A Phase 2a Trial of Immune Modulation in Combination With Ultrasound-mediated Blood Brain Barrier Opening in Patients With Newly Diagnosed Glioblastoma

Brain tumor treatment is hampered by the blood-brain barrier (BBB). This barrier prevents drugs carried in the bloodstream from getting into the brain. If the BBB can be opened, making it temporarily more permeable, drugs may able to better reach the brain tumor. In this trial we will implant a novel device with 9 ultrasound emitters, allowing temporary and reversible opening of the BBB to maximize brain penetration of drugs that modulate the immune system. The device will be implanted after radiation is completed. Immune modulating drugs will be given every 3 weeks in conjunction with activation of the device to open the BBB. The objectives of this trial are to establish whether it is safe and feasible to administer immune modulating drugs in this manner, and identify whether the treatment is effective in treating glioblastoma.

More details:

Eligible patients will undergo implant of the Soncloud-9 device within 1-5 weeks of completion of radiotherapy. About 1-3 weeks after surgery, patients will undergo sonication and intravenous administration of balstilimab, botensilimab and liposomal doxorubicin. Brain MRI will be done to quantify extent of blood brain barrier opening. The dose for balstilimab is 450 mg every 3 weeks. The dose for botensilimab is 1mg/kg every 6 weeks. The dose for liposomal doxorubicin is 30 mg every 3 weeks. Sonication and administration of study agents will continue every 3 weeks (21 days= 1 cycle) for a total of 9 cycles (approx. 6 months). Additional cycles may be considered if deemed beneficial and in the patient's best interest. Blood samples for circulating tumor DNA will also be collected before and after each sonication. The first 6 patients will comprise a safety run-in cohort with intensified safety monitoring through the end of the second cycle.

Overview

Balstilimab (AGEN2034; BAL) is an investigational, fully human IgG4 monoclonal antibody targeting programmed cell death protein 1 (PD-1). It has been studied as monotherapy and in combinations—particularly with the CTLA-4 inhibitors zalifrelimab and botensilimab—in cervical, colorectal and other solid tumors. It is not FDA-approved; a biologics license application for recurrent/metastatic cervical cancer was voluntarily withdrawn in 2021 after the FDA recommended against pursuing accelerated approval following full approval of pembrolizumab in the same setting. (investor.agenusbio.com)

Mechanism of action

Balstilimab binds PD-1 and blocks its interaction with PD-L1 and PD-L2, preventing inhibitory PD-1 signaling and potentially restoring cytotoxic T-cell activity against tumors. (cancer.gov)

Efficacy

Cervical cancer monotherapy: In a single-arm phase I/II study, 140 efficacy-evaluable patients with previously treated recurrent or metastatic cervical cancer received balstilimab 3 mg/kg every two weeks. The confirmed objective response rate (ORR) was 15.0% (95% CI, 10.0–21.8%), including 3.6% complete responses and 11.4% partial responses. The disease-control rate was 49.3%, and median response duration was 15.4 months. ORR was 20.0% in PD-L1-positive and 7.9% in PD-L1-negative disease. (pmc.ncbi.nlm.nih.gov)

Randomized cervical-cancer study: In the phase II RaPiDS trial, investigator-assessed ORR was 24% with balstilimab–zalifrelimab and 15% with balstilimab–placebo; blinded independent review found ORRs of 24% and 26%, respectively. Because the study was non-comparative and the investigator and independent assessments differed, it did not establish that adding zalifrelimab improved efficacy. (annalsofoncology.org)

Colorectal cancer combination: In a peer-reviewed phase I study of 101 response-evaluable patients with heavily pretreated, microsatellite-stable metastatic colorectal cancer, botensilimab plus balstilimab produced a 17% ORR (95% CI, 10–26%) and 61% disease-control rate. Median response duration was not reached and median progression-free survival was 3.5 months. These results apply to the combination and do not isolate balstilimab’s contribution. (pubmed.ncbi.nlm.nih.gov)

Safety

With cervical-cancer monotherapy, 71.4% of patients experienced treatment-related adverse events; commonly reported events included asthenia and diarrhea. The most frequent grade ≥3 treatment-related events were immune-mediated enterocolitis (3.1%) and diarrhea (1.9%). (pmc.ncbi.nlm.nih.gov)

In RaPiDS, treatment-related adverse events occurred in 71% of patients receiving balstilimab–zalifrelimab and 64% receiving balstilimab–placebo; grade 3 events occurred in 23% and 15%, respectively. No treatment-related deaths were reported. (annalsofoncology.org)

With botensilimab–balstilimab in metastatic colorectal cancer, 89% experienced treatment-related adverse events, most commonly fatigue (35%), diarrhea (32%) and fever (24%). Twelve percent discontinued because of treatment-related toxicity, and no grade 5 treatment-related events occurred. (pubmed.ncbi.nlm.nih.gov)

Further reading

• Balstilimab monotherapy in recurrent/metastatic cervical cancer — Gynecologic Oncology
• Balstilimab plus zalifrelimab in advanced cervical cancer — Journal of Clinical Oncology
• Botensilimab plus balstilimab in microsatellite-stable metastatic colorectal cancer — Nature Medicine
• NCI Drug Dictionary: balstilimab

Last updated: Sep 2026

Overview

Botensilimab (AGEN1181; BOT) is an investigational, fully human, Fc-engineered IgG1 monoclonal antibody targeting CTLA-4. It is being studied both alone and—more commonly—in combination with the investigational anti-PD-1 antibody balstilimab. Clinical development includes colorectal, ovarian, liver and other difficult-to-treat solid tumors. Evidence remains preliminary because most results come from small, non-randomized phase 1–2 studies; a randomized phase 2 colorectal-cancer trial is ongoing and has not posted results. (cancer.gov)

Mechanism of action

Botensilimab blocks CTLA-4-mediated suppression of T-cell activation. Its engineered Fc region has increased affinity for activating Fcγ receptors on antigen-presenting and natural-killer cells. Preclinical and translational studies indicate that this design may strengthen T-cell priming and memory, activate innate immunity, and promote depletion of intratumoral regulatory T cells. Whether these properties produce better clinical outcomes than established CTLA-4 antibodies has not yet been demonstrated in randomized trials. (pmc.ncbi.nlm.nih.gov)

Efficacy

The most extensive published evidence is for botensilimab plus balstilimab:

• Previously treated microsatellite-stable metastatic colorectal cancer: In a phase 1 trial, 101 response-evaluable patients had a confirmed objective response rate (ORR) of 17%, including one complete response, and a disease-control rate of 61%. Median progression-free survival was 3.5 months; median duration of response was not reached. All responses occurred in patients without active liver metastases. The study was single-arm, and efficacy data for 47 additional treated patients were immature at publication. (pmc.ncbi.nlm.nih.gov)
• Localized colorectal cancer: In the 24-patient phase 2 NEST study, neoadjuvant treatment produced major pathological responses in 41% of mismatch-repair-proficient tumors and 100% of mismatch-repair-deficient tumors. Treatment did not delay planned surgery. These results are promising but derive from a small, non-randomized study using pathological rather than long-term survival outcomes. (aacrjournals.org)
• Treatment-refractory ovarian cancer: A phase 1b cohort reported an ORR of 23% among 35 efficacy-evaluable patients, including one complete and seven partial responses; the 24-week clinical-benefit rate was 31%. (pubmed.ncbi.nlm.nih.gov)
• Treatment-refractory hepatocellular carcinoma: Among 18 efficacy-evaluable patients, the ORR was 17%, the 18-week clinical-benefit rate was 50%, median progression-free survival was 4.4 months, and median overall survival was 12.3 months. (pmc.ncbi.nlm.nih.gov)
• Responses have also been reported in phase 1 sarcoma cohorts, particularly angiosarcoma, but sample sizes were small and histologies heterogeneous. (pmc.ncbi.nlm.nih.gov)

Safety

Toxicity resembles that expected from CTLA-4/PD-1 checkpoint inhibition, particularly immune-mediated gastrointestinal and hepatic events. In the 148-patient metastatic colorectal-cancer safety population, 89% experienced a treatment-related adverse event; 31% had grade 3 and 1% had grade 4 events. Common events included fatigue (35%), diarrhea (32%) and fever (24%). Immune-mediated diarrhea/colitis occurred in 33%, including grade 3–4 events in approximately 11%. Twelve percent discontinued both drugs because of toxicity, and no treatment-related deaths were reported. (pmc.ncbi.nlm.nih.gov)

Further reading

• Phase 1 metastatic colorectal-cancer trial in Nature Medicine
• Mechanistic and translational study in Cancer Discovery
• NEST phase 2 neoadjuvant colorectal-cancer study
• Phase 1b ovarian-cancer study
• ClinicalTrials.gov randomized colorectal-cancer study

Last updated: Sep 2026

Inclusion Criteria:

* Have newly diagnosed pathologically proven glioblastoma, isocitric dehydrogenase-1/2 wild-type

* Tumor with methyl guanine methyl transferase (MGMT) gene promoter unmethylated

* Available paraffin embedded tumor tissue for the study

* Have completed standard radiotherapy with or without temozolomide

* 18 years of age or older

* Able to undergo contrast-enhanced MRI

* Have an Eastern Cooperative Oncology Group/World Health Organization performance status ≤ 2

* Size and location of the residual tumor and/or resection cavity must allow to be able to be covered by the sonication field

* Have not received any prior treatment with immunotherapeutic agents treatments for glioblastoma or other indications

* Have the ability to understand and willingness to sign a written informed consent prior to registration on study.

* Be willing and able to comply with the protocol.

* Have adequate organ and bone marrow function

* Agree to use adequate contraception if appropriate

Exclusion Criteria: Patients will be ineligible if they have:

* Multifocal tumor (unless all localized in a 50-mm diameter area accessible to ultrasound field) or tumor located in the posterior fossa.

* Uncontrolled epilepsy.

* Received other investigational agents within 2 weeks of registration

* Received prior therapy with or have history of allergic reactions attributed to compounds of similar chemical or biologic composition to agents used in this study.

* Contraindication to checkpoint inhibitor therapy (e.g., history of autoimmune disease)

* Uncontrolled illness

* History of active malignancy other than the brain tumor within 12 months prior to registration.

* Are pregnant or breastfeeding.