KL-50: A novel therapeutic agent targeting MGMT-deficient glioblastoma

Glioblastoma, isocitrate dehydrogenase (IDH) wild type, is the most common primary intracranial malignant tumor, with a median survival of less than 2 years.^1,2 While numerous studies have proposed novel treatment strategies, few have successfully translated into clinical practice. Recent efforts have focused on targeted therapies aimed at blocking glioblastoma recurrent drivers, immunotherapies with various strategies, or anti-angiogenic agents. The DNA alkylating agent temozolomide (TMZ), combined with radiotherapy and tumor-treating fields, is still the standard postsurgical treatment.³ However, most of the patients eventually acquire resistance to TMZ through mismatch repair (MMR) pathway deficiency.⁴ Only a few trials have tested strategies to optimize the use of currently available alkylating agents, increasing their activity by developing novel, more potent compounds, or combining them with DNA damage response inhibitors.⁵

Mechanistically, TMZ induces O⁶-methylguanine lesions, which lead to DNA double-strand breaks and subsequent tumor cell death mediated by the recruitment of MMR at the site of DNA damage in MGMT-deficient cells. Consequently, cells inactivating the MMR pathway through mutations or epigenetic mechanisms develop high resistance to TMZ.⁶ Lomustine (CCNU) is a nitrosourea that acts independently of MMR by introducing inter-strand DNA crosslinks and leads to carbamoylating amino acids.⁷ MMR-deficient cells can sometimes retain sensitivity to CCNU,⁶ but the toxicities (eg, hematological, lung) associated with CCNU limit its use in the clinic. A phase III trial combining CCNU with the standard treatment demonstrated improved overall survival in newly diagnosed MGMT-deficient glioblastoma.⁵ However, significant toxicities, with over half of the patients experiencing adverse events of grade 3 or higher, have limited its clinical utility.

Recently, KL-50, an imidazotetrazine-based compound, has emerged as a potential alternative with potent efficacy in MGMT-deficient models, regardless of their MMR status.⁸ To provide healthy cells with more time for MGMT to repair alkyl groups, the researchers innovatively substituted the chloroethyl group in CCNU with fluoroethyl—a slower cross-linking molecule. This modification allowed healthy cells to recover while still effectively targeting and damaging MGMT-deficient cancer cells. In long-term survival assays and extensive screening across a wide range of cancer cell lines, the molecule demonstrated selective toxicity toward MGMT-deficient cells. These results underscore its potential as a promising therapeutic agent for gliomas, as well as other cancers like melanoma and colorectal cancer. In this context, McCord and colleagues investigated the therapeutic potential of KL-50 in an in vivo MMR-deficient patient-derived xenograft (PDX) model.⁹ By orthotopically engrafting glioblastoma PDX models and inducing MMR deficiency through in vivo exposure to the TMZ chemotherapy, the authors successfully recapitulated the development of TMZ-induced MMR deficiency in a subset of offspring engrafts. This innovative model offers a promising platform for studying TMZ-induced resistance mechanisms and evaluating alternative therapies. KL-50 showed a major survival benefit in MGMT/MMR-deficient PDX mice, underscoring its potential efficacy in this challenging subset of glioblastoma. Additionally, in engineered MSH6 knock-out patient-derived cell lines, KL-50 demonstrated superior efficacy compared to both TMZ and CCNU, highlighting its potential to improve outcomes of patients with MGMT-deficient gliomas.

These results established a strong preclinical rationale for the further development of KL-50 as a therapeutic agent for MMR-deficient recurrent glioblastoma. The role of MMR deficiency in driving resistance to TMZ has been recognized for nearly 2 decades.¹⁰ However, it took a dedicated collaboration between chemists and cancer researchers to develop an effective strategy to overcome this key resistance mechanism in MGMT-deficient cells. This study by McCord and colleagues represents an important step toward addressing the unmet need for effective treatments in MGMT-deficient patients. These results highlight the importance of further research focused on optimizing existing treatments (eg, CCNU, radiation, tumor-treating fields treatment). Such efforts have the potential to significantly improve patient outcomes by identifying novel biomarkers and developing innovative strategies rooted in a deeper understanding of resistance mechanisms.

Together, KL-50 leverages a selective toxicity profile to target resistant tumor populations while sparing healthy cells. The introduction of fluoroethyl chemistry in KL-50 serves as an example of reducing off-target toxicities and addressing the unmet clinical need. Beyond its efficacy in glioblastoma, KL-50’s performance across diverse malignancies underscores its broader applicability in oncology. Future research should investigate biomarker-driven patient stratification and combinatorial regimens to maximize therapeutic outcomes. Furthermore, the PDX models used in these studies offer a platform to study resistance mechanisms and precision oncology approaches in glioblastoma treatment.

Conflict of interest statement

M.K. reports grants paid to his institution, or contracts from BMS, AbbVie, BioNTech, CNS Pharmaceuticals, Daiichi Sankyo Inc., Immorna Therapeutics, Immvira Therapeutics, JAX lab for genomic research, and Personalis Inc. M.K. also received consulting fees from AnHeart Therapeutics, George Clinical, Manarini Stemline, and Servier and is on a data safety monitoring board for BPG Bio. M.T. reports research grants from Sanofi. Consulting fees from Servier, Novocure, Resilience, and NH TherAguix. Honoraria for lectures/educational events from Servier, Novocure, and Ono. Advisory Board: Servier and Novocure. K.Z. declares no conflicts.

Authorship statement

Design and writing of the manuscript: K.Z. and M.K. with assistance from M.T. Revision and approval of the final version: all authors.

References