Induction nivolumab (Opdivo) plus ipilimumab (Yervoy) followed by chemoradiation (RT-CT) and nivolumab maintenance produced a 3-year progression-free survival (PFS) rate of 83% (95% CI, 67%-91%) and a 3-year overall survival (OS) rate of 90% (95% CI, 76%-96%) in patients with locally advanced cervical cancer (LACC), according to updated data from the phase 2 COLIBRI/GINECO-CE108b trial (NCT04256213) presented at the 2026 ESMO Gynaecological Cancers Congress.1
Moreover, at a median follow-up of 40 months (range, 38-44), patients with FIGO 2018 stage I-II disease achieved a 3-year PFS rate of 93% (95% CI, 59%-99%) and a 3-year OS rate of 100% (95% CI, 100%-100%), while those with stage III-IV disease achieved a 3-year PFS rate of 77% (95% CI, 56%-89%) and a 3-year OS rate of 85% (95% CI, 64%-94%). The updated safety analysis confirmed no new signals with extended follow-up.
Analysis of survival outcomes by hot/cold immune signature at baseline showed a 3-year PFS rate of 76% (95% CI, 56%-88%) and a 12-month PFS rate of 86% (95% CI, 67%-95%) in patients with cold tumors (n = 29). Respective rates in patients with hot tumors were 100% (95% CI, 100%-100%) and 100% (95% CI, 100%-100%). After immune checkpoint blockade (ICB) induction, 38% of baseline cold tumors converted to hot tumors. The post-ICB immune signature analysis showed a 3-year PFS rate of 60% (95% CI, 32%-80%) and a 12-month PFS rate, 73% (95% CI, 44%-89%) in those with cold tumors after ICB (n = 15). In those with hot tumors post-ICB (n = 22), respective rates were 95% (95% CI, 72%-99%) and 100% (95% CI, 100%-100%).
"The patients [whose tumors] remained cold—38% of the tumors remained cold—performed very low. The patients who had hot [tumors] at the beginning or [whose tumors] became hot after induction [therapy with] ICB performed quite well," -Florence Joly, MD, PhD, presenting author, head of the Department of Medical Oncology at Centre Léon Bérard and GINECO in France.
How was the COLIBRI trial designed?
COLIBRI-1 was a multicenter, single-arm pilot study enrolling women aged 18 years or older with histologically confirmed cervical adenosquamous carcinoma, FIGO 2018 stage IB3 to IVA disease, and ECOG performance status of 0 or 1.
The treatment schedule consisted of a neoadjuvant ICB phase in which patients received a single cycle of intravenous ipilimumab (1 mg/kg) on day 1 and nivolumab (3 mg/kg) on days 1 and 15. This was followed by 5 to 8 weeks of concurrent chemoradiation and a 4-to-6-week delay, with optional surgery for residual disease. Patients then received nivolumab 480 mg every 4 weeks for 6 months as maintenance therapy. Mandatory tumor biopsies were collected at baseline (before ICB) and after the neoadjuvant phase (before chemoradiation).
The study’s primary end point was the change in CD8+/FOXP3+ cell ratio from pre- to post-ICB therapy; the secondary endpoints reported here include safety, 3-year PFS and OS, and evolution of the tumor immune microenvironment and its correlation with outcomes.
What were the baseline characteristics of patients in the analysis population?
The median age in the analysis population (n = 40) was 55.0 years (range, 31.0-77.0). FIGO stage distribution included IB3 (3%), IIA2 (3%), IIB (30%), IIIB (3%), IIIC1 (25%), IIIC2 (25%), and IVA (13%). The majority of patients had squamous cell carcinoma (95%), and 93% had PD-L1 expression of at least 1 by SP263 assay. All 40 patients received the full neoadjuvant ICB induction and received EBRT per protocol, 36 received brachytherapy, and 39 entered the maintenance phase. Of those, 34 completed all 6 cycles of nivolumab maintenance.
Prior Findings From the Phase 2 COLIBRI Trial
- Initial findings published in Nature Communications showed a statistically significant increase in the ratio of effector CD8+ T cells to regulatory FOXP3+ T cells, with a mean increase of 0.87 cells/mm² (P = .0164) after 1 cycle of nivolumab plus ipilimumab.2
- The HOT score also increased significantly after one cycle of combination immunotherapy in an exploratory analysis (increase of 0.17; P < .0001)
- Objective response rates were 13% immediately after combination immunotherapy, 98% after chemoradiation, and 90% at treatment completion.
What did the immune biomarker analyses show?
To evaluate the tumor immune microenvironment, investigators used two complementary techniques on paired biopsies obtained at baseline and prior to the start of chemoradiation. The first method was 7-color multiplex immunofluorescence (multi-IF) tissue imaging applied to 32 biopsy pairs, enabling quantification of total and proliferating (Ki67+) CD8+ (CD3+CD8+FOXP3-) and total FOXP3+ (CD3+FOXP3+) cell densities; the CD8+/FOXP3+ ratio median of 4.43 served as the threshold for patient stratification. The second method was High-Throughput Genomic sequencing (HTG) transcriptomic analysis applied to 37 biopsy pairs, evaluating a 27-gene "HOT" immune signature to classify tumors as immunologically active ("HOT") or inactive ("COLD").
Analysis by the CD8+/FOXP3+ Treg ratio demonstrated that patients with a ratio above the median at baseline achieved a 3-year PFS rate of 89% (95% CI, 68%-99%) and a 3-year OS rate of 95% (95% CI, 68%-99%), compared with 75% (95% CI, 60%-95%) and 85% (95% CI, 60%-95%), respectively, for those with a ratio at or below the median. This separation was amplified when assessed using the post-ICB induction biopsy: patients with a high post-ICB ratio achieved a 3-year PFS rate of 94% (95% CI, 63%-99%) and a 3-year OS rate of 94% (95% CI, 63%-99%) vs 63% (95% CI, 52%-94%) and 81% (95% CI, 52%-94%) in those with a low ratio.
Among patients with stage III-IV disease, the immune microenvironment trajectory appeared particularly informative for outcomes: those whose tumors were persistently cold from baseline through the pre-RTCT biopsy demonstrated markedly worse PFS and OS compared with those who were hot at either time point or converted from cold to hot after ICB induction.
What did the safety analysis show?
During the neo-adjuvant ICB phase, 82.5% of patients experienced any adverse effect (AE), with any AE of CTCAE grade ≥2 occurring in 35% and any treatment-related AE (TRAE) of grade 3 or 4 in 2.5%. No AEs led to death in any phase of treatment, and no AEs led to discontinuation of ICB during the neo-adjuvant phase.
During chemoradiation, any AE was reported in 97% of patients, and grade 2 or higher AEs were observed in 85%. The incidence of grade 3/4 TRAEs was 30%; the majority (27%) were attributed to chemoradiation, with 7.5% possibly related to nivolumab and 7.5% possibly related to ipilimumab. One patient (2.5%) discontinued ICB during RTCT due to neutropenia.
During the maintenance phase (n = 39), 87% of patients experienced any AE, 41% experienced grade 2 or higher AEs. The incidence of grade 3/4 TRAEs was 20%, with 15% possibly related to nivolumab, 2.5% to ipilimumab, and 12.5% to chemoradiation. Two patients (5%) discontinued ICB during maintenance; the remaining high-grade events included lymphopenia, neutropenia, asthenia, musculoskeletal pain, cutaneous rash, proctitis, and liver enzymatic abnormalities.
What are the next steps for the COLIBRI program?
Joly and fellow investigators noted that the current findings are exploratory and that an extensive translational program is ongoing. Planned analyses include evaluation of additional tumor immune parameters, assessment of the systemic immune response, tumor genomic profiling, HPV characterization, and analysis of circulating tumor DNA for cancer-related alterations and minimal residual disease detection.
The COLIBRI-2 trial (NCT06715241) is currently enrolling and directly compares induction and maintenance of nivolumab plus relatlimab-rmbw (Opdualog) vs nivolumab alone, with the goal of evaluating whether a next-generation dual checkpoint combination can further augment immune activation and survival outcomes in LACC.3
Disclosures: Joly reported consulting/advisory board fees and lecture/symposium support from GSK, AstraZeneca, Roche, Esai, MSD, Astellas, Janssen, Ipsen, Bayer, Novartis/3A, Pfizer, Daichii, and Pharma&; travel expenses from Esai, MSD, Ipsen, GSK, and Novartis; non-personal interests (research funding) from AstraZeneca, GSK, Astellas, and BMS (industrial), and from Ligue contre le cancer, Inca, Arc, and Rubal Rose (academic); funding committee membership with Inca and La ligue Nationale contre le Cancer; and scientific society/clinical research intergroup affiliations with ASCO, ESMO, GCIG, GINECO, and GETUG.
References
- Joly F, Bello-Roufai D, Hardy-Bessard AC, et al. Long-term survival outcomes and immune subgroup analyses in the COLIBRI-1-GINECO study: Induction nivolumab plus ipilimumab (ICB) prior to and following chemoradiation (RT-CT) in locally advanced cervical cancer (LACC). Presented at: 2026 ESMO Gynaecological Cancers Annual Congress; June 18, 2026.
- Ray-Coquard I, Kaminsky-Forrett M, Ohkuma R, et al. Neoadjuvant immune checkpoint blockade before chemoradiation for cervical squamous carcinoma (GINECO window-of-opportunity COLIBRI study): a phase II trial. Published January 5, 2026. Accessed January 28, 2026. Nat Commun 17, 922 (2026). doi: 10.1038/s41467-025-67646-z
- A multicenter, seeking signal, randomised, open-label phase II of relatlimab and nivolumab vs nivolumab alone in locally advanced cervical cancers (COLIBRI-2). ClinicalTrials.gov. Updated January 22, 2026. Accessed June 18, 2026. https://clinicaltrials.gov/study/NCT06715241
