Transplantation and Myelofibrosis: A Little Medicine and a Little Art

Myelofibrosis is a chronic malignancy, the morbidity of which typically evolves over years. For the past decade, decision making for HCT often has been based on two scoring systems: the Dynamic International Prognostic Scoring System (DIPSS) and the DIPSS-Plus. These prognostic metrics were developed to track risk profile over time.^3,4  They have also helped establish the timeline for HCT in patients with myelofibrosis. An expert panel recommends consideration of HCT for patients younger than 70 years with DIPSS or DIPSS Plus scores of intermediate-2 or higher, and for intermediate-1 risk patients younger than 65 years if they have refractory/transfusion-dependent anemia, peripheral blasts higher than 2 percent, or adverse cytogenetics, mainly derived from the poor median overall survival of five or fewer years in this subset of patients.⁵  A retrospective comparison between HCT and nontransplantation therapies from two international multicenter databases, in patients younger than 65 years who never received JAK inhibitors, showed benefit of HCT approach in patients with intermediate-2 and high-risk DIPSS.⁶ 

Neither of these scoring systems, however, takes into account molecular mutations, which have been associated with more aggressive disease phenotypes or with early transformation to blast-phase disease, a dire development. To amend that, Mutation-Enhanced International Prognostic Score Systems (MIPSS70 and MIPSS70-Plus), and subsequently, MIPSS70-Plus version 2.0, have been described for patients 70 years or younger to include clinical factors along with cytogenetic and molecular mutations in a prognostic model.^7,8  Presence of one or two high-risk mutations such as ASXL1, EZH2, SRSF2, or IDH1/2, and absence of CALR type 1 mutation added additional risk factors in the MIPSS70-Plus model. In version 2.0, the U2AF1 Q157 mutation, a very high-risk cytogenetic category and gender-based hemoglobin stratification, was included in the scoring model. Using retrospective or registry datasets, efforts to apply these newer systems to the question of when to transplant, and in whom, have been published. For example, the former, a four-tiered MIPSS70-Plus classification, suggests watchful monitoring for low-risk disease while considering HCT for high- and very high-risk disease.⁸  Similarly, for the five-tier MIPSS70-Plus version 2.0, HCT is recommended for high- and very high-risk patients, while non-HCT management is recommended for very low-, low-, and intermediate-risk patients.⁷  Of note, these scoring systems were developed and validated in patients with “primary” myelofibrosis but are often applied to post-polycythemia vera or post-essential thrombocythemia myelofibrosis with similar principles. That being said, there is a specific prognostic scoring system that has been developed for patients with secondary myelofibrosis who are older than 65 years, with time to secondary myelofibrosis greater than 15 years, who have previous thrombosis and constitutional symptoms, hemoglobin lower than 10 g/dL, and circulating blasts of 1 percent or higher.⁹  This scoring system has not yet undergone rigorous validation with regard to transplant outcomes.

The utility of these scoring systems walks closely with identification of changes in clinical course of the patient. Transitions such as worsening transfusion requirements, increasing blasts, the appearance or deterioration of constitutional symptoms can be ominous signs and should be recognized early to enable timely initiation of treatments, including consideration for HCT. Data suggest that patients with three or more mutations may have limited response to JAK inhibitors, and hence, these patients may be the ones to start thinking of HCT at the first indication of clinical deterioration.¹⁰  Further, there are data that suggest that HCT may benefit patients with high-risk chromosomes¹¹  and somatic mutations.^12,13 

For HCT, “the older the age, the worse the outcomes” is a generally accepted phenomenon. This is especially relevant in myelofibrosis because the median age at diagnosis is older than 65 years. With recent improvements in supportive therapy and the popularity of reduced-intensity conditioning, the age limit for HCT continues to shift upward. In myelofibrosis, HCT with reduced-intensity conditioning has been reported in patients older than 70 years with some success.¹⁴  This also shows that chronological age itself might not be as informative as in combination with performance status and comorbidities.

Eventually, it’s about striking the balance between anticipated benefit in disease features and risks from transplant-related mortality based on an individual’s overall health and performance status. Fine tuning of this assessment relies on experience and appreciation of disease transitions. The art lies in close follow-up and enquiry into the symptoms, leading to some understanding of the disease biology in an individual patient.

As illustrated in this article, while expert panel recommendations have been established, many aspects related to timing and the point of maximal benefit remain shrouded in the “no data zone.” Additionally, myelofibrosis being a relatively uncommon malignancy, it is almost obligatory to conduct collaborative efforts.

Transplantation versus nontransplantation options. Most established recommendations are based on the data available from the pre–JAK inhibitor era. Whether the same holds true with improvements attributed to JAK inhibitors remains a question. A randomized study to compare HCT versus non-HCT options would be difficult to conduct. However, a current Center for International Blood and Marrow Transplant Research initiative is enrolling patients older than 55 years, to compare outcomes of patients undergoing HCT with those of an age-matched historical cohort (NCT02934477), which would shed some light towards this.

While on ruxolitinib. In real-world clinical scenarios, another challenging question is what is the best window to HCT when patients have improvement in symptoms while on JAK inhibitors. Is it when patients have the maximal response from JAK inhibitors or when they begin to lose the clinical response, such as at the return of the constitutional symptoms or splenomegaly? This can be feasibly studied in a randomized fashion as a collaborative effort.

Should you transplant if there are too many mutations? It also remains unknown whether the high risk portended by mutation status is overcome by HCT — information that can provide additional insight into identifying patients who may benefit from HCT consideration at an earlier course in the disease (or not) and that might offer a quantification of benefit from HCT in these patients.