Edited by: Assia Konsoulova, Complex Oncological Center (Burgas), Bulgaria
Reviewed by: Pengpeng Qu, Tianjin Central Hospital for Gynecology and Obstetrics, China; Takuro Ariga, University of the Ryukyus Hospital, Japan
*Correspondence: Hui Wang,
This article was submitted to Women’s Cancer, a section of the journal Frontiers in Oncology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
This study aimed to assess the effect of neoadjuvant chemotherapy (NACT) on the rate of lymph node metastasis (LNM) in FIGO stage IB1-IIB cervical cancer patients and compare the LNM between NACT plus surgery and surgery only.
We identified 34 eligible studies in PubMed, Web of Science, Cochrane Library, and EMBASE from inception to July 27, 2019. Data analyses were performed by Stata (version 13) and Revman (version 5.3).
In these 34 included studies, the pooled incidence of LNM was estimated as 23% (95% CI, 0.20-0.26; I2 = 79.6%,
Among cervical cancer in stage IB1-IIB, the preoperative NACT plus radical surgery resulted in a 23% probability of LNM, which was lower than those receiving radical surgery only. In stage IIA and IIB, the effect of NACT to reduce LNM was more obvious.
Despite the development of comprehensive treatment technology, the cases of cervical cancer have increased from 528,000 in 2012 to 570,000 in 2018, and the deaths have increased from 266,000 in 2012 to 311,000 in 2018 (
Since neoadjuvant chemotherapy was proposed and applied to the treatment of cervical cancer in the 1980s, more and more studies have focused on NACT (
Actually, there existed many risk factors affecting the prognosis of cervical cancer, among which lymph node metastasis (LNM) was one of the most important high risk factors (
Therefore, the aim of this study was to conduct a meta-analysis of the literature on NACT followed by radical surgery and to evaluate the effect of NACT on LNM in FIGO stage IB1-IIB cervical cancer.
This study was conducted on the basis of the preferred reporting items for systematic reviews and meta-analysis (PRISMA) statement and registered in Prospero (PROSPERO CRD42018117658) (
The eligible studies must meet the inclusion criteria as follows: (i) the patient was pathologically diagnosed as stage IB1-IIB cervical cancer; (ii) NACT was platinum based; (iii) the surgery was extensive RH; (iv) the study provided complete data, especially LN status.
Studies were excluded if they met any of these criteria: (i) The studies reported patients with other malignant diseases; (ii) the studies included patients receiving other treatments in addition to NACT and radical surgery; (iii) non-English literature; and (iv) reviews, meta-analysis, case reports, conference articles, and articles without clear data.
Two independent researchers (BC and LM) filtered all publications. Disagreements were determined by group discussion with a third researcher (HW).
Data were extracted by two independent researchers (BC and LM) and filled in standardized data-collection forms (
The quality of each identified study was assessed according to a modified version based on the Cochrane Collaboration’s risk of bias tool (
Meta analyses were conducted through Stata (version 13) and Revman (version 5.3). We estimated the LNM rate with Freeman-Turkey double arcsine transformation because there were a large proportion of data that were close to the margins of the possible interval (0% or 100%) (
Initially, we identified 3645 eligible studies. After screening, 34 studies were finally included in this meta-analysis, consisting of 3813 patients (
Flow diagram of literature selection process for the meta-analysis.
After receiving NACT in patients with stage IB1-IIB cervical cancer, the pooled incidence was 23% (95% CI: 0.20-0.26) with significant heterogeneity (I2 = 79.6%,
Forest plot of the estimated LNM rate in cervical cancer patients receiving NACT plus RH.
After removing the literature with high risk of bias, the forest plot of estimated LNM rate.
We took the following factors into consideration for the subgroup analysis: study types, the year of publication, patients’ continents, histological types, and FIGO stage (
Subgroup analysis of the 34 studies.
Subgroups | No. of trials | Rates1 | 95% CI | No. of patients | Heterogeneity |
---|---|---|---|---|---|
study types | |||||
retrospective study | 23 | 24% | (0.21-0-28) | 887 | I2 = 76.6%, |
prospective study | 3 | 28% | (0.15-0.41) | 313 | I2 = 85.5%, |
randomized controlled study | 8 | 19% | (0.13-0.25) | 2613 | I2 = 81.2%, |
Total | 34 | 23% | (0.20-0.26) | 3813 | I2 = 79.6%, |
year of publication | |||||
1995-1999 | 3 | 12% | (0.05-0.18) | 210 | I2 = 45.4%, |
2000-2004 | 2 | 19% | (0.12-0.27) | 101 | I2 = 0.0%, |
2005-2009 | 4 | 31% | (0.26-0.36) | 314 | I2 = 0.0%, |
2010-2014 | 18 | 22% | (0.19-0.26) | 1581 | I2 = 69.7%, |
2015-present | 7 | 28% | (0.20-0.35) | 1607 | I2 = 90.4%, |
Total | 34 | 23% | (0.20-0.26) | 3813 | I2 = 79.6%, |
continents of patients2 | |||||
Europe | 8 | 24% | (0.19-0.29) | 839 | I2 = 60.7%, |
Asia | 23 | 24% | (0.20-0.27) | 2711 | I2 = 79.9%, |
South America and North America | 2 | 8% | (0.03-0.13) | 118 | I2 = 0.0%, |
Total | 33 | 23% | (0.20-0.26) | 3668 | I2 = 79%, |
histological types | |||||
squamous cervical cancer | 5 | 18% | (0.10-0.25) | 878 | I2 = 79.3%, |
non-squamous cervical cancer | 3 | 16% | (0.03-0.29) | 112 | I2 = 69.7%, |
Total | 6 | 17% | (0.11-0.23) | 990 | I2 = 74.7%, |
FIGO stage3 | |||||
IB | 7 | 13% | (0.10-0.15) | 699 | I2 = 5.5%, |
IIA | 3 | 23% | (0.18-0.28) | 308 | I2 = 0.0%, |
IIB | 3 | 27% | (0.20-0.33) | 184 | I2 = 0.0%, |
Total | 7 | 17% | (0.13-0.21) | 1191 | I2 = 62.8%, |
1The relevant positive LN rates in subgroup or in total.
2The study by Eddy GL et al. was not included in the subgroup analysis of continents, because this study consisted of 62% were white,13% were black,19% were Hispanic,4% were Asian/Pacific islander and 1% were Other.
3The patients’ stages were determined according to the FIGO stage criteria at the time of diagnosis.
The sensitivity analyses were performed by excluding one study at a time and did not noticeably affect the results (
From the 34 included studies, we identified six two-arm RCTs including 1016 patients (
Characteristics of the 6 RCTs in the meta-analysis.
Author | Country | Continent of patients | Year of publication | Nos. of total patients | Age (medium, range) or (mean, SD) | Patients in NACT+surgery group | Patients in surgery group | ||
---|---|---|---|---|---|---|---|---|---|
Nos. of LNM | Nos. of Total | Nos. of LNM | Nos. of Total | ||||||
Cai et al. ( |
China | Asia | 2010 | 106 | 45.6 ± 22.4 | 5 | 52 | 16 | 54 |
Chen et al. ( |
China | Asia | 2008 | 142 | 44(25-74) | 18 | 72 | 30 | 70 |
Eddy et al. ( |
USA | 8 | 2007 | 288 | ≤30:14%,31-40:31%,41-50:32%,51-60:15%,≥61:7% | 47 | 145 | 56 | 143 |
Sardi et al. ( |
Argentina | South America and North America | 1997 | 201 | 38.5(24-63) | 8 | 98 | 32 | 103 |
Wen et al. ( |
China | Asia | 2012 | 60 | 44.53 ± 9.10 | 5 | 28 | 11 | 32 |
Yang et al. ( |
China | Asia | 2016 | 219 | 47(23-66) | 22 | 109 | 25 | 110 |
Forest plots for the LNM rate in the comparison between NACT plus RH and RH.
LNM is one of the most important risk factors affecting the prognosis of cervical cancer, so we explored the impact of NACT on LNM rates. In FIGO stage IB1-IIB cervical cancer, the positive LN rate of NACT followed by RH was 23% (95% CI: 0.20-0.26). The NACT plus RH brings more benefits in reducing LNM among the stage IIA and IIB patients. Through the comparison between NACT plus RH and RH, we further confirmed that the incidence of nodal metastasis was lower in the NACT plus RH group (RR=0.57, 95% CI: 0.39-0.83, I2 = 60.5%,
Based on this research, LNM occurs in 23% of patients with stage IB1-IIB cervical cancer after NACT plus RH (95% CI: 0.20-0.26, I2 = 79.6%,
The incidence of positive LNs from 1995 to 2019 show an upward trend through the subgroup analysis by the year of publication. The reason for the difference over time was speculated to be stage migration and development of diagnostic techniques. With the improvement of imaging technology and surgical methods, the accuracy and sensitivity of finding positive LNs were greatly improved. Similarly, a study based on squamous cell carcinoma of the anus noted an increase in observed positive LN rate over time, which was summarized as the Will Rogers phenomenon (
If we divide patients into three subgroups as IB, IIA, and IIB, the between-study heterogeneity is not significant. The estimated incidence was 13% (95% CI: 0.10-0.15) in stage IB, 23% (95% CI: 0.18-0.28) in stage IIA, and 27% (95% CI: 0.20- 0.33) in stage IIB. According to the previous literature including patients receiving surgery only, the nodal metastasis rates of the FIGO stage IB, IIA, and IIB were 11.5-22%, 26.7-33%, and 39.2%-63%, respectively (
When we consider the LNM rate in patients undergoing preoperative chemotherapy plus RH and RH only, the pooled RR of 0.57 (95% CI: 0.39-0.83, I2 = 60.5%,
However, some studies propose different perspectives. Gong L et al. provided a phase III trial based on locally advanced cervical cancer (LACC) patients to detect the incidence of nodal metastasis. Their analysis showed that the proportion of positive-node patients in the NACT plus RH group (27.8%) was similar to that in the RH group (28.8%) (
Recently, an RCT by Gupta S et al. caught great attention (
There also existed several limitations. First, to limit heterogeneity, subgroup analysis was performed based on six terms. Some subgroup analysis might be limited to only 3 studies with small sample sizes. Second, several important factors, such as depth of invasion and lymph-vascular space invasion (LVSI), can significantly affect the prognosis of patients with cervical cancer, but these included articles did not provide the incidence of LNM in patients grouped by depth of invasion or LVSI. Therefore, the subgroup analysis to explore the impact of infiltration depth and LVSI was not performed. Third, some studies used different evaluation methods. Four only provided the status of LNs by preoperative MRI. Another 30 studies determined the status of both para-aortic LNs and pelvis LNs based on surgical findings. When only including the studies reporting LNM based on surgical findings, the analysis demonstrated similar results (
The rate of positive LN in stage IB1-IIB cervical cancer patients was 23% after receiving NACT plus RH. The LNM rate in the NACT plus RH group was lower than the RH group. NACT plus RH showed the more obvious effect of eliminating positive LNs in patients with stage IIA and IIB compared with previously reported surgical patients. Therefore, NACT can be considered as a valuable and reasonable treatment option in patients with stage IB1-IIB cervical cancer.
All datasets presented in this study are included in the article/
BC, LW, CR, HS, WD, DZ, LM, and HW contributed to the conception and design of the study. BC, LM, and HW organized the databases and provided methodological supports. BC and LM performed the statistical analysis. BC wrote the draft of the manuscript. LM and HW contributed to the supervision of the study. All authors critically reviewed and revised the manuscript for important intellectual content and approved the final version of the manuscript before submission. All authors contributed to the article and approved the submitted version.
This work was supported by a key program from the National Natural Science Foundation of China (No.81830074).
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at: