Background

Nodal status has long been considered a key pathologic factor in cancer management and care. Previous studies have demonstrated an association between lymph node yield and status with regard to patient outcomes in colorectal cancer (CRC).^1–4 As such, it was first recommended at the World Congress of Gastroenterology in 1990 that at least 12 lymph nodes be examined after radical resection to optimally stage patients, and the impact of preoperative therapy on nodal yield was acknowledged.⁵ Subsequently, several major organizations, including the American Joint Committee on Cancer (AJCC), Union for International Cancer Control, and College of American Pathologists (CAP) have recommended at least 12 lymph nodes be examined for accurate staging in CRC becoming a well-accepted metric.

Since that time, several large population-based studies have demonstrating inadequate lymph node harvest and evaluation being undertaken in CRC.^6,7 This has been more prevalent with rectal cancer, and a number of studies have determined limitations to this, including variability related to surgical techniques, pathologic assessment, patient factors, institutions, and the effect of neoadjuvant therapy.^8–12

Regarding surgical technique, it is well accepted that a total mesorectal excision (TME) should be undertaken to help achieve optimal oncologic outcomes. Proficient TME allows for harvest of the traditional primary nodal drainage basin and optimizes the circumferential resection margin. If an incomplete TME is undertaken, this can account for “understaging” (Nx) with inadequate surgical lymph node yield, and could potentially negatively impact the outcome with disease recurrence.

Pathologic techniques have been implicated as a factor in lymph node yield due to inadequate specimen analysis and processing. Patient-related factors, such as age and obesity, have been associated with decreased lymph node yield.^13–16 Institutional-related factors demonstrate variation and high noncompliance with the 12-or-more lymph node metric.^17,18

In the past decade, the standard of care for rectal cancer has been based on the German Rectal Cancer Study Group trial.¹⁹ This phase III trial compared neoadjuvant chemoradiation with the then-standard postoperative chemoradiation therapy and demonstrated a statistically significant improvement in local recurrence with reduced treatment toxicity, but without a difference in overall survival (OS). Because neoadjuvant chemoradiation was established as the standard of care for clinical stage II/III rectal cancer, studies have demonstrated its' association with decreased lymph node yield.^20–22

Recently, long-term results from the EORTC 22921 study examined the utility of fluorouracil-based adjuvant chemotherapy after preoperative therapy for rectal cancer.²³ This study concluded that adjuvant chemotherapy demonstrated no statistically significant improvement in disease-free survival or OS in patients after neoadjuvant treatment. This naturally prompts the question as to the utility of the current standard of care, which is 6 months of perioperative systemic therapy.

Given the unclear utility of adjuvant therapy after preoperative therapy has been administered, the clinical importance of pathologic nodal yield and status must be questioned given their impact on the treatment decisions made. Hence, an established lymph node metric and nodal status for rectal cancer may or may not be clinically relevant in the era of neoadjuvant chemoradiation administered for locally advanced rectal cancer.

Patients and Methods

NCCN is a not-for-profit alliance comprising 27 of the nation's leading cancer centers whose goal is to improve the quality of multidisciplinary cancer care for patients. NCCN has compiled a prospective NCCN Oncology Outcomes Database for CRCs abstracted from 8 participating NCCN institutions: City of Hope Comprehensive Cancer Center (CCC), Memorial Sloan Kettering Cancer Center, The University of Texas MD Anderson Cancer Center, Dana-Farber Cancer Institute, Fox Chase Cancer Center, Roswell Park Cancer Center, Robert H. Lurie CCC at Northwestern University, and The Ohio State University CCC – James Cancer Hospital and Solove Research Institute. The data collection process was approved by the respective institutions' Institutional Review Boards.

A cohort was compiled of 1,680 patients with clinical stage II/III primary rectal cancer who underwent surgical intervention excluding pelvic exenteration, subtotal colectomy, total proctocolectomy, transanal excision, Hartmann's procedure, and rectosigmoid resections. The patient cohort was divided into 3 groups based on the type of neoadjuvant therapy received before surgery: no-neoadjuvant therapy (n=111), neoadjuvant chemoradiation (n=1,364), and neoadjuvant chemotherapy (n=205). This analysis is inclusive of data from September 2005 to September 2013.

The clinicopathologic characteristics were analyzed and compared using chi-square test for categorical variables and the Kruskal-Wallis test for continuous variables among the 3 groups. Lymph node analysis for group comparisons was undertaken using the Kruskal-Wallis test. Clinicopathologic characteristics effect on lymph node were analyzed using the Kruskal-Wallis or Wilcoxon test as appropriate.

OS was defined from date of diagnosis to date of death. Patients alive at the date of last observation were censored for survival analysis. Survival curves were estimated using the Kaplan-Meier method. Survival curves were compared using the log-rank tests for the univariate analysis. Multivariable Cox regression models were fit to the OS using all the variables with a P value of less than 0.2 in the univariate analysis. Variables with a P value greater than 0.05 were removed sequentially from the Cox regression model. All statistical analyses were conducted using SAS 9.2 (SAS Institute Inc., Cary, NC).

Results

The 3 groups' clinicopathologic characteristics were compared as illustrated in Table 1, showing heterogeneity between groups, with statistical significance demonstrated for median age, age at diagnosis, Charlson comorbidity index, pathologic TNM stage, type of surgery undertaken, median OS, lymph node yield, and node positivity.

The treatment groups demonstrated statistical significance with regard to median age and age at diagnosis. The no-neoadjuvant group had an increased median age of 64 years, compared with 57 and 56 years for neoadjuvant chemoradiation and the neoadjuvant chemotherapy groups, respectively (P<.0001). Likewise, the age of diagnosis in the no-neoadjuvant group demonstrated a higher proportion of patients aged 65 years or older at 45% compared with 26% and 28% in the neoadjuvant chemoradiation and neoadjuvant chemotherapy groups, respectively (P=.0001). In examining the comorbidities of the group, the no-neoadjuvant therapy group had a higher proportion of patients with a Charlson comorbidity index greater than 2, at 16% compared with 8% and 10% in the neoadjuvant chemoradiation and neoadjuvant chemotherapy groups, respectively (P=.02).

Statistically significant differences in pathologic staging were seen among the 3 groups, with the neoadjuvant chemoradiation group showing a greater proportion of patients with pathologic stage 0 disease: 20% compared with 5% and 9% in the no-neoadjuvant and neoadjuvant chemotherapy groups, respectively. Pathologic stage 3 disease was seen in 30% of patients in the neoadjuvant chemoradiation group compared with 60% and 15% in the neoadjuvant chemotherapy and no-neoadjuvant therapy groups, respectively.

The neoadjuvant chemoradiation group demonstrated the lowest median lymph node yield at 15 compared with 17 and 18 in the no-neoadjuvant and neoadjuvant chemotherapy groups, respectively (P<.0001). In analyzing the number of positive nodes between groups, a statistically significant difference was seen, with the neoadjuvant chemotherapy group demonstrating increased positive nodes at a mean of 2.2 ± 3.6 compared with 0.9 ± 2.1 and 0.8 ± 4.4 for the neoadjuvant chemoradiation and no-neoadjuvant groups (P<.0001). In comparing median OS, improved survivals of 42.7 and 37.3 months were seen in the neoadjuvant chemotherapy and neoadjuvant chemoradiation groups, respectively, compared with 26.6 months in the no-neoadjuvant therapy group (P<.0001).

As illustrated in Table 2, univariate analysis demonstrated that the factors associated with an effect on lymph node yield included tumor grade, clinical T stage, surgery type, and neoadjuvant chemoradiation. Multivariate analysis revealed that surgery type, tumor grade, clinical T stage, and neoadjuvant treatment were predictors of OS, with neither node number nor status being statistically significant. When examining nodal status irrespective of nodal yield, patients with positive lymph node status, although not statistically significant, had an increased OS of 38.4 months compared with 36.9 months for node-negative status (P=.09).

Subgroup analyses were undertaken examining the impact of both nodal yield and status on OS, as illustrated in Table 3 and Figure 1. Patients with a lymph node yield of less than 12 had an improved median OS of 43.3 months compared with 36.6 months in patients with 12 or more lymph nodes (P=.009). Further analysis of patients with a

Table 1.

Patient Clinicopathologic Characteristics

View Table

Discussion

Lymph node status remains an essential factor for cancer staging and guiding treatment decisions. In

Table 2.

Univariate and Multivariate Analysis of Clinical and Pathologic Factors Related to Median OS

View Table

CAP understands the importance of lymph node yield, as reflected in their suggested rectal specimen assessment protocols. This requires meticulous tissue assessment but is not necessarily standardized, with

Table 3.

Univariate Analysis Examining Impact of Nodal Yield and Status on Median OS

View Table

Neoadjuvant therapy is a well-accepted non-modifiable factor (with the exception of deferring it) affecting lymph node yield; this was again demonstrated in this analysis with statistical significance seen between groups, with the neoadjuvant chemoradiation group having the lowest median lymph node yield (n=15). In examining the pathologic staging of the groups, the neoadjuvant chemoradiation group not only had a greater stage 0 (pathologic complete response) rate at 20% versus 9% in the neoadjuvant chemotherapy group, but also comprised 30% with pathologic stage III disease versus 60% in the neoadjuvant chemotherapy group. This difference could be reflective of response to the neoadjuvant treatment given but may also relate to earlier clinical staging.

Again, EORTC 22921 concluded that fluorouracil-based adjuvant chemotherapy showed no statistically significant improvement in disease-free survival or OS in patients after neoadjuvant treatment. Among patients with clinical stage II/III rectal cancer who can tolerate multimodality therapy and receive neoadjuvant chemoradiation, the benefit of adjuvant therapy is unclear, and hence the clinical importance of nodal status, and particularly yield, remains in question.

This study has several limitations. This is a retrospective analysis undertaken using the NCCN Oncology Outcomes Database. The NCCN database

View Full Size

Overall survival stratified by nodal yield (<12 and ≥12) and status (negative/positive).

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 14, 12; 10.6004/jnccn.2016.0164

As is often encountered with many large population database studies, the NCCN database has inherent limitations regarding the data available for analysis. There exists missing or even desired but unavailable data for which data imputation and analysis is not reasonable to pursue for scientifically sound statistics and conclusions, thus leaving some questions unanswered. Finally, as in all retrospective analyses, the factors examined and analyzed cannot be considered causative, but rather associated with the obtained results and drawn following conclusions.

Acknowledging the limitations of this large, but retrospective analysis, this analysis demonstrated no association between increased lymph node harvest or nodal status on patient survival irrespective of the patient therapies undertaken. The current NCCN Guidelines discussion based on previous studies acknowledges the impact of neoadjuvant therapy on decreased lymph node yield.²⁴ This study interestingly demonstrated that patients with fewer than 12 lymph nodes harvested had improved OS irrespective of nodal status. Hence, a nodal metric in rectal cancer may not be clinically relevant in the setting of the current standard of care, with neoadjuvant treatment playing a major role in the multimodality treatment of rectal cancer.

Nodal yield and status after neoadjuvant therapy may more importantly be a surrogate marker of disease response and not ideal to base adjuvant therapy decisions on, particularly given the EORTC conclusions showing no demonstrable benefit to 5-FU–based adjuvant therapy. Preoperative clinical staging may in fact be more representative of cancer biology and therefore a more ideal characteristic on which to base adjuvant therapy decision-making to potentially impact and improve patient outcomes. That being said, in patients who were not significantly downstaged by neoadjuvant treatment and had positive lymph node status, one could anticipate potentially adverse outcomes based on their pathologic staging and lack of response to neoadjuvant therapy. Given the impact of neoadjuvant therapy on rectal cancer outcomes, ongoing (eg, NCCTG N1048 [PROSPECT]) and future prospective randomized trials are prudent and needed to further determine the relevance of lymph node yield and status to patient outcomes, and to optimize therapeutic approaches in order to advance the multimodality care of rectal cancer.