Experiment: time-to-resistance, not synergy index, as the readout for menin-inhibitor combinations
also called orthogonality panel · resistance-competent combination screen
Almost every study of drug combinations in this leukemia measures how many cells die in a few days. This experiment instead measures how long it takes the leukemia to come back, which is the thing patients actually care about.
The same thing, for a scientist
A head-to-head comparison of combination partners for menin inhibition in KMT2A::AFF1 models, with time-to-resistance under continuous drug pressure as the primary endpoint rather than a short-term synergy index. Arms pair a menin inhibitor with a same-layer chromatin agent (DOT1L or BET inhibitor), an orthogonal agent (venetoclax; a CD19xCD3 engager with human T cells), a driver-eviction agent (KAT6/KAT7 inhibitor), and monotherapy controls. Emergent resistant populations are characterised by MEN1 sequencing plus RNA-seq and ATAC-seq to classify escape as genetic, transcriptional or apoptotic-threshold. This is the experiment that discriminates deeper killing from durable control, and the field currently does not run it.
Design1 deeper
Models. KMT2A::AFF1 lines with retained menin dependency (SEM, SEMK2, MV4-11 as a KMT2A::AFF1 AML comparator) plus at least two patient-derived xenografts from untreated infant KMT2A-rearranged ALL. Untreated PDXs matter specifically because multiply pretreated samples may already be menin-resistant before any drug is applied.
Arms (menin inhibitor at a fixed, clinically achievable exposure in all combination arms):
- Menin inhibitor alone (reference)
- Menin + DOT1L inhibitor — same layer
- Menin + BET inhibitor — same layer
- Menin + venetoclax — orthogonal (apoptotic)
- Menin + CD19xCD3 engager with healthy-donor T cells — orthogonal (immune)
- Menin + CDK9 inhibitor, intermittent dosing — orthogonal-ish (output collapse)
- Menin + KAT6/KAT7 inhibitor — driver eviction
- Each partner alone
Primary readout. Time to regrowth to a defined threshold under continuous drug pressure, in vitro over 8-12 weeks and in vivo to a bioluminescence or peripheral-blast threshold. Not an IC50 shift, not a Bliss or Loewe index.
Secondary readouts. Depth of initial log-kill at day 7 (so the dissociation between depth and durability can be shown explicitly); MEN1 amplicon sequencing of emergent populations; RNA-seq and ATAC-seq of emergent populations; BH3 profiling of venetoclax-arm escapees; surface CD19 and CD33 density by flow in every arm at baseline and at escape.
Why the CD19 density measurement is not an afterthought
Menin inhibition drives differentiation. Nobody has published what that does to CD19 surface density in primary KMT2A-rearranged infant ALL, and the answer determines whether menin inhibitors and CD19-directed therapy should be given together, sequentially, or not at all together. This experiment measures it as a by-product at essentially zero marginal cost, and it may be the most immediately useful thing the panel produces.
Interpretation rules, set in advance
- If same-layer arms show the greatest day-7 log-kill but the shortest time-to-resistance, the orthogonality hypothesis is supported and the field's synergy-index literature is shown to be measuring the wrong thing.
- If same-layer arms extend time-to-resistance as much as orthogonal arms, the hypothesis is falsified.
- If the KAT6/KAT7 arm behaves like the orthogonal arms despite being a chromatin agent, the hypothesis must be restated in terms of point of attack rather than pathway layer.
- If escapees from the venetoclax arm are transcriptionally identical to escapees from monotherapy, the apoptotic layer is not independent and the whole framing fails.
| design | Head-to-head combination panel in KMT2A::AFF1 cell lines and untreated infant-derived PDXs, with menin inhibitor held at fixed clinically achievable exposure across eight arms spanning same-layer chromatin partners, orthogonal apoptotic and immune partners, a CDK9 partner and a driver-eviction (KAT6/KAT7) partner, plus monotherapy controls. |
|---|---|
| readout | PRIMARY: time to regrowth to a defined threshold under continuous drug pressure (8-12 weeks in vitro; bioluminescence or peripheral blast threshold in vivo). SECONDARY: day-7 log-kill; MEN1 amplicon sequencing of escapees; RNA-seq and ATAC-seq of escapees; BH3 profiling of venetoclax-arm escapees; CD19 and CD33 surface density by flow at baseline and escape. |
| controls | Menin inhibitor monotherapy as the durability reference; each partner as monotherapy; vehicle; a non-KMT2A-rearranged line to confirm genotype specificity; healthy-donor T cells alone for the engager arm. |
| cost tier | med |
| timeline | In vitro arms 4-5 months including escapee profiling; in vivo PDX arms 9-12 months |
| key design choice | Time-to-resistance replaces synergy index as the primary endpoint. This is the entire point of the experiment and the reason it has not already been done. |
| decision value | Determines which combination the lab and the field should prioritise in frontline infant trials, and produces the first measurement of CD19 surface density under menin inhibition in primary infant KMT2A-rearranged ALL. |
- A negative result in vitro does not exonerate a combination: 8-12 weeks of culture selects for culture-adapted escape, which may not be the escape a patient's marrow microenvironment supports.
- The immune arm is the weakest part of the design and is the arm whose result would most influence clinical decisions, which is an uncomfortable combination.
- Running this panel takes a year. The phase 3 trials that will set practice read out in 2029-2031, so the experiment is only decision-relevant if it starts now.
- Can continuous menin-inhibitor pressure be maintained long enough in a PDX to reach resistance without the mice failing first? If not, serial transplantation under drug is the fallback and it changes the interpretation.
- Is a CD19xCD3 engager arm with healthy-donor T cells a fair model of an infant's own exhausted T cells? Almost certainly not, and the arm will overstate immune durability; an autologous-T-cell variant would be better but is much harder.
- Should an arm use sequential rather than concurrent dosing? Sequencing may matter more than combination, and the panel as designed does not test it.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
Existing combination studies in KMT2A-rearranged leukemia report short-term synergy and enhanced killing rather than time-to-resistance, leaving durability untested: menin plus DOT1L inhibition enhances differentiation and cell killing in short-term assays across KMT2A-rearranged models.
Complementary activities of DOT1L and Menin inhibitors in MLL-rearranged leukemia · Dafflon C, Craig VJ, Méreau H, et al. · Leukemia 2017What this evidence cannot show
Cited to characterise the methodological gap this experiment addresses, not to support a biological claim. It is possible that durability data exist in unpublished form or in figures we did not read.
Evidence that argues against this (1)
- Time-to-resistance in vitro is not the durability endpoint the entry claims to rescue. Escape in this disease is substantially a lineage/identity phenomenon playing out in a patient with a marrow niche and an immune system - 64.3% of post-immunotherapy lineage switches carry a KMT2A rearrangement, at a median of 1.5 months. A cell-line panel cannot produce a lineage switch under immune pressure, so the experiment substitutes one surrogate for another rather than measuring what patients care about. Blood 2025
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
Acquired resistance to menin inhibition in KMT2A-rearranged ALL cells arises either through MEN1 mutation or independently of MEN1 mutation, so an experiment classifying escape must sequence MEN1 and profile transcription in parallel.
Distinct Responses to Menin Inhibition and Synergy with DOT1L Inhibition in KMT2A-Rearranged Acute Lymphoblastic and Myeloid Leukemia · Adriaanse FRS, Schneider P, Arentsen-Peters STCJM, et al. · International Journal of Molecular Sciences 2024What this evidence cannot show
In vitro resistance models; establishes that both routes exist without establishing their relative frequency, which is one of the things this experiment would measure.
Evidence that argues against this (1)
- The classification the design rests on may not be exhaustive. Sorting escape into MEN1-mutant and MEN1-independent is tractable by sequencing, but the clinically observed third route - a cell that stops being the disease the drug targets - is invisible to both categories and to a viability readout. An arm scored as 'MEN1-independent resistance' could be a lineage switch, and the experiment would not know. Blood 2025
- ← supportsSix countermeasures, no head-to-head — so rank their targets: CDK6 is 13th and CDK4 is 16,052nd
Supplies a cheap pre-ranking for the experiment that would actually compare these six head to head.
the head-to-head partner comparison this entry says the field has not run
time-to-resistance panel is the discriminating experiment
- tests →KAT7 / HBO1 — an enzyme, a leukaemia-stem-cell dependency, and an inhibitor already in a crystal
includes a KAT6/KAT7 driver-eviction arm as the boundary case for the orthogonality framing
the panel uses untreated infant-derived PDXs specifically to address this