Preoperative risk stratification of colorectal high-grade intraepithelial neoplasia based on endoscopic biopsy: superior performance of ¹⁸F-FDG positron emission tomography-computed tomography over contrast-enhanced computed tomography in guiding surgical decision-making

Introduction

Colorectal cancer (CRC) remains a major cause of cancer-related death worldwide (1), highlighting the importance of accurate preoperative diagnosis of colorectal lesions for guiding clinical staging and initial treatment planning (2). Colonoscopy is widely regarded as the most effective modality for detecting colorectal lesions, and endoscopic forceps biopsy (EFB) serves as the gold standard for the histological diagnosis of colorectal epithelial neoplasia. However, EFB results can occasionally be inadequate for definitive diagnosis due to technical limitations (3), and histopathologic discrepancies between EFB and surgically resected specimens pose significant challenges for both clinicians and pathologists (4,5).

According to the World Health Organization (WHO) classification, CRC is histologically characterized by the invasion of tumor cells into the submucosa (ISM). Neoplasms confined to the epithelial layer without submucosal invasion are classified as intraepithelial neoplasia—categorized as either low-grade or high-grade—and not as carcinoma (6). High-grade intraepithelial neoplasia (HGIN) specifically refers to lesions exhibiting high-grade dysplasia (HGD) that do not invade beyond the muscularis mucosa. However, due to factors such as sampling error, inadequate bowel preparation, technical limitations of equipment, and variability in endoscopist experience, EFB may fail to detect invasive components of tumors that have infiltrated the submucosa (3). Studies have shown that some lesions initially diagnosed as HGIN based on EFB are ultimately confirmed to be invasive CRC upon examination of surgical specimens—a phenomenon referred to in this study as pathological upstaging (3,7). Therefore, EFB should not be used as the sole basis for preoperative diagnosis, as reliance on biopsy findings alone may lead to significant underestimation of disease severity and subsequent undertreatment (8,9).

Advanced imaging modalities play a pivotal role in the preoperative assessment of CRC by providing critical insights into tumor invasion depth and lymph node involvement (10,11). Contrast-enhanced computed tomography (CECT) scans are routinely performed for a variety of clinical indications and provide a method for the diagnosis of CRC. In addition, CECT is an effective method of staging diagnosed CRC, which helps in the development of an appropriate treatment plan (12,13). Positron emission tomography-computed tomography (PET/CT) is an important tool for the staging, treatment monitoring, and restaging of patients with CRC. PET/CT’s capacity to detect occult metastases directly impacts surgical planning, particularly in determining the extent of lymphadenectomy and the need for multivisceral resection (14).

In this study, we retrospectively analyzed cases of colorectal HGIN diagnosed by EFB that were further subjected to PET/CT and CECT, comparing the diagnostic and evaluative efficacy of these two modalities in order to determine the most suitable examination for patients with this condition. In addition, we characterized the diagnostic discrepancies and conditions requiring supplemental imaging evaluation through the analysis of the clinicopathological features associated with the underdiagnosis of ISM in CRC biopsy specimens. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-852/rc).

Methods

We performed a retrospective cohort analysis of patients with a diagnosis of HGIN who underwent surgical resection and PET/CT and CECT at The First Affiliated Hospital of Soochow University from January 1, 2019, to January 1, 2024. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and was approved by the Ethics Committee of The First Affiliated Hospital of Soochow University (approval No. 2024324). The requirement for informed consent was waived due to the retrospective nature of the analysis.

To comprehensively evaluate the extent of colonic lesions, we included patients with preoperative endoscopic biopsy-confirmed colorectal HGIN who underwent definitive surgical radical resection (including those receiving radical surgery after prior endoscopic resection), ensuring that the postoperative intestinal pathology specimens encompassed the mucosa, submucosa, muscularis propria, and serosa. To guarantee consistent and rigorous pathological assessment, patients treated solely with endoscopic resection were excluded, as their pathological specimens typically lack tissue beyond the mucosa and partial submucosa. The specific inclusion criteria were as follows: a diagnosis of colorectal HGIN via endoscopic biopsy, radical surgical resection (including postendoscopic resection), availability of complete corresponding postoperative pathology data, and completion of both PET/CT and CECT examinations preoperatively with an interval of ≤2 weeks between scans. Meanwhile, the exclusion criteria were as follows: tumors managed only with endoscopic therapy without subsequent radical surgery, absence of complete postoperative pathology data after radical resection, administration of chemotherapy without subsequent radical surgery, and completion of only CECT or PET/CT scans (not both).

Preoperative biopsy

Endoscopists at The First Affiliated Hospital of Soochow University performed all preoperative biopsies in this study. Following a standard bowel preparation, total colonoscopy (from the anal verge to the cecal pole) was conducted on 83 lesions. The macroscopic features, including tumor size, growth pattern, and the number of lesions, were documents. All specimens were then analyzed through standardized histopathological methods. Finally, each patient underwent curative-intent surgical resection within 2 weeks (2–13 days) after biopsy.

PET/CT

All imaging was performed on an integrated PET/CT system (Discovery STE 16, GE HealthCare, Chicago, IL, USA). Prior to the intravenous administration of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG; 4.07–5.55 MBq/kg), patients fasted for a minimum of 6 hours. Blood glucose levels were checked 1 hour before tracer injection, and studies were deferred if glucose exceeded 150 mg/dL. Following an uptake period of 60±10 minutes after ¹⁸F-FDG injection, a low-dose CT scan was acquired from the skull base to the upper thighs for both attenuation correction and diagnostic purposes under the following parameters: 140 kV, 120 mA, a transaxial field of view of 70 cm, a pitch of 1.75, a rotation time of 0.8 s, and a slice thickness of 3.75 mm. PET imaging was performed in either the two- or three-dimensional mode, with an acquisition time of 3 minutes per bed position. Image reconstruction included a standard ordered subset expectation maximization algorithm provided by the device manufacturer. Both visual and semiquantitative assessments were carried out by two nuclear medicine physicians with 10 and 24 years of experience, respectively.

CECT

Chest and abdominal CT examinations were performed with multidetector row CT scanners from major vendors (Philips, GE HealthCare, and Siemens Healthineers). The scanning parameters were configured as follows: a detector collimation of 1–1.25 mm, a field of view ranging from 20 to 38 cm, a beam pitch between 0.800 and 1.396, a beam width of 10–40 mm, a matrix size of 512×512, a tube voltage of 100–120 kV, a tube current of 250–60 mA, and a section interval and thickness of 5 mm. After unenhanced imaging, the contrast agent iopromide (Ultravist 300, Bayer, Berlin, Germany) was administered intravenously via a power injector (CT Plus 150, Ulrich Medical, Ulm, Germany) at a flow rate of 3–4 mL/s, with a total dose of 100–120 mL.

Image and data analysis

Interpretation of ¹⁸F-FDG-PET/CT and CECT examinations was conducted separately by a nuclear medicine specialist and an oncologic radiologist. Both interpreters lacked access to findings from alternative imaging procedures or pathologic results. ¹⁸F-FDG uptake was deemed abnormal if it exceeded background levels and deviated from expected anatomical patterns. Lesions were identified through combined visual assessment and semiquantitative measurement of focally elevated ¹⁸F-FDG activity, with rigid standardized uptake value (SUV) thresholds being avoided. CECT assessments focused on colorectal abnormalities, lymph node characteristics, and metastatic disease in distant organs. Lymph nodes were classified as metastatic if they exhibited one or more of these morphologic features: effaced fatty hilum, contrast enhancement, irregular margins, cortical thickening, increased dimensions, or elevated count. The primary analytical approach involved assessment of individual lesions. A supplementary assessment evaluated overall patient status. Discrepancies in lesion quantification between the two imaging modalities were subsequently resolved. Validation of imaging findings consisted of either histopathologic confirmation or longitudinal clinical monitoring (mean duration: 12 months). Confirmation methods comprised pathologic examination of image-detected lesions or clinical surveillance incorporating tumor marker analysis and supplementary imaging [CT, PET/CT, magnetic resonance imaging (MRI), or ultrasound; for osseous metastases, follow-up with bone scintigraphy and/or CT/MRI].

Surgical decision-making

All surgical procedures were performed by board-certified colorectal surgeons. The operative approach (laparoscopic vs. open) and extent of resection (segmental colectomy vs. total mesorectal excision) were guided by preoperative imaging findings. Discordant PET/CT and CECT results were resolved through multidisciplinary tumor board review.

Clinicopathologic data

Clinicopathological data were prospectively recorded, encompassing age, gender, pathological findings of biopsy specimens, and pathological findings of surgically resected specimens (location of the tumor, mode of growth, maximum size of the tumor, histological classification, depth of infiltration, and metastatic status of lymph nodes and other organs). The tumor-node-metastasis (TNM) classification system for gastrointestinal malignancies was applied and included stage 0 (malignant cells confined exclusively to the mucosal epithelium—the innermost colonic/rectal layer), stage I [tumor invasion extending through the lamina propria into the submucosa (T1) or muscularis propria (T2), without regional nodal involvement (N0) or distant metastasis (M0)], stage II [neoplasms penetrating to the subserosa (T3) or perforating the visceral peritoneum/invading adjacent structures (T4) with an absence of regional lymph node metastasis or distant dissemination], stage III [regional lymph node metastasis (N1–N2) present irrespective of primary tumor depth], and stage IV (distant metastatic dissemination to organs or extra-regional lymph nodes, independent of T/N status).

Statistical analysis

Statistical analysis was performed with SPSS software 23.0 (IBM Corp., Armonk, NY, USA). Data were analyzed on an individual lesion basis as some patients had multiple lesions. Statistical analysis included descriptive and inferential statistics. Continuous data are expressed as the mean and standard deviation (SD). Categorical variables are expressed as frequencies (n) and percentages. Receiver operating characteristic (ROC) curves were used to assess and compare the efficacy of PET/CT and CECT in distinguishing primary lesions, lymph node metastases, liver metastases, omental metastases, and bone metastases, with the area under the curve (AUC) values for each modality being calculated separately. The results are presented as the odds ratio (OR) with 95% confidence intervals (CIs). The DeLong test was used for the comparative analysis of AUC values between PET/CT and CECT. Univariate analyses were performed via the Chi-squared or Fisher exact test for categorical variables and the Student t-test for continuous variables. Multivariate analysis was performed with multivariate logistic regression models to identify the clinicopathologic features associated with the underdiagnosis of ISM in CRC biopsy specimens. P<0.05 was considered statistically significance.

Results

Characteristics of the study group

During the study period, a total of 76 patients (83 lesions), 48 men and 28 women, with a mean age of 65.06±10.72 years, were included in our study. Of the 76 patients, 71 (93.4%) showed isolated lesions and 5 (6.5%) were found to have multiple lesions for a final total of 83 lesions. We excluded 466 patients who only underwent endoscopic resection, opted directly for chemotherapy, or who were lost to follow-up. Prior to surgical resection, 30 patients from this group were subjected to both PET/CT and CECT examinations. Figure 1 provides the flowchart of the study. The baseline characteristics of the colorectal HGIN lesions with examinations are listed in Table 1. Among the 83 lesions, the distribution was as follows: 19 in the ascending colon, 7 in the transverse colon, 15 in the descending colon, 34 in the sigmoid colon, and 8 in the rectum. The maximum tumor diameter ranged from 5 to 70 mm. As shown in Table 1, the upgrade rate of the pathological diagnosis was 69.8% (58/83). The most commonly identified submucosal invasive lesion was moderately differentiated adenocarcinoma, present in 39 out of the 83 (47.0%) lesions, followed by poorly-to-moderately differentiated adenocarcinoma in 16 (19.3%) lesions, mucinous adenocarcinoma in 2 (2.4%) lesions, and poorly differentiated adenocarcinoma in 1 (1.2%) lesion, as outlined in Table 1. Of the lesions with no upgraded pathologic diagnosis, 24 (28.9%) were diagnosed as HGIN and 1 (1.2%) as inflammatory bowel disease (IBD). We performed TNM staging on the lesions. Among the 83 lesions analyzed, after exclusion of one IBD lesion, stage III and stage IV lesions accounted for 21.9% and 31.7% of the total, respectively. Furthermore, among a subset of 30 lesions, after exclusion of one IBD lesion, stage III and stage IV lesions accounted for 24.1% and 37.9% of the total, respectively.

Figure 1 Flowchart of patient recruitment. CECT, contrast-enhanced computed tomography; EFB, endoscopic forceps biopsy; HGIN, high-grade intraepithelial neoplasia; PET/CT, positron emission tomography-computed tomography.

PET/CT and CECT results

Among the 83 lesions, 30 underwent PET/CT and CECT imaging assessments prior to surgical resection. The general characteristics of patients evaluated by PET/CT and CECT are presented in Table 1. In this cohort, PET/CT correctly detected the false-negative lesions of CT in 4 of 5 (80.0%) primary lesions, in 2 of 4 (50.0%) lesions of lymph node metastasis, in 1 of 2 (50.0%) lesions of liver metastasis, in 3 of 3 (100%) lesions of omentum metastasis, and in 2 of 4 (50.0%) lesions of bone metastasis (Table 2). An example of a false-negative lesion in the sigmoid colon diagnosed by CECT is presented in Figure 2. In addition, PET/CT detected 5 of 5 (100%) lesions with omentum metastasis and 8 of 9 (88.9%) lesions with liver metastasis. However, CECT detected 2 of 5 (40.0%) lesions with omentum metastasis and 7 of 9 (77.8%) lesions with liver metastasis. The sensitivity, positive predictive value (PPV), and negative predictive value (NPV) of PET/CT were higher than those of CECT for the detection of the primary lesion, lymph node metastasis, liver metastasis, omentum metastasis, and bone metastasis. Figure 3 presents a case of a lymph node metastasis in which CECT produced a false-negative result despite pathological confirmation, with the absence of corresponding histopathological images attributable to specimen fragmentation during lymph node dissection.

Figure 2 A 77-year-old woman underwent ¹⁸F-FDG PET/CT and CECT for an endoscopic finding of a sigmoid colon mass. PET/CT revealed an abnormal ¹⁸F-FDG uptake in the sigmoid colon, while CECT indicated no abnormal enhancement in the sigmoid colon. The patient underwent radical sigmoid and rectal surgery. (A-D) Abnormal ¹⁸F-FDG PET/CT uptake in the sigmoid colon (SUVmax: 29.44). (E,F) A 1.3 cm × 1.1 cm intraluminal lesion arising from the wall of the sigmoid colon with homogeneous soft tissue attenuation on a pre-enhanced scan (CT value: 31 HU) and homogeneous enhancement with contrast (CT value: 81 HU). (G) Endoscopic sigmoid colon lesion. (H) The pathologic pattern of high-grade intraepithelial neoplasia on endoscopic biopsy. (I) The pathologic pattern of moderately differentiated adenocarcinoma in a postoperative sigmoid colon lesion. Figures H and I show images magnified at 100×. The staining method used is hematoxylin and eosin staining. The red arrow points to the location of the lesion. ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose; CECT, contrast-enhanced computed tomography; CT, computed tomography; HU, Hounsfield unit; PET/CT, positron emission tomography-computed tomography; SUV, standardized uptake value.

Figure 3 A 53-year-old woman underwent ¹⁸F-FDG PET/CT with CECT for an endoscopic finding of a sigmoid colon mass. PET/CT revealed abnormal ¹⁸F-FDG uptake in the sigmoid colon and pelvic lymph nodes, while CECT indicated no abnormal enhancement of the patient’s pelvic lymph nodes. (A-C) Abnormal ¹⁸F-FDG PET/CT uptake in the sigmoid colon (SUVmax: 9.38). (D,E) Abnormal ¹⁸F-FDG PET/CT uptake of the 1.2 cm × 0.8 cm pelvic lymph node (SUVmax: 2.32). (F,G) Pelvic lymph nodes with homogeneous soft tissue attenuation on a pre-enhanced scan (CT value: 35 HU) and homogeneous enhancement with contrast (CT value: 51 HU). The red arrow points to the location of the lesion. ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose; CECT, contrast-enhanced computed tomography; CT, computed tomography; HU, Hounsfield unit; PET/CT, positron emission tomography-computed tomography; SUV, standardized uptake value.

During the comparative analysis of lesions examined with PET/CT and CECT, discrepancies in findings were recorded and categorized according to primary lesions, metastatic lymph nodes, and distant organ metastases (including of the liver, peritoneum, and bone) (Table 3). Each site of discrepancy was subsequently analyzed and followed up on (Table 4). Pathological confirmation revealed that PET/CT correctly identified 5 lesions that had been assessed as negative by CECT. Lymph node metastases were classified as regional or distant based on their anatomical distribution. In one instance, CECT detected regional lymph node metastasis that was interpreted as negative on PET/CT, and subsequent surgical pathology confirmed the PET/CT finding. Conversely, for two cases of distant lymph node metastasis deemed negative by CECT, PET/CT interpretations were positive, and follow-up imaging confirmed the PET/CT findings. Regarding distant organ metastases, one liver lesion interpreted as positive on CECT was read as negative on PET/CT, and pathology subsequently confirmed it was positive. However, for all other discrepant findings involving the liver, bone, and peritoneal lesions, the accuracy of PET/CT was confirmed through subsequent pathological examination, MRI, or scheduled CT follow-up of the patients.

ROC curves were analyzed to evaluate the diagnostic performance of PET/CT and CECT in distinguishing primary lesions and various metastatic sites, including the lymph nodes, liver, omentum, and bone. ROC curves were analyzed, and AUC values were computed for each imaging modality. Notably, PET/CT demonstrated significantly higher AUC values than did CECT for primary lesions, lymph node metastases, and peritoneal metastases (P<0.05; Figure 4).

Figure 4 ROC analyses comparing the diagnostic values of ¹⁸F-FDG PET/CT and CECT for the detection of (A) primary lesions, (B) lymph node metastasis, (C) liver metastasis, (D) bones metastasis, and (E) omentum metastasis with an EFB diagnosis of HGIN. The AUCs were separately calculated, and the statistical differences of AUC values between ¹⁸F-FDG PET/CT and CECT were then compared (P<0.05). ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose; AUC, area under the curve; CECT, contrast-enhanced computed tomography; CI, confidence interval; EFB, endoscopic forceps biopsy; HGIN, high-grade intraepithelial neoplasia; PET/CT, positron emission tomography-computed tomography; ROC, receiver operating characteristic.

Risk factors associated with histological upgrade

Univariate logistic regression analyses (Table 5) demonstrated that postoperative histological upgrade of colorectal lesions was associated with patient age >60 years (OR 3.111; 95% CI: 1.060–9.128; P=0.039), pedunculated lesion (OR 0.069; 95% CI: 0.017–0.283; P<0.001), surface ulceration (OR 8.211; 95% CI: 2.671–25.239; P<0.001), and maximum dimension >2 cm (OR 8.889; 95% CI: 2.880–27.431; P<0.001). Next, the potential factors with P<0.05 were subsequently included in a multivariate regression analysis to identify independent associated risk factors for histological upgrade (Table 5). The results revealed that lesion size >2 cm (OR 4.688; 95% CI: 1.111–19.778; P=0.035) and surface ulceration (OR 6.942; 95% CI: 1.360–35.439; P=0.020) were independent risk factors for histological upgrade, whereas a pedunculated lesion (OR 0.179; 95% CI: 0.032–0.994; P=0.049) was a protective factor for histological upgrading.

Discussion

HGIN, encompassing carcinoma in situ, intramucosal carcinoma, and HGD, is defined as the stage at which there is no involvement of the muscularis mucosa. Only when the neoplastic lesions invade through the muscularis mucosae into the submucosa can they be defined as invasive CRC (6). There is consensus that colorectal tumor lesions without ISM have little potential for lymph node metastasis, while submucosal invasive CRC has a significantly higher potential for lymphatic dissemination (15). Due to its low invasiveness, endoscopic submucosal dissection has become the recommended treatment modality for HGIN. Conversely, invasive CRC primarily requires surgical intervention as opposed to endoscopic resection. Therefore, securing an accurate diagnosis to guide the appropriate therapeutic strategy, whether endoscopic resection or surgery, is crucial for managing patients with CRC.

Although EFB can be used as a basic diagnostic tool for diagnosing colorectal HGIN, discordant results are often observed when the biopsy result from EFB and the final pathologic diagnosis from endoscopic resection or surgical resection are compared. The reported confirmation rate of biopsy-diagnosed HGIN progression to invasive CRC after surgical resection ranges from 58.7% to 79.3% (7,16,17). In our study, the upgrade rate of HGIN to invasive CRC was as high as 69.8% (58/83), indicating the need for aggressive management of biopsy-diagnosed HGIN. The discrepancy between postoperative pathology and endoscopic biopsy findings may be attributed to the following factors. First, CRC is a heterogeneous disease, and minor biopsy specimens cannot be considered representative of the entire lesion (18). Second, the atypia of adenoma and adenocarcinoma is too subtle for detection in a small biopsy specimen (19). Third, the damage to the muscularis mucosae by cancer cells contributes to the failure in the diagnosis of ISM (20). In summary, biopsy-based pathologic diagnosis is insufficient for assessing the severity of colorectal tumors. Patients requiring endoscopic resection or surgical treatment for cancer may receive treatment for benign disease based on an incorrect diagnosis, which may lead to additional treatment in the future. Therefore, it is imperative to devise strategies to supplement EFB to reduce discrepancies in escalation assessment. Notably, five biopsy-confirmed HGIN cases in our cohort demonstrated hepatic, pulmonary, and osseous metastases on combined PET/CT or CECT imaging. These findings prompted direct initiation of systemic chemotherapy, enabling avoidance of nonindicated surgical interventions. This underscores the critical role of comprehensive preoperative imaging in accurate disease staging and optimal therapeutic decision-making. Detailed clinical characteristics of these patients are documented in Table S1.

Compared with traditional imaging techniques, PET/CT can discern not only morphological changes but also changes in glucose metabolism, which improves the validity of early diagnosis and reveals active tumor tissue (21). The greatest strength of PET lies in its ability to cover the entire body and detect distant sites of disease. One meta-analysis found that PET/CT performs well in the preoperative assessment of primary lesions, with a specificity and AUC of 99% and 96%, respectively, significantly better than those CECT (22). In our study, the primary lesion on PET/CT showed increased ¹⁸F-FDG uptake and thickening of the bowel wall, whereas on CECT, it showed thickening of the bowel wall and different degrees of contrast enhancement. Moreover, the specificity of PET/CT and CECT was 82.1% and 96.4, respectively, while the AUC was 0.731 and 0.661, respectively, with the diagnostic efficacy of PET/CT for primary lesions being significantly higher than that of CECT (P<0.05). Figure 2 presents a case in which CT showed a soft tissue lesion in the sigmoid colon that was indistinguishable from the intestinal contents despite its homogeneous enhancement with contrast enhancement, resulting in a false-negative finding on CT. However, PET/CT showed markedly increased glucose metabolism in the lesion [maximum SUV (SUVmax): 29.44], which led to it being considered a malignant lesion (23). For detection of regional lymph node metastasis, one retrospective study reported that PET/CT had superior specificity compared to CT and comparable sensitivity (24,25). Figure 3 presents a case in which the pelvic lymph node shown by CT was enlarged, but its contrast enhancement was not significant, and the lymph node did not show obvious imaging signs of metastasis (26,27), resulting in a false-negative CT result. However, PET/CT showed increased glucose metabolism in this lymph node, and it was therefore considered to be a metastatic lymph node (28). Regarding distant metastases, it has been shown that PET/CT is more sensitive than is CT in detecting liver metastases and lung metastases and in identifying other sites of intra-abdominal disease (29,30). In our study, 43.3% of patients had lymph node metastasis and 36.7% had distant organ metastasis. The PPVs and NPVs for diagnosing lymph node metastasis and distant organ metastasis were higher with PET/CT than with CECT. Therefore, patients who have biopsy-proven HGIN but are confirmed to have invasive CRC by surgical specimen analysis can receive more accurate tumor staging via PET/CT, which will allow them to choose a more appropriate treatment. Notably, a relatively high proportion of cases with distant metastasis was observed in this study. This may be attributed to the fact that all included cases underwent radical surgery, whereas patients in other similar studies (16-18,31) received subsequent treatments including both endoscopic resection and radical surgery. However, to ensure the consistency and rigor of postoperative pathology, case selection in this study was restricted to surgically treated patients, a selection bias that might have contributed to the higher proportion of distant metastasis. Another contributing factor could be the small sample size of this study, which might also account for this elevation.

The use of PET/CT for screening is not clinically justified, as it is associated with high costs, complex examination protocols, and limited accessibility across healthcare facilities. Consequently, there is a compelling need to elucidate the clinicopathologic characteristics of HGIN pathologically upgraded to invasive CRC. Clinicians are reminded of the risk of HGIN being underdiagnosed endoscopically and that further PET/CT may be needed to assess the patient’s systemic condition. Several studies have examined the risk factors associated with CRC after surgery for HGIN lesions, including a lesion diameter of >1 cm, indurated ulcers, irregular contours, malformations, tumor necrosis, desmoplastic stroma, and adjacent mucosal invasion (16,17). Compared to other studies (17,31), which only categorized macroscopic shape, into elevated, flat, and depressed, our study categorized types in more detail into broad-based, pedunculated, flat, and depressed lesions. In our study, the multivariate analyses indicated that a lesion diameter of >2 cm (OR 4.688; 95% CI: 1.111–19.778; P=0.035), pedunculated shape (OR 0.179; 95% CI: 0.032–0.994; P=0.049) and surface ulceration (OR 6.942; 95% CI: 1.360–35.439; P=0.020) were significantly associated with histopathologic discrepancies in EFB. A pedunculated lesion is a protective factor for histologic escalation, meaning that the possibility of pathologic escalation of endoscopically diagnosed HGIN is reduced when the lesion has a pedunculated shape.

Despite the extensive clinical adoption of established imaging techniques such as CT and MRI, economic analyses involving PET/CT remain comparatively sparse. This scarcity likely reflects prevailing perceptions of PET/CT as high-cost diagnostic methods. Countering this view, von Schulthess et al. (32) proposed a significant cost-efficiency for PET/CT in oncological diagnostic pathways. Standard CRC imaging protocols (typically liver ultrasound, CT, and MRI) demonstrate critical shortcomings, particularly in restaging precision. Only 30–40% of liver resections achieve curative intent, leading to numerous palliative surgeries and avoidable healthcare expenditures. Compared to CT-dominant diagnostic strategies, PET/CT reduces surgical interventions by 88%, thereby lowering treatment costs (33). Additional study indicate that ¹⁸F-FDG PET can prevent 2.8% of redundant surgeries in CRC cases with post-treatment elevated carcinoembryonic antigen levels (34). In a follow-up study of patients undergoing ablation for CRC liver metastases, Schnitzer et al. (35) developed a decision-analytic model using Markov simulation to estimate lifetime costs and quality-adjusted life years (QALYs). The base-case analysis found a total cost of USD $28,625.08 and an effectiveness of 0.755 QALYs for CECT, compared to USD $29,239.97 and 0.767 QALYs for ¹⁸F-FDG PET/CT. The resulting incremental cost-effectiveness ratio for ¹⁸F-FDG PET/CT was USD $50,338.96 per QALY, demonstrating it to be a cost-effective alternative for follow-up imaging after percutaneous ablation of colorectal liver metastases.

This study involved several limitations that should be noted. First, we employed a single-center, retrospective design with a relatively modest sample of included lesions, which could have introduced statistical variability. Second, to ensure consistency and rigor in postoperative pathological assessment, we specifically enrolled only cases with biopsy-confirmed HGIN who underwent surgical resection, consequently excluding cases managed by endoscopic resection. These inclusion criteria might have introduced selection bias. To address these limitations, we plan to initiate a multicenter, prospective study that will include patients undergoing endoscopic resection, thereby expanding the target population and increasing the sample size. This will allow more robust identification of patient subgroups most likely to benefit from PET/CT evaluation, ultimately providing clearer guidance for clinical practice. Furthermore, to enhance the clinical and health economic relevance of our findings, we intend to develop a decision-analytic model to systematically evaluate the cost-effectiveness and utility of PET/CT versus that CECT. This analysis will provide a rigorous economic assessment of PET/CT within this specific diagnostic context, offering valuable evidence for both clinicians and health policy makers.

Conclusions

Our findings indicate that invasive CRC is at risk of being underdiagnosed as HGIN by EFB. Compared to CECT, PET/CT improves the endoscopic diagnosis of HGIN by better evaluating the patient’s condition from various aspects such as primary foci, lymph node metastasis, and distant organ metastasis. When endoscopic lesions present with a diameter greater than 2 cm and superficial ulceration, PET/CT examination may be warranted.

Acknowledgments

We thank the radiologists of The First Affiliated Hospital of Soochow University for technical support.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-852/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-852/dss

Funding: This study was funded by the National Natural Science Youth Foundation of China (No. 82402324).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-852/coif). All authors report that this study was funded by the National Natural Science Youth Foundation of China (No. 82402324). The authors have no other 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. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics committee of The First Affiliated Hospital of Soochow University (No. 2024324) and individual consent for this retrospective analysis was waived.

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. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127:2893-917.
2. Nash GM, Weiser MR, Guillem JG, Temple LK, Shia J, Gonen M, Wong WD, Paty PB. Long-term survival after transanal excision of T1 rectal cancer. Dis Colon Rectum 2009;52:577-82.
3. MacDonald AW, Tayyab M, Arsalani-Zadeh R, Hartley JE, Monson JR. Intramucosal carcinoma on biopsy reliably predicts invasive colorectal cancer. Ann Surg Oncol 2009;16:3267-70.
4. Sung HY, Cheung DY, Cho SH, Kim JI, Park SH, Han JY, Park GS, Kim JK, Chung IS. Polyps in the gastrointestinal tract: discrepancy between endoscopic forceps biopsies and resected specimens. Eur J Gastroenterol Hepatol 2009;21:190-5.
5. Park YJ, Kim GH, Park DY, Lee S, Lee MW, Lee BE, Song GA. Histopathologic discrepancies between endoscopic forceps biopsy and endoscopic resection specimens in superficial esophageal squamous neoplasms. J Gastroenterol Hepatol 2019;34:1058-65.
6. Assarzadegan N, Montgomery E. What is New in the 2019 World Health Organization (WHO) Classification of Tumors of the Digestive System: Review of Selected Updates on Neuroendocrine Neoplasms, Appendiceal Tumors, and Molecular Testing. Arch Pathol Lab Med 2021;145:664-77.
7. Tominaga K, Fujinuma S, Endo T, Saida Y, Takahashi K, Maetani I. Efficacy of the revised Vienna Classification for diagnosing colorectal epithelial neoplasias. World J Gastroenterol 2009;15:2351-6.
8. Aust DE, Rüschoff J. Polyps of the colorectum: non-neoplastic and non-hamartomatous. Pathologe 2011;32:297-302.
9. Aarons CB, Shanmugan S, Bleier JI. Management of malignant colon polyps: current status and controversies. World J Gastroenterol 2014;20:16178-83.
10. Olsen ASF, Gundestrup AK, Kleif J, Thanon T, Bertelsen CA. Accuracy of preoperative staging with multidetector computed tomography in colon cancer. Colorectal Dis 2021;23:680-8.
11. Nerad E, Lambregts DM, Kersten EL, Maas M, Bakers FC, van den Bosch HC, Grabsch HI, Beets-Tan RG, Lahaye MJ. MRI for Local Staging of Colon Cancer: Can MRI Become the Optimal Staging Modality for Patients With Colon Cancer? Dis Colon Rectum 2017;60:385-92.
12. Mettler FA Jr, Bhargavan M, Faulkner K, Gilley DB, Gray JE, Ibbott GS, Lipoti JA, Mahesh M, McCrohan JL, Stabin MG, Thomadsen BR, Yoshizumi TT. Radiologic and nuclear medicine studies in the United States and worldwide: frequency, radiation dose, and comparison with other radiation sources--1950-2007. Radiology 2009;253:520-31.
13. Nørgaard A, Dam C, Jakobsen A, Pløen J, Lindebjerg J, Rafaelsen SR. Selection of colon cancer patients for neoadjuvant chemotherapy by preoperative CT scan. Scand J Gastroenterol 2014;49:202-8.
14. Almuhaideb A, Papathanasiou N, Bomanji J. 18F-FDG PET/CT imaging in oncology. Ann Saudi Med 2011;31:3-13.
15. Schlemper RJ, Kato Y, Stolte M. Diagnostic criteria for gastrointestinal carcinomas in Japan and Western countries: proposal for a new classification system of gastrointestinal epithelial neoplasia. J Gastroenterol Hepatol 2000;15:G49-57.
16. Wei XB, Gao XH, Wang H, Fu CG, Zheng WQ, Zheng JM, Zhang W, Liu LJ. More advanced or aggressive colorectal cancer is associated with a higher incidence of "high-grade intraepithelial neoplasia" on biopsy-based pathological examination. Tech Coloproctol 2012;16:277-83.
17. Liu M, Lu ZH, Wang QX, Zheng W, Pei XQ, Han F, Zhou JH, Lin X, Wan DS, Li AH. Diagnostic value, safety, and histopathologic discrepancy risk factors for endoscopic forceps biopsy and transrectal ultrasound-guided core needle biopsy in rectum lesions. Ann Transl Med 2019;7:607.
18. Lim H, Jung HY, Park YS, Na HK, Ahn JY, Choi JY, Lee JH, Kim MY, Choi KS, Kim DH, Choi KD, Song HJ, Lee GH, Kim JH. Discrepancy between endoscopic forceps biopsy and endoscopic resection in gastric epithelial neoplasia. Surg Endosc 2014;28:1256-62.
19. Kato M, Nishida T, Tsutsui S, Komori M, Michida T, Yamamoto K, Kawai N, Kitamura S, Zushi S, Nishihara A, Nakanishi F, Kinoshita K, Yamada T, Iijima H, Tsujii M, Hayashi N. Endoscopic submucosal dissection as a treatment for gastric noninvasive neoplasia: a multicenter study by Osaka University ESD Study Group. J Gastroenterol 2011;46:325-31.
20. Farina AR, Mackay AR. Gelatinase B/MMP-9 in Tumour Pathogenesis and Progression. Cancers (Basel) 2014;6:240-96.
21. Zhu A, Lee D, Shim H. Metabolic positron emission tomography imaging in cancer detection and therapy response. Semin Oncol 2011;38:55-69.
22. Ye Y, Liu T, Lu L, Wang G, Wang M, Li J, Han C, Wen J. Pre-operative TNM staging of primary colorectal cancer by (18)F-FDG PET-CT or PET: a meta-analysis including 2283 patients. Int J Clin Exp Med 2015;8:21773-85.
23. Fletcher JW, Djulbegovic B, Soares HP, Siegel BA, Lowe VJ, Lyman GH, Coleman RE, Wahl R, Paschold JC, Avril N, Einhorn LH, Suh WW, Samson D, Delbeke D, Gorman M, Shields AF. Recommendations on the use of 18F-FDG PET in oncology. J Nucl Med 2008;49:480-508.
24. Sasaki K, Kawasaki H, Sato M, Koyama K, Yoshimi F, Nagai H. Impact of Fluorine-18 2-Fluoro-2-Deoxy-D-Glucose Uptake on Preoperative Positron Emission Tomography/Computed Tomography in the Lymph Nodes of Patients with Primary Colorectal Cancer. Dig Surg 2017;34:60-7.
25. Kwak JY, Kim JS, Kim HJ, Ha HK, Yu CS, Kim JC. Diagnostic value of FDG-PET/CT for lymph node metastasis of colorectal cancer. World J Surg 2012;36:1898-905.
26. Rollvén E, Abraham-Nordling M, Holm T, Blomqvist L. Assessment and diagnostic accuracy of lymph node status to predict stage III colon cancer using computed tomography. Cancer Imaging 2017;17:3.
27. de Vries FE, da Costa DW, van der Mooren K, van Dorp TA, Vrouenraets BC. The value of pre-operative computed tomography scanning for the assessment of lymph node status in patients with colon cancer. Eur J Surg Oncol 2014;40:1777-81.
28. Bae SU, Won KS, Song BI, Jeong WK, Baek SK, Kim HW. Accuracy of F-18 FDG PET/CT with optimal cut-offs of maximum standardized uptake value according to size for diagnosis of regional lymph node metastasis in patients with rectal cancer. Cancer Imaging 2018;18:32.
29. Kemeny NE, Gonen M. Hepatic arterial infusion after liver resection. N Engl J Med 2005;352:734-5.
30. Niekel MC, Bipat S, Stoker J. Diagnostic imaging of colorectal liver metastases with CT, MR imaging, FDG PET, and/or FDG PET/CT: a meta-analysis of prospective studies including patients who have not previously undergone treatment. Radiology 2010;257:674-84.
31. Wei XB, Xin L, Hao J. Malignant recurrence and distal metastasis after complete local resection of colorectal "high-grade intraepithelial neoplasia": incidence and risk factors. Int J Colorectal Dis 2014;29:1467-75.
32. von Schulthess GK, Steinert HC, Hany TF. Integrated PET/CT: current applications and future directions. Radiology 2006;238:405-22.
33. Park KC, Schwimmer J, Shepherd JE, Phelps ME, Czernin JR, Schiepers C, Gambhir SS. Decision analysis for the cost-effective management of recurrent colorectal cancer. Ann Surg 2001;233:310-9.
34. Lejeune C, Bismuth MJ, Conroy T, Zanni C, Bey P, Bedenne L, Faivre J, Arveux P, Guillemin F. Use of a decision analysis model to assess the cost-effectiveness of 18F-FDG PET in the management of metachronous liver metastases of colorectal cancer. J Nucl Med 2005;46:2020-8.
35. Schnitzer ML, Froelich MF, Gassert FG, Huber T, Gresser E, Schwarze V, Nörenberg D, Todica A, Rübenthaler J. Follow-Up 18F-FDG PET/CT versus Contrast-Enhanced CT after Ablation of Liver Metastases of Colorectal Carcinoma-A Cost-Effectiveness Analysis. Cancers (Basel) 2020;12:2432.