J Cancer Res Clin Oncol (2013) 139:1887–1898

DOI 10.1007/s00432-013-1509-y

ORIGINAL PAPER

Responsiveness of neoadjuvant chemotherapy before surgery

predicts favorable prognosis for cervical cancer patients:
a meta-analysis
Qing Ye • Hong-Xin Yuan • Hong-Lin Chen

Received: 28 August 2013 / Accepted: 30 August 2013 / Published online: 11 September 2013
Ó Springer-Verlag Berlin Heidelberg 2013

Abstract were 6.179 (95 % CI 3.390–11.264), 9.155 (95 % CI

Background Neoadjuvant chemotherapy (NAC) before 5.759–14.555), 8.431 (95 % CI 5.667–12.543), and 5.785
surgery has already shown the therapy effectiveness in (95 % CI 4.124–8.115) at 1-, 2-, 3-, and 5-year follow-up,
patients with cervical cancer. The present meta-analysis respectively. No obvious statistical heterogeneity was
was conducted to determine whether the response to NAC detected. Funnel plots and Egger’s tests did not reveal
predicts for prognosis. publication bias. Sensitivity analysis showed the results of
Methods Systematic computerized searches of the Pub- meta-analysis were robust.
Med and Web of Knowledge were performed. Prognosis Conclusion This meta-analysis confirms that response to
outcomes included progression-free survival (PFS), and NAC is an indicator for PFS and OS, and suggests that
overall survival (OS). The pooled odd ratio (OR) was patients-achieving response of NAC before surgery pre-
estimated by using fixed-effect model or random-effect dicts favorable prognosis for cervical cancer patients.
model according to heterogeneity between studies.
Results Eighteen studies with 1,785 patients were inclu- Keywords Cervical cancer
Neoadjuvant
ded. Cisplatin-based NAC treatments were most commonly chemotherapy
Anti-tumor response
Prognosis

used. The clinical response rate ranged from 48.4 to Meta-analysis
93.0 %, and the pathological response rate ranged from
27.6 to 30.6 %. The pooled ORs estimating the association
of PFS with NAC response were 5.707 (95 % CI Introduction
3.564–9.137), 6.798 (95 % CI 4.716–9.799), 6.327 (95 %
CI 4.398–9.102), and 5.214 (95 % CI 3.748–7.253) at 1-, Cervical cancer is the third most commonly diagnosed
2-, 3-, and 5-year follow-up, respectively, and the pooled cancer and the fourth leading cause of cancer death in
ORs estimating the association of OS with NAC response females worldwide, accounting for 9 % (529,800) of the
total new cancer cases and 8 % (275,100) of the total
cancer deaths among females in 2008. More than 85 % of
Q. Ye (&)
these cases and deaths occur in developing countries (Je-
Department of Obstetrics and Gynecology, Affiliated Hospital
of Nantong University, Xi Si Road 20#, Nantong City 226001, mal et al. 2011). Standard treatment strategies recommend
Jiangsu Province, People’s Republic of China by the National Comprehensive Cancer Network (NCCN)
e-mail: [email protected] (2013) and the International Federation of Gynecology and
Obstetrics (FIGO) include surgical therapy, adjuvant radi-
H.-X. Yuan
Department of Interventional Radiology, Affiliated Hospital ation, adjuvant chemotherapy, and systemic therapy (Wi-
of Nantong University, Xi Si Road 20#, Nantong City, Jiangsu ebe et al. 2012).
Province, People’s Republic of China Neoadjuvant chemotherapy (NAC), short-course che-
motherapy administered before surgical treatment, has
H.-L. Chen
Nantong University, Qixiu Road 19#, Nantong City, Jiangsu become an alternative approach for the treatment of cer-
Province, People’s Republic of China vical cancer since the 1990s. NAC can reduce tumor bulk

123
1888 J Cancer Res Clin Oncol (2013) 139:1887–1898

and positivity of lymph nodes and facilitate subsequent remission [title/abstract]) AND (prong* [title/abstract] OR
radical trachelectomy. NAC also increase radio-sensitivity survival [title/abstract]). In Web of Science citation data-
and decrease the hypoxic cell fraction. Moreover, NAC base, we select Science Citation Index Expanded (SCI-
treats the micrometastatic disease, preventing a significant EXPANDED) database and Conference Proceedings Cita-
proportion of relapses (González-Martı́n et al. 2008). More tion Index Science (CPCI-S) database. The search detail
and more evidence showed NAC before surgery can used as follows: TS = (cervical cancer) AND
improve the long-term prognosis of cervical cancer. A (TS = neoadjuvant chemotherapy OR TS = preoperative
meta-analysis of individual patient data from 21 random- chemotherapy) AND (TS = response OR TS = remission)
ised trials indicated a highly significant reduction in the AND (TS = prong* OR TS = survival). We also looked at
risk of death with NAC (2003) for locally advanced cer- posters from the annual meetings of the American Society
vical cancer [hazard ratio (HR) 0.65; 95 % confidence of Clinical Oncology (ASCO) (http://www.asco.org/
interval (CI) 0.53–0.80; P = 0.0004]. A recently updated ASCOv2/Meetings) and the European Society for Medi-
Cochrane systematic review included 6 randomized con- cal Oncology (ESMO) (http://www.esmo.org/) in the past
trolled clinical trials with 1,078 women and found that both 10 years. The references of all relevant studies were also
overall survival (HR 0.77; 95 % CI 0.62–0.96; P = 0.02) manually reviewed to supplement our searches. Only
and progression-free survival (PFS) (HR 0.75; 95 % CI studies published in English were included.
0.61–0.93; P = 0.008) were significantly improved with
NAC (Rydzewska et al. 2012). Study selection
Some studies also reported the achievement of anti-
tumor response was the strongest independent prognostic The relevant clinical studies were manually selected care-
variable for patients with cervical cancer treated with NAC fully based on the following criteria: (1) NAC had to be
before surgery. In the SNAP01 study, Buda et al. (2005) delivered before surgery for cervical cancer. (2) Anti-tumor
reported the average death rates were higher in the group response after NAC was evaluated using Response Eval-
that did not achieve optimal pathologic response (HR 5.88; uation Criteria in Solid Tumors guidelines or other criteria.
95 % CI 2.50–13.84; P \ 0.0001). In a retrospective study, The anti-tumor responses included complete response
Gadducci et al. (2010a) also found the pathological (CR), partial response (PR), stable disease (SD), and pro-
response was independent prognostic variables of recur- gressive disease (PD). The response rate was defined as
rence-free survival (HR 7.999; 95 % CI 1.916–33.394) and (CR ? PR)/(CR ? PR ? SD ? PD) * 100 %. (3) Prog-
overall survival (HR 6.007; 95 % CI 1.426–25.307) on nosis outcomes, such as progression-free survival (PFS),
multivariate analysis. Xiong et al. (2011) also reported disease-free survival (DFS), tumor-free survival (TFS),
response to NAC was the only factor associated with sur- disease-specific survival (DSS), and overall survival (OS),
vival (P = 0.036) on Cox proportional hazards model. were reported in the studies; and the outcomes were also
However, this conclusion all came from small sample compared between NAC responders and nonresponders.
studies. When the same patient population was used in several
The aim of current meta-analysis was to pool the papers, only the most recent study was included in the
existing data to determine if anti-tumor responses predict meta-analysis.
for outcomes in cervical cancer patients who have received
NAC before surgery. Data extraction

The following data were abstracted onto standardized

Methods forms: (1) basic information from papers, such as first
author, publication year, country; (2) characteristics of
Database and literature search patients, such as number, age, TNM stage, and histology;
(3) information of NAC, such as year, NAC program, and
Systematic computerized searches of the PubMed and Web response rate; (4) duration of follow-up; and (5) survival
of Knowledge were performed from their inception to July outcomes, such as 1-, 2-, 3-, and 5-year PFS/OS rate in
25, 2013. The following search terms were used: cervical NAC responders and nonresponders. If these data were not
cancer, neoadjuvant chemotherapy, preoperative chemo- directly available in the paper, we used Engauge Digitizer
therapy, response, remission, prognosis, and survival. The (version 4.1, free software downloaded from http://
search detail in PubMed was ‘‘uterine cervical neo- digitizer.sourceforge.net/) to read the survival curves
plasms’’[MeSH Terms] AND (‘‘neoadjuvant chemother- extracted from the included studies. DFS and TFS were
apy’’ [title/abstract] OR ‘‘preoperative chemotherapy’’ recognized as PFS, and DSS was recognized as OS. Data
[title/abstract]) AND (response [title/abstract] OR extraction was carried out independently by two reviewers.

123
J Cancer Res Clin Oncol (2013) 139:1887–1898 1889

Disagreements were resolved by discussion between the

Records identified through Pubmed Additional records
two.
searching (n=132) identified through
Records identified through Web of meeting abstract
Qualitative assessment
Science searching (n=127) (n =5)

The quality of studies was assessed according to the

Newcastle–Ottawa Scale (NOS) (Wells et al. 2013). The Records after duplicates removed (n = 187)
NOS contains eight items, categorized into three dimen-
sions including Selection (4), Comparability (1), and
Exposure (3). A high-quality study can be awarded a Records excluded by title and abstract review
Records Reviews (n=11)
maximum of one star for each numbered item within the
screened Case reports and case series (n=23)
Selection and Exposure categories. A maximum of two
(n =187) Non-English articles (n=18)
stars can be given for Comparability. The NOS ranges from
No available data (n=73)
zero up to nine stars.

Statistical analysis Full-text articles assessed Full-text articles excluded

(n=62) No available data (n=44)
The pooled ORs with 95 % CI for survival outcomes
compared with NAC responders and nonresponders were
calculated for each measurement. Each publication was Studies included in quantitative synthesis (meta-analysis) (n=18)

weighted according to the size of the sample. Statistical

heterogeneity was explored by v2 and inconsistency (I2) Fig. 1 Flow diagram showing selection of studies
statistics; P \ 0.05 for v2 or I2 value of 50 % or more
represented substantial heterogeneity (Higgins et al. 2003). India, Taiwan, and Australia, respectively. Thirteen studies
In the absence of significant heterogeneity, studies were included local advanced cancer for TNM Stage I–II, and
pooled using a fixed-effect model. If heterogeneity was the rest of 5 included cancers for TNM Stage I–III or I–IV.
observed, a random-effects model was used. Overall effects Two studies only included squamous cell carcinoma, a
were determined using the Z test. Visual inspection of a study only included non-squamous cell carcinoma, and the
funnel plot, the Egger’s regression test, and Begg’s other 15 studies included squamous cell carcinoma and
adjusted rank correlation test were performed to assess adenocarcinoma. NAC agents varied greatly, but most of
publication bias. Sensitivity analyses were conducted by them were cisplatin-based. The clinical response rate pro-
removing the largest sample study. All statistical analyses vided by 15 studies ranged from 48.4 to 93.0 %, and the
were performed with Stata software, version 12.0 (Stata pathological response rate provided by 3 studies ranged
Corp, College Station, Texas). Two-sided P \ 0.050 was from 27.6 to 30.6 %. The quality rating of the included
considered statistically significant. studies ranged from six to eight stars on the scale of nine.
Table 1 shows the main characteristics of the 18 included
studies, and Table 2 shows the survival outcomes com-
Results pared NAC responders with nonresponders.

Eligible studies Main results of progression-free survival

We identified 18 studies (Xiong et al. 2011; Shoji et al. A total of 10 studies provided 1-year PFS in NAC
2013; Li et al. 2013; Robova et al. 2013; Hu et al. 2012; responders and nonresponders. The 1-year PFS ranged
Huang et al. 2011; Park et al. 2009, 2011; Gadducci et al. from 72.2–100.0 % in NAC responders and 40.0–95.2 % in
2010b; Mori et al. 2010; Kumar et al. 2009; Choi et al. nonresponders. No heterogeneity was found between the
2008; Cai et al. 2006; Selvaggi et al. 2006; Fuso et al. studies (I2 = 0.0 %, P = 0.513). The pooled OR was
2005; Chen et al. 2002, 2008; MacLeod et al. 2001) that 5.707 (95 % CI 3.564–9.137; Z = 7.25, P = 0.000) by
met our inclusion criteria for meta-analysis. The detailed fixed-effect model (Fig. 2). The Begg’s test (Z = 0.00,
steps of our literature search are shown in Fig. 1. A total of P = 1.000) and the Egger’s test (t = 1.08, P = 0.310)
1,785 patients were used in the pooled analyses. Of the 18 suggested there was no significant publication bias. Fig-
studies, sample sizes ranged from 23 to 707. Six of these ure 3 shows the funnel plot of publication bias.
studies were conducted in China, 3 in Italy, 3 in Korea, 2 in A total of 11 studies provided 2-year PFS in NAC
Japan, and the rest of four conducted in Czech Republic, responders and nonresponders. The 2-year PFS ranged

123
 Table 1 Characteristics of the included studies
1890

Study (year) County No. of Age TNM Histology Year of NAC Response rate Duration of follow- NOS
patients stage NAC up

123
Shoji et al. Japan 23 50 (32–63) IB2– Non-sq 2002–2011 Paclitaxel ? carboplatin (TC) or docetaxel ? carboplatin (DC), 78.3 % (18/ Up to 90 months 7
(2013) IIB maximum 3 courses 23)
Li et al. China 154 41.6 ± 7.92 IB2– Sq and 2000–2011 Docetaxel ? paclitaxel ? nedaplatin (TP) or 72.1 % (111/ 1–12 years 8
(2013) IIA2 Ad bleomycin ? vincristine ? cisplatin (BVP), 2–3 cycles 154)
Robova Czech 151 45.7 IB Sq and 1998–2009 Cisplatin ? ifosfamide or cisplatin ? doxorubicine, 3 cycles 78.8 % (119/ 29–154 months 8
et al. Republic (20–70) Ad 151)
(2013)
Hu et al. China 707 NR IB1– Sq and 2002–2008 Cisplatin-based chemotherapy, 1–2 courses 79.3 % (541/ Up to 120 months 8
(2012) IIB Ad 682)
Xiong et al. China 60 49 (28–61) IB2– Sq and 2004–2009 Irinotecan ? cisplatin, 2 cycles 65 % (39/60) Up to 70 months 8
(2011) IIB Ad
Huang et al. China 52 46 (30–63) IB2– Sq and 2007–2009 Docetaxel ? cisplatin, 2 cycles 86.5 % (45/ Up to 40 months 7
(2011) IIB Ad 52)
Park et al. Republic 43 50 (30–78) IIB Sq and 1997–2007 Cisplatin ? etoposide, 3 courses 65.1 % (28/ Up to 125 months 8
(2011) of Korea Ad 43)
Gadducci Italy 43 43 (27–69) IB2– Sq and 1999–2009 Ifosfamide ? paclitaxel ? cisplatin, or paclitaxel ? cisplatin, 27.9 % (12/ 5 years 8
et al. IIB Ad 3 cycles 43)
(2010a, b) pathological
Mori et al. Japan 30 60 (29–74) IB2– Sq and 2002–2006 Paclitaxel ?carboplatin, 6 cycles 86.7 % (26/ 55.6 7
(2010) IIIB Ad 30) (26–83 months)
Park et al. South 43 52 (36–72) IB2– Sq and 2000–2002 Paclitaxel ? cisplatin, 3 cycles 93.0 % (40/ 5 years 7
(2009) Korea IIB Ad 43)
Kumar et al. India 49 50 (35–70) IB2– Sq and 2000–2006 Paclitaxel ? ifosfamide ? cisplatin, 2 or 3 courses 30.6 % (15/ 14 7
(2009) IVa Ad 49) (0.5–62.9 months)
pathological
Choi et al. Korea 29 47 (33–70) IB2– Sq NR Vincristine ? mitomycin-C ? cisplatin, 3 cycles 27.6 % (8/29) 48 (3–105 months) 7
(2008) IIB pathological
Chen et al. China 72 44 (25–74) IB2– Sq and 1999–2004 Cisplatin ? mitomycin-C ? 5-fluorouracil, 2 cycles 69.4 % (50/ 4–9 years 7
(2008) IIB Ad 72)
Cai et al. China 52 45.6 ± 22.4 IB Sq and 1999–2001 Cisplatin ? 5-Fu, 2 courses 84.6 % (44/ 62 (18–95 months) 7
(2006) Ad 52)
Selvaggi Italy 67 51 (27–75) IB2– Sq and 1992–2003 Cisplatin and vinorelbine; cisplatin and irinotecan; cisplatin and 91.0 % (61/ 93 (4–130 months) 7
et al. III Ad ifosphamide; and bleomycin, cisplatin, and methotrexate, 3 67)
(2006) (2–6) cycles
Fuso et al. Italy 73 Mean 49.3 IB2– Sq 1997–2004 Platinum, 3 cycles 64.4 % (47/ Up to 140 months 7
(2005) III 73)
Chen et al. Taiwan 31 49.3 ± 8.4 IB2– Sq and 1991–1999 Cisplatin ? vincristine ? bleomycin, 3 courses 48.4 % (15/ Up to 120 months 6
(2002) IIA Ad 31)
MacLeod Australia 106 NR IB–IV Sq and 1982–1995 Cisplatin ? vinblastine ? bleomycin; or Epirubicin ? cisplatin; 58.5 % (62/ Median 6.9 years 6
et al. Ad or Etoposide ? vincristine ? cyclophosphamide ? adriamycin, 106)
(2001) 3 courses

NAC neoadjuvant chemotherapy, Sq squamous cell carcinoma, Ad adenocarcinoma, NOS Newcastle–Ottawa Scale
J Cancer Res Clin Oncol (2013) 139:1887–1898
J Cancer Res Clin Oncol (2013) 139:1887–1898 1891

Table 2 Survival outcomes in NAC responders (CR ? PR) and nonresponders (SD ? PD)
Study (year) Outcomes NAC responders Nonresponders

Shoji et al. (2013) 1-year PFS rate 72.2 % (13/18) 40 % (2/5)

1-year OS rate 88.9 % (16/18) 80 % (4/5)
2-year PFS rate 55.6 % (10/18) 40 % (2/5)
2-year OS rate 66.6 % (12/18) 60 % (3/5)
3-year PFS rate 38.9 % (7/18) 20 % (1/5)
3-year OS rate 50 % (9/18) 20 % (1/5)
5-year PFS rate 27.8 % (5/18) 0 % (0/5)
5-year OS rate 33.3 % (6/18) 0 % (0/5)
Li et al. (2013) 1-year PFS rate 98 % (109/111) 93 % (40/43)
1-year OS rate 100 % (111/111) 93 % (40/43)
2-year PFS rate 96 % (107/111) 86 % (37/43)
2-year OS rate 100 % (111/111) 86 % (37/43)
3-year PFS rate 96 % (107/111) 86 % (37/43)
3-year OS rate 100 % (111/111) 86 % (37/43)
5-year PFS rate 94 % (104/111) 86 % (37/43)
5-year OS rate 96 % (107/111) 86 % (37/43)
Robova et al. (2013) 1-year PFS rate 98 % (117/119) 80 % (26/32)
1-year DSS rate 100 % (119/119) 95 % (30/32)
2-year PFS rate 93 % (110/119) 70 % (22/32)
2-year DSS rate 96 % (114/119) 71 % (23/32)
3-year PFS rate 90 % (107/119) 65 % (21/32)
3-year DSS rate 92 % (109/119) 70 % (22/32)
5-year PFS rate 86 % (102/119) 65 % (21/32)
5-year DSS rate 88 % (105/119) 65 % (21/32)
Hu et al. (2012) 1-year DFS rate 97 % (524/541) 87 % (123/141)
1-year OS rate 98 % (532/541) 93 % (131/141)
2-year DFS rate 96 % (519/541) 72 % (101/141)
2-year OS rate 98 % (530/541) 81 % (114/141)
3-year DFS rate 94 % (509/541) 68 % (96/141)
3-year OS rate 96 % (519/541) 70 % (99/141)
5-year DFS rate 92 % (498/541) 68 % (96/141)
5-year OS rate 95 % (514/541) 68 % (96/141)
Xiong et al. (2011) 1-year OS rate 97 % (38/39) 90 % (19/21)
2-year DFS rate 88 % (34/39) 51 % (11/21)
2-year OS rate 94 % (37/39) 51 % (11/21)
3-year OS rate 94 % (37/39) 42 % (9/21)
Huang et al. (2011) 1-year PFS rate 95.5 % (43/45) 57.1 % (4/7)
1-year OS rate 100 % (45/45) 85.7 % (6/7)
2-year PFS rate 90.3 % (41/45) 57.1 % (4/7)
2-year OS rate 100 % (45/45) 68.6 % (5/7)
Park et al. (2011) 1-year DFS rate 89.0 % (25/28) 49.0 % (7/15)
2-year DFS rate 82.0 % (23/28) 39.0 % (6/15)
3-year DFS rate 70.0 % (20/28) 31.0 % (5/15)
5-year DFS rate 66.4 % (19/28) 24.0 % (4/15)
Gadducci et al. (2010a, b) 5-year DFS rate 75 % (9/12) 19 % (6/31)
5-year OS rate 83 % (10/12) 37 % (11/31)

123
1892 J Cancer Res Clin Oncol (2013) 139:1887–1898

Table 2 continued
Study (year) Outcomes NAC responders Nonresponders

Mori et al. (2010) 1-year OS rate 96.1 % (25/26) 75 % (3/4)

2-year OS rate 92.3 % (24/26) 75 % (3/4)
3-year OS rate 92.3 % (24/26) 75 % (3/4)
5-year OS rate 88.4 % (23/26) 50 % (2/4)
Park et al. (2009) 5-year OS rate 91.2 % (31/34) 66.7 % (2/3)
Kumar et al. (2009) 1-year DFS rate 100.0 % (15/15) 72 % (24/34)
2-year DFS rate 100.0 % (15/15) 52 % (18/34)
3-year DFS rate 100.0 % (15/15) 39 % (13/34)
5-year DFS rate 100.0 % (15/15) 39 % (13/34)
Choi et al. (2008) 1-year PFS rate 100 % (8/8) 95 % (20/21)
2-year PFS rate 100 % (8/8) 71 % (15/21)
3-year PFS rate 100 % (8/8) 65 % (14/21)
5-year PFS rate 100 % (8/8) 65 % (14/21)
Chen et al. (2008) 1-year TFS rate 96 % (48/50) 64 % (14/22)
2-year TFS rate 91 % (46/50) 50 % (11/22)
3-year TFS rate 91 % (46/50) 50 % (11/22)
5-year TFS rate 83 % (42/50) 42 % (9/22)
Cai et al. (2006) 2-year OS rate 100 % (44/44) 87.5 % (7/8)
3-year OS rate 98 % (43/44) 62.5 % (5/8)
5-year OS rate 93 % (41/44) 50 % (4/8)
Selvaggi et al. (2006) 1-year OS rate 97 % (59/61) 67 % (4/6)
2-year OS rate 84 % (51/61) 17 % (1/6)
3-year OS rate 79 % (48/61) 17 % (1/6)
5-year OS rate 66 % (40/61) 17 % (1/6)
Fuso et al. (2005) 1-year OS rate 100 % (47/47) 72 % (19/26)
2-year OS rate 98 % (46/47) 58 % (15/26)
3-year OS rate 94 % (44/47) 54 % (14/26)
5-year OS rate 94 % (44/47) 46 % (12/26)
Chen et al. (2002) 1-year DFS rate 87 % (13/15) 88 % (14/16)
2-year DFS rate 80 % (12/15) 81 % (13/16)
3-year DFS rate 80 % (12/15) 81 % (13/16)
5-year DFS rate 80 % (12/15) 81 % (13/16)
1-year OS rate 87 % (13/15) 95 % (15/16)
2-year OS rate 80 % (12/15) 95 % (15/16)
3-year OS rate 80 % (12/15) 95 % (15/16)
5-year OS rate 80 % (12/15) 85 % (14/16)
MacLeod et al. (2001) 5-year OS rate 38 % (24/62) 18 % (8/44)
NAC neoadjuvant chemotherapy, CR complete response, PR partial response, SD stable disease, PD progressive disease, DFS disease-free
survival, PFS progression-free survival, TFS tumor-free survival, DSS disease-specific survival, OS overall survival

from 55.6–100.0 % in NAC responders and 40.0–96.0 % in A total of 9 studies provided 3-year PFS in NAC
nonresponders. No heterogeneity was found between the responders and nonresponders. The 3-year PFS ranged
studies (I2 = 0.0 %, P = 0.461). The pooled OR was from 38.9–100.0 % in NAC responders and 20.0–86.0 % in
6.798 (95 % CI 4.716–9.799; Z = 10.27, P = 0.000) by nonresponders. No heterogeneity was found between the
fixed-effect model. The Begg’s test (Z = 0.00, P = 1.000) studies (I2 = 11.5 %, P = 0.339). The pooled OR was
and the Egger’s test (t = 0.28, P = 0.783) suggested there 6.327 (95 % CI 4.398–9.102; Z = 9.94, P = 0.000) by
was no significant publication bias. The forest plot and fixed-effect model. The Begg’s test (z = 0.52, P = 0.602)
funnel plot are showed in ‘‘Appendix’’. and the Egger’s test (t = 0.73, P = 0.490) suggested there

123
J Cancer Res Clin Oncol (2013) 139:1887–1898 1893

Study %
and the Egger’s test (t = 1.42, P = 0.194) suggested there
ID OR (95% CI) Weight was no significant publication bias. The forest plot and
funnel plot are showed in ‘‘Appendix’’.
Shoji T, 2013 3.90 (0.49, 30.76) 6.33
The pooled PFS rates in NAC responders and nonre-
Li R, 2013 4.09 (0.66, 25.37) 7.56
Robova H, 2013 13.50 (2.58, 70.70) 5.01
sponders were listed in the Table 3 and Fig. 4. The results
Hu T, 2012 4.51 (2.26, 9.01) 44.63 showed NAC response is a prognostic indicator for PFS
Huang X, 2011 16.13 (2.05, 126.74) 2.24 in cervical cancer patients treated with NAC before
Park JS, 2011 9.52 (1.98, 45.75) 7.11
surgery.
Kumar JV, 2009 13.29 (0.73, 243.27) 3.50
Choi CH, 2008 1.24 (0.05, 33.69) 4.81
Chen H, 2008 13.71 (2.61, 72.12) 5.66 Main results of overall survival
Chen CA, 2002 0.93 (0.11, 7.59) 13.15
Overall (I-squared = 0.0%, p = 0.513) 5.71 (3.56, 9.14) 100.00
A total of 10 studies provided 1-year OS in NAC
responders and nonresponders. The 1-year OS ranged from
.00411 1 243
86.7–100.0 % in NAC responders and 73.1–93.8 % in
Fig. 2 Pooled OR of 1-year progression-free survival (PFS) com- nonresponders. No heterogeneity was found between the
pared NAC responders with nonresponders studies (I2 = 16.2 %, P = 0.294). The pooled OR was
6.179 (95 % CI 3.390–11.264; Z = 5.94, P = 0.000) by
fixed-effect model. The Begg’s test (z = 1.97, P = 0.049)
Funnel plot with pseudo 95% confidence limits and the Egger’s test (t = 1.40, P = 0.200) suggested there
was no significant publication bias. The forest plot and
0

funnel plot are showed in ‘‘Appendix’’.

A total of 11 studies provided 2-year OS in NAC
.5

responders and nonresponders. The 2-year OS ranged from

66.7–100.0 % in NAC responders and 16.7–93.8 % in
se(logOR)

nonresponders. No heterogeneity was found between the

1

studies (I2 = 42.2 %, P = 0.068). The pooled OR was

9.155 (95 % CI 5.759–14.555; Z = 9.36, P = 0.000) by
1.5

fixed-effect model. The Begg’s test (z = 1.09, P = 0.276)

and the Egger’s test (t = 1.40, P = 0.195) suggested there
was no significant publication bias. The forest plot and
2

0 5 10 15 funnel plot are showed in ‘‘Appendix’’.

OR A total of 10 studies provided 3-year OS in NAC
responders and nonresponders. The 3-year OS ranged from
Fig. 3 The funnel plot of meta-analysis for 1-year progression-free
survival (PFS) compared NAC responders with nonresponders 50.0–100.0 % in NAC responders and 20.0–93.8 % in
nonresponders. No heterogeneity was found between the
was no significant publication bias. The forest plot and studies (I2 = 37.6 %, P = 0.108). The pooled OR was
funnel plot are showed in ‘‘Appendix’’. 8.431 (95 % CI 5.667–12.543; Z = 10.52, P = 0.000) by
A total of 10 studies provided 5-year PFS in NAC fixed-effect model. The Begg’s test (z = 0.89, P = 0.371)
responders and nonresponders. The 5-year PFS ranged and the Egger’s test (t = 1.13, P = 0.290) suggested there
from 27.8–100.0 % in NAC responders and 0.0–86.0 % in was no significant publication bias. The forest plot and
nonresponders. No heterogeneity was found between the funnel plot are showed in ‘‘Appendix’’.
studies (I2 = 24.2 %, P = 0.221). The pooled OR was A total of 12 studies provided 5-year OS in NAC
5.214 (95 % CI 3.748–7.253; Z = 9.80, P = 0.000) by responders and nonresponders. The 5-year OS ranged from
fixed-effect model. The Begg’s test (z = 1.07, P = 0.283) 33.3–96.4 % in NAC responders and 0.0–86.0 % in

Table 3 Pooled progression-free survival (PFS) rate and overall survival (OS) rate in NAC responders and nonresponders
Outcomes Group 1-year 2-year 3-year 5-year

PFS rate (95 % CI) NAC responders 96.3 % (94.9–97.4 %) 93.5 % (91.8–95.0 %) 91.8 % (89.8–93.5 %) 88.9 % (86.7–90.8 %)
Nonresponders 81.5 % (77.0–85.6 %) 67.2 % (62.1–72.0 %) 64.1 % (58.7–69.3 %) 59.2 % (53.9–64.3 %)
OS rate (95 % CI) NAC responders 98.4 % (97.5–99.1 %) 96.2 % (94.9–97.3 %) 93.6 % (92.0–95.1 %) 87.7 % (85.6–89.6 %)
Nonresponders 90.0 % (86.1–93.1 %) 75.7 % (70.6–80.4 %) 68.2 % (62.6–73.4 %) 57.9 % (52.6–63.1 %)

123
1894 J Cancer Res Clin Oncol (2013) 139:1887–1898

100 (0 vs. 4.7 %, P = 0.022; 0 vs. 3.3 %, P = 0.052), deep

90
DFP (%)

stromal invasion (19.8 vs. 53.5 %, P = 0.000; 19.8 vs.

80
70
29.3 %, P = 0.08), lymph node metastasis (8.1 vs. 25.6 %,
60 P = 0.004; 8.1 vs. 17.3 %, P = 0.031), and the need of
50 adjuvant radiotherapy (5.5 vs. 30.2 %, P = 0.000; 5.4 vs.
0 year 1 year 2 year 3 year 4 year 5 year
15.3 %, P = 0.012), as compared with nonresponders and
Follow-up
primary surgery treatment. Chen et al. (2008) reported the
100 pelvic metastasis (P = 0.025) and parametrial infiltration
90 rates (P = 0.038) were significantly lower in NAC group
OS (%)

80 than in the primary surgery group. The same results were

70
confirmed by other studies (Xiong et al. 2011; Huang et al.
60
50 2011; Gong et al. 2012). These results may be the reason
0 year 1 year 2 year 3 year 4 year 5 year
that NAC achieved favorable prognosis.
Follow-up In our included studies, NAC agents varied greatly, but
Fig. 4 Pooled progression-free survival (PFS) rate and overall most of them were cisplatin-based. In the SNAP01 study,
survival (OS) rate in NAC responders and nonresponders Buda et al. compared a three-drug [paclitaxel, ifosfamide,
and cisplatin (TIP)] with a two-drug [ifosfamide and cis-
nonresponders. No heterogeneity was found between the platin (IP)] regimen followed by radical surgery in patients
studies (I2 = 24.8 %, P = 0.200). The pooled OR was with locally advanced squamous cell cervical carcinoma.
5.785 (95 % CI 4.124–8.115; Z = 10.17, P = 0.000) by The results showed the TIP regimen is associated with a
fixed-effect model. The Begg’s test (z = 0.34, P = 0.732) higher response rate (residual disease \3 mm stromal
and the Egger’s test (t = 0.21, P = 0.836) suggested there invasion) than the IP regimen (48 vs. 23 %; OR 3.22; 95 %
was no significant publication bias. The forest plot and CI 1.69–5.88; P = 0.0003), without a statistically signifi-
funnel plot are showed in ‘‘Appendix’’. cant effect on OS (HR 0.66; 95 % CI 0.39–1.10; P = 0.11)
The pooled OS rates in NAC responders and nonre- (Buda et al. 2005). No other randomized controlled clinical
sponders are listed in the Table 3 and Fig. 4. The results trials compared the effect of different NAC agents. In our
showed NAC response is a prognostic indicator for OS in included studies, NAC cycles ranged 2–6. Exist evidence
cervical cancer patients treated with NAC before surgery. showed chemotherapy cycle lengths of 14 days and shorter
HR 0.83 (95 % CI 0.69–1.00; P = 0.046) or cisplatin dose
Sensitivity analyses intensities greater than or equal to 25 mg/m2 per week (HR
0.91; 95 % CI 0.78–1.05; P = 0.20) tended to show an
After removing the largest sample study (Hu et al. 2012), advantage for NAC on survival. In contrast, trials using
the pooled OR of PFS and OS did not change much. The cycle lengths longer than 14 days (HR 1.25; 95 % CI
results of sensitivity analysis are presented in Table 4. 1.07–1.46; P = 0.005) or cisplatin dose intensities lower
Sensitivity analysis showed the results of meta-analysis than 25 mg/m2 per week (HR 1.35; 95 % CI 1.11–1.14;
were robust. P = 0.002) tended to show a detrimental effect of NAC
(2003) on survival. However, new randomized controlled
clinical trials are needed to assess different NAC agents,
Discussion cycle lengths, and dose intensities.
Our meta-analysis only compared NAC responders with
Our meta-analysis found response to NAC is an important nonresponders for long-term prognosis. Hu et al. also
indicator for PFS and OS at 1-, 2-, 3-, and 5-year follow-up compared NAC–nonresponders with primary surgery
for cervical cancer patients. Li et al. (2013) also reported treatment. They found the 5-year DFS rate was 70 % and
NAC decreased the ratio of lymphovascular space invasion 5-year OS rate 68.9 % in the clinical nonresponders

Table 4 Pooled OR of progression-free survival (PFS) and overall survival (OS) compared NAC responders with nonresponders in sensitivity
analysis
Outcomes 1-year 2-year 3-year 5-year

PFS (95 % CI) 6.671 (3.489–12.754) 5.645 (3.515–9.065) 5.517 (3.299–9.225) 5.052 (3.198–7.980)
OS (95 % CI) 7.478 (3.324–16.824) 8.158 (4.532–14.685) 7.379 (4.255–12.795) 4.536 (2.936–7.007)

123
J Cancer Res Clin Oncol (2013) 139:1887–1898 1895

subgroup. The prognosis outcomes were poorer in NAC– Study %

nonresponders compared with primary surgery treatment ID OR (95% CI) Weight

(P \ 0.05) (Hu et al. 2012). Chen et al. (2008) also

Shoji T, 2013 1.88 (0.25, 14.08) 6.66
reported NAC–nonresponders decreased the tumor-free Li R, 2013 4.34 (1.16, 16.23) 9.20
survival compared with primary surgery, although no sig- Robova H, 2013 5.56 (2.02, 15.26) 12.55

nificant difference was detected by log-rank test. The poor Hu T, 2012 9.34 (5.32, 16.39) 31.18
Xiong Y, 2011 6.18 (1.73, 22.03) 8.77
prognosis outcome may be due to delays of surgery. These Huang X, 2011 7.69 (1.25, 47.22) 2.94
results suggest that early surgery should be conducted in Park JS, 2011 6.90 (1.68, 28.41) 6.68

non-NAC responders for avoiding the delay of effective Kumar JV, 2009 27.65 (1.53, 499.14) 1.74
Choi CH, 2008 7.13 (0.36, 142.56) 2.39
treatment.
Chen H, 2008 11.50 (3.07, 43.05) 5.85
Hu et al. (2012) reported patients with earlier FIGO Chen CA, 2002 0.92 (0.16, 5.49) 12.04

stage or smaller tumor size have a better clinical response Overall (I-squared = 0.0%, p = 0.461) 6.80 (4.72, 9.80) 100.00

(P \ 0.001). While Chen et al. (2008) found squamous

tumors showed a more favorable response rate than non- .002 1 499

squamous tumors (76.7 vs. 33.3 %, P = 0.005). In our

Fig. 5 Pooled OR of 2-year progression-free survival compared
meta-analysis, we did not conduct the subgroup analysis by NAC responders with nonresponders
FIGO stage, tumor size, or histology because the included
studies did not provide enough data. However, no hetero-
geneity was found between the studies in our meta-ana- Funnel plot with pseudo 95% confidence limits
lysis. It seems NAC responders before surgery have

0
favorable prognosis regardless of FIGO stage, tumor size,
and histology. This conclusion should be confirmed by new .5

prospective studies.
se(logOR)

There are some limitations in this meta-analysis. First,

some of the included studies did not provide sufficient data
1

of time-to-event outcomes for meta-analysis directly. We

used Engauge Digitizer to extract data from survival
curves, which may result in inaccurate data. Second,
1.5

although no significant publication bias was found in the

0 10 20 30
meta-analysis, we only searched through English-language OR
reports, and it may have missed studies in our literature
Fig. 6 The funnel plot of meta-analysis for 2-year progression-free
review.
survival compared NAC responders with nonresponders
Despite of these limitations, our meta-analysis confirms
that response to neoadjuvant chemotherapy is an indicator
for PFS and overall survival, and suggests that patients-
achieving response of neoadjuvant chemotherapy before Study %

surgery predicts favorable prognosis for cervical cancer ID OR (95% CI) Weight

patients.
Shoji T, 2013 2.55 (0.23, 27.71) 4.44

Li R, 2013 4.34 (1.16, 16.23) 8.92

Acknowledgments This paper is dedicated to my wife’s mother. Robova H, 2013 4.67 (1.82, 11.99) 15.48
Two months ago, she was diagnosed with stage IIB cervical cancer Hu T, 2012 7.46 (4.51, 12.33) 41.79
and was clinical responsible for cisplatin-based NAC. God Bless.
Park JS, 2011 5.00 (1.30, 19.30) 8.63

Kumar JV, 2009 49.37 (2.72, 894.79) 1.23

Conflict of interest The authors had no conflict of interest to
Choi CH, 2008 8.79 (0.44, 174.04) 2.17
declare in relation to this article.
Chen H, 2008 11.50 (3.07, 43.05) 5.67

Chen CA, 2002 0.92 (0.16, 5.49) 11.67

Overall (I-squared = 11.5%, p = 0.339) 6.33 (4.40, 9.10) 100.00

Appendix .00112 1 895

Fig. 7 Pooled OR of 3-year progression-free survival compared

See Figs. 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 18. NAC responders with nonresponders

123
1896 J Cancer Res Clin Oncol (2013) 139:1887–1898

Funnel plot with pseudo 95% confidence limits

0 Study %
ID OR (95% CI) Weight

Shoji T, 2013 2.00 (0.14, 27.99) 8.81

Li R, 2013 19.27 (0.97, 381.29) 3.29
.5
se(logOR)

Robova H, 2013 19.59 (0.92, 418.77) 2.53

Hu T, 2012 4.51 (1.80, 11.33) 43.81
Xiong Y, 2011 4.00 (0.34, 46.95) 8.02
Huang X, 2011 21.00 (0.77, 571.90) 1.53
1

Mori T, 2010 8.33 (0.41, 170.67) 2.53

Selvaggi L, 2006 14.75 (1.62, 133.94) 3.03
Fuso L, 2005 77.31 (4.24, 1409.93) 1.93
Chen CA, 2002 0.43 (0.04, 5.35) 24.52
1.5

Overall (I-squared = 16.2%, p = 0.294) 6.18 (3.39, 11.26) 100.00

0 10 20 30 40 50
OR
.00071 1 1410

Fig. 8 The funnel plot of meta-analysis for 3-year progression-free

survival compared NAC responders with nonresponders Fig. 11 Pooled OR of 1-year overall survival compared NAC
responders with nonresponders

Funnel plot with pseudo 95% confidence limits

Study %

0
ID OR (95% CI) Weight

Shoji T, 2013 4.48 (0.21, 95.59) 1.91

.5
Li R, 2013 2.41 (0.76, 7.63) 11.92
Robova H, 2013 3.14 (1.29, 7.67) 16.76
se(logOR)

Hu T, 2012 5.43 (3.39, 8.70) 42.90

1

Park JS, 2011 5.81 (1.44, 23.36) 5.93

Gadducci A, 2010 20.83 (3.58, 121.13) 1.98
Kumar JV, 2009 49.37 (2.72, 894.79) 0.94
1.5

Choi CH, 2008 8.79 (0.44, 174.04) 1.66

Chen H, 2008 7.58 (2.43, 23.65) 7.09
Chen CA, 2002 0.92 (0.16, 5.49) 8.92
Overall (I-squared = 24.2%, p = 0.221) 5.21 (3.75, 7.25) 100.00
2

0 20 40 60 80
.00112 1 895 OR

Fig. 9 Pooled OR of 5-year progression-free survival compared Fig. 12 The funnel plot of meta-analysis for 1-year overall survival
NAC responders with nonresponders compared NAC responders with nonresponders

Study %

Funnel plot with pseudo 95% confidence limits ID OR (95% CI) Weight
0

Shoji T, 2013 1.33 (0.17, 10.25) 13.05

Li R, 2013 38.65 (2.13, 702.59) 2.00
Robova H, 2013 8.92 (2.74, 29.08) 12.70
Hu T, 2012 11.41 (5.50, 23.67) 30.66
.5
se(logOR)

Xiong Y, 2011 16.82 (3.20, 88.51) 6.11

Huang X, 2011 41.36 (1.75, 977.56) 0.85
Mori T, 2010 4.00 (0.27, 58.56) 3.34
Cai HB, 2006 17.80 (0.66, 479.13) 1.16
1

Selvaggi L, 2006 25.50 (2.68, 242.30) 2.49

Fuso L, 2005 33.73 (4.02, 283.42) 3.43
Chen CA, 2002 0.27 (0.02, 2.90) 24.21
Overall (I-squared = 42.2%, p = 0.068) 9.16 (5.76, 14.55) 100.00
1.5

0 10 20 30 40 50
OR .00102 1 978

Fig. 10 The funnel plot of meta-analysis for 5-year progression-free Fig. 13 Pooled OR of 2-year overall survival compared NAC
survival compared NAC responders with nonresponders responders with nonresponders

123
J Cancer Res Clin Oncol (2013) 139:1887–1898 1897

Funnel plot with pseudo 95% confidence limits

0 Study %
ID OR (95% CI) Weight

Shoji T, 2013 5.72 (0.27, 120.33) 1.87

.5

Li R, 2013 4.34 (1.16, 16.23) 7.20

Robova H, 2013 3.93 (1.57, 9.84) 14.59
se(logOR)

Hu T, 2012 8.92 (5.28, 15.08) 28.47

1

Gadducci A, 2010 5.45 (1.22, 24.43) 5.75

Mori T, 2010 7.67 (0.77, 76.45) 1.50
Park DC, 2009 5.17 (0.36, 75.13) 1.21
1.5

Cai HB, 2006 13.67 (2.23, 83.94) 1.73

Selvaggi L, 2006 9.52 (1.04, 86.91) 2.35
Fuso L, 2005 17.11 (4.22, 69.45) 3.69
Chen CA, 2002 0.57 (0.08, 4.01) 10.15
2

MacLeod C, 2001 2.84 (1.13, 7.14) 21.49

0 10 20 30 40
Overall (I-squared = 24.8%, p = 0.200) 5.79 (4.12, 8.11) 100.00
OR

Fig. 14 The funnel plot of meta-analysis for 2-year overall survival .00831 1 120
compared NAC responders with nonresponders
Fig. 17 Pooled OR of 5-year overall survival compared NAC
responders with nonresponders

Study % Funnel plot with pseudo 95% confidence limits

0
ID OR (95% CI) Weight

Shoji T, 2013 4.00 (0.37, 43.14) 4.90

Li R, 2013 38.65 (2.13, 702.59) 1.51 .5
Robova H, 2013 4.95 (1.84, 13.32) 18.26
se(logOR)

Hu T, 2012 10.01 (5.72, 17.50) 40.02

Xiong Y, 2011 24.67 (4.67, 130.35) 3.76
Mori T, 2010 4.00 (0.27, 58.56) 2.51
Cai HB, 2006 25.80 (2.24, 297.58) 1.21
1

Selvaggi L, 2006 18.46 (1.98, 172.17) 2.43

Fuso L, 2005 12.57 (3.10, 51.02) 7.21
Chen CA, 2002 0.27 (0.02, 2.90) 18.19
1.5

Overall (I-squared = 37.6%, p = 0.108) 8.43 (5.67, 12.54) 100.00

0 5 10 15 20
.00142 1 703 OR

Fig. 15 Pooled OR of 3-year overall survival compared NAC Fig. 18 The funnel plot of meta-analysis for 5-year overall survival
responders with nonresponders compared NAC responders with nonresponders

References
Funnel plot with pseudo 95% confidence limits
Buda A, Fossati R, Colombo N, Fei F, Floriani I, Gueli Alletti D,
0

Katsaros D, Landoni F, Lissoni A, Malzoni C, Sartori E, Scollo

P, Torri V, Zola P, Mangioni C (2005) Randomized trial of
neoadjuvant chemotherapy comparing paclitaxel, ifosfamide,
and cisplatin with ifosfamide and cisplatin followed by radical
.5

surgery in patients with locally advanced squamous cell cervical

se(logOR)

carcinoma: the SNAP01 (Studio Neo-Adjuvante Portio) Italian

Collaborative Study. J Clin Oncol 23(18):4137–4145
Cai HB, Chen HZ, Yin HH (2006) Randomized study of preoperative
1

chemotherapy versus primary surgery for stage IB cervical

cancer. J Obstet Gynaecol Res 32(3):315–323
Chen CA, Cheng WF, Wei LH, Su YN, Hsieh CY (2002) Radical
hysterectomy alone or combined with neoadjuvant chemother-
1.5

apy in the treatment of early stage bulky cervical carcinoma.

0 10 20 30 40 J Formos Med Assoc 101(3):195–202
OR Chen H, Liang C, Zhang L, Huang S, Wu X (2008) Clinical efficacy
of modified preoperative neoadjuvant chemotherapy in the
Fig. 16 The funnel plot of meta-analysis for 3-year overall survival treatment of locally advanced (stage IB2 to IIB) cervical cancer:
compared NAC responders with nonresponders randomized study. Gynecol Oncol 110(3):308–315

123
1898 J Cancer Res Clin Oncol (2013) 139:1887–1898

Choi CH, Song SY, Choi JJ, Park YA, Kang H, Kim TJ, Lee JW, MacLeod C, O’Donnell A, Tattersall MH, Dalrymple C, Firth I
Kim BG, Lee JH, Bae DS (2008) Prognostic significance of (2001) Locally advanced cervix cancer: chemotherapy prior to
VEGF expression in patients with bulky cervical carcinoma definitive surgery or radiotherapy. A single institutional expe-
undergoing neoadjuvant chemotherapy. BMC Cancer 8:295. rience. Australas Radiol 45(4):491–495
http://www.ncbi.nlm.nih.gov/pubmed/18847499 Mori T, Hosokawa K, Sawada M, Kuroboshi H, Tatsumi H, Koshiba
Fuso L, Mazzola S, Marocco F, Ferrero A, Dompè D, Carus AP, Zola H, Okubo T, Kitawaki J (2010) Neoadjuvant weekly carboplatin
P (2005) Pretreatment serum hemoglobin level as a predictive and paclitaxel followed by radical hysterectomy for locally
factor of response to neoadjuvant chemotherapy in patients with advanced cervical cancer: long-term results. Int J Gynecol
locally advanced squamous cervical carcinoma: a preliminary Cancer 20(4):611–616
report. Gynecol Oncol 99(3 Suppl 1):S187–S191 NCCN clinical practice guidelines in oncology (NCCN Guidelines)
Gadducci A, Cosio S, Zola P, Tisi G, Ferrero A, Piovano E, Cristofani cervical cancer. Available from: http://www.nccn.org/professionals/
R, Greco C, Sartori E (2010a) Pretreatment platelet and physician_gls/pdf/cervical.pdf. Assessed 25 July 2013
hemoglobin levels are neither predictive nor prognostic variables Neoadjuvant Chemotherapy for Locally Advanced Cervical Cancer
for patients with locally advanced cervical cancer treated with Meta-analysis Collaboration (2003) Neoadjuvant chemotherapy
neoadjuvant chemotherapy and radical hysterectomy: a retro- for locally advanced cervical cancer: a systematic review and
spective Italian study. Int J Gynecol Cancer 20(8):1399–1404 meta-analysis of individual patient data from 21 randomised
Gadducci A, Fabrini MG, Perrone F, Manfredi B, Fanucchi A, trials. Eur J Cancer 39(17):2470–2486
Laliscia C, Barsotti C, Scotti V, Genazzani AR, Greco C (2010b) Park DC, Suh MJ, Yeo SG (2009) Neoadjuvant paclitaxel and
Pattern of failures and clinical outcome of patients with locally cisplatin in uterine cervical cancer: long-term results. Int J
advanced cervical cancer treated with a tailored integrated Gynecol Cancer 19(5):943–947
therapeutic approach. Anticancer Res 30(9):3731–3735 Park JS, Jeon EK, Chun SH, Won HS, Lee A, Hur SY, Lee KH, Bae
Gong L, Lou JY, Wang P, Zhang JW, Liu H, Peng ZL (2012) Clinical SN, Yoon SC, Hong SH (2011) ERCC1 (excision repair cross-
evaluation of neoadjuvant chemotherapy followed by radical complementation group 1) expression as a predictor for response
surgery in the management of stage IB2-IIB cervical cancer. Int J of neoadjuvant chemotherapy for FIGO stage 2B uterine cervix
Gynaecol Obstet 117(1):23–26 cancer. Gynecol Oncol 120(2):275–279
González-Martı́n A, González-Cortijo L, Carballo N, Garcia JF, Robova H, Rob L, Halaska MJ, Pluta M, Skapa P, Strnad P, Lisy J,
Lapuente F, Rojo A, Chiva LM (2008) The current role of Komar M (2013) High-dose density neoadjuvant chemotherapy
neoadjuvant chemotherapy in the management of cervical in bulky IB cervical cancer. Gynecol Oncol 128(1):49–53
carcinoma. Gynecol Oncol 110(3 Suppl 2):S36–S40 Rydzewska L, Tierney J, Vale CL, Symonds PR (2012) Neoadjuvant
Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring chemotherapy plus surgery versus surgery for cervical cancer.
inconsistency in meta-analyses. BMJ 327:557–560 Cochrane Database Syst Rev 12:CD007406
Hu T, Li S, Chen Y, Shen J, Li X, Huang K, Yang R, Wu L, Chen Z, Selvaggi L, Loizzi V, Di Gilio AR, Nardelli C, Cantatore C, Cormio
Jia Y, Wang S, Cheng X, Han X, Lin Z, Xing H, Qu P, Cai H, G (2006) Neoadjuvant chemotherapy in cervical cancer: a 67
Song X, Tian X, Xu H, Xu J, Zhang Q, Xi L, Deng D, Wang H, patients experience. Int J Gynecol Cancer 16(2):631–637
Wang S, Lv W, Wang C, Xie X, Ma D (2012) Matched-case Shoji T, Takatori E, Saito T, Omi H, Kagabu M, Miura F, Takeuchi S,
comparison of neoadjuvant chemotherapy in patients with FIGO Sugiyama T (2013) Neoadjuvant chemotherapy using platinum-
stage IB1-IIB cervical cancer to establish selection criteria. Eur J and taxane-based regimens for bulky stage Ib2 to IIb non-
Cancer 48(15):2353–2360 squamous cell carcinoma of the uterine cervix. Cancer Chemo-
Huang X, Lan C, Huang H, Zhang Y, Huang H, Cao X, Huang Y, ther Pharmacol 71(3):657–662
Guo Y, Wan T, Liu J (2011) Neoadjuvant docetaxel combined Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M,
with cisplatin and followed by radical surgery for the treatment Tugwell P (2013) The Newcastle–Ottawa Scale (NOS) for
of locally advanced (stage IB2–IIB) cervical cancer: preliminary assessing the quality of non-randomized studies in meta-
results of a single-institution experience. Expert Opin Pharmac- analysis. Ottawa Health Research Institute. Available from:
other 12(2):165–173 http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.
Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D (2011) Assessed 25 July 2013
Global cancer statistics. CA Cancer J Clin 61(2):69–90 Wiebe E, Denny L, Thomas G (2012) Cancer of the cervix uteri. Int J
Kumar JV, Doval DC, Rao R, Rawal S (2009) A retrospective study Gynaecol Obstet 119(Suppl 2):S100–S109
of patients with locally advanced cancer of the cervix treated Xiong Y, Liang LZ, Cao LP, Min Z, Liu JH (2011) Clinical effects of
with neoadjuvant chemotherapy followed by radical surgery. Int irinotecan hydrochloride in combination with cisplatin as
J Gynecol Cancer 19(3):417–422 neoadjuvant chemotherapy in locally advanced cervical cancer.
Li R, Lu ST, Si JG, Liu B, Wang H, Mei YY, Linghu H (2013) Gynecol Oncol 123(1):99–104
Prognostic value of responsiveness of neoadjuvant chemother-
apy before surgery for patients with stage IB(2)/IIA(2) cervical
cancer. Gynecol Oncol 128(3):524–529

123