Proper surgical extent for clinical Stage I right colon cancer.

PURPOSE: Pre-operative evaluation identifying clinical-stage affects the decision regarding the extent of surgical resection in right colon cancer. This study was designed to predict a proper surgical resection through the prognosis of clinical Stage I right colon cancer. PATIENTS AND METHODS: We included patients who were diagnosed with clinical and pathological Stage I right-sided colon cancer, including appendiceal, caecal, ascending, hepatic flexure and proximal transverse colon cancer, between August 2010 and December 2016 in two tertiary teaching hospitals. Patients who underwent open surgeries were excluded because laparoscopic surgery is the initial approach for colorectal cancer in our institutions.
RESULTS: Eighty patients with clinical Stage I and 104 patients with pathological Stage I were included in the study. The biopsy reports showed that the tumour size was larger in the clinical Stage I group than in the pathological Stage I group (3.4 vs. 2.3 cm, P < 0.001). Further, the clinical Stage I group had some pathological Stage III cases (positive lymph nodes, P = 0.023). The clinical Stage I group had a higher rate of distant metastases (P = 0.046) and a lower rate of overall (P = 0.031) and cancer-specific survival (P = 0.021) than the pathological Stage I group. Compared to pathological Stage II included in the period, some of the survival curves were located below the pathological Stage II, but there was no statistical difference.
CONCLUSION: The study results show that even clinical Stage I cases, radical resection should be considered in accordance with T3 and T4 tumours.

INTRODUCTION

Despite innovative changes in surgical treatment and global standardisation, unexpected recurrence and metastasis have been reported in potentially curable colon cancer, even in the early stages of the disease. The surgical treatment for right-sided colon cancer has been standardised;[1] however, treatment for early colon cancer is controversial, varying from endoscopic management to radical surgical resection. Excessive lymphatic dissection in the early stages of the disease can cause post-operative complications, while insufficient resection in the advanced stages of the disease can increase the risk of recurrence. Therefore, predicting the exact cancer stage preoperatively is important for determining the extent of surgical resection. This study was designed to predict the proper surgical extent through the prognosis of clinically early right-sided colon cancer.

PATIENTS AND METHODS

The study included patients diagnosed with right-sided colon cancer classified as clinical Stage I and pathological Stage I. All patients underwent surgical treatment at OOO hospital and OOOO hospital between 2010 and 2016, and the parameters were prospectively collected. The disease was staged according to the eighth edition of the tumour, node and metastasis classification of colon cancer proposed by the Union for International Cancer Control (UICC) and the American Joint Committee on Cancer. Appendiceal, caecal, ascending colon, hepatic flexure colon and proximal transverse colon cancer were included as right-sided colon cancer. Clinical Stage I was defined as T1 and T2 on abdominal computed tomography (CT) with negative lymph node metastasis. All the included cases were confirmed to have no metastasis on abdominal and chest CT. Radical lymph node dissection was performed on all patients, which refers to the case in which the principal nodes were included according to the Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines.[2] In these institutions, open surgeries performed in <5% of patients; therefore, this study only included patients who underwent laparoscopy.

The following pre-operative data were evaluated: age, gender, body mass index (BMI), American Society of Anaesthesiologists (ASA) score, past medical history, previous abdominal surgery and pre-operative carcinoembryonic antigen (CEA) level. Further, the following intra-operative parameters were assessed: tumour location, extent of surgical resection, operating time, and estimated blood loss. The following pathological factors were evaluated: tumour size, length of proximal/distal margins, number of retrieved/pathological lymph nodes, presence of vascular/lymphatic/neural invasion and type of differentiation. In addition, the following post-operative outcomes were assessed: post-operative CEA level, date of first flatus/defecation/diet, length of hospital stay and type of post-operative complications (the Clavien–Dindo classification). The 5-year disease-free survival (DFS), overall survival, and cancer-specific survival (CSS) were analysed based on the presence or absence of local and systemic recurrence and metastasis. The patients visited the hospitals quarterly for the first 2 years and semiannually thereafter for 3 years. During follow-up, serum CEA tests and abdomen and chest CT were performed semiannually and total colonoscopy was performed annually. Recurrence was defined as a pathologically identified or radiologically proven local or systemic metastasis. The study was approved by the Institutional Review Board of OOO Hospital (CNUH-2019-119).

Statistical analysis

The groups were compared using Student's t-test or the Mann–Whitney U-test for continuous data and the Chi-square test or Fisher's exact test for categorical data. The recurrence and survival rates were estimated using Kaplan–Meier curve analyses. Statistical analyses were performed using SPSS (IBM, SPSS Statistics, Version 20, Armonk, NY, USA). A P ≤ 0.05 was considered statistically significant.

RESULTS

Patients’ characteristics and operative outcomes

Eighty patients were diagnosed with clinical Stage I right-sided colon cancer, and 104 were diagnosed with pathological Stage I right-sided colon cancer. There was no difference in the groups in terms of age, gender, BMI, ASA score, past medical history (hypertension and diabetes mellitus) and previous abdominal surgery. Further, the tumour location, extent of surgical resection, operation time and estimated blood loss did not differ between the groups. None of the cases required conversion to open surgery [Table 1].

Table 1

Patients’ characteristics and operative outcomes

	Clinical Stage I (n=80), n (%)	Pathologic Stage I (n=104), n (%)	P
Median age, years (range)	68 (44-80)	66.4 (47-81)	0.275
Sex
Male	49 (61.2)	58 (55.8)	0.455
Female	31 (38.8)	46 (44.2)
BMI (range)	24.4 (19.4-32.8)	24.4 (19.5-33.5)	0.988
ASA
I	14 (17.7)	18 (17.6)	0.563
II	58 (73.4)	79 (77.5)
III	7 (8.9)	5 (4.9)
Past medical history
HTN	40 (50)	55 (52.9)	0.698
DM	21 (26.2)	27 (26)	0.965
Previous abdominal surgery	13 (16.2)	17 (16.3)	0.986
Pre-operative CEA (SD)	3.9 (6.2)	3.0 (3.6)	0.224
Tumour location
Appendix	0	2 (1.9)	0.809
Cecum	5 (6.2)	6 (5.8)
Ascending colon	46 (57.5)	58 (55.8)
Hepatic flexure	14 (17.5)	19 (18.3)
Proximal transverse colon	15 (18.8)	19 (18.3)
Operative extent
Right hemicolectomy	69 (86.2)	97 (93.3)	0.112
Extended right hemicolectomy	11 (13.8)	7 (6.7)
Operative duration, min (range)	145 (60-290)	144 (70-360)	0.911
Estimated blood loss, ml (range)	156.6 (10-800)	136.7 (10-500)	0.429
Open conversion	0	0

BMI: Body mass index, ASA: American Society of Anaesthesiologists, HTN: Hypertension, DM: Diabetes mellitus, CEA: Carcinoembryonic antigen, SD: Standard deviation

Pathologic outcomes

The tumour size was larger in the clinical Stage I group than in the pathological Stage I group (2.3 vs. 3.4 cm; P < 0.001). The frequency of neural invasion was higher in the clinical Stage I group than in the pathological Stage I group (15% [12 cases] vs. 5.8% [6 cases]; P = 0.037). Of the 184 patients included in the study, 154 (83.7%) patients had undergone radical lymph node dissection, including the principal node [Table 2].

Table 2

Pathologic outcomes

	Clinical Stage I (n=80), n (%)	Pathologic Stage I (n=104), n (%)	P
Case including principal node	68 (85)	86 (82.7)	0.674
Tumour size, cm (range)	3.4 (0.5-9)	2.3 (0.5-6)	<0.001
Proximal margin, cm (range)	13.9 (4-35)	15.2 (4-38)	0.208
Distal margin, cm (range)	12.3 (1-38)	11.4 (1-38)	0.403
Retrieved LN, n (range)	35.4 (10-98)	35.7 (3-89)	0.898
<12	1 (1.2)	4 (3.8)	0.283
Pathologic LN, n (range)	0.49 (0-13)	0	0.023
Invasion
Vascular invasion	10 (12.5)	5 (4.8)	0.059
Lymphatic invasion	10 (12.5)	6 (5.8)	0.108
Neural invasion	12 (15)	6 (5.8)	0.037
Differentiation
Well-differentiation	31 (38.8)	53 (51)	0.099
Moderate differentiation	35 (43.8)	29 (27.9)	0.025
Poorly differentiation	6 (7.5)	5 (4.8)	0.445

LN: Lymph node

Short-and long-term outcomes

There was no difference between the groups in terms of short-term post-operative outcomes. The length of the hospital stay was 8 days. There was no difference in post-operative complications, such as surgical site infection, intra-abdominal bleeding, anastomotic bleeding, anastomosis leakage, ileus and urinary retention, between the groups. No death was recorded till post-operative day 30 [Table 3].

Table 3

Post-operative outcomes

	Clinical Stage I (n=80), n (%)	Pathologic Stage I (n=104), n (%)	P
Post-operative details
Post-operative CEA (SD)	1.95 (1.4)	1.72 (1.1)	0.276
First flatus, day (range)	3 (1-6)	2.9 (1-7)	0.397
Defecation, day (range)	4.6 (2-9)	4.4 (2-9)	0.264
Diet, day (range)	2.7 (1-6)	2.5 (1-7)	0.325
Hospital stay, day (range)	7.7 (6-15)	8.1 (5-34)	0.330
Complications
SSI	1 (1.2)	0	0.253
Intraabdominal bleeding	0	1 (1)	0.379
Anastomotic bleeding	1 (1.2)	1 (1)	0.852
Anastomosis leakage	1 (1.2)	1 (1)	0.852
Ileus	0	1 (1)	0.379
Urinary retention	2 (2.5)	2 (1.9)	0.790
ClavienDindo classification
Grade I	3 (3.8)	2 (1.9)	0.450
Grade II	1 (1.2)	4 (3.8)	0.283
Grade IIIb	1 (1.2)	0	0.253
Post-operative mortality	0	0
Recurrence
Locoregional	2 (2.5)	3 (2.9)	0.082
Distant metastasis	3 (3.8)	0	0.874
Liver	2 (2.5)	0	0.046
Lung	1 (1.2)	0
Median follow-up period (months)	34.2 (3.0)	33.1 (2.2)	0.757
5 years disease-free survival, rate (%)	94.3	94.9	0.386
5 years overall survival, rate (%)	85.4	95.3	0.031
5 years cancer-specific survival, rate (%)	92	99	0.021

Grade II: Requiring pharmacological treatment, IIIb: Intervention under general anaesthesia, CEA: Carcinoembryonic antigen, SSI: Surgical site infection, SD: Standard deviation

The median follow-up period after surgery was 33.6 months, which did not differ between the groups. The rate of local recurrence did not differ between the groups (P = 0.874), but the rate of distant metastasis statistically differed (3 [2 hepatic and 1 pulmonary] cases in the clinical Stage I group vs. 0 cases in the pathological Stage I group; P = 0.046). No statistical difference was seen in the 5-year DFS between the clinical Stage I group and the pathological Stage I group (94.3 vs. 94.9; P = 0.386); however, statistical difference was seen in the 5-year OS (85.4 vs. 95.3; P = 0.031) and CCS (92 vs. 99, P = 0.021). Compared to pathological Stage II, some of the survival curves of clinical Stage I were located below the pathological Stage II, although there did not statistically differ (DFS: 94.3 vs. 95.4%; P = 0.140, OS: 85.4 vs. 87.2; P = 0.292, CSS: 92 vs. 96%; P = 0.163) [Table 3 and Figure 1].

Figure 1

Oncologic outcomes between clinical Stage I, pathological Stage I, and pathological Stage II: (a) recurrence-free survival, (b) overall survival and (c) cancer-specific survival

DISCUSSION

The study revealed significant differences in several parameters between clinical and pathological Stages I right-sided colon cancer. The clinical Stage I group had a larger tumour size, higher frequency of neural invasion (which denotes a poorer prognosis), and lower survival rate in some survival analyses than the pathological Stage I group. In particular, survival curves of clinical Stage I were similar to those of pathological Stage II.

The ESMO clinical practice guidelines recommend wide surgical resection and anastomosis for Stage I colon cancer.[3] The National Comprehensive Cancer Network does not provide clear recommendation for early-stage colon cancer.[4] The JSCCR guidelines 2016 recommend that at least D2 dissection for clinical T2 cancer and that D3 dissection may be needed in approximately 1% of case with a possibility of principal lymph node metastasis. However, this guideline provides insufficient evidence regarding the extent of lymph node dissection in cT2 cancers.[2] Although there are recent expert consensus recommendations, complete mesocolic excision in early colon cancer is controversial.[5]

In a study on the necessity of IMA ligation in T1 left-sided colon cancer including 121 patients, 12 had a nodal involvement, of whom 11 were diagnosed with lymph node metastases limited to the pericolic area. Therefore, IMA, the principal lymph node, can be preserved in T1 tumours.[6] However, in the case of right-sided colon cancer, lymphadenectomy is technically challenging and there is a variable distribution of blood vessels and lymph nodes. For all colorectal cancers, several studies on early stage have been published. In these studies, 13% of patients with T1 colorectal cancer had lymphatic metastasis, and lymphovascular invasion (P = 0.005), sm3 depth (P = 0.001) and lower rectum (P = 0.007) were identified as risk factors.[7] In a study on submucosal invasive colorectal cancer (SICC), lymph node metastasis was found in 14.3% of patients, and sm3, poorly differentiated cancer (P = 0.028) and tumour cell dissociation (P = 0.045) were identified as risk factors. According to the authors, local excision should be considered carefully.[8] In another pathological study on SICC, lymph node metastasis was found in 14.3% of patients, and lymphatic invasion (P < 0.01), tumour differentiation (P < 0.01), and tumour budding (P < 0.01) were significantly associated with lymph node metastasis.[9] Some of the authors included in our study also recommended that lymph node metastasis was higher in the case of lymphovascular invasion (P < 0.001) or perineural invasion (P = 0.004) in T1 and T2 colorectal cancer.[10] A systemic review and meta-analysis reported a 13% risk of lymph node metastasis in T1 cancers (95% confidence interval: 11.5%–15.4%) and the lowest risk of nodal metastasis in case of a depth of invasion ≤1,000 μm from the muscularis mucosa, good differentiation, absence of lymphatic and vascular invasion and the absence of tumour budding. However, it revealed the risk was not zero (1.9%), and the quality of evidence was poor.[11]

The reliability of CT in evaluating the staging preoperatively is questionable. In a study that reviewed 150 CT scans, CT failed to identify 18% of primary cancers, with an overall accuracy of only 63%. Overstaged and understaged clinical T stage were 23.7% and 48.3%, and clinical nodal stage were 28.7% and 53.0%, respectively.[12] Thus, MRI may be a more valuable tool than CT in the pre-operative staging of colon cancer.[13]

In a recent study by these authors, only 43.8% of clinical Stage I right-sided colon cancers were diagnosed as pathological Stage I. Risk factors for migration to overstaging include tumour size (P = 0.010) and the number of retrieved lymph nodes (P = 0.046).[14] In a study comparing complete (D3) and standard (D2) mesenteric excision with Stage I and II cancers, the D3 group had a higher 3-year OS (88.1% vs. 79.0%, P = 0.003) and DFS (82.1% vs. 74.3%, P = 0.026) than the D2 group.[15] In another study, the complete mesocolic excision (CME) group had better 4-year DFS rates (85.8% vs. 73.4%, P = 0.0014) and a higher UICC Stage I disease (100% vs. 89.8%, P = 0·046) than the non-CME group after propensity score matching.[16] This should be clarified and more accurate in upcoming studies.[17]

This study has some limitations. The radiological reports of pre-operative images may have different results depending on the institution. However, it is not thought to be beyond the range of standard radiologic reports because the institutions are responsible for more than 500 cases of colorectal cancer surgeries per year. The small sample size included in this study and retrospective design can also be limitations. However, the data were collected prospectively. Therefore, a large-sized multicentre trial is required to confirm the study results.

CONCLUSION

Even in the case of clinical Stage I, radical resection should be considered in accordance with T3 and T4 tumours.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.