Retrospective Study Shows HSCT Consolidation After Blast Reduction Improves OS in Chronic Phase–Reverted MPN

Patients with myeloproliferative neoplasms in accelerated or blast phase (MPN-AP/BP) who revert to chronic phase (cMPN) after blast-reduction therapy, as well as those with complete response (CR) or CR with incomplete hematologic recovery (CRi) after blast reduction, experience improved overall survival (OS) outcomes after hematopoietic stem cell transplant (HSCT) consolidation therapy, according to findings from a single-center, retrospective analysis evaluating outcomes with intensive and nonintensive blast-reduction strategies in patients with MPN-AP/BP, which were published in Blood Advances.¹

This study, which used clinically relatable response criteria developed at the Princess Margaret Cancer Centre, as well as the European LeukemiaNet (ELN) 2022 acute myeloid leukemia (AML) response criteria, found that patients who received intensive blast-reduction therapy achieved a best overall response rate (ORR) of 77% (n = 62/81) vs 39% (n = 21/54) in those who received nonintensive therapy. CR/CRi and cMPN reversions were observed in 24 and 38 patients in the intensive group and 4 and 17 patients in the nonintensive group, respectively.

Although allogeneic HSCT is the only therapy associated with long-term survival improvements for patients with MPN-AP/BP, this treatment strategy is typically reserved for patients who have achieved disease control. Other blast-reduction strategies include induction chemotherapy, as well as hypomethylating agents (HMAs)—such as azacitidine (Vidaza)—as monotherapy or in combination with agents such as venetoclax (Venclexta).²

“However, the optimal blast-reduction strategy and depth of disease clearance required before HSCT are unknown,” lead study author Marta B. Davidson, MD, PhD, FRCPC, of the Princess Margaret Cancer Centre in Toronto, Ontario, Canada, and coauthors, wrote in the paper.¹ “Moreover, a lack of standardized response criteria to evaluate the treatment of MPN-AP/BP poses challenges for understanding treatment efficacy between reported studies.”

Furthermore, the authors noted that standardized AML response criteria, such as those from the revised ELN guidelines for AML, may not be directly applicable to patients with MPNs. However, many MPN treatment centers determine patient HSCT eligibility based on ELN response criteria, which may exclude several patients who could benefit from consolidative HSCT. The absence of validated response criteria for MPN-AP/BP prompted investigators to develop the Princess Margaret Cancer Centre response criteria.

This retrospective analysis evaluated patients with BCR::ABL1–negative MPN-AP/BP (n = 138) treated with either intensive (n = 81) or nonintensive (n = 57) blast-reduction treatment at the Princess Margaret Cancer Centre between January 1998 and April 2022. Eligible patients included those with confirmed BCR::ABL1–negative MPN, including polycythemia vera (PV), essential thrombocythemia (ET), primary myelofibrosis, post-PV myelofibrosis, post-ET myelofibrosis, or unclassifiable MPNs; evidence of disease transformation to AP or BP; and treatment with intensive induction chemotherapy or nonintensive chemotherapy.

Patients received either standard induction therapy with daunorubicin plus cytarabine arabinoside (ara-C); intensive blast-reduction treatment with ara-C plus fludarabine, idarubicin, and granulocyte-colony stimulating factor (FLAG-IDA) or ara-C plus mitoxantrone, etoposide, and cytarabine (NOVE-HiDAC); or nonintensive HMA monotherapy with azacitidine (n = 35) or decitabine (n = 1), or azacitidine plus venetoclax (n = 18).

In this analysis, best ORR was defined as CR, CRi, and reversion to cMPN. Patients with stable disease (SD) or progressive disease (PD) were considered nonresponders, per the Princess Margaret response criteria.

The authors noted that at Princess Margaret Cancer Centre, patients with MPNs are eligible for HSCT if they have a suitable matched sibling, haploidentical, or unrelated donor; and have achieved at least a CR/CRi or cMPN. HSCT is typically restricted to patients who are a maximum of 70 years of age; however, no upper age limit for HSCT candidacy has been defined.

Best overall response after blast-reduction therapy served as the primary end point of this study. Other end points of interest included OS—defined as the time from AP/BP transformation until death or last follow-up—and the proportion of patients undergoing HSCT.

Patients in the intensive and nonintensive groups had median ages of 60 years (range, 22-76) and 70 years (range, 42-87), respectively. More patients in the nonintensive group had an ECOG performance status (PS) of 2 or higher than in the intensive group (25% vs 9%). The nonintensive group also had more patients with AP disease (44% vs 11%).

The subgroups of patients who received intensive induction therapy composed of daunorubicin plus ara-C (n = 31) or FLAG-IDA/ara-C plus NOVE-HiDAC (n = 50) achieved overall best response rates of 74% and 78%, respectively. However, 29% of patients who received induction daunorubicin plus ara-C required second inductions vs 4% of those who received FLAG-IDA/ara-C plus NOVE-HiDAC. After the induction phase, which consisted of initial induction followed by first reinduction, investigators observed similar responses in patients regardless of the initial chemotherapy regimen.

Of the patients who received lower-intensity regimens, 54 had evaluable treatment responses. The ORR in patients who received HMA monotherapy was 33% (n = 12/36); 11 of those responses resulted in cMPN reversion. The ORR in patients who received azacitidine plus venetoclax was 50%.

Per univariate analysis, among patients in the intensive group, TP53 and RAS pathway mutations were associated with inferior treatment response, as was poorer ECOG PS. The significance of these findings was maintained in a multivariate analysis, where the respective odds ratios (ORs) were 12.42 (95% CI, 2.06-74.85; P = .006), 8.23 (95% CI, 1.11-60.92; P = .04), and 15.89 (95% CI, 1.37-184.26; P = .009).

In the nonintensive group, a univariate analysis demonstrated that variables significantly associated with treatment response were ECOG PS of 2 or more; hemoglobin levels; and blast percentage in BP. In the multivariate analysis, only ECOG PS remained significant (OR, 8.65; 95% CI, 1.32-56.88; P = .02).

To evaluate the optimal pre-HSCT blast-reduction strategy, investigators compared the proportion of patients undergoing HSCT after a frontline blast-reduction strategy. This patient population was limited to patients 70 years of age or younger who had responded to frontline blast reduction, including those who received a second induction regimen. In the intensive group, 73% of patients were eligible for HSCT, including 68% of those who received daunorubicin plus ara-C and 76% of those who received FLAG-IDA/ara-C plus NOVE-HiDAC (P = .45). Among these patients, 54% underwent HSCT, including 48% of patients in the daunorubicin group and 58% of those in the FLAG-IDA/ara-C plus NOVE-HiDAC group. A total of 32% of patients who were otherwise eligible for HSCT did not undergo transplant because of early relapse. Furthermore, donors were identified for all but 1 patient.

In the nonintensive group, 17% and 28% of patients who received HMA monotherapy or azacitidine plus venetoclax would have been eligible for HSCT, respectively. Among these patients, 67% of those who received HMA monotherapy and 100% of those who received azacitidine plus venetoclax underwent HCT.

No statistically significant difference was observed regarding median time to HSCT between the intensive and nonintensive groups, at 145 days (range, 54-455) and 102 days (range, 51-260), respectively (P = .41). Additionally, there was no statistically significant difference regarding median time to HSCT between the induction regimens, and no survival difference was observed between blast-reduction strategies among patients who underwent HSCT.

Investigators performed an analysis to understand how Princess Margaret response criteria compared with traditional AML response criteria in the intensive group. When the Princess Margaret treatment responses were reclassified according to the ELN 2022 AML response criteria, all CR and CRi classifications stayed consistent, and 12 patients with SD and PD were reclassified as having no response. Among the 38 patients with cMPN, 24% were reclassified to CR, and 16% were reclassified to CRi based on the absence of circulating blasts and bone marrow blast clearance to less than 5%. Of the other patients with cMPN, 3% were reclassified to morphologic leukemia-free state, and 58% were classified as having no response or refractory disease.

At a median follow-up of 40.3 months, the median OS was 21.9 months (95% CI, 12.2-95.1) after transformation to MPN-AP and 8.8 months (95% CI, 7.97-10.27) after transformation to MPN-BP. Among patients in the intensive group, investigators observed no OS difference between patients achieving a best response of cMPN vs CR/CRi per the Princess Margaret criteria. Responders achieved improved OS vs patients who achieved a best response of SD. A landmark analysis demonstrated that the OS benefit was limited to responders who received HSCT consolidation after blast-reduction therapy. No OS difference was observed when ELN 2022 AML response criteria were used to distinguish between responders and nonresponders, or when cMPN responders were reclassified by ELN 2022 AML response criteria to either responders or nonresponders.

Among patients in the nonintensive group, investigators observed a statistically significant OS difference between those who achieved cMPN vs SD per the PM criteria. When ELN 2022 AML response criteria were used to assign responses, there was no OS difference between responders and nonresponders. Moreover, when cMPN responders were reclassified by ELN 2022 AML response criteria to either responders or nonresponders, no OS difference was observed.

A univariate analysis showed that treatment response, hemoglobin levels, transformation type (AP vs BP), platelet counts, ECOG PS, albumin levels, peripheral blood and bone marrow blast percentages, ELN risk, RAS pathway mutations, TP53 mutations, and number of mutations were all associated with survival. In a multivariate analysis, transformation type, treatment response, TP53 mutation status, and number of mutations remained significant.

Of the 21 patients with TP53 mutations, 13 received an intensive blast-reduction strategy, resulting in CR, cMPN, SD, and early death in 3, 3, 5, and 2 patients, respectively. Additionally, 8 patients received a nonintensive blast-reduction strategy, 2 of whom responded (1 cMPN and 1 CR). Among the 6 patients with TP53 mutations who received a transplant, 2 underwent HSCT after achieving best responses of SD following blast reduction, and 1 underwent HSCT following an early relapse. HSCT was not associated with a survival benefit in this population.

Investigators conducted a paired sample analysis to understand the association between blast-reduction strategies and Princess Margaret–defined criteria. Mutation burden at the time of MPN-AP/BP diagnosis was compared with post-treatment time points in 17 patients who received induction chemotherapy, 5 who received azacitidine plus venetoclax, and 5 who received HMA monotherapy, with samples collected approximately 30 days after induction, after 1 to 2 cycles, and after 3 to 13 cycles, respectively. Molecular responses included clearance (n = 3), partial clearance (n = 7), persistence (n = 6), emergence (n = 3), and mixed response (n = 7).

Patients with paired samples before and after therapy generally had significant residual mutational burden regardless of response to blast-reduction therapy. Full or partial clearance was observed in 38% of patients, and the mutations that persisted post-treatment included genes beyond those previously associated with AML clonal remissions, such as DNMT3A, ASXL1, TET2, and DTA.

Beyond nonresponders who exclusively had either a mixed response or mutational persistence (n = 4), no clear associations between Princess Margaret–defined treatment responses and molecular response patterns emerged. Each of the defined molecular response patterns were observed among the 12 patients with CR/CRi: 2 clearances, 2 partial clearances, 3 persistences, 2 emergences, and 3 mixed responses. A similar distribution of molecular response patterns was observed in patients with cMPN.

“Together, these findings indicate that mutational burden does not align with the observed clinicopathologic response after blast reduction treatment in patients with AP/BP,” the authors explained.

In 5 patients in whom AP/BP mutations persisted after initial blast-reduction therapy, mutational analysis was also conducted on samples collected after HSCT, at which point full mutational clearance was observed in all patients with evaluable samples.

“Our study highlights that both intensive and nonintensive blast-reduction modalities have limited disease-modifying and clonal clearance capability on their own, and consolidative HSCT is required to impart long-term disease control and survival benefit,” the authors concluded. “There is a need for the MPN community to revisit the criteria best suited for assigning responses and informing clinical decision-making in patients with MPN-AP/BP.”

References

1. Davidson MB, Kennedy JA, Capo-Chichi JM, et al. Outcomes of intensive and nonintensive blast-reduction strategies in accelerated and blast-phase MPN. Blood Adv. 2024;8(5):1281-1294. doi:10.1182/bloodadvances.2023011735
2. Gangat N, Guglielmelli P, Szuber N, et al. Venetoclax with azacitidine or decitabine in blast-phase myeloproliferative neoplasm: a multicenter series of 32 consecutive cases. Am J Hematol. 2021;96(7):781-789. doi:10.1002/ajh.26186