The court has seen the evidence, venetoclax looks wildly guilty of helping many AML patients, the verdict has been mostly favorable, and then the appeal arrives: a tiny population of leukemia stem cells claiming, with suspicious confidence, that the whole case was built on the wrong witness.
The Usual Suspects Were Not So Usual
Acute myeloid leukemia, or AML, is a fast-moving blood cancer where immature cells crowd the bone marrow and mess with normal blood production. Think of the marrow as a factory that should be making red cells, white cells, and platelets, but the night manager got replaced by a copier jam with ambitions.
Venetoclax changed the AML story because it targets BCL-2, a protein that helps cancer cells dodge apoptosis, the body’s built-in “please exit the premises” program. In older or medically fragile patients, venetoclax plus azacitidine improved outcomes enough to become a major treatment option, as shown in the VIALE-A trial (DOI: 10.1056/NEJMoa2012971).
But AML has a recurring villain problem. Even when treatment clears most leukemia cells, leukemic stem cells can hang around like the last cockroach in a luxury apartment. They self-renew, survive stress, and can restart the disease later. Rude? Yes. Biologically impressive? Also yes.
Four Little Escape Artists
In the new Cell Stem Cell study, Waclawiczek and colleagues profiled leukemic stem cells from more than 150 AML patients and found four distinct LSC subtypes, each resembling a different stage of normal blood development (DOI: 10.1016/j.stem.2026.04.012).
That matters because the cells’ developmental “flavor” helped determine which survival protein they leaned on. Some depended heavily on BCL-2, making venetoclax a good hit. Others leaned more on BCL-xL or MCL-1, which is basically cancer saying, “Nice lock. I use the side door.”
The team used a measurement called the MAC-score, a ratio comparing BCL-2 with resistance-associated proteins BCL-xL and MCL-1. Earlier work from the same group showed that this score, measured in LSCs by flow cytometry, could predict response to azacitidine/venetoclax (DOI: 10.1158/2159-8290.CD-22-0939; PMCID: PMC10236156).
Translation: instead of asking “Does this patient have AML?” doctors may need to ask, “Which survival scam is this patient’s AML currently running?”
The Plot Twist Has Platelets
The biggest resistance pattern involved LSCs shifting toward a megakaryocytic/erythroid progenitor-like state, or MEP-LSC. These are cells leaning toward blood lineages that normally help make platelets or red blood cells. In that state, the leukemia cells became less dependent on BCL-2 and more dependent on BCL-xL.
This fits with previous research showing that erythroid and megakaryocytic differentiation can create BCL-xL dependency and venetoclax resistance in AML (DOI: 10.1182/blood.2021011094; PMCID: PMC10651789).
So venetoclax is still doing its job. The target moved. It is less “the drug failed” and more “the leukemia changed outfits and slipped past security wearing a fake mustache.”
The Rare Weirdos Matter Too
The researchers also found a rarer resistant subtype: mature monocytic/dendritic-like LSCs, including LAMP5-positive monocytic AML. These cells were not just mildly annoying. They appeared to drive venetoclax resistance in specific cases.
Other recent work has also argued that monocytic leukemia stem cells can behave differently from more primitive LSCs and may require different therapeutic pressure (DOI: 10.1158/2159-8290.CD-22-1297; PMCID: PMC10527971).
Here is the useful part: these subtypes were not just labels for a very fancy filing cabinet. They pointed to vulnerabilities. MEP-LSCs showed sensitivity to BCL-xL inhibition. MoDe-LSCs showed sensitivity to MEK1/2 inhibition. Different escape route, different roadblock.
Why This Could Actually Change Things
Venetoclax resistance is already a major clinical headache, and reviews keep circling the same theme: AML cells can escape by shifting metabolism, mutation patterns, maturation state, or anti-apoptotic protein dependence (DOI: 10.1111/bjh.19314; DOI: 10.3389/fonc.2025.1577908).
This study gives that chaos a more usable map. If validated in clinical trials, LSC subtyping could help doctors choose treatment based not only on gene mutations, but also on what the leukemia stem cells are currently addicted to. BCL-2? Venetoclax makes sense. BCL-xL? Maybe add or switch strategy. MEK signaling? Different trap.
Cancer biology, naturally, refuses to be tidy. But this paper makes one idea hard to ignore: AML relapse may not be random bad luck. Sometimes it is a stem cell changing its identity just enough to survive the drug meant to kill it.
The appeal is still pending. But the prosecution now has better evidence.
References
-
Waclawiczek A, Leppä A-M, Renders S, et al. Leukemic stem cell subtypes determine venetoclax resistance and therapeutic vulnerabilities in AML. Cell Stem Cell. 2026. https://doi.org/10.1016/j.stem.2026.04.012
-
DiNardo CD, Jonas BA, Pullarkat V, et al. Azacitidine and Venetoclax in Previously Untreated Acute Myeloid Leukemia. N Engl J Med. 2020;383:617-629. https://doi.org/10.1056/NEJMoa2012971
-
Waclawiczek A, Leppä A-M, Renders S, et al. Combinatorial BCL2 Family Expression in Acute Myeloid Leukemia Stem Cells Predicts Clinical Response to Azacitidine/Venetoclax. Cancer Discovery. 2023;13:1408-1427. https://doi.org/10.1158/2159-8290.CD-22-0939
-
Kuusanmäki H, Dufva O, Vähä-Koskela M, et al. Erythroid/megakaryocytic differentiation confers BCL-XL dependency and venetoclax resistance in acute myeloid leukemia. Blood. 2023;141:1610-1625. https://doi.org/10.1182/blood.2021011094
-
Zhang H, Savage S, Schultz AR, et al. A Novel Type of Monocytic Leukemia Stem Cell Revealed by the Clinical Use of Venetoclax-Based Therapy. Cancer Discovery. 2023. https://doi.org/10.1158/2159-8290.CD-22-1297
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.