Introduction
Glioblastoma (GBM) is the most frequent type of primary malignant brain tumor, with an estimated 13,000 new cases in the US in 2023 [
1]. Despite standard surgical resection followed by radiation therapy (RT) and temozolomide (TMZ), the median overall survival (OS) for patients with unmethylated
O6-methylguanine-DNA methyltransferase (MGMT) is 12.6 months and with methylated MGMT is 23.4 months, respectively [
2,
3]. Although immunotherapy has improved the outcomes associated with many types of cancer, most clinical trials examining different approaches for GBM have been disappointing [
4‐
8]. The unique features of the immune tumor microenvironment (TME) in the central nervous system (CNS) system and the overwhelming immunosuppression associated with GBM make therapeutic manipulation of the immune system a formidable challenge. To be effective, immunotherapy-targeting GBM needs to overcome systemic and local suppression, cross the blood-brain barrier (BBB), and induce a therapeutic anti-tumor effect.
Glioblastoma cells overexpress epidermal growth factor receptor (EGFR) to varying degrees and have been a target for previous immunotherapies, though with unsatisfactory results [
9‐
11]. Adoptive immune cell therapy has the potential to overcome the problem of delivery and adaptive mechanisms of resistance thought to be, at least in part, responsible for the failure of therapeutically targeting EGFR in GBM [
12]. Anti-CD3 monoclonal antibody activated T cells (ATC) armed with anti-CD3 x anti-EGFR bispecific antibody (EGFR BiAb) target T cells to tumor-associated antigen (TAA) on solid tumors in a non-MHC restricted manner [
13]. Our preclinical studies showed that EGFRBi-armed T cells (EGFR BATs) lyse both established and patient-derived GBM cell lines, and induce multiple rounds of proliferation, endogenous tumor-specific cytotoxicity, and Th
1 cytokine release without affecting T cell viability. The range of antitumor and immunostimulatory complementary functions may overcome the limitations of other types of cancer immunotherapy [
11,
14].
This phase I study in patients with astrocytoma grade 4 (AG4) [
15] examines whether infusions of EGFR BATs in combination with standard radiation and chemotherapy are safe and feasible and if the therapy induces systemic anti-tumor immunity. This study shows that a total of 80 × 10
9 EGFR BATs is safe without cytokine release syndrome (CRS) and induces anti-tumor immunity.
Discussion
To circumvent and overcome the limitations of monoclonal antibody therapy targeting EGFR alone in patients with GBM, we designed this phase I study based on our preclinical data that showed that the arming of ATC with EGFR BiAb exerts potent in vitro anti-tumor activity [
11]. The combination of intravenous administration of EGFR BATs after RT/TMZ in patients with newly diagnosed AG4 was safe using two different regimens: every 4 weeks in combination with TMZ and weekly as the only adjuvant treatment after standard RT-TMZ. The starting total dose of 80 × 10
9 EGFR BATs given in eight divided doses was selected based on previous trials using EGFR BATs to start the dose escalation with the highest safe dose possible to maximize the trafficking of BATs across the BBB [
22]. Expanding cells from single apheresis was not sufficient to attain the second dose level. The safe and feasible phase II recommended dose of 80 × 10
9 cells in eight divided infusions triggered adaptive and endogenous anti-tumor cellular immune responses in PBMC of patients with AG4 with the potential to translate into therapeutic benefit.
A concern with this immunotherapy strategy is that it might cause a CRS similar to that seen after chimeric antigen receptor (CAR)T cell infusions, but we found no DLTs, and most of the observed adverse events resolved within 24–48 h after the infusion. This side effect profile was similar to our previous experience using BATs in patients with other types of cancer [
21‐
24]. A patient developed a symptomatic intratumoral hemorrhage, but the tumor had bled before immunotherapy. The safety profile of EGFR BATs in patients with GBM will be confirmed in a larger phase II cohort.
Patients with GBM are immunosuppressed with baseline lymphopenia, have T-cell dysfunction, and exhibit higher proportions of Tregs in peripheral blood and the tumor microenvironment [
25,
26]. Although the T cells in the apheresis product before RT/TMZ did not contain Tregs, T cell expansion was insufficient to fully explore dose level 2. However, EGFR BATs products from patients with AG4 and healthy controls showed comparable in vitro cytotoxicity (data not shown). Despite impaired ex vivo proliferative responses, ATC exhibited cytotoxic function after arming with EGFRBi.
Although our preclinical studies showed that EGFR BATs maintain their cytotoxicity after radiation and TMZ, it is possible that TMZ blunted the anti-tumor effect of immunotherapy [
11]. Because the benefit of TMZ in patients with unmethylated MGMT GBM is controversial and its withholding for this group has been proposed [
27,
28], we expanded the study to include three patients with unmethylated MGMT who received weekly EGFR BATs without adjuvant TMZ. The results of this cohort provided information for the design of a phase II study to include patients with unmethylated MGMT GBM.
Multiple infusions of EGFR BATs significantly increased GBM-specific CTL and NK activity, and serum concentrations of Th
1 cytokines and chemokines. The evaluation of anti-tumor-specific cytotoxicity was limited to using tumor cell lines because we could not grow cell lines from the patient’s tumors. These findings are consistent with our earlier studies showing that PBMC isolated from patients after multiple infusions of BATs exhibit significant anti-tumor cytotoxicity and IFN-γ EliSpots responses to other types of cancer cell lines [
14,
20,
22,
29,
30]. A potential advantage of our strategy is that the ability of EGFR BATs to recruit and activate endogenous immune cells in the TME may enhance systemic-specific cellular and humoral tumor immunity [
18]. In combination, these data will inform clinical trials that examine the efficacy of this immunotherapy in patients with GBM.
Enrollment was limited by our exclusion of 20% of the patients with high IgE alpha-gal in serum that cross-react to cetuximab, a component of EGFRBi known to increase the risk of anaphylactic reactions [
17,
31]. Before instituting the exclusion, an earlier trial using EGFR BATs in seven patients with unresectable and metastatic pancreatic cancer was safe without anaphylactic reactions [
22]. In our clinical pipeline, a molecularly engineered recombinant anti-CD3 x anti-EGFR BiAb (rEGFRBi) with deletion of the sequence responsible for allergic reactions, has markedly enhanced cytotoxicity, and induced Th
1 cytokine secretion when engaging multiple EGFR + tumor cell lines including GBM [
32].
Although by IHC we detected EGFR in all nine samples examined (one sample was not available), there remains a concern that low or nil EGFR expression on GBMs would preclude targeting by EGFR BATs. Our preclinical studies showed that HER2 BATs could efficiently kill MCF-7 cells, a cell line with negative to low HER2 expression by flow cytometry [
20]. In our phase I trial including patients with HER2-negative metastatic breast cancer (MBC) [
21], we observed a near complete response in a patient with HER2-negative liver metastases and found a lack of correlation between HER2 status and survival. These results suggest that armed T cells may kill tumors with low or near nil TAA expression in the clinical setting.
Our study shows that EGFR BATs were produced from PBMC of patients with AG4 before receiving concomitant RT/TMZ, but that in some instances, T cell expansion was limited by tumor-induced immunosuppression or other inherent T cell dysfunction. We demonstrated that up to eight infusions of 10 × 10
9 EGFR BATs were feasible, safe, and well tolerated in both every 4-week and weekly regimens while inducing cellular and cytokine/chemokine anti-glioma immune responses. These results support further studies in patients with GBM to demonstrate the efficacy of the safe and feasible dose of 80 × 10
9 EGFR BATs. Future trials will involve an optimized rEGFRBi with the potential of enhanced T cell-mediated antitumor cytotoxicity [
32], labeling of T cells to track trafficking into the GBM microenvironment, and combining EGFR BATs with other immune stimulatory agents.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.