Abstract
Background:
Patients with epithelial ovarian cancer (EOC) receiving neoadjuvant platinum-based chemotherapy (NACT) who remain ineligible for complete interval cytoreductive surgery (ICS) due to poor chemosensitivity (CA-125 KELIM™ score <1.0) have a poor prognosis (~20% 5-year survival). A weekly dose-dense carboplatin–paclitaxel regimen may improve outcomes in this high-risk subgroup.
Objectives:
To demonstrate the superiority of a salvage weekly dose-dense carboplatin–paclitaxel regimen over continuation of the standard 3-weekly regimen in poor-prognosis EOC patients after 3–4 cycles of standard NACT.
Design:
SALVOVAR is a pragmatic, open-label, multicenter, international, randomized phase III trial.
Methods and analysis:
Patients with stages III–IV high-grade EOC are eligible if they present (1) an unfavorable standardized KELIM score <1.0, and (2) a disease not amenable to complete ICS after 3–4 cycles of standard 3-weekly carboplatin–paclitaxel. Patients are randomized (1:1) to either the experimental arm (dose-dense carboplatin AUC5 day 1 plus paclitaxel 80 mg/m2 on days 1, 8, and 15, every 3 weeks) or the control arm (continuation of the standard regimen) for 3 cycles. Bevacizumab use is allowed at investigator discretion. Stratification factors include planned bevacizumab administration, BRCA mutation status, and KELIM strata. The two co-primary endpoints are (1) improvement in late complete cytoreduction rates (from 5% in the control arm to 20% in the experimental arm), and (2) overall survival (target hazard-ratio, 0.61). Total 250 patients will be randomized. Secondary endpoints include objective response rate, progression-free survival, and safety. Additional planned analyses include quality-of-life, cost-effectiveness, surgical standardization, human sciences, and biology studies.
Ethics:
The protocol was approved by the national ethics committee and health authorities.
Discussion:
SALVOVAR will evaluate whether chemotherapy densification improves outcomes in poorly chemosensitive advanced EOC. If positive, this pragmatic strategy could be implemented in large-scale studies, independent of resource setting.
Trial registration:
ClinicalTrials.gov NCT06476184 (June-2024). Available at: https://clinicaltrials.gov/study/NCT06476184
Plain language summary
Introduction:
Some women with advanced ovarian cancer get chemotherapy before surgery. If the cancer doesn’t respond well and doctors can’t remove it completely, the outlook is poor: only about 1 in 5 women live for 5 years. The KELIM score, which measures how quickly the CA-125 blood marker drops during treatment, helps show how sensitive the cancer is to chemotherapy. A low score (<1.0) means the cancer isn’t responding well. New studies suggest that giving chemotherapy weekly instead of every three weeks may help improve results in these high-risk patients.
Aim:
The SALVOVAR trial is testing whether stronger chemotherapy can help women with advanced ovarian cancer that does not respond well to treatment and cannot be fully removed with surgery after standard chemotherapy.
Method:
This is a large international study (phase III) for women with advanced high-grade ovarian cancer (stage III and IV). To join, patients must have already received 3 to 4 rounds of standard chemotherapy before surgery, have a low KELIM score (<1.0), and have cancer that cannot be fully removed by surgery.
Participants are randomly assigned to one of two groups:
- Experimental group: weekly carboplatin and paclitaxel (dose-dense chemotherapy)
- Standard group: chemotherapy every 3 weeks
Doctors may also add bevacizumab if they think it is appropriate.
The main goals are to see how many women can later have complete surgery after chemotherapy and to measure overall survival. Other goals include tumor response, time until the cancer gets worse, treatment safety, and how patients feel during treatment.
Discussion:
SALVOVAR will test if giving chemotherapy more often (dose-dense) helps high-risk patients do better. If it works, this simple treatment could be used everywhere, even in places with fewer medical resources.
Keywords
Background
In Europe, the annual prevalence of epithelial ovarian cancer (EOC) is about 12 cases per 100,000.1 –3 Approximately, 75% of patients are diagnosed with advanced stages III–IV disease. EOC is the fifth most common cause of cancer death in women, accounting for more deaths than any other cancer of the female reproductive system.2,4
The standard management relies on medical-and-surgical treatment backbone, where all components play major prognostic roles in the success of the first-line treatment and patient survival. 5 A key objective of the treatment is to achieve a complete cytoreductive surgery with no visible residual postoperative lesions (completeness of cytoreduction 0 (CC0), or nul residual (R0)). 6 If possible, surgery should be performed initially (as a primary cytoreductive surgery), followed by adjuvant chemotherapy with the 3-weekly carboplatin-paclitaxel regimen (carboplatin AUC 5–6 and paclitaxel 175 mg/m2, every 3 weeks) for 6 cycles.4,7 Alternatively, neoadjuvant chemotherapy (NACT) using the same carboplatin-paclitaxel regimen should be administered for 3–4 cycles to reduce tumor burden, to facilitate a complete interval cytoreductive surgery (ICS), followed by postoperative chemotherapy.8,9 It is estimated that about 50% of patients are treated with this strategy in Europe and North America.10,11
As complementary medical treatments, the addition of bevacizumab—an anti-angiogenic monoclonal antibody—to adjuvant chemotherapy followed by maintenance therapy, and/or maintenance treatment with a poly(ADP-ribose) polymerase inhibitor (PARPi) for 2–3 years may be considered, depending on the disease profile and patient characteristics.4,12
The subpopulation of patients who are treated with NACT and who do not respond sufficiently to chemotherapy (identified by an unfavorable CA-125 KELIM™ score <1.0, a pragmatic indicator of the tumor primary chemosensitivity) leading to a disease not amenable to a complete ICS, is known to have a particularly poor prognosis. They represent about 50% of those treated with NACT, and their overall survival (OS) is less than 20% at 5 years, according to several trials performed before the PARPi era.13,14 Unfortunately, these patients are unlikely to benefit from the medical progress brought about by the development of PARPi, as these drugs have been assessed only in patients who had a favorable response to platinum-based chemotherapy. 15
Consequently, any solution that may increase the chemotherapy efficacy in this subpopulation of patients would be highly relevant to improving the patient’s prognosis. In that context, chemotherapy intensification may be a promising approach. This strategy has been successfully implemented in other poor prognostic cancers, such as nonseminomatous germ cell tumors, 16 bladder cancer, 17 or breast cancer. 18
In ovarian cancer, the post hoc analysis of ICON-8 phase III trial dataset suggested that the patients belonging to the poor prognostic subgroup (unfavorable KELIM score <1.0, and an incomplete ICS) had benefited from the weekly-dose carboplatin-paclitaxel regimen (carboplatin AUC 5–6 every 3 weeks with weekly paclitaxel 80 mg/m2) in terms of progression-free survival (PFS; median PFS, 9.2 vs 5.6 months, hazard-ratio, 0.63, 95% confidence interval (CI) 0.45–0.88), and OS (median, OS, 31.7 vs 20.3 months, hazard-ratio: 0.73, 95% CI 0.52–1.03) compared to the standard 3-weekly regimen.14,19
Further evidence recently came from the ICON8B trial, which demonstrated that a weekly dose-dense regimen combined with bevacizumab outperformed the standard 3-weekly regimen with bevacizumab in high-risk ovarian cancer patients, including those treated with NACT, before the widespread use of PARPi. 20 However, the higher toxicity profile of intensified chemotherapy (e.g., myelotoxicity, neurotoxicity, nausea-vomiting) necessitates careful selection of patients most likely to benefit from this strategy.
The SALVOVAR trial was designed to investigate the benefit of adjusting the chemotherapy dose and dosing schedule to the weekly dose-dense regimen in patients belonging to the subgroup of patients with a poor prognostic advanced ovarian epithelial cancer due to an unfavorable response to 3–4 cycles of NACT and a disease not amenable to complete ICS.
Methods
Trial design
SALVOVAR is a pragmatic academic randomized (1:1) open-label multicenter phase III trial sponsored by the French collaborative group ARCAGY-GINECO (Paris, France). SALVOVAR is planned to be conducted in nine European countries (France, United Kingdom, The Netherlands, Italy, Czech Republic, Slovakia, Slovenia, Hungary, and Israel) and in Japan, in collaboration with the European Network for Gynecological Oncological Trial groups (ENGOT) and the Gynecologic Cancer InterGroup (GCIG; List of institutions presented in the Supplemental Material). SALVOVAR is funded by a European Union HORIZON-MISS-CANCER-2022-01 research program grant. The reporting of this study adheres to the SPIRIT 2025 guidelines (Supplemental Material). 21 The present article describes version 1.0 of the protocol.
Study population
Main inclusion criteria
The enrolled patients must be aged ⩾18 years; with an Eastern Cooperative Oncology Group performance status of 0 or 1; and pathology confirmed high-grade epithelial (serous, endometrioid, or carcinosarcoma with a ⩾30% epithelial tumor component) ovarian, primary peritoneal, or fallopian-tube carcinoma; and staged III or IV according to FIGO staging. They should have received 3 or 4 neo-adjuvant cycles of the standard 3-weekly carboplatin-paclitaxel regimen in first-line setting. Moreover, they must exhibit both main poor prognostic features, with (1) an unfavorable standardized KELIM score <1.0 and (2) a disease not amenable to a complete CC0-R0 ICS. It is estimated that about 35% of all patients treated with NACT will present these two main inclusion criteria.
The standardized KELIM score can be easily calculated for each patient treated with a NACT on the online calculator for identifying those presenting an unfavorable score <1.0 (https://www.biomarker-kinetics.org/).
The other main inclusion criteria comprise adequate bone marrow, kidney, and liver functions allowing for potential weekly-dense chemotherapy. Patients must provide written informed consent, obtained by the investigator, and be able to comply with the protocol.
Main exclusion criteria
The exclusion criteria include other histologic subtypes with distinct chemosensitivity profiles and prognoses, which could bias trial outcomes (low-grade endometrioid, clear cell, mucinous, or sarcomatous histology, or mixed tumors containing any of these histologies, or low-grade or borderline ovarian tumor); a contraindication to investigational drugs; a previous treatment with bevacizumab or weekly carboplatin-paclitaxel during the initial NACT; a primary platinum-refractory disease; a concomitant cancer; a treatment with other investigational agents in clinical trials; any clinically significant uncontrolled condition; known psychiatric disorder that would interfere with trial compliance; along with pregnant or lactating patients.
Treatment plan
Chemotherapy doses and dosing schedules during the randomized chemotherapy phase
Enrolled patients will be randomly allocated with a 1:1 ratio to the experimental arm (salvage dose-dense regimen, with carboplatin AUC 5 on day 1 combined with weekly paclitaxel 80 mg/m2 on day 1, day 8, and day 15, given every 3-week cycles; potentially combined with daily sub-cutaneous administrations of Granulocyte Colony Stimulating Factors (GCSF) 30 MUI from days 3 to 5, and from days 10 to 12, and from days 17 to 19, at investigator discretion) or the control arm (carboplatin AUC 5 combined to paclitaxel 175 mg/m2, at the same dose as given during the neo-adjuvant phase, given every 3-week cycle, potentially combined with GCSF or Peg-GCSG at investigator discretion), for 3 cycles.
Bevacizumab addition is advised in countries where it is available per standard of care. The dose of bevacizumab will be given at the investigator’s discretion, as per the local standard of care (15 or 7.5 mg/kg, every 3 weeks), given in combination with chemotherapy for 2–3 cycles, and potentially as a maintenance treatment following the chemotherapy for 15 months.
The randomization will be stratified by the three following covariates: planned bevacizumab administration (yes, vs no); BRCA mutation (yes, vs no/unknown; and KELIM score strata (very unfavorable <0.7, vs moderately unfavorable (0.7–1.0)).
Patients are randomly allocated to the treatment arms using a minimization procedure implemented automatically through the Ennov Clinical® program by the sponsor, ensuring balanced distribution across key stratification factors.
Subsequent disease management
At the end of this randomized chemotherapy phase, the feasibility of a late complete cytoreductive surgery will be assessed and performed if considered feasible.
At the investigator’s discretion, administration of additional chemotherapy cycles is possible, given either before or after a potential late cytoreductive surgery.
Subsequent maintenance with bevacizumab and/or a PARPi may be administered upon the local standard of practice and at the investigator’s discretion (Figure 1).

Trial design. Debulking surgery is equivalent to cytoreductive surgery.
Efficacy assessments
Imaging assessments should be performed with thorax-abdomen-pelvis CT-scanner (or MRI or PET-CT if clinically indicated) at enrollment, after 3 cycles, and then ideally every 3 months for 3 years and then every 6 months for 2 years, according to local standard practice.
The study flowchart is presented in the Supplemental Material.
Study objectives and endpoints
The primary objective is to demonstrate the superiority of the experimental with the salvage weekly dose-dense chemotherapy compared to the control arm consisting of the continuation of the standard 3-weekly carboplatin-paclitaxel regimen in terms of efficacy, with 2 co-primary endpoints: (1) percentage of patients achieving late complete cytoreductive surgery with an expected increase from 5% to 20%, and (2) OS with an expected improvement of 49% (hazard ratio (HR) = 0.61) from a median OS of 20.0–32.8 months.
The secondary objectives are to compare the efficacy of the two regimens in terms of radiologic response according to Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v1.1) (objective response rate (ORR) and PFS). Moreover, the impact of each regimen on subsequent PARPi use (proportion of patients receiving PARPi during maintenance) and associated efficacy outcomes (PFS and OS) will be evaluated. In addition, the effect of bevacizumab use on treatment efficacy will be analyzed with respect to ORR, PFS, OS, and the rate of complete late cytoreductive surgery. Finally, safety profiles will be assessed in both arms according to the NCI CTCAE version 5.0.
Statistics and sample size
Sample size calculation
The study is calibrated for a dual primary endpoint: (1) first primary endpoint: rate of patients undergoing late complete cytoreductive surgery after the three randomized cycles; and (2) second primary endpoint: OS. To control for type 1 error of 5%, a significance level of 1% will be used for the first primary endpoint and 4% will be used for the second primary endpoint. The study will be considered positive if either endpoint reaches statistical significance.
The study was calibrated to detect a treatment effect HR of 0.61 under the proportional hazard assumption in OS, translating into an improvement in median OS from 20.0 months (control arm) to 32.8 months (experimental arm) with a 1:1 randomization. A total of 138 deaths observed during the trial would have 80% power to show statistically significant OS at a 4% two-sided alpha level. Considering a recruitment duration of 24 months and an 18-month follow-up for the last included patient (total duration of the trial: 42 months), approximately 250 patients will be randomized in the study (125 patients in each arm).
Using this sample size, the study will have 85% power at type I alpha error of 1% to detect a 15% increase in the rate of patients with late complete cytoreductive surgery (from 5% in the control arm to 20% in the experimental arm).
Considering that 35% of patients treated with NACT should fit with the SALVOVAR inclusion criteria, 715 patients should be identified at the NACT step.
The final analysis of the late cytoreductive surgery endpoint will occur first (approximately 27 months after the first inclusion). At that time, an interim analysis of OS will also be conducted using a Lan-DeMets spending function approach to adjust the significance level (alpha).
Statistical analysis plan
Analyses will be performed according ITT population (all patients randomized). Baseline characteristics will be described to verify that treatment groups are comparable according to risk levels (qualitative variables using frequency and percentage distributions; quantitative data with mean, standard deviation, median, minimum, and maximum values). Median follow-up will be calculated using reverse Kaplan–Meier method.
The first primary endpoint (percentage of late complete cytoreductive surgery) will be calculated in each arm by the number of patients for whom a complete cytoreductive surgery have been performed after three additional cycles of CT divided by the number of randomized patients in that arm (ITT population). It will be reported with its 99% CI and compared between arms using a χ2 test (a p-value <0.001 will be considered as statistically significant).
The second primary endpoint is OS, defined as the time from randomization to death from any cause. OS will be estimated using the Kaplan–Meier method with its two-sided CI. Comparisons between study arms will be performed using a stratified two-sided log-rank test at a 4% significance level, supported by a stratified Cox regression model (stratified on randomization strata: planned bevacizumab administration; BRCA mutation; and KELIM score). We assume that OS status will be known for all patients. No imputation of missing data is planned.
A Statistical Analysis Plan describing the analysis sets, the derived variables, and the statistical analysis to be produced for “SALVOVAR” study will be approved before the database lock.
Statistical analyses will be performed using SAS® software version 9.4 or later (SAS Institute Inc., 2023, USA).
Timelines and study duration
The protocol was first approved by the French ethics committee on May 7, 2024, and by the national health-authorities on May 7, 2024. The national health authorities and ethics committees of the other recruiting countries subsequently approved the protocol.
The enrolment period is planned for 24 months. The patient participation in the study will end at the trial closure (planned for 5 years in total) or at patient death, whichever occurs first.
Substantial protocol changes will be documented in an amendment or revised protocol and approved by the Ethics Committee and, when applicable, the regulatory authority before implementation.
Independent monitoring
A Data Safety Monitoring Board, composed of the sponsor, one statistician, and two clinical experts, will assess patient safety. Regular monitoring visits will be conducted to support investigators, verify protocol adherence, data accuracy, and study drug accountability. Source data verification, including confirmation of informed consent, will require direct access to original records.
SALVOVAR comprehensive project to improve the management of patients with advanced ovarian cancer
Beyond the phase III trial (Work-package 1) and the related statistics (Work-package 2), SALVOVAR is a larger academic collaborative comprehensive project meant to assess other dimensions of the disease management of these patients:
- Work-package 3: Determinants of the treatment decision-making process in the context of therapeutic uncertainty will be assessed from both clinician and patient perspectives. Questionnaires and semistructured interviews will be conducted at inclusion, after the randomized chemotherapy phase, and during follow-up to explore perceptions regarding key treatment decisions, including the administration of bevacizumab, late cytoreductive surgery, and maintenance therapy. The findings will help inform strategies to better integrate patient preferences into therapeutic decision-making and enhance shared decision-making in cancer care.
- Work-package 4: Definition of consensual standardized criteria for a nonresectable disease at ICS based on the ESGO Ovarian Cancer Operative Report. An expert surgeon group, composed of 14 members, has been built with surgeons from the different participating countries to validate the criteria of a disease not amenable so that the enrolled population of patients is homogenous. E-learning has been built to train the other surgeons participating in the trial (https://www.oncostream.com/fr/playlists/289-salvovar). Moreover, the completeness of the late cytoreductive surgery for the patients who will have been operated on after the randomized chemotherapy phase will be centrally confirmed.
- Work-package 5: Inventory of BRCA mutational and homologous recombination deficiency assays used in real-life settings and the related outcomes. An online questionnaire will be sent to all the participating sites and countries involved in the GCIG network to understand the heterogeneity of assays available, covered, and used worldwide.
- Work-package 6: Assessment of Patient-Reported Outcome (PRO) data will be collected about symptomatic toxicity and quality of life. The PRO-CTCAEs will be administered at randomization and at each chemotherapy cycle. The EORTC QLQ-C30 and EORTC QLQ-OV28 questionnaires will be given on chemotherapy administration days of cycles 1 and 3 during treatment phase, and every 3 months thereafter, for up to 2 years irrespective of disease progression.
- Work-package 7: Cost-utility and cost-effectiveness analyses will be conducted to compare the experimental strategy with the standard strategy. These evaluations will be performed exclusively on patients enrolled in France in the SALVOVAR trial, as medico-economic assessments are country-specific. However, an analysis of the generalizability and transferability of the economic findings to other countries will also be undertaken. A budget impact analysis will complement the efficiency evaluation, aiming to estimate the financial impact on the French Health Insurance system of implementing the experimental strategy more broadly within the French healthcare context.
- Work-package 8: Integration of patient associations and communication/dissemination. The Spanish ovarian cancer patients ASACO and Cancer Patients Europe are involved in all steps of the comprehensive project to ensure that the cancer patient perceptions are integrated into the organizational aspects of the project (informed consent, questionnaires, interviews, etc.). Moreover, educating the academic and pharmaceutical scientific community about the prognostic of this subgroup of patients and the need for an innovative approach has been a significant commitment, concretized by a specific internet site and explicative videos (https://www.salvovar.eu/).
Discussion
There is an unmet medical need for innovative therapeutic strategies that can improve the treatment efficacy in advanced EOC patients who respond insufficiently to neoadjuvant chemotherapy to benefit from a complete ICS. If this effort is not achieved, the survival gap, and therefore the inequity, between these patients and those who respond well to platinum-chemotherapy may potentially increase, as the first group may not benefit from the progress related to PARPi emergence.
In this context, any therapeutic strategy that can enhance chemosensitivity in poorly responsive diseases is highly relevant.
SALVOVAR is investigating the efficacy of chemotherapy intensification based on a solid rationale, since several studies suggested that this approach could improve ovarian cancer survival.14,19,20 If SALVOVAR is positive, this chemotherapy adjustment could be implemented in all centers and countries, regardless of income level, as it would be feasible at low cost.
The SALVOVAR trial is designed to ensure that the analyzed patient population will be homogenous from a surgical point of view in terms of disease nonresectability. The project has also planned to assess the cost-effectiveness and the impact of the chemotherapy intensification strategy on the patient quality of life. Subgroup analyses will evaluate the impact of prior ICS attempt, bevacizumab exposure, disease stage, and resectability, and delayed cytoreductive surgery on OS. All these data will inform about the overall benefit/harm of this approach with original sciences and human ancillary projects.
In addition, SALVOVAR aims at understanding the heterogeneity in the BRCA/HRD assays used worldwide and how to better integrate patients in the treatment decision-making process.
In this regard, this academic, pragmatic trial—funded by the European Commission—addresses the need for pragmatic, affordable, and practice-changing real-life oncology trials, as emphasized by Leary et al. 22
We envision that SALVOVAR might be the first step in future studies to chemosensitize ovarian cancer using other therapeutic approaches, such as innovative drugs acting on immune checkpoints or cell cycle checkpoint inhibitors. 23
Supplemental Material
sj-docx-1-tam-10.1177_17588359251396777 – Supplemental material for SALVOVAR: a pragmatic randomized phase III trial comparing the SALVage weekly dose-dense regimen to the standard 3-weekly regimen in patients with poor prognostic OVARian cancers
Supplemental material, sj-docx-1-tam-10.1177_17588359251396777 for SALVOVAR: a pragmatic randomized phase III trial comparing the SALVage weekly dose-dense regimen to the standard 3-weekly regimen in patients with poor prognostic OVARian cancers by Claire Chator, Nozomu Yanaihara, Sylvie Chabaud, Gabriella Maria Parma, Alexandra L. Dima, Andrew Clamp, Gwenaël Ferron, Judith R. Kroep, Alexandra Leary, David Cibula, Frederic Fiteni, Ilan Bruchim, Hassan Serrier, Susannah Carroll, Anne-Sophie Belmont, Julien Peron and Benoit You in Therapeutic Advances in Medical Oncology
Supplemental Material
sj-docx-2-tam-10.1177_17588359251396777 – Supplemental material for SALVOVAR: a pragmatic randomized phase III trial comparing the SALVage weekly dose-dense regimen to the standard 3-weekly regimen in patients with poor prognostic OVARian cancers
Supplemental material, sj-docx-2-tam-10.1177_17588359251396777 for SALVOVAR: a pragmatic randomized phase III trial comparing the SALVage weekly dose-dense regimen to the standard 3-weekly regimen in patients with poor prognostic OVARian cancers by Claire Chator, Nozomu Yanaihara, Sylvie Chabaud, Gabriella Maria Parma, Alexandra L. Dima, Andrew Clamp, Gwenaël Ferron, Judith R. Kroep, Alexandra Leary, David Cibula, Frederic Fiteni, Ilan Bruchim, Hassan Serrier, Susannah Carroll, Anne-Sophie Belmont, Julien Peron and Benoit You in Therapeutic Advances in Medical Oncology
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
