Cervical cancer is still the second most common malignancy and second most common cause of cancerrelated death in women worldwide. Approximately 15% of all cervical cancers occur in women under the age of 40. Traditional treatment for cervical cancer consists of radical surgery or radiotherapy; however, neither method spares fertility, and either can lead to psychosexual dysfunction and decreased quality of life. This issue is becoming more critical because an increasing number of cervical cancers are diagnosed in nulliparous patients of reproductive age. Over the past decade, the treatment of cervical cancer has placed an increased emphasis on fertility preservation, prompting the question of whether it is possible to preserve genital functions without increasing the risk of recurrence, affording patients the opportunity to bear children.
Neoadjuvant chemotherapy (NACT) prior to surgery is an alternative option that does not affect genital functions. Successful pregnancy following NACT demonstrates that NACT and adjuvant chemotherapy do not always affect fertility or eliminate the chance for pregnancy. The ability of NACT to reduce tumour volume and virtually sterilise micrometastases may be due to an uncompromised tumour blood supply and a population of chemosensitive tumour cells. Thus, NACT allows for a less-extensive dissection of the cervix instead of radical hysterectomy while preserving the effectiveness of the treatment.
Despite high response rates, the use of NACT for cervical cancer treatment still remains controversial. In part, this is due to the fact that patient responses to NACT were both positive and detrimental (increased or decreased survivals). However, the discrepancies in these findings are likely due to patient selection, NACT scheme and length of NACT treatment. NACT followed by surgery is regarded as an effective option and also provides opportunities of fertility-sparing treatment for patients with International Federation of Gynecology and Obstetrics (FIGO) stage IB-IIA tumours of >2 cm.However, various risk factors have been shown to contribute to disease recurrence and cancerrelated death following NACT treatment.It remains unknown whether the clinical and pathological risk factors were significantly decreased after NACT. Therefore, correlations between these risk factors, chemoresponsiveness and patient outcomes require further investigation.
In this study, we explored short/long-term outcomes, chemoresponsiveness and risk factors to identify beneficial characteristics that may help predict a positive response to NACT treatment in patients with FIGO stage IB1-IIB cervical cancer in order to establish selection criteria, especially for fertility preservation.
2. Materials and methods
Clinical data from 10,897 patients was recruited from the cervical cancer database v1.10 (http://clinicaltrials. gov; NCT01267851). This study was reviewed by the Ethics Committee of Huazhong University of Science and Technology, and informed consent was obtained from each patient.
The inclusion criteria were as follows: patients with FIGO stage IB1-IIB cervical cancer (squamous, adeno or adenosquamous and small cell carcinoma); patients treated with or without NACT followed by radical hysterectomy and bilateral pelvic lymphadenectomy; patients who did not receive primary radiotherapy, preoperative radiotherapy or concurrent chemoradiotherapy and patients with no existing complicating disease or prior malignant disease.
Three thousand eight hundred and ninety six patients with FIGO stage IB1-IIB were treated with or without NACT from January 2002 to December 2008. Generally, the mean tumour size and the percentage of patients with stage IIB in the NACT group were larger than that in the primary surgery treatment (PST) group (4.1 ± 1.3 versus 2.9 ± 1.3, P < 0.001; 31.8% versus 5.8%, P < 0.001). To eliminate bias, a matched casecontrol study (1:1) for age at diagnosis, FIGO stage, tumour size (cm) and histological type was designed to evaluate the efficacy of NACT (Fig. S1). All patients received radical hysterectomy with pelvic lymphadenectomy, and para-aortic lymphadenectomy was performed in patients with suspicious para-aortic lymph node metastasis.
2.2. NACT treatment
Cisplatin-based chemotherapy was the standard treatment for NACT in this study, and the NACT regimens used in our study were shown in Table S1. NACT was generally administered in 1–2 courses depending on tolerance and response (n = 648; 91.65%), and small amount of patients received additional 1–2 cycles (n = 16, 2.26%). The maximal duration was no more than 60 days to avoid a delay in curative treatment. For the fertility-sparing option, 1–2 courses of NACT were followed by less-extensive surgery and two or more courses of adjuvant chemotherapy.
2.3. Evaluation of short-term response
In accordance with the World Health Organization (WHO) criteria, the clinical response to NACT was evaluated at the initial diagnostic procedure and before surgery. A complete response (CR) was defined as the complete disappearance of all clinically detectable disease. A 50% or more decrease in tumour size constituted a partial response (PR). Stable disease (SD) was defined as less than 50% reduction in tumour size, and progressive disease (PD) was regarded as an increase of tumour volume or appearance of new lesions. NACT responders were defined as the patients with CR or PR, and NACT non-responders included patients with SD or PD. Pathological optimal response (pOR) was defined as complete disappearance of tumour(s) in the cervix with negative nodes (pCR) or residual disease with less than 3 mm stromal invasion (PR1). Pathological suboptimal response included persistent residual disease with more than 3 mm stromal invasion on the surgical specimen (PR2). Two pathologists participated in the pathological evaluation.
2.4. Evaluation of high-risk factors (HRFs) and intermediate-risk factors (IRFs)
HRFs and IRFs were evaluated after surgery. HRFs included positive lymph nodes, positive resection margin and parametrial invasion. IRFs included large tumour size (the longest diameter on a surgical specimen >4 cm) and lymphovascular space invasion (LVSI) (>1/3 stromal invasion).
2.5. Follow-up study
Follow-up examination for all patients was suggested to be every three months for the first year and every 6 months for the next four years. In our database, follow-up was not feasible for a small proportion of patients due to loss of contact, and these data were excluded from the survival analysis.
2.6. Statistical analysis
Age at diagnosis (year), tumour sizes (cm), operation duration (hour) and blood loss volume (ml) were compared in the NACT and PST groups with Student’s ttest. In the clinical and pathological response analysis, the association between NACT efficacy and several
clinical characteristics (FIGO stage, tumour size and histological types) were evaluated using the Pearson chi-squared test with linear-by-linear association. Other statistical analyses of frequency data were performed via the chi-squared test or Fisher exact test. Disease-free survival (DFS) and overall survival (OS) were estimated by Kaplan–Meier analysis. The log-rank test was used to compare survival curves.
Statistical analyses were performed by using SPSS 13.0 software package. Values are presented as mean ± standard deviation unless otherwise noted. Statistical tests were considered significant when P values were less than 0.05.
3.1. Patient characteristics
The stratified characteristics in the NACT group (n = 707) and the PST group (n = 707) were wellmatched (Table S2). The follow-up rate was 78.4% (1108/1414), and the median follow-up time was 38 months (range, 4–108).
3.2. Short-term responses to NACT
A clinical response occurred in 79.3% (541/682) of patients, including 12.6%, (86/682) CR and 66.7% (455/682) PR, and the pathological response was 14.9% (105/707) (Table S3). Moreover, better clinical response was observed in patients with earlier FIGO stage or smaller tumour size (P < 0.001; Table 1).
Reduced IRFs were clearly detected in the NACT group compared to the PST group (P < 0.01, respectively; Table 2), especially for patients with FIGO stage IB1, with a tumour size of 64 cm, and with squamous cell carcinoma (Table 2). However, a decrease in HRFs was not detected after NACT, even in subpopulations stratified by FIGO stage, tumour size or histological type (P > 0.05; Table 2). For patients with bulky tumours in FIGO stages IB2, IIA2 and IIB, tumour size was significantly decreased after NACT (P < 0.001; Table 2).
3.3. Long-term survival analysis
Matched case–control comparison (417/417) was performed to evaluate long-term survival, and patients without follow-up data and their matched cases were excluded. The 5-year DFS rate in the NACT group was significantly improved in comparison to the PST group (88.4% versus 83.1%; P = 0.021), whereas there was an increasing trend of 5-year OS rate (P = 0.075; Fig. 1A and B). Compared with the PST group, the mortality (7.4% versus 11.5%; P = 0.044) and recurrence (9.6% versus 15.3%; P = 0.012) rates were decreased in the NACT group.
To analyse the clinical response to NACT, patients in the NACT were divided into a clinical responder subgroup (n = 330) and a clinical non-responder subgroup (n = 76), while the response of 11 patients remained unknown. Compared with patients in the PST groups and the clinical non-responder subgroup, the clinical responder subgroup displayed favourably increased DFS and OS rates (P < 0.05; Fig. 1C and D). The poorest outcomes were observed in the clinical non-responder subgroup, with a 70% 5-year DFS rate and a 68.9% 5-year OS rate (P < 0.05; Fig. 1C and D).
Next, patients in the NACT and PST groups were further stratified by HRFs or IRFs, respectively. The 5-year DFS and OS rates of NACT responders, with or without HRFs, were significantly increased over those of the NACT non-responders (P < 0.01; Fig. S2C and D). Alternatively, neither the 5-year DFS rate nor the 5-year OS rate significantly differed among four subgroups stratified by IRFs (P > 0.05; Fig. S3A-B). The prognosis in NACT non-responders with<2 IRFs was the poorest of the four subgroups, with a 73.4% 5-year DFS rate and a 71.8% 5-year OS rate, both of which significantly differed from the NACT responders with <2 IRFs (P < 0.001; Fig. S3C and D).
In FIGO stage IB2, the 5-year DFS and OS rates were significantly increased in the NACT group compared to those of the PST group (P = 0.015 and 0.044; Fig. 2), whereas the mortality (4.1% versus 11.0%; P = 0.027) and recurrence (4.1% versus 12.3%; P = 0.011) rates, operation duration (mean value, 3.6 h versus 3.8 h; P < 0.05) and blood loss volume (mean value, 517 ml versus 559 ml; P = 0.145) were decreased. Moreover, the 5-year DFS and OS rates of clinical responders were significantly improved compared to those in the PST and clinical non-responder subgroups for patients with tumours of 2–5 cm in FIGO IB (P < 0.05, Fig. 3), whereas the mortality (3.9% versus 10.8% versus 17.2%; P = 0.015) and recurrence rates (5.2% versus 15.6% versus 17.2%; P = 0.006) were distinctly decreased.
3.4. Representative NACT cases followed by the fertilitysparing option
A 31-year-old G2P0A2 patient desiring fertility preservation was diagnosed with FIGO stage IA1 cervical cancer in 2007. Following a course of TP regimen NACT, a simple trachelectomy and laparoscopic pelvic lymphadenectomy were performed, and all lymph nodes were found to be negative (0/30). Subsequently, two courses of TP regimen chemotherapy were applied. The patient conceived naturally 19 months after surgery and delivered by caesarean section at 37+2 weeks. The newborn exhibited no evidence of abnormalities, with an Apgar score of 8/9, and was completely normal after 43 months. Another 28-year-old G2P0A2 woman was diagnosed with FIGO stage IA2 cervical cancer in 2005. After one cycle of CP regimen NACT, large colonization was undertaken, followed by two courses of TP regimen chemotherapy. The patient conceived naturally after 20 months and delivered a 3300-g female with an Apgar score of 8/9 via caesarean section at 39+1 weeks. Throughout the 67-month follow-up, the mother remained disease-free and the child was developing normally.
In this study, NACT displayed efficient short-term responses, whereas clinical and pathological responses increased along with a reduction in tumour size and FIGO stage. Also, the decrease observed in IRFs was found to be significant. Previously, lymph node metastasis (LNM) was accepted as the most important adverse prognostic factor for cervical cancer. However, the incidence of LNM, parametrial invasion and positive surgical margin was slightly lower after NACT in this study (P > 0.05). These observations suggest that HRFs are not predominantly influenced by NACT in the shortterm.
Generally, NACT treatment significantly improved the 5-year DFS rate, whereas the mortality and recurrence rates decreased. Further analysis revealed that the 5-year DFS and OS rates of NACT responders were increased, whereas those of NACT non-responders were dramatically worse. These observations may be directly correlated with chemoresponsiveness, operability and outcomes in cervical cancer. Thus, the clinical response to NACT may be a significant prognostic factor of survival, which provides meaningful and rapid information about the efficacy of treatment that can be used by the patient and clinician to follow, or renounce, a conservative strategy.
Several studies have explored NACT effectiveness, alone, in patients with HRFs, although no clear consensus has emerged.Here, we found that the 5-year DFS and OS rates greatly improve in patients with or without HRFs in the NACT responder subgroups compared to those in NACT non-responder subgroups. Similarly, 5-year DFS and OS rates of the NACT responders with<2 IRFs were also increased. These observations suggest that chemoresponsiveness prior to HRFs and IRFs is more important for NACT long-term outcomes. Additionally, the poorest outcome was observed in NACT non-responders with <2 IRFs subgroup but not in NACT non-responders with ≥2 IRFs subgroup. It was previously reported that NACT could make a small metastasis hard to detect by making a tumour smaller, resulting in poor outcomes of patients with fewer IRFs.Therefore, this harmful, concealing effect of NACT may negate the beneficial effects of NACT.
It is important to establish selective criteria for patients who would benefit most from NACT, especially young patients who desire to preserve fertility. Firstly, NACT is beneficial to a majority of clinical responders because of its high clinical response rate, even for patients with HRFs and IRFs. The short-term responses to NACT should be regarded as an important prognostic factor for survival. NACT non-responders are unsuitable for NACT followed by fertility-sparing options. Secondly, NACT enhances the long-term outcome for patients with bulky tumours, especially for FIGO IB2. In this study, the 5-year DFS and OS rates were distinctly increased, whereas the mortality and recurrence rates declined. At the same time, operation duration and blood volume were decreased because of the reduction in tumour size after NACT. Therefore, patients with FIGO IB2, especially NACT responders, greatly benefit from NACT. Furthermore, NACT improves operability for patients with bulky tumours in FIGO IIA2 and IIB by decreasing tumour size (Table 2), providing proper treatment for inoperable patients. Finally, appropriate candidates for fertilitysparing surgery generally include patients with FIGO stages IA1, IA2 and IB1 tumours of less than 2 cm only. In this study, the long-term outcomes of NACT clinical responders were significantly improved in IB tumours, 2–5 cm in size, whereas the mortality and recurrence rates were also distinctly decreased. According to the principle of NACT followed by fertility-sparing surgery plus adjuvant chemotherapy, it is possible to meet the more extensive requirements of patients with larger tumours. Adjuvant chemotherapy may be helpful because the harmful concealing effects of NACT could make pathological risk factors undetected after surgery in partial patients.
Although our investigation is still limited by a lack of prospective randomisation and multicentre verification, the current study suggests that NACT may be a successful treatment option for patients with invasive cervical cancer. Fertility preservation for young patients is definitely possible after induction NACT and fertility-sparing surgery, affording a better quality of life and preserving fertility without worsening the prognosis. Although very preliminary, our results open the way to a new concept in the management of invasive cervical cancer in young women desiring to preserve fertility.