The research team, led by Saisai Li, Yuanyuan Chen, and senior authors Jing Wu and Yunxiang Zhang at the Shanghai Institute of Hematology, Ruijin Hospital affiliated with Shanghai Jiao Tong University School of Medicine, retrospectively analyzed five newly diagnosed AML patients aged 16 or older who were treated at their institution between 2020 and 2024. Each of these cases was identified through targeted RNA sequencing and confirmed by polymerase chain reaction (PCR), a molecular approach that reads the leukemia’s messenger RNA to catch fusion transcripts that conventional karyotyping may miss. To strengthen their analysis, the investigators combined their five molecularly confirmed cases with nine previously reported adult cases of t(10;17)(p15;q21)-associated AML, producing an integrated cohort of 14 patients — small by the standards of leukemia research, but substantial for a fusion this uncommon.

What makes this study particularly valuable to practicing hematologists is the consistency of the immunophenotype it describes. All five institutional cases displayed a strikingly uniform cell-surface protein signature — what the authors call a RAM-like immunophenotype. In technical terms, the leukemic cells showed bright, aberrant expression of CD7, a marker normally associated with T lymphocytes rather than myeloid cells, along with bright CD56, a neural cell adhesion molecule often linked to natural killer cells. The cells were also positive for CD33 and CD117, both canonical myeloid markers, but showed dim CD45 — the pan-leukocyte marker — and absent or reduced HLA-DR and CD38. The acronym RAM itself refers to this combination: CD56-positive, CD45-dim, CD38-dim, and HLA-DR-negative. This profile mimics what has been described in certain other AML entities, but its consistency across ZMYND11::MBTD1-positive patients gives clinicians a visual clue on flow cytometry plots that should, the authors argue, trigger suspicion and prompt RNA-based fusion testing when the diagnosis is otherwise murky.

The genetics of this leukemia carry their own lessons. Of the five patients at the Shanghai institution, conventional karyotyping — the decades-old technique of staining and visually inspecting chromosomes under a microscope — detected the t(10;17) translocation in only one case. That means in four out of five patients, the defining genetic event of their leukemia would have gone unnoticed by standard cytogenetic analysis. Cryptic or subtle rearrangements like this one are precisely the kind of molecular lesion that next-generation sequencing platforms, particularly those examining RNA fusion transcripts, are designed to catch. The study therefore adds to a growing body of evidence that cytogenetics alone underestimates the prevalence of certain fusions and that molecular testing should complement, not merely follow, classical chromosome analysis in diagnostically challenging cases of AML.

The biological significance of the fusion itself is only beginning to be understood. ZMYND11 encodes a protein with a bromodomain-like module involved in reading histone marks and modulating transcriptional elongation, while MBTD1 is a methyl-lysine reader protein associated with chromatin regulation. The fusion of these two chromatin-interacting genes — a hallmark of so-called histone-acetyltransferase (HAT)-related and chromatin-dysregulation leukemias — plausibly rewires gene expression programs during leukemogenesis, though the exact mechanism by which the fusion drives AML remains an open question that larger cohorts and functional studies will need to answer.

Clinically, the picture that emerges is cautiously nuanced. In the institutional cohort of five patients, the complete remission (CR) rate after a single induction cycle of chemotherapy was 40 percent — meaning two of five patients achieved CR with standard initial treatment. However, with sequential therapy, the best CR rate climbed to 80 percent, suggesting that although initial responses may be imperfect, this leukemia can ultimately be brought under control with additional intensive treatment. Induction regimens in this setting included idarubin plus cytarabine (the traditional “7+3”-style backbone) as well as the lower-intensity combination of venetoclax plus azacytidine, reflecting the modern era of AML therapy in which BCL-2 inhibition has reshaped treatment options, particularly for older or less fit patients.

In the integrated cohort of 14 patients, the median age was 51 years and 57 percent were female. When the authors stratified patients according to the 2022 European LeukemiaNet (ELN) risk classification — the internationally recognized framework that assigns AML patients to favorable, intermediate, or adverse risk categories based on genetics — 85.7 percent fell into the intermediate-risk group. That is a noteworthy observation: the ELN system relies heavily on recurrent cytogenetic and molecular abnormalities to assign risk, and a fusion as rare as ZMYND11::MBTD1 is not yet formally incorporated into those tables. The predominance of intermediate-risk assignments reflects this gap and underscores why studies like this one matter; without outcome data on specific rare lesions, clinicians cannot know whether such patients should be treated more or less aggressively than the classification suggests.

Survival analysis, restricted to the 13 patients for whom overall survival (OS) data were evaluable, yielded a median OS of 34.0 months for the integrated cohort. Among those who underwent allogeneic hematopoietic cell transplantation (allo-HCT) — a procedure in which a patient receives blood-forming stem cells from a matched donor after myeloablative or reduced-intensity conditioning — median OS was 37.0 months, compared with 23.0 months for those who did not receive a transplant. Three-year OS was 80.0 percent in the transplanted group versus 33.3 percent in the non-transplanted group. These numbers are superficially encouraging for transplantation, but the authors are careful and appropriately cautious in their interpretation. The comparison is limited by the small sample size and, critically, by substantial treatment-selection heterogeneity — meaning that patients selected for transplantation may differ systematically from those who were not, in age, fitness, remission status, or measurable residual disease burden. Such confounding makes it impossible to conclude from these data alone that transplantation itself caused the better outcomes, and the authors explicitly state that the findings are preliminary and hypothesis-generating rather than evidence of a definitive transplant benefit.

This restraint is a model of scientific communication in the era of rare-disease reporting. With a total of 14 patients spread across multiple institutions and time periods, any apparent association between a treatment and survival could easily reflect selection bias rather than biological or therapeutic truth. The authors emphasize that neither an independent adverse prognostic effect of the fusion nor a transplant benefit can be established from their data, and they call for larger multicenter cohorts to properly define the prognostic significance and therapeutic implications of this rare fusion. In the interim, their study serves a different but no less important purpose: it alerts the diagnostic community to the existence and recognizable face of this leukemia subtype.

The practical implications for pathology and hematology laboratories are concrete. When flow cytometry reveals the RAM-like pattern — bright aberrant CD7 and CD56, dim CD45, dim or absent CD38, and loss of HLA-DR — in a case of AML, particularly one where conventional cytogenetics fails to identify a defining abnormality, the authors suggest that RNA-based fusion testing should be considered. Targeted RNA sequencing is increasingly available in major centers and can be performed on bone marrow or peripheral blood specimens, offering a molecular diagnosis even when chromosomes look deceptively normal under the microscope. Early molecular identification matters because it may eventually guide risk stratification, transplant decision-making, and enrollment into clinical trials tailored to fusion-positive leukemias.

The study was approved by the Ruijin Hospital Ethics Committee of Shanghai Jiao Tong University School of Medicine, with informed consent waived owing to the retrospective, anonymized design. Funding came from the National Natural Science Foundation of China (grant 82400187). The work reflects the collaboration of investigators at Ruijin Hospital, the Affiliated Hospital of Qingdao University, and Ren Ji Hospital in Shanghai. As an open-access publication, it is available to clinicians and researchers worldwide who may encounter similar cases.

For now, ZMYND11::MBTD1-rearranged AML remains a rare entity — too rare for any single center to accumulate the numbers needed for definitive prognostic answers. But rare leukemias are precisely where international collaboration, registry building, and molecular screening pay the greatest dividends. This study takes an early and necessary step: it names the disease, describes its face under the flow cytometer, points out where the standard genetic test fails, and offers preliminary outcome data that will inform, rather than conclude, the conversation. As more cases are identified through RNA sequencing and shared across borders, the true clinical personality of this fusion-driven leukemia — and whether transplantation genuinely improves its course — will come into view.