Enhancing postoperative recurrence assessment in gastric and colorectal cancer patients with intraperitoneal fluorouracil implants: overcoming the diagnostic challenge of fluorouracil implant-related tumor-like lesions

Introduction

Colorectal cancer and gastric cancer are two of the most prevalent cancers globally, ranking third (10.0%) and fifth (5.6%) in terms of incidence, and second (9.4%) and fourth (7.7%) in terms of mortality rates, respectively (1). Subclinical implantation metastasis formed by the tumor invasion of the serosa or the surgical-induced shedding of cancer cells is the main cause of local recurrence and liver metastasis after surgery (2). Intraoperative intraperitoneal chemotherapy using sustained-release fluorouracil implants has emerged as a novel strategy for preventing liver metastasis and local recurrence in digestive tract tumors (3). The fluorouracil can be released continuously, maintaining high local drug concentration for about one month postoperatively (4,5), and has demonstrated safety advantages over conventional infusion techniques (3). In addition to its confirmed safety, this method has also been shown to improve the postoperative survival rate and progression-free survival rate of cancer patients, significantly reducing the risk of peritoneal recurrence (2-4,6). Despite these promising outcomes, existing research and clinical applications of this method have been limited to certain regions, particularly China, and global dissemination has been relatively limited to date. Thus, there is a need for broader international research and the recognition of potential benefits of intraoperative intraperitoneal chemotherapy using sustained-release fluorouracil implants.

The early detection of recurrent disease or secondary primary tumors during intensive follow up can improve patient survival (7). Abdominopelvic computed tomography (CT) and magnetic resonance imaging (MRI) are recommended as essential monitoring methods for patients with colorectal and gastric cancer after treatment. MRI can be used to assess tumor activity and the treatment response, providing more detailed information on disease progression (8,9).

Fluorouracil implantation is gradually becoming widely adopted for chemotherapy in gastric cancer, colorectal cancer, breast cancer, and pancreatic cancer (3). These implants may be retained and present as tumor-like lesions in postoperative follow-up imaging, potentially leading to false-positive interpretations as metastatic sites, affecting patient management. Unnecessary surgical resections of fluorouracil implantation-related lesions have been reported (10,11). Therefore, an ability to accurately differentiate between these lesions and true malignancies is crucial for optimal patient management. Fluorouracil implants placed on the hepatic diaphragmatic surface appear as low-density lesions without significant enhancement on imaging (11). Previous studies have focused on a small number of cases with localized lesions, and longitudinal data on lesion evolution over time is lacking. Therefore, a comprehensive imaging study involving a larger patient cohort, deeper lesion characterization, and detailed follow-up observations was needed to enhance radiologists’ and clinicians’ understanding of these lesions, and reduce the risk of false-positive interpretations. We hypothesized that providing radiologists with training on fluorouracil implant imaging characteristics and accurate surgical record documentation on the fluorouracil implant location and quantity would improve their diagnostic accuracy. The goal of the present study was to improve the diagnostic accuracy of radiologists in evaluating patients with gastric or colorectal cancer who had received fluorouracil implants. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-24-2033/rc).

Methods

Patient selection

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and was approved by the Ethics Committee of Tongji Hospital (No. TJ-IRB202407012). The requirement of written informed consent was waived due to the retrospective and observational nature of the study. Patients who received intraperitoneal fluorouracil implants during surgery between January 2017 and June 2023 were included in the study. A retrospective analysis of their clinical and imaging data was conducted up to December 2023 to complete a six-month event window. Patients were included in the study if they met the following inclusion criteria: (I) had gastric or colorectal cancer and had received intraoperative intraperitoneal chemotherapy using sustained-release fluorouracil implantation; and (II) had undergone at least two post-operation MRI or CT follow-up examinations up to December 2023. Patients were excluded from the study if they had been diagnosed with malignancies other than gastric or colorectal carcinoma (e.g., sarcoma) (Figure 1A).

Figure 1 Study flow chart. (A) Flow chart of the patient inclusion and exclusion process for the first part of the study, focusing on the retrospective analysis of the clinical and imaging features. (B) The inclusion of CT images and the process in the second part of the study for the three-stage assessment of tumor-like lesions, conducted by two junior radiologists. CT, computed tomography; MDT, multidisciplinary team; MRI, magnetic resonance imaging.

This study comprised two parts. The first part of the study involved the summarizing of imaging characteristics of fluorouracil implant-related lesions by the multidisciplinary team (MDT), which served as the reference standard for the evaluation in the second part of the study. The second part of the study involved a three-stage assessment of tumor-like lesions conducted by two junior radiologists (S.Z. and X.C.), each with two years of experience.

Imaging techniques

Patients in the present study underwent routine abdominopelvic CT scans for postoperative evaluation and long-term follow up, including unenhanced CT and dynamic contrast-enhanced computed tomography (CECT). Unenhanced MRI [routine T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI)], diffusion-weighted imaging (DWI), and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) were primarily used to evaluate the liver or pelvic metastases (12-14). The scanning systems (vendors) and CECT/DCE-MRI protocols are detailed in Table S1.

Retrospective analysis and observation index

Two senior radiologists (Y.S. and X.H.) from the MDT identified the presence of the fluorouracil implant-related lesions and localized them to specific imaging slices (after a full discussion with surgeons). They also summarized the baseline imaging characteristics and follow-up changes based on a comprehensive review of all the CT and MRI examinations. The documented imaging features included size, morphology, density (CT attenuation), and signal intensity (on T1WI, T2WI, and DWI). Clinical data, including age, gender, primary tumor type, and tumor staging, were also recorded. The dates of surgery were recorded to analyze the number of patients who had undergone intraoperative intraperitoneal chemotherapy in recent years. Additionally, the study assessed disease-free survival (DFS) and documented the specific sites of recurrence (4). The metastasis or recurrence rates of the stage III gastric cancer and colorectal cancer patients in this study were compared with the published rates of patients who had not received fluorouracil implants (2,4).

Fluorouracil implant-related tumor-like lesion evaluation

Two junior radiologists first independently documented the locations of the tumor-like lesions, and classified the lesions as benign or malignant, and then rated their diagnostic confidence as high, medium, or low. Cases were excluded from the analysis if the surgical records lacked information about 5-fluorouracil implants or if the recorded information was incomplete. The assessment was conducted into three stages, which were defined based on the radiologists’ typical image reading process. A total of 344 CT examinations were performed among eligible patients. Of these, 240 examinations were randomly selected and evenly assigned to three groups (n=80 per group) corresponding to the three assessment stages (Figure 1B): stage 1: pre-training without surgical information; stage 2: post-training without surgical information; and stage 3: post-training with surgical information and prior CT/MRI images.

For the training, representative CT images that reflected the characteristics of 5-fluorouracil implants were selected by the MDT from the remaining 104 examinations. The second stage of the evaluation took place one week after the training. If there was a difference in the diagnostic confidence and benign-malignant assessment between the two radiologists, the radiologists conducted a joint review and discussed the issue until a consensus was reached. Across the three reading stages, the accuracy of the diagnosis was assessed by evaluating the malignancy or benignity of fluorouracil implant-related lesions based on MDT-confirmed reference standards and their precise localization in the imaging sequences. Any classification of a benign implant-related lesion as malignant was considered a diagnostic error.

Statistical analysis

Descriptive statistics were used for the clinical and imaging characteristics of patients, and the results are presented as the number (percentage). Continuous variables with a normal distribution are reported as the mean ± standard deviation. Non-normally distributed data (CT attenuations) were analyzed using percentiles along with the minimum and maximum values. Pearson’s Chi-squared test, Fisher’s exact test, or an independent samples t-test was used as appropriate to compare the recurrence rates between the groups and imaging characteristics. DFS was estimated using the Kaplan-Meier method. Diagnostic accuracy and diagnostic confidence were compared using Pearson’s Chi-squared test and the Wilcoxon rank-sum test. In the second part of the study, the intra-reader reliability of the diagnostic confidence and benign-malignant assessment by the two radiologists was conducted using the intraclass correlation coefficient (ICC). All statistical analyses were performed using SPSS 25.0 software package (SPSS Inc., Chicago, IL, USA).

Results

Retrospective analysis of clinical and imaging features

Patient characteristics

In total, 168 fluorouracil implants, which had been confirmed to have been placed during surgery in 164 patients (mean age: 56.17±10.62 years; 59 females and 105 males), were included in the study. Of the 164 patients, 111 (67.68%) had gastric cancer, and 53 (32.32%) had colorectal cancer (Figure 1). Each patient had 1–2 fluorouracil implants placed during surgery. There has been a gradual increase in the number of cases undergoing intraoperative fluorouracil placement since 2017 (Figure 2). In 30.49% of the surgical records, information on the location and quantity of fluorouracil implants was missing. One patient, whose surgical record lacked any mention of the use of fluorouracil implants, underwent unnecessary surgery, as a fluorouracil-related tumor-like lesion was misdiagnosed as a recurrent lesion (Figure 3).

Figure 2 Intraoperative intraperitoneal chemotherapy using fluorouracil implantation. (A) The number of cases undergoing intraoperative fluorouracil placement since 2017. (B,C) Gauze-wrapping of sustained-release fluorouracil implants prior to intraperitoneal placement.

Figure 3 Implant-related tumor-like lesion misdiagnosed as recurrence. A 46-year-old, female, gastric cancer patient with an implant (as indicated by the yellow arrows) located at the tumor bed (A,B). The surgical records lacked relevant information. Four years later, the rounded low-density lesion (as indicated by the yellow arrows) closely associated with the gastric wall was seen on the CT images (C,D), suggesting recurrence. Despite the MDT consultation advice that it was a postoperative implant, surgical resection was performed at the patient’s request. Postoperative pathology (E: H&E staining, ×40; F: H&E staining, ×100) revealed a capsule containing drug granules and gel-like material, consisting of collagen fibrosis and adipose tissue. It was ultimately confirmed to be an implant from a previous surgery. CT, computed tomography; H&E, hematoxylin and eosin; MDT, multidisciplinary team.

The clinical information of all the patients is summarized in Table 1. The gastric cancer patients had a mean DFS of 50.5 months, and a median follow-up time of 20 months. The colorectal cancer patients had a mean DFS of 18.7 months, and a median follow-up time of 12 months. Recurrence occurred in 17.12% of the gastric cancer patients and 24.53% of the colorectal cancer patients. The data of the stage III gastric cancer patients in the present study were compared with published data on the other patients (2, 4), and the results showed that the patients in the present study had significantly lower rates of in situ recurrence and peritoneal metastasis (4) (9.3% vs. 26.42%, P=0.033; 6.98% vs. 35.85%, P=0.001). There was no statistically significant difference in the tumor stage distribution between the colorectal cancer patients in the present study and the published group (2). A further comparison revealed no difference between the two groups in terms of peritoneal metastasis (7.55% vs. 7.69%, P=1); however, the in situ recurrence rate was lower in the colorectal cancer patients in the present study than the published group (2) (0% vs. 19.23%, P=0.001) (Table S2).

Imaging key features

The baseline characteristics of and follow-up changes in the fluorouracil implant-related tumor-like lesions are outlined in Tables 2,3. The locations of fluorouracil implants included the subdiaphragmatic region, paracolic gutter, and tumor bed. The CT attenuation values of the fluorouracil implants on the initial postoperative CT images ranged from 24 to 151 Hounsfield units (HU). In the visual assessment, 97.62% of lesions showed iso- to hyper-density compared to muscle, while the remaining showed hypo-density. Additionally, 65.24% of the lesions exhibited heterogeneous density (Figure 4). Subsequent follow-up revealed that the densities of fluorouracil implants remained unchanged in 47.02% (n=79) of the cases, decreased in 42.86% (n=72) of the cases, and showed marked calcification in 10.12% (n=17) of cases. Marginal calcification (Figure 4) was observed in 53.57% of the lesions (n=90). Cases with marginal calcification had longer follow-up durations (24.52±1.632 vs. 14.69±1.352 months, P<0.0001). In total, 31 patients underwent MRI. Most lesions exhibited hypo-intensity on T1WI and hyper-intensity on T2WI (26/31, 83.87%). On DWI (b=1,000), the lesions appeared hyper-intense in 76.67% (n=23) of the cases, while the corresponding apparent diffusion coefficient (ADC) mapping showed no diffusion restriction (Figure 5). No contrast enhancement was observed on either CT or MRI (Figure 6).

Figure 4 Implant-related tumor-like lesions on CT images. (A,B) A 63-year-old, female, colon cancer patient with an implant located near the tumor bed. On the CT images, the lesion (as indicated by the yellow arrow) appeared to be heterogeneous, with slightly low density (A), and a surrounding foreign body reaction. Three months later (B), shrinkage of the lesion (as indicated by the yellow arrow) was observed. (C,D) A 37-year-old, female, colon cancer patient with an implant located in the left paracolic gutter. On the initial postoperative CT images, the lesion appeared to be equal density, with a significant surrounding foreign body reaction (as indicated by the yellow arrow) (C). After eight months, significant marginal calcification (as indicated by the yellow arrow) was observed, with a noticeable reduction in the surrounding foreign body reaction (D). CT, computed tomography.

Figure 5 Implant-related tumor-like lesions on magnetic resonance images. A 71-year-old, male, colon cancer patient with an implant located in the subdiaphragmatic region. The lesion (as indicated by the yellow arrow) exhibited hypo-intensity on T1WI (A) and hyper-intensity on T2WI (B). On DWI (C) and ADC mapping (D), the lesion (as indicated by the yellow arrow) showed no significant restricted diffusion; rather, the ADC mapping showed increased signal intensity compared with normal liver parenchyma, reflecting increased water diffusion. ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

Figure 6 Example of a true-positive interpretation before training in stage 1 of the study. A 59-year-old, male patient with stage III colorectal cancer had a fluorouracil implant placed in the pelvic tumor bed area. The lesion was recorded as malignant by two reviewers. (A) Arterial phase contrast-enhanced CT showed a heterogeneous nodule (as indicated by the yellow arrow) adjacent to the rectum with 66 HU. (B) Venous phase contrast-enhanced CT showed the same nodule (as indicated by the yellow arrow) with 64 HU; there was no significant change in attenuation between the phases. AP, arterial phase; CT, computed tomography; HU, Hounsfield units; VP, venous phase.

The shapes of the lesions were predominantly oval, and some irregularities were observed. No significant difference in lesion size was found between the gastric and colorectal cancer patients. According to the initial measurements, the lesions had an average length of 21.63±4.54 mm and an average short diameter of 14.05±3.3 mm. A size reduction was observed in the majority of the lesions (113, 67.26%) (Figure 4). Cases with lesion size reduction had a longer follow-up period (21.22±1.36 vs. 15.89±1.96 months, P=0.0266); however, the degree of size reduction was not correlated with the follow-up time (r=0.064, P=0.512). No statistically significant difference was observed in the location distribution between lesions that showed shrinkage and those that did not (P=0.633).

During the follow up, foreign body reactions were observed in 144 lesions (85.71%), characterized by centrally located fluorouracil implants, which were surrounded by an annular area of fat attenuation on CT (or hypointense signal on T2 fat-suppressed MRI), and further encircled by a blurred, slightly hyperdense rim on CT or a hyperintense rim on T2 fat-suppressed MRI. These imaging changes could be detected as early as two weeks after surgery (Figure 4). As the follow-up time increased, foreign body reactions gradually faded and eventually resolved, concurrent with the progressive calcification of the lesions.

Fluorouracil implant-related tumor-like lesion evaluation

The intra-reader agreement of the two radiologists was excellent in terms of the diagnostic confidence and benign-malignant assessment (Table S3). After training, the diagnostic accuracy (91.25%) and confidence (most often classified as medium) were improved in stage 2 compared to stage 1 (67.5%, most often classified as low) (accuracy: P<0.001; confidence: P<0.001) (Table 4). When surgical information was available, the accuracy (100%) and diagnostic confidence (most often classified as high) were improved in stage 3 compared to stage 2 (accuracy: P=0.007; confidence: P<0.001). Examples of the erroneous judgments made by radiologists in stages 1 and 2 are shown in Figures 6,7.

Figure 7 Example of a true-positive interpretation after training in stage 2 of the study. A 49-year-old, male patient with stage Ⅲ colon cancer had a fluorouracil implant placed on the left side of the root of the inferior mesenteric artery. (A,B) Six-week postoperative CT imaging showed multiple nodules near the abdominal aorta. Differentiation was challenging because the nodules included both the implant (as indicated by the yellow arrows) and new lesions (as indicated by the red arrows), suggesting recurrence. (C) Preoperative CT imaging showed no nodular lesions. (D) Two-week postoperative CT imaging showed a new oval-shaped lesion (as indicated by the yellow arrow), corresponding to the fluorouracil implant. By comparing the images over time, the implant was identified as the lesion (as indicated by the yellow arrow), which first appeared at two weeks postoperatively, and then decreased in size with the surrounding foreign body reaction by six weeks. Conversely, the nodules consistent with recurrence (red arrow) appeared de novo at six weeks. AP, arterial phase; CT, computed tomography; VP, venous phase.

Discussion

This study investigated the imaging characteristics of fluorouracil implant-related tumor-like lesions in a relatively larger cohort of gastrointestinal cancer patients than previously reported. Additionally, it was the first to systematically characterize the temporal evolution of these lesions through longitudinal imaging analysis, filling a critical gap in the existing literature. We showed that enhancing radiologists’ training on the specific imaging features of these lesions, along with detailed documentation of implant locations and quantities in surgical records, significantly improved their diagnostic confidence and accuracy. These findings offer valuable insights for radiologists and clinicians into distinguishing between fluorouracil implants and recurrent or metastatic tumors, ultimately improving patient management.

Sustained-release fluorouracil implants have shown promise in reducing postoperative recurrence and metastasis, and improving survival rates in gastrointestinal cancers, and pancreatic and liver cancers (2-5,15-17). In this study, the fluorouracil implants were shown to inhibit both in situ and peritoneal recurrence in patients with stage III gastric cancer, while a reduction in in situ recurrence was also observed in patients with colorectal cancer. These findings align with previous research, including that of Xu et al., who showed the efficacy of fluorouracil implants in reducing peritoneal metastasis in gastric cancer, and Yuan et al., who reported a significant reduction in in situ recurrence in colorectal cancer. Together, these studies underscore the potential of sustained-release fluorouracil implants as an effective adjuvant therapy for minimizing postoperative recurrence and metastasis across various gastrointestinal cancers. Fluorouracil implants are commonly prepared by wrapping sustained-release fluorouracil in gauze and then positioning the implants at various locations (e.g., the subdiaphragmatic region, paracolic gutter, and tumor bed). In clinical practice, surgeons decide where to place the fluorouracil implants based on the location of the tumors (2,3,6).

In clinical practice, misdiagnosis caused by gauze left behind during abdominal surgery occasionally occurs. These fiber-wrapped lesions can resemble tumor-like lesions on imaging (18,19). Similarly, gauze-wrapped fluorouracil implants may present as tumor-like lesions; thus, radiologists need to recognize their distinct imaging features to avoid misdiagnosis. Our study provided a comprehensive imaging analysis and examination of the follow-up data of 168 patients, further emphasizing the importance of accurate interpretation to minimize the risk of misdiagnosis.

To improve radiologists’ and clinicians’ understanding of fluorouracil implant–related tumor-like lesions, the MDT outlined the imaging features and temporal changes that occur during follow-up periods. The density of these lesions on unenhanced CT images is variable relative to the muscle; on MRI, they always show typical hypo-intensity on T1WI and hyper-intensity on T2WI. DWI is most frequently used to distinguish between benign and malignant lesions. Notably, lower ADC values derived from DWI series have been used to identify metastatic lymph nodes in patients with gastrointestinal malignancies (20,21). Unlike the malignant lesions, these lesions showed mild hyper-intensity or hyper-intensity on DWI (b=1,000); however, these lesions showed no restricted diffusion on DWI and ADC mapping. The corresponding ADC mapping showed a hyper-intense or isointense signal compared with normal liver parenchyma, reflecting unrestricted diffusion. In addition, contrast enhancement was not observed on CT/MRI. Therefore, using a combination of complementary imaging modalities helps distinguish between fluorouracil-related lesions and metastatic lesions.

In addition to the direct manifestations of the lesions, a foreign body reaction was observed in most of the lesions in this study. The foreign body reaction is characterized by centrally located fluorouracil implants, surrounded by an annular area of fat attenuation on CT (or hypointense signal on T2 fat-suppressed MRI), which is further encircled by a blurred, slightly hyperdense rim on CT or a hyperintense rim on T2 fat-suppressed MRI. Pathologically, this corresponds to a granulomatous reaction, which is the main response of the body’s non-specific immune system to implanted foreign bodies. This reaction may persist as a capsule or resolve due to the degradation of the foreign body (22). The foreign body reaction gradually resolves as the lesion calcifies. If the combined relevant surgical information is provided, it is not difficult to determine whether a lesion is caused by a sustained chemotherapy implantation.

Size reduction and calcification were observed in more than half of the follow-up images of the patients, and the longer the follow-up time, the more these changes were observed. In addition, the variety in the density/signal intensity, calcification, and the degree of foreign body reaction might be related to differences in the package material. The surface properties of fluorouracil implants play a crucial role in modulating the foreign body reaction (23). Only part of the fluorouracil implant is wrapped in absorbable gauze. However, variations in its utilization could stem from differences in drug approvals, and reimbursement strategies across years and regions.

By comparing the accuracy and diagnostic confidence of the radiologists in differentiating between the tumor-like lesions across the three stages, we confirmed that training the radiologists on the aforementioned radiological characteristics increased their accuracy and diagnostic confidence. However, despite the training, the diagnostic accuracy remained at 91.25% in the absence of relevant surgical information. Even when radiologists considered the possibility of fluorouracil implants, their diagnostic confidence was often insufficient, particularly for implants located in the tumor bed, which were challenging to distinguish from recurrent or metastatic lymph nodes. In clinical practice, radiologists typically rely on surgical records to gather information. In 30.49% of the surgical records, there was a lack of documentation in relation to the location and quantity of fluorouracil implants. Notably, one patient who lacked relevant surgical information had the fluorouracil implant surgically removed unnecessarily (Figure 3). Thus, fluorouracil implant location and quantity information needs to be recorded in surgical records if radiologists are to make accurate assessments.

The limitations of the study include its retrospective single-center design, which rendered it susceptible to selection bias. Additionally, due to variations in the fluorouracil implant locations there was a lack of cases with metastases or recurrences at the same site that could serve as controls in the evaluation of the second part of the study, which restricted the assessment of other diagnostic parameters. Consequently, only diagnostic accuracy could be evaluated. Nonetheless, this aligned with the aim of the study, as it showed that radiologists do not interpret fluorouracil implant–related tumor-like lesions as malignant during follow-up imaging studies.

Conclusions

Fluorouracil implants remain permanently in the body. Radiologists should be aware of this phenomenon and be familiar with its imaging characteristics. Providing radiologists with training on fluorouracil implant imaging characteristics, and accurate surgical record documentation of fluorouracil implant location and quantity are crucial in enabling precise radiological assessments and streamlining diagnosis.

Acknowledgments

We would like to thank all the participants and volunteers for their cooperation and efforts.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-2033/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-24-2033/dss

Funding: This work was supported by the teaching program of Huazhong University of Science and Technology (grant No. 2024PY08 to Y.S.), the Local Science and Technology Development Fund Project Guided by Central Government, China (grant No. YDZJSX 2025D074 to Y.S.), and the Scientific Research Projects from Wuhan Municipal Health Commission (grant No. WX23Q40 to L.H.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-2033/coif). L.H. reports that she received a grant from Scientific Research Projects from Wuhan Municipal Health Commission (Grant No. WX23Q40). Y.S. reports that she received grants from a teaching program of Huazhong University of Science and Technology (Grant No. 2024PY08) and the Local Science and Technology Development Fund Project Guided by Central Government, China (Grant No. YDZJSX 2025D074). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and was approved by the Ethics Committee of Tongji Hospital (No. TJ-IRB202407012). The requirement of written informed consent was waived due to the retrospective and observational nature of the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Yuan H, Zheng B, Tu S. Clinical research of intraperitoneal implantation of sustained-release 5-fluorouracil in advanced colorectal cancer. World J Surg Oncol 2015;13:320. [Crossref] [PubMed]
3. Ge J, Liu T, Lei T, Li X, Song K, Azizi S, Liu H, Tang M. Retrospective Cohort Study of Intraoperative Administration of Sustained-Release 5-Fluorouracil Implants in Advanced Gastric Cancer Patients. Front Pharmacol 2021;12:659258. [Crossref] [PubMed]
4. Xu Y, Zhang R, Li C, Sun Z, Deng J, Wang X, Ding X, Wang B, Xue Q, Ke B, Zhan H, Liu N, Liu Y, Wang X, Liang H, Xue Y, Xu H. Intraperitoneal Chemotherapy Using Fluorouracil Implants Combined With Radical Resection and Postoperative Adjuvant Chemotherapy for Stage III Gastric Cancer: A Multi-Center, Randomized, Open-Label, Controlled Clinical Study. Front Oncol 2021;11:670651. [Crossref] [PubMed]
5. Li JQ, Yang JC, Liang JX, Wang SL. Pharmacokinetic study and clinical evaluation of a slow-release 5-fluorouracil implant in pancreatic cancer patients. Anticancer Drugs 2016;27:60-5. [Crossref] [PubMed]
6. Wang ZJ, Tao JH, Chen JN, Mei SW, Shen HY, Zhao FQ, Liu Q. Intraoperative intraperitoneal chemotherapy increases the incidence of anastomotic leakage after anterior resection of rectal tumors. World J Gastrointest Oncol 2019;11:538-50. [Crossref] [PubMed]
7. Renehan AG, Egger M, Saunders MP, O'Dwyer ST. Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. BMJ 2002;324:813. [Crossref] [PubMed]
8. Yuval JB, Patil S, Gangai N, Omer DM, Akselrod DG, Fung A, Harmath CB, Kampalath R, Krehbiel K, Lee S, Liu PS, Millet JD, O'Malley RB, Purysko AS, Veniero JC, Wasnik AP, Garcia-Aguilar J, Gollub MJ. MRI assessment of rectal cancer response to neoadjuvant therapy: a multireader study. Eur Radiol 2023;33:5761-8. [Crossref] [PubMed]
9. Hao L, Li Y, Xiong Z, Jiang Y, Hu X, Hu D, Li Z, Shen Y. Imaging Phenotypes and Evolution of Hepatic Langerhans Cell Histiocytosis on CT/MRI: A Retrospective Study of Clinical Cases and Literature Review. Bioengineering (Basel) 2023.
10. Bai DS, Jin SJ, He R, Jiang GQ, Yao J. Granuloma induced by sustained-release fluorouracil implants misdiagnosed as a hepatic tumor: A case report. Oncol Lett 2014;8:742-4. [Crossref] [PubMed]
11. Li Y, He K, Hao X, Morelli JN, Shen Y, Hu X, Hu D, Li Z. Gastrointestinal tumor-related perihepatic fluorouracil encapsulated lesions and liver metastases: a diagnostic imaging study based on contrast-enhanced computed tomography and magnetic resonance imaging. Quant Imaging Med Surg 2023;13:7236-46. [Crossref] [PubMed]
12. Lordick F, Carneiro F, Cascinu S, Fleitas T, Haustermans K, Piessen G, Vogel A, Smyth ECESMO Guidelines Committee. Electronic address: clinicalguidelines@esmo. Gastric cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2022;33:1005-20. [Crossref] [PubMed]
13. Morgagni P, Bencivenga M, Carneiro F, Cascinu S, Derks S, Di Bartolomeo M, et al. International consensus on the management of metastatic gastric cancer: step by step in the foggy landscape : Bertinoro Workshop, November 2022. Gastric Cancer 2024;27:649-71. [Crossref] [PubMed]
14. Van Cutsem E, Cervantes A, Adam R, Sobrero A, Van Krieken JH, Aderka D, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol 2016;27:1386-422. [Crossref] [PubMed]
15. Chen J, Zhang J, Wang C, Yao K, Hua L, Zhang L, Ren X. Safety of implanting sustained-release 5-fluorouracil into hepatic cross-section and omentum majus after primary liver cancer resection. Int J Immunopathol Pharmacol 2016;29:475-9. [Crossref] [PubMed]
16. Wu YZ, Wu M, Zheng XH, Wang BZ, Xue LY, Ding SK, Yang L, Ren JS, Tian YT, Xie YB. No long-term survival benefit with sustained-release 5-fluorouracil implants in patients with stages II and III gastric cancer. World J Gastroenterol 2022;28:5589-601. [Crossref] [PubMed]
17. Li L, Li C, Zhou J. Effective sustained release of 5-FU-loaded PLGA implant for improving therapeutic index of 5-FU in colon tumor. Int J Pharm 2018;550:380-7. [Crossref] [PubMed]
18. Cheon JW, Kim EY, Kim KY, Park JB, Shin YK, Kim KY, Chae HD. A case of gossypiboma masquerading as a gastrointestinal stromal tumor. Clin Endosc 2011;44:51-4. [Crossref] [PubMed]
19. Kim SW, Kim SW, Song SK. Gastric pseudotumoral lesion caused by a fish bone mimicking a gastric submucosal tumor. J Gastric Cancer 2014;14:204-6. [Crossref] [PubMed]
20. Cheng J, Wang Y, Deng J, McCarthy RJ, Wang G, Wang H, Ye Y. Discrimination of metastatic lymph nodes in patients with gastric carcinoma using diffusion-weighted imaging. J Magn Reson Imaging 2013;37:1436-44. [Crossref] [PubMed]
21. Cho EY, Kim SH, Yoon JH, Lee Y, Lim YJ, Kim SJ, Baek HJ, Eun CK. Apparent diffusion coefficient for discriminating metastatic from non-metastatic lymph nodes in primary rectal cancer. Eur J Radiol 2013;82:e662-8. [Crossref] [PubMed]
22. Carneiro BC, Cruz IAN, Chemin RN, Rizzetto TA, Guimarães JB, Silva FD, Junior CY, Pastore D, Ormond Filho AG, Nico MAC. Multimodality Imaging of Foreign Bodies: New Insights into Old Challenges. Radiographics 2020;40:1965-86. [Crossref] [PubMed]
23. Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2008;20:86-100. [Crossref] [PubMed]