Neoadjuvant Chemotherapy Followed by Surgery/Chemoradiotherapy Versus Radical Chemoradiotherapy in Stage IB2, IIA2, IIB Locally Advanced Cervical Carcinoma: A Retrospective Multicenter Study of 388 Patients

Purpose: The present study aimed to compare the oncologic outcomes and side effects between neoadjuvant chemotherapy followed by surgery/chemoradiotherapy and radical chemoradiotherapy for locally advanced cervical carcinoma (LACC). Methods: We conducted a retrospective review of patients with LACC (IB2, IIA2 and IIB stages) from six hospitals between June 2007 and January 2017. Results: A total of 388 patients were included, in which 278 patients received radical chemoradiotherapy (Standard group), and 110 patients received neoadjuvant chemotherapy. 65 patients of the 110 received radical hysterectomy (Surgery group), and 45 received chemoradiotherapy (Neo-Ra group). The 5-year overall survival (OS) in Surgery group (92.5%) was similar to that of the Standard group (84.9%), but both groups had higher OS rates than Neo-Ra group (75.6%). The 5-year disease-free survival (DFS) and progression-free survival (PFS) showed no differences among the three groups, respectively. There were no signicant differences for grades 1-3 gastrointestinal and genitourinary toxicities among the three groups. No patient had grade 4 adverse effects. In multivariate analysis, tumor regression (CR vs. PR+SD+PD), pathological type (squamous cancer vs. non-squamous cancer) and lymph node metastasis (positive vs. negative) were considered as independent predictors of OS. Furthermore, besides above factors, the maximum diameter of tumors and adjuvant chemotherapy were also considered as signicant prognostic factors for DFS and PFS. Conclusion: These ndings showed that neoadjuvant chemotherapy followed by radical hysterectomy as a feasible and reliable therapy, resulting in encouraging oncologic outcomes and low side effects when compared to those obtained by standard chemoradiotherapy for IB2, IIA2, IIB stage cervical carcinoma patients. of bleomycin, cisplatin, and vincristine administration prior to radical surgery and/or radiotherapy. The results showed a positive impact on the survival rate in 30 women with FIGO stage IIB-IVA disease, concluding that the NACT regimen was ineffective in converting patients from inoperable to operable state. So, based on these studies, the doctors should make a discrete decision in choosing NACT for beyond IIB stage. There are many chemotherapy regimens available till date. However, the guidelines failed to identify the combination that offers the best result. All the regimens achieved comparable results with many combinations and with cisplatin as the main agent 25


Introduction
Cervical carcinoma is a public health issue globally, and it is currently the fourth most common cancer in women. About 80% of cervical cancers occur in individuals of low-and middle-income countries, where the health systems cannot provide well-known effective prevention network. Furthermore, cervical cancer remains the third main cause of death due to its malignancy in women worldwide 1,2 . Since 1999, cisplatin-based concurrent chemoradiotherapy (CCRT) is the standard primary treatment used for treating patients with locally advanced cervical carcinoma (LACC) with 3-and 5-year survival rates of 69% and 62% for IB to IVA stages, respectively 3 . The bene ts of CCRT have been evaluated by several randomized trials including the Morris et al. study, which reported conduction of radiotherapy with cisplatin and 5 uorouracil. The 5-year overall survival (OS) rate signi cantly reached to 67% when compared with 40% with radiotherapy alone. Furthermore, the rates of both distant metastases and locoregional recurrences were signi cantly higher among patients treated with radiotherapy alone 4 . Surgery, which is the primary mode of treatment for cervical carcinoma, is increasingly being used in China. For LACC FIGO stages IB2, IIA2 and IIB, the use of neoadjuvant chemotherapy (NACT) followed by radical surgery could be an alternative choice to chemoradiotherapy for treating LACC based on a 5-year OS rate range of 45-83% [5][6][7] . For NACT followed by CCRT approach in LACC, A meta-analysis study by conducted Tierney et al. included 2074 patients from 18 trials with NACT followed by local therapy (mainly radiotherapy), and the results revealed neither higher response rates nor longer survival rates 8 .
Narayan et al. have reported NACT followed by CCRT demonstrated a signi cantly superior disease-free survival (DFS) when compared to de nitive CCRT in bulky stage IB2 and locally advanced (stages II-IVA) in cervical carcinoma 9 . The INTERLACE is the only randomized phase 3 trial that evaluated NACT followed by CCRT versus CCRT alone and is still recruiting patients 10 .
In contrast, a direct comparison of NACT followed by surgery versus CCRT is still lacking as the standard treatment offered considerable e cacy. The purpose of this retrospective multicenter study was to compare the clinical outcomes and side effects of cervical cancer patients who underwent platinumbased NACT followed by radical hysterectomy or CCRT versus CCRT alone. Also the prognostic factors that have more effect for over a long period of time with regard to survival outcomes were analyzed using univariate and multivariate analyses.

Patient characteristics
Between June 2007 and January 2017, 388 patients with LACC who underwent treatment in 6 cancer centers were retrospectively analyzed. The patient and tumor characteristics are summarized in

Late toxicities
Acute toxicities during radiotherapy were well tolerated by entire group of patients. Hematological and gastrointestinal toxicities were most commonly observed adverse events. The most common side effect that occurs due to late toxicity was gastrointestinal toxicity, which occurred in 24.5% of patients. Most of them were grade 1 or 2 adverse events. The grade 1-3 gastrointestinal toxicities were seen in 23% cases in Surgery group, 18% cases in Neo-Ra group, and 26% cases in Standard group, respectively, showing no signi cant difference. No patient had grade 4 adverse events. Genitourinary toxicities were the second most common late side effects, which occurred in 14% of patients. Most of them were grade 1-2, and grade 1-3 genitourinary toxicities were seen in 11% cases in Surgery group, 4% cases in Neo-Ra group, and 16% cases in Standard group, respectively, which showed no signi cance either (Table 2). Other less common late side effects included vaginal dehiscence and dyspareunia, but are less observed.

Survival analysis
The 5-year OS was observed in 92.5% patients in Surgery group, and in 84.9% in Standard group, which showed no signi cant differences, but both groups had higher rates than that of 75.6% patients in Neo-Ra non-squamous cancer) and lymph node metastasis (positive vs. negative) were found as independent predictors of OS. Furthermore, besides the above factors, the maximum diameter of tumors and adjuvant chemotherapy were considered as signi cant prognostic factors for DFS and PFS (Table 3-5).

Discussion
The present study demonstrated that NACT followed by surgery resulted in similar outcomes in terms of 5-year OS, PFS and DFS rates compared with radical chemoradiotherapy for LACC. Subgroup analysis demonstrated that NACT has an excellent high conversion rate, which converted patients with IB2, IIA2 stage from inoperable to operable state, and their 5-year OS, DFS and PFS showed no signi cant differences among any groups. So, NACT followed by radical hysterectomy was considered feasible and reliable therapy, especially for IB2, IIA2 stage LACC patients, and also salvage radiotherapy remains to be effective even if the response of NACT was not satisfactory. But for patients with stage IIB, the NACT conversion rate of patients from inoperable to operable states was much lower than those with stage IB2, IIA2. The study revealed that only less than half of the patients bene tted from NACT with stage IIB, and unfortunately the remaining patients (Neo-Ra group) who cannot receive the surgery had the lowest OS. Actually, patients who did not receive surgery because of bad pathological response had a 7.4-fold higher risk of recurrence and a 4.7-fold higher risk of death than those who received surgery according to our study results. In the present study, no signi cant differences were observed in grade 1-3 late gastrointestinal genitourinary toxicities among the three different treatment groups. However, NACT followed by radical surgery have been used to reduce the long-lasting adverse effects of radical radiotherapy, because of adjuvant radiotherapy after surgery has the lower prescription irradiation dose compared with CCRT. NACT in LACC patients showed no meaningful improvement in the decrease of gastrointestinal and genitourinary late toxicities based on our studies.
The standard treatment for CCRT followed by brachytherapy for LACC acts as an essential independent treatment factor for improving the pelvic control 12,13 . Whereas NACT followed by radical hysterectomy is performed in some institutions of Europe and Asia. NACT before local therapy showed unexpected results in reducing tumor mass before de nitive therapy. The effect of NACT can be evaluated immediately. Most of the chemotherapy is administered after surgery when patients lacked a measurable disease. When chemotherapy precedes surgery, one might measure the primary tumor directly or evaluate the surgical specimen for histological evidence of response to therapy 14  Several studies have demonstrated that positive lymph nodes, large tumor burden, close surgical margins, advanced stage and parametrial in ltration are associated with a greater risk of local recurrence or death for LACC 26-28 , and the FIGO clinical stage has no prognostic signi cance. Previous FIGO staging system does not take into account the in uence of clinicopathological factors. In particular, numerous investigators have identi ed pelvic and/or para-aortic lymph node metastases as the most important in uential factor for survival 29,30 . In this scenario, our analysis con rmed the literature data that a statistically signi cant difference can be observed for lymph node metastasis. The FIGO Committee added IIIC1P and IIIC2P to FIGO 2018 staging criteria by taking into consideration the effect of lymph node metastasis on prognosis 31 . In this trial, the addition of adjuvant chemotherapy improved the DFS and PFS. Despite the proven e cacy of chemotherapy that is delivered concurrently with conventional radiation, the literature data on adjuvant treatments are still confusing and heterogeneous, lacking the results to make more di cult in selecting the appropriate therapy. A larger survival advantage occurs when adjuvant chemotherapy is administered after CCRT 32-34 . But for those who had adjuvant chemotherapy in chemo-surgical treatment, optimal responders after surgery for IB2-IIB stage required no further treatment. Additional cycles of chemotherapy could be of bene t for patients with suboptimal response and intra-cervical residual disease 35 .
However, there are some limitations to our study. Firstly, due to the retrospective nature of the study, there is a possibility of lack of some relevant clinical data, and the inclusion of patients treated for over a long period of time, which might in turn affect disease staging as well as supportive therapy. Another major limitation of this study was that the Neo-Ra group all included patients with selective resistance to chemotherapy, leading to worst oncologic outcomes. Some studies have obtained the good response to NACT, making subsequent local treatment (radiation or surgery) more effective 36 , while bad response of NACT provided evidence after platinum cross-resistance with radiotherapy, inducing the development of radioresistant cellular clones 18 .

Conclusion
The results of the present retrospective analysis showed that chemo-surgical approach as a feasible and reliable therapy, resulting in encouraging oncologic outcomes and low side effects when compared to those obtained by standard chemoradiotherapy for IB2, IIA2, IIB stage cervical carcinoma patients.