Will Targeting EVA1 With ADCs Eliminate Glioblastoma-Initiating Cells?

Toru Kondo, PhD

At the 16th Annual World Antibody Drug Conjugate (ADC) Summit, all the latest research surrounding ADCs was shared and discussed by various experts across the field. During one session, Toru Kondo, PhD, a distinguished professor from the Institute for Genetic Medicine at the University of Hokkaido in Hokkaido, Japan, presented a seminar titled “Evaluating EVA1-ADC as a Novel Therapeutic Strategy for Eliminating Glioblastoma- Initiating Cells”.

The seminar highlighted data on utilizing the previously unexplored EVA1 membrane protein as a target for eliminating glioblastoma-initiating cells. At the Summit, Kondo spoke with CancerNetwork® about this research.

What separates EVA1 as a target is that normal stem cells don’t express it; thus, it may lead to fewer adverse events when compared with a popular target, such as CD133, which is expressed in normal stem cells.

Kondo hopes that this target may eventually be explored in humans in phase 1 or phase 2 trials. “We imagine that if we can get a good antibody to the EVA1, this antibody can be used for the elimination of glioblastoma-initiating cells,” said Kondo.

What was the rationale/background for your presentation on evaluating EVA1-ADC as a novel strategy for eliminating glioblastoma-initiating cells?

We are looking for which factors [are] upregulating or downregulating the glioblastoma-initiating cell. We used a microarray to check the expression levels of glioblastoma-initiating cells and non-tumor cells, such as normal stem cells or brain cells. We found that many factors upward and downward regulate. Among them, we focused on one membrane protein named EVA1…. From this data, we imagine that if we can get a good antibody to the EVA1, this antibody can be used for the elimination of glioblastoma-initiating cells. That means therapy for glioblastoma.

In the context of ADC-targeting glioblastoma-initiating cells, how important is it to have high-affinity antibodies for the success of the conjugate?

For instance, CD133 is a very famous antigen, but in my opinion, CD133 is expressed in normal neural stem cells, and also in other stem cells. That means if we target CD133, we are very much afraid of the [adverse] effects because, as you can understand, if we combine the anti–CD133 MMAE drug, these ADCs can target normal stem cells and kill the stem cells. That means that we cannot survive for a long time, so in contrast, EVA1 is not expressed in normal neural stem cells. Some of the differentiation cells express EVA1; however, normal stem cells don't express EVA1, so therefore the [adverse] effects are very low. Therefore, we believe that [this strategy is] the best strategy to kill the cancer cells alone.

What was the biggest takeaway from your seminar?

The big takeaway is that we found that the new antibody, which can specifically kill glioblastoma-initiating cells, but not normal cells…and normal stem cells. We found that brown adipocytes express EVA1; therefore, our ADC can eliminate brown adipocytes. This is a negative point.

What has been the non-clinical performance of B2E5-ADC in EVA1 expressing cell lines and glioblastoma-initiating cells?

At the moment, we don't have a big run. Therefore, now this experiment is stopped at the stage of the pre-clinical, but once we can get a good background, we want to proceed with this experiment so that this antibody can be used for humans in [phase] 1 or 2 [trials] in the clinical [setting].

Reference

Kondo T. Evaluating EVA1-ADC as a novel therapeutic strategy for eliminating glioblastoma-initiating cells. Presented at the 2025 World Antibody Drug Conjugate Conference; San Diego, CA; November 3-6, 2025.