	ET: JAK2 48%, CALR 35%, MPL 1% PV: JAK2 100% MF: JAK2 39%, CALR 48%	ET: 16% PV: 27% MF: 5%	ET: 16% PV: 22%	ET: 2% PV: 4% MF: 10%
Barziali	ET (54), PV (37), MF (15), MPN-u (3)	18 - 45 (24 - 38)	ET: JAK2 75% PV: JAK2 91% MF: JAK2 60%	Entire cohort: 37% with atleast one thrombotic event	ET: 15% PV: 5%	ET: 0% PV: 0% MF: 13%
Sobas	ET (318), PV (81), MPN-other (45)	2 - 25 (20.4)	ET: JAK2 49%, CALR 15% PV: 92%	ET: 10% PV: 16%	ET: 7% PV: 9%	Entire cohort: 0.3%
Stein	PV (120)	29 - 41.4 (36.5)	JAK2 98%	27%	15%	2%
Palandri	ET (197)	18 - 40 (32)	JAK2 63%, CALR 24%, MPL 2%	10%	6%	0.01%
England	ET (317), PV (165), MF (59), pre-MF (43), MPN-u (25)	18 - 45* (33.4 - 35.9) *17 patients <18 years	ET: JAK2 45%, CALR 30%, MPL 3%, TN: 8% PV: JAK2 90%, TN 2% MF: JAK2 34%, CALR 42%, MPL 3%, TN 8%	ET: 10% PV: 16% MF: 7%	ET: 22% PV: 28%	ET: 4% PV: 7% MF: 15%

Abbreviations: MPN: myeloproliferative neoplasm, ET: essential thrombocythemia, PV: polycythemia vera, MF: myelofibrosis, MPN-u: MPN unclassifiable, pre-MF: pre-fibrotic MF, post-MPN MF: post-myeloproliferative neoplasm myelofibrosis, post-MPN BP: post myeloproliferative neoplasm blast phase

Epidemiology

Reports are varied regarding the prevalence of MPN AYA, with some describing a smaller proportion of MPN cases of less than 10% (17, 18, 19, 20), and others suggesting that closer to 20% of MPN cases may be attributable to this population (15, 21, 22). A population-based study in Sweden demonstrates that patients between 15 and 39 years old have an incidence of MPN of 0.67 per 100,000 person-years (23). Out of the classical MPNs, ET is the most common to be diagnosed in the AYA population, with several cohorts of AYA MPN patients reporting ET to encompass between 51 and 72% of their population. PV is the second most common, followed by PMF which comprises the minority of MPN AYA cases (9, 10, 11, 12, 13, 16). Paired with the fact that ET is the most commonly diagnosed MPN in AYA, and has a female predominance in general, MPN AYA cohorts are mostly female (65 – 71%) compared to older MPN cohorts (46 – 58%) (9, 10). The median age of diagnosis in several AYA MPN cohorts ranged from 20 to 36 years old (9, 13, 14, 16).

Clinical Characteristics

Similar to older patients with MPN, many AYA patients are asymptomatic at the time of diagnosis. Half of the patients in a cohort of 120 PV patients younger than 40 years old were incidentally diagnosed (14). Others, however, may demonstrate symptoms prompting a workup leading to their diagnosis. Sobas and colleagues reported that more than one third of their AYA MPN patients (diagnosed before 25 years old) had viscosity-symptoms at the time of diagnosis. Although symptom reporting by MPN subtype is limited, patients with PV may be more symptomatic compared with ET, with 9.2% of patients with PV reporting constitutional symptoms at diagnosis compared to only 3.9% of ET patients (13). One group reported higher rates of splenomegaly in their AYA PV and ET patients (age less than or equal to 40 years) when compared to patients older than 60 years old (PV 57% vs 21%; ET 28% vs 13%) (11). This is notable as some studies have identified splenomegaly as a risk factor for complications including transformation, haemorrhage (13), and thrombosis (12) in AYA patients. Although the potentially higher rates of splenomegaly in the AYA population has not been consistently demonstrated in other cohorts (10), this does warrant further attention. Bleeding seems to be a rare complication prior to or at diagnosis for AYA patients, although admittedly there is less reporting of such events in the literature. Less than 2% of one MPN AYA cohort had bleeding events at or before diagnosis (10). On the other hand, thrombotic events at or prior to diagnosis range from 14 to 27% in PV, 6 to 13% in ET, and 7 to 16% in PMF (10, 11, 13, 14, 16).

Regarding baseline hematologic parameters, several groups reported that their AYA PV and ET patients demonstrated higher platelet (plt) counts compared to older patients, however haemoglobin (hgb) levels were comparable (10, 11, 14). These higher plt counts may be due to the increased incidence of CALR positivity in AYA, given this mutation’s association with elevated plt levels (24, 25). For patients with PMF, Boddu and colleagues reported that their AYA patients (age 16 – 39 years) tended to be less anemic and have a lower tendency for leukocytosis compared to older adults (10). Szuber et al reported similar findings for their AYA PMF patients, and additionally found that their young patients demonstrated higher median plt counts (11). The higher blood cell parameters that younger MPN patients demonstrate may at least be partially explained by the age-related decline in hematopoiesis (26) that these patients may not yet have experienced, however differing molecular patterns may contribute.

Molecular Profile

Compared to pediatric (typically defined as age 0 – 18 years) (27) MPN patients, the molecular profile of AYA patients are more similar to that of older patients (28). However, there are still notable differences, particularly when it comes to the incidence of the CALR mutation and triple negative (TN) phenotype. Szuber et al described that the CALR mutation was more common in their AYA ET population compared to older adults (35% vs 19%); this finding was also demonstrated in their MF population (AYA 48% vs older 13%) (11). Harris and colleagues also reported that in their MPN AYA population of patients less than 40 years old, CALR exon 9 mutations were more common in AYA versus older adults (20.5% vs 10.5%) (9). The TN phenotype, defined as the lack of JAK2, CALR, MPL mutations, may also be seen in the AYA population, although the incidence is typically lower compared to pediatric MPN, where rates can be as high as 73% (29). Harris et al reports that triple negativity was more common in AYA compared to older adults (13.3% vs 4.5%), although the authors did note that most TN MPN patients were in their pediatric group of patients (9). This finding was consistent with a second study attributing the highest proportion of their TN AYA cases to children (13). Similar to what is seen in older adults (30, 31), JAK2 remains the primary driver mutation in AYA patients, with a reported incidence of either JAK2V617F or JAK2 exon 12 in more than 90% of patients (9, 10, 11, 12, 13, 14, 16) with PV and 40 to 75% in ET (9, 10, 12, 13, 15, 16). Several studies report lower JAK2 variable allele frequency (VAF) in AYA patients with PMF or PV compared older patients (10, 14).

Thrombosis and Bleeding

Thrombotic events can occur at high rates during follow up for AYA patients with MPN. One cohort of AYA patients actually experienced higher rates of venous thrombosis compared older adults (AYA 19.8% vs older 10.7%). The authors felt that this was due to the higher number of splanchnic vascular events in younger patients (9). Barzilai and colleagues report that of the venous thrombotic events that occurred during follow up in their MPN patients aged 18 to 45 years old, 77% of those were attributable to SVT (12). A high incidence of SVT was similarly reported by Stein et al in their cohort of PV patients, with SVT rates of 13% in AYA patients (less than or equal to 45 years old) compared to 2% in older patients (14). Arterial thrombotic events occur less often in AYA compared to older adults, with an incidence of 13.5% in MPN AYA versus 16.6% in older adults (9). Similarly, another study described lower rates of arterial thromboses at the time of PV diagnosis and during follow up for ET and PMF in the AYA population compared to their older counterparts (11). This may be due to a lower incidence of comorbid cardiovascular disease in AYA patients (32), however, there may also be inherently different disease biology at play.

Despite these varied patterns of thrombotic events in AYA patients compared to older patients, Boddu and colleagues report that their MPN AYA cohort did not differ in terms of thrombosis free survival (TFS) when compared to their non AYA MPN patients (10). This is noteworthy given the burden that thrombosis can impart in a young patient. Given the high morbidity that may be associated with thrombosis, attention should be made to identify which patients may be at the highest risk for such events. In the study of 109 AYA MPN patients by Barzilai et al, male sex, hgb > 14 g/dL, JAK2 mutation, and spleen size > 12 cm were predictive of thrombosis in univariable analysis (12). The JAK2 mutation was similarly predictive of thrombosis in the cohort of 444 young MPN patients by Sobas et al (13). These findings are perhaps not surprising given the recognition of the JAK2 mutation portending adverse thrombotic risk in older adult MPN patients (2, 33). Reports of bleeding in MPN AYA during follow up are less common, however several studies describe rates between 0% and 13% (10, 12, 15).

Disease Progression and Survival

Harris and colleagues demonstrated similar rates of post-ET PV between AYA and older adults (20% vs 26% respectively); this progression occurred over similar lengths of time in both groups (median 10 years in AYA vs 8 years in older) (9). AYA PV patients experienced transformation to MF in 0 to 24% of cases, whereas progression to post-ET MF occurred in 0 to 16% of patients based upon several studies (9, 10, 11, 13, 14). The high rates of fibrotic transformation that was one cohort demonstrated (ET 16% and PV 22%) was thought to be due to these patients’ extended survival (11). The median length of time to progression to MF was prolonged in AYA patients for both post-ET MF (AYA 22 years vs older 12 years) and post-PV-MF (AYA 24 years vs older 10 years) in the cohort by Harris and colleagues (9). Progression to AML occurs less frequently than the development of post-ET/PV MF. However, the development of post-MPN AML still occurs at similar rates to what is seen in the adult population, with an incidence of 0 to 4% for PV, 1 to 4% for ET, and 0 to 10% for PMF (9, 10, 11, 14). Risk factors for transformation have also been evaluated, with the progression to post-PV MF influenced by the presence of an abnormal karyotype, and leukocytosis contributing to the development of post-ET AML. Additionally, the presence of constitutional symptoms and the ASLX1 mutation independently predicted for leukemia-free survival in patients with PMF (11).

Despite the fact that disease progression remains a significant risk in AYA patients, overall survival (OS) appears to be superior when compared to non-AYA (9, 10, 11). Median OS, defined as time from the date of diagnosis to date of death or last contact, was 35 years for ET, 37 years for PV, and 20 years for PMF AYA patients in one group which followed patients over a median follow up of 7 to 13 years (11). This prolonged survival that MPN AYA patients experience is perhaps not surprising, although should be highlighted by the fact that the most common causes of death in several studies were as a result of the underlying MPN, transformation to AML, progressive MF or because of transplant-related complications (9, 13, 16). Nine percent of 609 AYA MPN patients in a Canadian cohort died during the median 9 years of follow-up with the most common causes of death as the following: progression to accelerated/blast phase 29%, progression to MF 27%, complications of allogeneic stem cell transplant (allo-SCT) 14%, infection 13%, and bleeding 7% (16). And while AYA patients generally experience prolonged survival, these data further support the crucial need for pursuing disease modifying agents in MPN.

Management

A summary of treatment recommendations for AYA PV, ET, and PMF may be found in Table 2.

Table 2.

Proposed Treatment Strategies for MPN AYA Patients

Risk Category	PV	ET	MF
AYA Low-Risk PV: no thrombotic event ET: no thrombotic event MF: adequate hematologic parameters, no high risk molecular mutations, normal karyotype	Hct control <45% Daily low-dose aspirin Clinical trial if available	Daily low-dose aspirin in select patients (JAK2 mutated) Clinical trial if available	Observe Clinical trial enrollment or earlier use of JAKi if constitutional symptoms or cytopenias
AYA High-Risk PV: thrombotic event ET: thrombotic event MF: presence of cytopenias, high risk molecular mutations, or abnormal karyotype	Hct control <45% Daily low-dose aspirin Interferon*	Daily low-dose aspirin Interferon*	Allo-SCT

^*Consideration to administer in low risk patients for possible disease modifying benefits

Abbreviations: ET: essential thrombocythemia, PV: polycythemia vera, MF: myelofibrosis, Hct: haematocrit, JAKi: JAK inhibitor, allo-SCT: allogeneic stem cell transplant

Polycythemia Vera (PV)

The European Leukemia Net (ELN) provides guidelines for treatment of PV on the basis of recurrent thrombotic risk, where high-risk is defined as age greater than 60 years old or history of a thrombotic event, and low-risk defined as the lack of both risk factors. Therefore, unless they have had a prior thrombotic event, AYA patients with PV will fall into the low-risk category per the ELN guidelines (2). Management of low-risk patients with PV entails haematocrit (Hct) control to less than 45% and low dose aspirin (2, 34). As with all patients with MPN, attention should be made regarding the impact that phlebotomy may have on inducing iron deficiency (28). Prior to initiating antiplatelet treatment, testing for acquired von Willebrand syndrome (aVWS) may be considered to identify patients at higher risk for bleeding complications, keeping in mind that aVWS may develop even in the absence of extreme thrombocytosis (35, 36). Although the theoretical risk of increased risk of bleeding while on aspirin in patients with aVWS exists, it is worth noting that there, to our knowledge, has not been recent data to support this.

If an AYA patient with PV is high-risk due to a prior thrombosis, treatment should include low dose aspirin, Hct control, and cytoreduction per ELN (2). While hydroxyurea (HU) is standardly used in older patients with MPNs, its use is cautioned against by both the ELN and NCCN guidelines in younger individuals. If an AYA patient with PV requires cytoreduction, guidelines recommend recombinant or pegylated interferon alpha (INF-α) as first line therapy (2, 37). For high-risk PV, IFN-α in the form of either peginterferon alfa-2a or ropeginterferon alfa-2b-njft are considered to be first line agents (34). Notably, the use of INF-α has been associated with achievement of molecular response (38, 39, 40, 41, 42, 43, 44, 45) and improvement in bone marrow fibrosis (44, 46). In a retrospective study of 470 patients with PV by Abu-Zeinah, INF-α improved myelofibrosis free survival (MFS) and OS, with the MFS benefit seen even in low-risk patients (47). Results from a recently described cohort of 348 AYA ET and PV patients diagnosed before 25 years of age by Beauverd et al demonstrated that cytoreduction with INF-α improved 10 and 20 year MFS compared to HU, anagrelide, or watchful waiting. This study was the largest contemporary AYA cohort focusing on the impact of cytoreduction, and given these findings, early consideration of a disease modifying agent such as INF-α is warranted given the long length of disease duration these patients will likely experience (48). Although the relationship between allele burden and clinical outcomes in MPN is not fully elucidated, several studies have demonstrated a reduction in thrombosis or disease progression in patients with lower JAK2V167F VAF (49, 50, 51). And as INF-α has demonstrated the ability to reduce allele burden in MPN (40, 52, 53), VAF monitoring may be a useful marker for clinical response in the future, however this is neither routinely done nor proven to be a validated end point.

In the event that first line cytoreductive treatment with IFN-α is either not tolerated or effective in maintaining counts or symptoms, then ruxolitinib may be considered (54). Ruxoliltinib, when compared to standard therapy in patients intolerant or resistant to hydroxyurea, has been shown to improve Hct control, spleen size, symptoms (55), and also has demonstrated superior event-free survival (EFS; defined by major thrombosis, hemorrhage, transformation, or death), progression-free survival (PFS) and OS in patients achieving a molecular response in the MAJIC-PV study (56). A potential limitation to utilizing ruxolitinib as second line for AYA PV is that ruxolitinib has been evaluated in hydroxyurea intolerant patients, and many AYA patients may have had IFN-α as their first line agent. Therefore, the use of ruxolitinib as second line for AYA is extrapolated to this setting, however may serve as an effective alternative for those who may not tolerate IFN-α or have high symptom burden.

Essential Thrombocythemia (ET)

Management for AYA patients with ET is based on risk stratification guidelines such as the revised International Prognostic Score for Thrombosis in ET (IPSET-thrombosis). If they have had no prior thrombosis, most AYA patients would fall into the very low or low-risk subgroups based on age less than 60, with the groups differentiated by the presence or absence of the JAK2V617F mutation (33). Although there have been no prospective studies specifically evaluating aspirin in low-risk ET patients, retrospective studies demonstrated a reduction in thrombotic events particularly in JAK2 positive patients when utilized (57, 58). The open-label randomized PT1 trial comparing HU plus aspirin versus aspirin alone in ET patients 40 to 59 years of age without thrombotic risk factors or extreme thrombocytosis demonstrated an overall low incidence of vascular events in both groups (0.93 per 100 patient-years), thus informing our knowledge of the use of aspirin in an intermediate risk group (59). Some advocate for twice daily aspirin for certain patients (JAK2, cardiovascular risk factors present) due to the increased plt turnover and compromised inhibition of plt cyclo-oxygenase that may occur in patients with ET (3, 60). And in contrast, for TN young patients with no prior thrombotic events or cardiovascular disease, some omit aspirin altogether given the very low thrombotic risk that these patients demonstrate (61).

CALR-ET patients may demonstrate a higher risk of bleeding when on low dose aspirin compared to those with a JAK2 mutation. In a retrospective study of 433 patients with low-risk ET, the incidence of major bleeding in CALR-mutated patients (n = 271, age 12 – 59 years) was higher while on aspirin versus observation (12.9 vs 1.8 bleeding events per 100 person-years, p = 0.03); antiplatelet therapy did not affect thrombotic risk (58). Thus, given the increased prevalence of CALR in AYA patients, the use of aspirin should be carefully considered. Regarding cytoreductive therapy for AYA ET patients, if it is needed, INF-α (specifically, peginterferon alfa-2a) should be administered as first line therapy (2, 37). However, while there may be a particular consideration for initiating therapy in those with CALR-mutated ET given the potentially increased risk of progression to post-ET MF that they demonstrate (48), it should be noted that molecular response may not be as robust in CALR-mutated patients receiving IFN-α compared to JAK2-mutated patients. A retrospective analysis of 38 patients with either JAK2 or CALR-mutated MPN receiving IFN-α demonstrated that JAK2 patients experienced a greater reduction in VAF compared to CALR patients, although there were no significant differences in haematological responses (62). Therefore, while CALR-mutated patients may achieve a haematologic response with IFN-α, its effects on inducing a molecular response is less clear.

For second or third-line treatment of ET, anagrelide is sometimes utilized in older adults who are intolerant or unresponsive to hydroxyurea or IFN-α (2). However, the use of anagrelide has been associated with the development of fibrotic progression (63, 64), particularly in CALR-mutated patients (65, 66, 67), thus if utilized in AYA patients, should be administered with caution. An additional area to consider when caring for these patients is regarding ideal plt goal, given the potential for CALR-driven thrombocytosis. As multiple studies have not demonstrated a correlation between the degree of thrombocytosis with thrombosis (68, 69, 70, 71, 72), we do not generally target a specific plt count with the goal of thrombosis reduction unless the patient is symptomatic or at increased risk of bleeding.

Regarding the previously mentioned cytoreductive agents for ET and PV, it is important to consider that there may be differential impacts on thrombotic risk reduction. While HU is not standardly recommended for AYA patients, this agent does show a particular impact on the reduction of recurrent arterial thrombosis in patients with PV and ET (73, 74). As AYA patients have less arterial events compared to older patients, the effect of HU may be less impactful in this population. Specific data is more limited in terms of the impact on arterial versus venous thrombosis reduction with IFN-α, however multiple studies have demonstrated low rates of thrombotic events in patients with PV or ET treated with IFN-α (between 0.42 and 1.2 per 100-person years) (75, 76). The MAJIC-PV study demonstrated ruxolitinib's ability to significantly improve thromboembolic-EFS compared to best alternative available therapy (56). Overall, the use of surrogate endpoints such as haematologic response as a marker of thrombosis may lead to difficulty in ascertaining true therapeutic effects of these agents (77), thus further elucidation of the differential impact of cytoreductive agents are needed to determine the optimal thrombotic reducing agent for the AYA patient.

Myelofibrosis (PMF)

As with ET and PV, prognostic models may be utilized to guide risk-informed management of patients with PMF. These models delineate patients into high, intermediate or low-risk based on factors including age, karyotype, the presence of high-risk mutations, clinical symptoms, and haematologic parameters (3). The Dynamic International Prognostic Scoring System (DIPSS) and Mutation-Enhanced International Prognostic Score System (MIPSS) are commonly used, of which the latter incorporates high-risk molecular mutations (ASXL1, EZH2, SRSF2, and IDH1/2) into risk stratification (78). The incidence of high-risk molecular mutations may be lower in younger patients, with one group reporting an incidence of 19% ASXL1, 6% SRSF2, and 3% for U2AF1 in their MF patients younger than 40 years of age (comparatively, for patients older than 60 years: 45% ASXL1, 18% SRSF2 and 19% U2AF1) (11). Nonetheless, the presence of one of these adverse factors may place an asymptomatic AYA patient with mild cytopenias into a higher risk MIPSS category (and subsequently inferior OS), thus emphasizing the importance of obtaining molecular testing for these patients.

Due to age, many AYA patients with MF fall into the lowest risk groups if they lack the previously mentioned high risk mutations or cytopenias. If asymptomatic, these patients may be observed (3). However, if constitutional symptoms or aberrant blood cell counts are present, clinical trial enrollment should be prioritized; otherwise one can consider any of presently available therapies. These include JAK inhibitors (JAKi) which aim to improve symptoms and spleen size, HU if elevated blood counts are present, or anemia-directed therapies (3). Currently available JAKi include ruxolitinib (79, 80), fedratinib (81), pacritinib (82), and momelotinib (83), of which cytopenias or the presence of splenomegaly guide the section of a particular agent. Although JAKi therapy can improve symptom burden and spleen size, they have not yet been shown to have clear disease modifying capabilities (84), which is a crucial unmet need. There is limited data on the use of JAKi in AYA, however in a cohort of MF patients aged 18 through 50 years old, ruxolitinib was the second most commonly utilized agent, which on average was initiated after a median of 27.2 months and continued for approximately 44 months (85). Further data surrounding the clinical characteristics and outcomes of JAKi therapy in AYA patients is an area which requires further research.

In DIPSS intermediate-2 and high-risk MF patients, consensus recommendations advise proceeding with allo-SCT given the improved survival over standard treatment (86, 87) and the existence of this as the only curative option (88). Earlier allo-SCT may lead to superior transplant outcomes (89), however, it is important to note that there is a risk of early transplant-related mortality and morbidity. In a cohort of 609 MPN AYA patients (≤ 45 years old) in Canada, of the 56 patients underwent allo-SCT for either MF or post-MPN AML, 14% of patients died from complications of transplant (16). Furthermore, after undergoing allo-SCT, patients carry the risk of developing chronic graft versus host disease (GvHD) for many years, and thus the survival benefit that allo-SCT confers must be carefully weighed against these factors, especially in an AYA patient who has the potential for significant longevity (90).

Novel approaches for the treatment of PMF include combination therapies, CALR-directed agents, and type II JAKi. Combination approaches of JAKi with BCL-X_L inhibition in the form of navitoclax or the BET inhibitor pelabresib, demonstrate possible disease modifying capabilities (91, 92, 93, 94, 95, 96), thus serving as promising options. Additionally, efforts have been made to identify type II JAK2i, which may demonstrate enhanced efficacy through greater reduction in mutant clones compared to type I JAKi (97). Novel therapeutics targeting the CALR mutation are also of great interest. INCA033989 is a monoclonal antibody with high affinity for the mutant CALR protein (CALRmut), with data demonstrating that targeting CALRmut leads to normalization of thrombopoietin receptor (TPO-R) signaling in vivo (98); this is currently being investigated in a phase I trial (NCT06034002). Additionally, a first in class T-cell redirecting bispecific antibody targeting CALRmut, JNJ-88549968, has demonstrated promising activity in mouse models (71) and is under active exploration (NCT06150157). Therapeutics targeting CALR are of great interest for the treatment of PMF, and particularly for AYA patients given their increased incidence with this mutation.

Pregnancy Considerations

With the complexity of care that is involved in MPN pregnancies, it is crucial that patients are under the monitoring of a haematologist and high-risk maternal fetal medicine physician. The most common MPN in women of childbearing age is ET (99). A UK based prospective study demonstrated an incidence of MPN pregnancies of 3.2 per 100,000 pregnancies (100). Patients with MPN may have higher rates of pregnancy complications, including preterm delivery and small-for-gestational-age infants (100). Compared to an 80% life birth rate in the general United States population (101), a study of pooled outcomes of 461 pregnancies in ET patients (median age at pregnancy 29 years old) demonstrated that only 50 to 70% of women experienced successful live births. Women in this study experienced first-trimester losses in 25 to 40% of pregnancies, with late pregnancy loss occurring in 10% of cases (102). Gangat and colleagues synthesized results from 4 studies to produce live birth, maternal thrombosis, and haemorrhage rates of 70.2%, 2.5%, and 4.5% respectively in ET patients (99). A retrospective study of 121 pregnancies in ET by How et al revealed that prior pregnancy loss was associated with increased pregnancy complications in ET, a finding that had not been previously demonstrated in the MPN population (103).

The JAK2V617 mutation may have an impact on pregnancy outcomes, although reports have been inconsistent. In two cohorts of women with ET, those who were JAK2 positive experienced more pregnancy complications and fetal losses compared to those who did not carry the mutation (104, 105). Other studies, however, suggest that the JAK2 mutation does not affect pregnancy outcomes (103, 106, 107). Rumi and colleagues suggest that the presence of the CALR mutation may even be associated with improved pregnancy outcomes (107). Several other factors have been associated with elevated risk in an MPN pregnancy, including extreme thrombocytosis with platelets greater than 1500 x 10⁹/L, history of thrombosis or bleeding, or previous pregnancy complication (108, 109). In line with the natural decrease in plt count that occurs during pregnancy, Hobbs and colleagues described a decrease of almost 43% in plt count from baseline to delivery in their cohort of ET patients, and furthermore, found that a greater decrease may reduce the risk of developing a pregnancy complication (103).

Figure 1 depicts a summary of treatment considerations during pregnancy for MPN AYA patients.

Figure 1. Treatment Considerations During Pregnancy for MPN AYA Patients.

Abbreviations: ASA: aspirin; LMWH: low-molecular weight heparin; IFN: interferon

Commonly utilized agents during pregnancies in MPNs are aspirin, heparin, and INF-α. As aspirin has been endorsed by the US Preventative Services Task Force (USPFTF) to prevent preeclampsia in high-risk pregnancies in the general population (110), its use has subsequently been explored in MPN patients. A meta-analysis by Maze et al of 22 studies reporting on 1210 pregnancies in MPNs demonstrated that aspirin use during pregnancy was associated with an improved live birth rate (odds ratio, OR = 8.6), although it did not affect thrombotic outcomes (111). When using aspirin during pregnancy, is crucial to assess for aVWS to mitigate any added risk of bleeding, with experts suggesting testing VWF parameters early in pregnancy and again in the third trimester (108). Overall, the use of low dose aspirin in the absence of increased risk of bleeding is generally recommended for MPN patients during pregnancy (108).

Low molecular weight heparin (LMWH) is often employed prophylactically during high-risk pregnancies, possibly due to its proposed anti-inflammatory and antiangiogenic effects (112). A meta-analysis evaluating the risk of venous thromboembolism (VTE) during ET pregnancies found that the risk of VTE during the antepartum period was 2.5%, however post-partum, that risk increased to 4.4% (113). Results from the meta-analysis by Maze et al did not reveal any improvement in live birth rates or thrombotic events when LMWH was utilized (111). A common practice is to employ close observation in the antepartum period, however in high-risk pregnancies with prior thrombosis, LMWH should be offered, and consideration for administration should be made for patients with a previous poor pregnancy outcome. Given the increased risk of thrombosis in the post-partum period, LMWH during this time should likely be offered for all patients barring any high risk of bleeding (108).

Cytoreductive agents have also been explored during pregnancy in MPN. A case series on the use of INF-α during pregnancy in ET reported no grade 3/4 adverse events or drug discontinuation. Furthermore, there were no major bleeding or thrombotic events during pregnancy, delivery, or in the post-partum period in patients receiving INF-α (114). A meta-analysis by Maze et al demonstrated that the use of INF-α in MPN pregnancies resulted in an improved live birth rate (OR 9.7), and this benefit was demonstrated with and without concurrent aspirin or heparin (both unfractionated and LMWH) (111). Overall, in patients who require cytoreduction based on risk stratification, INF-α (specifically peginterferon alfa-2a) is the preferred agent to administer during pregnancy. Additionally, consideration to the use of INF-α should be made for patients who are traditionally low-risk however had had a prior pregnancy-related event (115).

Regarding MPN medication effects on fertility, the majority of data surrounds the use of hydroxyurea, from which data has largely been derived from the sickle cell anemia population. It has been demonstrated that hydroxyurea may significantly decrease sperm count in men, however this usually reversible when the drug is discontinued (116). While hydroxyurea is not preferred for MPN patients, this should be kept in mind if prescribed. There is a lack of clinical studies evaluating the effect of ruxolitinib on fertility or safety during pregnancy, and therefore is generally avoided. And while we have previously discussed that IFN- α appears to be safe in pregnancy and is the preferred agent in the event cytoreduction is needed, its effects on fertility are unknown (117).

The use and safety of OCPs in females with MPN is another area of great importance. A retrospective analysis to determine the existence of any association between the use of estrogen-based hormonal therapy (EBHT) in the form of either hormonal replacement or OCP and thrombosis was carried out in 305 women with ET. Gangat et al found an increased risk of venous thrombosis in women receiving estrogen-containing OCP compared to those not on OCP (23% vs 7%, respectively) (118). In light of this, non-estrogen based forms of contraception may be preferred for females with MPN (99).

Lastly, pertaining to hct goal, the ideal target during pregnancy in PV may differ from the traditional goal of less than 45% for the general MPN population. During pregnancy, there is a greater expansion in plasma volume compared to red blood cell volume, therefore predisposing one to lower hct. In MPN pregnancies, instead of maintaining a goal of <45%, it may be more appropriate to target a lower hct goal that would be physiologic during pregnancy (108, 115).

Quality of Life

Understanding symptom burden is critical to the management of MPN patients, and accordingly, efforts have been made to explore the impact of symptoms on quality life in the general MPN population (119, 120, 121, 122). The MPN Landmark survey demonstrated that 60% of low-risk ET and PV patients reported adverse effects on quality of life, with one-quarter of patients missing a day of work in the past month due to symptoms (120). A parallel that could be made to the AYA population involves the impact that their disease may have not only on missed work days, but also on missing school or college.

A multicenter cross-sectional study in China compared patient reported outcomes (PROs) in MPN AYA compared to older adults through utilization of the Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score (MPN-SAF TSS, or MPN-10). The MPN-10 is an MPN specific questionnaire used to assess symptoms including fatigue, early satiety, and pain, where higher scores indicate more severe symptoms (123). This study demonstrated that although AYA patients reported lower MPN-10 scores, they endorsed a higher proportion of adverse effects of their disease and therapy on daily life. They also reported more concerns with fertility, drug advances/prices, and treatment efficacy and adverse effects. As part of the short-form health survey, AYA patients reported the lowest mental health score out of all age groups (124). These results suggest that young patients with MPN experience disease burden that may differ from older patients. Management of MPNs should prioritize patient’s concerns when possible, which may encompass modification of activities of daily living in order to promote optimal work-life balance (125). A further contributor to these psycho-social burdens is drug cost. Similar notions have been made surrounding the cost of tyrosine kinase inhibitors for chronic myeloid leukemia (126), which similar to MPNs, often necessitates extended treatment. This potentially lifelong cost of treatment for MPN AYA patients is especially relevant in the advent of novel therapies. The impact of financial pressures and mental health considerations should not be underscored in this population.

Social Media

As many AYA patients grew up in the era of social media, they may turn to platforms such as X and Facebook for information and community support. Recently, there has been interest in exploring the role of social media in AYA patients with cancer (127). The disease-specific hashtag, #mpnsm (Myeloproliferative Neoplasms on Social Media) by Naveen Pemmaraju, MD (@doctorpemm) has served as a venue to improve engagement among patients, clinicians, and organizations (128). Utilization of online MPN-specific questionnaires can not only highlight gaps in patients’ disease knowledge, but also determine differences in social media habits between patients and physicians, which may help guide personalized educational campaigns (129). These avenues function as important modalities for AYA patients to engage with physicians, researchers, and one another, which is particularly important when navigating a rare disease that is less often promoted in the media. The utilization of social media serves as a platform to promote clinical trial enrollment in a way that may reach younger patients more so than the traditional routes of engagement.

Figure 2 represents a suggested agenda for the care of AYA patients with MPN.

Figure 2. Agenda for Next Steps in the Care of MPN AYA Patients.

Conclusions

AYA patients comprise a distinct population within MPNs, displaying key features which set them apart from older adults. In particular, there is a higher incidence of the CALR mutation, with the ensuing implications that this may have, including the potential for more extreme thrombocytosis or bleeding events. As a result of their younger age at diagnosis and therefore longer duration of disease, they remain at high risk of thrombotic events, particularly splenic venous thrombosis, as well as transformation to more aggressive disease. Attention must be made to identify efficacious, well-tolerated, and risk-adapted strategies. The role of cytoreduction in AYA patients with high symptom burden or uncontrolled counts is an area that requires further research. As disease modification becomes a possibility, it is imperative that AYA patients are enrolled in clinical trials and considered for such therapeutic approaches. Symptom burden and disease related effects on mental health remain hugely important for these patients, as there is the potential for significant impacts on quality of life. Efforts should be made toward establishing risk stratification guidelines centered around AYA patients to enable uniform practices across institutions.

Data Availability Statement:

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author