Clinical practice recommendations for myelofibrosis management in Latin America from an expert consensus group Open Access

Myelofibrosis (MF) is a clonal myeloproliferative neoplasm (MPN) characterized by bone marrow fibrosis, extramedullary hematopoiesis, splenomegaly, cytopenias, and constitutional symptoms.¹ Primary MF is a rare disease, with an estimated incidence of ∼0.47 cases per 100 000 people. However, this was estimated in 2014 using studies from North America, Europe, and Australasia; as such, the true incidence in the Latin America (LATAM) region is not known.² Due to the lack of data, MF outcomes across the LATAM region cannot be accurately estimated. However, the Colombian Registry of MPNs recently reported in 2017 that MF characteristics in Colombia may differ from those reported in other series, highlighting the importance of local data collection.³ 

Health care professionals face a number of challenges that affect MF care in the LATAM region, such as limited availability of therapies, and difficulty in accurate diagnosis due to a lack of access to driver mutation tests at many treatment centers.⁴ Furthermore, disparities in drug availability, particularly the lack of access to newer Janus kinase inhibitors (JAKis) contribute to suboptimal survival and symptom benefit for patients with MF in the LATAM region.⁴ An international consensus group recently published global recommendations, based on a global systematic literature review (SLR) and clinical experience, for MF management in routine clinical practice to help health care professionals overcome challenges and optimize care for patients with MF around the world.⁵ However, there remains a lack of unified clinical recommendations tailored to the differences in standard practice, drug availability, and nuances in patient needs across the LATAM region. In response, an expert panel of 8 LATAM hematologists conducted a consensus program to address these LATAM-specific unmet needs.

The program focused on 5 key themes:

• Defining the thresholds for anemia and when to initiate or modify treatment.

• Defining when to initiate or modify treatment for thrombocytopenia.

• JAKi therapy: selection, initiation, and switching.

• Defining the most appropriate risk stratification model for MF in the LATAM region.

• Defining allogeneic hematopoietic stem cell transplantation (allo-HSCT) eligibility in the LATAM region.

The aim was to propose practical, context-specific solutions to improve outcomes for patients with MF in the LATAM region. Although these recommendations aim to guide physicians’ approaches to MF in the context of resource constraint and access to newer therapies, expanding access to these treatments remains crucial for making a substantial impact on outcomes of patients with MF.

The authors, 8 expert hematologists from Argentina, Colombia, and Brazil, comprised the steering committee (SC) for this program. An extended faculty (EF; N = 20 [hematologists, n = 19; transplant specialist, n = 1]) was invited to participate in the voting rounds, with representatives from Argentina, Brazil, Chile, Colombia, Costa Rica, Paraguay, and Uruguay.

The SC drafted and refined 15 consensus questions across 5 themes. A LATAM-specific SLR was then conducted to provide an evidence base (Figure 1), which was combined with the SC’s clinical experience to draft the recommendations. Alongside the LATAM-specific SLR, sources identified in the global SLR that was conducted to support the global consensus by Koschmieder et al⁵ were also utilized as additional evidence for this LATAM-specific consensus program. The EF voted on the recommendations using an online voting platform; a 9-point scale was used, where 1 indicates strongly disagree and 9 indicates strongly agree. If an EF member scored a recommendation of ≤6, they were able to provide feedback for the SC’s consideration. Consensus was achieved when 75% of the voters agreed in the range of 7 to 9.

All data supporting the findings of this study are included in the article and/or its supplemental Materials.

All 15 recommendations achieved a high level of consensus after the first voting round, with 14 recommendations reaching 96% to 100% consensus and 1 recommendation achieving 92% consensus (Tables 1-5). Following EF feedback, the SC agreed that 1 recommendation (please see Question 14 in Table 5), which achieved 96% consensus after the first round of voting, would benefit from further refinement. The recommendation under Question 14 was revised and resubmitted for a second voting round, after which 100% consensus was achieved.

Please refer to Table 1 for recommendations under this theme.

Anemia is a common complication of MF and is associated with symptoms such as fatigue, dizziness, and weakness, which affect health-related quality of life.^6,7 Blood transfusions are frequently used to manage anemia in MF, although this is associated with iron overload, which may lead to organ damage.^8,9 Furthermore, some patients may refuse blood transfusion for personal reasons.¹⁰ Iron chelation therapy, such as deferasirox, can be used to mitigate iron overload. Although there are limited data on its safety and effectiveness in MF, the multicenter, retrospective RUX-IOL study demonstrated that 47.8% of patients with MF who were treated with ruxolitinib and deferasirox achieved an iron chelation response (ICR). The importance of achieving an ICR was shown in this study, in which an ICR at any time predicted improved survival.¹¹ Transfusion dependence may carry a significant economic burden for patients, caregivers and the health care system itself, and could be up to 1.8 times higher for transfusion-dependent patients compared with transfusion-independent patients.¹² The cost implications of this should therefore be considered when making treatment decisions.

Despite this, as allo-HSCT is the only curative option for MF, physicians should always consider whether a patient is a potential candidate, especially when initiating blood transfusions, as a high transfusion burden is associated with poor allo-HSCT outcomes. This may be due to increased sensitization to human leukocyte antigen (HLA), thereby increasing the risk of graft failure.¹³ However, it may be that transfusion burden is simply a surrogate for advanced disease or late diagnosis.¹³ 

• MF-related anemia?

• Treatment-related anemia?

• Patients not yet requiring transfusion?

The National Comprehensive Cancer Network (NCCN) guidelines advise the use of JAKis for the treatment of MF-related symptomatic splenomegaly and/or constitutional symptoms, even in patients with anemia.¹⁴ However, before administering JAKis to these patients, physicians should consider that patients may experience anemia that is not a direct result of MF itself, but rather from concomitant conditions, such as iron or B₁₂ deficiency, hemolysis, or inflammation from unrelated comorbidities.¹⁵ It is crucial that physicians thoroughly investigate other potential causes of anemia and correctly rule these out. Furthermore, European LeukemiaNet recommendations suggest that the choice of a specific drug should be based on overall toxicity and expected risk of this in the patient.¹⁶ 

The newer JAKi therapies, fedratinib, pacritinib, and momelotinib, are not yet widely available in the LATAM region, with ruxolitinib being the most commonly used JAKi in the LATAM region. However, even ruxolitinib remains inaccessible to some patients in LATAM. Although JAKis can be beneficial in controlling MF-related symptoms, they can also result in treatment-related anemia.^1,6 In particular, it is important to note that treatment with both fedratinib and ruxolitinib often lead to a temporary drop in hemoglobin (Hb) levels, with levels generally recovering after 24 weeks to a steady state for ruxolitinib,^17,18 and a partial recovery after 16 weeks for fedratinib.¹⁹ Momelotinib and pacritinib have the potential to improve MF-related anemia in patients with low Hb levels^20,21; however, as stated before, these options are not yet widely available in the LATAM region.

Alternative treatments, either alone or as an add-on therapy to ruxolitinib, include erythropoiesis-stimulating agent (ESAs), low-dose corticosteroids, thalidomide (with or without prednisolone), and danazol,^22-24 although these therapies are not widely reimbursed for anemia in patients with MF in the LATAM region. Splenectomy is rarely indicated as a viable option for MF-related anemia. When considering splenectomy, it should be cautiously evaluated and restricted to patients with severe refractory anemia and massive splenomegaly who demonstrate no response to pharmaceutical management, and no alternative treatment options are available, because of the high morbidity and mortality rate it confers.²⁵ 

The end point to measure overall response to JAKis is typically spleen volume reduction. Clinically, treatment success for anemia in patients with MF may be measured by improvements in Hb levels, lower transfusion dependency, splenic size reduction (as splenomegaly can be induced by extramedullary erythropoiesis, which prompts destruction of red blood cells [RBCs]), and overall symptomatic relief.¹⁵ Conversely, indicators of failure include persistent anemia despite treatment, persistent RBC transfusion requirement, increasing spleen size, or worsening of symptoms.

Limited access to newer treatments, such as pacritinib and momelotinib, in LATAM countries adversely affects anemia management in MF, particularly for subgroups of patients who are nonresponsive to the approved therapies. Pacritinib is a JAK2i, which demonstrated its effectiveness in the PERSIST-1 and PERSIST-2 trials in managing MF-related splenomegaly, and also reducing patients’ RBC transfusion requirement.^21,26 Momelotinib is a JAK1/2 and activin A receptor type 1 inhibitor, which improved anemia measures in the SIMPLIFY-1, SIMPLIFY-2, and MOMENTUM trials.^20,27 In the absence of pacritinib and momelotinib, patients receiving other JAKis may benefit from add-on treatments, including danazol, ESAs, immunomodulatory imide drugs, low-dose corticosteroids, and RBC transfusions to control anemia symptoms.

There are also broader cost and availability issues in LATAM; the high cost of newer therapies can be a significant barrier when health care resources and funding are limited. Expensive medications may not be covered by public health systems or insurance, making them inaccessible to many patients. Efforts to expand access to newer therapies, such as pacritinib and momelotinib, and reduce costs at a national level are essential for improving patient outcomes.

Please refer to Table 2 for recommendations on this theme.

For patients with platelet counts ≥50 × 10⁹/L, low-dose ruxolitinib, fedratinib, pacritinib, and momelotinib are all suitable options to reduce spleen size and symptoms. Both pacritinib and momelotinib have demonstrated efficacy for patients experiencing thrombocytopenia. Data supporting their use in these populations were obtained with pacritinib for patients with platelet counts <50 × 10⁹/L in the PERSIST trial and momelotinib for patients with platelet counts >25 × 10⁹/L in the MOMENTUM trial.^20,21 It should be noted that ruxolitinib starting dose is defined according to the patient’s baseline platelet count, as it can cause thrombocytopenia.^17,18 It is also important to consider that when reducing ruxolitinib dose, spleen volume reduction will likely be less pronounced, and this should be factored in when evaluating treatment response.

MF-related thrombocytopenia management is affected by the limited availability of pacritinib and momelotinib, similar to MF-related anemia. Becuase pacritinib and momelotinib show beneficial symptom control in patients with thrombocytopenia, in the future it will be critical for access to these treatments to be expanded in the LATAM region.

Splenectomy has the potential to increase platelet count, but this is associated with significant morbidity.²⁸ Therefore, splenectomy should only be considered in very specific cases, at the discretion of the treating physician where there are no alternative options available.

Please refer to Table 3 for recommendations on this theme.

NCCN guidelines and European LeukemiaNet recommendations advise initiating JAKi therapy based on symptom burden, splenomegaly, and risk category.^14,16 JAKi therapy is indicated for patients with intermediate-2 and high-risk MF when splenomegaly or symptoms need to be treated, and also in certain circumstances for lower risk, symptomatic patients with MF. Observation is suitable for most patients with asymptomatic disease.¹⁴ 

Our recommendation refers to multiple prognostic scoring tools, which may be used to assess patients with MF to guide treatment decisions. Notably, the Mutation and Karyotype-Enhanced International Prognostic Scoring System (MIPSS70+ version 2.0) helps identify high-risk patients who may benefit most from allo-HSCT. For patients already receiving JAKi therapies, disease course should be monitored to identify patients who may benefit from a treatment shift. Responses to ruxolitinib can be anticipated using the RR6 (response to ruxolitinib after 6 months) model (Figure 2), ensuring patients continue to benefit from this therapy, while guiding when to switch treatment strategy.²⁹ 

For patients switching from JAKi therapy, namely ruxolitinib, tapering JAKi treatment beforehand is important to minimize withdrawal syndrome. Patients discontinuing treatment should be carefully monitored, and any other therapy should be initiated as close to discontinuation as possible. Particularly for patients proceeding to allo-HSCT, a gradual tapering approach helps prevent rebound splenomegaly and symptom flare-up, while maintaining control of symptoms during conditioning for transplant.³⁰ 

Clear criteria are needed to determine when to adjust therapy, consider alternative treatments, or discontinue JAKi therapy altogether. Criteria used in clinical trials may not always align with real-world clinical settings, as they focus on strict definitions of spleen response within set timeframes. A more flexible and patient-centered definition of JAKi failure and suboptimal response is provided in our recommendation (Table 3).

In general, JAKi failure may be defined as clinical outcome not aligning with intention of treatment. For example, the spleen size and/or symptoms reductions after treatment with pacritinib or momelotinib may be defined as a treatment response, but if there was no anemia improvement, where this was the intention of treatment, the treating physician may consider this a failure. Similarly, although anemia improvement is desirable, a lack of anemia response may not constitute treatment failure if this was not the goal of treatment. It is essential to consider what the treatment goals are for each individual patient. Treatment goals differ from case to case, particularly because of the heterogeneous nature of MF and variations in symptom profiles. Common goals may include slowing/delaying disease progression, preserving quality of life and restoring healthy blood counts,³¹ and these should be taken into consideration when selecting therapies and evaluating response. Moreover, it is important to note that discordance in treatment goals between patients and physicians is not uncommon; therefore, fostering open patient–physician communication is key to improving patient outcomes and overall treatment satisfaction.³¹ 

Ongoing monitoring of spleen size, symptoms, and body weight is essential to determine treatment efficacy, identify disease progression, and guide therapeutic adjustment. The appearance of high-risk mutations (eg, ASXL1, SRSF2, IDH1/2, EZH2, and U2AF1) may also be monitored, particularly in patients who may benefit from earlier consideration of allo-HSCT,³² although there are currently no published recommendations or guidelines around monitoring these mutations. Next-generation sequencing (NGS) testing is not currently widely available for clinical practice in LATAM, which limits its implementation.

As previously mentioned, ruxolitinib and fedratinib can cause an initial drop in Hb levels that can be managed with dose reduction or interruption.^17-19 In patients with worsening anemia and/or thrombocytopenia, a switch to momelotinib or pacritinib (in regions where these are available) may be considered after 6 months of ruxolitinib treatment. However, it should be noted that although momelotinib and pacritinib are generally well tolerated in patients with thrombocytopenia, they can also lead to thrombocytopenia as a treatment-related side effect.^20,21 

To avoid switching JAKi therapy, or when other JAKis are not available, danazol, blood transfusions, immunomodulatory imide drugs, and ESAs can be used as add-on therapies in the LATAM region to control cytopenias and their concomitant symptoms. Although luspatercept is not widely used as an add-on therapy, a recent phase 2 trial showed that luspatercept improved anemia and reduced transfusion burden in patients treated with ruxolitinib.³³ However, data are not yet mature enough to draw significant efficacy conclusions, and luspatercept use is further limited by local regulatory approvals in the LATAM region.

• Restricted access?

• Unrestricted access?

As most LATAM countries are still only able to access ruxolitinib, our recommendation focuses on clinical trial enrollment and ruxolitinib rechallenge following ruxolitinib failure. For regions with unrestricted access, a switch to alternative JAKis, such as fedratinib, momelotinib, or pacritinib, should be considered for patients with severe anemia or thrombocytopenia.

For patients discontinuing ruxolitinib, physicians should be mindful of the potential for ruxolitinib discontinuation syndrome. To minimize this risk, vigilant surveillance of patients is required upon discontinuing ruxolitinib, which should be carefully tapered.³⁰ 

Please refer to Table 4 for recommendations on this theme.

• Restricted access to full molecular testing?

• Unrestricted access to full molecular testing?

The International Prognostic Scoring System and Dynamic International Prognostic Scoring System (DIPSS) models rely on clinical and hematologic parameters, such as age and blood counts, and are suitable for use in settings with limited diagnostic resources.³⁴ DIPSS Plus includes cytogenetic and transfusion status, improving DIPSS accuracy.³⁵ Models incorporating NGS, such as MIPSS70+ version 2.0 and Genetically Inspired International Prognostic Scoring System, incorporate mutational and cytogenetic data to improve prognostic accuracy. These models are suitable for regions with access to advanced molecular/cytogenic profiling, providing more tailored risk stratification.³⁴ For selective younger patients whose treatment decisions could be guided by molecular testing results, NGS should be performed if available.

The HCT-CI (Hematopoietic Cell Transplantation–specific Comorbidity Index), DRI (Disease Risk Index), and EBMT (European Society for Blood and Marrow Transplantation) risk scores offer complementary risk assessments that collectively enhance predictions of nonrelapse mortality, overall survival, and relapse risk for patients undergoing allo-HSCT.³⁶ 

The MTSS (Myelofibrosis Transplant Scoring System) accurately predicts transplant-related mortality and overall survival, guiding clinicians at transplant referral with a “Go,” “Slow Go,” or “No Go” decision based on patient factors (age, performance status, and HLA matching) and disease characteristics (eg, ASXL1 mutation or blood counts).³⁴ 

Please see Question 11 for elaboration on the prognostic scores to be used in this recommendation.

Prognostic scores that use clinical data, calculated using the risk stratification models outlined in Question 11 above, should be performed at each visit following the initiation of therapy. For patients who are experiencing progression or worsening of disease, models that also incorporate molecular and/or cytogenetic data, such as DIPSS Plus and/or MIPSS70+ version 2.0, may be considered.

Please refer to Table 5 for recommendations on this theme.

The NCCN Guidelines for Hematopoietic Cell Transplantation provide a comprehensive framework for identifying eligible candidates and integrating clinical, genetic, and psychosocial factors in the evaluation process.³⁷ Our recommendation focuses on key patient-related and disease-related clinical criteria. We also recommend that the specific capabilities of the treatment center are considered during the decision-making process.

The management of JAKi therapy in patients with MF considered for transplantation should be individualized, considering the disease state, response to JAKi therapy, and potential risks during the transplant process, to optimize timing for discontinuation and potential reintroduction after transplant. However, it should be acknowledged that continuing JAKi therapy after transplantation is not standard practice in all regions.

Effectively engaging patients in their own care is essential for shared decision-making between health care professionals and patients and will ultimately contribute to improving health care outcomes. Recognizing this, it was critical to ensure the inclusion of patients’ perspectives on these recommendations outside of a purely medical standpoint.

Patient representatives emphasized that the recommendations developed are key to providing patients with a full picture of the treatment options available to them, to allow greater collaboration with their health care team and engagement in making a fully informed treatment decision. Patients also highlighted the importance of patient advocacy groups, which are critical for providing educational support (such as through online resources, phone support, support for families and carers, and events connecting networks of patients) for patients to better understand their disease and potential treatment pathways, as well as advice on how they can play a more active role in their health care decisions.

Although international guidelines exist for managing MF, they do not fully address the unique challenges in LATAM. Variability in health care systems, economic conditions, limited resources, regulatory environments, and unequal access to advanced diagnostics and treatments across the region make it necessary to develop tailored, region-specific guidance. Significant challenges exist in accessing JAKis and other novel treatments due to regulatory delays and high drug costs, contributing to suboptimal disease management and health-related quality of life for patients with MF. Moving forward, more LATAM-specific clinical trials and research are essential to generate data relevant to the local population, which can guide future decision-making and improve outcomes.

This consensus initiative was launched to create context-specific recommendations that consider the region’s unique economic, health care, and regulatory considerations to ensure that treatment recommendations are both practical and effective. Prioritizing access to innovative therapies is key, with a focus on reducing health care disparities across the region and worldwide, and improving care for all patients, regardless of socioeconomic status. The high level of consensus achieved across all recommendations strengthens the validity of these LATAM region–specific recommendations for managing MF. In the future, an opportunity lies ahead to develop more widespread recommendations for the broader region by including expert perspectives from areas such as Central America, Mexico, and the Caribbean.

The authors thank the members of the extended faculty for their invaluable participation in the voting process: Claudia Agudelo, Carolina Barreto, Carlos Bermudez, Emiliano Carricondo, Miguel Castro Ríos, Diana Cuervo, Vaneusa Funke, Paula Heller, Amado Karduss, Beatriz Moiraghi, Katia Pagnano, Mariel Perez, Ana Inés Prado, Guillermo Quintero, Mariela Rodriguez, Christine Rojas, Carmen Rosales, Mariana Stevenazzi, Ximena Valladares, and Verónica Vallejo. Their expertise and thoughtful contributions were instrumental in shaping the consensus and ensuring the rigor and relevance of the recommendations. Medical writing support was provided by Lily Rowsell and Benedict Macintyre of Bedrock Healthcare Communications Ltd, United Kingdom, and was funded by GlaxoSmithKline. Assistance with the systematic literature review was provided by AccuScript and supported by GlaxoSmithKline.

This study was funded by GlaxoSmithKline.

GlaxoSmithKline had no involvement in any aspect of the design or execution of the program, maintaining the full independence of the steering committee.

Contribution: All authors contributed equally to the development of this consensus program, actively participated in the formulation, review, and revision of the manuscript, and approved the final version for submission.

Conflict-of-interest disclosure: A.I.V. has received honoraria from Novartis, Bristol Myers Squibb, and GlaxoSmithKline, as a speaker and in consultancy fees. C.S. has declared clinical research with Libbs, AbbVie, Novartis, and Johnson & Johnson; speaker fees from Libbs, AbbVie, Pint Pharma, Novartis, and BeiGene; and participated in advisory boards with Novartis, GlaxoSmithKline, and Pint Pharma. F.S. has received honoraria from Novartis, Bristol Myers Squibb, and GlaxoSmithKline as a speaker and in consultancy fees. F.P.S.S. has received speaker fees from Johnson & Johnson, Bristol Myers Squibb, Novartis, Merck, AbbVie, Roche, Amgen, Pfizer, Astellas, BeiGene, Servier, GlaxoSmithKline, and Teva; participated in advisory boards with Johnson & Johnson, Bristol Myers Squibb, Novartis, Merck, AbbVie, Roche, Amgen, Pfizer, Astellas, BeiGene, Servier, GlaxoSmithKline, AstraZeneca, Teva, and Knight Industries; and received research funding from AbbVie and Keros. J.A.O.I. has participated as a medical lecturer and adviser for AbbVie, Amgen, AstraZeneca, Johnson & Johnson, Novartis, Pfizer, Roche and Takeda; and is a member of Asociación Colombiana de Hematología y Oncología (Colombian Association of Hematology and Oncology), Grupo de Estudio Latinoamericano de Linfoproliferativos (Latin American Lymphoproliferative Study Group), American Society of Hematology, and Global Consortium of Lymphomas. M.d.L.Á.V.R. has received honoraria for participation in the advisory board and for the preparation of this manuscript from GlaxoSmithKline. R.T. has received an honorarium from GlaxoSmithKline for participation in this consensus program. V.A.P. has received an honorarium from GlaxoSmithKline for participation in this consensus program and from Novartis for advisory board meeting participation and consultancy fees.

Correspondence: Renato Tavares, Clinica Medica, Universidade Federal de Goiás, Alameda das Espatodias, Q44, L09, Residencial Aldeia do Vale, Goiania, Goias. Brazil, CEP: 74680-160, Brazil; email: renatosampaiotavares@gmail.com.