Oncofusion

start / graph / trial-augment-101

Trialstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q6b-therapeutics2

AUGMENT-101 (NCT04065399) — the registration trial, and what 22.8% actually means

also called AUGMENT-101 · NCT04065399 · SNDX-5613 registration trial

AUGMENT-101 is the trial that got the first menin inhibitor approved. Roughly two thirds of patients responded in some way, but only about one in five reached the strict definition of remission the approval was based on.

The same thing, for a scientist

AUGMENT-101 is an open-label phase 1/2 dose-escalation and expansion study of revumenib across 22 sites in five countries, enrolling patients aged 30 days and older with relapsed/refractory KMT2A-rearranged acute leukemia or NPM1-mutant AML. In the registration-enabling phase 2 KMT2A-rearranged portion, 94 patients were treated (median age 37 years, range 1.3-75). Among 57 efficacy-evaluable patients the CR+CRh rate was 22.8% (95% CI 12.7-35.8), exceeding the 10% null hypothesis (p=0.0036); overall response rate was 63.2% (95% CI 49.3-75.6) with 15 of 22 assessed (68.2%) having no detectable residual disease. Grade ≥3 events included febrile neutropenia 37.2%, differentiation syndrome 16.0% and QTc prolongation 13.8%. The trial remains open with planned enrolment of 447.

Go deeper
The numbers, and the gap between response and remission2 deeper
EndpointValue
Treated (safety)94
Efficacy-evaluable (centrally confirmed KMT2A-r)57
CR + CRh (primary)22.8% (95% CI 12.7-35.8), p=0.0036 vs 10% null
Overall response rate63.2% (95% CI 49.3-75.6)
MRD-negative among assessed15/22 (68.2%)
Grade ≥3 febrile neutropenia37.2%
Grade ≥3 differentiation syndrome16.0%
Grade ≥3 QTc prolongation13.8%
Median age37 years (range 1.3-75)

Enrolment ran 1 October 2021 to 24 July 2023 1.

The 63.2% versus 22.8% gap is the whole clinical story. Two thirds of patients have a measurable anti-leukemic effect; fewer than a quarter achieve a remission deep and durable enough to count. Most of the difference is patients who clear blasts but do not recover counts — which is what you would expect from a differentiation agent acting on a marrow that has no normal haematopoiesis left in it. It is also why the drug's natural home is frontline and combination, not monotherapy salvage.

The MRD result cuts the other way and is the most encouraging number in the trial: 68% of assessed responders had no detectable residual disease. When it works, it works deeply. That is what makes it a credible bridge to transplant and a credible partner for an immune agent.

Paediatric eligibility — the detail that matters most for this lab

Eligibility was age 30 days and older, and the dose was weight-adjusted (163 mg every 12 hours, or 95 mg/m² if weight under 40 kg, with a strong CYP3A inhibitor) 1. The age range of treated patients ran down to 1.3 years, and the author list includes paediatric investigators from Memorial Sloan Kettering, the Princess Máxima Center and the University of Iowa.

So children were on the registration trial, which is why the approval covers them. But the median age was 37 and no paediatric subgroup efficacy analysis appears in the primary report. The approval that covers infants rests on a trial whose median patient was middle-aged. That is the structural feature of this whole field and it is worth stating plainly rather than treating as a technicality.

AUGMENT-101 is still listed as RECRUITING as of 2026-09-18 with a planned total enrolment of 447 across 57 sites, and the NPM1-mutant cohort is ongoing 2.

Differentiation syndrome and QTc — the two mechanism-linked toxicities

Grade ≥3 differentiation syndrome in 16% is not a side effect in the ordinary sense; it is the drug working. A menin inhibitor forces differentiation of a large blast population, and the resulting cytokine release, capillary leak, respiratory distress and organ dysfunction scale with how well the drug is doing its job. It can be fatal. Any combination that increases efficacy should be expected to increase it.

QTc prolongation at grade ≥3 in 13.8% is compounded by the fact that revumenib is deliberately co-administered with a strong CYP3A inhibitor to achieve exposure. In an infant also receiving azole antifungal prophylaxis and possibly venetoclax, the CYP3A arithmetic becomes the practical limiting factor on combination design.

Facts and figures
nctNCT04065399
phase1/2
statusRECRUITING as of 2026-09-18 (registration-enabling KMT2A-rearranged portion reported; NPM1-mutant cohort ongoing)
open nowtrue
pediatric eligibletrue
minimum age30 days
n planned447
n treated registration portion94
n efficacy evaluable57
locations count57
sponsorSyndax Pharmaceuticals
dosingrevumenib 163 mg (95 mg/m² if weight <40 kg) every 12 hours with a strong CYP3A inhibitor, 28-day cycles
median age years37
enrolment period2021-10-01 to 2023-07-24 (registration portion)
readoutPositive on primary endpoint; supported the first menin inhibitor approval
safety grade3plus
qtc_prolongation_pct13.8qtc_prolongation_pct: 13.8differentiation_syndrome_…16.0differentiation_syndrome_pct: 16febrile_neutropenia_pct37.2febrile_neutropenia_pct: 37.2
primary result
{
  "cr_crh_pct": 22.8,
  "ci95": [
    12.7,
    35.8
  ],
  "p": 0.0036,
  "null_hypothesis_pct": 10
}
secondary results
{
  "orr_pct": 63.2,
  "orr_ci95": [
    49.3,
    75.6
  ],
  "mrd_negative": "15/22 (68.2%)"
}
safety grade3plus — the numbers behind the chart
{
  "febrile_neutropenia_pct": 37.2,
  "differentiation_syndrome_pct": 16,
  "qtc_prolongation_pct": 13.8
}
age range years
[
  1.3,
  75
]
How this could bite us back
  • Differentiation syndrome at grade 3 or worse in 16% is on-mechanism and potentially fatal; it scales with efficacy, so a better combination is also a more dangerous one.
  • QTc prolongation in 13.8% is compounded by deliberate CYP3A boosting, and stacking azole prophylaxis plus venetoclax in an infant makes the interaction arithmetic the real limit on combination design.
  • Selection of MEN1 interface mutants: this trial's own single-agent salvage design is the setting in which clinical resistance mutations were first identified.
  • Endpoint framing: a 22.8% CR+CRh rate against a 10% null is a low bar, appropriate for a desperate population but easily mistaken for evidence of a transformative agent when quoted without the denominator.
  • Paediatric extrapolation: an approval covering one-year-olds resting on a trial with a median age of 37 and no published paediatric subgroup is the central evidentiary weakness in paediatric menin-inhibitor use.
Still open
  1. What was the response rate in the paediatric patients on this trial? The age range goes down to 1.3 years, paediatric investigators are on the author list, and no paediatric subgroup analysis has been published — yet the approval covers children from one year of age.
  2. What is the duration of response and overall survival? Not stated in the abstract and not asserted here. Given a 22.8% CR+CRh rate, durability determines whether this is a bridge or a treatment.
  3. Why do two thirds of patients respond but under a quarter achieve CR+CRh? If the gap is count recovery rather than leukemia control, the endpoint may be understating the drug and combinations that support haematopoiesis would look very different.
  4. How many of the non-responders were resistant at baseline rather than acquiring resistance? The patient-derived xenograft evidence that multiply pretreated disease is already menin-resistant predicts a substantial fraction, and this trial has the samples to check.
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In the registration-enabling phase 2 portion of AUGMENT-101, 94 patients aged 30 days and older with relapsed/refractory KMT2A-rearranged acute leukemia were treated with revumenib 163 mg (95 mg/m² if under 40 kg) every 12 hours with a strong CYP3A inhibitor; among 57 efficacy-evaluable patients the CR+CRh rate was 22.8% (95% CI 12.7-35.8, p=0.0036 against a 10% null), overall response 63.2% (95% CI 49.3-75.6), 15 of 22 assessed (68.2%) had no detectable residual disease, and grade 3 or worse adverse events included febrile neutropenia 37.2%, differentiation syndrome 16.0% and QTc prolongation 13.8%.

    Menin Inhibition With Revumenib for KMT2A-Rearranged Relapsed or Refractory Acute Leukemia (AUGMENT-101) · Issa GC, Aldoss I, Thirman MJ, et al. · Journal of Clinical Oncology 2025
    PMID 39121437DOI
    What this evidence cannot show

    Single-arm, open-label, no control. The efficacy population (57) is much smaller than the treated population (94) because it requires central confirmation of KMT2A rearrangement, which selects. Median age 37 years with no paediatric subgroup efficacy analysis reported, despite the approval covering children from one year. Duration of response and overall survival are not stated in the abstract and are not asserted here. The trial was industry-sponsored and several authors are Syndax employees.

    Evidence that argues against this (2)
    • The registration response rate is not the durable-benefit rate, because revumenib exerts enough selection pressure in patients to breed escape. In the phase 1 portion of the same programme, somatic MEN1 mutations at the revumenib-menin interface were recovered from patients with acquired resistance; these mutants attenuate drug binding while leaving the natural menin-MLL1 interaction intact and prevent inhibitor-induced eviction of menin and MLL1 from chromatin. The authors describe this as the first demonstration that a chromatin-targeting drug drives the evolution of escape mutants that restore chromatin occupancy - which means the 22.8% CR+CRh is a snapshot of a population under active selection, not a stable effect. Perner F et al. 2023
    • Pooled across the class, single-agent menin inhibition performs close to the AUGMENT-101 figure and clearly worse than combination. A systematic review and meta-analysis of 14 studies and 784 treated adults with relapsed/refractory AML found a pooled ORR of 54.6% (95% CI 46.4-62.6, I2=63.5%), pooled CR 29.3% and CR+CRh 28.5%, but with monotherapy CR of only 19.5% against 43.3% for menin inhibitor plus hypomethylating agent plus venetoclax (p=0.002). Differentiation syndrome occurred in 14.6% and treatment-related mortality in 5.0%. The implication is that AUGMENT-101's result is near the ceiling for monotherapy and that the drug's real value may lie only in combination - which the registration data do not establish. Alhajahjeh A et al. 2026
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.trial_registry

    AUGMENT-101 (NCT04065399) is a phase 1/2 open-label dose-escalation and dose-expansion study of revumenib in relapsed/refractory leukemias including KMT2A-rearranged and NPM1-mutant disease, sponsored by Syndax Pharmaceuticals, with planned enrolment of 447 across 57 sites, started 5 November 2019, and listed as RECRUITING as of 2026-09-18.

    NCT04065399
    What this evidence cannot show

    Registry snapshot of 2026-09-18; sponsor-maintained records update irregularly. The 447 figure is planned enrolment across all cohorts, not the registration cohort.

    Evidence that argues against this (1)
    • A registry record is a list of PRESPECIFIED outcomes, not of everything the trial measured, and AUGMENT-101's list contains no survival endpoint at all. Raw ClinicalTrials.gov v2 API, retrieved 2026-09-19: the primary outcomes are, for phase 1, dose-limiting toxicities, treatment-emergent adverse events, Cmax, Tmax and AUC0-t; for phase 2 cohorts 2A-2C, CR+CRh rate and treatment-emergent adverse events; and for cohort 2D, Cmax and AUC0-tau. Neither overall survival nor event-free survival appears. Both FDA approvals of revumenib therefore rest on a trial that never prespecified a survival endpoint. The record is also still RECRUITING with primary completion pushed to 2027-12-15 and hasResults FALSE, so it cannot be cited for results either. Syndax Pharmaceuticals 2019
Connected to