Outcomes of venetoclax + HMA therapy in patients with MPN-BP

Blast phase (BP) transformation, or secondary acute myeloid leukemia (AML), is the main clinical challenge associated with the treatment of myeloproliferative neoplasms (MPN).¹ The 20-year estimated incidence of the evolution to AML varies by MPN subtype: 9% for primary myelofibrosis (PMF), 4% for polycythemia vera (PV), and 3% for essential thrombocythemia (ET).

Past experience has shown that many patients with MPN-BP are unfit for intensive chemotherapy due to age and comorbidities. Overall survival (OS) is poor in this patient population, although the 3-year survival rate in patients who receive allogeneic hematopoietic stem cell transplant (allo-HSCT) is higher than in those who do not undergo allo-HSCT. It is important to note, however, that most patients with MPN-BP are over 65 years of age, and many suffer from comorbidities that preclude intensive chemotherapy. For these patients, HMA-based treatment regimens offer a therapeutic benefit, and it has been suggested that adding venetoclax to HMA may also offer a survival benefit for patients with MPN-BP.

Gangat N. et al., retrospectively evaluated the outcomes for 32 consecutive patients with MPN-BP treated with venetoclax + HMA at treatment centers in the US, Italy, and Canada, and recently published their results in the American Journal of Hematology.¹

Study design and methods

Patients received venetoclax in a 3-day ramp-up during Cycle 1; either 75 mg/m² azacitidine on Days 1−7 or 20 mg/m² decitabine on Days 1−5 were also administered. The treating physician determined the treatment cycle delays/interruptions and dose reductions, as well as timing of bone marrow biopsy (after either Cycle 1 or Cycle 2).

Laboratory findings, results of cytogenetic and molecular studies, and details related to chronic phase MPN (type, driver mutation profile, treatment) were reviewed.

Results

Patient characteristics

Of the 32 patients with MPN-BP who received venetoclax + HMA, 23 were treated in the firstline setting and nine were treated in the relapsed/refractory setting; only one underwent prior allo-HSCT. Clinical characteristics at the time of leukemic transformation, stratified by complete response (CR)/CR with incomplete recovery (CRi) are shown in Table 1.

Table 1. Clinical characteristics stratified by achievement of CR or CRi*

HMA treatment details and predictors of response

Cycle delays/interruptions occurred in 11 patients, and dose reductions of venetoclax and HMA occurred in 14 (44%) and seven (22%) patients, respectively. About half of treated patients (n = 17) experienced pancytopenia related to therapy, and the incidence was similar regardless of MPN type; in 10 of these patients, this was complicated by neutropenic fever or sepsis. In addition, three patients each had hepatic function abnormalities and gastrointestinal toxicity (anorexia, fatigue, and diarrhea) associated with therapy. One patient died of intracranial hemorrhage. HMA treatment details, response rates and outcomes, stratified by CR/CRi, are shown in Table 2.

Table 2. HMA treatment details, response rates, and overall outcome*

Response was evaluable in all patients:

• CR/CRi was documented in 14 (44%) patients, 10 (31%) of whom had CR and 4 (13%) of whom had CRi; the median time to response in these patients was 1.5 months (range, 1–6 months) and the median duration of response was 3.5 months (range, 0.5–10 months).
  • Notably, six (43%) of the 14 patients who achieved CR/CRi were able to undergo allo-HSCT.
  • Relapse occurred in four (29%) of the 14 responders, including one that occurred after allo-HSCT.
• Two (6%) patients had morphologic leukemia-free state (MLFS), and four (13%) patients had partial response (PR).
• Six patients had residual morphological features of MPN, four of whom had CR/CRi.

Regarding OS:

• Median OS was 8 months.
  • Median survival was longer in patients achieving CR/CRi.
• Patients who were bridged to allo-HSCT trended toward prolonged survival (p = 0.17).
  • Of these six patients, four have died: three from transplant-related complications and one from relapse.
• Median follow-up was 5.5 months from initiation of treatment with venetoclax + HMA; in that time, 23 patients have died.
  • Cause of death is known in 20 patients: disease progression (n = 10), infection (n = 9), and bleeding (n = 1).

• The predictors of shortened survival included complex karyotype (p = 0.05) and N/KRAS mutations (p = 0.06).

When compared to historical controls from a database of patients with MPN-BP treated with HMA alone or intensive chemotherapy, CR/CRi rates and median survival were more favorable with venetoclax + HMA compared with HMA alone; however, there was no significant difference in OS and, in fact, longer-term survival was associated with intensive chemotherapy compared with either HMA alone or venetoclax + HMA.

Predictors of response and survival were observed:

• There was no difference in response rates following venetoclax + HMA therapy between untreated and relapsed/refractory settings.
• CR/CRi was more likely to occur in the absence of pre-leukemic PV phenotype (p < 0.01), complex karyotype (p < 0.01), or N/KRAS mutations (p = 0.03).
  • Furthermore, complex karyotype and N/KRAS mutations independently predicted poor response and shortened survival (p < 0.01), and all patients with pre-leukemic PV phenotype had complex karyotype, explaining the poor response among these patients.
  • None of the other mutations impacted response or survival.
• None of the four patients who had prior anagrelide therapy, and one of six patients who had prior ruxolitinib therapy, achieved CR/CRi.
• JAK2 mutation, lower hemoglobin, higher platelet count, and exposure to hydroxyurea were associated with lower responses, whereas, age, leukocyte count, blast percentage, venetoclax exposure, prior HMA exposure, and type of HMA did not appear to have any effect on treatment response.

Conclusion

For fit patients with MPN-BP, management is centered on induction chemotherapy followed by allo-HSCT, as this is the only treatment that has demonstrated long-term survival in MPN-BP (in a small number of patients). In the study by Gangat et al., the authors demonstrated CR/CRi rates with venetoclax + HMA similar to rates that have been previously reported in AML in frontline and relapsed/refractory settings. They also identified N/KRAS mutations and complex karyotype as independent predictors of poor response. The results of this study suggest that venetoclax + HMA combination has potential value as a bridging therapy to allo-HSCT in patients with MPN-BP and needs further evaluation.