Prognostic Value of Preoperative Circulating Tumor Cell Counts in Patients with UICC-Stage I-IV Colorectal Cancer

The detection of CTCs in peripheral blood is one of the most promising approaches to identify disseminated disease in colorectal cancer (CRC). This study aims to evaluate the prognostic relevance of preoperative CTCs using the Cellsearch® system (CS)in patients, who underwent resection with curative intent of different stages of colorectal cancer (UICC I-IV). CTC analysis was performed in 68 CRC patients at UICC stages I-IV immediately before surgery. Data were correlated with clinicopathological parameters and patient outcomes. One or more CTCs/7.5 mL were detected in 45.6% (31/68) of patients. CTCs were detected in all stages of the Union of International Cancer Control (UICC), in stage I (1/4, 25%), in stage II (4/12, 33.3%), in stage III (5/19, 26.3%) in stage IV (21/33, 63.6%).The detection of CTCs was associated to the UICC stage (p = 0.035) and to the presence of distant overt metastases (p = 0.014). The presence of ≥ 1 CTCs/ 7.5 ml correlated signicantly with shorter progression-free (p = 0.013) and overall survival (p = 0.014). Multivariate analyses showed that preoperative CTCs are an independent prognostic indicator for overall survival (HR, 2.68; 95% CI, 1.05–6.92 7; p = 0.039, ≥ 1 CTC). In conclusion, detection of CTCs is an independent and strong prognostic factor in CRC, which might improve the identication of high-risk patients in future clinical trials.


Introduction
Colorectal cancer (CRC) is the 3rd most commonly diagnosed cancer and the 4th leading cause of death in the world with 1.3 million new cases annually (1). Two-thirds of the patients with CRC present with localized and potentially curable disease at diagnosis (TNM Stage I-III) (2). In these patients, surgery remains the most important treatment modality.
Currently, the overall 5-year survival is 65% (1). This increased in the last four decades after the introduction of screening programs and the concept of adjuvant chemotherapy (3,4). Surgical technique also evolved with the introduction of total mesorectal excision (TME) for rectal cancer and central venous ligature (CVL) with complete mesocolic excision (CME) for colon cancer, where sharp dissection in the embryological planes increases lymph node yields, subsequently improving staging and survival (5,6). Nevertheless, tumor recurrence or spread to distant sites and formation of metastases still occur in 20% of the patients and is the leading cause of death in these patients (7). Even in patients with apparently early stages TNM (I-II), local recurrence or distant metastases occur despite proper treatment (8).
Tumor persistence and progression occur mainly due to circulating tumor cells (CTCs) which are seeded from the primary tumor and target distant organs, where they eventually mature and cause a secondary metastasis (9). Liquid biopsy for identi cation of CTC preoperatively is of proven utility in predicting prognosis in breast, colon, and prostate cancer. (10)(11)(12).
The 8th AJCC Cancer Staging Manual expanded the de nitions of Tis, T4a, and M1 and nodal micrometastasis in CRC (13). However, unlike in breast cancer, new staging categories like M0 (i+), in which CTC or disseminated tumor cells in bone marrow are detected, are still lacking (14).
Identifying CTCs preoperatively in liquid biopsies could provide information to solve common dilemmas in CRC (15), like patient selection for adjuvant chemotherapy in stage II CRC as well as selecting patients for D2 or D3 lymphadenectomy in right hemicolectomy.
The isolation and molecular analysis of CTCs in peripheral blood is one of the most promising approaches to identify disseminated disease, in order to upgrade or downgrade the multimodal therapy. The CellSearch® (CS) (Menarini, Silicon Biosystems, Bologna, Italy), a known method for quanti cation of CTCs based on the expression of the epithelial cell adhesion molecule (EpCAM) and of keratin, is the rst standardized system approved by the U.S. Food and Drug Administration for capturing and detection of CTCs derived from metastatic breast and prostate cancer as well as metastatic CRC (10,16).
The aim of this study is to assess the preoperative value of CTCs in different stages of CRC on the overall survival and progression-free survival using CS.

Study Design
This prospective study was conducted at the University Hospital Hamburg-Eppendorf in Germany and enrolled 68 patients with different stages of colorectal cancer, who underwent surgical resection for CRC. Informed consent was obtained from all patients. The study was approved by the medical ethics committee of the Chamber of Physicians of Hamburg.
This study included only patients who underwent primary resection for colorectal cancer as well as patients after neoadjuvant radiotherapy for rectal cancer. Peripheral blood samples for CTC analysis were collected immediately before surgery and postoperatively during the rst 96 hours. Follow up was conducted according to S3 German Guidelines (17). Events considered were death, local recurrence, and distant metastasis. Overall survival was the time from operation to death or last follow-up; and progression-free survival was de ned as the time from operation to the diagnosis of tumor recurrence.

CTC Analysis
CTC analysis was performed using CS as previously described (18). Blood samples (7.5 mL) were collected in CellSave preservative tubes, stored at room temperature and processed within 96 hours after blood collection, according to the manufacturer's instructions. The accuracy and reproducibility of CS have been described previously (18,19). The presence of a nucleus, cytokeratin expression, round or oval cell morphology, and absent CD45 expression were the criteria for CTCs (18).
Statistical Analysis SPSS statistic software version 25 was used. Histological characteristics were expressed as descriptive statistics. The x 2 was used to investigate the association between CTCs and histopathological parameters. Survival rates were determined using the Kaplan-Meier method and were compared using the log-rank test. A multivariate analysis of factors that might in uence OS was performed using the Cox proportional hazards regression model. The results were presented as hazard ratios with 95% CI. All comparisons were two-tailed. A p-value of less than 0.05 was considered statistically signi cant.
All authors had access to the study data and had reviewed and approved the nal manuscript.

Results
Patient Characteristics and CTC Detection CTC analysis was performed in blood samples from 68 patients preoperatively. 38 patients had CTC analysis within the rst 96 hours after the procedure. Forty-three patients were nodal positive. Distant metastasis was present in 34 patients at the time of operation (Table 1). Of the 68 patients, 44 were males. The median age was 64.7 years (range, 18-88 years). Applying a cut-off of ≥ 1 CTC, 31 out of 68 (45%) patients were CTC-positive preoperatively and 13 out of 38 (19.1%) patients were postoperatively CTC-positive. In 5 out of 13 patients, CTCs were not present preoperatively.
We assessed the correlation between CTC detection preoperatively with sex, age and the following histopathologic parameters:

Univariate and Multivariate Analysis of Survival
The median survival time was 32 months. Kaplan-Meier's univariate analysis showed that patients with ≥ 1 CTC had signi cantly shorter progression-free (p = 0.008) as well as overall survival (p = 0.008) compared to CTC-negative patients (Fig. 1).
Nine clinicopathological factors were analyzed using Cox-Regression analysis. Only 4 factors correlated with survival in univariate analysis (Table 3). These included age, metastatic disease, UICC stage, preoperative CTC detection (p < 0.05). Gender, postoperative CTC detection, and other factors were not related to survival in the univariate analysis. The four factors signi cantly related to survival in univariate analysis were evaluated in the multivariate analysis, which showed that only advanced age and preoperative CTC detection were independent prognosticators of an unfavourable OS (Table 3).

Discussion
Despite advances in multimodal treatment, CRC is the second most common cause of cancer death worldwide (1). CTCs play a pivotal role in disease progression, metastasis, and recurrence (20). Aside from TNM classi cation and residual tumor status, novel tools and staging systems are needed for adequate prognostic staging and for guiding multimodal therapy (21). CTC identi cation has proven to be an important prognostic factor in breast (22). For this reason, a decade ago, 7th AJCC Cancer Staging Manual introduced the new category M0(i) for breast cancer, which is de ned by the presence of circulating or disseminated tumor cells not exceeding 0.2 mm detectable in bone marrow, circulating blood or other non-regional tissues of non-metastatic patients. In contrast, such a category is still lacking in the current CRC Staging system (23) due to the fact that the assessment of CTCs in CRC is still controversial (16) and different technical platforms have provided con icting results (24)(25)(26). Here, we applied the FDA-cleared Cellsearch® system, which uses immunomagnetic enrichment (EpCAM) and immunocytochemical (cytokeratin, CD45, DAPI) analysis. Compared to immunocytochemistry, CS is standardized, more time e cient and provides better CTC enrichment (16,18,27). In our cohort, we found CTCs in 45.6% % of the patients before surgical resection. CTCs were present in all stages of CRC. Even in early stage CRC, UICC I (T1-2, N0, M0), one of four patients had detectable CTCs.
Up to now, there is no consensus regarding the threshold used to de ne CTC positivity in CRC (16). Cohen et al have used the cut-off of ≥ 3 CTC/7.5 ml for de ning CTC positivity in metastatic CRC (12), while others have shown that cut-offs ≥ 1 CTC/7.5 ml and ≥ 2 CTC/7.5 ml were also associated with poor prognosis in CRC (28,29).
We used a strict cut-off of ≥ 1 CTC/7.5 ml. CTC detection before surgery was associated signi cantly with a shorter progression-free (P = 0.008) and overall survival (P = 0.008). Multivariate analyses identi ed CTCs as a strong, independent, prognostic indicator for overall survival.
Many studies have shown inferior survival for right-sided tumors (32,33), so we assessed the effect of tumor location on CTC enumeration. Right-sided colon cancer (RCC), left-sided colon cancer (LCC) and rectal cancer (RC) were considered independently, in light of the "three entities" hypothesis in colorectal cancer (34)(35)(36)

Conclusion
Preoperative CTC detection is an important prognostic marker for survival in CRC, which can be further developed as enrichment tool to study a high-risk population of CRC patients in clinical trials.

Declarations
Author contributions M.R and K.P. conceived the idea and designed the study. J.M. and K.K. contributed to the collection of the samples and clinical data. S.R., J.T., K.P analyzed the blood samples for CTC-detection. J.I., A.K. and N.M. supervised the study. T.A. analyzed the data and wrote the main manuscript text. M.R., T.A., K.P. S.R. were major contributors in writing the manuscript. All the authors reviewed the paper.
Additional information