Patterns of response to venetoclax + hypomethylating agents in IDH-mutated AML

Isocitrate dehydrogenase 1/2 (IDH1/2) mutations occur in around 20% of patients with acute myeloid leukemia (AML) and, as recently reported on the AML Hub, venetoclax (VEN) in combination with a hypomethylating agent (HMA), such as decitabine or azacitidine, is a promising therapeutic option for IDH-mutated AML.

Here, we are pleased to summarize a study by Danielle Hammond and colleagues, presented during the 62nd American Society of Hematology (ASH) Annual Meeting and Exposition.¹ In this retrospective analysis, the patterns of response to VEN + HMA, including measurable residual disease (MRD) by flow cytometry (FC) and next generation sequencing (NGS), were evaluated in patients with IDH-mutated AML.

Study design

A single-institution, retrospective analysis of 65 patients with newly diagnosed or relapsed/refractory (R/R) IDH1/2-mutated AML who had been treated with VEN in combination with HMA.

Primary analyses:

• Responses according to modified 2017 European LeukemiaNet criteria, incorporating MRD assessment by FC (sensitivity 10^-3–10^-4).
• NGS MRD by mutant IDH variant allele frequency assessment of bone marrow samples pre-treatment and at time of best response.

Secondary analysis:

• Comparative response in a subset of patients treated with IDH inhibitor (IDHi)-based regimens.

Key findings

Selected patient, disease, and treatment characteristics are shown in Table 1.

• Almost three-quarters of patients had an IDH2 mutation; commonly co-mutated genes included SRSF2, RAS-pathway, and NPM1.
• Decitabine was the HMA administered in 85% of cases.
• Common reasons for VEN + HMA discontinuation (n = 57) were relapse (30%), resistant/inadequate response (18%), stem cell transplant (25%), and patient/physician choice (20%).
• Approximately half of patients had received previous therapies; 15% had received prior IDHi treatment, and 40% had received treatment with IDHi following VEN + HMA.

Table 1. Selected patient, disease, and treatment characteristics¹

AML, acute myeloid leukemia; del, deletion; ELN, European LeukemiaNet; FLT3, FMS-like tyrosine kinase 3; HMA, hypomethylating agent; IDH, isocitrate dehydrogenase; IDHi, IDH inhibitor; ITD, internal tandem duplication; NPM1, nucleophosmin-1; RAS, rat sarcoma; VEN, venetoclax.
Characteristic	N = 65
Median age, years (range)	69 (32–83)
2017 ELN prognostic group, %    Favorable    Intermediate    Adverse	22 25 45
Cytogenetics, %    Diploid    Complex    Del(17p)    -7/-7q and/or -5/-5q	37 18 3 14
IDH mutational status, %    IDH1    IDH2	29 71
Co-mutations, %    SRSF2    NPM1    FLT3-ITD    RAS-pathway    TP53	43 28 9 31 16
Prior therapies for AML, %    0    1    2    ≥ 3	55 15 14 15
Median number of VEN + HMA cycles (range)	3 (1–34)
IDHi therapy    None    Prior IDHi treatment    Subsequent IDHi treatment	47 15 40

Response rates, MRD, and overall survival

Best bone marrow response rates after a median of one treatment cycle (range, 1–5) are shown in Figure 1.

CR, complete response; CRi, CR with incomplete hematologic recovery; MLFS, morphologic leukemia-free state.

Overall, 67.7% of patients achieved a complete response (CR) or CR with incomplete hematologic recovery (CRi) after VEN + HMA treatment. Of all patients achieving CR/CRi, 90% were also MRD-negative by FC, whereas only 59% were MRD-negative by NGS. Median duration of response was 18.0 months (95% CI, 13.7–NE).

Responses were deeper and more durable for patients receiving frontline VEN + HMA therapy compared with the R/R setting (CR/CRi 86.1% vs 44.8%, and median duration of response 24.1 months vs 15.1 months). After frontline therapy, 90% of CR/CRi responses were FC MRD-negative, compared with 42% MRD negativity by NGS. MRD responses were less robust in the R/R setting, with just 62.5% and 50% of CR/CRi responses also negative by FC and NGS MRD, respectively.

Median overall survival (OS) for all patients treated with VEN + HMA was 22.4 months (95% CI, 16.4–NE), and was not reached (95% CI, 42.1–NE) for patients who had received frontline VEN + HMA. There was a significant association between improved OS and depth of response by bone marrow assessment (p < 0.001) and MRD negativity by FC (p < 0.001) (Table 2). In contrast, MRD negativity by mutant IDH NGS was not associated with a survival benefit (p = 0.385), however mutant IDH variant allele frequency data was only available in 60% of responding patients and analysis of additional samples is underway.

Table 2. Overall survival by depth of response¹

	Median OS, months (95% CI)
CR, complete response; CRi, CR with incomplete hematologic recovery; FC, flow cytometry; mIDH, mutant isocitrate dehydrogenase; MLFS, morphologic leukemia-free state; MRD, measurable residual disease; NGS, next generation sequencing; NE, not evaluable; NR, not reached; OS, overall survival. *Includes only patients with a CR/CRi and evaluable MRD at time of best response by FC (n = 40) or NGS (n = 27).
All patients	22.4 (16.4–NE)
By marrow assessment
CR	NR (NE–NE)
CRi	42.2 (20.1–NE)
MLFS	13.4 (7.9–NE)
No response	5.3 (3.5–NE)
By FC MRD assessment*
FC MRD-negative	NR (42.2–NE)
FC MRD-positive	15.7 (2.8–NE)
By mIDH NGS MRD assessment*
mIDH NGS MRD-negative	NR (NE–NE)
mIDH NGS MRD-positive	NR (20.2–NE)

Response by co-mutation pattern

NPM1 mutations were enriched in patients with superior bone marrow responses and MRD-negative status. Conversely, RAS-pathway and TP53 mutations were enriched in patients with poorer responses and MRD-positive status.

Impact of IDHi treatment

Of 29 patients treated with an IDHi-based regimen, eight patients were treated with IDHi initially, and then switched to VEN + HMA therapy, mainly due to lack or loss of response, and 21 received IDHi as a salvage regimen after VEN + HMA.

A bone marrow response was seen in seven of the eight patients (88%) who received VEN + HMA as salvage therapy, and in 11/21 (52%) patients who received IDHi as salvage therapy after VEN + HMA; the highest response in this setting was seen when an IDHi was added to the VEN + HMA backbone.

Conclusion

VEN + HMA is an effective treatment option for IDH-mutated AML, particularly for patients with co-mutated NPM1 and in the frontline setting, where CR/CRi rates reached 86.1%. MRD negativity by FC was frequent and predictive of superior OS, however, IDH1/IDH2 mutations by NGS were still detectable in around half of patients at best response, highlighting a potential role for combination therapy with an IDHi. Further research into combination vs sequential therapy with VEN + HMA and IDHi-based regimens is warranted.