Although the prognosis of locally advanced cervical cancer (LACC) has improved dramatically, survival for those with stage ⅢB-ⅣA disease or lymph nodes metastasis (LNM) remains poor. If treated with conventional radiotherapy alone, approximately 46%78% stage ⅢB-ⅣA disease would relapse at 5 years, and the outcomes for those with LNM seemed to be worse. Even after concurrent chemoradiotherapy (CCRT) became the standard of care for LACC, nearly 50% of this subgroup of patients would develop treatment failure. Thus, this subgroup of patients was classified as “high-risk”.In recent decades, intensity-modulated radiotherapy (IMRT) has become the mainstream of treatment for patients with prostate, rectal, neck, and several other malignancies. Theoretically, its incorporation into the treatment of LACC might yield an improved loco-regional control as well, as it allows for an escalated prescription dose to target volumes while sparing normal tissues from excessive radiation. However, several concerns still exist, including prolonged treatment time, geographical target miss, organ motion and set-up errors, etc. Therefore, further evidence supporting its regular use in the treatment of LACC is warranted.Potential benefit of adjuvant chemotherapy following CCRT could firstly be implied in the reports published by Morris et al., who revealed that patients prescribed with more cycles of higher dose chemotherapy demonstrated a diminished risk of distant metastasis (DM). Nevertheless, investigations carried out thereafter were mainly phase Ⅱ trials composed of heterogeneous patients treated with various cytotoxic combinations, thus conflicting and ambiguous results were usually reported.We therefore initiated a non-randomized prospective phaseⅡ study to investigate whether the incorporation of IMRT and adjuvant paclitaxel plus cisplatin (TP) regimen, which had been revealed to be the most, or at least one of the most active cytotoxic combinations in the treatment of advanced or recurrent cervical cancer, would improve the treatment outcome of high-risk LACC patients. The primary endpoints of this study were disease-free survival (DFS) and overall survival (OS). The secondary endpoints included the patterns of failure and toxicity profiles.Patients were enrolled if they had biopsy proven stage ⅢA-ⅣA squamous cervical cancer or stage ⅡB disease with metastatic regional nodes. Tumor staging was defined according to the International Federation of Obstetrics and Gynecology (FIGO) system. Lymph nodes were classified as metastatic based on their radiographic findings (≥1.0cm in the short-axis dimension). Eligibility criteria also included: age≤70 years, Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤1, no previous history of chemotherapy or radiotherapy, sufficient bone marrow, adequate renal and hepatic functions. Patients with synchronous malignancies, distant metastases, known hypersensitivity to cisplatin or paclitaxel, or those with poorly controlled medical conditions would be excluded.Treatment schedule was outlined in figure 1. IMRT was delivered with dynamic multi-leaf collimators using 6MV photon beams. A 5mm slice thickness CT simulation was carried out in the supine position. A comfortably full bladder and empty rectum were required, and both oral and intravenous administrations of contrast agents were used.Planning was performed using the Pinnacle3 9.0 treatment planning systems (Philips Healthcare, Andover, MA, USA). Clinical target volume (CTV) was defined as the gross tumor plus areas potentially containing microscopic disease, generally consisting of a 1–2 cm margin around the cervix, uterus, parametria, presacral space, lymph drainage area, and superior third of the vagina. The common iliac, external iliac, internal iliac and obturator nodal volume was contoured based on the contrast-enhanced vessels with a 7mm circumferential margin. In patients with bulky pelvic LNs (>2cm in shortest axis) or involved common iliac nodes or beyond, it would be contoured to the level of renal arteries (extended field-MIRT, EF-IMRT); otherwise, it would be 2-3cm above the aortic bifurcation.The CTV was expanded by 5mm uniformly to create the planning target volume (PTV). Prescription for PTV ranged from 45.050.0Gy at 1.8Gy2.0Gy/fraction in 25 fractions. Involved nodes were contoured separately and were defined as GTV-nodes. A tailored margin of 3mm was added to GTV-nodes to generate PTV-nodes, which were treated with a simultaneous integrated boost (SIB) technique to a total dose of 50.0–65 Gy at 2.0- 2.6Gy/fraction in 25 fractions.Brachytherapy was initiated when 27.030.0 Gy of external beam was delivered to PTV. The aim of brachytherapy boost was to deliver cumulative EQD2 doses (combined external beam radiotherapy (EBRT) and brachytherapy delivered in 2Gy equivalent doses) of ≥80 Gy to point A for stage ⅡB-ⅢA disease and ≥90 Gy for stage ⅢB-ⅣA disease. Specifically, a total dose of 2835Gy high-dose- rate (HDR) brachytherapy was prescribed to point A in 45 weekly fractions using an iridium- 192 source. An additional fraction of 5-7Gy brachytherapy would be delivered if residual cervical tumor was suspected by pelvic examination or MRI.Chemotherapy consisted of 46 cycles of concurrent cisplatin infusions and 2 cycles of adjuvant TP regimen. Concurrent weekly intravenous cisplatin at 30mg/m2 was initiated on the first day of radiotherapy for over 1-hour during EBRT. Adjuvant chemotherapy was scheduled within 4 weeks after the completion of CCRT and repeated 3 weeks later. Paclitaxel 150 mg/m2 was given as a 3-hour infusion on day1, followed by cisplatin 35 mg/m2 with 1-hour infusion on day1-2 (70 mg/m2 in total). Discontinuation of chemotherapy was allowed in the event of grade 3-4 hematological or gastrointestinal toxicities. It would be resumed when patients’ absolute neutrophil count recovered to ≥1500/mm3 and their platelet count improved to ≥100,000/mm3; however, doses of all agents should be subsequently reduced by 20%.The patients were assessed for toxicities twice per week during treatment. Complications occurred within 90 days of the initiation of chemoradiation were classified as acute complications, whereas those occurred afterwards were classified as late complications. The severity of acute complications was classified according to the Common Terminology Criteria for Adverse Events (CTCAE) v4.0. Late complications were graded according to the Radiation Therapy Oncology Group (RTOG) Late Radiation Morbidity Scoring Scheme.Post-treatment response was assessed based on pelvic examination and pelvic MRI or CT 4 weeks after the completion of treatment schedule. Continued surveillance was conducted at 3-month intervals for 2 years, every 6 months during the next 3 years, and annually thereafter. At each follow-up visit, pelvic examination including Pap smear and HPV detection was routinely performed, whereas imaging including ultrasound, chest x-ray, CT, or MRI were prescribed at physician’s discretion.Patterns of failure were analyzed in terms of loco-regional recurrence (LRR) and distant metastasis (DM). LRR was defined as persistent disease or any recurrence in cervix, uterus, vagina, adjacent pelvic structures, or regional lymph nodes including pelvic or para-aortic LNs. DM was defined as recurrence occurred in non-regional LNs or visceral metastases.Between July 2010 and January 2013, 52 eligible patients were enrolled. All the 52 patients completed intended radiation schedule. A median of 50Gy was prescribed to the PTV (range, 45-50Gy), whereas PTV-nodes were boosted to a median of 60Gy (range, 50.0-65Gy). EF-IMRT was administered in 22 patients, including 10 with positive PALNs and 12 treated prophylactically.A median of 5 fractions (range, 4-6) of brachytherapy boost were prescribed. The median cumulative EQD2 doses prescribed to point A was 93.8 Gy (range, 83.993.8Gy) for patients with stage IIB disease and 99.6 Gy (range,93.8109.5Gy) for those with ⅢB-ⅣA disease, respectively.The median overall treatment time (OTT), including the duration of EBRT and brachytherapy, was 46.5 days (range, 42-66 days). Radiotherapy interruption exceeding 3 days was documented in 5 patients, among which 3 were related to delayed recovery of myelosuppression and 2 were related to acute gastrointestinal toxicities.A total of 257 cycles of concurrent cisplatin were administered with a median of 5 cycles (range, 1-6 cycles), and the majority (79%) of patients completed 5-6 cycles. Two patients discontinued concurrent cisplatin within 2 cycles, one related to grade 3 vomiting at first cycle and the other related to grade 3 myelosuppression occurred at 2nd cycle of chemotherapy.A total of 50 patients received adjuvant chemotherapy, while above mentioned 2 patients who discontinued concurrent chemotherapy declined further medical intervention and withdrew from the trial. Thirty-nine (78%) patients completed planned 2 cycles of adjuvant chemotherapy. Two patients received only 1 cycle due to grade 3 thrombocytopenia. As for the remaining 9 patients, serum squamous cell carcinoma antigen (SCCA) levels did not return to normal until 3rd cycle was administered. Median interval between the completion of CCRT and the initiation of adjuvant chemotherapy was 26 days (range, 2335 days).Disease-free survival (DFS), loco-regional recurrence free survival (LRRFS), distant metastasis free survival (DMFS), and overall survival (OS) was defined from the time of diagnosis to the time of first evidence of relapse or death from any cause. Patients without documented evidence of recurrence were censored at the date of last follow up visit. Cumulative survival rate was calculated with the Kaplan-Meier method using SPSS ver. 17.0 (SPSS Inc., Chicago, IL, USA). Toxicities are reported as counts with percentages.