A method for identifying neoantigens through isolation of circulating tumor cells using apheresis among patients with advanced-stage cancer

Daiki Kobayashi¹Takuya Kosumi²Queenie Lai Kwan Lam³Shigeharu Fujita³Yasuki Hijikata⁴Kaori Takeda⁵Tomoya Narita⁶Naomi Yamashita⁶Guilhem Richard⁷Anne Searls De Groot⁷Naohide Yamashita^8*

• ¹Tokyo Medical University Ibaraki Medical Center, Ibaraki, Japan
• ²Clinic Grandsoul Nara, Nara, Japan
• ³Meiko CIT Clinic, Tokyo, Japan
• ⁴Hijikata Clinic, Nagoya, Japan
• ⁵Biomedica Solution Inc, Osaka, Japan
• ⁶University of Musashino, Tokyo, Japan
• ⁷EpiVax, Inc, Providence, Rhode Island, United States
• ⁸Novacellum Inc., Tokyo, Japan

Background: Immune checkpoint inhibitors show limited efficacy in tumors with low tumor mutational burden, partly due to insufficient neoantigen presentation.Methods: We developed a novel approach for neoantigen identification using circulating tumor cells (CTCs) isolated via leukapheresis and flow cytometry. Peripheral blood mononuclear cells (PBMCs) were collected from 11 stage IV cancer patients and 2 healthy volunteers. CTCs were enriched by depleting CD45⁺ hematopoietic cells and selecting CD45⁻Vimentin⁺ cells, which were confirmed cytologically to contain malignant cells. Hematopoietic lineage analysis showed that over 50% of the CTC fraction consisted of non-hematopoietic cells. DNA extracted from both the CTC and normal hematopoietic fractions underwent exome sequencing. Neoantigens were identified using the Ancer® bioinformatics platform. Results: In representative patients with gastric and salivary gland cancers, 94,636 and 46,423 CTCs were isolated, respectively. DNA yields were sufficient for exome sequencing without amplification or extensive cell culture. A total of 102 (patient with gastric cancer) and 108 (patient with salivary gland cancer) neoantigens were identified in each subject, including high-ranking T-cell epitopes derived from single nucleotide variants and frameshift mutations. According to the same procedures we could successfully identify a large number of neoantigens from the CTCs of all stage IV cancer patients. This confirms the feasibility of identifying individual patient-specific neoantigens from CTCs without requiring tumor biopsies. Conclusions: This is the first study to demonstrate successful neoantigen identification using non-amplified CTCs isolated by apheresis and flow cytometry. The approach provides a minimally invasive, scalable alternative for neoantigen discovery and may better capture tumor heterogeneity compared to single-site biopsies. This method holds promise for enabling rapid, personalized immunotherapy strategies, including peptide vaccines, dendritic cell vaccines, and mRNA-based treatments.