PDF full text in English, along with translations in Chinese, Hindi, Indonesian, Arabic, Spanish, Portuguese, and French.
Glioblastoma is the deadliest type of brain tumor with limited effective therapies over decades. Its growth and recurrence rely heavily on glioblastoma stem cells (GSCs), which resist standard chemo-radiotherapy. Bone Morphogenetic Protein 4 (BMP4) can push these malignant stem cells to mature and lose their tumor-forming ability in lab and animal models. This paper reports the world’s first human Phase 1 trial testing locally injected human recombinant BMP4 (hrBMP4) as a differentiation-based anti-cancer therapy for recurrent glioblastoma.
This international multi-center open-label trial used a standard 3+3 dose-escalation design, enrolling 15 adult patients whose glioblastoma had returned after standard chemoradiation. Eligible patients had decent daily function (Karnofsky score >70) and limited brain swelling. Doctors implanted fine catheters into and around tumors for Convection Enhanced Delivery (CED)—a slow, pressure-driven infusion that spreads drugs deep into brain tissue better than regular injections. Five dose groups (0.5 mg to 18 mg hrBMP4, three patients each) received continuous infusion over 4–6 days. Researchers tracked safety, maximum tolerable dose (MTD), blood hrBMP4 levels, MRI tumor changes, survival, and quality of life for up to 57 months.
The treatment proved very safe overall. No drug-related severe toxicities or dose-limiting side effects emerged, so MTD was not reached. Most mild-to-moderate symptoms (headache, vomiting, weakness) stemmed from catheter surgery or steroid drugs, not hrBMP4. Only three patients had temporarily elevated blood hrBMP4 that returned to normal within four weeks without harm.
Notably, 3 out of 15 patients showed lasting tumor shrinkage, including two complete local tumor responders with exceptional long-term survival. Patient 2 received the lowest 0.5 mg hrBMP4 dose for a second recurrent temporal glioblastoma. After infusion, the treated tumor slowly shrank steadily over 12 months with no new tumor growth. At the 57-month follow-up, the patient remained clinically stable with improved movement and balance, and never needed extra chemo or surgery afterward. Patient 15 got the highest 18 mg hrBMP4 dose for a treatment-resistant parietal lobe recurrence. The lesion regressed gradually over 18 months and achieved full radiological clearance. Thirty months post-treatment, this injected tumor area stayed completely gone, even though a separate untreated frontal tumor later progressed. The third responder only had partial tumor shrinkage before later progression. Another key finding: tumor regrowth almost always happened outside brain regions covered by hrBMP4 infusion, while drug-permeated zones rarely saw new tumor growth. Median overall survival across all patients hit 7 months, longer than the typical 5.6 months from common second-line lomustine chemotherapy.
This pioneering first-in-human study confirms local hrBMP4 via CED is safe and tolerable for recurrent glioblastoma. The two long-term complete remissions strongly validate differentiation therapy targeting GSCs. Its main limitations are incomplete tumor coverage by CED catheters and lack of pre-trial molecular patient screening. The promising anti-tumor signals support launching larger Phase 2 trials to formally test hrBMP4’s therapeutic power, with technical improvements to boost drug coverage and predictive biomarkers to select patients most likely to respond.
Reviews
There are no reviews yet.