Introduction

Bladder cancer is classified into two groups based on the depth of local invasion: non-muscle-invasive bladder cancer (NMIBC), which does not invade the muscle layer, and muscle-invasive bladder cancer (MIBC), which does. Patients with NMIBC can be treated by surgical removal of the bladder region via transurethral resection of the bladder tumor (TURBT). Radical cystectomy with complete organ removal is the standard treatment for patients with MIBC1. Total cystectomy is highly invasive, with postoperative changes in urinary function inevitably leading to a significant reduction in the QOL. There are many cases in which surgery is not performed because of clinical problems, such as old age or complications. In addition, the outcomes of total cystectomy are not satisfactory, with a 5-year survival rate of 50–72%, although there is a trend toward improvement with adjuvant chemotherapy2,3.

As an alternative to surgery, cryoablation has spread rapidly in recent years as a minimally invasive treatment. Cryoablation is a method in which a frozen area, called an ice ball, is created around a frozen needle to necrose a tumor. Recently, cryoablation has been utilized for the treatment of various types of cancer4,5.

However, cryoablation has not been used in clinical practice to treat bladder cancer. A validation study using a mouse model of orthotopic bladder cancer is required to demonstrate whether cryoablation is a safe and effective treatment option for bladder cancer. Several animal studies have shown that transurethral cryoablation can be safely performed on normal bladder tissue6,7,8, whereas there are few reports concerning extravesical cryoablation in animal models of bladder cancer. Cryoablation exerts a synergistic effect by exposing cancer-associated antigens to the immune system without thermal denaturation, thereby promoting immune activation and enhancing anti-tumor responses9. By leveraging this property, cryoablation for metastatic MIBC may help reduce metastatic lesions.

In this study, we investigated cryoablation targeting tumor sites in the urinary bladder, with a particular focus on both local tumor control and immunostimulatory effects, supporting the potential for bladder preservation. We believe this experimental model closely mirrors clinical scenarios and may contribute to future innovations in bladder cancer therapy.

Materials and methods

Initiation of bladder tumors in mice

Six-week-old female C57BL/6 mice (Shimizu Laboratory Supplies Co. Ltd., Kyoto, Japan) were maintained in a temperature-controlled environment with 12:12 light/dark cycle and provided with unrestricted access to drinking water supplemented with 0.05% N-butyl-N-(4-hydroxybutyl) nitrosamine (BBN, Tokyo Chemical Industry Co Ltd, Tokyo, Japan) in order to induce bladder carcinogenesis. BBN has previously been used as a carcinogen to efficiently induce bladder cancer in laboratory animals10. All animal procedures were approved by the Animal Care and Use Committee of the Kyoto Prefectural University of Medicine and performed in accordance with the Guidelines for Animal Care of the Kyoto Prefectural University of Medicine. This study is reported in accordance with the ARRIVE guidelines (https://arriveguidelines.org).

Cryoablation procedure

Cryoablation surgery was performed 26–28 weeks after initiating carcinogen administration. BBN-treated mice were anesthetized using either isofluorane inhalation or an intraperitoneal injection of a mixed anesthetic solution (0.75 mg/kg domitor, 4 mg/kg midazolam, and 5 mg/kg betorphol). A lower midline abdominal incision was made to expose the bladder. The mice were then randomly assigned to either the cryoablation or sham (control) group. In the cryoablation group, a metal rod (approximately 4 × 4 mm, stainless steel) pre-cooled in liquid nitrogen at − 196 °C was applied externally to the bladder as a cryoprobe. The cryoablation procedure consisted of two cycles of freezing and thawing, each involving 30 s of freezing followed by 30 s of thawing. In the sham group, animals were exposed to the open abdomen for the same duration as the cryoablation group without probe contact. The cryoablation group and sham group treated with BBN were termed “Treated” and ”Non-treated” groups, respectively. By comparing these two groups, we evaluated the efficacy and safety of cryoablation as a therapeutic approach for bladder cancer.

Validation of cryoablation efficacy

This procedure is illustrated in Fig. S4A. This experiment aims to clearly demonstrate the efficacy of cryoablation under identical conditions by comparing differences in adjacent cells within bladder tissue with and without surgery. The bladder was bisected by inserting a dividing line from the apex to the urethra. The urethra was preserved, and cryoablation was performed on only one of the divided bladder fragments. We ensured the anesthesia remained effective for 8 h to prevent the animal from experiencing pain. Inhalation of isoflurane anesthesia was administered approximately every hour, with visual confirmation of breathing and reflex movements. The bladder was removed 8 h postoperatively and cryopreserved. We validated the efficacy of cryoablation by comparing tissue changes between treated and untreated areas of the bladder using hematoxylin and eosin (H&E) staining.

Validation of cryoablation safety

This procedure is illustrated in Fig. 1B. After anesthesia, cryoablation was performed on the apex of the bladder and the wound was closed. At several time points (3 h and POD 1, 3, 7, 14, and 28), the mice were sacrificed and their bladders were harvested. The safety of cryoablation was validated by assessing tissue changes in the bladder after the procedure. The maximum and minimum diameters and depths of congestion and tumor volume were measured postoperatively using calipers.

Body weight, water intake, urine output, and urinary frequency were measured at each time points shown in figures. All assessments were also performed between groups with and without BBN treatment (Fig. S3).

Histopathology

Excised bladders were cryopreserved, sectioned at a thickness of 5 μm using a cryostat, and stained with H&E to enable microscopic evaluation of cellular and tissue morphology.

Voided stain-on-paper (VSOP) analysis

The VSOP method, a simple and non-invasive method for measuring small volumes of urination in mice, was used to analyze the voiding volume and urination frequency before and after surgery (Fig. S2)9,11.

Voiding function in control groups, both BBN-treated and non-BBN-treated mice, was monitored using the VSOP method from 20 weeks after the start of rearing until the time of surgery. In the operated groups, mice were evaluated using VSOP on POD 1, 2, 3, 7, 14, and 28. Mice were individually housed in metabolic cages for 2 h with free access to food and water. Filter papers (FILTER PAPER QUALITATIVE ADVANTEC 240 mm; Tokyo Roshi Kaisha, Ltd., Japan) were placed beneath the mesh floors to collect urine spots. The stained areas on the filter papers were transferred onto separate sheets and quantified using Adobe Photoshop 2020 (Adobe, San Jose, CA, USA). To calculate voided volumes, a calibration curve was established using known volumes (2–500 µl) of normal saline applied using a micropipette (PIPETMAN®: M&S Instruments Inc., Osaka, Japan). Additionally, the correlation between voided volume and body weight was analyzed both before and after surgery.

Immunohistochemistry staining

BBN-treated (n = 21) and BBN-untreated (n = 18) mice were divided into four groups; cryoablation non-treated group (POD -) and cryoablation treated groups (POD 1, 7, and 28), respectively (eight groups with four or five mice per group).

For immunohistochemical analysis, frozen specimens were cut into 5 μm sections using a cryostat. Consecutive tissue sections were obtained along the long axis of the bladder. Slides were thawed at room temperature and fixed in 10% neutral buffered formalin. Antigen retrieval was achieved by immersion of tissue sections in 0.01 mol/L (pH 6.0) citrate, and endogenous peroxidase activity was blocked with H2O2.

Immunohistochemical staining was performed using a Dako LSAB + System-HRP kit (Agilent, CA, United States). Briefly, the slides were incubated overnight at 4 ˚C with mouse monoclonal antibodies against CD4 (ab183685; Abcam, Bristol, UK) and CD8 (ab209775; Abcam). Negative controls were defined by the omission of primary antibody from the same microslide section. After washing, secondary antibodies were added to each section and slides were exposed to 3, 3-diaminobenzidine chromogenic mixture and counter-stained with H&E.

Quantitative analysis of lymphocyte infiltration was performed on immunostained samples using automated cell counting based on pixel values within selected areas in Adobe Photoshop 2020. The slide field of view was divided to ensure the entire tumor area at the site of the cryoprobe was included, and CD4⁺ and CD8⁺ immunopositive cells were counted using a 100× optical microscope. The positive index was calculated as the ratio of marker-positive cells to the total number of cells in each field, and the mean value was determined across multiple fields per slide. This analysis was repeated twice using the same method for consistency. Finally, the positive index was normalized to the corresponding index obtained from negative control samples processed identically but without primary antibodies.

Statistical analysis

Results are shown as means ± standard error (SE). Differences were analyzed using 2-tailed Student’s t-test with p < 0.05 considered statistically significant.

Results

Physiological effects of BBN administration in mice

This study is designed to assess the efficacy and safety of cryoablation therapy using a mouse model of MIBC (Fig. 1A). In the first phase, we established a mouse model of MIBC. BBN has been reported to induce bladder cancer in approximately 90% of mice around 20 weeks post-administration10. BBN treatment induced MIBC in 86.7% of mice in our initial trials (T2 stage or higher in Table S1, Fig. S1). Body weight, volume of water consumed, and urination patterns determined using the VSOP method were investigated for each mouse from 20 weeks after the first BBN administration to cryoablation surgery (Fig. S2). Body weights tended to be slightly lower in BBN-treated mice than in non-BBN-treated mice (Fig. S3A), whereas water consumption was higher in BBN-treated mice (Fig. S3B). The total volume of urination was significantly lower and the frequency of urination was higher in BBN-treated mice (Fig. S3C,D).

Therapeutic efficacy of cryoablation on bladder cancer cells

The efficacy of cryoablation was evaluated by comparing cell morphology with and without cryoablation within the same bladder tissue. Cryoablation was performed on a tissue fragment of the bisected bladder, as shown in Fig. S4A. This experiment is an independent technique from other experiments because the cryoprobe is applied to the epithelial/mucosal side of the bladder. Eight hours after surgery, the bladder was dissected for observation. Macroscopic observation of frozen sites in the cryoablation group revealed reddish discoloration and stasis (Fig. 2A). Stasis may reflect vascular injury.

H&E staining revealed cancer cells at the bladder margins on both sides of the bisected bladder in the control group that did not undergo cryoablation surgery (Fig. S4B). In the cryoablation-treated group, cancer cells were found at the bladder margins on the non-treated side (Fig. S4B). In contrast, the loss of smooth muscle cell nuclei and decreased staining were prominent on the treated side (Fig. S4B).

The disappearance of cell nuclei and the formation of tissue cavities following cryoablation surgery are indicative of cell death and cytoplasmic shedding, respectively. Therefore, cryoablation appears to be effective in destroying cancer cells in the bladder.

Validation of the safety of cryoablation

In this study, the observation period was not extended beyond 1 month because the mice were used for downstream analysis. Cryoablation-induced mortality was not recognized in either BBN-untreated or BBN-treated mice during the 1-month observation period. In the cryoablation non-treated group, there were no mortality cases within the 1-month observation period, similar to the cryoablation treated group. Several mortality cases occurring after surgery were found to be similar in frequency between the cryoablation treated group and the non-treated group—suggesting they were likely due to anesthesia or abdominal procedures. Therefore, these events were excluded from the results. Microscopic observation revealed obvious stasis at the apex of the bladder along the ablation area until POD 7 (Fig. 2A). The degree of congestion caused by cryoablation was similar from 3-h after surgery to POD 3, followed by a decreasing trend up to POD 7, and almost no stasis after POD 14 (Fig. 2A).

Next, histopathological findings involving bladder tissue at 3-h after cryoablation were compared between the ablated and non-ablated areas. H&E-stained images revealed a significant decrease in, and loss of, smooth muscle cell nuclei and unstained voids at ablation sites (Fig. 2B). The histopathological findings of H&E staining of the bladder tissue over time after cryoablation were similar to those of macroscopic observations; the surgical areas were clear until POD 7 and unclear at POD 14 (Fig. 2C). In the acute phase, between 3-h after cryoablation and POD 3, there was a marked loss of smooth muscle cell nuclei and staining of the bladder muscle layer at ablation sites (Fig. 2C). The cavities appeared to diminish POD 7 and had almost completely disappeared after POD 14 (Fig. 2C).

To closely examine the bladder muscle layer at the surgical site, the distribution of α-SMA (α-smooth muscle actin) was assessed using immunohistochemistry. α-SMA is widely used as a marker for smooth muscle cells constituting the bladder muscle layer12. α-SMA expression at the ablation sites was barely detectable on POD 7 but increased by POD 14 and 28 (Fig. 2D). This pattern likely reflects removal of the cryonecrotic tissue by the immune system, followed by replacement with adjacent tissue.

To quantitatively represent changes in tumor volume following cryoablation, the tumor shrinkage rate compared to the preoperative tumor volume (mm3) was shown temporally (Fig. 2E, Table S2). Compared to the POD 0 group, significant tumor shrinkage was recognized in the POD 7, 14, and 28 groups (Fig. 2E). Meanwhile, tumor weight could not be quantified as it was technically difficult to completely surgically remove the invasive tumor from the mouse bladder.

To determine the physiological effects of cryoablation, body weight, water consumption, urinary output, and frequency of urination were examined. There were no significant differences in any of the parameters between the cryoablation treated and non-treated groups (Fig. 3). Body weight decreased until POD 2, but increased on POD 3 (Fig. 3A). There was no difference in body weight change between the cryoablation treated group and non-treated group at any time point from immediately after surgery to POD 28 (Fig. 3A). Water consumption was lower in the treated group than in the non-treated group on POD 2 and 3, but no difference was found between the two groups at other time points (Fig. 3B). The volume and frequency of urination examined using the VSOP method (Fig. S2). Urine output was lower in the treated group compared to the non-treated group on POD 7, leading to a marked increase in urine output in the treated group between POD 7 and 14 (Fig. 3C). This observation suggests an association with tissue changes following cryoablation. Frequency of urination showed no significant difference between the treated group and the non-treated group at any time point (Fig. 3D).

Immune activity after cryoablation

To investigate the immune activity caused by cryoablation, we employed immunostaining for CD4+, CD8+, and T cell surface proteins that determine the properties of T cells. The mean CD4+ and CD8+ indexes at the cryoablation site within the bladder were compared between the cryoablation treated group (POD 1, 7, and 28) and the non-treated group (POD) (Fig. 4). The results showed that in BBN-treated MIBC model mice, CD4+ and CD8+ cells were significantly increased in the POD 28 group compared to the POD− group (Fig. 4A). The same study was performed in BBN-untreated tumor negative mice, which showed increased expression of only CD4+ cells in the POD 28 group compared to the non-treated group (Fig. 4A). In the spleen at POD28, CD4+ T cell numbers showed no significant difference between the treated and non-treated groups (Fig. S5). These results may suggest that a local immune response rather than a systemic antitumor immune response is occurring at POD28.

Discussion

This study performed extravesical cryoablation in a mouse model of orthotopic bladder cancer and showed that (i) it can effectively eliminate cancer cells, (ii) it can be performed safely, and (iii) it has no perioperative mortality. All experiments in this study concluded after a one-month observation period; it is considered necessary to accumulate further long-term survival data and other information to evaluate the therapeutic effect. Although the extravesical approach used in this study does not directly replicate current clinical practice, it does provide a robust and reproducible experimental platform to evaluate the oncological, functional, and immunological impact of bladder cryotherapy. Our data support the feasibility of bladder-sparing cryotherapy and lay the groundwork for the future development of transurethral cryoablation systems, potentially in combination with immune checkpoint inhibitors for the treatment of MIBC.

In this study, cryoablation was performed using an extravesical approach via laparotomy. This strategy was intentionally selected based on technical and experimental considerations inherent to murine orthotopic bladder cancer models. In mice, the small bladder volume, thin bladder wall, and limited urethral caliber make stable and reproducible transurethral cryoprobe placement technically challenging and highly variable. An extravesical approach allowed direct visualization of bladder, precise targeting of tumor-bearing regions, and controlled application of freezing cycles, thereby enabling reliable assessment of tissue injury, safety, and immunological responses. Although an extravesical approach is not routinely used for primary bladder cancer treatment in current clinical practice, it is occasionally employed in selected surgical or hybrid procedures, particularly in experimental or salvage settings. Importantly, the aim of this study was not to propose laparotomy-based cryoablation as a definitive clinical technique, but rather to establish proof-of-concept evidence that localized cryoablation can safely eradicate bladder tumor tissue while preserving bladder function and inducing immune activation.

From a translational perspective, the findings of this study are highly relevant to the future development of transurethral cryoablation devices. Several studies have demonstrated the feasibility of transurethral cryoablation in large animal models, such as porcine and canine bladders, using balloon-based systems6,7,8. The present murine model provides mechanistic insight into tissue responses, functional outcomes, and immune activation following focal cryoablation, which are essential prerequisites for adapting cryotherapy to transurethral clinical applications. Then, ice-ball formation and freezing depth were controlled by using a standardized metal probe size and fixed freeze-thaw cycles. Although minor variability in the extent of congestion and tissue necrosis was observed, the depth of cryo-induced injury was largely confined to the muscle layer and showed consistent temporal resolution across animals. This reproducibility suggests that, even in small-animal models, controlled cryoablation can achieve predictable tissue effects when procedural parameters are standardized. Additionally, tumor location is expected to be a critical determinant of cryoablation feasibility and efficacy. Lesions located at the bladder apex or dome are likely more amenable to focal ablation, whereas tumors near the trigone or ureteral orifices may pose a higher risk of functional complications. This study primarily targeted accessible tumor regions to minimize confounding factors related to urinary outflow obstruction. Future studies incorporating spatial tumor mapping and image-guided cryoablation will be required to address location-specific limitations.

Our results demonstrated that only long-term bladder cancer model mice survivors (POD 28) exhibited a significant increase in CD4+ and CD8+ cell expression compared with pre-operative cases (Fig. 4). The increase in CD4+ and CD8+ in POD28 is particularly notable in BBN-treated mice, indicating that this immune response is especially pronounced in tumors. However, a slight increase in CD4+ was also detected in BBN-untreated mice, indicating that it is not completely tumor-specific. Nevertheless, there is a clear correlation between tumor shrinkage and immune activity, as tumor volume decreases and CD4+/CD8+ increases over time after cryoablation (Figs. 2E and 4A). In addition, CD4+ population in the spleen at POD28 show no difference between cryoablation treated and non-treated groups (Fig. S5). These results suggest that at 28 weeks post-tissue damage, a local immune response likely occurs rather than a systemic antitumor immune response. Considering that these T cells were only detected in the damaged tissue after 28 weeks, they appear to be tissue-resident memory T cells (TRM) rather than circulating T cells. TRM cells are known to localize in the bladder epithelium and tumor tissues, where they exhibit hallmark features such as CD69 and CD103, enabling them to remain anchored in the tissue and maintain prolonged immune surveillance13. These cells respond to tumor-specific antigens by producing effector cytokines like IFN-γ and TNF-α, which stimulate cytotoxic activity and enhance the recruitment of other immune cells, such as CD8+ T cells and NK cells, to the tumor site14. Future studies such as cytokine profiling and marker quantification may clarify the immune response system induced by cryotherapy. The challenges in immunotherapy for bladder cancer are predicting drug efficacy, discovering biomarkers, and improving the clinical efficacy of combination immunotherapy. Novel cancer immunotherapy, which combines cryotherapy with conventional immunotherapy such as intravesical BCG therapy and immune checkpoint inhibitors (ICIs), may be a new treatment for MIBC beyond conventional chemotherapy.

Fig. 1
Fig. 1
Full size image

This study’s flow and methodology: (A) Flowchart of this study. (B) Schematic for cryoablation of the bladder apex for safety verification experiments.

Fig. 2
Fig. 2
Full size image

Pathological bladder changes in safety validation experiments: (A) Macroscopic findings of the bladder after cryoablation. The illustration corresponds to a macroscopic finding taken 3-h after cryoablation. The ablation site is clear, as shown in red in the illustration. POD, postoperative day. (B) Comparison of bladder muscle layer tissue in the ablation treated area and non-treated area. The illustration corresponds to an H&E-stained image of the whole bladder of 3-hours postoperatively. The area enclosed by the black dotted line indicates the ablation area. The right-side two images are enlarged views of the boxed areas in H&E-stained imaging of the whole bladder (a, ablation treated area; b, non-treated area). Scale bars in H&E-stained images indicate 100 μm (left image) and 25 μm (right images). (C) H&E-stained images of the bladder after cryoablation. Scale bars indicate 50 μm. (D) Distribution of α-SMA in the bladder muscle layer after cryoablation visualized after immunohistochemistry. Brown staining indicates α-SMA-positive cells. Scale bars indicate 50 μm. (E) Tumor shrinkage rate at each postoperative time point relative to preoperative tumor volume. A comparative test with the POD 0 group showed significant tumor shrinkage in the POD 7, 14, and 28 groups. *p < 0.0001.

Fig. 3
Fig. 3
Full size image

Influence of cryoablation on physiological effects in mice: (A) Body weight changes after cryoablation surgery. (B) Changes in amount of water consumption. (C) Voiding volume. (D) Frequency of urination. *p < 0.05.

Fig. 4
Fig. 4
Full size image

Immunohistological findings of CD4+ and CD8+ T cells after cryoablation: (A) The positive index of CD4+ and CD8+ cells. The number of CD4+ and CD8+ cells was counted using cryoablation-treated samples immunostained with anti-CD4 and anti-CD8 antibodies. Positive index ​​is represented as the mean of positive cell numbers in images of each group. A comparative test with the each of POD - group. *p < 0.05, ** p < 0.01. (B) A few cases of immunostaining using antibodies against CD4 and CD8. Left images: cryoablation non-treated (POD−) group of BBN-treated mice, right image: POD28 group of BBN-treated mice. Scale bars indicate 25 μm.