Analysis of clinical and pathological factors in patients undergoing surgery for rectal cancer: a single-centre experience with 692 patients in China

Background: The management of rectal carcinoma has substantially over the so as warrant further study. The present study aimed to assess the associations of clinical and pathological factors in rectal cancer patients undergoing radical surgery. Methods : From October 2015 to February 2019, all rectal cancer patients treated with radical surgery without neoadjuvant therapy were identified. The analysis was performed with data obtained from the prospectively collected database. Predictive factors for lymph node metastasis were analysed. Results: In total, 692 patients with a median age of 61.64 years (range: 22-89) were included. There was no significant difference in onset age between male and female patients (61.75±11.10 vs 61.43±11.92, P=0.723). Tumour location (P=0.004), perineural invasion (PNI) (P=0.000), lymphovascular invasion (LVI) (P=0.000), tumour deposit (TD) (P=0.000), and differentiation grade (P=0.000) were significantly related to pathologic T stage in univariate analysis, while sex was not (p=0.192). explore the correlations of clinical and pathological factors in patients undergoing surgery for rectal cancer.

Conclusion: There was no significant difference between male and female patients in onset age.
Tumour location, PNI, LVI, TD, and differentiation grade were significantly related to pathologic T stage. Patients with the presence of LVI and TD, higher T stage, and poorly differentiated histology have a significantly higher chance of LNM.
Surgical resection remains the most effective therapy for rectal cancers. Pathologic findings in surgical resection specimens are the best predictor of prognosis. The main postoperative pathological information includes tumour differentiation grade, macroscopic type, lymphovascular invasion (LVI), perineural invasion (PNI), tumour deposit (TD), primary tumour stage, and lymph node stage. AJCC staging and NCCN guidelines have been updated in recent decades. For example, the classification of tumour nodules as TDs versus LNs has been debated in the past. Since the AJCC 7th (2010) edition, the new nodal subclassification category N1c is used if there are TDs but no concurrent positive LNs [8]. The percentage of LN metastasis has also changed in the new classifications. Therefore, the inherent relationships of these pathologic factors warrant further study. The present study aimed to explore the correlations of clinical and pathological factors in patients undergoing surgery for rectal cancer.

Materials And Methods Patients
We enrolled 1061 patients with primary rectal cancer who had undergone radical resection at the First Affiliated Hospital of Nanjing Medical University between October 2015 and February 2019. The inclusion criteria for the study were as follows: (1)

Surgery and histopathological examinations
The 692 patients underwent curative TME. Samples were obtained from opposite side of the resected specimens and fixed in formalin for 24 h after surgery. The specimens were then sliced transversely at increments of 5 mm. The slices were embedded in paraffin, sectioned, and examined histologically after haematoxylin and eosin (HE) staining. The results were evaluated by experienced pathologists.
The histological grade, presence of LN metastasis, LNI and PNI were all evaluated.

Statistical analysis
The study was designed to identify clinicopathologic variables that could be associated with the pathological T and N stages. Categorical variables such as patient demographics (such as sex) and tumour characteristics (such as tumour location, differentiation grade, PNI, LVI, TD, and pathologic T and N staging) were compared with chi-square tests, while for continuous variables such as patient age, Student's t tests and analysis of variance (ANOVA) were used. Multiple covariate analysis was performed using the stepwise regression hazards regression model. The hazard ratio (HR) and P value with the 95% confidence interval (CI) were calculated for the variable groups. All P values were 2-tailed, and P values less than 0.05 were considered statistically significant. Statistical analyses were performed with SPSS version 19.0 (SPSS, Chicago, IL).

Patient Demographics And Clinicopathologic Data
The patient characteristics are shown in Table 1. There were 245 females (35.4%) and 447 males (64.6%), with a mean age of 61.64 ± 11.39 (range: 22-89) years. The average age at the onset of rectal cancer was 61.43 ± 11.92 years for female patients and 61.75 ± 11.10 years for male patients.
Clinicopathologic variables that correlate with the T stage We analysed the clinicopathological factors that could be correlated with the pathological T stage, as shown in Table 3. On univariate analysis, tumour location (P = 0.004), PNI (P = 0.000), LVI (P = 0.000), TD (P = 0.000), and differentiation grade (P = 0.000) were significantly correlated with the pathological T stage, but sex was not (p = 0.192). Regarding tumour location, there were no differences between patients with T4 disease and those with T1 (P = 0.726), T2 (P = 0.486) and T3 (P = 0.836) disease.

Clinicopathologic variables that may be correlated with LNM
We also analysed the clinicopathologic factors that could be correlated with pathologic lymph node metastasis (LNM), as shown in  As shown in Table 2, Fig. 2 and Fig. 3, we further confirmed the correlation between the primary tumour (T) and nodule status (N) by a binary logistic regression model. Compared with T1, the HRs of LNM for T2, T3, and T4 patients were 1.309 (95% CI: 0.559-3.065, P = 0.535), 3.126 (95% CI: 1.380-7.082, P = 0.006) and 4.730 (95% CI: 1.499-14.927, P = 0.008), respectively. More advanced T stages (T3 and T4) were significantly associated with a higher incidence of LNM.

Discussion
The results of the present study demonstrated the relationships among clinical and pathological factors in patients undergoing surgery for rectal cancer. These clinicopathological factors included sex, age, tumour location, differentiation grade, LVI, PNI, TD, and pathological T and N stages.
Our study demonstrated that the number of male rectal cancer patients was approximately 1.8 times that of female patients. However, no significant difference in age at onset was observed between female and male rectal cancer patients.
LVI is defined as the involvement of tumours in vascular and lymphatic structures [9][10]. In our study, 20.2% of 692 patients had LVI, which was in agreement with previous studies that reported that the incidence of LVI in rectal cancer patients who underwent curative surgery was 20-30% based on pathological evaluation [11][12][13][14].
The regional LN is the most common site of tumour metastasis. To accurately evaluate LNM, as many LNs as possible should be assessed to determine the N stage. Both the total number of regional LNs removed and the number of positive LNs involved should be reported. The AJCC 7th and 8th editions state that it is important to obtain and examine at least 12 LNs [8]. The prior 6th edition suggested obtaining a range from 7 to 14 LNs. In this study, an average of 18.56 ± 5.68 (from 10-49) LNs were obtained and examined.
We determined that LVI, pathological T stage, differentiation grade, and TD were significantly related to N based on univariate and multivariate analyses. The presence of LVI and TD, poor differentiation and advanced T stage (T3 and T4) were significantly associated with LNM. PNI was also an independent risk factor for N in univariate analysis but not multivariate analysis. The HRs were calculated for every risk factor in our study and are shown in  [22]. Another study reported that the overall incidence of LNM was 12.7% for colorectal cancer (5.6% for T1 and 14.5% for T2; p = 0.021) [23]. In Patel's study [24] Fig. 2 and Fig. 3). This conclusion was confirmed by univariate and multivariate analyses (Table 4). Compared with T1, the hazard ratios of LNM for T3 and T4 were 3.126 (95% CI: 1.380-7.082, P = 0.006) and 4.730 (95% CI: 1.499-14.927, P = 0.008), respectively, but a significantly increased risk was not observed for T2 patients (HR = 1.309, 95% CI: 0.559-3.065, P = 0.535).
Therefore, increased depths of tumour penetration are associated with a greater incidence of LNM.

Conclusion
The present study indicated that the presence of LVI, PNI, TD, low pathological differentiation grade, and advanced pathological T stage (T3 and T4) were independent factors associated with a higher incidence of LNM on univariate and multivariate analyses.

Consent for publication
All authors of this paper have read and and complied with author guidelines. The contents of this manuscript have not been copyrighted or published previously. The contents of this manuscript are not under consideration for publication elsewhere. The contents of this manuscript will not be copyrighted, submitted, or published elsewhere while acceptance by the manuscript is under consideration. All authors approve to publish this article in Journal of Experimental & Clinical Cancer Research.

Availability of supporting data
Not applicable

Competing interests
The authors have declared that no competing interests exist. The mean number of positive lymph nodes for each T stage.