Prognostic factors in patients with high-risk stage II colon cancer after curative resection: a post hoc analysis of the JFMC46-1201 trial

Purpose The goal of the current study was to identify prognostic factors for disease-free survival (DFS) and overall survival (OS) in high-risk stage II colon cancer. Methods The subjects were patients with histologically confirmed stage II colon cancer undergoing R0 resection who met at least one of the following criteria: T4, perforation/penetration, poorly differentiated adenocarcinoma, mucinous carcinoma, and < 12 examined lymph nodes. Patients self-selected surgery alone or a 6-month oral uracil and tegafur plus leucovorin (UFT/LV) regimen. Serum CEA mRNA at ≥ 24 h after surgery and < 2 weeks after registration was also examined as a potential prognostic factor for stage II colon cancer. This study is registered with UMIN-CTR (protocol ID: UMIN000007783). Results 1880 were included in the analysis to identify prognostic factors for DFS and OS in patients with high-risk stage II colon cancer. In multivariate analyses, gender, depth of tumor invasion, extent of lymph node dissection, number of examined lymph nodes, and postoperative adjuvant chemotherapy (POAC) emerged as significant independent prognostic factors for DFS. Similarly, multivariate analysis showed that age, gender, depth of tumor invasion, perforation/penetration, extent of lymph node dissection, number of examined lymph nodes, and POAC were significant independent prognostic factors for OS. Univariate analyses showed no significant difference in DFS or OS for CEA mRNA-positive and mRNA-negative cases. Conclusion This study showed that gender, depth of tumor invasion, extent of lymph node dissection, number of examined lymph nodes, and lack of use of POAC were significant independent prognostic factors in stage II colon cancer.


Introduction
Colorectal cancer (CRC) is the third most common malignancy and the second most frequent cause of cancer-related mortality worldwide [1].In the absence of conclusive randomized controlled trial data, clinical guidelines published by the American Society of Clinical Oncology (ASCO), the National Comprehensive Cancer Network (NCCN), and the European Society for Medical Oncology (ESMO) have suggested a benefit of 260 Page 2 of 9 postoperative adjuvant chemotherapy (POAC) in stage II colon cancer with risk factors for recurrence [2][3][4][5].Clinical guidelines from the Japanese Society for Cancer of the Colon and Rectum also recommend POAC for stage II CRC with risk factors for recurrence, but not for cases without these risk factors [6].However, such risk factors have yet to be clearly established [7].
We previously performed a prospective observational study (JFMC46-1201) to judge the efficacy of POAC with oral uracil and tegafur plus leucovorin (UFT/LV) for stage II colon cancer with potential risk factors for recurrence [8].Here, we describe a post hoc analysis of this study, with the goal of identifying prognostic factors for disease-free survival (DFS) and overall survival (OS).In addition to prevalent clinicopathological factors, serum positivity for carcinoembryonic antigen (CEA) mRNA has been proposed as a surrogate marker for circulating tumor cells (CTCs) [9][10][11][12], and the presence of CTCs in peripheral blood at ≥ 24 h after surgery has been reported to be an independent risk factor [13].Therefore, in the current study, serum CEA mRNA at ≥ 24 h after surgery and < 2 weeks after registration was also examined as a potential prognostic factor for stage II colon cancer.

Patients
Details of the study protocol have been reported elsewhere [7].The subjects were patients with histologically confirmed stage II colon cancer undergoing R0 resection who met at least one of the following criteria: T4, perforation/ penetration, poorly differentiated adenocarcinoma, mucinous carcinoma, and < 12 examined lymph nodes.The patients were aged 20-80 years and had ECOG PS of 0 or 1. Patients were registered after the confirmation of pathological findings.Patients self-selected surgery alone (Group A) or a 6-month oral UFT/LV regimen (Group B).The study was performed in accordance with the ethical standards of the 1975 Declaration of Helsinki and its later amendments.This study is registered with UMIN-CTR (protocol ID: UMIN000007783).

Clinicopathological analysis
The following clinicopathological factors were analyzed: age, gender, location (right-side/left-side), depth of tumor invasion (TNM classification [14]), presence of perforation/ penetration, histology of the primary tumor [15], extent of lymph node dissection [15], number of examined lymph nodes, number of patients at each center, and use of POAC.
D3 lymph node dissection means the complete removal of the pericolic, intermediate, and main lymph nodes with a high vascular tie at the root of the feeding artery.D2 lymph node dissection means the complete removal of the pericolic and intermediate lymph nodes without a high vascular tie.D1 lymph node dissection means the complete removal of the pericolic lymph nodes.D0 lymph node dissection means the incomplete removal of the pericolic lymph nodes.Since the JFMC46-1201 prospective observational study was a multicenter trial, the evaluation of lymphatic or vascular or perineural invasion may have varied [16], and thus, these factors were excluded from the study.Microsatellite instability (MSI) was also excluded because evaluation of MSI status was not covered by Japanese health insurance during the study period.

Carcinoembryonic antigen (CEA) mRNA in peripheral blood
Peripheral blood samples were collected at ≥ 24 h after surgery and < 2 weeks after registration.To prevent contamination with epithelial cells, the samples were obtained through a catheter inserted into the peripheral vessel and the first 10 ml of blood was discarded.As described elsewhere [9], total RNA was isolated from 7 ml of peripheral blood using an RNeasy Mini Kit (Qiagen, Germantown, MD, USA) and redissolved in diethyl pyrocarbonatetreated water. 1 μg of isolated total RNA was reversetranscribed into complementary DNA (cDNA) using Random primer (Sigma-Aldrich, Darmstadt, Germany) and an M-MLV Reverse Transcriptase (Thermo Fisher Scientific, Waltham, MA, USA).Each quantitative RT-PCR (qRT-PCR) contained 2.5 µl of cDNA product, 2.5 µl of PCR buffer, 2.5 µl of deoxynucleotide triphosphate (dNTP), 3.5 µl of 25 mM MgCl 2 , 1.25 µl of forward and reverse primers, 1.25 µl of probe and 0.125 µl of Taq Gold probe (Thermo Fisher Scientific, Waltham, MA, USA), made up to a total volume of 25 µl with sterile water.qRT-PCR was performed using a 10 µM forward primer, 10 µM reverse primer and 2 µM probes, with cycling conditions of 95 °C for 10 min, followed by 53 cycles of 95 °C for 25 s, 65 °C for 1 min, and 72 °C for 30 s.The forward and reverse primers and probe (Human qRT-PCR step for CEA No. 4391675-T, Nihon Gene Research Laboratory Inc., Miyagi, Japan) are synthesized to yield a band of 135 bp of CEACAM5 (Nt.1811-1945).The CEA mRNA was analyzed by performance of the GAPDH competitive PCR as an internal control.Primers for GAPDH were as follows: forward: 5′-TGA AGG TCG GAG TCA ACG G-3′ and reverse: 5′-AGA GTT AAA AGC AGC CCT GGT G-3′; probe for GAPDH is as follows: 5′-TTT GGT CGT ATT GGG CGC CTGG-3′.The CEA mRNA/GAPDH mRNA ratio was used to determine the relative CEA mRNA expression level.In the current study, more than 50 copies/μg of the CEA mRNA/GAPDH mRNA ratio was determined as positive.

Treatment and surveillance
In Group A, postoperative follow-up procedures included clinical assessment and serum CEA measurements every 3 months and chest and abdominal CT every 6 months until 5 years postoperatively or until recurrence, another malignancy or death.Colonoscopy was performed at one and three years after the operation.In Group B, UFT/ LV was started within 8 weeks after surgery, with five courses given as UFT (300 mg/m 2 /day) plus LV (75 mg/ day) administered orally in 3 doses per day at approximately 8-h intervals.Treatment was administered daily for 28 days followed by a 7-day rest period (daily treatment regimen) or daily for 5 days followed by a 2-day rest period (5-day treatment plus 2-day rest regimen; no treatment on Saturdays and Sundays).For both regimens, one course lasted 5 weeks, and 5 courses were given.After completion of UFT/LV therapy, patients were followed up using the same schedule as that for Group A.

Statistical analysis
In the current study, DFS is defined as survival from registration for colon cancer until first recurrence of the disease or secondary cancer or mortality from any cause.OS is defined as survival from registration for colon cancer until mortality from any cause.Multivariate analysis was performed using Cox proportional hazard regression by backward elimination method from prognostic factors (age, gender, location, depth of tumor invasion, perforation/penetration, poorly differentiated component, mucinous component, number of examined lymph node, number of participating patients at each institution, and POAC) with an exclusion criterion of p = 0.05.All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), with differences considered significant at p < 0.05.

Prognostic factors for DFS and OS
In multivariate analyses, gender (male) (HR:

Discussion
Among patients diagnosed with adenocarcinoma of the colon, 25-40% have stage II disease, and these cases have a relatively good prognosis, with 5-year OS rates of 72-85% [17] compared with stage III cases [18,19].Therefore, there appears only to be a small benefit of POAC, but this may be more important in cases with high-risk features [20].This indicates the importance of identification of patients who may benefit from POAC [21].Many studies have examined methods for better identification and treatment of stage II cases with a risk of recurrence, but it has been difficult to find reliable prognostic factors [20].This makes risk stratification difficult, which causes physicians to individualize treatment and produces widely varied clinical practice [22].For these reasons, we examined prognostic factors in stage II colon cancer in the cohort from our previous prospective clinical study.The current study showed that gender (male), depth of tumor invasion (T4), extent of lymph node dissection (D0-2), number of examined lymph nodes (< 12), and POAC were significant independent prognostic factors for DFS.Clinical guidelines published by ASCO [2] and ESMO [4] also mention the depth of tumor invasion (T4) and number of examined lymph nodes (< 12) as risk factors for recurrence.A poorly differentiated component and a mucinous component were not risk factors for DFS and OS in the current study.This may be explained by the cases with these  components having significantly lower rates of T4 invasion depth and examined lymph nodes < 12 (data not shown).In a large cohort in the National Cancer Database, male sex was found to be an independent risk factor for recurrence in stage II CRC cases that did not receive adjuvant systemic therapy [23].Cases with and without POAC were included in the current study, but male sex was still identified as a risk factor for DFS and OS in stage II colon cancer.In the current study, multivariate analyses revealed that the extent of lymph node dissection was a risk factor for the DFS and OS in stage II colon cancer, as was the number of examined lymph nodes.D3 lymph node dissection is recommended for T3 or T4 primary tumor in the Japanese guidelines [6], in contrast to the clinical guidelines in western countries [2][3][4][5].In the current study, D0-2 lymph node dissection was performed in only 15% of patients.Although this was presumably due to perforation/penetration, D3 lymph node dissection seems to be ideal for improving the prognosis of patients with stage II colon cancer.POAC with UFT/LV was found to be useful to improve DFS and OS, and several studies in a large cohort in the National Cancer Database have also shown that POAC is associated with a survival benefit for patients with high-risk features [24,25].In these studies, the kinds of POAC may have varied.Therefore, the current study is significant in terms of the consistent regimen of oral UFT/LV for stage II colon cancer with high-risk features.In addition, since the incidence of adverse events was low, oral UFT/LV appears to be safe and well-tolerated as POAC [8].CEA mRNA in peripheral blood ≥ 24 h after curative resection of CRC has been found to be significantly correlated with poor long-term survival [9][10][11][12][13].However, there was no correlation between a CEA mRNA-positive status and DFS or OS in the current study.The CEA mRNA-positive rate was 23.4%, which was similar to that of 24.1% in our previous study of the correlation between recurrence and CEA mRNA status in stage II, in which we also found no significant correlation with stage II cases in a small cohort (n = 87) [9].Thus, the current study in a large multicenter cohort confirmed the lack of significance of CEA mRNA as a prognostic factor in stage II colon cancer.Therefore, measurement of CEA mRNA may not be beneficial for prediction of recurrence in stage II colon cancer.Circulating tumor DNA (ctDNA) has recently been reported to be a powerful tool to help guide treatment decisions in colorectal cancer.In patients with high-risk stage II or stage III disease, Kotani et al. [26] found that those with a ctDNApositive status 4 weeks after surgery derived significant benefit from postoperative adjuvant chemotherapy, while those with a ctDNA-negative status 4 weeks after surgery did not.It was suggested that ctDNA positivity after curative surgery or therapy, which indicated minimal residual disease (MRD), was strongly associated with a poor prognosis in patients with surgically resectable CRC.Therefore, postsurgical ctDNA status is likely to be a future predictor of a benefit of postoperative adjuvant chemotherapy, instead of CEA mRNA.Finally, there are several limitations in the current study that are inherent to the non-randomized design.First, the decision regarding the use of POAC with UFT/LV was made for each case based on individual patient/physician discussions.Second, some risk factors, such as MSI, were not included, but MSI status influences the benefit of POAC with fluoropyrimidine alone in colon cancer [27].We also excluded lymphatic, vascular, or perineural invasion as a risk factor because its evaluation was likely to be inconsistent among the participating institutions.In fact, discrepancies in the observed rates in lymphovascular invasion in colorectal cancer have been widely reported, with Davenport et al. [28] suggesting that interobserver agreement on lymphovascular invasion is poorer than that for other clinicopathological factors.

Conclusions
This study showed that gender (male), depth of tumor invasion (T4), extent of lymph node dissection (D0-2), number of examined lymph nodes (< 12), and lack of use of POAC were significant independent prognostic factors in stage II colon cancer.

Table 2
Details of the events of recurrence/secondary cancer/mortality for disease-free survival (DFS)

Table 3
Details of the events of mortality for overall survival (OS)

Table 4
Prognostic factors for disease-free survival (DFS) and overall survival (OS) a Prognostic factors were selected by multivariate analysis with an exclusion criterion of p = 0.05 b D3: complete removal of pericolic, intermediate, and main lymph nodes with a high vascular tie at the root of the feeding artery.D2: complete removal of pericolic and intermediate lymph nodes without a high vascular tie.D1: complete removal of pericolic lymph nodes.D0: incomplete removal of pericolic lymph nodes