A troublesome burden, the amplification of EGFR in glioblastoma!

See the article by van den Bent et al in this issue, pp. 684–693.

The INTELLANCE 2/EORTC 1410 randomized phase II study of Depatux-M is reported in this issue of the Journal. It is a well-designed and completed phase II study done by Dr Martin van den Bent and colleagues with the European Organisation for Research and Treatment of Cancer (EORTC).¹ Utilizing a tumor-specific antibody–drug conjugate (ABT-806 and the toxin monomethylauristatin-F), patients were randomized to either Depatux-M monotherapy, Depatux-M in combination with temozolomide (TMZ), or a control arm (TMZ or lomustine) at first recurrence of glioblastoma. The primary endpoint of the study was overall survival (OS) in the intent-to-treat population. Given the antibody–drug targeting to epithelial growth factor receptor (EGFR), all patients were centrally screened for EGFR amplification and the presence of EGFR variant (v)III mutation. However, this screening was from tissue at diagnosis, not at recurrence when patients enrolled on the study. Unfortunately, the study did not meet its primary endpoint at the primary analysis, but a trend toward a survival advantage did exist for the arm of Depatux-M when combined with TMZ. This was noted to have reached statistical significance (P = 0.017) in the long-term analysis with 2-year survival in the combination arm at 19.8%, the Depatux-M monotherapy arm 10%, and the control arm 5.2%. The subgroup analysis was most revealing for the combination arm time of relapse being >16 weeks, far outperforming the control arm (TMZ or lomustine) with 28.6% versus 3.9% alive at 24 months, suggesting a mechanism involving more than simply a retention of TMZ sensitivity.² The INTELLANCE 2 study highlights the challenges in neuro-oncology of targeted therapy and a disease with significant spatial as well as temporal heterogeneity.³

The identification of alterations in EGFR is among the earliest molecular events described in glioblastoma, both related to the disease and researcher’s efforts at molecular understanding of the disease. Extensive efforts have led to in-depth understanding of the role of EGFR in cellular growth, proliferation, and resistance.⁴ Likewise, significant investments in time, energy, and financial resources have sought to use this bench understanding to advance bedside opportunities through tyrosine kinase inhibitors, vaccine-based approaches, and targeted toxins (Table 1). Unfortunately, the general outcome has typically been similar to this study, where the primary endpoint is not met but a suggestion of activity within a subset appears to exist. It is clear that there are often flaws to study designs or simply challenges we lack the means or will to overcome. An example is: using the tissue from diagnosis for confirmation of EGFR amplification when temporal changes in expression and amplification of EGFR in glioblastoma is well described.⁵

Table 1.

Active and completed multi-institutional clinical trials of EGFR-targeted therapies in the past 10 years

Drug	Phase/Design	Target	Population	Endpoints	Results
Neratinib7 NCT02977780	II, randomized, adaptive platform trial, experimental arms (neratinib, abemaciclib, CC-115), control arm (TMZ)	EGFR amplification (neratinib), CDK4/6 amplification (abemaciclib), PI3K/ Akt/mTOR (CC-115)	Newly diagnosed glioblastoma, IDH1 wild-type, unmethylated MGMT promoter	OS	Active, recruiting
Rindopepimut (ReACT) NCT014983288	II, double-blind, randomized, experimental arm (bevacizumab+rindopepimut), control arm (bevacizumab+KLH)	EGFRvIII (rindopepimut), VEGF (bevacizumab)	First or second recurrence, bevacizumab naïve, EGFRvIII-positive, n = 73	PFS6	28% for rindo+bev and 16% in control arm (P = 0.12)
Rindopepimut (ACT IV)9 NCT01480479	III, double-blind, randomized, experimental arm (adjuvant TMZ+rindopepimut), control arm (adjuvant TMZ+KLH)	EGFRvIII	Newly diagnosed glioblastoma, EGFRvIII positive, completed standard chemoradiation, n = 745	OS	Terminated
Depatux-M INTELLANCE 2 NCT02343406	II, randomized, open-label, experimental arms (Depatux-M, Depatux- M+TMZ), control arm (TMZ or lomustine)	EGFR amplification (anti-EGFR ADC)	First recurrence, EGFR amplification, n = 260	OS	2-year survival 19.8% in Depatux- M+TMZ, 5.2% in control arm P = 0.017
Depatux-M INTELLANCE 1 NCT02573324	III, double-blind, randomized, experimental arm (Depatux-M + TMZ concurrent with RT followed by adjuvant Depatux-M +TMZ), control arm (RT+TMZ+placebo then TMZ+placebo)	EGFR amplification (anti-EGFR ADC)	Newly diagnosed glioblastoma, EGFR amplification	OS	Terminated
Afatinib10	I/II, phase II 3 arms; experimental arms (afatinib+TMZ, afatinib alone), control arm (TMZ)	EGFR amplification	First recurrence, n = 39 (AFA/TMZ), n = 39 (TMZ), n = 41 (AFA)	PFS6	23% TMZ, 3% AFA, 10% AFA+TMZ

Abbreviations: ADC, antibody–drug conjugate; AFA, afatinib; CDK, cyclin dependent kinase; KLH, keyhole limpet hemocyanin; PI3K/Akt/mTOR, phosphatidylinositol-3-kinase/protein kinase B/mammalian target of rapamycin; PFS6, six-month progression-free survival; VEGF, vascular endothelial growth factor.

We share the conclusions of the authors: there appears to be a role for a toxin targeted to EGFR therapy in a subset of glioma. It is noteworthy that grade 3 or 4 ocular toxicities were similar in phase I and II studies of Depatux-M, and the required dose reductions may have hampered the drug efficacy. Moreover, caution should be exercised not to overinterpret the data from the subgroup analysis in the Depatux-M plus TMZ combination arm due to the small sample and the lack of biological plausibility; as there was no improvement of OS in the Depatux-M monotherapy arm in the subgroup with the O⁶-methylguanine-DNA methyltransferase (MGMT) promoter unmethylated compared with the control arm.¹ The authors comment on the futility analysis of INTELLANCE 1, a companion phase III study in newly diagnosed glioblastoma casting a negative light on the potential positive findings within a subset of recurrent glioblastoma.⁶

In summary, this excellent and well-designed study shed light on the challenges of targeted therapy in glioblastoma. The study may promote the efforts to develop combinatorial therapies to overcome the resistance to anti-EGFR therapies in glioblastoma. Whenever feasible, verification of target expression in recurrent glioblastomas should be pursued prior to enrollment in clinical trials utilizing targeted therapies.