Reduction of cancer cell quantity and viability in autologous blood salvaged from patients with liver cancer: efficacy of intraoperative cell salvage coupled with leukocyte depletion filtration Authors: Jinhuo Wang, Zhenzhou Li, Yong Cheng, Laiwei You, Zhanyu Cheng, Mandi Wu, Yuming Sun, Lei Chen, Jianrong Guo Published: 2026 ABSTRACT AbstractObjectiveThis in vitro study aims to evaluate the effects of intraoperative cell salvage (ICS) combined with leukocyte depletion filtration (LDF) on hepatocellular carcinoma (HCC) cell number and viability in salvaged autologous blood from patients undergoing liver cancer surgery.MethodsTwenty patients undergoing open radical resection for primary liver cancer with ICS were enrolled in the study.Blood samples of 20 ml each were procured at three distinct stages: from the surgical field(S1),post ICS treatment(S2),and post ICS treatment combined with leukocyte depletion filter(LDF) filtration(S3).Within these 20-ml blood samples,10 ml underwent cancer cell enrichment procedures, followed by identification and enumeration of cancer cells using immunofluorescence staining.The remaining 10 ml of blood samples were cultured for a duration of three weeks, with subsequent assessment of cell viability using immunofluorescence techniques subsequent to enrichment procedures.ResultsHCC cells were identified in samples obtained from S1(19/20),S2(18/20),and S3 (16/20) without a significant difference in the detection rate(P > 0.05).However, a significant reduction in HCC cells count was observed in samples from S2 and S3 when compared to S1(P < 0.05). Notably, no statistically significant differences were noted between HCC cells counts in samples from S2 and S3(P > 0.05). Following three weeks of culture, optical microscopy revealed the presence of liver cancer cell clusters exclusively in S1 samples, while such clusters were absent in samples from S2 and S3. Further examination under fluorescence microscopy indicated the presence of epithelial-mesenchymal hybrid-type HCC cells(S1: 400, S2: 14) and mesenchymal-type HCC cells(S1: 100, S2: 21) in both S1 and S2 samples, whereas no HCC cells were detected in S3 samples.Specifically, HCC cells in S1 samples manifested as liver cancer cell clusters, whereas such clusters were notably absent in samples from S2 and S3.ConclusionFollowing treatment with ICS alone or in combination with LDF (ICS-LDF), both the number and viability of hepatocellular carcinoma (HCC) cells in salvaged autologous blood were markedly decreased, with no cell cluster formation, which lowered the risk of salvaging cancer cells to some extent. Nevertheless, LDF failed to completely remove HCC cells from all samples, and its filtration efficiency may be compromised when the HCC cell number exceeds a certain threshold.Keywords: Cancer cells, Cancer metastasis, Intraoperative cell salvage (ICS), Leukocyte depletion filter (LDF), Liver cancer Introduction In recent years, concerns regarding blood scarcity and transfusion safety have increased [1]. Despite advancements in liver cancer resection techniques leading to reduced perioperative blood loss, patients undergoing such procedures still face the risk of anemia ranging from 20% to 30% [2]. This anemia stems from various factors such as the rich blood supply of the liver, underlying abnormalities such as liver dysfunction and blood coagulation disorders in patients with liver cancer, as well as complications arising from treatments like chemotherapy or radiotherapy, which can impair erythropoietin response and induce bone marrow suppression. Liver cancer surgery, characterized by tumors with robust blood supply and significant intraoperative blood loss, consumes approximately 30% of the total perioperative blood requirements [3]. The escalating severity of blood shortage is evident. Allogeneic blood transfusion, while beneficial for enhancing circulation and oxygen delivery, carries substantial risks due to its immunosuppressive effects, closely linked to tumor recurrence. Reports indicate that allogeneic blood transfusion, whether administered preoperatively, intraoperatively, or postoperatively, significantly accelerates tumor recurrence and raises mortality rates [4]. Various modes of allogeneic blood transfusion, encompassing whole blood, red blood cells, platelets, and plasma components, exhibit a dose-response relationship with tumor recurrence [5]. Transfusion-related immunomodulation (TRIM) emerges as a significant contributor to malignant tumor metastasis, recurrence, or death, both directly through soluble factors and indirectly through immunomodulatory effects [4, 6]. Allogeneic blood transfusion potentially fosters proliferation and dissemination of cancer cells within minimal residual lesions, while also increasing concentrations of pro-inflammatory cytokines and prostaglandin E in the tumor microenvironment and systemic circulation, thereby tilting the immune balance toward suppression. Numerous cellular mechanisms contribute to impaired immune function and subsequent evasion of tumor cells, including diminished natural killer (NK) cell activity, reductions in cluster of differentiation 4 (CD4+), CD8 + T cells, and B lymphocyte proliferation, as well as compromised maturation and antigen presentation of inducible regulatory T cells (iTregs) and dendritic cells [7]. Macrophages, pivotal regulators of immune responses, face dysfunction-induced adverse reactions, further complicating immune surveillance against tumors [8]. Consequently, autologous blood transfusion emerges as an indispensable strategy for preserving patients’ well-being. Autologous blood transfusion encompasses three main categories: stored autologous blood transfusion, diluted autologous blood transfusion, and intraoperative cell salvage (ICS), among which ICS stands out as a widely used approach. ICS involves collection of blood from the surgical field, subsequent processing involving anticoagulation, filtration, washing, and concentration, culminating in reinfusion back into the patient [9]. This method not only conserves precious blood resources but also effectively reduces the complications associated with allogeneic blood transfusion [10]. Nevertheless, controversies have persisted regarding the use of ICS in patients with malignant tumors, stemming from concerns about potential tumor metastasis resulting from the reinfusion of malignant tumor cells. In 1986, malignant tumors were designated as contraindications for ICS [11]. However, the notion proposing that ICS-mediated reinfusion of tumor cells may contribute to tumor metastasis and recurrence remains primarily theoretical and lacks robust evidential support. Extensive literature review has revealed that ICS does not heighten the postoperative recurrence rate among patients with liver cancer or those undergoing liver transplantation [12]. Conversely, ICS has been associated with enhanced survival rates, liver synthesis function, postoperative prognosis in patients with liver cancer, and reduced mortality rates, contrary to traditional assumptions. ICS has been used in combination with radiation therapy or LDF filters in several countries, and patients who received ICS during cancer surgery had similar or better outcomes compared to patients with or without allogeneic transfusions [13]. Patients undergoing cancer surgery for which ICS is applicable are divided into three categories: patients with localized cancer without metastasis but undergoing surgery in a cancer-free area, in which case ICS can be used; patients with cancer and metastasis, in which case ICS can be used because metastasis has already occurred; and patients with localized cancer and in which the surgery is a resection of the cancerous area, in which the applicability is still controversial. The application of ICS in patients in this scenario deserves further exploration in recent years with the increase in blood supply constraints and a renewed awareness of the safety of blood transfusion [14]. In this study, surgical field blood was collected from hepatocellular carcinoma (HCC) patients, enriched, and cultured, followed by in vitro evaluation of the effects of intraoperative cell salvage (ICS) alone or combined with leukocyte depletion filtration (LDF) on HCC cell number and viability in salvaged autologous blood. Our objective is to establish a theoretical foundation for the safe perioperative use of ICS in patients with malignant tumors. Materials and methods This study included 20 patients who underwent open radical resection of primary liver cancer at the Eastern Hepatobiliary Surgery Hospital Affiliated to the Naval Medical University between December 2021 and February 2022. Inclusion criteria encompassed patients with confined tumor, absence of distant metastasis, and preoperative Barcelona Clinic Liver Cancer (BCLC) staging A–C. Exclusion criteria comprised of patients with hematologic disorders, those who had undergone radiotherapy, chemotherapy, or tumor intervention therapy, patients with intraoperative blood loss less than 200 ml, or those who declined participation in the study. Adherence to the Declaration of Helsinki was ensured throughout all study procedures. Approval for this study was obtained from the ethics committee of the hospital (Approval No.: glll-44, 2021, Public Hospital Ethics Review), and all patients provided informed consent. The blood salvage collecting kit was installed before the surgery in accordance with the instructions provided by the Cell Saver Elite+ blood recycle system (Haemonetics, USA). Negative pressure within the range of 120 to 150 mmHg (1 mmHg = 0.133 kPa) was applied for blood aspiration. At the initiation of surgery, the blood collection container was initially pre-filled with 200 ml of heparin saline solution (30 IU/ml). The drip rate of the anticoagulant heparin saline was adjusted to approximately 5 ml/min, or 2 ml/min in case of mild bleeding. Upon exposure of the liver, the collection of all blood emanating from the surgical field was initiated. This collection process was continued until the primary surgical procedures were completed and hemostasis was achieved. Subsequently, the collected blood underwent standardized processing, including washing, centrifugation, and recovery. Recovered blood was considered acceptable if it exhibited a hematocrit within the range of 30%–60% and a volume exceeding 100 ml. The recovered blood was filtered using a LDF filter with a pore diameter of 40 μm (The employed LDF filter was Model DD-GLF-10, manufactured by Shanghai Dediag Biotechnology Co., Ltd.). The LDF filter was attached to the outlet of the red blood cell collection bag. It is important to emphasize that all blood salvage activities were strictly designated for research purposes and were not intended for transfusion. Blood samples of 20 ml each were collected from the surgical field (S1), from blood in the blood collection bags subjected to ICS treatment without LDF filtration (S2), and from blood subjected to a combined treatment of ICS and LDF (ICS-LDF) (S3). Within the 20-ml blood samples, 10 ml underwent HCC cell counting, while the remaining 10 ml underwent cell culture. The blood was preserved in vacuum blood collection tubes containing ethylenediaminetetraacetic acid dipotassium salt dihydrate (K2EDTA). All blood samples were refrigerated at 4 ℃ and processed for enrichment within two hours of collection. The identification and counting of HCC cells involved the separation of cells based on their physical properties, with larger HCC cells being retained by filtration membranes. Immunofluorescence staining and identification used 4’,6-diamidino-2-phenylindole (DAPI), vimentin, cytokeratin/glypican-3 (CK/GPC3), and CD45 antigen (leukocyte common antigen). Liver cancer cells were identified as cells exhibiting CK+/GPC3+/Vimentin+/DAPI+/CD45- characteristics. Lymphocytes were identified as CD45 + cells. Epithelial-mesenchymal hybrid-type HCC cells were classified as cells displaying CK+/GPC3+/vimentin+/DAPI+ characteristics. Mesenchymal-type HCC cells were categorized as cells revealing vimentin+/DAPI+ characteristics. Epithelial-type HCC cells were designated as cells with CK+/GPC3+/DAPI+ characteristics. Staining of the slide was conducted using a CircucelNA-10 fully automatic special immunofluorescence staining machine (Shanghai Dediag Biotechnology Co., Ltd.), and scanned using a BX43F fluorescence microscope automatic scanner (OLYPUS Corporation, Japan) equipped with HDStar191030 software (four-color fluorescence) for counting purposes. For the HCC cell viability assay, 5 aliquots of HCC cells, each containing more than 20 HCC cells per 10 ml, were extracted from the S1 samples. After the removal of red blood cells, the remaining cells in the aliquots were resuspended and passed through a 70 μm sieve for filtration. The enriched HCC cells were subsequently seeded onto low-adhesion cell culture plates (Greiner bio-one: 657970) for cultivation. Concurrently, HCC cells from the S2 and S3 samples were cultured synchronously for three weeks. After cultivation, the cells were collected in centrifuge tubes and stained following the instructions provided for the CircuCelNA-10 fully automatic special staining machine and cell filter, using DAPI, vimentin, CK/GPC3, and CD45 for staining. The membrane from the cell filter was then removed, and slides were prepared and sealed. The area containing the cell suspension was scanned using the NovaScan-01 fluorescence microscope (Shanghai Dediag Biotechnology Co., Ltd.). During the scanning process, observations were made to determine the presence or absence of cell clusters, cell types, and counting. Liver cancer tissue specimens excised during the surgery were promptly delivered to the pathology department within 30 min. Two experienced clinical pathologists staged and classified the liver cancer tissue specimens in accordance with the Edmondson grading system, and documented their pathological characteristics. All the data collected in this study were analyzed using SPSS 26.0 software. Normally distributed measurement data were expressed as mean±standard deviation (SD), while non-normally distributed measurement data were expressed as median (interquartil range), and the comparisons were examined by Student-t test and Mann-Whitney test (non-parametric distribution). HCC cell counts in different blood samples were compared using the Kruskal-Wallis test followed by Dunn’s post hoc test. The categorical data were expressed as n (%), and the differences between the two groups were examined by chi-square analysis or Fisher’s exact test. P < 0.05 was considered statistically significant. Results Twenty patients were included, 8 females and 12 males, and the results of specific patient demographic and tumor characteristics are shown in Table 1. Table 1Patient demographic and tumor characteristicsPatient demographic and tumor characteristicsHCC (n = 20)Age (median [IQR])62 (48 ~ 76)Female (%)12 (75%)Preoperative AFP level (median [IQR]), ug/dL17.0 (1.6 ~ 19.0)Estimated blood loss (median [IQR]), ml475 (300 ~ 1000)Maximum tumor diameter (median [IQR]), mm55 (30 ~ 130)Number of blood transfusions (%)8 (40%)moderate polarization (%)6 (40.0)vascular invasion (%)6 (30%)tumor rupture (%)0 (0.0)Peripheral Infiltration (%)0 (0.0)BCLC classification(A/B/C)1/13/6Edmondson classification(Ⅰ/Ⅱ/Ⅲ)4/12/4Note: 1. Edmondson’s classification: Grade I: the best differentiated cells of this type, with the cancer cells arranged in the form of thin beams; Grade II: the nuclei of the cells are larger and densely stained, with abundant eosinophilic cytoplasm; Grade III: the nuclear staining of the cancer cells is more significant than that of Grade II, with the emergence of tumor giant cells; Grade IV: the tumor cells are the least differentiated, with the nuclei of the tumor cells being strongly densely stained, occupying the majority of the cells, with the cytoplasm often lacking, and lacking the connection between the cells2. BCLC classification: Stage A (single tumor, or less than three tumors with diameter less than 3 cm); Stage B (single tumor with diameter more than 5 cm; or 2 ~ 3 tumors, at least one lesion with diameter more than 3 cm; or any tumor with diameter more than 3 tumors); Stage C (vascular invasion or extrahepatic metastasis) Patient demographic and tumor characteristics Note: 1. Edmondson’s classification: Grade I: the best differentiated cells of this type, with the cancer cells arranged in the form of thin beams; Grade II: the nuclei of the cells are larger and densely stained, with abundant eosinophilic cytoplasm; Grade III: the nuclear staining of the cancer cells is more significant than that of Grade II, with the emergence of tumor giant cells; Grade IV: the tumor cells are the least differentiated, with the nuclei of the tumor cells being strongly densely stained, occupying the majority of the cells, with the cytoplasm often lacking, and lacking the connection between the cells 2. BCLC classification: Stage A (single tumor, or less than three tumors with diameter less than 3 cm); Stage B (single tumor with diameter more than 5 cm; or 2 ~ 3 tumors, at least one lesion with diameter more than 3 cm; or any tumor with diameter more than 3 tumors); Stage C (vascular invasion or extrahepatic metastasis) The total HCC cell counts in S1, S2, and S3 samples were 1111, 479, and 313, respectively. Regarding the presence of HCC cells in S1, S2, and S3 samples, 19 (95%), 18 (90%), and 16 (80%) samples were found to contain HCC cells, respectively. There was no statistically significant difference in the detection rate of HCC cells among samples S1, S2, and S3 (P > 0.05). However, compared to S1 samples, both S2 and S3 samples exhibited a lower HCC cell count (P < 0.05). Additionally, there was no statistically significant difference in HCC cell count between S2 and S3 samples (P > 0.05, refer to Table 2; Fig. 1). Table 2Comparison of HCC cell count and detection rate among S1, S2, and S3 samples (n = 20)Blood samplesHCC cell count[n, M (Q1, Q3)]Detection rate of HCC cells (%)S1 samples61.00(26.00,79.50)95S2 samples14.50(5.25,41.75)90S3 samples3.50(1.25,25.50)80Z/value17.852.26P value< 0.0010.322Note: S1 represents blood samples from the surgical field. S2 represents the blood samples in the blood collection bags following ICS treatment and without LDF filtration. S3 represents the blood samples after LDF filtration. HCC signifies hepatocellular carcinoma Comparison of HCC cell count and detection rate among S1, S2, and S3 samples (n = 20) Note: S1 represents blood samples from the surgical field. S2 represents the blood samples in the blood collection bags following ICS treatment and without LDF filtration. S3 represents the blood samples after LDF filtration. HCC signifies hepatocellular carcinoma Immunofluorescence staining images of DAPI, vimentin, and Ck/GPC3 in HCC cells of S1 samples (×400 magnification). Note: a represents epithelial-mesenchymal hybrid HCC cells; b represents mesenchymal tumor cells; and c represents epithelial-type HCC cells. DAPI stands for 4’,6-diamidino-2-phenylindole; Vimentin stands for vimentin; Ck/GPC3 stands for cytokeratin/glypican-3; and S1 samples are surgical blood specimens.’ Following three weeks of culture, all S1 samples were observed to contain HCC cells under an optical microscope. These cells displayed adherent growth to the container wall, maintaining intact morphology, clear boundaries, a transparent appearance, and were present in large numbers, forming numerous liver cancer cell clusters. In contrast, no obvious liver cancer cell clusters were found in the S2 and S3 samples (see Fig. 2). We conducted cell counting in all fields of view under a fluorescence microscope, revealing 400 vs. 14 and 100 vs. 21 epithelial-mesenchymal hybrid-type HCC cells and mesenchymal-type HCC cells in S1 and S2 samples, respectively. No HCC cells were observed in S3 samples. In the S1 samples, HCC cells were primarily found within liver cancer cell clusters, whereas in the S2 and S3 samples, no liver cancer cell clusters were detected (refer to Fig. 3). Before ICS treatment, HCC cells exhibited a strong tendency towards clustering, with all tumor cells transitioning into epithelial-mesenchymal hybrid-type HCC cells during culture. Following ICS treatment, HCC cells displayed a preference for single-cell growth. After ICS-LDF treatment, no HCC cell proliferation was detected. Growth of HCC cells in samples S1, S2, and S3 following three weeks of cultivation (×40 magnification under an optical microscope). Note: Growth of HCC cells in samples S1, S2, and S3 following three weeks of cultivation (×40 magnification) Immunofluorescence staining images of HCC cells in S1, S2, and S3 samples after three weeks of cultivation (×400 magnification). Note: S1 represents blood samples from the surgical field. S2 represents the blood samples in the blood collection bags following ICS treatment and without LDF filtration. S3 represents the blood samples after LDF filtration. The acronym DAPI stands for 4’,6-diamidino-2-phenylindole. Vimentin stands for vimentin. Ck/GPC3 stands for cytokeratin/glypican-3. Representative images of samples with detectable cells are shown. Note that no cells were detected in some S3 specimens Discussion Patients with liver cancer frequently present with concurrent anemia and coagulation dysfunction attributed to disruptions in hepatic function. Liver cancer surgery, characterized by its complexity and potentially substantial intraoperative hemorrhage, often requires multiple blood transfusions. Numerous studies have underscored allogeneic blood transfusion as an independent risk factor for adverse postoperative outcomes in patients with liver cancer [15]. Consequently, optimization of perioperative blood management has a high importance in the context of liver cancer resection surgery. Prior research revealed that ICS has been implicated as a potential contributor to tumor recurrence and metastasis. However, recent retrospective analyses have not found an association between ICS and cancer recurrence. Therefore, in this study, we assessed the quantity and viability of liver cancer cells subsequent to ICS or a combined ICS-LDF regimen. The objective of this trial was to assess the safety profile of ICS when administered to patients with malignant tumors. Due to the heterogeneous nature of HCC and its propensity for epithelial-mesenchymal transition (EMT) during metastasis, the detection of circulating tumor cells (CTCs) using epithelial cell adhesion molecule (EpCAM), a marker for epithelial cells, may be limited. Following the enrichment of HCC cells, cell identification was conducted through immunofluorescence staining based on vimentin and glypican-3 (GPC3). Notably, GPC3 exhibits a high expression level in over 85% of HCC tumor specimens but is not expressed in normal liver tissue, liver tissues with benign liver diseases, liver cirrhosis, or hepatitis. Also, GPC3 expression is associated with prognosis [16]. Vimentin, a 57 kDa intermediate filament protein, serves as a principal cytoskeletal component of mesenchymal cells and represents a primary marker for mesenchymal liver cancer cells [17]. Previous studies have reported a high positivity rate of MXR7 (GPC3) cDNA in tumor tissues of patients with HCC [18]. Therefore, in this study, DAPI, vimentin, Ck/GPC3, and CD45 were used as immunofluorescent staining markers to facilitate the identification of HCC cells. The filtration method, with an effective enrichment rate ranging between 85% and 100%, is currently the most efficacious approach for tumor cell enrichment [19]. Significantly, this method is independent of antigen expression, thereby ensuring the normal viability of tumor cells and laying the groundwork for subsequent culture and passage. While most studies use blood specimen volumes ranging between 4.0 and 7.5 ml, we used a volume of 10 ml, enabling the collection of a higher number of tumor cells and thus enhancing the clinical relevance of our findings. Also, we concurrently detected both mesenchymal and epithelial cancer cells. Our findings demonstrate that LDF effectively eliminates liver cancer cells from blood samples containing fewer than 6 liver cancer cells per 10 ml. However, the efficacy of LDF diminishes when confronted with higher cell counts, indicative of a threshold beyond which the combination of ICS-LDF may not effectively filter liver cancer cells. However, LDF significantly reduces the risk of reintroducing liver cancer cells. The primary contention surrounding the application of ICS for patients with HCC revolves around safety concerns—specifically, whether it is safe to transfuse autologous blood salvage containing HCC cells back to the patients. The current focal point of research lies in the clearance of HCC cells and the reduction of their quantity in autologous blood salvage. LDF, a filtration device resembling a membrane with a pore diameter of 40 μm, is primarily used for removing white blood cells from the bloodstream. However, debates persist regarding its impact on the prognosis of patients with malignant tumor undergoing ICS treatment. Based on findings from a clinical retrospective study, the transfusion of recycled blood following ICS-LDF treatment did not escalate the risk of cancer recurrence or cancer-related mortality. Notably, the survival rate of patients in the ICS-LDF group surpassed that of the control group [20]. However, upon further observation of the application of LDF in patients with malignant tumors, diverse outcomes emerge. The application did not demonstrate a conspicuous advantage of LDF in reducing tumor metastasis or recurrence [21]. In the context of this study, HCC cells were detected in both blood samples from the surgical field and post-ICS processing, with their total count decreasing from 1111 in the former to 479 in the latter. Subsequently, blood samples after ICS processing underwent LDF filtration, resulting in the detection of HCC cells in 16 blood samples, with their total count decreasing from 479 to 313. Notably, despite a substantial decrease in the total count of HCC cells following ICS and LDF treatment, no significant difference was observed between S1, S2, and S3 samples. Therefore, although ICS-LDF may not entirely eliminate HCC cells, it represents a viable option for reducing the HCC cell count. In blood samples following ICS treatment, a significant proportion of tumor cells, accounting for 62%, incurred lethal damage, resulted in notable morphological changes. These cells exhibited a loss of viability and a reduced capacity to form metastatic foci [22, 23]. Although detectable, the residual tumor cells likely experienced a loss of proliferative potential [24]. Additionally, tumor cell clusters, known for their enhanced immune resistance and metastatic capabilities when compared to individual tumor cells, have drawn considerable attention in recent years. The formation of circulating tumor microemboli (CTM) primarily hinges upon molecular adhesion mechanisms. CTM predominantly expresses cell-cell adhesion proteins such as PLak hemoglobin, intercellular adhesion molecule-1 (ICAM1), and proteins associated with tight junctions and bridging grains. Additionally, CTCs exhibiting stem cell traits can secrete the adhesion protein ICAM1, thereby promoting CTM formation. Diminished ICAM1 expression significantly impedes CTC migration and CTM formation, ultimately restraining the ability of CTCs to infiltrate blood vessels and establish distant tumors. In this study, both untreated HCC cells and those subjected to ICS-LDF treatment were cultured to assess the impact of ICS-LDF on HCC cell viability. Results indicate that untreated HCC cells thrived in Dulbecco’s Modified Eagle Medium (DMEM), exhibiting robust proliferative and adhesive abilities, tight cell-cell connections, and distinct clustering, with all cells undergoing EMT after three weeks of culture. Conversely, HCC cells treated with ICS or ICS-LDF displayed single-cell growth, reduced cell adhesion, and less pronounced clustering. These findings align with recent clinical retrospective analyses, revealing that ICS-LDF does not elevate the risk of metastasis or recurrence in patients with malignant tumors [25]. Drawing on data from international clinical studies, it is postulated that the beneficial effect of ICS-LDF on the prognosis of patients with malignant tumors may be attributable to the following mechanisms: Most patients with primary tumors have tumor cells in the circulation early in life, and tumor tissue can shed large numbers of tumor cells [26, 27]. Post-surgical monitoring indicates a significant increase in the concentration of peripheral blood tumor cells, a rise several-fold greater than that observed following ICS treatment alone [28]. However, the number of tumor cells in salvaged autologous blood following ICS processing is almost negligible compared to pre-existing circulating tumor cells. Most patients with primary tumors have tumor cells in the circulation early in life, and tumor tissue can shed large numbers of tumor cells [26, 27]. Post-surgical monitoring indicates a significant increase in the concentration of peripheral blood tumor cells, a rise several-fold greater than that observed following ICS treatment alone [28]. However, the number of tumor cells in salvaged autologous blood following ICS processing is almost negligible compared to pre-existing circulating tumor cells. Throughout the ICS process, residual malignant tumor cells undergo shear forces such as attraction, centrifugation, filtration, and washing, resulting in their predominance as individual cells. Upon entry into the bloodstream, these individual cells are swiftly cleared, rendering them less likely to survive. Throughout the ICS process, residual malignant tumor cells undergo shear forces such as attraction, centrifugation, filtration, and washing, resulting in their predominance as individual cells. Upon entry into the bloodstream, these individual cells are swiftly cleared, rendering them less likely to survive. When compared to allogeneic blood transfusion, ICS has a lesser impact on the immune function of the body. Upon entering the bloodstream, tumor cells are readily cleared by the immune system. The remaining small proportion of tumor cells, even if not eliminated, are more likely to enter a dormant state in a location distant from the original tumor and are subsequently cleared by NK cells. Consequently, their capacity to cause recurrence and other complications is greatly diminished. Thus, the likelihood of tumor metastasis recurrence attributable to ICS is exceedingly low. When compared to allogeneic blood transfusion, ICS has a lesser impact on the immune function of the body. Upon entering the bloodstream, tumor cells are readily cleared by the immune system. The remaining small proportion of tumor cells, even if not eliminated, are more likely to enter a dormant state in a location distant from the original tumor and are subsequently cleared by NK cells. Consequently, their capacity to cause recurrence and other complications is greatly diminished. Thus, the likelihood of tumor metastasis recurrence attributable to ICS is exceedingly low. Consistently, several studies have indicated the favorable outcomes of ICS and LDF for patients with malignancies. Kumar et al. [29, 30] demonstrated in metastatic spine tumor surgery (MSTS) that ICS-LDF effectively eliminates tumor cells from salvaged autologous blood. Furthermore, autologous blood transfusion in this setting is not only safe but may also confer superior overall survival and a lower risk of tumor progression compared with allogeneic transfusion, thereby providing both laboratory and clinical evidence to support the broader application of this technique in MSTS and musculoskeletal oncological surgery. Frietsch et al. [31] conducted a meta-analysis of safety of ICS in cancer surgery and concluded that outcome was similar or better in patients receiving ICS during cancer surgery compared to patients with allogeneic blood transfusion or nontransfused patients. Tan et al. [32] analyzed the presence of tumor cell in HCC liver transplantation autologous blood using microfluidics technology and found that the risk of tumor cell reintroduction using intra-operative blood salvage autotransfusion in HCC liver transplantation patients can be reduced with a single LDF. This study showed that either ICS alone or ICS-LDF could lead to notable reduction in both the quantity and viability of liver cancer cells within salvaged autologous blood. The above findings sufficiently confirmed safety of ICS + LDF. However, this study has several limitations. Only 20 patients were enrolled, and the study was designed as a preliminary exploratory investigation using a within-subject self-control approach. The assessment primarily relied on tumor cell counts and immunofluorescence staining to infer the decline in tumor cell viability, without performing functional assays or in-depth mechanistic studies on the tumor cells. Additionally, the salvaged autologous blood was used exclusively for experimental analysis and not reinfused into patients; consequently, the clinical outcomes of enrolled patients were unaffected. Future large-scale randomized controlled clinical trials and basic mechanistic studies are needed to evaluate the clinical efficacy of salvaged autologous blood following ICS combined with LDF. Conclusion Collectively, both ICS alone and ICS combined with LDF significantly reduced the number and viability of residual hepatocellular carcinoma (HCC) cells in salvaged autologous blood, with these cells existing primarily as single cells rather than forming clusters. Although LDF may not consistently eliminate all HCC cells—particularly when cell counts exceed a certain threshold—its use nonetheless reduces reinfusion risk to some extent. The potential theoretical risks of ICS in HCC patients must be carefully weighed against the risks of allogeneic blood transfusion. Further preclinical and clinical studies are warranted to evaluate the feasibility of ICS-LDF as an intraoperative strategy for massive hemorrhage in liver cancer patients. Such research is essential to provide objective and compelling evidence supporting the clinical translation and broader application of ICS-LDF. Acknowledgements We would like to acknowledge the hard and dedicated work of all the staff that implemented the intervention and evaluation components of the study. Abbreviations Intraoperative cell salvage Leukocytes depletion filter hepatoma carcinoma cell Barcelona Clinic Liver Cancer 4’,6-diamidino-2-phenylindole Cytokeratin Glypican 3 microvascular invasion Circulating Tumor Microemboli Intercellular Cell Adhesion Molecule-1 Dulbecco’s Modified Eagle’s Medium Authors’ contributions Conception and design of the research: Zhenzhou Li ; Jianrong GuoAcquisition of data: Jinhuo Wang; Zhenzhou Li; Yong Cheng; Laiwei You; Zhanyu Cheng; Mandi WuAnalysis and interpretation of the data: Jinhuo Wang; Zhenzhou LiStatistical analysis: Jinhuo Wang; Zhenzhou Li ; Laiwei You; Zhanyu Cheng; Mandi Wu;Yuming Sun; Lei ChenObtaining financing: Jianrong Guo; Yong ChengWriting of the manuscript: Jinhuo Wang; Zhenzhou LiCritical revision of the manuscript for intellectual content: Yuming Sun; Lei Chen; Jianrong GuoAll authors read and approved the final draft. Funding National Natural Science Foundation of China(No.82470236); Shanghai Pudong New Area health system key subject group construction project(PWZxq2022-05). Data availability All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author. Declarations Ethics approval and consent to participate This study was conducted with approval from the Ethics Committee of Shanghai Gongli Hospital(Approval number:2022-44). This study was conducted in accordance with the declaration of Helsinki. Written informed consent was obtained from all participants. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes The original article has been updated: Funding section has been updated. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Zhenzhou Li and Yong Cheng contributed equally to this work. Change history 9/16/2026 The original article has been updated: Funding section has been updated. References Associated Data Data Availability Statement All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.