The investigators combined single-cell transcriptomics with surface proteomics (CITE-seq) to profile malignant populations in paediatric acute myeloid leukaemia. Their integrated analytic framework separated malignant from non-malignant cells and extracted unbiased gene programmes. The cohort showed three recurrent leukaemia stem cell (LSC) states: primitive (P-LSC), monocytic (M-LSC) and erythroid-like (E-LSC). These states carry distinct transcriptional signatures and immunophenotypes. Patients frequently harboured more than one LSC state at diagnosis, with proportions varying by individual sample. The study team observed that these are stem-cell states rather than discrete, patient-specific clones—conserved programmes that recur across the dataset.
Quantitative analysis of surface antigens across the defined LSC states revealed that no single antigen covered all three states. The authors reported CLEC12A expression across both P-LSC and M-LSC compartments, but noted that CLEC12A was largely absent from E-LSC cells. In contrast, CD84 provided complementary coverage: it marked cells in the P and M groups and served as the primary target within the E-LSC population. The investigators therefore proposed a dual-targeting strategy combining CLEC12A and CD84 to increase LSC coverage across heterogeneous cell states. The profiling described relative antigen expression patterns; the source did not report any therapeutic dosing, antibody affinities, or engineered receptor constructs.
The authors compared antigen expression on malignant LSC states with expression on normal haematopoietic stem cells. Data indicated that the CLEC12A/CD84 combination offered broad LSC coverage while sparing healthy haematopoietic stem cells, according to the reported surface-proteome measurements. This selective expression pattern informed the authors’ prioritisation logic for candidate immunotherapy targets in the paediatric setting. The manuscript did not include functional cytotoxicity assays or clinical safety datasets in patients; therefore the reported sparing is derived from single-cell antigen-profiling rather than from in vivo or clinical exposure outcomes.
To assess robustness, investigators validated the CLEC12A/CD84 antigen strategy across independent single-cell and bulk cohorts. Validation analyses confirmed the recurrent presence of the three LSC programmes and the complementary antigen distribution reported in the discovery dataset. The authors stated that both single-cell and bulk-level data supported CLEC12A and CD84 as a paired strategy to increase antigen coverage across LSC states in paediatric AML. The source did not specify the composition, size, or clinical characteristics of the independent cohorts in this summary; those cohort details appear in the full manuscript supplementary material as referenced by the authors.
The study team framed their results as a paediatric-specific framework for antigen prioritisation, intended to inform rational design of next-generation CAR T-cell combinations. They proposed that combining targets to capture multiple conserved stem-cell programmes could mitigate antigen escape driven by LSC heterogeneity. The report stops short of presenting engineered CAR constructs, in vitro cytotoxicity against primary samples, or clinical trial outcomes. The source did not report any completed or ongoing clinical trials testing CLEC12A/CD84 dual targeting. Specific operational details for translating this antigen pair into a CAR T clinical protocol—such as construct design, co-expression strategies, safety switches, or manufacturing considerations—were not provided in the article.
The authors acknowledged heterogeneity within paediatric AML at the stem-cell programme level and provided molecular and antigenic maps; still, several practical questions remain. The source did not report functional efficacy or safety data from in vivo models or human trials, so antigen-expression correlations cannot be equated with clinical efficacy. The report did not include dosing, pharmacodynamics, or toxicity endpoints for cellular therapies targeting CLEC12A or CD84. The manuscript did not present long-term patient follow-up data tied to antigen expression patterns. Unanswered items that remain relevant for clinical translation include whether dual targeting reduces relapse driven by E-LSC dominance, how antigen density thresholds influence CAR activity on each LSC state, and what safety margins will be required to avoid on-target, off-leukaemia effects. These specific clinical endpoints and regulatory milestones were not addressed in the source and will require prospective preclinical and early-phase clinical evaluation.
The authors declared no competing interests in the preprint. The posted record and supplementary materials provide the analytic workflows and cohort-level results underlying these conclusions; investigators and clinical developers will need to expand on functional testing, safety assessment, and trial design before patient-directed immunotherapies based on these antigen choices can be implemented.