Study design

This multi-center, prospective, open-label, phase III randomized controlled trial will be conducted at four medical institutions in China. The study evaluates two EBRT regimens combined with concurrent chemotherapy, followed by brachytherapy. In the standard treatment arm, CFRT delivers 45 Gy in 25 fractions, with or without a simultaneous regional lymph node boost to 60 Gy. The experimental arm utilizes MHRT, delivering 43.35 Gy in 17 fractions, with or without a regional lymph node boost to 54.40 Gy. Both regimens are administered at five fractions per week. To evaluate whether the 3-year progression-free survival (PFS) rate in the experimental group is non-inferior to that in the control group, this study adopts a non-inferiority design. Based on previous studies, a 3-year PFS rate of 70% is expected in both groups, with a non-inferiority margin set at 12% [16, 17]. According to the National Comprehensive Cancer Network (NCCN) guidelines for cervical cancer (Version 1.2025), concurrent immunotherapy is permitted during radiotherapy for patients classified as FIGO 2014 Stage IIIA, IIIB, or IVA, as well as select FIGO 2018 Stage III–IVA cases [3].

This study has been approved by the Ethics Committee of Peking Union Medical College Hospital (Approval No. K6852) and registered on ClinicalTrials.gov (NCT06641635).

EndpointsPrimary endpoints

The primary endpoint of this study is 3-year PFS, which is defined as the proportion of patients who remain free from disease progression or death at three years after randomization.

Secondary endpoints

3-month complete response rate (CRR), 3-year overall survival (OS), 3-year locoregional progression-free survival (LPFS), 3-year metastasis-free survival (MFS), 3-year cervical cancer-specific survival (CSS), tumor regression after EBRT, acute and late treatment-related toxicities, and quality of life (QoL).

Eligible criteriaInclusion criteria 1.

Patients should provide written informed consent voluntarily, at least 30 days prior to enrollment.

2.

Age between 18 years and 75 years at the time of diagnosis.

3.

Histologically confirmed cervical cancer, with histological subtypes limited to squamous carcinoma, adenocarcinoma or adenosquamous carcinoma.

4.

FIGO stage IB1 to IIIB cervical cancer; FIGO stage IIIC1 cervical cancers with a maximum metastatic lymph node diameter of less than 2 cm and no involvement of common iliac chain.

5.

Planned definitive chemoradiotherapy with EBRT and concurrent weekly cisplatin (± immunotherapy), followed by brachytherapy.

6.

Eastern Cooperative Oncology Group performance status of 0–1 and the ability to tolerate supine positioning for at least 30 min.

Exclusion criteria 1.

History of surgery for cervical cancer, excluding pelvic lymphadenectomy, lymph node dissection, or cervical conization.

2.

Prior history of abdominal or pelvic radiation.

3.

Pregnancy or lactation.

4.

Active infections and fever.

5.

Severe comorbidities and seropositive status that may significantly impact adherence to the clinical trial, including but not limited to unstable cardiovascular disease, renal dysfunction, chronic hepatitis, poorly controlled diabetes mellitus, psychiatric disorders and acquired immunodeficiency syndrome.

Randomization

After screening for eligibility, patients will be randomized into two groups (1:1) in balanced permuted blocks: the MHRT group (experimental group) or the CFRT group (control group). Stratification factors for randomization will include FIGO stage for cervical cancer (IB to IIB vs. IIIA to IIIC1) and immunotherapy status (yes vs.no) and a block size of 4 will be utilized. The randomization process will be carried out by statisticians from the School of Public Health at Soochow University, using SAS Proc Plan for generating the block randomization scheme. This study is an open-label trial for both investigators and participants.

RadiotherapyTarget volumes definition

The approach has been extensively described in previous studies [12]. Before simulation and treatment session, patients will be instructed to empty their bladder and rectum one hour and forty minutes prior to their appointment, and then to consume 450–500 ml of water within 10 min, adjusted based on their height and weight. All patients underwent a single CT simulation in the supine position with thermoplastic immobilization and received intravenous contrast administration when no contraindications were present.

Clinical target volumes (CTV) will be contoured separately. In the MHRT group, CTV delineation followed the guidelines for oART, comprising the clinical target volume of the uterus (CTV-U), cervix (CTV-C), and lymph nodes (CTV-N). CTV-U encompasses the entire uterine body, while CTV-C includes the vagina, uterine cervix, and parametria. CTV-N encompasses the pelvic lymphatic drainage regions, including the common, internal, and external iliac, obturator, and presacral lymph nodes. Gross tumor volume node (GTVnd) includes any involved pelvic lymph nodes with a diameter of ≥ 1 cm on computed tomography (CT) or magnetic resonance imaging (MRI), or those exhibiting positive uptake on positron emission tomography-computed tomography (PET-CT) [18]. PET-CT imaging is recommended when accessible, particularly for patients with equivocal or suspicious lymph nodes on CT or MRI, to aid in the identification of pelvic and para-aortic nodal metastasis.

A 5 mm margin will be added to the CTV-C, CTV-U, and CTV-N to create the corresponding planning clinical target volumes. A 3 mm margin will be applied to metastatic lymph nodes to define the planning gross tumor volume of node [12, 19]. The margin of EBRT may be adjusted for individual patients based on the first five fractions of treatment, taking CTV-U mobility into account. This margin strategy is tailored to leverage the advantages of oART over conventional non-adaptive radiotherapy.

In the CFRT group, the CTV includes the uterus, cervix, parametrium, upper vagina, and pelvic lymphatic drainage regions, encompassing the internal iliac, external iliac, obturator, presacral, and common iliac lymph nodes. The GTVnd includes pelvic lymph nodes with a short-axis diameter ≥ 1 cm on CT or MRI, or those exhibiting positive uptake on PET-CT [18]. The specific radiation target area may be adjusted based on factors such as positioning errors, image guidance techniques, and prior clinical experience at each center, such as positioning errors, image guidance techniques, and previous experience. The CTV should adequately account for the mobility of the uterus, cervix, and vagina to ensure proper coverage. A 15 mm margin is added around the uterus and cervix, and a 6–8 mm margin is applied in all other directions to delineate the planning target volume (PTV) [20, 21].

Based on the NCCN guideline recommendations, pelvic and extended field radiotherapy (EFRT) is indicated for patients with documented common iliac and/or para-aortic lymph node involvement [3]. In our study, patients with multiple pelvic lymph node metastases received standard pelvic-field radiotherapy for cervical cancer, as EFRT is not routinely required for this patient population.

Dose prescription

For MHRT group, the prescribed dose for planning clinical target volumes is 43.35 Gy in 17 fractions, once daily, 5 days a week. The prescribed dose for planning gross tumor volume of node is 54.4 Gy in 17 fractions with a simultaneous integrated boost. Assuming an α/β ratio of 10 Gy for tumor tissue and 3 Gy for late-responding normal tissue [22], the equivalent dose in 2-Gy fractions (EQD2) was calculated to be 45.34 Gy for MHRT and 59.84 Gy for the lymph node boost, both delivered over 25 fractions.

For CFRT group, the prescribed dose of PTV is 45 Gy in 25 fractions over five weeks and simultaneous or sequential lymph node boost, a total cumulative dose of 60 Gy in 25 fractions. The corresponding EQD2 values are 44.25 Gy for PTV and 62 Gy for the lymph node boost.

Treatment planning and dose constraints

In the MHRT group, patients receive daily oART. In the CFRT group, patients are treated with fixed-field intensity-modulated radiation therapy (IMRT), volumetric modulated arc therapy (VMAT), or tomotherapy. It is required that at least 95% of the PTV receives 100% of the prescribed dose, and less than 1% of the PTV receives more than 110% of the prescribed dose. The detailed dose constraints for treatment planning are provided in Table 1.

Table 1 Dose-volume constraints of OARsoART procedure

The oART procedure has been thoroughly described in previous studies [23]. For each treatment fraction, the first iterative cone beam computed tomography (iCBCT) was acquired after bladder and rectal preparation. Based on this scan, the CTVs and OARs were automatically contoured and subsequently reviewed and manually adjusted by a radiation oncologist. Both the adapted plan and the scheduled plan were then generated and evaluated according to target coverage and dose constraints, with the plan best meeting the clinical goals selected for treatment. After plan approval, a second iCBCT was performed to verify the positions of the CTVs and OARs. Treatment proceeded if intrafractional motion remained within acceptable limits; otherwise, additional iCBCT scans and re-contouring were conducted. The daily oART workflow is illustrated in Fig. 1.

Fig. 1

Workflow for daily online adaptive radiotherapy. CT, computed tomography; oART, online adaptive radiotherapy; MHRT, moderately hypofractionated radiotherapy; PCTV, planning clinical target volume; PGTVnd, planning gross tumor volume of node; iCBCT, iterative cone beam computed tomography; OARs, organs at risk

Brachytherapy

Following EBRT, all patients will undergo three-dimensional MRI- or CT-guided high-dose-rate intracavitary brachytherapy, combined with interstitial implantation when clinically indicated. MRI-guided brachytherapy is recommended for patients. The treatment will consist of either five fractions of 6 Gy or four fractions of 7 Gy, prescribed to the high-risk CTV, encompassing the entire cervix and any residual tumor, in accordance with each center’s clinical practices. The quality of radiotherapy will be assessed by a qualified central vendor both prior to study initiation and for each patient’s treatment plan throughout the study. Treatment planning will follow Groupe Européen de Curiethérapie– European Society for Radiotherapy and Oncology and NCCN guidelines to ensure adequate target coverage and OARs sparing [3, 24]. The planning aims include achieving a cumulative total EQD2 to the high-risk CTV D90 (minimum dose to 90% of the volume) of ≥ 80 Gy, taking into account contributions from both EBRT and brachytherapy. For OARs, the following cumulative dose constraints will be applied. Soft constraints: bladder D2cc < 80 Gy, rectum D2cc < 65 Gy, sigmoid colon D2cc < 70 Gy, and small bowel D2cc < 70 Gy. Hard constraints: bladder D2cc < 90 Gy, rectum D2cc < 75 Gy, sigmoid colon D2cc < 75 Gy, and small bowel D2cc < 75 Gy. Contouring and dose reporting will adhere to International Commission on Radiation Units and Measurements Report 89 standards [25].

Concurrent chemotherapy ± PD-1 inhibitor

All patients will receive concurrent weekly cisplatin (40 mg/m² per week) as a single agent for a minimum of three weeks. For patients unable to tolerate chemotherapy, alternative regimens may include reduced-dose cisplatin, paclitaxel, or carboplatin. According to the NCCN guidelines, patients classified as FIGO 2014 Stage IIIA, IIIB, IVA or select FIGO 2018 Stage III–IVA is permitted to receive concurrent chemotherapy ± pembrolizumab.3 For enrolled patients in this study, pembrolizumab may be added to patients with FIGO 2018 stage III cervical cancer based on clinical considerations [26]. Considering drug accessibility, other approved immunotherapy agents may serve as alternative treatment options. Cisplatin, other alternative chemotherapy regimens, or immunotherapy will all be administered intravenously.

Follow up

Assessments for primary and secondary endpoints will be conducted before treatment and repeated weekly throughout the treatment course. Patients demonstrating persistent central residual tumor within 3 months post-radiotherapy may be candidates for a three-dimensional image-guided brachytherapy boost. In such cases, both the boost dose and total cumulative radiation dose must be meticulously recorded and calculated. Follow-up examinations will occur every three months during the first two years and every six months during the third year.

Tumor response will be assessed according to Response Evaluation Criteria in Solid Tumors version 1.1 by investigators [27], with disease progression primarily monitored through regular physical exams, hematological and biochemical laboratory tests, squamous cell carcinoma antigen levels, and pelvic MRI or chest/abdomen/pelvis CT. If disease progression is suspected during follow-up, PET-CT will be performed when available to evaluate the extent of recurrence and guide subsequent treatment. All cases of disease progression will be thoroughly documented.

Acute toxicity will be assessed from the start of treatment until three months post-treatment, while late toxicity will be evaluated from three months onward during routine follow-up visits. Acute toxicity will be assessed using the Common Terminology Criteria for Adverse Events version 5.0 (supplementary file 1), while late toxicity will be evaluated according to the Radiation Therapy Oncology Group criteria. Patient-reported toxicity and QoL will be measured using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 and the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Cervical Cancer Module 24. The flowchart of the study process is shown in Fig. 2.

Fig. 2

Flowchart of the process of the study phases. FIGO, International Federation of Gynecology and Obstetrics; LN, lymph node; CIN, common iliac nodes; ECOG, Eastern Cooperative Oncology Group; IMRT, Intensity modulated radiotherapy; oART, online adaptive radiotherapy; CT, computed tomography; MRI, Magnetic resonance imaging; HDR, high dose rate; ICBT, intracavity brachytherapy; ISBT, interstitial brachytherapy; CRR, complete response rate; OS, overall survival; LPFS, locoregional progression-free survival; MFS, metastasis-free survival; CSS, cervical cancer-specific survival

Sample size calculation

To assess whether the PFS at 3 years in the experimental group is non-inferior to that in the control group, a non-inferiority design will be employed. The enrollment period is set at 24 months, followed by a 36-month follow-up period, with a one-sided α = 0.025 and β = 0.20. Informed by the literature, the 3-year PFS rates for both the experimental and control groups are approximately 70%16,17. The non-inferiority margin for 3-year PFS is set at 12%, which corresponds to a hazard ratio (HR) non-inferiority margin of 1.527. According to the survival analysis non-inferiority sample size formula for two groups, the sample size per group is approximately 195. Considering a 10% dropout rate, the required sample size per group is adjusted to 220, resulting in a total sample size of approximately 440 participants.

This clinical trial is planned to be conducted across four medical institutions. A total of 440 patients will be enrolled across these institutions, with 220 assigned to the experimental group and 220 to the control group.

Statistical analysis

The full analysis set, comprises all randomized patients, is the primary evaluation population for all efficacy endpoints, including the primary endpoint. The per-protocol set includes all patients in the full analysis set population who have received the planned therapy in its entirety and for whom documentation related to the primary endpoint is complete. Analyses of the per-protocol set population serve as sensitivity analyses to assess the robustness of the results from the intention-to-treat analysis. Patients in the full analysis set in who have initiated the planned therapy are included in the safety set, which serves as the primary evaluation population for toxicity and other safety endpoints.

The primary endpoint, PFS, along with other long-term survival outcomes, OS, LPFS, MFS, CSS, will be analyzed using the Kaplan-Meier method. Differences in survival outcomes between MHRT and CFRT groups will be assessed using the log-rank test. The false positive rate will be set at 0.025 for one-sided tests and 0·05 for two-sided tests. Non-inferiority of PFS will be declared if the upper limit of the one-sided 97.5% confidence interval (CI) for HR is less than 1.527. HR and 97.5% CI will be estimated using a Cox proportional hazard model, with treatment group as a factor. Differences in the CRR will be evaluated using the Chi-square test. Toxicities will be described as categories and grades by investigators, and will also be reported as scores from questionnaires for QoL by patients. And it will be compared using Rank Sum Test if necessary. Comparisons of tumor regression after EBRT and toxicities between the two groups will be performed using the independent sample t test, Mann-Whitney U test, or Rank Sum Test and Chi-square test, depending on the normality, homogeneity of variance, and data types. Subgroup analyses will be conducted based on the same stratification factors applied for randomization. The same statistical methods will be utilized to evaluate differences in primary and secondary endpoints across the subgroups, as previously outlined.

To assess futility, interim analyses will be conducted using Bayesian predictive probability in a non-inferiority design. Statistical analyses and programming will be performed using SAS version 9.4.

Data quality assurance (QA) and monitoring

To ensure data quality, treatment consistency, and patient safety, a standardized QA program has been implemented across all participating sites. A unified radiotherapy protocol outlines the definitions of target volumes, dose prescriptions, and constraints for OARs.

A QA team—comprising radiation oncologists, medical physicists, and radiation therapists—was established at the lead center. Prior to trial initiation, each site was required to submit simulation CT scans and iCBCT images from two eligible but non-enrolled patients: one with lymph node boost and one without. These submissions included delineated target volumes and OARs, as well as both conventional and moderately hypofractionated adaptive radiotherapy plans. Additionally, brachytherapy contours and plans were also submitted. All Digital Imaging and Communications in Medicine files were reviewed and approved by the QA team.

During the trial, the first 20 EBRT and brachytherapy plans will be audited, followed by monthly random checks of two cases per site. In the MHRT arm, daily online adaptive planning follows standardized algorithms with periodic audits. QA site visits and virtual meetings ensure protocol adherence and resolve deviations.

In addition to radiotherapy-specific QA, overall study conduct is regularly reviewed by a monitor from independent data monitoring committee 1–4 times a month for each center as part of quality assurance process. Enrollment, reported adverse events and long-term survival will be continuously monitored to uphold ethical standards and protect the safety and interests of the participants.