Prognostic significance of clinical and pathological stages on locally advanced rectal carcinoma after neoadjuvant chemoradiotherapy

Objective To investigate prognostic significance of clinical and pathological stages in patients with locally advanced rectal carcinoma treated with neoadjuvant chemoradiotherapy (neo-CRT) and total mesorectal excision. Patients and methods 210 patients with locally advanced rectal carcinoma (cT3-4 or cN+) treated with neo-CRT followed by total mesorectal excision. Treatment outcomes were compared according to clinical and pathological stage. Overall survival (OS), disease free survival (DFS) among patients with different clinical stage and pathological stage after neo-CRT. Results The median follow-up time was 47 months (range, 14–98 months). Clinical T stage was associated with 5 year OS (p = 0.042) and 5 year DFS (p = 0.014) while clinical N stage was not associated with 5 year OS (p = 0.440), 5 year DFS (p = 0.711). Pathological T stage was associate with 5 year OS (p = 0.001) and 5 year DFS (p = 0.046); and N stage was associated with 5 year OS (p = 0.001), 5 year DFS (p = 0.002). The pathological stage was further classified into three groups: ypT0–2N0 in 91 patients (43.3 %), ypT3–4N0 in 69 patients (32.9 %) and ypT0–4N+ in 50 patients (23.8 %). While pathological stage (ypT0–2 vs ypT3–4N0 vs ypT0–4N+) was associated with 5 year OS (87.9 %, 75.5 %, 56.7 %, p = 0.000), 5 year DFS (74.5 %, 77.4 %, 50.5 %, p = 0.003). Multivariate analysis showed that ypN stage was an independent prognostic factor for patients 5 year DFS. Conclusions Pathological stage is strongly associated with treatment outcomes in patients with locally advanced rectal carcinoma treated with neo-CRT followed by total mesorectal excision, which may be used as guidance for further individualized treatment.

therapy. It has been reported [9][10][11] that posttreatment pathological TNM stage is more closely related to the prognosis than pretreatment clinical stage in these patients. Patients with locally advanced rectal cancer may demonstrate a different treatment response to Neo-CRT ranging from pathological complete response (pCR) to resistance to Neo-CRT. Studies have reported that pathological TNM stage is of prognostic significance in treatment outcomes after Neo-CRT when compared to clinical TNM stage. Bujko et al. [12] reported that prognostic effect of adjuvant chemotherapy in patients with ypT0-4N0 disease after Neo-CRT and TME is not convincing, suggesting that adjuvant chemotherapy should be given individually according to pathological TNM stage.
The purpose of the study was to investigate treatment response in patients with locally advanced rectal cancer treated with Neo-CRT and TME and to compare prognostic effects of clinical and pathological TNM stages.

Materials and methods
From March 2003 to December 2010, 210 patients with locally advanced rectal cancer (cT3-4 or cN+) were treated with Neo-CRT followed by TME with or without adjuvant chemotherapy at the Sun Yat-Sen University Cancer Center. Diagnosis was pathologically confirmed prior to Neo-CRT. Pretreatment clinical stage was determined by physical examination, ultrasound colonoscopy/ colonoscopy, computerized tomography and/or magnetic resonance imaging.
In the present study, preoperative radiotherapy (RT) was performed with three-dimensional conformal radiation therapy (3D-CRT) technique with 6-8 MV X-ray and 3-4 fields at the full pelvis. The target volume definition was followed by the International Commission of Radiation Units 50 report recommendations. The delineation of clinical target volume (CTV) included primary rectal carcinoma, both ends of the affected rectum, the surrounding tissues of the affected rectum, the mesorectal area, the presacral lymph nodes, the obturator lymph nodes, and the iliac lymph nodes. For patients with stage T4 rectal carcinoma with bladder involvement, the delineation of CTV also included external iliac lymph nodes. The planned target volume (PTV) was designated as 8-10 mm margin from the CTV. For the dose prescription was as follows: 100 % of the prescription dose covered at least 95 % of the PTV; 95 % of the prescription dose covered at least 100 % of the PTV. The reference point was set as the intersection of the central axes of the three or four beams. The median radiation dose to PTV was 46.0 Gy (ranging from 30.0 to 50.0 Gy), 2 Gy per fraction, 5 fractions per week. The dose to the OARs was aimed to be as low as possible and must at least comply with the following constraints: bladder >50Gy in <50 % volume; Dmean of small bowel < 46Gy, small bowel > 50Gy in <5 % volume. Five patients who were operated on R1 or R2 resection were treated with postoperative radiotherapy with median dose of 36.0 Gy (range from 30.0 Gy to 40.0 Gy) delivered by 3D-CRT technique.
Two regimens of chemotherapy were administered during radiotherapy either the FOLFOX regimen (Fluorouracil, 3.0 g/m 2 , IV continuous infusion for 48 h on day 1; Leucovorin calcium, 200.0 mg/m 2 , IV bolus on day 1; Oxaliplatin, 100.0 mg/m 2 , IV on day 1; two cycles at an interval of 3 weeks), or the XELOX regiment (Capecitabine, 1000.0 mg/m 2 , on d1-14; Oxaliplatin, 100.0 mg/m 2 , IV on day 1; two cycles at an interval of 3 weeks). The postoperative adjuvant chemotherapy was either one of FOLFOX, XELOX or Capecitabine alone with median cycles of 2 (range from 2 to 6 cycles).
Radical surgery was performed according to the principles of total mesorectal excision. Postoperative pathological examination was performed according to the criteria developed by the AJCC/UICC (2002).
Treatment response including tumor regression response to Neo-CRT was investigated according to pathological TNM staging classification after radical surgery. Treatment outcome was analyzed according to clinical and pathological TNM stages in terms of overall survival (OS), disease free survival (DFS), local recurrence free survival (LRFS) and distant metastasis free survival (DMFS) among patients with different clinical stage (IIA, IIB, IIIA, IIIB and IIIC) and pathological stage (ypT0-2N0, ypT3-4N0 and ypT0-4N+).
All statistical analysis was performed using SPSS v17.0 software. p value of <0.05 was considered to be of statistical significance. t-test or the Wilcoxon rank sum test was used for continuous variables and the chi-square test for categorical variables. Kaplan-Meier analysis was performed for clinical outcome, and the log-rank test was used for comparison between clinical and pathological outcomes curves. Univariate and multivariate analysis of the prognostic factors were performed using Cox proportional hazard models.

Patient characteristics
The patient characteristics are shown in Table 1. There were totally 210 patients including 149 males (71 %) and 61 females (29 %) with a median age of 56 years (range, 15-80 years). Patients with clinical stage IIA, IIB, IIIA, IIIB and IIIC were 31, 38, 4, 54 and 83, respectively. The median distance of carcinoma to the anus was 5 cm (1-16 cm). The median CEA level was 4.5 ng/mL (ranging from 0.2 ng/ml to 249.6 ng/mL). The tumor location of low-lying (distance between anal verge and lower edge of tumor ≤ 5 cm), middle-third (5 cm < distance between anal verge and lower edge of tumor ≤ 10 cm) and upper-third (distance btw anal verge and lower edge of tumor > 10 cm) patients were 130, 78 and 2, respectively.
Five patients who were operated on R1 or R2 resection due to bladder/prostate involvement received postoperative radiotherapy with median dose of 36 Gy (range from 30 to 40 Gy) delivered by 3D-CRT technique. Among patients with ypT0-4N0 (160 patients), 115 received postoperative chemotherapy.
The complication after radical surgery included postponed incisions heal in 7 patients (3.3 %), wound infection in 2 patients (1.0 %), anastomotic leak in 2 patients (1.0 %) and pelvic sepsis in 1 patient. All of them were healed with best supportive care. The artificial anus was unobstructed.
Eleven patients developed single locoregional tumor recurrence that occurred at the first 3 years; 2 developed locoregional tumor recurrence and distant metastasis in the lung or bone; 29 patients developed distant metastasis with 14 in the lung, 11 in the liver, 2 in the bone, other 2 in the the abdominal or supraclavicular lymphadenopathy. Among them, 5 showed recurrence at the primary site, 4 at the iliac, 2 at the perirectal region, one at the primary site with lung metastasis and the other one at the iliac with lung and bone metastases. Twenty-nine patients showed distant metastasis alone after neo-CRT and TME. Among them, 14 patients showed distant metastasis at the lung, 11 at the liver, 2 at the bone, and 2 at lymph nodes at other site. Five patients who were treated with Neo-CRT of 46 Gy and operated on R1 or R2 resection due to bladder/prostate involvement received postoperative radiotherapy with median dose of 36 Gy. Two of them died of the disease and the other 3 remains disease free survival.
At the end of the study, 43 patients had died (20.4 %) of recurrent or metastatic disease in the lung, liver and locoregional recurrence.

Univariate analysis and multivariate analysis of prognostic factors for patients with locally advanced rectal carcinoma
Univariate analysis (Table 2) showed that pT, pN and pathological stages were associated with OS, DFS, LRFS and DMFS after Neo-CRT for patients with locally advanced rectal carcinoma. Clinical T stage was associated with DFS and distance between anal verge and lower edge of tumor was associated with LRFS. Multivariate analysis (Table 3) showed that only ypN stage was an independent prognostic factor for OS, DFS, LRFS and DMFS; ypT was associated with OS and DMFS. Patients with pN+ disease showed a high recurrence rate than those with pN0 disease (HR: 2.239;95 % CI:1.286-3.898).   Pathological stage was independent prognostic factors for OS and DMFS.

Discussion
Our study has demonstrated that neoadjuvant chemoradiotherapy is associated with tumour downstage with complete pathologic response rate of 24.8 %, elevated radical resection rate, sphincter preservation and no increase in surgical complication after total mesorectal excision in patients with locally advanced rectal carcinoma. Pathological TNM stage is strongly associated with treatment outcome in these patients after neo-CRT and may be of better prognostic significance than clinical TNM in terms of OS, DFS. Distant metastasis remains the main obstacle of the successful treatment for patients with locally advanced rectal carcinoma.
In recent years, studies have been conducted to investigate the influence of the postoperative pathological stage on the prognosis of locally advanced rectal carcinoma. Quah et al. [11] found that the postoperative pathological stage was related to the prognosis of 331 patients with locally advanced rectal carcinoma receiving neoadjuvant chemotherapy. In this study, pathological stage (ypT1-2N0, ypT3-4N0, and ypN+) was significantly correlated with DFS (P = 0.003) and OS (P = 0.000), and patients with stage ypT0N0 or ypT1-2N0 rectal carcinoma had a better prognosis. Park et al. [13] reported that in patients with stage ypT0N0, ypT1-2N0, and ypT3-4/N+ rectal carcinoma, the 5 year OS was 93.4 %, 87.0 %, and 77.3 % (P = 0.002), respectively. The 5 year DFS was 90.5 %, 78.7 %, and 58.5 %, respectively (p < 0.001) in 725 patients with locally advanced rectal carcinoma, and OS and DFS were significantly different among these patients.
Thus, these investigators proposed that the pathological stage after neoadjuvant chemotherapy was an important predictor, and recommended postoperative adjunctive therapeutics according to the response of rectal carcinoma patients to chemoradiotherapy, which was also known as individualized therapy. For example, a wait-and-see strategy can be applied in patients with stage ypT0N0 rectal carcinoma; intensified adjunctive chemotherapy is employed in patients with stage ypT3-4/N+ colorectal carcinoma. Kim et al. [14] reported that the 5 year DFS was 65.2 % and 35.7 % in patients with stage N0 and N+ rectal carcinoma, respectively, among 114 patients with locally advanced rectal carcinoma (p = 0.002), suggesting that pathological N stage is an important predictor [15].
Our results in univariate and multivariate analyses of prognostic factors also suggest that the pathological stage after neoadjuvant chemoradiotherapy significantly correlates with the prognosis of rectal carcinoma. In patients with ypT0-2N0, ypT3-4N0, and ypT0-4N+ rectal carcinoma, the 5 year OS was 87.9 %, 75.5 % and 56.7 % (p = 0.000), respectively; and the 5 year DFS was 74.5 %, 77.4 % and 50.5 % (p = 0.003). Stages ypT and ypN are important predictors for disease-free survival. Clinical TNM stage was not related to overall survival and distant metastatic free survival although cT stage was associated only with disease free survival in univariate analysis. Thus, our findings together with previously reported findings indicate that the postoperative pathological stages reflect the prognosis of patients with locally advanced rectal carcinoma better than does clinical stage before treatment, and it is more rational to choose and administer individualized adjuvant therapy according to the postoperative pathological stages [15].
Since pelvic lymph node status after neoadjuvant chemoradiotherapy is an important prognostic factor in patients with locally advanced rectal carcinoma, Jwa et al. [16] developed nomogram to improve prediction of lymph node status after neo-CRT by analyzing pretreatment ypT stage, patient age, tumor differentiation and clinical N stage before Neo-CRT, lymphovascular invasion and perineural invasion.
Stratified adjuvant therapy, performed according to pathological stage, has been reported in patients with locally advanced rectal carcinoma after neoadjuvant chemotherapy. Focal dissection can be performed in patients with locally advanced rectal carcinoma achieving pathological complete response (pCR) after neoadjuvant chemotherapy, and the therapeutic efficacy is similar to that after radical surgery [17,18]. Moreover, this also avoids the complications and sequelae of radical surgery, and significantly improves quality of life. Govindarajan et al. [19] found that postoperative adjunctive chemotherapy had no influence on the 5 year DFS in patients with locally advanced colorectal carcinoma at pathological stage ypT0-2N0 after preoperative chemoradiotherapy. Thus, the value of postoperative adjunctive chemotherapy should be further investigated in future randomizedcontrolled trials.
In the present study, we found that the major cause of treatment failure in patients with locally advanced rectal carcinoma was distant metastasis. Patients with stage ypT3-4N0 and ypT0-4N+ rectal carcinoma had the highest rate of distant metastasis (35.7 %), which was markedly greater than that of patients with stage ypT0-2N0 rectal carcinoma (P < 0.001). Thus, an effective way to improve therapeutic efficacy is to reduce distant metastasis. For patients with rectal carcinoma at stage ypT3-4N0 or stage ypT0-4N+, postoperative adjunctive therapy should be changed to prolong the duration of treatment or increase the dose of chemotherapeutics, or to apply targeted therapy, which may reduce the rate of distant metastasis and improve the survival rate.
Similar to previous studies [11], we found that clinical stage before treatment failed to reflect the prognosis of patients with locally advanced rectal carcinoma. These were several underlying reasons. First, the pretreatment evaluation of clinical stage is different in different studies. The accuracy and specificity of the depth of colorectal carcinoma invasion (T stages) for endoscopic ultrasonography (EUS) are 94 % and 86 %, respectively [20,21]; and for MRI 94 % and 69 %, respectively [22][23][24][25]. The sensitivity and specificity of determining the relationship between carcinoma and surrounding tissues for EUS are 90 % and 75 %, respectively; for MRI 82 % and 76 %, respectively; and for CT 55 % and 74 %, respectively [20,26]. Thus, T stage and N stage may not accurately reflect the disease condition. Second, after combined use of chemotherapy and radiotherapy, additional factors may influence the response to treatment, such as carcinoma size, dose of radiation and chemotherapeutic regimen [27]. In addition, some molecular biological markers (such as epidermal growth factor receptor, thymidylate synthase, p21, and CEA) may also influence the response of colorectal carcinoma to chemoradiotherapy [28,29].
Neoadjuvant chemoradiotherapy may reduce the pathological stage of rectal carcinoma to a different extent in different patients. In the present study, postoperative pathology confirmed that the pathological stage of rectal carcinoma was decreased in 132 patients (62.9 %) after neoadjuvant chemoradiotherapy, of whom patients with rectal carcinoma at stage ypT0-2N0 accounted for 43.3 %. This finding is consistent with previous reports [13,[30][31][32]. Patients with rectal carcinoma at stage ypT0-2N0 had a better prognosis.
In conclusion, neo-CRT and total mesorectal excision is an effective treatment modality for patients with locally advanced rectal carcinoma. The pathological stage is an important prognostic factor, which may be used as guidance for further individualized adjuvant therapy. Distant metastasis remains obstacle for the successful treatment of the disease. Further investigation therapeutic options are still needed to improve clinical outcomes of locally advanced rectal carcinoma.