Oncology · Medical Literature
Tumor-targeted interferon-α gene therapy for glioblastoma: a phase 1 trial.
Abstract
Glioblastoma (GBM) is an immunologically cold brain tumor with poor outcome, characterized by a myeloid-driven immunosuppressive microenvironment. Here we report an interim analysis of a first-in-human phase 1/2a dose-escalation study evaluating Temferon-a genetically engineered autologous stem cell transplant designed to deliver interferon-α2 by means of myeloid progeny recruited to the GBM TME and locally activate antitumor immunity. Twenty-four newly diagnosed patients with GBM and unmethylated MGMT promoter were treated across eight cohorts following surgical resection and radiotherapy, testing Temferon doses ranging from 0.5 × 106 to 4.0 × 106 CD34+ cells kg-1 and different conditioning regimens (BCNU or busulfan, with or without thiotepa). The primary endpoint was safety and tolerability within 90 days after infusion. Secondary endpoints included long-term safety, dose and conditioning regimen selection, Temferon engraftment, clinical response, quality of life and survival. No dose-limiting toxicities were observed up to the highest dose level of temferon tested. Busulfan conditioning was selected for further development. Adverse events included laboratory abnormalities, cytopenias and infections consistent with autologous stem cell transplant. Median overall survival and progression-free survival were 16.7 months and 8.1 months from diagnosis, respectively, with most patients maintaining good performance status and quality of life. Genetically engineered cells were detected long term in the bone marrow and the blood, where minimal amounts of interferon-α were measured. Temferon is a safe and tolerable immunotherapeutic strategy in patients with newly diagnosed GBM. ClinicalTrials.gov: NCT03866109 .
Clinical relevance
Is Temferon — an autologous hematopoietic stem cell product genetically engineered to deliver tumor-targeted interferon-α2 via myeloid progeny — safe, tolerable, and biologically active in newly diagnosed glioblastoma (GBM) patients with unmethylated MGMT promoter following surgical resection and radiotherapy? __SNAPSHOT__ {"studyDesign":"First-in-human, open-label, non-randomized, phase 1/2a dose-escalation trial (interim analysis); NCT03866109","setting":"Multicenter; Italy (including IRCCS Ospedale San Raffaele); academic medical centers","sampleSize":"n=24 treated patients across 8 cohorts (3 patients per cohort)","population":"Adults with histologically confirmed, newly diagnosed supratentorial GBM and unmethylated MGMT promoter who had undergone complete or partial tumor resection, were candidates for radiotherapy, had Karnofsky Performance Score ≥70, and life expectancy ≥6 months, with adequate cardiac, renal, hepatic, and pulmonary function","intervention":"Temferon: autologous CD34+ hematopoietic stem and progenitor cells (HSPCs) transduced with a lentiviral vector encoding IFN-α2 under a Tie-2 promoter/enhancer, delivered as a single infusion following conditioning chemotherapy (BCNU or busulfan ± thiotepa), after surgical resection and radiotherapy. Doses ranged from 0.5 × 10⁶ to 4.0 × 10⁶ CD34+ cells/kg across 8 cohorts.","comparator":"Not applicable (single-arm, non-randomized study); historical control data for unmethylated MGMT GBM referenced for contextual survival comparison","primaryOutcome":"Safety and tolerability within 90 days after Temferon infusion, including dose-limiting toxicities and adverse events","followUp":"Short-term primary safety endpoint: 90 days post-infusion; long-term secondary endpoints up to 2 years; some patients with extended follow-up beyond 3 years from surgery in long-term follow-up substudy (TEM-LT)","fundingSource":"Industry (Genenta Science); government/academic co-sponsorship not explicitly reported in abstract"}
Key findings
PRIMARY ENDPOINT — SAFETY: The trial met its primary endpoint. No dose-limiting toxicities (DLTs) were observed at any dose level, including the highest tested dose of 4.0 × 10⁶ CD34+ cells/kg. Adverse events were consistent with the expected toxicity profile of autologous hematopoietic stem cell transplantation (HSCT) and included laboratory abnormalities, cytopenias, and infections. Busulfan-based conditioning was selected over BCNU for further development based on efficacy and tolerability data from dose-escalation cohorts. ENGRAFTMENT AND BIOLOGICAL ACTIVITY: Genetically engineered cells were detected long-term in bone marrow and peripheral blood. Systemic IFN-α levels in the blood remained minimal, consistent with the tumor-selective Tie-2 promoter mechanism. Cerebrospinal fluid (CSF) analysis showed measurable IFN-α activity over time, suggesting progressive infiltration of Temferon-derived myeloid progeny into the tumor. In patients undergoing second surgery, tumor microenvironment (TME) reprogramming was documented, with increased inflammatory macrophages, CD8+ T cell accumulation, and upregulation of IFN-α and inflammatory gene signatures across myeloid, T cell, and stromal compartments. A ~3-fold increase in predicted tumor-reactive CD8+ T cells bearing expanded clonotypes was observed in Temferon-treated versus control patients (not reported as a formal statistical comparison in the abstract). SURVIVAL (SECONDARY ENDPOINT — DESCRIPTIVE): Median overall survival (OS) was 16.7 months and median progression-free survival (PFS) was 8.1 months from diagnosis. Two-year OS rate was approximately 29% (reported as 29% at one data cut and 30% [6/20 evaluable patients] at another). For historical context, median OS in unmethylated MGMT GBM with standard of care is approximately 12.7–15 months, and 2-year OS is approximately 14%. No formal statistical comparisons to historical controls were conducted or reported. Two patients achieved late partial responses more than one year after treatment without additional therapy, and two patients survived beyond 3 years from initial surgery. Most patients maintained good performance status and quality of life. No NNT/NNH calculable from available data.
Practice implications
This interim phase 1 analysis establishes that Temferon can be manufactured from autologous CD34+ HSPCs, successfully engrafted, and administered safely at doses up to 4.0 × 10⁶ cells/kg without dose-limiting toxicity — a meaningful safety milestone for a novel gene-cell therapy platform in a critically ill population. The tumor-selective IFN-α delivery mechanism (Tie-2 promoter restricting expression to myeloid cells in the TME) appears to function as designed, producing local but not systemic IFN-α exposure and thereby avoiding the well-known systemic toxicities of exogenous IFN-α therapy. Evidence of TME immune reprogramming — including CD8+ T cell infiltration, macrophage polarization shift, and IFN-α gene signature upregulation — supports the biological rationale and mechanism of action. WHAT WAS NOT ESTABLISHED: Efficacy was not formally established. The observed median OS of 16.7 months and 2-year OS of ~29% are numerically higher than historical unmethylated MGMT GBM benchmarks (~12.7–15 months and ~14%, respectively), but these comparisons are hypothesis-generating only. The absence of a concurrent randomized control arm, the small sample size (n=24), patient selection effects (KPS ≥70, adequate organ function, able to undergo HSCT), and the multiple conditioning regimen variations across cohorts preclude causal inference about survival benefit. The two patients with late partial responses and 3-year survival are notable but cannot be attributed to Temferon without controlled data. This summary is based on a published abstract and press/sponsor communications, as the full manuscript data were not available for review; reported figures should be interpreted accordingly. Patients with poor performance status, significant comorbidities precluding HSCT, or methylated MGMT promoter were not studied. PRACTICE IMPLICATION: No change in clinical practice is warranted at this time. Findings strongly support progression to a randomized phase 2/3 trial. __LIMITATIONS__ No randomized control arm — survival data compared only to historical, non-contemporaneous benchmarks, introducing substantial selection and lead-time bias,Very small sample size (n=24 across 8 heterogeneous cohorts, 3 patients per cohort) with limited statistical power for any efficacy inference,Multiple conditioning regimens (BCNU vs. busulfan ± thiotepa) varied across cohorts, confounding cross-cohort comparisons and making dose-response attribution difficult,Interim analysis only — long-term safety and efficacy data remain incomplete for most patients,Highly selected patient population (KPS ≥70, adequate multiorgan function, HSCT-eligible) limits generalizability to the broader GBM population,Industry sponsorship (Genenta Science) introduces risk of reporting bias; full peer-reviewed manuscript data not accessible for independent verification at time of this summary,Surrogate biological endpoints (TME reprogramming, CSF IFN-α, T cell clonotype expansion) are mechanistically compelling but not validated as predictors of clinical benefit in GBM,Two long-term survivors (≥3 years) cannot be causally attributed to Temferon without concurrent controls; spontaneous long-term survival occurs in a small proportion of GBM patients,Open-label design precludes blinding of outcome assessment,No formal statistical comparison between Temferon survival outcomes and historical controls was performed or reported __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ In this first-in-human phase 1 interim analysis (n=24), Temferon — a tumor-targeted IFN-α gene-cell therapy — was safe and well tolerated at all dose levels tested, with no dose-limiting toxicities, durable engraftment, and evidence of local TME immune reprogramming in newly diagnosed unmethylated MGMT GBM. Observed median OS of 16.7 months and 2-year OS of ~29% are numerically encouraging relative to historical benchmarks for this poor-prognosis subgroup, but the absence of a randomized comparator arm precludes any efficacy conclusion. A randomized controlled trial is required before this approach can be considered for clinical adoption.