N-methyl-N-nitrosourea-induced forestomach neoplastic lesions formed by week 13 following a single intraperitoneal administration in rasH2 mice Authors: Takamasa Numano, Ryo Inoue, Tatsuya Miyake, Masahiko Yasuda, Misa Mochizuki, Takashi Katsumata, Izumi Owada, Chiyoko Nishime, Taichi Yamamoto, Masami Suzuki Published: 2026 ABSTRACT Abstract CByB6F1-Tg (HRAS)2Jic mice (rasH2 mice) are widely used in 26-week short-term carcinogenicity assessments. To ensure phenotypic stability and carcinogenic sensitivity, a 26-week N-methyl-N-nitrosourea (MNU)-induced carcinogenic susceptibility monitoring study was conducted for quality assurance. The study included (i) comprehensive whole-organ monitoring after colony renewal, and (ii) simplified annual monitoring on production colonies focusing on forestomach tumor formation as the most sensitive endpoint. Retrospective analysis of our 26-week historical data indicated a high incidence of forestomach tumors in mice that were moribund or sacrificed before 13 weeks. Therefore, we compared the 13- and 26-week observation periods to evaluate whether 13 weeks was sufficient for simple carcinogenicity monitoring. RasH2 mice were intraperitoneally administered MNU (75 mg/kg) and observed for 13 and 26 weeks. Both groups exhibited small and large forestomach nodules, including hemorrhagic lesions. The corresponding histopathological features were squamous cell papillomas and carcinomas. The incidence of squamous cell papilloma or carcinoma was 100.0% in both sexes at 13 weeks, which was comparable to that at 26 weeks (100.0% in males and 93.3% in females). These results demonstrated that MNU-induced forestomach tumor can be reliably detected within 13 weeks, indicating that a shortened monitoring period of 13 weeks is sufficient to monitor carcinogenic susceptibility for quality assurance using genotoxic carcinogens in rasH2 mice.Keywords: rasH2 mouse, carcinogenic susceptibility, monitoring system, N-methyl-N-nitrosourea (MNU), forestomach tumor Introduction The CByB6F1-Tg (HRAS)2Jic mouse (rasH2 mouse) model, which harbors multiple copies of the human c-Ha-ras proto-oncogene, was established by the Central Institute for Experimental Medicine and Life Science (CIEM) and is a widely accepted tool for short-term studies of the carcinogenicity of pharmaceuticals1–3. Recognized for its appropriate sensitivity to both genotoxic and non-genotoxic carcinogens, the rasH2 mouse model was validated through extensive multi-site collaborative studies organized by the International Life Sciences Institute and its Health and Environmental Sciences Institute from 1997 to 20014, 5. These validation efforts included comparisons with conventional two-year rodent bioassays, which demonstrated that the rasH2 mouse model could reliably detect carcinogenic potential within a much shorter study duration. Following this, data accumulated from studies conducted over time at different facilities consistently demonstrated the high reliability and stability of the rasH2 mouse model, supporting its value as a cornerstone in modern carcinogenicity study strategies6–9. RasH2 mouse model has recently been specifically recommended in the International Council for Harmonisation (ICH) S1B guideline for assessing pharmaceuticals carcinogenic potential10, and use of rasH2 mice in carcinogenicity studies has increased accordingly11, 12. The quality and genetic integrity of rasH2 mice were rigorously maintained to support their continued use in regulatory science. The murine model is currently being bred at the following tightly regulated production sites: CLEA Japan (Fujinomiya, Shizuoka, Japan) and Taconic Biosciences (Germantown, NY, USA; Tornbjerg, Denmark)13. To prevent genetic drift and to maintain consistency across breeding locations, colonies are systematically refreshed every five years. Furthermore, all mice are screened by polymerase chain reaction (PCR) to confirm the presence of transgenes and ensure genetic uniformity14. For quality assurance, a 26-week carcinogenic susceptibility monitoring study using N-methyl-N-nitrosourea (MNU), a known genotoxic carcinogen, as a positive control, is routinely conducted. This protocol includes (i) comprehensive monitoring (full-volume monitoring) across all organs after colony renewal and (ii) simplified monitoring (simple monitoring) focusing on forestomach tumor formation as the most sensitive endpoint, performed annually on production colonies. Previous studies reported that forestomach tumors in rasH2 mice following MNU administration frequently formed as early as 13 weeks15, 16. This result suggests carcinogenic susceptibility can be assessed at 13 weeks using forestomach tumor development as an endpoint. In this study, we aimed to evaluate whether forestomach tumors in rasH2 mice could be reliably detected 13 weeks after MNU administration, potentially allowing for shortened and simplified monitoring of carcinogenic susceptibility. Materials and Methods Retrospective analysis of historical data in carcinogenic susceptibility studies Histopathological data from 26-week carcinogenic susceptibility studies performed at CIEM (2019–2024) using MNU-treated rasH2 mice (150 males, 150 females) were analyzed. The incidence of forestomach tumors (squamous cell papilloma and carcinoma) was calculated overall and in mice that were moribund or died by week 13. Outline of animal experiments Four experiments were performed: experiments 1 and 2 were designed to compared forestomach tumor incidence at 13 and 26 weeks after MNU administration, and while experiments 3 and 4 assessed the occurrence of tumors in the forestomach at 13 weeks after MNU administration. Animal experiments were conducted at two facilities: experiments 1 and 3 at CIEM and experiments 2 and 4 at CLEA, Japan. Preparation of MNU solution MNU was purchased from MedChemExpress (Monmouth Junction, NJ, USA). MNU was received as a powder and refrigerated below −20ºC. The MNU solution was prepared at 7.5 mg/mL in citrate buffer and adjusted to pH 4.5. The prepared MNU solution was stored at room temperature and protected from light before use. Animals and husbandry Five- to six-week-old rasH2 mice of both sexes were obtained from CLEA Japan, Inc. (Fujinomiya, Japan). The mice were housed in plastic cages on soft woodchip bedding in an animal room maintained under standard conditions (room temperature set at 20–26ºC; relative humidity set at 30–70%, with a 12-h light/dark cycle). The CA-1 basal diet (CLEA Japan) and tap water or well water were provided ad libitum throughout the experimental period. The mice were quarantined, and their body weights and health conditions were monitored. After confirming their normal status, the mice were included in the experiment at the age of 7 weeks. This study protocol was approved by the Institutional Animal Care and Committee of the CIEM (approval numbers: AIA240039 and AIA240069) and CLEA Japan (Doujitu: 2460-158, 2461-074). Experimental design Following acclimation, male and female mice were assigned to 13- and 26-week observation groups for experiments 1 and 2, and to 13-week observation groups for experiments 3 and 4. In experiments 1 and 2, 30 male and 30 female mice (60 per study) were used. In experiments 3 and 4, 10 male and 10 female mice were used (20 per study). On the day of initiation, all mice in experiments 1–4 received a single intraperitoneal injection of MNU solution (7.5 mg/mL, 75 mg/kg). The mice were then observed for the designated periods. General observation and body weights The general condition of mice was assessed once daily during the observation period. The body weights of all surviving mice were measured individually using an electronic balance on the day of grouping and MNU administration, and weekly thereafter. Body weight on the day of MNU administration was measured to calculate dose (75 mg/kg). Body weight was measured when the mice were euthanized or died during the rearing period. Macroscopic and histopathological examinations Following the observation period, all surviving mice were placed under deep isoflurane anesthesia and exsanguinated by cutting the abdominal aorta or axillary vein. Following exsanguination, a gross pathological examination of all organs and tissues of the whole body was performed. Stomachs were injected with 10% neutral buffered formalin (NBF) solution, fixed for several minutes, cut open along the greater curvature, and photographed. Stomach samples were collected and fixed in 10% NBF. After fixation, tissues were embedded in paraffin and processed for hematoxylin and eosin (H&E) staining and histopathological analysis. In a conventional 26-week monitoring study, the carcinogenic sensitivity of rasH2 mice was considered appropriately maintained if the combined incidence of forestomach tumors, either squamous cell papilloma or carcinoma, exceeded 80%. Similarly, in this study, colonies were considered appropriate when the combined incidence of forestomach squamous cell papilloma and carcinoma was >80%. The number of nodules observed in the forestomach was counted in experiments 1 and 2 for both the 13-week and 26-week cohorts. Photographic images of the forestomach were captured using ImageJ (version 1.54g 2 September 2025)17 and the number of nodules was recorded. Masses that spread throughout the stomach were counted as single nodules. Statistical analysis Kalan–Meier survival analyses were performed to evaluate differences between the 26-week groups in experiments 1 and 2, and group comparisons were assessed using the log-rank test. These comparisons included those between experiments 1 and 2 within the same sex as well as comparisons between males and females within each experiment. The incidence of neoplastic changes was analyzed using Fisher’s exact probability test for the following comparisons: (1) between the 13- and 26-week groups in experiments 1 and 2, respectively; (2) between the 13- and 26-week groups in experiments 1 and 2; and (3) among the 13-week groups in experiments 1, 2, 3, and 4. Results Retrospective analysis of forestomach neoplastic lesions from carcinogenic susceptibility monitoring studies conducted between 2019 and 2024 In total, 150 mice of each sex were used for 26-week carcinogenic susceptibility. Most mice treated with MNU developed forestomach neoplastic lesions, and the combined incidences of squamous cell papilloma and carcinoma in the forestomach were 97.3% and 98.7% in males and females, respectively (Table 1). Of these, 29 and 41 male and female mice were moribund or had died by week 13, respectively. In moribund and dead mice, the combined incidence of squamous cell papilloma and carcinoma was 93.1% in males and 95.1% in females (Table 2). In a retrospective analysis of MNU-treated rasH2 mice observed for 26 weeks, the mean survival rate was 2.7% in males (range: 0.0–6.7%) and 7.3% in females (range: 0.0–16.7%). The corresponding survival curves from the five 26-week carcinogenic susceptibility studies used for the retrospective analysis are presented in Supplementary Fig. 1. Table 1. Incidence of Forestomach Tumors in 26-week Carcinogenicity Susceptibility Studies in Mice Administered 75 mg/kg MNU Conducted at CIEM Over the Past 5 Years (2019–2024)SexNo. of animalsPapilloma, squamous cellCarcinoma, squamous cellPapilloma or carcinomaMale150138 (92.0)13 (8.7)146 (97.3)Female150144 (96.0)4 (2.7)148 (98.7)Numbers in parentheses indicate incidence (%). MNU: N-methyl-N-nitrosourea; CIEM: Central Institute for Experimental Medicine and Life Science. Incidence of Forestomach Tumors in 26-week Carcinogenicity Susceptibility Studies in Mice Administered 75 mg/kg MNU Conducted at CIEM Over the Past 5 Years (2019–2024) Numbers in parentheses indicate incidence (%). MNU: N-methyl-N-nitrosourea; CIEM: Central Institute for Experimental Medicine and Life Science. Table 2. Incidence of Forestomach Tumors Observed in Moribund or Deceased Mice by Week 13 in 26-week Carcinogenic Monitoring Studies Administered 75 mg/kg MNU Conducted at CIEM over the Past 5 Years (2019–2024)SexNo. of animals*Papilloma, squamous cellCarcinoma, squamous cellPapilloma or carcinomaMale2925 (86.2)3 (10.3)27 (93.1)Female4137 (90.2)2 (4.9)39 (95.1)Numbers in parentheses indicate incidence (%). *: Total number of moribund or deceased animals by week 13. MNU: N-methyl-N-nitrosourea; CIEM: Central Institute for Experimental Medicine and Life Science. Incidence of Forestomach Tumors Observed in Moribund or Deceased Mice by Week 13 in 26-week Carcinogenic Monitoring Studies Administered 75 mg/kg MNU Conducted at CIEM over the Past 5 Years (2019–2024) Numbers in parentheses indicate incidence (%). *: Total number of moribund or deceased animals by week 13. MNU: N-methyl-N-nitrosourea; CIEM: Central Institute for Experimental Medicine and Life Science. Survival rate and body weight curve in experiments 1–4 The survival curves, survival rates, and body weight changes observed in the four experiments are presented in Fig. 1, Supplementary Table 1, and Supplementary Fig. 2, respectively. All four experiments yielded right-skewed survival curves. However, Kaplan–Meier survival analysis demonstrated a significant difference between male mice in the 26-week groups in experiments 1 and 2, as determined by the log-rank test (p<0.01). However, no significant differences were observed between females in the corresponding groups, and no significant differences were detected between males and females within the 26-week group in either experiment. Moribund or deceased mice were observed 7 weeks after MNU administration. Survival rates in the 13-week groups from the four experiments ranged from 70.0–100.0% in males and 70.0–93.3% in females (Fig. 1, Supplementary Table 1). In experiment 2, the survival rate at 13 weeks in the 26-week group was low in males, and all mice of both sexes were either moribund or had died during the observation period (Fig. 1 and Supplementary Table 1). In Experiment 1, survival at 13 weeks was high, but only three males survived the 26-week study period (Fig. 1, Supplementary Table 1). Survival rates at 13 weeks in Experiments 1–4 were consistent with retrospective data from MNU-treated rasH2 mice observed for 26 weeks. Survival curves for N-methyl-N-nitrosourea (MNU)-treated rasH2 mice. Survival curves of male and female rasH2 mice treated with 75 mg/kg MNU and autopsied at 13- and 26-weeks at the CIEM and CLEA, Japan (A and B). Dead or moribund mice were observed after 8 weeks post-MNU administration. Only two males in the 26-week CIEM group survived until study completion. Survival curves for the 13-week groups at CIEM and CLEA Japan are shown in C and D. Dead or moribund mice were observed after 7 weeks post-MNU administration. a) Male and female mice in the 13-week groups of experiments 1 and 2 were necropsied 13 weeks after MNU administration. In the 26-week observation groups in experiments 1 and 2, the survival rates of MNU-treated mice were 20.0% and 0.0% for males (3/15 and 0/15 mice, respectively) and 0.0% for females (0/15 mice in both experiments). The main causes of deterioration and death induced by MNU were malignant lymphoma of the thymus, and in some cases, squamous cell carcinoma of the forestomach or splenic hemangiosarcoma. In the 26-week group, most deaths were attributed to malignant lymphoma: 83.3% (10/12) in males and 76.9% (10/13) in females in experiment 1, and 80.0% (12/15) in males and 66.7% (10/15) in females in experiment 2. In addition, in the 13-week experiments 1–4, moribund or deceased mice were predominantly affected by malignant lymphoma. All nine moribund or deceased males were diagnosed with thymic malignant lymphoma, and among the 11 affected females, 9 had thymic malignant lymphoma and 2 had squamous cell carcinoma of the forestomach. Therefore, the primary causes of deterioration and death at 13 weeks were consistent with those observed in the 26-week group. Pathological forestomach findings Macroscopically, small nodules ranging from approximately 1–5 mm were observed in the forestomaches of mice in the 13- and 26-week groups (Fig. 2A). The corresponding histopathological features, squamous cell papillomas, were observed as well-circumscribed masses (Fig. 2B). In some cases in the 13- and 26-week groups, large nodules exhibiting hemorrhage with penetration extending to the serosal surface were observed (Fig. 3A). Histopathological examination revealed squamous cell carcinoma with basal membrane penetration and muscle layer infiltration (Fig. 3B, 3C). Squamous cell papilloma in forestomach. A: Small nodular lesions observed in the forestomach 13 weeks post- N-methyl-N-nitrosourea (MNU) administration. White scale bar=1.0 cm. B: Corresponding histopathological features of squamous cell papilloma. Scale bar=1.0 mm. Squamous cell carcinoma in forestomach. A: A large nodular lesion exhibiting serosal hemorrhage was identified in the forestomach 13 weeks post-N-methyl-N-nitrosourea (MNU) administration. White scale bar=1.0 cm. B and C: Nodular lesion histopathologically identified as squamous cell carcinoma. Scale bar=1.0 mm. The combined incidence of squamous cell papilloma and carcinoma was 100.0% in males and females in the 13-week group and 100.0% in males and 93.3% in females in the 26-week group (Table 3). No statistically significant differences were observed in the incidence of forestomach tumors between groups. Table 3. Neoplastic Findings in the Forestomach of rasH2 Mice Administered 75 mg/kg MNUSexExperimental No.Observation period (weeks)1326No. of animalsPapilloma, squamous cellCarcinoma, squamous cellPapilloma or carcinomaNo. of animalsPapilloma, squamous cellCarcinoma, squamous cellPapilloma or carcinomaMale11515 (100.0)0 (0.0)15 (100.0)1515 (100.0)0 (0.0)15 (100.0)21515 (100.0)1 (6.7)15 (100.0)1515 (100.0)1 (6.7)15 (100.0)31010 (100.0)0 (0.0)10 (100.0)----41010 (100.0)0 (0.0)10 (100.0)----Female11514 (93.3)1 (6.7)15 (100.0)1514 (93.3)0 (0.0)14 (93.3)21514 (93.3)2 (13.3)15 (100.0)1514 (93.3)0 (0.0)14 (93.3)31010 (100.0)0 (0.0)10 (100.0)----41010 (100.0)0 (0.0)10 (100.0)----Numbers in parentheses indicate incidence (%). MNU: N-methyl-N-nitrosourea. Neoplastic Findings in the Forestomach of rasH2 Mice Administered 75 mg/kg MNU Numbers in parentheses indicate incidence (%). MNU: N-methyl-N-nitrosourea. The number of forestomach nodules in each mouse in the 13- and 26-week cohorts is shown in Fig. 4. The number of forestomach nodules did not increase when the observation period was extended from 13 to 26 weeks. For squamous cell carcinoma, four cases were observed by week 13 and one case post-week 13 (Table 3). No increase in the incidence of squamous cell carcinoma was observed during the prolonged observation period. Number of forestomach nodules in each mouse. Scatter plots showing the number of pro-ventricular nodules and the week of moribundity or death in male (A) and female (B) necropsy groups after 13 and 26 weeks in experiments 1 and 2. Discussion This study demonstrated that MNU-induced forestomach tumors in rasH2 mice could be reliably detected 13 weeks after administration. A retrospective analysis of 26-week carcinogenic susceptibility monitoring studies conducted at the CIEM over the past five years demonstrated a high incidence of forestomach tumors in moribund or dead mice prior to 13 weeks (93.1% in males and 95.1% in females). Reports indicate that forestomach tumors can be induced within 12–13 weeks of MNU administration at doses of 50–75 mg/kg, with an incidence >90%15, 16. These results are comparable to those obtained in the present study, suggesting that a 13-week observation period may be sufficient for carcinogenic susceptibility monitoring studies (simple monitoring) focusing on forestomach tumor development. MNU administration reduced the survival rate of rasH2 mice, consistent with previous findings. In contrast, vehicle-treated rasH2 mice in CIEM background data maintained high 26-week survival, with 97.8% survival in males (range: 86.7–100.0%) and 95.6% survival in females (range: 86.7–95.6%), indicating that the reduced survival observed in the present experiments was attributable to MNU administration rather than general colony conditions. Our background data (internal background data, unpublished) further showed that the average survival rates of MNU-treated rasH2 mice observed for up to 26 weeks were 1.6% (range: 0.0–10.0%) for males and 4.9% (range: 0.0–20.0%) for females. Similarly, Suguro and colleagues reported a survival rate of approximately 20% following MNU administration18. The survival rates observed in this study fell within historical and published ranges, indicating that high mortality is typical of MNU-treated rasH2 mice and does not deviate from expected responses. In our previous studies (internal background data, unpublished), malignant lymphoma of the thymus was the predominant cause of death in MNU-treated rasH2 mice (male 91.5%; range: 86.2–96.7% and female 84.5%; range: 72.4–96.6%), whereas forestomach squamous cell carcinoma occurred less frequently. Consistently, malignant lymphoma was the principal cause of death in this 26-week study and 13-week studies, indicating a similar mortality pattern despite the shorter observation period. Although the incidence of malignant lymphoma observed in experiments 1 and 2 was slightly lower than historical mean values, particularly in females in experiment 2, it remained within the observed range of variability. This mortality pattern was therefore comparable to that observed in previous 26-week carcinogenic susceptibility studies using MNU-treated rasH2 mice. Similarly, Takaoka et al. reported malignant lymphoma as the most common neoplasm in dead or moribund rasH2 mice treated with MNU, followed by squamous cell carcinoma, with splenic hemangiosarcoma also noted16. Taken together, the causes of death observed in this study are consistent with those previously reported. A significant difference in survival was observed between male mice in the 26-week groups in experiments 1 and 2. Malignant lymphoma was the primary cause of death in both experiments, with comparable incidence rates. No significant difference in survival was observed between male and female mice as determined by the log-rank test, and the principal causes of death were the same. In this study, rasH2 mice were treated with a single intraperitoneal dose of MNU (75 mg/kg) and were observed for 13 and 26 weeks. Histopathological examinations revealed that the incidence of forestomach tumors was 100.0% in both males and females in the 13-week group, which was comparable to that observed in the 26-week group (100.0% in males and 93.3% in females). These results are consistent with those of our retrospective analysis. No clear association was observed between the time of moribundity/death and the number of forestomach nodules across the cohorts. Histopathological examination revealed no increase in squamous cell carcinoma incidence in the 26-week group compared with that in the 13-week group. MNU-induced forestomach carcinogenesis is primarily driven by point mutations in the introduced c-Ha-ras gene, as demonstrated by mutation analyses19. In this study, forestomach tumors were induced by a single initial dose of MNU at the start of the experiment and were therefore presumed to be caused by similar mutations. From the perspective of the carcinogenic mechanism, the experimental results were considered consistent, as the number of forestomach nodules and incidence of squamous cell carcinoma did not vary owing to differences in the observation periods. Therefore, a 13-week observation period was sufficient to evaluate carcinogenic susceptibility in this monitoring study. Based on these findings, the following approaches can be employed for future carcinogenic susceptibility monitoring and quality assurance. Following colony renewal, a full-volume monitoring study following the standard protocol for short-term carcinogenicity studies (26 weeks) using a single dose of MNU (75 mg/kg) will assess tumor type and incidence across organs. Subsequently, routine monitoring in production colonies will be conducted using a simplified 13-week study after a single MNU dose (75 mg/kg), with all mice euthanized and subjected to histopathological examinations of the forestomach according to established criteria (see “Macroscopic and histopathological examinations”). Carcinogenicity studies have been conducted to identify the tumorigenic potential of pharmaceuticals in animals and to assess associated risks in humans. Such studies are generally required for drugs intended for long-term or frequently repeated intermittent use in clinical settings. The assessment was performed based on dose–response relationships and appropriate durations of exposure, integrating both the exposure level and duration. Although this study demonstrated that MNU-induced carcinogenicity could be detected within 13 weeks, carcinogenicity assessment in drug development aims to define the margin of safety based on estimated human exposure, including duration19, and these findings are limited to MNU-induced forestomach tumors and may not apply to other agents or target organs. Therefore, the study period should continue to follow the standard 26-week protocol for short-term carcinogenicity. Overall, this study demonstrated that carcinogenic susceptibility focusing on forestomach tumor development in rasH2 mice could be reliably evaluated within a 13-week observation period. Consistent forestomach tumor incidence was observed across studies conducted at CIEM and CLEA Japan. Histopathological findings at 13 weeks were comparable to those at 26 weeks and aligned with those of a retrospective analysis of historical 26-week data, further supporting the reproducibility and robustness of the abbreviated protocol. These results indicate that a simplified 13-week monitoring study provides a reliable and efficient approach for assessing carcinogenic susceptibility induced by MNU in rasH2 mice. Author Contributions TN, RI, TM, TK, and IO performed animal experiments and contributed to data acquisition, analysis, and interpretation. MY and MM analyzed and interpreted the pathological data. TN and RI drafted the manuscript. CN, TY, and MS conceived the study, supervised the study design, and provided advice for interpretation of the experimental data. All authors have read and approved the final version of the manuscript and agree to be accountable for all aspects of the research to ensure that the accuracy or integrity of any part of the work has been appropriately investigated and resolved. MS provided final approval for the version to be published. Disclosure of Potential Conflicts of Interest TN, RI, MY, MM, CN, TY, and MS are researchers at CIEM who developed the rasH2 mice. TM, TK, and IO are employees of CLEA, Japan, which produces rasH2 mice. Supplementary Material Acknowledgments The authors express their gratitude to Kenji Kawai (CIEM), Yoko Kamai (CIEM), and Rui Nagakubo (CLEA, Japan) for their technical assistance. References Associated Data Supplementary Materials