Clinical relevance of perioperative changes in circulating tumor cells in resectable colorectal cancer.

Introduction
Colorectal cancer (CRC) is one of the most common gastrointestinal cancers, ranking third in cancer incidence globally and the second leading cause of cancer-related deaths, with nearly 1.8 million newly diagnosed patients and one million deaths each year [1, 2]. With the continuous development of medical diagnosis and treatment technologies, remarkable progress has been achieved in colorectal cancer. However, under current medical conditions, after surgery, chemoradiotherapy, and other treatments, nearly 10% of stage I, 20% of stage II, and more than 40% of stage III colorectal cancer patients experience recurrence and/or metastasis within 5 years, even if the lesion has been completely removed and there is no visible metastasis [3]. This indicates that latent metastasis may have already occurred during the discovery and treatment of the tumor or that tumor cells with the potential to proliferate and metastasize enter the bloodstream from the primary tumor site during surgical resection, leading to tumor recurrence and metastasis [4].
At present, tumor recurrence and metastasis are primarily monitored using B-ultrasonography, CT, MRI, serum tumor markers and histopathological examination. Serological tumor marker examination can indirectly reflect the status of tumor treatment; however, it cannot directly provide information regarding tumor progression. It also has little correlation with pathophysiology and its specificity remains insufficient [5]. Pathological examinations can provide good biological information and directly reflect the tumor cytology and molecular biological characteristics. However, because tumor cells constantly change during the progression process, the characteristics of relapsed and metastatic tumor cells are not exactly the same as those of the primary tumor, and their responses to drugs are also different. Because obtaining pathological specimens after tumor recurrence and metastasis is often challenging, histopathology cannot serve as a real-time monitoring method [6]. Therefore, there is an urgent need to develop new diagnostic and monitoring methods that can provide real-time information regarding treatment efficacy.
Circulating tumor cells (CTCs) are highly viable tumor cells with metastatic potential that spontaneously separate from the primary site or metastases, or are dislodged during treatment and enter peripheral blood circulation [7]. The entry of tumor cells into circulation is a critical step in the metastatic cascade, which is a multi-step process in which cancer cells that separate from the primary tumor and enter the bloodstream. CTCs in the blood of patients with tumors are regarded as potential seeds of metastasis and are one of the most direct factors contributing to tumor metastasis and recurrence [8]. CTCs are released into the peripheral blood from the primary tumor, and their quantity and morphological characteristics are directly related to the tumor stage, diagnosis, and prognosis. Hematological indicators such as CTCs and ctDNA, which can be collected repeatedly, are therefore more suitable for dynamic monitoring of therapeutic efficacy and drug resistance, while molecular analyses of CTCs can further aid in identifying novel therapeutic targets and resistance mechanisms, thereby supporting real-time precision therapy [9–12]. Therefore, the utilization of CTCs in blood for early diagnosis and treatment monitoring of diseases is of considerable clinical value. CTCs have been detected in the peripheral blood of patients with various malignancies, including colorectal, breast, prostate, and lung cancers [13–16].
The number of CTCs in the blood is extremely low; therefore, achieving high-purity CTC isolation from whole blood is an important prerequisite. In recent years, research on CTC separation based on microfluidic technology has advanced rapidly. Based on the principles employed in the separation process, they can be classified as labeled and unlabeled separation [17, 18]. Progress in separation techniques has enabled the isolation of circulating cells and molecular analyses to detect the spread of cancer cells during tumorigenesis. CTC detection of CTCs holds broad prospects in clinical applications. A potential stratification strategy can be devised based on CTC counts to identify patients who may benefit from more intensive treatments [19]. In cases with the same tumor stage, further stratification can be carried out according to the differences in CTC counts, guiding the determination of the treatment plan and indicating the stratification of patient prognoses [20].
Given the above-mentioned issues, we employed microfluidic chip technology to isolate tumor cells from the peripheral blood samples of patients with colorectal cancer before and after surgery. We quantified the captured CTCs to assess their clinical significance and aimed to establish the relationship between CTC changes and clinicopathological features and prognosis of colorectal cancer.

Materials and methods

Study design and sample collection
In this prospective study, peripheral blood samples were collected from colorectal cancer patients before and after surgery to detect circulating tumor cells (CTCs). The study aimed to investigate the dynamic changes of CTCs before and after colorectal cancer surgery, analyze their correlation with lymph node metastasis to determine the optimal threshold for predicting lymph node metastasis, and perform an exploratory analysis on the potential prognostic value of CTCs. A total of 103 patients with colorectal cancer admitted to the Anorectal Surgery Department of Jiangyin People’s Hospital between January 2023 and December 2024 were recruited. The inclusion criteria were as follows: (a) newly diagnosed with colorectal cancer by pathological biopsy. (b) No prior treatment, such as neoadjuvant chemotherapy or radiotherapy before specimen collection. (c) The patient had no history of other malignancies. (d) Underwent radical resection of colorectal cancer at our hospital. Exclusion criteria included: (a) Patients with autoimmune diseases. (b) Patients with a body temperature exceeding 37.3° C. (c) Patients with abnormally elevated white blood cells. (d) Patients with poor treatment compliance. (e) History of chronic inflammatory bowel disease or acute inflammatory/infectious disease. Among the 103 initially screened patients, 10 refused postoperative CTC testing, 8 samples failed CTC detection due to hemolysis, insufficient blood volume, or microfluidic tube blockage, and 4 were excluded due to incomplete clinical data collection. Eventually, 81 patients were enrolled in the study. This study was approved by the Ethics Committee of Jiangyin People’s Hospital, and all participants provided written informed consent (2023018). Paired preoperative and postoperative blood samples were collected from 81 patients. Peripheral blood samples were collected between 1 and 7 days before surgery (T0) and seven days after surgery (T1). Blood collection for T0 was completed over 24 h before surgery to minimize the influence of perioperative intervention.

CTC detection
5 ml of peripheral venous blood was collected from the upper extremities at T0 and T1. Peripheral blood samples from patients with a pathological diagnosis of adenoma were used as controls.
To ensure efficient and consistent separation of the spiral microfluidic chip, red blood cells (RBCs) were lysed to eliminate the interference of their large numbers on separation. Briefly, a 10 × sample volume of ammonium chloride (ACK) lysis buffer (Thermo Fisher Scientific) was added to the peripheral blood sample and incubated on a shaker for 5 min to lyse the RBCs. The samples were then centrifuged at 300 × g for 5 min, the RBC fragments were removed by clearing the supernatant, and the cell sediment was resuspended in a 5 × sample volume of phosphate-buffered saline (PBS, Thermo Fisher Scientific) to prepare the cell sample. The prepared cell sample was then injected into a spiral microfluidic chip with a trapezoidal cross-Sect [21] for label-free separation of CTCs. Finally, the CTCs in the collected solutions were identified and counted using immunofluorescence staining.

Immunofluorescence staining
After microfluidic chip processing, the collected samples were concentrated and transferred onto polysine adhesion slides (P4981; Thermo Fisher Scientific). The slide was then placed in a wet box and incubated for 30 min to ensure adhesion of all cells. Subsequently, the cells were fixed using a −20 °C methanol solution for 5 min. Next, a blocking solution composed of 3% bovine serum albumin (BSA, Sigma-Aldrich), 3% goat serum (Thermo Fisher Scientific), and 3% fetal bovine serum (FBS, Thermo Fisher Scientific) was applied, and the slide was incubated for 30 min for blocking. Following the blocking step, the cells were incubated with fluorophore-labeled Anti-Pan Cytokeratin (CK, Thermo Fisher Scientific) antibodies and fluorophore-labeled Anti-CD45 (Thermo Fisher Scientific) antibodies at 4 °C for 12 h. Finally, an anti-fade mounting medium containing DAPI (Vector Laboratories) was added to stain the cell nuclei. After staining, the slides were covered with a coverslip (Matsunami) and observed under a fluorescence microscope (ECLIPSE 80i; Nikon, Tokyo, Japan). Circulating tumor cells (CTCs) and white blood cells (WBCs) were identified and counted. CTCs typically exhibit high CK expression and a negative CD45 signal, whereas WBCs exhibit the opposite characteristics. Thus, cells with a CK+/CD45 -/DAPI + phenotype are defined as CTCs, and cells with a CK -/CD45+/DAPI + phenotype are identified as WBCs.

Determination of Circulating tumor cell-positive cells

Criteria for Determining CTCs

Nuclear Morphology: The nucleus of CTCs is irregularly nodular and lobulated. In contrast, abnormal neutrophils have rod-shaped and lobulated nuclei, whereas monocytes have horseshoe-shaped, kidney-shaped, or oval-shaped nuclei, and abnormal lymphoid cells have distinct nuclear structures. Comparison of cell nuclei with those of the surrounding normal white blood cells during slide reading is an important method for determining cell atypia.

Nucleoplasmic Ratio: The Nucleoplasmic ratio is greater than 0.8.

Nuclear Diameter: The nuclear diameter is greater than 24 μm.

Nuclear Staining Characteristics: The nucleus showed hyperchromatin and uneven staining (because the chromatin of CTCs increased with coarse particles, resulting in nuclear hyperchromatin). Additionally, the nuclear membrane thickened, became concave, and wrinkled, forming an irregular zigzag shape.

Nuclear Abnormalities: These include phenomena such as chromatin migration within the nucleus, nucleolar enlargement, or abnormal nuclear division.

The final confirmation of the CTCs was performed by the Microfluidic Laboratory of Southeast University, which issued a detailed report. Results were independently validated by two qualified laboratory professionals. If results are inconsistent, further review will be conducted by a third reviewer.

Clinical data
Relevant clinicopathological data of the patients were retrieved using an electronic medical record management system. These data included sex, age, tumor size, tumor site, pathological tissue subtype, degree of tumor differentiation, depth of invasion, lymphatic vascular invasion, nerve invasion, lymph node metastasis, TNM stage, and tumor markers such as CEA and CA199.

Statistical analyses
The Shapiro-Wilk normality test was used to examine whether the difference in CTCs before and after surgery followed a normal distribution. Because the differences did not adhere to a normal distribution, the Wilcoxon signed-rank test was used to compare the differences in CTCs before and after surgery. The paired chi-square test was used to compare the positive and negative rates of CTCs before and after surgery. Continuous variables were transformed into categorical variables and presented as absolute frequencies and percentages. Univariate and multivariate logistic analyses were conducted to identify risk factors for regional lymph node metastasis in patients with resectable colorectal cancer. For univariate analysis, the chi-square test was used. Variables with a P-value less than 0.2 in the univariate analysis were incorporated into the multivariable analysis. Multivariate logistic regression analysis was performed to elucidate the association between baseline variables and regional lymph node metastasis. The cut-off value for the continuous variable selected in the multivariable logistic regression analysis was estimated using the Receiver Operating Characteristic (ROC) curve analysis. The best cut-off value was determined based on the maximum Youden index. Disease-free survival (DFS) was analyzed using Kaplan–Meier survival curves and the log-rank test. All statistical analyses were performed using SPSS software (version 26.0 for Mac). Receiver operating characteristic (ROC) curves were plotted using GraphPad Prism (version 9.0; Mac). Statistical significance was defined as a two-sided P-value less than 0.05.

Results

Microscopic morphology of CTCs detected by microfluidic chips
CTCs identified using the microfluidic chip system were shown in Fig. 1. Fluorescence microscopy images depict malignant tumor cells isolated from the peripheral blood of patients with CRC. The cells were stained with DAPI (blue), CK (red), and CD45 (green).

Comparison of CTCs at T0 and T1
Currently, there is no standardized positivity threshold for CTC detection. Previous studies utilizing CellSearch technology in colorectal cancer have identified CTCs ≥ 3 to define positivity, which correlates with pathological characteristics, drug sensitivity, and prognosis. Accordingly, this study adopted a positive threshold CTCs ≥ 3 to define positivity for analysis [22]. In terms of CTC counts, the Shapiro-Wilk test showed that the difference between these two groups did not follow a normal distribution (P < 0.001), and a non-parametric test for paired samples was employed. The median number of CTCs was 16 in the preoperative patients and 4 in the postoperative patients. The results indicated a statistically significant reduction in CTCs after surgery (P < 0.001) (Table 1). Interestingly, the number of CTCs in eight patients increased after surgery. In these patients, CTC levels increase after surgery could be related to intraoperative manipulation, tumor compression, or venous contamination. The CTC test results of 20 controls with adenoma were negative.

Regarding the positive rate of CTCs, 23.4% of the patients were negative, and 76.6% were positive preoperatively at TO. The proportion of negative CTCs had increased to 43.2%, whereas the proportion of positive CTCs had decreased to 56.8% at T1. McNemar’s test (paired chi-square test) revealed a statistically significant difference in the proportion of negative CTCs before and after surgery (P < 0.001) (Table 2).

Relationship between lymph node metastasis and clinicopathological features in colorectal cancer patients
Among 81 patients with colorectal cancer, 36 had lymph node metastasis, whereas 45 did not. The chi-square test was used for univariate analysis. The results demonstrated that patients with T3-4 stage tumors, nerve invasion, vascular invasion, and positive preoperative CTCs were more prone to developing local lymph node metastasis, and these differences were statistically significant (P < 0.05). In contrast, factors such as sex, age, tumor size, tumor morphology, pathological type, degree of differentiation, tumor site, CEA, CA199, smoking, drinking, hematochezia, and local lymph node metastasis were not significantly correlated (P > 0.05) (Table 3).

Multivariate logistic regression analysis
In the univariate analysis, patients with stage T3-4 tumors, nerve invasion, vascular invasion, and positive preoperative CTCs were more likely to have local lymph node metastasis. To avoid missing important factors, variables with P < 0.2 were included in the multivariate logistic regression analysis.
Multifactorial analysis revealed that vascular invasion (compared to no vascular invasion; odds ratio (OR), 20.528; 95% confidence interval (CI), 4.296–98.086; P < 0.001) and positive preoperative CTCs (compared to negative preoperative CTCs; OR, 8.017; 95% CI, 1.138–56.482; P = 0.037) were significantly associated with regional lymph node metastasis in patients with CRC (Table 4).

ROC curve analysis for predicting lymph node metastasis
The ROC curve of CTCs for predicting regional lymph node metastasis in patients with CRC indicated an area under the curve (AUC) of 0.835 (95% CI: 0.743–0.927), optimal cutoff value of 15.5, sensitivity of 83.3%, and specificity of 75.6% (Fig. 2).

Prognostic analysis based on CTC status
Further prognostic analysis based on CTC status was performed. Comparison of postoperative CTC status revealed that patients with persistent CTC positivity after surgery had significantly shorter PFS than those who were CTC-negative (Fig. 3). Meanwhile, there was no statistically significant difference in PFS between patients who were CTC-positive before surgery but converted to negative postoperatively and those who were CTC-negative at both time points (Fig. 3).

Discussion
Current applications of CTCs in colon cancer primarily focus on the diagnosis and monitoring of progression. As tumor-derived cells carry specific molecular alterations, CTCs serve as valuable markers for micrometastases, enabling precise risk stratification of early-stage colon cancer [23]. Currently, there is no uniform positivity threshold for CTCs detection. This study adopted CTCs ≥ 3 as the positive threshold for analysis according to previous literature. The complexity of CTC-positive interpretation stems mainly from the diversity of CTCs detection methods, which may have different sensitivities and specificities, leading to differences in the interpretation of the results. In addition, the disease differences among the patients included in the study (such as tumor type and stage) and the different detection time points (such as preoperative and postoperative) also affect the CTCs detection results and their interpretation [24]. In a study by Sotelo et al., which examined CTCs and CRC prognosis, the CellSearch System was used to quantify CTCs in 472 patients, including166, 93, 57, and 34 patients with CTC ≥ 1, ≥ 2, ≥ 3, and ≥ 5, respectively. Among patients with recurrent disease, the proportion of patients with CTCs ≥ 1 was not significantly different from those without recurrence. In the multivariate analysis, CTCs ≥ 1 was not an independent prognostic factor for DFS (HR 0.97, P = 0.87) or OS (HR 0.96, P = 0.89) [25]. Bork et al. included 287 patients with CRC and used the CellSearch system to detect CTCs in the blood. They found that CTCs detection in blood (≥ 1 CTCs per 7.5 ml of blood) was associated with poorer overall survival (OS) (49.8 months vs. 38.4 months; P < 0.001) [26]. These experiments differed from the CTCs interpretation standard used in this study because they adopted different CTCs detection methods, and the blood collection volume and operation methods were different from those used in this study. Therefore, they could not provide reference standards for this study. In summary, there is no clear positive criterion for CTCs, and the results may vary owing to different detection methods and time points.
The results of this study further indicated that, at T0, the proportion of CTC-positive cases decreased to 56.8%. This finding suggests that with excision of the primary lesion, CTCs in the peripheral blood either decrease or disappear, highlighting the efficacy of surgical tumor treatment. In some cases, resection of the primary tumor and surgical removal of metastases can significantly enhance patient survival. Nevertheless, a significant number of patients experience relapse. Future research should focus on characterizing individual disseminated cancer cells to identify targeted molecular therapies and the mechanisms of immune escape [27].
In a study involving 131 patients with renal cell carcinoma, alterations in circulating tumor cells, mesenchymal tumor cells, and CTC white blood cells (CTC-WBCs) during the perioperative period were explored. Blood samples were collected using the CanPatrol technique before and three months after surgery. Contrastingly, postoperative total CTCs (≥ 6), positive mesenchymal - like CTCs (mCTCs), and positive CTC - WBC were significantly associated with recurrence and metastasis, and served as independent predictors of progression - free survival (PFS) deterioration [28]. Gazzaniga et al. measured perioperative CTC levels in colorectal cancer patients and discovered that the preoperative positivity rate was 37.2%, while the postoperative positivity rate was 60.5%, a statistically significant difference. Our results were consistent with these findings [29]. Sastre et al. found that patients with CRC with positive CTCs after surgery were more prone to recurrence and metastasis than those with negative CTCs [30]. Notably, in our study, the number of CTCs increased postoperatively in some patients. This can be attributed to several factors. Deep tumor infiltration and large tumor size may inevitably cause tumor compression during surgery, leading to the release of tumor CTCs into the bloodstream. Additionally, the random error inherent in single CTC detection may have contributed to the observed increase in CTC numbers after surgery.
The results of this study indicated that among 81 patients with CRC, 36 had lymph node metastasis, whereas 45 did not. The analysis revealed that patients with stage T3–4 tumors, nerve invasion, vascular invasion, and positive preoperative CTCs were more prone to developing local lymph node metastasis. In contrast, factors such as sex, age, tumor size, tumor morphology, pathological type, degree of differentiation, tumor site, CEA, CA199, smoking, drinking, hematochezia, and local lymph node metastasis exhibited a weaker correlation, which might be attributed to tumor progression. As the depth of invasion increases along with nerve and vessel invasion, an increasing number of tumor cells evade clearance. Moreover, a higher CTC positivity rate was observed, which increased the risk of tumor metastasis. CTC detection is instrumental in predicting the pathological stage of colorectal tumors. The higher the clinical stage of CRC, the greater the likelihood of a positive CTC result. Previous studies have demonstrated that colorectal cancer patients with an advanced tumor stage, high pathological grade, deep invasion, lymph node metastasis, vascular invasion, and nerve invasion have a higher CTC level, which is associated with a poor prognosis [31, 32]. The positive rate and count of CTCs correlated with N stage. This is because lymph node metastasis, to a certain extent, reflects the degree of distant tumor invasion and is directly related to tumor stage.
In this study, the initial threshold of ≥ 3 CTCs as the positive threshold was based on the previous literature using the CellSearch platform [22], and this cutoff has been widely applied in prognostic assessments. However, our ROC analysis indicated that the optimal cutoff value for predicting lymph node metastasis was 15.5 CTCs (AUC = 0.835), suggesting that, when using microfluidic technology, a higher CTC burden might be more strongly associated with the risk of metastasis. To verify the robustness of the results, a sensitivity analysis was conducted. It was found that when ≥ 3 and ≥ 15.5 CTCs were used respectively, the predictive power of CTCs in the multivariate model remained significant (p < 0.05), but the latter has a higher specificity. Future research should further optimize the CTC threshold standards for this platform.
In this study, multivariate analysis revealed that vascular invasion and positive preoperative CTCs were significantly associated with regional lymph node metastasis in colorectal cancer patients. Previous research have shown that the CTC count can mirror the tumor load and rises in tandem with the advancement of the tumor stage. Additionally, the number of CTCs correlates with tumor size, T stage, and lymph node metastasis, rendering it a potential marker for determining tumor invasion or metastasis [33]. In one study, serum samples were collected from 97 patients with CRC and 30 healthy volunteers. A CTC count of ≥ 2 CTCs per 7.5 ml was defined as positive. Among the patients with CRC, 34 tested positive for CTCs, and the positive patients showed a significant association with tumor stage (stage II, 20.7%; stage III, 24.1%; stage IV, 60.7%; P = 0.005) [34]. Similarly, Sotelo et al. discovered in their study that CTCs are closely linked to the depth of tumor invasion and are significantly associated with the presence of lymph node metastasis [25]. CTC detection is instrumental in predicting the pathological stage of colorectal tumors. Specifically, the higher the clinical stage of the CRC, the greater the likelihood of a positive CTC result.
To further explore the correlation between CTCs and lymph node metastasis in CRC, ROC curve analysis indicated that the ROC curve of CTCs for predicting regional lymph node metastasis in CRC patients demonstrated an AUC of 0.835. This finding suggests that CTCs have a relatively high value for predicting regional lymph node metastasis in patients with CRC. One of the crucial criteria for the pathological staging of CRC is lymph node metastasis, which typically implies poor prognosis. In clinical practice, treatment plans are usually formulated based on lymph node status [35, 36]. Positive lymph node metastasis indicated that the tumor had advanced to stage III or higher. Adjuvant chemoradiotherapy at stage III or higher can significantly improve patient prognosis [37]. Therefore, accurate preoperative prediction of lymph node metastasis or tumor stage can optimize the treatment plan to a certain extent. Our results support this conclusion.
Moreover, the number of CTCs in most patients decreases following surgical treatment. For patients with persistently positive CTCs after surgery, close follow-up is recommended. Preoperative serum CTC levels are closely associated with the pathological stage of CRC. According to the American Joint Committee on Cancer (AJCC) staging system, patients with lymph node metastasis are classified as stage III or IV, whereas those without lymph node metastasis are classified as stage I or II. Currently, it is recommended that patients with stage III (lymph node-positive) CRC undergo surgical resection followed by adjuvant chemotherapy. Nevertheless, only 20% of patients benefit from adjuvant chemotherapy and 80% are exposed to unnecessary drug toxicity [38]. The preoperative pathological staging of CRC can be evaluated by detecting peripheral blood tumor markers. Combined detection can enhance the predictive ability of CRC pathological staging, thereby guiding clinicians in formulating individualized treatment plans. With continuous progress in medicine, individualized precision therapy represents a current trend. Preoperative, simple, and readily available methods for predicting individualized treatment of patients with CRC at different stages will undoubtedly provide more options and hope to patients.
However, this study had several limitations. First, the spiral microfluidic chip used in current study may lose the small-sized CTCs, which is the inherent limitation of current size-based separation techniques. In our study, we have set the size threshold of our spiral microfluidic chip to be 12 μm to recover more CTCs, which means that the CTCs with sizes larger than 12 μm could be successfully recovered. As all the samples in our studies were processed using the same separation technique, the loss of small-sized CTCs would not affect the statistical conclusions in our study. Indeed, it is still challenging to recover all CTCs in the sampled blood using any separation technique. Second, given the sparse and uneven distribution of CTCs in peripheral blood and the controversial duration of their presence in the body, the difficulty and randomness of capturing CTCs through a single blood collection increase, leading to random detection errors. Thus, it is advisable to combine CTC detection with liquid biopsy methods such as circulating tumor DNA (ctDNA) to mitigate the limitations of CTC detection. Third, this study classified CTCs solely by number, without conducting a specific phenotype classification. Recent research has indicated that CTCs have multiple phenotypes, and the relationship between different CTC categories and tumor metastasis and staging requires further investigation [39]. Therefore, a classification based solely on numbers may not be the optimal approach. This study is also limited by its relatively small sample size, short follow-up duration, and exploratory nature of the prognostic analysis, all of which warrant cautious interpretation of the results and highlight the need for further validation in large-scale, prospective studies with extended observation periods.

Conclusions
This study summarized and analyzed the value of postoperative CTCs in resectable colorectal cancer. Specifically, we investigated the relationship between changes in CTCs during and after surgery and the clinicopathological features. Monitoring CTCs changes during and after surgery is important for making informed decisions regarding postoperative adjuvant treatment.