This article is mentioned in:

Introduction

Despite the availability of effective screening methods, cervical cancer remains the first or second leading cause of cancer-related mortality among women in developing countries (1). For patients with International Federation of Gynaecology and Obstetrics (FIGO) stage IB/IIA cervical cancer, radical surgery (RS) or radiotherapy, with or without concurrent chemotherapy, may be the primary treatment option. However, patients with a lesion >4 cm in greatest dimension exhibit a poorer prognosis compared to those with a tumor ≤4 cm in greatest dimension, regardless of treatment selection (2). Therefore, the optimal management for patients with stage IB2/IIA2 cervical cancer remains controversial.

Neoadjuvant chemotherapy (NAC) prior to RS has been widely used in bulky cervical cancer since the mid-1980s, with the purpose of reducing tumor size, lowering the difficulty of operation, reducing the risk factors of recurrence and, thus, improving prognosis. However, NAC has not been considered to be the standard treatment option, since different studies reported conflicting results. Although neoadjuvant cisplatin and 5-flourouracil (NAPF) treatment prior to RS was previously compared to primary surgical treatment (PST) in patients with cervical cancer (3), to the best of our knowledge, it has not been investigated in patients with early-stage bulky disease only. Based on these facts, we performed this study to evaluate the short- and long-term efficacy of NAPF in patients with FIGO stage IB2/IIA2 cervical cancer.

Materials and methods

Study population

A database of 2,284 cervical cancer patients who were admitted to Peking Union Medical College Hospital (Beijing, China) between January, 2002 and December, 2011 was reviewed. Approval by the Institutional Review Board of Peking Union Medical College Hospital was obtained in advance and the requirement of patient informed consent was waived, since this study was retrospective. Patients with primary, previously untreated, histologically confirmed squamous cell carcinoma or adenocarcinoma of the cervix, with FIGO stage IB2/IIA2 disease, were considered eligible for inclusion this study. The patients were also required to have a performance status of 0 or 1, normal organ function and no complicating diseases that would affect survival. Patients who had received primary radiotherapy were excluded.

Treatment

Neoadjuvant chemotherapy consisted of 2–3 cycles of cisplatin (70 mg/m2 on day 1) with 5-flourouracil (1,000 mg/m2 for 4 consecutive days) every 3 weeks. The response to NAPF was evaluated by tumor size measured at initial diagnosis and immediately prior to surgery, according to the Response Evaluation Criteria in Solid Tumors (RECIST) (4). Complete response (CR) was defined as no visible lesion, partial response (PR) as ≥30% decrease of the greatest dimension, progressive disease (PD) as ≥20% increase of the greatest dimension or appearance of new lesions and stable disease (SD) as the status between PR and PD. Following NAPF, the patients were clinically reassessed and those with normal organ function and good performance status were identified as eligible for RS.

Type III radical hysterectomy plus bilateral pelvic lymphadenectomy was performed within 2 weeks after diagnosis in the PST group and within 3 weeks after the last administration of chemotherapy in the NAPF group. Para-aortic lymphadenectomy was also performed when there was a suspicion of metastasis. Furthermore, postoperative concurrent chemoradiation or radiotherapy alone was administered to patients with at least one high-risk factor (HRF) or at least two intermediate-risk factors (IRFs) identified in surgical specimens. HRFs included lymph node metastasis, parametrial invasion, or positive resection margin, whereas IRFs included large tumor size (≥4 cm), deep stromal invasion (≥1/2), lymphovascular space invasion, or poor differentiation.

Statistical analyses

Statistical analyses of categorical variables were performed using the Chi-square test, with continuity correction, or the Fisher’s exact test. The operation time and estimated blood loss were compared with the Student’s t-test. The Kaplan-Meier analysis with the log-rank test and the multivariate Cox’s proportional hazards analysis were used to evaluate the prognostic factors affecting survival. All data were analyzed with the SPSS 17.0 software package (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Results

Patient characteristics

A total of 195 patients were enrolled in this study, with 103 patients in the NAPF and 92 patients in the PST group. The patient characteristics are summarized in Table I. There were no significant differences in age at diagnosis, FIGO stage, histology and grade between the two groups.

Table I Patient characteristics.

Table I

Patient characteristics.

Characteristics	NAPF	PST	P-value
Age (years)
≥40	62	60	0.469
<40	41	32
FIGO stage
IB2	78	62	0.197
IIA2	25	30
Histology
Squamous cell carcinoma	91	82	0.863
Adenocarcinoma	12	10
Grade
I–II	48	47	0.532
III	55	45

[i] NAPF, neoadjuvant cisplatin and 5-flourouracil; PST, primary surgical treatment; FIGO, International Federation of Gynaecology and Obstetrics.

Clinical response to NAPF

The overall response rate to NAPF was 61.2%, including CR in 2 patients and PR in 61 patients. A total of 40 patients were assessed as having SD, whereas no PD was observed. The NAPF group was then subdivided into intravenous chemotherapy (IVC) and intra-arterial chemotherapy (IAC) groups and the response rates of the two subgroups were not found to differ significantly.

Surgical profiles and complications

Of the 103 patients in the NAPF group, 3 patients received concurrent chemoradiation due to the toxicity of chemotherapy or other reasons. A total of 2 patients in the NAPF and 2 patients in the PST group were diagnosed with inoperable disease during surgery and the radical procedure was abandoned. The remaining 188 patients underwent RS and were included in the following analysis.

The operative time between the two groups did not exhibit a statistically significant difference (194.2±33.8 vs. 192.3±33.5 min, respectively; P=0.678). Intra-operative complications (ureter or bladder injury) occurred in 7.1% of the patients in the NAPF and 3.4% of those in the PST group, although the difference was insignificant (P=0.402). However, the estimated blood loss in the NAPF group was significantly higher compared to that in the PST group (620.1±394.9 vs. 434.8±233.7 ml, respectively; P=0.000). Analyses comparing these aspects between the IVC and IAC groups were performed and no difference was observed.

Intermediate- or high-risk factors

The patients in the NAPF group exhibited a lower rate of each IRF, whereas a significant difference was only observed in tumor size (22.5 vs. 93.3%, P=0.000). The frequencies of the HRFs were similar in the two groups (Fig. 1). To identify whether responders had significantly lower rates of IRFs or HRFs, we subdivided the NAPF group into the subgroups of responders and non-responders. Following subdivision, a remarkable decrease in an additional IRF, deep stromal invasion, was detected in responders compared to non-responders (46.7 vs. 76.3%, respectively; P=0.004) and the PST group (46.7 vs. 70.0%, respectively; P=0.004), whereas the reduction of HRFs in responders was insignificant (Fig. 2).

Figure 1 Comparison of intermediate- and high-risk factors between neoadjuvant chemotherapy (NAC) with cisplatin and 5-flourouracil and primary surgical treatment (PST) groups in International Federation of Gynaecology and Obstetrics stage IB2/IIA2 cervical cancer (*P<0.01).

Figure 2 Comparison of intermediate- and high-risk factors among responders, non-responders and the primary surgical treatment (PST) group (*P<0.01).

There was no difference in the rate of adjuvant radiotherapy between the NAPF and PST groups (94.9 vs. 95.6%, respectively; P=1.000). The frequency of adjuvant radiotherapy tended to be lower in responders compared to that in non-responders, althought the difference was not statistically significant (93.3 vs. 97.4%, respectively; P=0.679).

Survival analysis

During this procedure, 14 patients were excluded due to loss of contact. The overall follow-up rate was 92.6% and the median follow-up time was 41 months (range, 6–114 months). As shown in Fig. 3A and B, the progression-free survival (PFS) was similar in the two groups. Similar results were observed regarding overall survival (OS). Although non-responders tended to be associated with worse outcomes compared to the other two groups, responders exhibited no improvement of PFS and OS compared to the PST group (Fig. 3C and D).

Figure 3 (A and B) Comparison of progression-free survival (PFS) and overall survival (OS) between the neoadjuvant chemotherapy (NAC) with cisplatin and 5-flourouracil and the primary surgical treatment (PST) groups. (C and D) Comparison of PFS and OS among responders, non-responders and the PST group.

Furthermore, a multivariate Cox’s proportional hazards analysis was performed, including the following variables: age at diagnosis, FIGO stage, histology, number of IRFs, number of HRFs and treatment selection (Table II). Only the presence of at least one HRF was found to be an independent predictor of poor PFS and OS (P=0.000 and 0.001, respectively).

Table II Multivariate Cox’s proportional hazards analysis for prognostic factors.

Table II

Multivariate Cox’s proportional hazards analysis for prognostic factors.

	Progression-free survival			Overall survival
Prognostic factors	Adjusted HR	95% CI	P-value	Adjusted HR	95% CI	P-value
Age (years)
<40	Reference			Reference
≥40	0.748	0.358–1.564	0.440	0.803	0.348–1.852	0.607
FIGO stage
IB2	Reference			Reference
IIA2	0.848	0.359–2.002	0.707	0.883	0.343–2.271	0.796
Histology
Squamous cell carcinoma	Reference			Reference
Adenocarcinoma	1.314	0.454–3.799	0.614	1.579	0.530–4.702	0.412
No. of IRFs
≤1	Reference			Reference
≥2	1.185	0.287–4.894	0.815	1.716	0.327–8.995	0.523
No. of HRFs
0	Reference			Reference
≥1	4.094	1.888–8.881	0.000	4.343	1.789–10.542	0.001
Treatment
PST	Reference			Reference
Responders	0.793	0.266–2.364	0.678	1.018	0.313–3.307	0.976
Non-responders	1.802	0.769–4.221	0.175	2.127	0.799–5.661	0.131

[i] HR, hazard ratio; CI, confidence interval; FIGO, International Federation of Gynaecology and Obstetrics; IRFs, intermediate-risk factors; HRFs, high-risk factors; PST, primary surgical treatment.

Recurrence

Of the remaining 174 cases, 27 patients developed recurrences (15 patients in the NAPF and 12 in the PST group). The recurrence rates of the two arms did not differ significantly (16.7 vs. 14.3%, respectively; P=0.665). Furthermore, there was no significant difference in the rate of distant recurrence (14.4 vs. 9.5%, respectively; P=0.319) or pelvic failure (2.2 vs. 4.8%, respectively; P=0.616) between the two groups.

Discussion

Although NAC prior to RS has been widely applied in patients with early-stage or locally-advanced cervical cancer since the mid-1980s, only 6 randomized controlled trials (RCTs) comparing NAC plus RS to RS alone have been conducted and reported controversial results (3,5–9). A total of 3 RCTs concluded that NAC decreased HRFs (3,5,6), whereas the other 3 RCTs reported the opposite (7–9). The Eddy et al GOG trial (7) and the Katsumata et al JCOG trial (8) demonstrated no survival benefit with NAC, whereas the remaining trials reported improved survival. In addition, inconsistent outcomes were also reported by non-RCTs (10–13). One matched-case comparison of NAC prior to surgery and PST in FIGO stage IB/IIA cervical cancer concluded that NAC prior to surgery conferred no survival benefit over PST and may lead to poor prognosis in FIGO stage IIA disease, despite the reduction in IRFs and adjuvant radiotherapy (10), whereas another matched-case study reported that NAC, compared to primary surgery, significantly decreased IRFs and increased the 5-year disease-free survival (DFS) and OS rates in patients with stage IB2 disease (11).

The factors affecting the efficacy of NAC in cervical cancer have not been determined. A Cochrane meta-analysis of the above-mentioned 6 RCTs demonstrated that total cisplatin dose, chemotherapy cycle length and FIGO stage could not sufficiently explain the differences among individual trials (14). However, the rate of postoperative radiotherapy within each of the individual trials ranged between 36.1 and 100% and this difference may contribute to the variation in the individual trial results (14). Since the rate of adjuvant radiotherapy was high in the present study, there is a possibility that the differences in PFS and OS between the two groups caused by NAPF are diminished by adjuvant radiotherapy. Furthermore, the failure of NAC in reducing HRFs, which were shown to be the independent prognostic factor for poor PFS and OS in the present study, may reflect the role of potential underlying factors. Recently, there has been increasing interest in the identification of biomarkers able to predict response to treatment and survival of patients with cervical cancer, such as the clusterin protein, which was reported by Watari et al (15) to be significantly associated with poor response to platinum-based NAC.

Although the improvement of survival in responders was of no statistical significance compared to that in non-responders and the PST group in our study, previously published studies reported better prognosis for responders. Hu et al (11) reported that the 5-year DFS and OS rates in responders were favorably increased in comparison to those in non-responders and the PST group. A retrospective study suggested that the response to NAC was an independent prognostic factor for prolonged survival in patients with bulky stage IB2/IIA2 cervical cancer (16). Whether the prognosis of NAC-responders is significantly better compared to that of patients receiving primary surgery requires further investigation.

Analyses of surgical profiles revealed that the estimated blood loss was significantly increased in the NAPF compared to that in the PST group. A possible explanation for this phenomenon is the occurrence of necrosis and subsequent angiogenesis following chemotherapy.

In this study, NAPF followed by RS was compared to PST in women with early-stage bulky cervical cancer. We observed that neoadjuvant chemotherapy with the NAPF regimen prior to RS did not improve prognosis, despite the reduction in tumor size or deep stromal invasion (only in responders). The strengths of this study included sole neoadjuvant regimen, homogeneous subjects and high follow-up rate. However, large RCTs may be required to compare NAC followed by RS to RS alone. Furthermore, comparison of NAC prior to RS and cisplatin-based chemoradiation is warranted, since the latter is currently the standard treatment for cervical cancer. To the best of our knowledge, three prospective randomized trials on this subject (EORTC55994, NCT00193739 and NCT01000415) have been initiated and the results are keenly awaited.

References