Lateral lymph node dissection reduces local recurrence of locally advanced lower rectal cancer in the absence of preoperative neoadjuvant chemoradiotherapy: A systematic review and meta-analysis

DOI: https://doi.org/10.21203/rs.3.rs-42723/v2

Abstract

Background: The role of lateral lymph node dissection (LLND) in the treatment of locally advanced lower rectal cancer remains controversial. The present study was conducted to compare total mesorectal excision (TME) with or without LLND among patients with lower rectal cancer in clinical stage II/III.

Methods: PubMed, Embase, Ovid, Cochrane Library, Google Scholar, and the ClinicalTrials.gov databases were systematically searched for publications that compared TME with or without LLND among patients with lower rectal cancer in clinical stage II/III. Subgroup analysis was performed based on whether preoperative neoadjuvant chemoradiotherapy (nCRT) was undertaken. The hazard ratios (HR), relative risk (RR), and weighted mean difference (WMD) were pooled.

Results: Twelve studies that included 4458 patients were identified in the current meta-analysis. Collected data demonstrated that TME with LLND was associated with significantly longer operation time (WMD 90.73 min, P<0.001), more intraoperative blood loss (WMD 303.20 mL, P<0.001), and postoperative complications (RR=1.35, P=0.02). Urinary dysfunction (RR 1.44, P=0.38), sexual dysfunction (RR 1.41, P=0.17), and postoperative mortality (RR=1.52, P=0.70) were similar between the two groups. No statistically significant differences were observed in OS (HR 0.93, P=0.62), DFS (HR 0.99, P=0.96), total recurrence (RR 0.98, P=0.83), lateral recurrence (RR 0.49, P=0.14) or distant recurrence (RR 0.95, P=0.78) between the two groups regardless the use of nCRT. LLND significantly reduced local recurrence rate of patients who did not receive nCRT (RR 0.71, P=0.004), while the difference was not significant when nCRT was performed (RR 0.70, P=0.36).

Conclusions: Our study found out LLND could not significantly improve survival in locally advanced lower rectal cancer but could reduce the local recurrence in the absence of preoperative nCRT. The advantage of controlling local recurrence might be replaced with nCRT.

Registration: The protocol for this meta-analysis was registered prospectively with PROSPERO (CRD42020135575) on May 16, 2019.

1. Background

Total mesorectal excision (TME) technique has significantly improved the pathological and oncological outcomes and has become the standard surgical procedure for rectal cancer. While approximately 14%-30% of patients with lower rectal cancer in clinical stage II/III develop lateral pelvic lymph nodes (LLNs) metastases, which is outside the surgical field of TME and associated with a greater incidence of local recurrence and decreased survival[1, 2]. In Japan, lateral lymph node dissection (LLND) has been recommended as the standard treatment for patients with lower rectal cancer in clinical stage II/III since the 1970s [3, 4]. While, in Western countries, LLNs metastases were considered to be a sign of distant metastasis and could not be eliminated by surgery only. Therefore, preoperative nCRT (neo-adjuvant chemoradiotherapy) instead of LLND has become the standard regimen for the treatment of locally advanced lower rectal cancer in Western countries [5, 6]. However, studies have shown that preoperative nCRT couldn't completely eradicate the metastatic LLNs, suggesting nCRT followed by TME and LLND may be more effective in the management of locally advanced lower rectal cancer [7, 8].

The efficiency and safety of LLND in locally advanced lower rectal cancer remain controversial. Several studies have reported that LLND could significantly reduce the local recurrence of patients with rectal cancer and prolong survival time [9, 10]. However, some other studies indicated that LLND has no benefits in improving survival or reducing recurrence rates, and it also significantly increases urinary and sexual dysfunction [11, 12]. Two previous meta-analyses performed approximately ten years ago indicated LLND had no advantage in controlling recurrence or improving survival and appeared to be associated with increased urinary and sexual dysfunction [13, 14]. However, neither of these two meta-analyses explicitly restricted tumor anatomical location and clinical stage, leading to upper rectal cancers and early-stage rectal cancers were included in their studies. Actually, the application of LLND in upper-third or early-stage rectal carcinoma has practically been abandoned since 2000, and LLND was primarily performed for locally advanced lower rectal cancer at present [4]. In addition, nCRT is currently the primary treatment regimen for locally advanced rectal cancer, and neither of these two meta-analyses separately assessed the effects of LLND on patients who have received preoperative nCRT.

Several studies over the past decade, including large RCTs and well-designed cohort studies, were performed to clarify the value of LLND in stage II/III lower rectal cancer but provided controversial results [9, 15, 16]. A new meta-analysis is required to integrate the results of previous studies and make a clearer conclusion. Therefore, we performed the current meta-analysis to assess the efficacy and safety of LLND in locally advanced lower rectal cancer. We also evaluated the value of LLND in patients who underwent preoperative nCRT by subgroup analyses. The protocol for this meta-analysis was registered prospectively with PROSPERO (CRD42020135575).

2. Methods

2.1. Literature search:

A systematic search of all peer-reviewed literature was performed in electronic databases, including MEDLINE (via PubMed), Embase, Ovid, and the Cochrane Library up to December 22, 2019. Related literature in the first ten pages of the Google Scholar database was also reviewed. The following MeSH search headings and their synonyms were used: "total mesorectal excision", "lateral lymph node dissection", "extended lymphadenectomy", "lateral pelvic wall lymph-node dissection", "rectal neoplasms", "rectal cancer", "comparative study" and "treatment outcome". The related-articles function was used to broaden the search, and reference lists of relevant studies and related systematic reviews were screened manually. Full-text review was performed after a screening of the title and abstract. If multiple comparisons were used in the study, data were extracted only from groups that meet inclusion criteria.

2.2. Selection criteria:

All comparative studies evaluating the efficiency or safety of TME combined with LLND versus TME alone in the treatment of stage II/III lower rectal cancer were included. Studies with the following inclusion criteria were eligible: (1) Patients with locally resectable clinical stage II/III rectal cancer without evidence of metastatic disease at the time of surgery and tumor location within 8 cm from the anal verge, or the major part of tumor located at or below the peritoneal reflection; (2) Patients between the two groups with similar clinical characteristics and therapeutic schemes. The following exclusion criteria were used: (1) Tumor lesions located in the upper third of the rectum, or the major part of tumor located above the peritoneal reflection; (2) Patients with distinct therapeutic protocols or significantly different clinical characteristics between the two groups; (3) Patients with distant metastasis at the time of treatment or other malignant diseases or fixed tumors; And (4) Animal studies, letters, comments, and editorials. In cases of considerable overlap in subjects between studies published on a single clinical trial, the most recent or most informative study was included, and the results were used complementary.

2.3. Data extraction:

Two reviewers independently performed data extraction and study quality assessment. Consistent extraction data between reviewers were used directly for the final analysis, and disagreements between reviewers were discussed and resolved via consensus. The primary endpoints were 5-year overall survival (OS) and disease-free survival (DFS). Secondary endpoints including total recurrence, local recurrence, lateral recurrence, distant recurrence, operation time, intraoperative blood loss, postoperative complications and mortality, urinary dysfunction, and sexual dysfunction.

2.4. Study quality assessment

The Newcastle-Ottawa scale criteria recommended by the Cochrane Library for including trials were used to evaluate the quality of the cohort studies. The quality of RCTs was measured by using the Cochrane Collaboration's risk for bias assessment tool. Two reviewers assessed the quality of the studies. Where discrepancies arose, papers were re-examined, and the consensus was reached via discussion.

2.5. Statistical analysis

The meta-analysis was performed consistent with the recommendations from the Cochrane Collaboration and the Quality of Reporting of Meta-analyses (QUORUM) guidelines [17, 18]. Hazard ratios (HRs) and the respective 95% confidence intervals (CIs) were assessed as effect measures for time-to-event data (5-year OS and DFS). If HRs were not reported but adequate information (e.g., Kaplan-Meier plots) was available, the estimation methods described by Parmar and Tierney were used to estimate HRs and the respective 95% CIs [19, 20]. Risk ratio (RR) was used as the summary statistic for statistical analyses of dichotomous variables, and weighted mean difference (WMD) was used to analyze continuous variables. P-values for the overall effects were calculated based on a two-sided Z-test for independent samples for effect measures on a log scale. A P-value < 0.05 was considered statistically significant. Meta-analytic results were graphically displayed in Forest plots.

Heterogeneity was tested using chi-squared analyses and defined as present in cases of a P-value < 0.10. I2 > 40% was considered statistically significant heterogeneity, and the random-effects model was used to calculate overall effect estimates after examining the causes of heterogeneity. Otherwise, the fixed-effects model was used. Subgroup analysis was performed based on whether preoperative nCRT was undertaken. Review Manager version 5.3 was used for the meta-analysis (Copenhagen, the Nordic Cochrane Centre) [21].

3. Results

3.1. Study selection A total of 1564 citations were identified using the predefined search strategy (Fig. 1). After screening the titles and abstracts, 1499 of the studies were excluded due to lack of relevance. Sixty-five articles were further evaluated for eligibility. Among these publications, 54 studies were excluded due to the following reasons: 32 studies did not meet selection criteria; 2 studies were meta-analyses; 14 studies were reviews; 3 studies data were not extractable; 3 studies with overlapping data. Four studies based on one same randomized trial were included because they reported different outcomes. Full manuscripts were available for 11 studies, and the results of one RCT were available as a conference proceeding presented on the 2017 ECCO European Cancer Congress [15]. Following these exclusions, 12 studies published from 2001 to 2019 and involving a total of 4458 patients (1952 in the TME+LLND group and 2506 in the TME alone group) fulfilled the selection criteria were included in the current meta-analysis. The flow diagram is shown in Fig. 1. 3.2. Characteristics of the included studies Six of the included studies were RCTs, and the other six studies were non-RCTs. According to the Cochrane bias assessment, all of the RCTs mentioned "randomization", but only four studies (based on the same research) reported the generation of an adequate randomized sequence and mentioned that the allocation procedure was not masked to investigators or patients. Another two RCTs failed to report the randomization procedure or mentioned whether blinding was adopted [see Additional file 1 and Additional file 2]. The six non-RCTs were all cohort studies, including five retrospective studies and one prospective study with prospectively collected data. The quality of the non-RCTs was evaluated using the Newcastle-Ottawa criteria. As shown in Table 1, the total number of stars of the six non-RCTs was not less than seven for each study. The basic information of the eligible studies is listed in Table 2. Study outcomes are shown in Table 3.
Table 1

Scores of 6 cohort studies using Newcastle-Ottawa Criteria

Study

Selection

Comparability

Outcomes

Total

 

Representativeness of the exposed cohort

Selection of the nonexposed cohort

Ascertainment of exposure

Demonstration that outcome of interest was not present at the start of the study

Comparability of cohorts on the basis of the design or analysis

Assessment of outcome

Was follow-up long enough for outcomes to occur

Adequacy of the follow-up of cohorts

 

Fujita, S. 2003

1

1

1

1

0

1

1

1

7

Kusters, M. 2009

1

1

1

1

2

0

1

1

8

Watanabe, T. 2002

1

1

1

1

1

1

1

1

8

Oki, Eiji 2019

1

1

1

1

1

1

1

1

8

Ozawa, H. 2016

1

1

1

1

2

0

1

1

8

Ogura 2019

1

1

1

1

0

1

1

1

7


Table 2. Characteristics of the 12 included studies

Study

Study year(region)

Research type

Group name

Sample size

Mean age(years)

Sex ratio(M:F)

Median follow-up time

Tumor location

Clinical stage

Preoperative therapy regimen (n)

Postoperative therapy regimen (n)

Indication of LLND

Type of LLND

Type of TME

Matching criteria

Nagawa, H

2001(Japan)

RCTs

TME+LLND

23

59.1(±10.1)

17:6

N/A

Middle, Low

Stage B, C

50Gy (2Gy/d*5d/wk*5wk) (23)

5-FU-based chemotherapy (23)

Random controlled
allocation

Bilateral LLND

LAR, APR

A, B, C, D, E, F, H, I

TME alone

22

60.1(±8.8)

16:6

50Gy (2Gy/d*5d/wk*5wk) (22)

5-FU-based chemotherapy (22)

NO

Fujita, S

2003(Japan)

Retrospective

TME+LLND

204

57(±10)

133:71

59 months

Middle, Low

TNM II/III

NO

NO

No lateral lymph nodes metastases

Bilateral LLND

LAR, APR

A, B, C, D, E, F, G, H, I

TME alone

42

64(±12)

24:18

NO

Kusters, M

2009(Japan & Netherlands)

Retrospective

TME+LLND

324

58(±11)

215:109

7.9 years

Middle, Low

TNM II/III

NO

Postoperative chemoradiotherapy (27)

Non restriction but the same inclusion criteria in the two arms

Unilateral and bilateral LLND

LAR, APR

B, C, D, E, F, G, I

TME alone

376

64(±11)

234:142

7.0 years

Postoperative chemoradiotherapy (61)

NO

Watanabe, T

2002(Japan)

Retrospective

TME+LLND

75

N/A

N/A

N/A

Middle, Low

Stage B, C

50Gy(2Gy/d*5d/wk*5wk) (75)

NO

No lateral lymph nodes metastases

Bilateral LLND

LAR, APR, or Hartmann

A, B, C, D, E, G, H, I

TME alone

40

N/A

N/A

50Gy(2Gy/d*5d/wk*5wk) (40)

NO

Fujita, S

2012(Japan)

RCTs

TME+LLND

351

61(54-67)Ѱ

236:115

N/A

Middle, Low

TNM II/III

NO

5-FU-based chemotherapy (163)

Random controlled
allocation

Bilateral LLND

LAR, APR, or Hartmann

A, B, C, D, E, F, G, H, I

TME alone

350

62(55-68)Ѱ

236:114

5-FU-based chemotherapy (153)

NO

Dev, K

2017(India)

RCT

TME+LLND

163

N/A

N/A

N/A

Middle, Low

TNM II/III

25Gy(5gy*5) (163)

NO

Random controlled
allocation

Bilateral LLND

TME without explaining the details

A, B, C, D, E, F, G, H, I

TME alone

77

N/A

N/A

25Gy(5gy*5) (77)

NO

Saito, S

2016(Japan)

RCTs

TME+LLND

351

61(55-66)Ѱ

N/A

N/A

Middle, Low

TNM II/III

NO

5-FU-based chemotherapy (163)

Random controlled
allocation

Bilateral LLND

LAR, APR, or Hartmann

A, B, C, D, E, F, G, H, I

TME alone

350

62(56-69)Ѱ

N/A

5-FU-based chemotherapy (153)

NO

Fujita, S

2017(Japan)

RCTs

TME+LLND

351

61(26-75)@

236:115

72.2 months

Middle, Low

TNM II/III

NO

5-FU-based chemotherapy (163)

Random controlled
allocation

Bilateral LLND

LAR, APR, or Hartmann

A, B, C, D, E, F, G, H, I

TME alone

350

62(26-75)@

236:114

5-FU-based chemotherapy (153)

NO

Ito, Masaaki

2018(Japan)

RCTs

TME+LLND

351

61(26-75)@

236:115

N/A

Middle, Low

TNM II/III

NO

5-FU-based chemotherapy (163)

Random controlled
allocation

Bilateral LLND

LAR, APR, or Hartmann

A, B, C, D, E, F, G, H, I

TME alone

350

62(26-75)@

236:114

5-FU-based chemotherapy (153)

NO

Oki, Eiji

2019(Japan)

Prospective

TME+LLND

215

60.7(±9.4)

159:56

5 years

Middle, Low

TNM II/III

NO

5-FU-based chemotherapy (215)

Random controlled
allocation

Bilateral LLND

LAR, APR, Hartmann, or Others

B, C, D, E, F, G, H, I

TME alone

230

63.5(±8.9)

151:79

5-FU-based chemotherapy (230)

NO

Ozawa, H

2016(Japan)

Retrospective

TME+LLND

499

N/A

356:143

N/A

Middle, Low

TNM II/III

NO

Postoperative chemotherapy (193)

Non restriction but the same inclusion criteria in the two arms

Bilateral LLND

LAR, APR, or Others

A, B, C, D, E, F, I

TME alone

499

N/A

334:165

Postoperative chemotherapy (207)

NO

Ogura

2019(Japan)

Retrospective

TME+LLND

98

N/A

N/A

56.5 (55.0-58.1) @

Middle, Low

TNM II/III

45‐50.4 Gy/25Gy+Oxaliplatin-based/5-FU-based chemotherapy (98)

Partial patients received adjuvant chemotherapy

Non restriction but the same inclusion criteria in the two arms

Bilateral LLND

LAR, APR, Hartmann, ISR, or TPE

C, D, E, F,

TME alone

870

N/A

N/A

45‐50.4 Gy/25Gy+Oxaliplatin-based/5-FU-based chemotherapy (870)

NO

For matching criteria A=year, B=sex, C=tumor location, D= neo-adjuvant, E=adjuvant, F=lateral lymph-node status G=lymph and vessel invasive H=tumor differentiation I=lateral lymph node status;Ω: Astler-Coller staging system; @: Values are presented as the median (range); Ѱ: Values are presented as the median (IQR); ※: Values are presented as the mean ± standard deviation. FU: fluorouracil; Gy: LAR: Low anterior resection; APR: Abdominoperineal resection; Hartmann: Hartmann’s procedure.

 

Table 3

Results of meta-analysis comparing TME + LLND versus TME alone

 

Number of studies

TME + LLND patients

TME patients

Total patients

HR/RR/WMD (95% CI)

P value

Study heterogeneity

             

χ²

df

p-value

Survival

                   

5-year survival

4

1088

1101

2189

0.93*(0.71–1.22)

0.62

6

3

50%

0.11

5-year disease-free survival

5

868

684

1552

0.99*(0.74–1.34)

0.96

9.93

5

50%

0.08

Recurrence

                   

Total recurrence

4

653

454

1107

0.98(0.81–1.18)

0.83

2.37

4

0%

0.67

Local recurrence

7

1290

1930

3220

0.71(0.56–0.89)

0.003

9.22

7

24%

0.24

Lateral recurrence

3

773

1596

2369

0.49(0.18–1.28)

0.14

5.87

2

66%

0.05

Distant recurrence

5

615

1204

1819

0.95(0.68–1.34)

0.78

8.84

5

43%

0.12

Peri-operative outcomes

                   

Length of operation (min)

4

716

479

1195

97.03(75.35-118.72)

P < 0.001

82.14

3

96%

P < 0.001

Blood loss (mL)

4

716

479

1195

303.20(156.82–449.58)

P < 0.001

201.99

3

99%

P < 0.001

Peri-operative mortality

2

578

414

992

1.52(0.18–12.65)

0.7

0.47

1

0%

0.49

Postoperative complications

3

578

414

992

1.35(1.05–1.74)

0.02

0.53

2

0%

0.77

Functional outcomes

                   

Urinary dysfunction

2

374

372

746

1.44(0·63–3.28)

0·38

4.93

1

80%

0.03

Sexual dysfunction

2

108

92

200

1.41(0.87–2.31)

0·17

2.23

1

55%

0.13

HR = hazard ratio. RR = risk ratio. WMD = weighted mean difference. df = degrees of freedom. *HR. †WMD.


3.3. Primary endpoints: 5-year OS and DFS.

Four studies, with a total of 2189 patients, were pooled into the analysis of 5-year OS [9-12]. The results demonstrated no significant difference in 5-year OS between the LLND group and TME alone group (HR 0.93, 95% CI 0.71–1.22, P=0.62) with moderate heterogeneity (I²=50%, P=0.11). Subgroup analysis showed no significant difference in 5-year OS between the two groups no matter nRCT was undertaken or not (HR=1.41, 95% CI 0.56-3.55, P=0.47 vs HR=0.90, 95% CI 0.68-1.20, P=0.42). The details are shown in Fig. 2a.

Five studies, including 1552 patients, were pooled into the analysis of 5-year DFS [2, 9, 11, 12, 22]. The results indicated no significant difference in 5-year DFS between the two groups (pooled HR 0.99, 95% CI 0.74–1.34, P=0.96) with moderate between-study heterogeneity (I²=50%, P=0.08). Subgroup analysis showed no significant difference in 5-year DFS between these two groups regardless of the application of nCRT (HR=0.71, 95% CI 0.40-1.25, P=0.23 vs HR=1.08, 95% CI 0.75-1.55, P=0.69). The details are shown in Fig. 2b.

3.4. Secondary endpoints: total, local, lateral, and distant recurrence, operation time, intraoperative blood loss, postoperative complications, perioperative mortality, sexual and urinary dysfunction.  

            Four studies with a total of 1107 patients were eligible for the analysis of 5-year total recurrence [2, 9, 11, 22]. No significant difference in total recurrence was found between the two groups (RR 0.98, 95% CI 0.81–1.18, P=0.83) with no between-study heterogeneity (I²=0%, P=0.67). Subgroup analysis showed no significant difference in 5-year DFS between the two groups regardless of the application of nCRT (RR=1.46, 95% CI 0.76-2.81, P=0.25 vs RR=0.94, 95% CI 0.77-1.14, P=0.53). The details are shown in Fig. 3a.

         Seven studies with a total of 3220 patients were pooled into the analysis of 5-year local recurrence [2, 6, 9, 11, 12, 16, 22]. The results indicated the LLND group had significantly lower 5-year local recurrence than the TME alone group (RR 0.71, 95% CI 0.56–0.89, P=0.003) with low between-study heterogeneity (I²=24%, P=0.24). Subgroup analysis found the LLND group had a significantly lower incidence of local recurrence than the TME alone group when preoperative nCRT was not performed (RR 0.71, 95% CI 0.56–0.89, P=0.004). However, the difference was not significant once preoperative nCRT was introduced (RR 0.70, 95% CI 0.32–1.51, P=0.36). The details are shown in Fig. 3b.

Data on 5-year lateral recurrences were extracted from three studies with 2369 patients [6, 9, 16]. The results demonstrated no significant difference in lateral recurrence between the two groups (RR 0.49, 95% CI 0.18–1.28, P=0.14) with moderate heterogeneity (I²=66%, P=0.05). Subgroup analysis indicated no significant difference in lateral recurrence between the two groups regardless of the introduction of nCRT (RR=0.72, 95% CI 0.27-1.97, P=0.53 vs RR=039, 95% CI 0.08-1.89, P=0.24). The details are shown in Fig. 3c.

Five-year distant recurrence was reported in five studies that investigated 1819 patients [2, 11, 12, 16, 22]. The results demonstrated no significant difference in distant recurrence between the two groups (RR 0.95, 95% CI 0.68–1.34, P=0.78) with moderate between-study heterogeneity (I²=43%, P=0.12). Subgroup analysis revealed no significant difference in distant recurrence between the two groups regardless of the application of preoperative nCRT (RR=0.74, 95% CI 0.41-1.33, P=0.32 vs RR=1.14, 95% CI 0.89-1.47, P=0.29). The details are shown in Fig. 3d.

Four studies were included in the meta-analysis that assessed the length of operation in 1195 patients [2, 11, 15, 23]. Results demonstrated a significant difference that favored the TME alone group (WMD 90.73 min, 95% CI 75.35-118.72, P<0.001) with significant between-study heterogeneity (I²=96%, P<0.001). The details are shown in Fig. 4a.

Four studies were included in the meta-analysis to assess intraoperative blood loss in 1195 patients [2, 11, 15, 23]. Results indicated the TME alone group showed significantly lower intraoperative blood loss than the LLND group (WMD 303.20 mL, 95% CI 156.82–449.58, P<0.001) with high between-study heterogeneity (I²=99%, P<0.001). The details are shown in Fig. 4b.

Three studies assessed 992 patients and reported postoperative complications [2, 11, 23]. The LLND group was associated with a higher rate of postoperative complications than the TME alone group (RR=1.35, 95% CI 1.05-1.74, P=0.02) with no between-study heterogeneity (I²=0%, P=0.77). The details are shown in Fig. 5a.

Perioperative mortality was reported in three studies that investigated 992 patients [2, 11, 23]. The data extracted from one of the studies were not suitable for meta-analysis because no events occurred in either group [11]. Ultimately, two studies, including 947 patients, were pooled into analysis [2, 23]. The results indicated no significant difference in perioperative mortality between the two groups (RR=1.52, 95% CI 0.18-12.65, P=0.70) with no between-study heterogeneity (I²=0%, P=0.49). The details are shown in Fig. 5b.

Two RCTs studies assessed 200 patients and reported sexual dysfunction [11, 24]. Results indicated no significant difference in sexual dysfunction between the two groups (pooled RR 1.41, 95% CI 0.87–2.31, P=0.17) with moderate between-study heterogeneity (I²=55%, P=0.13). The details are shown in Fig. 5c.

Two RCTs studies assessed 746 patients and reported urinary dysfunction [11, 25]. Our results demonstrated no significant difference in urinary dysfunction between the two groups (pooled RR 1.44, 95% CI 0.63–3.28, P=0.38) with high between-study heterogeneity (I²=80%, P=0.03). The details are shown in Fig. 5d.

4. Discussion

The present study is the first meta-analysis evaluating the efficiency and safety of LLND in stage II/III of lower rectal cancer. The aggregated data demonstrated LLND did not decrease the total or distant recurrence but significantly reduced the local recurrence of patients with lower rectal cancer in clinical stage II/III at the cost of longer operation time, greater intraoperative blood loss, and higher incidence of postoperative complications. The benefit in controlling local recurrence did not translate to a better 5-year DFS or OS. Notably, LLND neither posed an additional risk on postoperative mortality, nor did it increase the risk of sexual and urinary dysfunction. Subgroup analysis showed LLND significantly reduced the local recurrence in patients who received no preoperative nCRT but not in the patients who did. Our results also suggested LLND has no benefit in prolonging the survival time of patients with rectal cancer regardless of the application of preoperative nCRT.

Our results are different from that of the previous two meta-analyses performed by Georgiou et al. and Chen et al. approximately ten years ago [13, 14]. These two meta-analyses indicated LLND neither reduced tumor recurrence nor prolonged survival time but significantly affected urinary and sexual function. However, confined by the limited quality of the studies included in their meta-analyses, there were inherent flaws in their results. For example, the clinical characteristics were significantly different between the two groups, and the LLND group had more advanced tumors, i.e., larger (higher T stage) [26], node-positive [26, 27], and more aggressive pathology [28] compared to the TME alone group. Furthermore, upper rectal cancers and early-stage rectal tumors (T1) were also included in their studies [28-32]. However, the Japanese guidelines have recommended the application of LLND limited to stage II/III of lower rectal cancer since the 2000s, regardless of lymphatic node metastasis [4]. Besides, in the study performed by Chen and colleagues, the time-to-event data were analyzed as dichotomous outcomes instead of the generally recommended method of log HRs and its standard error [14]. Therefore, there were certain limitations in applying their results to guide the application of LLND in clinical practice. Our study included more high-quality trials than the previous two meta-analyses, and all of the included studies had similar clinical features between the two groups. In addition, only studies with clinical stage II/III of lower rectal cancers were included in our study. Therefore, our study provides more powerful and valid results than the previous two meta-analyses.

Our results demonstrated LLND significantly reduced local recurrence of the patients who didn't receive preoperative nCRT, but the difference was not significant when nCRT was performed. These results indicated the advantage of LLND in controlling local recurrence might be replaced with preoperative nCRT. Caution should be taken when interpreting these results because no subgroup analysis was performed based on the pretreatment size of LLNs. Previous studies indicated that patients with positive LLNs have a higher rate of local recurrence, and nCRT followed by TME only was not sufficient to completely eradicate the metastatic LLNs and avoid local recurrence [33, 34]. Akiyoshi et al. reported 30-40% of patients with pretreatment-positive LLN developed local recurrence even after nCRT, and it reduced to almost zero when additional LLND was performed [35]. Ogura et al. also noted that 25.6% of patients with pretreatment-positive LLN developed local recurrence even after receiving nCRT and radical resection, and it reduced to 5.7% when extra LLND was performed [16]. Therefore, for patients with pretreatment-positive LLNs, nCRT followed by TME alone may not be sufficient, and selective LLND should probably be considered [7, 33]. The value of selective LLND in patients who received nCRT remains controversial. A current phase III Chinese randomized controlled trial (NCT02614157) to demonstrate the safety and efficacy of selective LLND after nCRT in the treatment of advanced lower rectal cancer bearing enlarged LLNs is being performed and may provide more reliable evidence [36].

The pattern of local recurrence can be divided into three categories: central pelvis recurrence, anastomosis recurrence, and lateral recurrence. The current study found LLND reduced the incidence of lateral recurrence, but without significant difference. The reason may be the different local recurrence patterns in the patients from our studies. Several studies have shown that the most common site of local recurrence varies among patients in different regions. A Dutch trial indicated the most common site of recurrence was the central pelvis, and only 24% of the local recurrence originated from the lateral pelvis in the TME alone group [6]. Besides, a study from Sweden also demonstrated lateral recurrence was not a major cause of local recurrence, and only 6% (2/33) of patients with local recurrence exhibited lateral pelvic recurrence [37]. However, a study by Nagasaki et al. from Japan suggested the most common site of local recurrence was the lateral pelvis, and approximately 50% of the patients with local recurrence developed lateral recurrence [38]. In addition, a study by Kim et al. from Korea also demonstrated approximately 65% (42/65) of patients with local recurrence developed lateral pelvic recurrence even after receiving nCRT and radical dissection [39]. Analogously, Fujita et al. reported a much higher rate of lateral pelvic recurrence (57%) in the TME alone group than Kusters and colleagues (24%) in the current meta-analysis, which may be the reason for the high between-study heterogeneity [6, 9]. Therefore, patients in East Asia tend to have a higher incidence of lateral pelvic recurrence, and LLND may play a more important role in East Asian patients than patients in Europe.

We also found that LLND couldn't improve the 5-year OS or DFS of patients with rectal cancer regardless of the application of preoperative nCRT. The results indicated LNNs metastases might be a sign of systemic disease with a dismal prognosis rather than a regional disease and couldn't be eliminated by surgery only [5]. Previous studies have demonstrated that the 5-year OS of patients with lateral lymph node enlargement is still poor (20-45%) even though local control has been achieved by the application of LLND [1, 40, 41]. Oki et al. also indicated LLND brings no benefits in improving 5-year DFS or OS of patients receiving no preoperative nCRT [12]. Besides, the Japanese randomized trial also demonstrated LLND couldn't prolong the survival time of patients with rectal cancer [9]. Actually, the latest tumor node metastasis classification (AJCC 8th edition) has classified LLNs involvement as distant disease, and preoperative nCRT followed by TME has been recommended as the standard treatment regimens [42]. However, whether LLND provides additional benefits to patients who have received nCRT remains controversial.

 The current meta-analysis found no significant differences in 5-year OS and DFS between the two groups after receiving preoperative nCRT. It is worth noting that these results were obtained without limiting the pretreatment size of LLNs, and studies with negative LLNs were also included in our study. While, it has been reported that the pretreatment size of LLNs was significantly associated with survival outcomes, and patients with positive LLNs have significantly worse survival rates [39, 43]. MERCURY study demonstrated patients with enlarged LLNs had significantly lower 5-year DFS than that of the patients with negative LLNs (42% vs 70.7%) [43]. Kim et al. also identified LLNs short-axis diameter ≥10 mm was significantly associated with lower 5-year OS and DFS, even after nCRT and TME [39]. A subgroup analysis based on the pretreatment size of LLNs was planned during the design phase of the present meta-analysis, but no sufficiently detailed information was provided in the included trials to perform this subgroup analysis. Therefore, whether LLND provides additional survival benefits to patients with pretreatment-positive LLNs who received nCRT remains unknown, and the phase III Chinese randomized controlled trial (NCT02614157) mentioned above may provide strong evidence [36].

Our data also demonstrated TME followed by LLND required longer operation time and resulted in greater blood loss than TME alone, and these results are expected. It is not difficult to understand that LLND combined with TME required more operation time because LLND is a meticulous and systemic procedure. Two trials performed approximately 2000 indicated the mean difference in intraoperative blood loss was greater than 500 mL between the two groups [2, 11]. However, two recent RCTs by Fujita et al. and Dev et al. showed the mean differences were 239 ml and 70 ml, respectively [15, 23]. A reasonable explanation may be due to the improvements in surgical techniques, and blood loss may have been minimized compared with the earlier studies.

Our study also found that LLND was associated with more frequent and severe postoperative complications, but LLND did not increase the risk of postoperative mortality. These results should be interpreted with caution when applied to clinical practice, as all three trials included in the current meta-analysis reported extremely low incidences of postoperative mortality in both groups [2, 11, 23]. Notably, the aggregated data also demonstrated LLND did not bring additional risks on sexual or urinary dysfunction. Therefore, the potential damage to urinary and sexual function cannot be a stumbling block to prevent the application of LLND for the treatment of rectal cancer.

The limitations of the current study should not be neglected because only six RCTs and six non-RCTs were included in our meta-analysis. Four of the six RCTs studies reported different outcomes based on the same randomized trial,and the results of another study were extracted from conference proceedings [15]. However, all of the included studies were of high quality (achieving more than seven stars) according to the Newcastle-Ottawa Scale (for non-RCTs)[44] or the Cochrane Collaboration's risk of bias tool (for RCTs) [45]. The follow-up times were different across studies, but the time was sufficient for outcomes to occur, and subjects lost to follow-up were unlikely to introduce bias. Despite meeting the inclusion criteria, clinical heterogeneity was present due to the different pretreatment statuses of LLNs between the included studies, which may have introduced bias. The specific surgical methods and quality were different across studies, which presents another possibility to introduce bias. Despite these limitations, the current study provides the most comprehensive and up-to-date information on the frequently discussed value of the routine use of LLND in the treatment of stage II/III lower rectal cancer.

5. Conclusions

In conclusion, our research demonstrated the general application of LLND in locally advanced lower rectal cancer couldn't improve 5-year OS or DFS but only could reduce the local recurrence of patients who received no nCRT, at the cost of longer operation time, greater blood loss and more complications. Furthermore, the routine application of LLND brought no additional benefits in controlling recurrence or improving survival in patients who had received preoperative nCRT. LLND should be selectively performed in patients with more risk factors, for example, patients with positive lateral lymph nodes. Future studies are needed to demonstrate whether selective LLND provides additional benefits for patients who have received preoperative nCRT.

Abbreviations

TME: Total Mesorectal Excision; LLNs: Lateral pelvic Lymph Nodes; LLND: Lateral Lymph Node Dissection; nCRT: Neoadjuvant Chemoradiotherapy; HR: Hazard Ratio; RR: Relative Risk; WMD: Weighted Mean Difference; OS: Overall Survival; DFS: Disease-Free Survival; RCT: Randomised Controlled Trial; CI: Confidence Interval;

Declarations

Acknowledgments

Not applicable.

Authors' contributions

XG, CW, and YYY made substantial contributions to conception and design, acquisition, analysis, and the interpretation of data; XG, CW, YYY and LY participated in drafting the article; XG, LY, and ZGZ participated in revising it critically for important intellectual content; All of the authors gave final approval of the version to be published.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or nonprofit sectors.

Availability of data and materials

No additional data is available.

Ethics approval and consent to participate

All analyses were based on previously published studies. Therefore, no ethical approval or patient consent was required.

Consent for publication

Not applicable.

Competing interests

There are no conflicts of interest of any authors in relation to the submission of this manuscript.

References

  1. Sugihara K, Kobayashi H, Kato T, et al. Indication and benefit of pelvic sidewall dissection for rectal cancer. Dis Colon Rectum. 2006;49(11):1663-72.
  2. Fujita S, Yamamoto S, Akasu T, Moriya Y. Lateral pelvic lymph node dissection for advanced lower rectal cancer. Br J Surg. 2003;90(12):1580-5.
  3. Takahashi T, Ueno M, Azekura K, Ohta H. Lateral node dissection and total mesorectal excision for rectal cancer. Dis Colon Rectum. 2000;43(10 Suppl):S59-68.
  4. Watanabe T, Muro K, Ajioka Y, et al. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2016 for the treatment of colorectal cancer. Int J Clin Oncol. 2018;23(1):1-34.
  5. Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med. 2006;355(11):1114-23.
  6. Kusters M, Beets GL, van de Velde CJ, et al. A comparison between the treatment of low rectal cancer in Japan and the Netherlands, focusing on the patterns of local recurrence. Ann Surg. 2009;249(2):229-35.
  7. Akiyoshi T, Ueno M, Matsueda K, et al. Selective lateral pelvic lymph node dissection in patients with advanced low rectal cancer treated with preoperative chemoradiotherapy based on pretreatment imaging. Ann Surg Oncol. 2014;21(1):189-96.
  8. Ishihara S, Kawai K, Tanaka T, et al. Oncological Outcomes of Lateral Pelvic Lymph Node Metastasis in Rectal Cancer Treated With Preoperative Chemoradiotherapy. Dis Colon Rectum. 2017;60(5):469-76.
  9. Fujita S, Mizusawa J, Kanemitsu Y, et al. Mesorectal Excision With or Without Lateral Lymph Node Dissection for Clinical Stage II/III Lower Rectal Cancer (JCOG0212): A Multicenter, Randomized Controlled, Noninferiority Trial. Ann Surg. 2017;266(2):201-7.
  10. Ozawa H, Kotake K, Hosaka M, Hirata A, Sugihara K. Impact of Lateral Pelvic Lymph Node Dissection on the Survival of Patients with T3 and T4 Low Rectal Cancer. World J Surg. 2016;40(6):1492-9.
  11. Nagawa H, Muto T, Sunouchi K, et al. Randomized, controlled trial of lateral node dissection vs. nerve-preserving resection in patients with rectal cancer after preoperative radiotherapy. Dis Colon Rectum. 2001;44(9):1274-80.
  12. Oki E, Shimokawa M, Ando K, et al. Effect of lateral lymph node dissection for mid and low rectal cancer: An ad-hoc analysis of the ACTS-RC (JFMC35-C1) randomized clinical trial. Surgery. 2019;165(3):586-92.
  13. Georgiou P, Tan E, Gouvas N, et al. Extended lymphadenectomy versus conventional surgery for rectal cancer: a meta-analysis. Lancet Oncol. 2009;10(11):1053-62.
  14. Cheng H, Deng Z, Wang ZJ, Zhang W, Su JT. Lateral lymph node dissection with radical surgery versus single radical surgery for rectal cancer: a meta-analysis. Asian Pac J Cancer Prev. 2011;12(10):2517-21.
  15. Dev K, Gurawalia J, Krishnamurthy S, Kumar VK, Ramachandra C. Role of lateral lymph node dissection in improving survival in low rectal cancer. A single institute, prospective study. European Journal of Cancer. 2017;72(Supplement 1):S65-S.
  16. Ogura A, Konishi T, Cunningham C, et al. Neoadjuvant (Chemo)radiotherapy With Total Mesorectal Excision Only Is Not Sufficient to Prevent Lateral Local Recurrence in Enlarged Nodes: Results of the Multicenter Lateral Node Study of Patients With Low cT3/4 Rectal Cancer. J Clin Oncol. 2019;37(1):33-43.
  17. Clarke M, Horton R. Bringing it all together: Lancet-Cochrane collaborate on systematic reviews. Lancet. 2001;357(9270):1728-.
  18. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283(15):2008-12.
  19. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials. 2007;8:16.
  20. Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med. 1998;17(24):2815-34.
  21. Review Manager (RevMan) [Computer program]. Version 5.3. Copenhagen. The Nordic Cochrane Centre, The Cochrane Collaboration; 2014.
  22. Watanabe T, Tsurita G, Muto T, et al. Extended lymphadenectomy and preoperative radiotherapy for lower rectal cancers. Surgery. 2002;132(1):27-33.
  23. Fujita S, Akasu T, Mizusawa J, et al. Postoperative morbidity and mortality after mesorectal excision with and without lateral lymph node dissection for clinical stage II or stage III lower rectal cancer (JCOG0212): results from a multicentre, randomised controlled, non-inferiority trial. Lancet Oncol. 2012;13(6):616-21.
  24. Saito S, Fujita S, Mizusawa J, et al. Male sexual dysfunction after rectal cancer surgery: Results of a randomized trial comparing mesorectal excision with and without lateral lymph node dissection for patients with lower rectal cancer: Japan Clinical Oncology Group Study JCOG0212. Eur J Surg Oncol. 2016;42(12):1851-8.
  25. Ito M, Kobayashi A, Fujita S, et al. Urinary dysfunction after rectal cancer surgery: Results from a randomized trial comparing mesorectal excision with and without lateral lymph node dissection for clinical stage II or III lower rectal cancer (Japan Clinical Oncology Group Study, JCOG0212). Eur J Surg Oncol. 2018;44(4):463-8.
  26. Matsuoka H, Masaki T, Sugiyama M, Atomi Y. Impact of lateral pelvic lymph node dissection on evacuatory and urinary functions following low anterior resection for advanced rectal carcinoma. Langenbecks Arch Surg. 2005;390(6):517-22.
  27. Suzuki K, Muto T, Sawada T. Prevention of local recurrence by extended lymphadenectomy for rectal cancer. Surg Today. 1995;25(9):795-801.
  28. Kobayashi H, Mochizuki H, Kato T, et al. Outcomes of surgery alone for lower rectal cancer with and without pelvic sidewall dissection. Dis Colon Rectum. 2009;52(4):567-76.
  29. Hasdemir O, Col C, Yalcin E, Tunc G, Bilgen K, Kucukpinar T. Local recurrence and survival rates after extended systematic lymph-node dissection for surgical treatment of rectal cancer. Hepatogastroenterology. 2005;52(62):455-9.
  30. Koyama Y, Moriya Y, Hojo K. Effects of extended systematic lymphadenectomy for adenocarcinoma of the rectum--significant improvement of survival rate and decrease of local recurrence. Jpn J Clin Oncol. 1984;14(4):623-32.
  31. Col C, Hasdemir O, Yalcin E, Yandakci K, Tunc G, Kucukpinar T. Sexual dysfunction after curative radical resection of rectal cancer in men: the role of extended systematic lymph-node dissection. Med Sci Monit. 2006;12(2):CR70-4.
  32. Kyo K, Sameshima S, Takahashi M, Furugori T, Sawada T. Impact of autonomic nerve preservation and lateral node dissection on male urogenital function after total mesorectal excision for lower rectal cancer. World J Surg. 2006;30(6):1014-9.
  33. Kim TH, Jeong SY, Choi DH, et al. Lateral lymph node metastasis is a major cause of locoregional recurrence in rectal cancer treated with preoperative chemoradiotherapy and curative resection. Ann Surg Oncol. 2008;15(3):729-37.
  34. Kim DJ, Chung JJ, Yu JS, Cho ES, Kim JH. Evaluation of lateral pelvic nodes in patients with advanced rectal cancer. AJR Am J Roentgenol. 2014;202(6):1245-55.
  35. Akiyoshi T, Matsueda K, Hiratsuka M, et al. Indications for Lateral Pelvic Lymph Node Dissection Based on Magnetic Resonance Imaging Before and After Preoperative Chemoradiotherapy in Patients with Advanced Low-Rectal Cancer. Ann Surg Oncol. 2015;22 Suppl 3:S614-20.
  36. Wei M, Wu Q, Fan C, et al. Lateral pelvic lymph node dissection after neoadjuvant chemo-radiation for preoperative enlarged lateral nodes in advanced low rectal cancer: study protocol for a randomized controlled trial. Trials. 2016;17(1):561.
  37. Syk E, Torkzad MR, Blomqvist L, Ljungqvist O, Glimelius B. Radiological findings do not support lateral residual tumour as a major cause of local recurrence of rectal cancer. Br J Surg. 2006;93(1):113-9.
  38. Nagasaki T, Akiyoshi T, Fujimoto Y, et al. Preoperative Chemoradiotherapy Might Improve the Prognosis of Patients with Locally Advanced Low Rectal Cancer and Lateral Pelvic Lymph Node Metastases. World J Surg. 2017;41(3):876-83.
  39. Kim MJ, Kim TH, Kim DY, et al. Can chemoradiation allow for omission of lateral pelvic node dissection for locally advanced rectal cancer? J Surg Oncol. 2015;111(4):459-64.
  40. Sato H, Maeda K, Maruta M, Masumori K, Koide Y. Who can get the beneficial effect from lateral lymph node dissection for Dukes C rectal carcinoma below the peritoneal reflection? Dis Colon Rectum. 2006;49(10 Suppl):S3-12.
  41. Ueno H, Mochizuki H, Hashiguchi Y, et al. Potential prognostic benefit of lateral pelvic node dissection for rectal cancer located below the peritoneal reflection. Ann Surg. 2007;245(1):80-7.
  42. Weiser MR. AJCC 8th Edition: Colorectal Cancer. Ann Surg Oncol. 2018;25(6):1454-5.
  43. Group MS, Shihab OC, Taylor F, et al. Relevance of magnetic resonance imaging-detected pelvic sidewall lymph node involvement in rectal cancer. Br J Surg. 2011;98(12):1798-804.
  44. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603-5.
  45. Higgins JP, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. 2011;343:d5928.