Correspondence on: "Targeting the HERV-K102 envelope elicits pyroptosis and represents a novel therapeutic strategy for acute myeloid leukemia" Authors: Maajid Mohi Ud Din Malik, Rajesh Prasad Jayaswal Published: 2026 ABSTRACT AbstractZong et al. reported that CRISPR-Cas9-mediated deletion of the HERV-K102 envelope (K-Env) induces apoptosis and pyroptosis in acute myeloid leukemia cells via S100A9/NLRP3 inflammasome activation. We examine four methodological concerns: the absence of evidence using a pathway-specific inhibitor to distinguish pyroptosis from secondary necrosis; the lack of a mechanistic rescue experiment confirming S100A9 as the proximal driver; the non-specificity of primers at the mRNA level; and the inability of the immunocompromised xenograft model to capture immune evasion by K-Env. The prognostic and therapeutic implications of the study require qualification. Dear Editor, The recent study by Zong et al. [1] represents a substantive contribution to knowledge on the biology of human endogenous retroviruses (HERVs) in hematological malignancy. By demonstrating that CRISPR-Cas9-mediated deletion of the HERV-K102 envelope (K-Env) suppresses cell proliferation in acute myeloid leukemia (AML), induces apoptosis, and triggers pyroptosis via the S100A9/NLRP3 inflammasome axis, the authors open an avenue that had received comparatively little attention. This work is relevant due to the growing interest in HERV-K both as a diagnostic biomarker and an immunotherapeutic target [2, 3]. However, some issues related to the methodology need to be discussed before these data can be integrated into a specific therapeutic paradigm. First, the prognostic analysis deserves consideration. Analysis of HERV-K102 expression, as measured in TCGA-derived AML datasets, showed no significant association with overall survival (p = 0.6381). This finding is reported in a supplementary figure, with little discussion by the authors. However, this is not a side observation. Thus, if elevated K-Env expression provides no independent prognostic value in the moderately powered cohort analysis (n = 132 with adequate follow-up), then the biological role of HERV-K102 as a driver of AML progression may be more contextual than what is implied in the report. However, antibody-based therapeutic strategies against HERV-K102 suggested in earlier studies [3, 4] rely on an invariant tumor-promoting role. The patient survival data do not support that assumption and merited a direct mention instead of being relegated to supplemental material. Characterization of pyroptosis is the most significant mechanistic assertion of the report, and this is where the chain of evidence is at its weakest. LDH release, caspase-1 activity and GSDMD cleavage are all increased following pyroptosis, but none of these markers is exclusive to it. The authors themselves recognize that all three markers are also present in apoptosis and secondary necrosis. The verification step through pharmacological inhibition using pathway-specific inhibitors (e.g. caspase-1 or disulfiram for GSD81 pore formation) was skipped. Therefore, the observed profile of cell death following K-Env knockout cannot be unambiguously attributed to canonical pyroptosis in the absence of this class of chimeric discrimination. This gap in the evidence is important, since pyroptosis and apoptosis have distinct biological characteristics, and this has therapeutic implications involving increased inflammatory sequelae and risk of a “cytokine storm”, which the authors highlight as a safety concern [5]. A causal link between K-Env loss and S100A9 upregulation is presumed, not proven. At that time, S100A9 is the most significantly upregulated transcript in both THP-1 and MV4-11 cells, as evidenced by RNA-seq after K-Env removal. This is also demonstrated at the protein level using Western blot. However, no rescue experiment was conducted. A restoration of K-Env expression in knockout cells to confirm the reversal of S100A9 induction and pyroptosis occurrence would have greatly strengthened the causal argument. Moreover, no experiments to specifically knock down or inhibit S100A9 in a K-Env knockout background were performed to demonstrate that S100A9 is indeed a selective trigger of NLRP3 inflammasome activation rather than a co-regulated bystander. The NLRP3 pathway is activated in myeloid cells by a wide variety of stressors [6], and the observed transcriptomic activation may have been partially mediated by general cellular stress induced by the CRISPR editing process itself. The use of cells transduced with a control vector as a negative control addresses some concerns but does not fully compensate for the lack of a mechanistic rescue design. The specificity of the primers is a technical limitation of the study not sufficiently considered by the authors in their interpretation. HERV-K102 mRNA quantification was performed using qRT-PCR primers derived from a universal HERV-K sequence common to type 1 (HERV-K103, with a 292 bp deletion in the 5’ env region) and type 2 (HERV-K107, with an intact envelope gene) [7] proviruses, although they have different biological functions. As HERV-K102 is a type-2 provirus and the knockout was targeted toward a single locus, the data on mRNA-level knockdown shown in Fig. 2 A of the report by Zong et al. (qRT-PCR of HERV-K102 mRNA expression in MV4-11 and THP-1 cells transduced with mock, gRNA1, and gRNA2 constructs) might not be specific to K-Env but could instead represent multi-loci suppression, as envelope sequences overlap. Protein-level validation by Western blot, which the authors appropriately prioritize, partially compensates for this limitation, but the discordance between gRNA1 and gRNA2 at the mRNA level, with that of gRNA1 showing no statistically significant reduction in one cell line, is not adequately explained. NSG mice, a severely immunocompromised strain lacking T, B, and NK cells, was used for the in vivo validation. This model is appropriate for xenograft engraftment of human-derived AML lines but unsuitable for evaluating the immunomodulatory consequences of K-Env ablation. HERV-K envelope proteins are increasingly recognized as mediators of immune evasion, and the transmembrane subunit of HERV-K Env has been shown to induce CD8 + T cell anergy in AML [8]. De-repression of immune recognition following K-Env knockout, which is potentially the most therapeutically exploitable effect of this manipulation, cannot be observed in an immunocompromised host. Tumor growth suppression is documented in Fig. 5 of the report through bioluminescent imaging of NSG mice intravenously inoculated with MV4-11-Cas9-mock or MV4-11-Cas9-K-Env knockout cells, with doxycycline-induced K-Env ablation, monitored over 14 days. In this context, this could be solely attributed to cell-intrinsic effects (proliferation arrest and pyroptosis), and cannot be generalized to an immunocompetent setting. This distinction matters considerably when proposing K-Env as an immunotherapy target. A broader concern is related to the safety profile of therapeutic pyroptosis induction. Pyroptosis, by its nature, is a lytic and pro-inflammatory form of cell death. Widespread activation across the tumor mass releases large quantities of interleukin (IL)−1β, IL-18, and HMGB1 into the systemic circulation [5, 9]. In AML – a disease frequently complicated by cytopenias, infection, and organ dysfunction – the risk of triggering a syndrome resulting from macrophage activation or cytokine release is clinically non-trivial. The therapeutic window between tumor cell pyroptosis and systemic inflammatory toxicity was not explored in this study and must be defined before K-Env ablation strategies can be responsibly translated to clinical settings. These reservations do not diminish the conceptual value of the work. The demonstration that an endogenous retroviral envelope protein suppresses inflammasome activity in leukemic cells is mechanistically interesting and, to the credit of the authors, the cell line panel is reasonably diverse, the inducible in vivo model is methodologically sound, and the transcriptomic dataset is rich enough to support follow-up studies. In order to further substantiate these findings, future studies should: (i) confirm pyroptosis specificity through experiments with pharmacological inhibitors; (ii) address confounding pathway hierarchy by incorporating S100A9-directed knockdown in the K-Env knockout background; (iii) validate the in vivo data in a humanized or syngeneic murine model for AML able to recapitulate adaptive immune responses; and (iv) delineate locus-specific quantification of HERV-K102 transcription using targeted long-read sequencing or digital droplet PCR approaches. However, HERV-K102 still holds promise as a clinical target to treat AML [10]. More broadly, whether the mechanism driving this effect is primarily pyroptotic, apoptotic, or immunomodulatory, and whether it can be exploited without exacerbating systemic inflammation, are the unanswered questions of the current dataset. Acknowledgements None Authors’ contributions MMDU conceived the correspondence, conducted the critical appraisal, and prepared the initial draft. RPJ reviewed the mechanistic arguments, contributed to the in vivo modelling discussion, and revised the manuscript for scientific accuracy. Both authors approved the final version submitted. Funding No external funding was received for the preparation of this correspondence. Data availability No datasets were generated or analysed during the current study. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.