YGYNO-975298; No.

of pages: 7; 4C:
Gynecologic Oncology xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

Gynecologic Oncology
journal homepage: www.elsevier.com/locate/ygyno

1 Is there a role for postoperative treatment in patients with stage Ib2–IIb

2 cervical cancer treated with neo-adjuvant chemotherapy and radical

F
3 surgery? An Italian multicenter retrospective study

O
4Q1 F. Landoni a, E. Sartori b, T. Maggino c, P. Zola d, V. Zanagnolo a, S. Cosio e, F. Ferrari b, E. Piovano d, A. Gadducci e,⁎
5 a
Department of Gynecology, Cervical Cancer Center, European Institute of Oncology, Milan, Italy

O
6 b
Department of Gynecology and Obstetrics, University of Brescia, Brescia, Italy
7 c
Unit of Gynecology and Obstetrics, Umberto I Hospital, Venice-Mestre, Italy
8 d
Department of Surgical Science, University of Turin, Italy

R
9 e
Department of Clinical and Experimental Medicine, Division of Gynecology and Obstetrics, University of Pisa, Pisa, Italy
10

P
11 H I G H L I G H T S
12
13 • The achievement of an optimal pathological response on surgical specimen is a strong predictor of a better clinical outcome.
D
14 • Additional cycles of chemotherapy could be of benefit only for patients with suboptimal response and intra-cervical residual disease.
15 • Adjuvant chemotherapy or radiotherapy does not seem to improve clinical outcome with extra-cervical residual disease versus no further treatment.
16
E
17
T

a r t i c l e i n f o a b s t r a c t
18
19 Purpose. Neoadjuvant chemotherapy [NACT] followed by radical hysterectomy is an alternative therapeutic 31
C

Article history:
20 Received 21 October 2013 option to concurrent chemotherapy–radiotherapy for locally advanced cervical cancer. However there are very 32
21 Accepted 9 December 2013 few data about the effectiveness of any post-operative treatment in this clinical setting. The purpose of this study 33
22 Available online xxxx
E

was to correlate the patterns of recurrence and the clinical outcomes of cervical cancer patients who received 34
24
23
25 NACT, with postoperative adjuvant treatment. 35
26 Keywords:
Patients and methods. This retrospective multicenter study included 333 patients with FIGO stage Ib2–IIb cervical 36
R

27 Cervical cancer
28 Chemotherapy
cancer who underwent platinum-based NACT followed by radical surgery. Pathological responses were 37
29 Adjuvant treatment retrospectively assessed as complete; optimal partial; and suboptimal response. Overall optimal response rate 38
30 was the sum of complete and optimal partial response rates. 39
R

Survival
Results. On the whole series, recurrence-free survival was significantly longer in patients who achieved an over- 40
all optimal response than in those who did not (p b 0.0001), and in patients who received adjuvant chemotherapy 41
O

compared to those who did not (p = 0.0001). On multivariate analysis, consolidation therapy (p = 0.0012) was 42
the only independent prognostic variable for recurrence-free survival; whereas FIGO stage (p = 0.0169) and 43
consolidation therapy (p = 0.0016) were independent prognostic variables for overall survival. 44
C

Conclusion. Optimal responders after chemo-surgical treatment for FIGO stage Ib2–IIb cervical cancer do not 45
need any further treatment. Additional cycles of chemotherapy could be of benefit for patients with suboptimal 46
response and intra-cervical residual disease. Both adjuvant chemotherapy and adjuvant radiation treatments do 47
N

not seem to improve the clinical outcome of patients with extra-cervical residual disease compared to no further 48
treatment. 49
© 2013 Published by Elsevier Inc. 50
U

54 52
51
53
55 Introduction [1–3]. A larger survival advantage appears to occur when adjuvant 59
chemotherapy is administered after CCRT [3–5]. Conversely, adjuvant 60
56 Concurrent chemotherapy and radiotherapy [CCRT] is the standard hysterectomy in complete responders after CCRT gives no survival 61
57 of care for locally advanced cervical cancer, able to achieve a 6% benefit [6]. 62
58 improvement in a 5-year survival compared to radiotherapy alone Neoadjuvant chemotherapy [NACT] followed by radical hysterectomy 63
is an alternative therapeutic option in this clinical setting, and the meta- 64
⁎ Corresponding author at: Department of Clinical and Experimental Medicine,
Division of Gynecology and Obstetrics, University of Pisa, Via Roma 57, 56127 Pisa,
analysis of five randomized trials has shown that chemo-surgical treat- 65
Italy. Fax: + 39 050992354. ment significantly improves survival when compared to radiotherapy 66
E-mail address: a.gadducci@med.unipi.it (A. Gadducci). alone [7–14]. Benedetti-Panici et al. [9] reported that the survival benefit 67

0090-8258/$ – see front matter © 2013 Published by Elsevier Inc.

http://dx.doi.org/10.1016/j.ygyno.2013.12.010

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 2 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx

68 for NACT arm versus radiotherapy arm was significant for stage Ib2–IIb patients who did not complete the planned cycles of NACT or who did 98
69 but not for stage III disease. A multicenter randomized trial is currently not undergo radical surgery after NACT because of progression of dis- 99
70 comparing NACT plus radical surgery versus CCRT in patients with ease or poor general conditions were not included in the present study. 100
71 stage Ib2–IIb cervical cancer (EORTC protocol 55994). Most authors Patients' information from the hospital records, including surgical 101
72 found that the pathological response on surgical specimen is a strong notes and pathological reports, was collected using a common form 102
73 predictive factor for the clinical outcome of patients treated with NACT with standardized items and was stored in a database shared by all 103
74 and radical hysterectomy [7,11,15–17]. For instance, in the SNAP01— the participants in the study. 104
75 Italian Collaborative Study, women who achieved an optimal pathologi- Patient characteristics at initial diagnosis (such as date, age, FIGO 105
76 cal response had a hazard ratio [HR] of death of 5.88 (95% CI = 2.50– stage, histological type, and tumor size), NACT regimen, type of radical 106
77 13.84; p b 0.0001) compared to those who did not [11]. hysterectomy, and pathological responses on surgical specimens were 107
78 Conversely, there are very few data about the effectiveness of any reported for each case. 108
79 post-operative treatment in this clinical setting. Pre-treatment evaluation included history, physical examination, 109
80 The purpose of this retrospective multicenter study was to correlate vaginal–pelvic examination, colposcopy, biopsy, complete blood analy- 110
81 the patterns of recurrence and the clinical outcomes of cervical cancer sis, chest X-ray, and abdominal–pelvic computed tomography [CT] scan 111

F
82 patients who underwent platinum-based NACT followed by radical and Magnetic Resonance Imaging [MRI]. Cystoscopy and/or proctoscopy 112
83 hysterectomy with postoperative adjuvant treatment. were performed if there was a clinical or on CT and/or on MRI suspicions 113

O
of bladder or rectal involvement. Further investigation was performed 114
when indicated. 115
84 Materials and methods Physical and vaginal–pelvic examination and abdominal–pelvic CT 116

O
scan and MRI were repeated 3–4 weeks after the completion of chemo- 117
85 This is a retrospective study approved by the participating centers of therapy. All patients underwent type II–III radical hysterectomy accord- 118
86 the Institutional Review Board. The study included 333 patients with ing to the Piver–Rutledge classification with pelvic lymphadenectomy 119

R
87 FIGO stage Ib2–IIb cervical squamous cell or adenocarcinoma who within 3–6 weeks from the last cycle of chemotherapy. 120
88 underwent platinum-based NACT followed by radical hysterectomy Pathological responses were retrospectively assessed as follows: 121

P
89 with pelvic lymphadenectomy at the Department of Gynecology and complete response was defined as the complete disappearance of the 122
90 Obstetrics of the University of Pisa and Turin between 1992 and 2011 tumor in the cervix with negative nodes; optimal partial response was 123
91 and at the Department of Gynecology and Obstetrics of the University defined as a persistent residual disease with b 3 mm stromal invasion 124
92 of Brescia and at the Department of Gynecologic Oncology of the
D
including in situ carcinoma on the surgical specimen and negative 125
93 European Institute of Milan between 1999 and 2011. This was the treat- nodes; and suboptimal response consisted of persistent residual disease 126
94 ment strategy chosen for patients with stage Ib2–IIb disease 70-year as with N 3 mm stromal invasion on the surgical specimen and negative 127
E
95 upper limit of age, and good performance status who signed an nodes (intra-cervical residual disease), or positive nodes with positive 128
96 informed consent form. Patients with special histology (small cell carci- or negative parametria and/or surgical margins (extra-cervical residual 129
T

97 noma, glassy cell carcinoma and neuroendocrine tumors) as well as disease with positive nodes), or positive parametria and/or surgical 130
margins with negative nodes (extra-cervical residual disease with neg- 131
C

t1:1 Table 1 ative nodes). Overall optimal response rate was the sum of complete 132
t1:2 Recurrence rates and sites according to postoperative treatment. and optimal partial response rates. 133
t1:3 Postoperative treatment Patients Recurrences Overall Those patients who did not achieve complete, optimal partial or sub- 134
E

optimal response were defined as non-responders. 135

t1:4 P EX P + EX
Postoperative management was individually established on the 136
t1:5 Optimal responders (n. 63)a
R

basis of histological findings on surgical specimen, patient age and gen- 137
t1:6 No further treatment 36 4 (11.1%) 4 (11.1%)
t1:7 Chemotherapy 21 1 (4.7%) 1 (4.7%) eral conditions, and standard therapeutic strategy of each center, after 138
t1:8 an exhaustive discussion with the patient herself by a multidisciplinary 139
R

EBRT ± BRT 6 2 (33.3%) 2 (33.3%)

t1:9 team. 140
t1:10 Suboptimal responders, intra-cervical residual disease (n. 124)a
t1:11 No further treatment 47 3 (6.3%) 1 (2.1%) 4 (8.5%) The patients were periodically followed-up with clinical and radio- 141
O

t1:12 Chemotherapy 34 4 (11.8%) 4 (11.8%) logical examinations until they died or until June 2012. The median 142
t1:13 CCRT ± BCT 20 4 (20.0%) 2 (10.0%) 1 (5.0%) 7 (35.0%) follow-up of survivors was 66.5 months (range, 8–212 months). 143
t1:14 EBRT ± BCT 23 2 (8.7%) 1 (4.3%) 3 (13.0%)
C

t1:15
t1:16 Sub-optimal responders, extra-cervical residual disease with positive nodes (n = 75) Statistical methods 144
t1:17 No further treatment 10 1 (10.0%) 1 (10.0%)
t1:18 Chemotherapy 4 1 (25.0%) 1 (25.0%)
N

The statistical package SAS, release 6.7, was used for computations. 145
t1:19 CCRT ± BCT 34 5 (14.7%) 6 (17.6%) 1 (2.9%) 12 (35.4%)
t1:20 EBRT ± BCT 27 9 (33.3%) 3 (11.1%) 12 (44.4%)
The time from the first cycle of NACT to the detection of recurrence 146
t1:21 was defined as recurrence-free survival. The time from the first cycle 147
U

t1:22 Sub-optimal responders, extra-cervical residual disease with negative nodes (n = 18) of NACT to death or last observation was defined as overall survival. 148
t1:23 No further treatment 1
The cumulative probability of recurrence-free survival and overall 149
t1:24 Chemotherapy 4 1 (25.0%) 1 (25.0%)
t1:25 CCRT ± BCT 7 1 (14.2%) 1 (14.2%) 2 (28.5%) 4 (57.1%) survival was estimated by the product-limit method. The log-rank test 150
t1:26 EBRT ± BCT 6 was used to compare the homogeneity of recurrence-free survival and 151
t1:27 survival functions across strata defined by categories of prognostic 152
t1:28 No responders (n. 49)
t1:29 No further treatment 7 2 (28.6%) 2 (28.6%) 4 (57.1%) variables. 153
t1:30 Chemotherapy 3 A multiple regression analysis based on the Cox proportional hazard 154
t1:31 CCRT ± BCT 25 7 (28.0%) 3 (12.0%) 2 (8.0%) 12 (48.1%) model was used to jointly test the relative importance of variables as 155
t1:32 EBRT ± BCT 14 5 (35.7%) 1 (7.1%) 6 (42.8%)
predictors of recurrence-free survival and overall survival. 156
t1:33 P, pelvic; EX, extrapelvic; EBRT, external beam irradiation; BCT, brachytherapy; CCRT,
t1:34 concurrent chemoradiation. Results 157
a
The present analysis did not include the patient who had adjuvant chemotherapy
followed by EBRT (sub-optimal responder with intra-cervical disease) and 3 patients
who had BCT alone (one optimal responder and two sub-optimal responders with intra- FIGO stage was Ib2 in 179 patients, IIa in 44, and IIb in 110 subjects. 158
t1:35 cervical disease). Histological types were squamous cell carcinoma in 268 patients and 159

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx 3

a) Recurrence-free survival
100

90

80
Estimated survival function
70

60 no therapy (pts 36, median not

reached9
50
Chemotherapy (pts 21, median
40 not reached)

F
Radiotherapy (pts 6, median not
30
reached)

O
20 Log-rank test: 4,3158, p value= 0,1156

10

O
0
0 19.1 31.5 43.4 52.6 60.3 120

R
months

P
b) Overall survival
100 D
90
Estimated survival function

80
E

70
T

60
C

50

40
E

Log-Rank test: 2.8557 ,p value=0.2398

30

20
R

10
R

0
0 29.1 33.9 48.6 58.6 75.5
Months
O

no therapy (pts 36, median not reached) Chemotherapy (pts 21, median not reached)
C

Radiotherapy (pts 6, median not reached

Fig. 1. Recurrence-free survival (a) and overall survival (b) in overall optimal responders by post-operative treatment.
N

160 adenocarcinoma or adenosquamous carcinoma in 65. NACT regimen external pelvic irradiation [EBRT] with or without brachytherapy 175
U

161 was platinum- and paclitaxel-based in 284 patients and platinum- (BCT), 88 (26.4%) received cisplatin-based CCRT on the pelvis with or 176
162 based in 49. Type of radical hysterectomy according to the Piver– without BCT (three of them with additional irradiation on the aortic 177
163 Rutledge classification was type II in 88 patients and type III in the area), three (0.9%) had BCT alone, 1 patient (0.3%) underwent two 178
164 remaining 225. All patients underwent systematic pelvic lymphadenec- cycles of adjuvant chemotherapy with the induction regimen followed 179
165 tomy. Aortic node dissection was performed in selected cases of patients by EBRT, and 101 patients (30.3%) had no further treatment. 180
166 with positive pelvic nodes at frozen sections. At the time of the present analysis, 79 (23.7%) out of the 333 patients 181
167 Pathological response was complete in 30 patients, optimal partial in relapsed after a median time of 14.9 months (range, 4.5–123 months). 182
168 34, and suboptimal in 220, whereas no response was found in 49. As far Recurrent disease was pelvic in 50 (63.3%) cases, extra-pelvic (aortic 183
169 as sub-optimal responders are concerned, 127 patients had intra- or distant) in 22 (27.8%), and both pelvic and extra-pelvic in 7 (8.9%). 184
170 cervical residual disease, 75 had extra-cervical residual disease with According to pathological response, tumor relapsed in 7 (10.9%) out 185
171 positive nodes, and 18 had extra-cervical residual disease with negative of the 64 overall optimal responders (6 pelvic recurrences and 1 extra- 186
172 nodes. pelvic recurrence) and 50 (22.7%) out of the 220 sub-optimal responders 187
173 After surgery, 66 patients (19.8%) received two cycles of adjuvant (30 pelvic recurrences, 15 extra-pelvic recurrences, and 5 pelvic plus 188
174 chemotherapy with the induction regimen, 74 (22.2%) underwent extra-pelvic recurrences). Among the sub-optimal responders, tumor 189

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 4 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx

a) Recurrence-free survival
100

90

Estimated survival function 80

70
no therapy (pts 47, median not
60
reached)
50 Chemotherapy (pts 34, median
not reached)
40
Radiotherapy (pts 43, median not
30
reached)

F
20 Log-Rank test: 6.1094, p value:0.047

10

O
0
0 9.8 14.9 19.1 28.9 33.2 43.4 52.6 60.3 91.9

O
Months

b) Overall survival

R
100

P
90

80
D
Estimated survival function

70
E

60 no therapy (pts 47, median not

reached)
T

50 c hemotherapy (pts 34, median

not reached)
C

40
radiotherapy (pts 43, median not
reached)
E

30
Log-rank test 9.3457, p value= 0.0093
20
R

10
R

0
0 11.9 18.8 22.5 31.5 41.5 52.4 59.4 75.5
O

Months

Fig. 2. Recurrence-free survival (a) and overall survival (b) in sub-optimal responders with intra-cervical residual disease by postoperative treatment.
C
N

190 relapsed in 19 (15.0%) out the 127 patients with intra-cervical residual disease). The former patient developed a distant failure, whereas none 207
191 disease (13 pelvic recurrences, 5 extra-pelvic recurrences, and 1 pelvic of the latter ones relapsed. 208
192 plus extra-pelvic recurrence), 26 (34.7%) out of the 75 patients who The risk of recurrence in overall optimal responders was the same in 209
U

193 had extra-cervical residual disease with positive nodes (15 pelvic re- patients who received any adjuvant treatment (3 out of the 27, 11.1%) 210
194 currences, 9 extra-pelvic recurrences, and 2 pelvic plus extra-pelvic and in those who did not (4 out of the 36, 11.1%). 211
195 recurrences), and 5 (27.8%) out of the 18 patients who had extra- Recurrence-free survival and overall survival of overall optimal 212
196 cervical residual disease with negative nodes (2 pelvic recurrences, 1 responders were similar in patients who had adjuvant chemotherapy, 213
197 extra-pelvic recurrence, and 2 pelvic plus extra-pelvic recurrences). adjuvant CCRT or EBRT ± BCT, or no adjuvant treatment (Fig. 1a and b). 214
198 Twenty-two (44.9%) out of the 49 non-responders recurred (14 pelvic Among the sub-optimal responders with intra-cervical residual dis- 215
199 recurrences, 6 extra-pelvic recurrences, and 2 pelvic plus extra-pelvic ease, a distant recurrence (with or without pelvic failure) occurred in 216
200 recurrences). none of the 34 patients who had adjuvant chemotherapy versus 4 out 217
201 Table 1 showed the patterns of recurrences according to postopera- of the 43 (9.3%) patients who underwent adjuvant CCRT or EBRT ± 218
202 tive treatment. BCT. Recurrence-free survival and overall survival of sub-optimal re- 219
203 The present analysis did not include the patient who had adjuvant sponders with intra-cervical residual disease were significantly better 220
204 chemotherapy followed by EBRT (sub-optimal responder with intra- in patients who received adjuvant chemotherapy (p = 0.047 and 221
205 cervical disease) and the 3 patients who had BCT alone (one overall op- p = 0.0093, respectively) (Fig. 2a and b) Among the sub-optimal re- 222
206 timal responder and two sub-optimal responders with intra-cervical sponders who had extra-cervical residual disease with either positive 223

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx 5

a) Recurrence-free survival
100

90

Estimated survival function

80

70

60

50

40

30 Log-rank test: 3.3913, p= 0.1835

F
20

10

O
0
0 6.5 9.7 11.1 16.1 19.1 23.5 30.5 33.9 46.1 54.1 62.7 91.9

O
Months
no therapy (pts 11, median 21.8 months)

R
Chemotherapy (pts 8, median 77.9 months)
Radiotherapy (pts 74 median not reached)

P
b) Overall survival
100

90
D
80
Estimated survival function

E
70
T

60

50
C

40
E

30 Log-Rank 5.1102 ; p= 0.0777

20
R

10
R

0
0 10 14.6 20 27.3 31.2 34.9 47.5 54.2 63.8 120
O

Months

no therapy (pts 11, median 21.8 months)

C

chemotherapy (pts 8, median not reached)

Radiotherapy (pts 74, median not reached)
N

Fig. 3. Recurrence-free survival (a) and overall survival (b) in sub-optimal responders with extra-cervical residual disease (with either positive or negative nodes) by post-operative
treatment.
U

224 or negative nodes, a distant recurrence (with or without pelvic failure) chemotherapy compared to those who did not (p = 0.0001) (Table 2). 234
225 occurred in none of the 8 patients who had adjuvant chemotherapy ver- Overall survival was significantly longer in patients with FIGO stage Ib2– 235
226 sus 13 out of the 74 (17.5%) patients who underwent postoperative IIa than in those with stage IIb disease (p = 0.0072), in patients who 236
227 CCRT or EBRT ± BCT. Recurrence-free survival and overall survival of received platinum/paclitaxel-based NACT than in those who received 237
228 sub-optimal responders who had extra-cervical residual disease with platinum-based NACT (p = 0.0464), in patients who achieved an over- 238
229 either positive or negative nodes were not significantly different accord- all optimal response than in those who did not (p b 0.0001), and in 239
230 ing to post-operative management (Fig. 3a and b). patients who received adjuvant chemotherapy compared to those 240
231 On the whole series, recurrence-free survival was significantly lon- who did not (p b 0.0001) (Table 3). 241
232 ger in patients who achieved an overall optimal response than in On multivariate analysis, consolidation therapy (p = 0.0012) was 242
233 those who did not (p b 0.0001), and in patients who received adjuvant the only independent prognostic variable for recurrence-free survival; 243

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 6 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx

t2:1 Table 2 received 3 cycles of cisplatin (100 mg/m2 day 1) + bleomycin 262
t2:2 Variables predictive of recurrence-free survival and overall survival by univariate analysis. (15 mg days 1 and 8) + methotrexate (300 mg/m2 day 8 with 263
t2:3 Variables Pts 5-year RFS p value 5-year OS p value leucovorin rescue) every 3 weeks prior to radical hysterectomy, postop- 264
erative treatment consisted of two additional cycles of the same chemo- 265
t2:4 Age
t2:5 ≤46 years 177 73.9% 0.2045 77.5% 0.2471 therapy in patients with histologically proven positive nodes regardless 266
t2:6 N46 years 156 66.4% 73.4% of surgical resection margin status, EBRT in those with positive surgical 267
t2:7 margins and negative nodes, and no further treatment in those with a 268
t2:8 Histology
t2:9 Squamous cell 268 70.1% 0.9070 76.3% 0.4689 pathological complete response or in those with negative surgical mar- 269
t2:10 Adenocarcinoma 65 71.5% 71.1% gins and negative nodes. 270
t2:11 Giaroli et al. [16], who employed cisplatin (50 mg/m2 day 1) + vin- 271
t2:12 FIGO stage
t2:13 Ib–IIa 223 72.7% 0.0755 78.6% 0.0072 cristine (1 mg/m2 day 1) + bleomycin (25 mg/m2 day 1–3) every 272
t2:14 IIb 110 65.7% 68.9% 10 days for 3 cycles as NACT regimen, reported that most of their 169 273
t2:15 patients underwent postoperative pelvic EBRT, whereas 7 women 274
t2:16 CT regimen
t2:17 Platinum/TAX 284 71.6% 0.1492 77.0% 0.0464 only with para-aortic metastases received adjuvant chemotherapy 275

F
t2:18 Platinum-based 49 63.3% 67.3% with cisplatin + methotrexate + vinblastine + cyclophosphamide. 276
t2:19 In an Italian multicenter study comparing cisplatin-based NACT 277
t2:20 Pathological response

O
t2:21 Optimal 64 87.4% b0.0001 96.0% b0.0001 followed by radical hysterectomy versus radiotherapy in 441 women 278
t2:22 Intracervical residual disease 127 78.5% 81.6% with stage Ib2–III cervical carcinoma, patients enrolled in the chemo- 279
t2:23 Extracervical residual diseasea 93 59.4% 63.8% surgical arm received EBRT if surgical margins were positive and no 280

O
t2:24 No response 49 47.5% 53.7%
additional treatment if both surgical margins and lymph nodes were 281
t2:25
t2:26 Consolidation therapyb negative or no residual tumor was detected on surgical sample [9]. 282
t2:27 No therapy 101 77.9% 0.0001 84.2% b0.0001 In the case of nodal involvement, the choice of adjuvant treatment 283

R
t2:28 CCRT or EBRT ± BCT 162 59.2% 62.7%
was based on the institution's policy (i.e., chemotherapy, EBRT, or no 284
t2:29 Chemotherapy 66 85.5% 91.6%
further therapy). About one third of the failures showed a distant com- 285

P
t2:30 Pts, patients; RFS, recurrence-free survival; OS, overall survival; TAX, paclitaxel; CCRT, ponent with no significant difference between chemo-surgery arm and 286
t2:31 concurrent chemoradiation; EBRT, external beam irradiation; BCT, brachytherapy.
a radiotherapy arm with regard to the pattern of recurrence [9]. These re- 287
t2:32 With either positive or negative nodes.
b
The patient who had chemotherapy followed by EBRT and the 3 patients who had BCT sults, as well as the similar data reported by Sardi et al. [8] appeared to
D 288
t2:33 alone were not included in this analysis. suggest that the relatively short duration of NACT might be not enough 289
to control distant micro-metastases. 290
In the SNAP01 trial, comparing NACT with ifosfamide (5 g/m2 24- 291
E
hour infusion) + cisplatin (75 mg/m2) (IP regimen) versus ifosfamide 292
244 whereas FIGO stage (p = 0.0169) and consolidation therapy (p = (5 g/m2 24-hour infusion) day 1 + paclitaxel (175 mg/m2 day 2) + 293
T

245 0.0016) were independent prognostic variables for overall survival cisplatin (75 mg/m2 day 2) (TIP regimen) every 3 weeks for three 294
246 (Table 4). cycles, women who achieved an optimal pathological response were 295
C

scheduled to receive two additional cycles of chemotherapy after sur- 296

247 Discussion gery with the same NACT regimen, whereas those women with positive 297
nodes, parametrial involvement, cut-through or suboptimal response 298
E

248 NACT followed by radical hysterectomy is an interesting alternative were candidates for EBRT or CCRT. Tumor relapsed in 30 out of the 299
249 therapeutic option to CCRT for patients with stage Ib2–IIb cervical 108 patients (27.8%) of IP arm and 25 out of the 96 (26.0%) of TIP arm, 300
R

250 cancer [7–17]. The achievement of an optimal pathological response and recurrent disease was local and distant ± local in 17 (56.7%) and 301
251 on surgical specimen is a strong predictor of a better clinical outcome 12 (40.0%), respectively, of the former (missing in 1), and in 14 302
(56.0%) and 11 (44.0%), respectively, of the latter. In the SNAP02 trial, 303
R

252 [7,11,15–17]. In the present investigation, patients who did not obtain
253 an overall optimal response had a 2.259-fold higher risk of recurrence comparing NACT with TIP versus paclitaxel (175 mg/m2) + cisplatin 304
254 and a 5.392-fold higher risk of death than those who obtained an overall (75 mg/m2) (TP) every 3 weeks for three cycles, the criteria for postop- 305
O

255 optimal response. erative treatment were the same as SNAP1 trial [13]. Disease recurred in 306
256 A few papers appear to suggest a survival advantage for cervical car- 27 out of the 80 patients (33.8%) of TP arm and 20 out of the 74 (27.0%) 307
257 cinoma patients who received adjuvant chemotherapy after definitive of TIP arm, and failure site was local and distant ± local in 10 (37.0%) 308
C

258 CCRT [3–5]. Conversely, very few data are currently available as for ad- and 9 (33.3%), respectively, of the former (missing in 8), and in 5 309
259 juvant post-operative therapy in patients treated with NACT and radical (25.0%) and 9 (45.0%), respectively, of the latter (missing in 6). However 310
N

260 hysterectomy (7, 9, 11, 13, 16). In the early experience of Benedetti in both SNAP trials there were relevant discrepancies between the 311
261 Panici et al. [7], including 75 women with stage Ib2–III disease who scheduled postoperative regimen and the effectively used postoperative 312
treatment, and neither studies investigated the relationship between 313
U

postoperative treatment and the pattern of failures. 314

In the present study, tumor relapsed in 10.9% of overall optimal 315
t3:1 Table 3 responders, 22.7% of sub-optimal responders and 44.9% of non- 316
t3:2 Variables predictive of recurrence-free survival and overall survival by Cox proportional responders. Among the overall optimal responders, the risk of recur- 317
t3:3 hazard model.
rence, recurrence-free survival and overall survival was the same in 318
t3:4 Variable Parameter Standard Wald χ2 HR 95% CI p patients who had any adjuvant treatment and in those who had not, 319
estimated error value and recurrent disease was almost exclusively pelvic. Among the sub- 320
t3:5 Recurrence-free survival optimal responders, a distant recurrence (with or without pelvic fail- 321
t3:6 Consolidation 0.42579 0.13173 10.4474 1.531 1.182–1.982 0.0012 ure) occurred in none of the 42 patients who had adjuvant chemother- 322
therapy apy versus 17 (14.5%) out of the 117 who received postoperative CCRT 323
t3:7
t3:8 Overall survival or EBRT ± BCT, whereas pelvic failure developed in 6 (14.2%) of the 324
t3:9 Stage 0.55188 0.2308 5.7040 1.737 1.104–2.731 0.0169 former and 25 (21.3%) of the latter. It is noteworthy that recurrence- 325
t3:10 Consolidation 0.47512 0.15089 9.9152 1.608 1.196–2.162 0.0016 free survival and overall survival of sub-optimal responders with 326
therapy
intra-cervical residual disease were significantly better in patients 327

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010
 F. Landoni et al. / Gynecologic Oncology xxx (2013) xxx–xxx 7

328 who had adjuvant chemotherapy (p = 0.047 and p = 0.0093, respec- References 374
329 tively). Conversely, the clinical outcome of sub-optimal responders [1] Rojas-Espaillat LA, Rose PG. Management of locally advanced cervical cancer. Curr 375
Q2
330 who had extra-cervical residual disease with either positive or negative Opin Oncol 2005;17:485–92. 376
331 nodes was not significantly different according to post-operative man- [2] Chemoradiotherapy for Cervical Cancer Meta-analysis Collaboration (CCCMAC). Re- 377
ducing uncertainties about the effects of chemoradiotherapy for cervical cancer: indi- 378
332 agement. The small number of women who had no adjuvant treatment vidual patient data meta-analysis. Cochrane Database Syst Rev 2010;20:CD008285. 379
333 (n. 11) or who had adjuvant chemotherapy (n. 8) versus those who [3] Vale C, Tierney JF, Stewart L, et al, Chemoradiotherapy for Cervical Cancer 380
334 had adjuvant CCRT or EBRT (n = 74) does not allow to drawn any defin- Meta-analysis Collaboration. Reducing uncertainties about the effects of chemora- 381
diotherapy for cervical cancer: a systematic review and meta-analysis of individual 382
335 itive conclusion. It is possible to speculate that patients with unsatisfacto- patient data from 18 randomized trials. J Clin Oncol 2008;26:5802–12. 383
336 ry response to NACT have limited benefit from both further cycles of the [4] Dueñas-González A, Zarbá JJ, Patel F, et al. Phase III, open-label, randomized study 384
337 same induction regimen or the administration of CCRT or EBRT since comparing concurrent gemcitabine plus cisplatin and radiation followed by adju- 385
338 chemo-resistant tumors are often radio-resistant too. A phase III random- vant gemcitabine and cisplatin versus concurrent cisplatin and radiation in patients 386
with stage IIB to IVA carcinoma of the cervix. J Clin Oncol 2011;29:1678–85. 387
339 ized trial of the Gynecologic Oncology Group (GOG) has recently demon- [5] Abe A, Furumoto H, Nishimura M, et al. Adjuvant chemotherapy following concur- 388
340 strated that the addition of bevacizumab to chemotherapy (cisplatin rent chemoradiotherapy for uterine cervical cancer with lymphadenopathy. Oncol 389
341 50 mg/m2 + paclitaxel 135–175 mg/m2 or topotecan 0.75 mg/m2 day Lett 2012;3:571–6. 390
391

F
[6] Morice P, Rouanet P, Rey A, et al. Results of the GYNECO 02 study, an FNCLCC phase
342 1–3 + paclitaxel 175 mg/m2 day 1 every 3 weeks) is associated with a III trial comparing hysterectomy with no hysterectomy in patients with a (clinical 392
343 3.7-month increase in median overall survival (HR = 0.71, 97.6% and radiological) complete response after chemoradiation therapy for stage IB2 or 393

O
344 CI = 0.54–0.95, p = 0.0035) when compared to chemotherapy alone II cervical cancer. Oncologist 2012;17:64–71. 394
[7] Panici Pl Benedetti, Scambia G, Baiocchi G, et al. Neoadjuvant chemotherapy and 395
345 in patients with recurrent cervical cancer [18]. New adjuvant treatments, radical surgery in locally advanced cervical cancer. Prognostic factors for response 396
346 i.e. the combined use of bevacizumab plus chemotherapy, could be tested and survival. Cancer 1991;67:372–9. 397

O
347 in sub-optimal responders to NACT who have extra-cervical residual [8] Sardi JE, Giaroli A, Sananes C, et al. Long-term follow-up of the first randomized trial 398
using neoadjuvant chemotherapy in stage Ib squamous carcinoma of the cervix: the 399
348 disease. final results. Gynecol Oncol 1997;67:61–9. 400

R
349 The weaknesses of the investigation are represented by its retro- [9] Benedetti-Panici P, Greggi S, Colombo A, et al. Neoadjuvant chemotherapy and rad- 401
350 spective nature, by the lack of a centralized pathological review by a sin- ical surgery versus exclusive radiotherapy in locally advanced squamous cell cervical 402
cancer: results from the Italian multicenter randomized study. J Clin Oncol 403
351 gle pathologist with expertise in gynecologic pathology, and by the lack 404

P
2002;20:179–88.
352 of standardization in the postoperative treatment. Moreover it is diffi- [10] Neoadjuvant Chemotherapy for Cervical Cancer Meta-Analysis Collaboration 405
353 cult to draw meaningful conclusions about the relative roles of adjuvant (NACCCMA). Neoadjuvant chemotherapy for locally advanced cervix cancer. 406
354 chemotherapy and radiotherapy when the criteria for choice of one ver- Cochrane Database Syst Rev 2004;2:CD001774. 407
[11] Buda A, Fossati R, Colombo N, et al. Randomized trial of neoadjuvant chemotherapy
D 408
355 sus the other are not a priori defined for the different types of re- comparing paclitaxel, ifosfamide, and cisplatin with ifosfamide and cisplatin follow- 409
356 sponders. However, the strengths of the study are represented by the ed by radical surgery in patients with locally advanced squamous cell cervical carci- 410
357 large number of patients and by the description of rate and pattern of noma: the SNAP01 (Studio Neo-AdjuvantePortio) Italian Collaborative Study. J Clin 411
E
Oncol 2005;23:4137–45. 412
358 recurrence according to pathological response and postoperative [12] Gonzalez-Martin A, Gonzalez-Cortijo L, Carballo N, et al. The current role of neoadju- 413
359 treatment. vant chemotherapy in the management of cervical carcinoma. Gynecol Oncol 414
T

360 The results of the present retrospective analysis appear to sug- 2008;110(3 Suppl 2):S36–40. 415
[13] Lissoni AA, Colombo N, Pellegrino A, et al. A phase II, randomized trial of neo- 416
361 gest that chemo-surgical approach is an effective therapeutic option adjuvant chemotherapy comparing a three-drug combination of paclitaxel, 417
C

362 for FIGO stage Ib2–IIb cervical cancer, even though data from ran- ifosfamide, and cisplatin (TIP) versus paclitaxel and cisplatin (TP) followed by rad- 418
363 domized clinical trials are not available yet to compare such a treat- ical surgery in patients with locally advanced squamous cell cervical carcinoma: 419
the Snap-02 Italian Collaborative Study. Ann Oncol 2009;20:660–5. 420
364 ment modality to CCRT as standard of care for locally advanced
E

[14] Cho YH, Kim DY, Kim JH, et al. Comparative study of neoadjuvant chemotherapy be- 421
365 cervical cancer [1–3]. Optimal responders do not need any further fore radical hysterectomy and radical surgery alone in stage IB2–IIA bulky cervical 422
366 treatment. Additional cycles of chemotherapy, as opposed to adju- cancer. J Gynecol Oncol 2009;20:22–7. 423
R

[15] Kim DS, Moon H, Kim KT, et al. Two-year survival: preoperative adjuvant chemo- 424
367 vant CCRT or EBRT, could be of benefit for patients with suboptimal therapy in the treatment of cervical cancer stages Ib and II with bulky tumor. Gyneco 425
368 response and intra-cervical residual disease. Both adjuvant chemo- Oncol 1989;33:225–30. 426
369 therapy and adjuvant CCRT or EBRT do not seem to improve the [16] Giaroli A, Sananes C, Sardi JE, et al. Lymph node metastases in carcinoma of the cer- 427
R

vix uteri: response to neoadjuvant chemotherapy and its impact on survival. 428
370 clinical outcome of patients with extra-cervical residual disease
Gynecol Oncol 1990;39:34–9. 429
371 compared to no further treatment. [17] Colombo N, Gabriele A, Lissoni A, et al. Neoadjuvant chemotherapy (NACT) in locally 430
O

advanced uterine cervical cancer (LAUCC): correlation between pathological re- 431
sponse and survival. Proc Am Soc Clin Oncol 1998;17:352a [abstr 1359]. 432
[18] Tewari KS, Sill M, Long HL, et al. Incorporation of bevacizumab in the treatment of 433
372 Conflict of interest statement 434
C

recurrent and metastatic cervical cancer: a phase III randomized trial of the Gyneco-
373 The authors declare that there are no conflicts of interest. logic Oncology Group. J Clin Oncol 2013;31 [Suppl.; abstr 3]. 435
436
N
U

Please cite this article as: Landoni F, et al, Is there a role for postoperative treatment in patients with stage Ib2–IIb cervical cancer treated with neo-
adjuvant chemotherapy and radical surgery?..., Gynecol Oncol (2013), http://dx.doi.org/10.1016/j.ygyno.2013.12.010