Harnessing hemostasis to transform surgical scars into CAR-T immunological niches

Surgical resection remains one of the most powerful tools in the treatment of solid tumors, yet it is also a moment of profound biological vulnerability. For many aggressive malignancies, including pancreatic ductal adenocarcinoma and triple-negative breast cancer, surgery rarely equates to cure (1). Despite achieving complete macroscopic tumor removal, microscopic residual disease at surgical margins and occult disseminated tumor cells frequently remain undetected, ultimately giving rise to local recurrence or distant metastases that manifest months to years following the initial resection (2). Current adjuvant therapies aim to address this risk, but their limitations are well known: Systemic chemotherapy and radiotherapy lack specificity and impose cumulative toxicity, while immune checkpoint inhibitors benefit only a subset of patients and often fail in immunologically “cold” tumors (3). The postoperative setting thus represents a narrow but critical window during which durable, spatially precise immune intervention could have outsized clinical impact.

Chimeric antigen receptor (CAR)-T cell therapy has revolutionized the treatment of hematologic malignancies, delivering deep and durable remissions in diseases once considered incurable (4). In contrast, comparable success in solid tumors has remained elusive. This disparity reflects a convergence of biological and technical barriers. Dense stromal architecture restricts T cell infiltration, immunosuppressive tumor microenvironments dampen effector function, and systemic administration results in inefficient tumor localization (3, 5). These challenges are further compounded by the scarcity of truly tumor-exclusive antigens. Many promising solid tumor targets, including B7-H3 (CD276), are markedly overexpressed in malignant tissues yet retain low-level expression in normal organs, creating a narrow therapeutic window in which systemically circulating CAR-T cells can induce on-target, off-tumor toxicity (6). To address these limitations, multiple innovative strategies have been advanced into early-phase clinical trials (7). Regional delivery approaches have sought to bypass trafficking barriers and limit systemic exposure. For example, intrapleural administration of mesothelin-targeted CAR-T cells in pleural mesothelioma and malignant pleural effusions (NCT02414269, NCT04577326) achieved detectable tumor infiltration and early signals of clinical activity, while intracranial delivery of IL13Rα2-directed CAR-T cells in recurrent glioblastoma produced transient tumor responses with manageable local toxicity (NCT02208362) (8, 9). In parallel, efforts to enhance systemic CAR-T efficacy have focused on modifying the immunosuppressive tumor microenvironment. Combination strategies pairing CAR-T cells with immune checkpoint blockade—including pembrolizumab with mesothelin-targeted CAR-T cells in pancreatic cancer (NCT06054308) have demonstrated feasibility but yielded only modest improvements in clinical outcomes, accompanied by manageable yet notable immune-related adverse events (10). Given persistent safety concerns, clinical development has also emphasized toxicity mitigation. Approaches include cautious dose-escalation protocols starting at low cell doses (as low as 10⁶ cells) to reduce the risk of cytokine release syndrome, prophylactic administration of tocilizumab to block IL-6 signaling (11), and incorporation of inducible safety switches such as iCasp9 or RQR8 in GD2- and CD19-targeted CAR-T trials to enable rapid elimination of CAR-T cells in the event of severe toxicity (12–14). Despite these advances, overall clinical efficacy in solid tumors has remained disappointing. Across trials in pancreatic and breast cancers, objective response rates typically remain below 25%, with low median progression-free survival rates—underscoring the persistent need for strategies that improve both spatial precision and therapeutic durability (15).

Wang et al. report a conceptually innovative solution that harnesses hemostatic biology to address this challenge (16) (Fig. 1). By conjugating tumor-specific CAR-T cells to platelets via biorthogonal click chemistry, the authors exploit platelets’ innate wound-homing properties to redirect immune effector cells precisely to surgical sites—transforming postoperative scars from passive remnants of tumor removal into active immunological battlegrounds. This strategy recognizes a fundamental limitation plaguing CAR-T therapy in solid tumors: Current approaches effectively target the right antigen but fail to concentrate therapeutic cells in the right anatomical location (17). The platelet–CAR-T platform introduces spatial precision as a core design principle, leveraging platelets as biological navigators that respond to tissue injury through conserved adhesion mechanisms, including binding to exposed collagen and von Willebrand factor at damaged endothelium (18–20). Crucially, in the postsurgical context, these platelet accumulation sites coincide precisely with the anatomical locations harboring residual microscopic disease—the surgical margins where cancer cells are most likely to persist.

Fig. 1.

Platelet-conjugated CAR-T cells against post-surgical tumor recurrence. The figure was created using BioRender.com.

By chemically conjugating autologous platelets to CAR-T cells using bioorthogonal click chemistry, the authors create a hybrid immune cell that retains both platelet wound tropism and CAR-mediated tumor recognition (16). In postsurgical mouse models of pancreatic cancer and triple negative breast cancer (TNBC), these platelet–CAR-T conjugates demonstrated enhanced accumulation at resection sites and increased infiltration into residual tumor tissue compared with conventional CAR-T cells. This targeted redistribution is not a trivial improvement in biodistribution; it directly addresses one of the major inefficiencies of systemic CAR-T therapy, in which the vast majority of infused cells fail to reach sites of disease.

Importantly, the contribution of platelets extends beyond physical guidance. Once activated within the inflammatory postoperative environment, platelets release a complex mixture of chemokines, cytokines, and growth factors capable of modulating immune responses. Transcriptomic and functional analyses in this study reveal that platelet-derived signals reshape CAR-T cell state, reducing markers of exhaustion while enhancing programs associated with tissue infiltration, persistence, and effector function. Platelet-conjugated CAR-T cells proliferated more robustly at tumor sites and produced higher levels of key effector cytokines, translating into sustained local immune pressure. One of the most intriguing observations is the promotion of memory T cell differentiation. Platelet–CAR-T-treated animals showed enrichment of central and effector memory subsets, suggesting that this strategy supports long-term immune surveillance rather than transient cytotoxic bursts. In the postsurgical context, this distinction is critical. The goal is not merely to eliminate residual tumor cells immediately after surgery but to maintain vigilance against future tumor outgrowth and metastatic seeding. By fostering immunological memory at the site most prone to recurrence, platelet–CAR-T cells may offer protection that extends well beyond the immediate perioperative period.

These biological advantages translated into compelling therapeutic outcomes. Across multiple preclinical models, platelet–CAR-T therapy significantly prolonged survival, reduced local recurrence, and suppressed metastatic dissemination. In both pancreatic cancer and TNBC models, efficacy exceeded that of conventional CAR-T therapy. Particularly notable was the finding that half-dose platelet–CAR-T achieved outcomes comparable to full-dose standard CAR-T treatment. This dose-sparing effect carries important implications for safety, manufacturing, and access. CAR-T–associated toxicities, including cytokine release syndrome, often scale with cell dose, and reducing the number of infused cells could improve tolerability in postoperative patients while easing manufacturing constraints.

Safety remains a central consideration. The humanized mouse model experiments provide crucial preliminary safety data. Full-dose platelet–CAR-T and conventional CAR-T treatments induced transient toxicity signals—decreased activity, ruffled fur, and temporary blood pressure drops—during the first week posttreatment, with subsequent recovery to baseline. Importantly, half-dose platelet–CAR-T largely avoided these acute effects while maintaining efficacy. Interleukin-6 levels, a key biomarker of cytokine release syndrome, showed temporary elevation followed by normalization, suggesting manageable inflammatory responses rather than life-threatening cytokine storms. Histological examination of major organs revealed no significant pathology, and liver/kidney function markers remained within normal ranges across all treatment groups. The humanized mouse models employed here represent important steps toward clinical translation.

Hu et al. report a conceptually innovative solution that harnesses hemostatic biology to address this challenge (16) (Fig. 1).

From a translational perspective, platelet conjugation undeniably adds manufacturing complexity to an already intricate CAR-T production process. The workflow requires coordinating autologous platelet isolation with T cell engineering, executing multistep chemical modifications under sterile conditions and maintaining reproducibility across GMP-compliant batches—challenges that could impact production timelines, costs, and regulatory approval pathways. However, the cell therapy manufacturing landscape is evolving rapidly, with automated closed-system platforms increasingly capable of handling complex multistep processes while ensuring sterility and reproducibility. Beyond its immediate translational promise, this work offers a broader conceptual advance. It reframes platelets—often viewed as passive participants or even facilitators of cancer progression—as programmable allies in immunotherapy. More broadly, it demonstrates the power of integrating physiological processes such as hemostasis and wound healing into the design of immune therapies. Rather than circumventing the body’s innate wound healing machinery, this approach harnesses it to achieve therapeutic precision. By hijacking hemostasis to turn surgical scars into immunological niches, platelet-engineered CAR-T cells offer a compelling vision for the future of postsurgical cancer therapy. This work represents an important step toward that goal, illuminating new paths for preventing recurrence at one of cancer’s most vulnerable moments.