Available online at www.sciencedirect.

com

Gynecologic Oncology 110 (2008) S36 – S40

www.elsevier.com/locate/ygyno

The current role of neoadjuvant chemotherapy in the

management of cervical carcinoma
Antonio González-Martín a,b,⁎, Lucía González-Cortijo a , Natalia Carballo c , Juan F. Garcia d ,
Fernando Lapuente e , Alejandro Rojo d , Luis M. Chiva e
a
Medical Oncology Department, M. D. Anderson International España, Madrid, Spain
b
Medical Oncology Service, Hospital Universitario Ramón y Cajal, Madrid, Spain
c
Radiation Oncology Department, M. D. Anderson International España, Madrid, Spain
d
Pathology Department, M. D. Anderson International España, Madrid, Spain
e
Gynecologic Oncology Department, M. D. Anderson International España, Madrid, Spain
Received 14 May 2008
Available online 30 June 2008

Abstract

The role of neoadjuvant chemotherapy (NACT) in cervical cancer has been a matter of investigation over the last 20 years. A systematic review
and meta-analysis of individual patient data (IPD) demonstrated that NACT followed by surgery is superior to radiotherapy alone in terms of
overall survival. However, in spite of the results of the meta-analysis, NACT has not been adopted as the new standard of care. In the present
paper, we review the reasons why NACT is still considered an investigational approach in cervical cancer.
© 2008 Elsevier Inc. All rights reserved.

Keywords: Neoadjuvant chemotherapy; Cervical carcinoma; Chemotherapy; Meta-analysis

Introduction The rationales for the use of neoadjuvant chemotherapy

(NACT) are several. Tumor-size reduction may facilitate
Cervical cancer is still the second most common malignancy subsequent local therapy, whether radiotherapy or surgery. This
and second most common cause of cancer-related death in reduction can transform inoperable tumors into radically resect-
women worldwide [1]. The geographic distribution of cervical able ones. Also, NACT has been suggested to increase radio-
cancer cases is associated with the level of development; up to sensitivity and decrease the hypoxic cell fraction. Moreover,
80% of all incidental cases of cervical cancer occur in resource- NACT, treats the micrometastatic disease, preventing a significant
poor populations, with the highest incidence rates observed in proportion of relapses. Finally, response to NACT has been
sub-Saharan Africa, Melanesia, Latin America and the Car- identified as an important prognostic factor in several studies [3,4].
ibbean, South-Central Asia, and Southeast Asia. By contrast, in On the other hand, some concerns have been associated with
industrialized nations, the widespread use of the Papanicolau the use of NACT. In patients who do not respond to chemotherapy,
(Pap) screening test has reduced cervical cancer rates by nearly the administration of curative treatment will have been delayed
80% in the last 50 years [2]. This distribution of cervical cancer unnecessarily. Moreover, some chemotherapy agents could have
cases may have some impact on management of the disease cross-resistance with radiotherapy, inducing the development of
because of the difficulties in access to standard therapies in some radioresistant cellular clones.
parts of the world. In spite of these concerns, chemoradiation is now considered
the treatment of choice for patients who are not candidates for
⁎ Corresponding author. Servicio de Oncología Médica, Hospital Ramón y
initial surgical therapy. Large retrospective case series have
Cajal, Ctra Colmenar Viejo Km 9100, 28034 Madrid, Spain. Fax: +34
shown that NACT is being used in up to 25% of some groups of
913368263. patients, such as those with International Federation of
E-mail address: [email protected] (A. González-Martín). Gynecology and Obstetrics (FIGO) disease stage IB2 [5]. In
0090-8258/$ - see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.ygyno.2008.05.012
 A. González-Martín et al. / Gynecologic Oncology 110 (2008) S36–S40 S37

this paper, we will review the evidence for the efficacy of Table 1
NACT and the most recent published data. Overall survival (OS) by frequency of chemotherapy and cisplatin dose intensity
in comparison 1 [6]

Meta-analysis of individual patient data Variable Trials HR (95% CI) p value Heterogeneity 5-year OS
p value

Since the first publication by Friedlander in 1983 on the use Frequency of chemotherapy
N14 days 11 1.25 (1.07–1.46) 0.005 0.23 A8%
of primary chemotherapy in cervical carcinoma, countless
≤14 days 6 0.76 (0.62–0.92) 0.005 0.19 z7%
phase II trials and several phase III trials have been published.
However, the small size of some trials, the inclusion of different Cisplatin dose intensity
disease stages in the studies, and the use of different strategies b25 mg/m2 7 1.35 (1.11–1.64) 0.002 0.74 A11%
have produced conflicting or disappointing results. In this ≥25 mg/m2 11 0.91 (0.78–1.05) 0.2 0.001 z3%
context, the combination of data from individual clinical trials
in a meta-analysis might give sufficient statistical power to
determine whether NACT is useful. tasis-free survival. In contrast, trials using shorter cycle lengths
In 2003, Tierney et al. published an individual patient data had a pooled HR of 0.76, equivalent to a 7% absolute
meta-analysis initiated and coordinated by the United Kingdom improvement in overall 5-year survival. Results for disease-
Medical Research Council Clinical Trials Unit and carried out free, locoregional disease-free, and metastasis-free survival
by the Neoadjuvant Chemotherapy for Cervical Cancer Meta- similarly suggested a benefit for short-cycle chemotherapy. A
analysis Collaboration. The meta-analysis included all relevant comparable, but less clear, result was observed when grouping
trials published up to that date [6]. trials according to the planned cisplatin dose intensity. Those
Two separate treatment comparisons were included: trials using a dose intensity less than 25 mg/m2 per week were
associated with an HR of 1.35, which translates into an 11%
1. Comparison 1, in which NACT followed by local treat- reduction in 5-year overall survival. In contrast, cisplatin dose
ment was compared with the same local treatment (mainly intensities of cisplatin 25 mg/m2 or more per week suggested a
radiotherapy) alone, and potential 3% improvement in 5-year overall survival, but this
2. Comparison 2, in which a combination of NACT followed was not statistically significant (HR 0.91; 95% CI 0.78–1.05). It
by surgery (with or without radiotherapy) was compared is not possible to give an accurate explanation for the improved
with the (then) more standard radiotherapy alone. responses to NACT without improved survival, but it has been
suggested that chemotherapy may select for radioresistant
Comparison 1 was based on 2074 patients from 18 trials, cellular clones due to cross-resistance between certain che-
representing 92% of patients from eligible randomized trials. motherapy agents and radiotherapy. Another explanation is that
Data on survival were available for all 18 trials, including 1084 more than 14 days of chemotherapy administration could lead to
patient deaths. The median follow-up across all trials was an accelerated regrowth of surviving cell clones, thus lessening
5.7 years for surviving patients. The majority of trials compared the effect of subsequent radiotherapy.
NACT followed by radiotherapy with the same radiotherapy Comparison 2 of the meta-analysis compared NACT followed
alone. Almost 70% of patients had advanced disease (stage II by surgery (with or without subsequent radiotherapy) to radical
or III), and the status of lymph nodes was unknown in 60% of radiotherapy alone. This analysis comprised 5 randomized trials
patients. with a total of 872 patients. The planned total dose of cisplatin
The comparison 1 showed that the addition of NACT to local was between 100 and 300 mg/m2 in 10- to 21-day cycles;
therapy (mainly radiotherapy) did not have any impact on external radiotherapy and intracavitary radiotherapy doses in the
overall survival (hazard ratio [HR] 1.05; 95%confidence radiotherapy alone arm were similar across trials (45–60 and 25–
interval [CI]: 0.94–1.19), disease-free survival (HR 1.00; 95% 40 Gy, respectively). The patient population in this comparison
CI: 0.88–1.14), or locoregional disease-free survival (HR 1.03; had less advanced disease, with approximately one third of
95% CI: 0.9–1.17). However, a highly significant level of patients having stage IB and one third having stage II disease. It is
statistical heterogeneity was evident for each of the outcomes noteworthy that chemotherapy was generally intense and of short
measured (e.g. heterogeneity p value for survival was 0.0003). duration.
Such observations may indicate that it is inappropriate to The results of comparison 2 suggest a highly significant
combine the trials in this way and in fact suggest that these trials effect of NACT, with an HR of 0.65 ( p = 0.00004), which
may not have addressed exactly the same question. For this translates into an absolute gain in 5-year overall survival of 14%
reason, the authors conducted a new analysis, grouping the trials (from 50% to 64%).
according to the frequency of the chemotherapy cycles and the
cisplatin dose intensity (Table 1). Grouping trials by these Factors contributing to reluctance to adopt NACT
variables provided the best explanation for the heterogeneity
seen among the trials. For trials with treatment cycles longer Comparison 2 of the meta-analysis showed a significant
than 14 days, the pooled HR was 1.25, equivalent to an absolute increase in overall survival with the administration of neoadju-
decrease of 8% in 5-year overall survival. A decrease was also vant chemotherapy before surgery. However, NACT has not been
observed in disease-free, locoregional disease-free and metas- adopted as the standard of care and is still considered
S38 A. González-Martín et al. / Gynecologic Oncology 110 (2008) S36–S40

investigational. Several explanations for this hesitancy to stan- Table 2

dardize its use can be given. High-risk surgical–pathological findings in GOG-141 trial [8]
Finding Number of patients (%)
Potential confounding factors in meta-analysis RHPPL NACT + RHPPL All
(N = 143) (N = 145) (N = 288)
First of all, the number of patients (872) and events (368) in
Extrauterine disease
comparison 2 was not large, and the results need to be interpreted Positive pelvic nodes 56 (39) 47 (32) 103 (36)
with caution. Furthermore, in two trials, almost all patients Positive paraaortic nodes 20 (14) 15 (10) 35 (12)
received pelvic radiotherapy, and in another two trials, it was Parametrial involvement 28 (20) 24 (17) 52 (18)
given to approximately 30% of the patients in the experimental Intraperitoneal disease 9 (6) 7 (5) 16 (6)
arm. Therefore, postoperative radiotherapy may act as a Uterine disease
N1/3 cervical penetration 26 (18) 31 (21) 57 (20)
confounding factor. Moreover, it is noteworthy that almost half CLS involvement 62 (43) 57 (39) 19 (41)
of the patients included in the comparison 2 of the meta-analysis Positive surgical margin 15 (10) 13 (9) 28 (10)
came from one trial from Italy [7]. When analyzing the results of RHPPL: radical hysterectomy and pelvic-paraaortic lymphadenectomy; NACT:
this Italian trial, we have to take into account that 25% of patients neoadjuvant chemotherapy; CLS: capillary-lymphatic space.
did not complete the planned therapy and that a benefit in survival
was observed only for disease stages IB to IIB, but not for stage
III. More importantly, the median total dose of radiotherapy at surgery performed (79% in the chemotherapy arm vs. 78% in
point A in the control arm (70 Gy) was lower than the the surgery-only arm) or in the surgical–pathological findings
radiotherapy dose considered standard now (85–90 Gy). More- (Table 2). Moreover, no significant differences were observed in
over, no paraaortic irradiation was administered, despite 60% of the progression-free survival rate (RHPPL: 60.4% and 53.8%;
patients having had stage IIB–III disease).Three additional NACT+RHPPL: 59.7% and 56.2%, at 3 and 5 years, respec-
arguments can be added to explain why the meta-analysis tively) and overall survival rate (RHPPL: 69.3% and 60.7%; and
generally has not been considered definitive evidence for NACT: NACT + RHPPL: 67.7% and 63.3%, at 3 and 5 years, respec-
First, the heterogeneity of the stages included in the trials; second, tively) Although not definitive, the results of this study have led
the chemotherapy regimens not including modern, more-active the GOG to recommend against adding NACT to RHPPL in
drugs; and third (and probably most important), the radiotherapy future randomized trials for stage IB2 cervical cancer patients.
in the control arm being inferior to today's standard of care of
chemoradiation. Trials included in the meta-analysis employed old chemotherapy
regimens
Results of the GOG-141 trial do not support NACT
The great majority of trials included in the meta-analysis
One of the criticisms of the meta-analysis was the hetero- employed cisplatin-based regimens, which are now considered
geneity of stages included and the low number of patients with out-of-date: and more importantly, regimens were not selected
stage IB2 disease, for which, theoretically, NACT could be from phase III trials. Owing to the absence of controlled trials
more useful. The value of NACT in stage IB2 cervical cancer with different chemotherapy regimens in the neoadjuvant setting,
was specifically addressed by the recently published Gyneco- the selection of a NACT schedule needs to be extrapolated from
logic Oncology Group (GOG)-141 trial [8]. This trial was begun the results of chemotherapy in a different setting.
in 1996 to determine if NACT using vincristine and cisplatin The GOG has carried out three consecutive phase III trials
prior to radical hysterectomy and pelvic/paraaortic lymphade- comparing cisplatin doublets against cisplatin monotherapy for
nectomy (RHPPL) could improve progression-free survival and metastatic or recurrent cervical carcinoma (Table 3) [9–11]. In
overall survival, as well as operability, with acceptable levels of these trials, cisplatin was combined with ifosfamide (GOG-110),
toxicity. A total of 288 patients with stage IB2 cervical cancer paclitaxel (GOG-169), and topotecan (GOG-179). All the trials
were randomized to receive neoadjuvant chemotherapy (cis- showed that cisplatin doublet chemotherapy produced signifi-
platin 50 mg/m2 and vincristine 1 mg/m2 every 10 days for cantly higher rates of response and progression-free survival
3 cycles) followed by RHPPL or RHPPL alone. Adjuvant than did cisplatin monotherapy. Moreover, the combination of
radiotherapy was planned in the case of lymph node involve- cisplatin and topotecan was associated with a statistically sig-
ment or parametrial infiltration. Having met 70% of the accrual nificant benefit in overall survival, without any impact on quality
goal, the trial was closed early owing to poor accrual and the of life [12]. These results should be taken into account in the
off-protocol use of postoperative radiotherapy in approximately design of new trials of NACT.
12% of patients. Some published phase II trials that incorporated newer drugs
The response rate to neoadjuvant chemotherapy was 52% [13–18] showed high rates of response (78–95%) and
(15% complete responses, including 5% pathological complete resectability (76–100%), with a pathological complete response
responses; and 37% partial responses), which was inferior to the rate varying from 11% to 20% (Table 4).
response rate observed by the GOG with the same regimen in a An Italian trial compared an ifosfamide–cisplatin doublet to
previous phase II trial. In spite of the response rate, no significant an ifosfamide–cisplatin–paclitaxel (TIP) triplet in the neoadju-
differences were observed between study arms in the rate of vant setting [4]. A total of 204 patients with disease stages
 A. González-Martín et al. / Gynecologic Oncology 110 (2008) S36–S40 S39

Table 3 U.S. National Cancer Institute issued an alert that chemor-

Gynecologic Oncology Group (GOG) randomized trials with cisplatin-based adiotherapy should be considered for all patients with cervical
doublets in advanced cervical carcinoma
cancer who are not candidates for surgical therapy. Moreover,
GOG-110 [9] GOG-169 [10] GOG-179 [11] another meta-analysis that included 4580 randomized patients
Cisplatin Cisplatin + Cisplatin Cisplatin + Cisplatin Cisplatin + with cervical cancer confirmed that concomitant chemotherapy
ifosfamide paclitaxel topotecan and radiotherapy improves overall (HR 0.71; p b 0.0001;
RR (%) 19 31⁎ 19 36⁎ 13 27 a absolute benefit 12%) and progression-free survival (HR 0.61;
PFS (mo) 3.2 4.6⁎ 2.8 4.8⁎ 2.9 4.6 a p b 0.0001; absolute benefit 16%) and reduces local and distant
OS (mo) 8 8.3 8.8 9.7 6.5 9.4 a
recurrence. The greatest benefit is seen in patients with stage I or
RR: response rate; PFS: progression-free survival; OS: overall survival. II disease [19].
a
Statistically significant difference. One trial launched by the European Organization for
Research and Treatment of Cancer (EORTC) is now comparing
NACT followed by surgery with concomitant chemoradiother-
IB2–IVA were randomized to receive one of the regimens for apy The EORTC 55994 trail is recruiting patients with FIGO
three cycles before surgery. Postoperative radiotherapy was stage IB2, IIA, or IIB cervical cancer, with adequate perfor-
administered for lymph node infiltration, pathologically pos- mance status (0–2), an age of 18–75 years, and any histology
itive margins, suboptimal response, or parametrial invasion. (squamous carcinoma, adenocarcinoma, or adenosquamous
The primary endpoint was response rate, defined as the sum of carcinoma). Patients are randomized to receive neoadjuvant
pathological complete remissions and excellent partial remis- cisplatin-based chemotherapy followed by hysterectomy or to
sions (residual tumor with stromal invasionb3 mm). The re- receive chemoradiotherapy. The neoadjuvant cisplatin-based
sponse rate was statistically significantly higher with the TIP chemotherapy must include a minimum total dose of 225 mg/
regimen (48% vs. 23% with the doublet), but TIP was asso- m2, with a dose intensity of at least 25 mg/m2 per week, and the
ciated with higher hematological toxicity. Unfortunately, no final dose should be administered no later than the eighth week.
statistically significant differences in progression-free survival The chemoradiotherapy arm consists of the standard dosage
or overall survival were detected, although a trend in favor of the cisplatin at 40 mg/m2 per week for 6 weeks concomitantly with
TIP regimen was observed. However, patients who achieved external beam radiotherapy (45–50 Gy).
a complete response or an excellent partial response had a This trial was activated in 2002 with a planned accrual of 686
significantly longer duration of survival than those with a lesser patients. So far, around half of the patients have been recruited,
response. and the study is ongoing. We await the results with interest,
since this is the only trial comparing NACT followed by surgery
Meta-analysis control arm was not given today's standard with chemoradiotherapy.
chemoradiotherapy: relevance of ongoing EORTC 55994 trial
Conclusions
The most important limitation and criticism of the meta-
analysis is that the radiotherapy in the control arm was inferior NACT is feasible and produces impressive responses in
to chemoradiotherapy, which is considered the standard of care cervical carcinoma, as has been demonstrated by several phase
now. In the last 10 years, data from 9 randomized controlled II and phase III trials. Unfortunately, the results of an individual
trials of treatment for cervical cancer have been published, 6 of patient data meta-analysis did not support the administration of
which showed the benefit of concomitant cytotoxic chemother- NACT before radiotherapy alone. On the other hand, the same
apy with radiation (chemoradiotherapy). In February 1999, the meta-analysis suggested that NACT followed by surgery

Table 4
Phase II trails trials of new neoadjuvant chemotherapy regimens in cervical cancer
Ref. No. of patients Stage Chemotherapy RR (CR) Resectability (pCR)
2 2
[13] 41 IB2–IIIB Cisplatin 100 mg/m day 1 + Gem 1000 mg/m days 1–8 95% (7.5%) 77% (14%)
Every 21 days × 3 cycles
[14] 43 IB2–IIB Paclitaxel 60 mg/m2 + cisplatin 60 mg/m2 90% (39%) 100% (11%)
Every 10 days × 3 cycles
[15] 43 IB2–IIIB Paclitaxel 175 mg/m2 + Carbo AUC 6 95% (9%) 95% (17%)
Every 21 days × 3 cycles
[16] 42 IB2–IVA Paclitaxel 175 mg/m2 + Epi 100 mg/m2 + cisplatin 100 mg/m2 78% (19%) 76% (19%)
Every 21 days × 2–3 cycles
[17] 38 IB2–IIIB Paclitaxel 175 mg/m2 + IFX 5 g + cisplatin 50 mg/m2 84% (28%) 89% (16%)
Every 21 days × 3 cycles
[18] 58 ≥IB2 CDDP 80 mg/m2 day 1 + VNB 25 mg/m2 days 1–8 Not reported 81% (20%)
Every 21 days × 3 cycles
RR: response rate; CR: complete response; pCR: pathological complete response; Gem: gemcitabine; Epi: 4-epirubicin; IFX: ifosfamide; VNB: vinorelbine; AUC:
area under the curve.
S40 A. González-Martín et al. / Gynecologic Oncology 110 (2008) S36–S40

improves overall survival compared with nonstandard radio- [7] Benedetti-Panici P, Greggi S, Colombo A, et al. Neoadjuvant chemotherapy
therapy. However, the inferiority of the control arm (radio- and radical surgery versus exclusive radiotherapy in locally advanced
cervical cancer: results from the Italian multicenter randomised study. J Clin
therapy alone) compared with the current standard of care of Oncol 2002;20:179–88.
chemoradiotherapy precluded the adoption of NACT before [8] Eddy G, Bundy B, Creasman W, et al. Treatment of “bulky” stage IB
surgery as the standard of care. An ongoing trial by the EORTC cervical cancer with or without neoadjuvant vincristine and cisplatin prior
Gynecologic Group has been designed to compare NACT to radical hysterectomy and pelvic/para-aortic lymphadenectomy: a phase
III trial of the gynecologic oncology group. Gynecol Oncol 2007;106:
followed by surgery to standard chemoradiotherapy, and the
362–9.
final results are eagerly awaited. In the meantime, NACT should [9] Omura GA, Blessing JA, Vacarello L, et al. Randomized trial of cisplatin
still be considered investigational. versus cisplatin plus mitolactol versus cisplatin plus ifosfamide in
Several trials have shown that patients achieving a patho- advanced squamous carcinoma of the cervix: a Gynecologic Oncology
logical complete response to NACT do experience a significant Group study. J Clin Oncol 1997;15:165–71.
improvement in rates of overall survival. Based on this obser- [10] Moore DH, Blessing JA, McQuellon MP, et al. Phase III study of cisplatin
with or without paclitaxel in stage IVB, recurrent, or persistent squamous
vation, new drugs and regimens must be explored in order to cell carcinoma of the cervix: a gynecologic oncology group study. J Clin
increase the rate of pathological complete response. Oncol 2004;22:3113–9.
[11] Long III HJ, Bundy BN, Grendys Jr EC, et al. Randomized phase III trial of
Conflict of interest statement cisplatin with or without topotecan in carcinoma of the uterine cervix: a
The authors have no conflicts of interest to declare. Gynecologic Oncology Group study. J Clin Oncol 2005;23:4626–33.
[12] Monk B, Huang HQ, Cella D, et al. Quality of life outcomes from a
randomized phase III trial of cisplatin with or without topotecan in
References advanced carcinoma of the cervix: a Gynecologic Oncology Group Study.
J Clin Oncol 2005;23:4617–25.
[1] WHO. World Health Organization Report: Comprehensive Cervical Cancer [13] Dueñas-González A, López-Graniel C, González-Enciso A, et al.
Control: a Guide to Essential Practice. World Health Organization Web site; Concomitant chemoradiation versus neoadjuvant chemotherapy in locally
2006. Available at: www.who.int/reproductive-health/publications/cervical_ advanced cervical carcinoma: results from two consecutive phase II
cancer_gep/index.htm. Accessed on February 2008. studies. Ann Oncol 2002;13:1212–9.
[2] Parkin M, Bray F. The burden of HPV-related cancers. Vaccine 2006;24 [14] Park DC, Kim JH, Lew YO, et al. Phase II trial of neoadjuvant paclitaxel
(suppl 3):S11–25. and cisplatin in uterine cervical cancer. Gynecol Oncol 2004;92:59–63.
[3] Benedetti Panici P, Greggi S, Scambia G, et al. Long term survival [15] Dueñas-González A, López-Graniel C, González-Enciso A, et al. A phase
following neoadjuvant chemotherapy and radical surgery in locally II study of multimodality treatment for locally advanced cervical cancer:
advanced cervical cancer. Eur J Cancer 1998;34:341–6. neoadjuvant carboplatin and paclitaxel followed by radical hysterectomy
[4] Buda A, Fossati R, Colombo N, et al. Randomized trial of neoadjuvant and adjuvant cisplatin chemoradiation. Ann Oncol 2003;14:1278–84.
chemotherapy comparing paclitaxel, ifosfamide, and cisplatin with [16] D'Agostino G, Distefano M, Greggi S, et al. Neoadjuvant treatment of
ifosfamide and cisplatin followed by radical surgery in patients with locally advanced carcinoma of the uterine cervix with epirubicin,
locally advanced squamous cell cervical carcinoma: the SNAP01 (Studio paclitaxel and cisplatin. Cancer Chemother Pharmacol 2002;49:256–60.
Neo-Adiuvante Portio) Italian Collaborative Study. J Clin Oncol 2005;23: [17] Zanetta G, Lissoni A, Pellegrino A, et al. Neoadjuvant chemotherapy with
4137–45. cisplatin, ifosfamide and paclitaxel for locally advanced squamous-cell
[5] Ryu HS, Kang SB, Kim KT, et al. Efficacy of different types of treatment in cervical cancer. Ann Oncol 1998;9:977–80.
FIGO stage IB2 cervical cancer in Korea: results of a multicenter retrospective [18] Di Vagno G, Cormio G, Pignata S, et al. Cisplatin and vinorelbine as
Korean study (KGOG-1005). Int J Gynecol Cancer 2007;17:132–6. neoadjuvant chemotherapy in locally advanced cervical cancer: a phase II
[6] Neoadjuvant Chemotherapy for Locally Advanced Cervical Cancer study. Int J Gynecol Cancer 2003;13:308–12.
Metaanalysis Collaboration. Neoadjuvant chemotherapy for locally advanced [19] Green JA, Kirwan JM, Tierney JF, et al. Survival and recurrence after
cervical cancer: a systematic review and meta-analysis of individual patient concomitant chemotherapy and radiotherapy for cancer of the uterine
data from 21 randomised trials. Eur J Cancer 2003;39:2470–86. cervix: a systematic review and meta-analysis. Lancet 2001;358:781–6.