Early Efficacy With Fedratinib in MDS/MPN and CNL Warrants Its Continued Evaluation

Patients with myelodysplastic syndrome (MDS)/myeloproliferative neoplasm (MPN) overlap syndromes and chronic neutrophilic leukemia (CNL) experienced spleen and symptom responses with the JAK2 inhibitor fedratinib (Inrebic) regardless of diagnosis, molecular profile, or treatment history, according to David A. Sallman, MD, who added that this indicates the viability of using JAK inhibition to ameliorate symptoms in a population that lacks effective options and is typically excluded from clinical trials.

At the 2023 ASH Annual Meeting, preliminary data from a phase 2, multi-institutional, investigator-initiated trial (NCT05177211) evaluated the efficacy of fedratinib in patients with aytpical chronic myeloid leukemia, CNL, unclassifiable MDS/MPN, and MDS/MPN with ring sideroblasts and thrombocytosis. The results showed an overall response rate (ORR) of 36% (n = 4/11) and a spleen response rate of 20% (n = 2/10). In the 10 patients evaluable for symptom response, the response rate was 40%.

Eight patients remain on treatment, with a median time on treatment of 7.6 months. The median overall survival (OS) was not reached for these patients. Three patients discontinued treatment due to disease progression after 11 months, an inability to keep up with trial visits, and allogeneic hematopoietic cell transplant. The agent’s safety profile in MDS/MPN appeared consistent with previous reports in myelofibrosis.¹

“If you have a patient with MDS/MPN and a clinical phenotype similar to your patients with classic primary myelofibrosis, there can be efficacy with the JAK inhibitor fedratinib in this setting,” said Sallman, an assistant member in the Department of Malignant Hematology at Moffitt Cancer Center in Tampa, Florida.

In an interview with OncLive®, Sallman discussed unmet needs and clinical trial challenges for patients with MDS/MPN overlap syndrome; expanded on the potential for fedratinib to address symptoms in this population; and detailed current and future research directions in MDS, including those presented at the 2023 ASH Annual Meeting.

OncLive: What unmet needs exist for patients with MDS/MPN overlap syndromes, and how might fedratinib ameliorate these?

Sallman: MDS/MPN patients are a group of patients who have no good therapeutic options. Because of this [disease] overlap, they’re very much excluded from the vast majority of MDS trials, and they are separately rejected from the vast majority of MPN trials. It’s a heterogeneous patient group with a lot of different molecular subgroups that can fit within [this disease categorization]. However, a lot of these patients will have different symptoms. Especially in the MDS/MPN category, many of these patients can have significant symptom burden and symptomatic splenomegaly that is much more similar to [that seen in] the MPN space.

The main crux of the trial was to ask: Can we take JAK inhibitors, which have changed the treatment landscape for patients with primary myelofibrosis and now apply that in the setting of this overlap syndrome, where these patients really don’t have options? These would all be off label [uses].

Could you briefly discuss the methodology utilized in this investigation?

We took the FDA-approved dose of fedratinib and applied that in this patient population. Patients were required to either have symptomatic splenomegaly as quantified by imaging, or to have significant [symptom] burdens. There are symptom scores that these patients have at baseline [that were then assessed] sequentially over time. Other than that, the study is relatively standard. There are serial laboratory counts, spleen imaging, and symptomatic question scores, as well as bone marrow biopsies to evaluate for evidence of efficacy.

What key findings were presented from this study at the 2023 ASH Annual Meeting? Were these outcomes consistent with what was expected?

The findings show that [fedratinib] can work. The agent [produced] quite comparable response rates to [those of] patients who have symptomatic primary myelofibrosis, although the trial had relatively small numbers [of patients]. Patients had both symptom and spleen responses, [which] have been durable to date.

The trial has been panning out exactly as we really expected it to. We’d love to have predictive biomarkers to identify the patients who best responded [and whether] there [are optimal] molecular subgroups of patients [for this therapy], because fedratinib not only inhibits JAK2 but FLT3 and [other targets]. The number of patients [on the trial] are too small to answer those questions definitively. The trial is continuing to develop into a multicenter, expansion [portion], so hopefully we can tease out more of those data over time.

What next steps are planned for this research in MDS?

It’s a very challenging space to think about a pivotal registrational path for, because this group of patients has a high degree of heterogeneity. If the data are further supported by larger cohorts of patients—even in this setting, given the clear lack of options—this could support off-label utilization or considerations. It would take a very large study for this to go forward. Again, it’s a challenging group of patients, so that may not be possible.

What other MDS research from the meeting did you find particularly interesting or impactful?

There are several key initiatives in MDS looking to validate response criteria classifications. There were several oral presentations at the 2023 ASH Annual Meeting [based on data from] the Validate Database. There has been an update in the International Working Group 2023 response criteria for MDS where [the criteria for a] complete remission [CR] changed from a hemoglobin [threshold] of 10 [g/dL] to 11 [g/dL or higher]. Now we have quantified a CR_H [CR with partial hematologic recovery] or CR_L [CR with limited count recovery] response, which are really marrow responses in the setting of hematopoietic recovery. This is what is most important for MDS patients. What’s key from the data, and what we showed, is that if you achieve a CR or a composite CR, the long-term survival of these patients is relatively identical to patients who achieve other responses and clearly better than [those who] do not respond. This matters going forward with drug development. When you’re reporting efficacy [data], a composite CR_H or CR_L is a meaningful response. This is even important in discussions with the FDA because some drugs may have been challenged with only having marrow responses. We need to quantify the other quality responses that should count in that setting. There was a total of 6 abstracts from this international consortium.

The other oral presentation was looking at outcomes [in MDS based on p53 mutation status]. The takeaway is that this [patient] group is doing poorly independent of any [effective] therapy. There were no differences [in outcomes] whatsoever between azacitidine and decitabine. Although there’s a ton of discussion on the impact of allelic status in the setting of hypomethylating agent [HMA] therapy, the majority of these patients are high risk, and we don’t see any prediction of outcomes with HMA based on the allelic status. [Accordingly], it’s most useful in predicting [responses in] patients who can have that more indolent course. These are typically going to be patients without excess blasts and without complex karyotype. [For] most patients who we’re treating in this study, the impact of [allelic status] either in the setting of HMA therapy or transplant is likely low. [This research provides us with] a nice, large dataset to answer some myths in that p53-mutant patient population.

Lastly, I presented an investigator-initiated [phase 1b/2] trial [NCT04798339] of canakinumab with darbepoetin alfa [in lower-risk MDS]. Canakinumab is a monoclonal antibody that leads to selective depletion of interleukin-1 beta [IL-1β], which is the N-cytokine most critical in inflammasome activation, which is central to MDS pathogenesis. We [are investigating whether this can be] a disease modification therapy. We do see on-target effects as far as decreasing ASC specks and some other factors associated with inflammasome activation, although unfortunately we are not seeing responses in that setting. It may be that we have to target multiple pathways; not just IL-1β but NLRP3, and you have to have somewhat of a combinatorial approach to achieve that. That’s the path forward. [The agent is] very safe, but although the study is ongoing in an expansion phase, canakinumab does not appear to have significant activity, at least in all-comer or lower-risk MDS population.

Could you spotlight any highly anticipated upcoming or ongoing trials in MDS? How might such research have the potential to shift practice?

At the 2023 ASH Education Session on Malignant Hematology,which I chaired and co-presented during, we highlighted MDS and how we are going to manage it. In the higher-risk setting, we’re all looking forward to seeing the data readout of the phase 3 VERONA trial [NCT04401748]. We’ve now had a lot of trials in high-risk MDS [that failed to meet their primary end point], and we want to have a new standard of care. If that trial [does not meet its primary end point], we must start from scratch and dramatically rethink how we design pivotal clinical trials in MDS. They probably can’t be all-comer patient populations, and we [may] need to think about [evaluating] p53 wild-type groups separately. Again, the data [for HMA and venetoclax (Venclexta)] looked pretty good, although [it remains to be seen] if the true CR rate of 30% and median survival of 26 months is better [than our current standard]. Patients who can get to transplant with venetoclax do particularly well, but that’s really the dataset we’re eagerly awaiting, because it is going to greatly influence how we think about prospective frontline trials in the high-risk setting.

Reference

Kuykendall AT, Pettit KM, Singh A, et al. A phase 2 study of fedratinib in patients with MDS/MPN and chronic neutrophilic leukemia. Blood. 2023;142(suppl 1):73. doi:10.1182/blood-2023-182879