IDO1 staining has often been treated as a proxy for tryptophan catabolism in tumors, but enzyme expression may not prove metabolite production at the tumor site. Using a monoclonal antibody against L-kynurenine, the team profiled FFPE tissue microarrays from colorectal and breast cancers by IHC and compared metabolite staining with IDO1 expression in colorectal cancer. L-kynurenine was present in 20.3% of colorectal and 26.1% of breast tumors, while IDO1 and L-kynurenine did not strictly overlap — a practical warning for IDO/TDO biomarker strategies.
This PLOS ONE paper was co-authored by Alban Bessede and supported by Immusmol, the mother company of Explicyte, which conducted early biomarker work in tryptophan metabolism. The team addressed a translational biomarker problem that remains relevant for immuno-oncology drug development: how to identify tumors in which tryptophan catabolism is active locally. Rather than relying only on IDO1, IDO2, or TDO2 enzyme expression — or systemic kynurenine/tryptophan ratios — the study tested whether L-kynurenine itself could be detected directly in FFPE tumor tissue.
L-kynurenine was conjugated to BSA and used to immunize BALB/c mice over a 2-month period. Hybridoma clones were screened by ELISA, and clone 3D4-F2 showed the strongest affinity for BSA-conjugated L-kynurenine, at 5 × 10^-10 M. The antibody showed no detectable reactivity against related kynurenine-pathway metabolites, including 3-hydroxykynurenine, anthranilic acid, kynurenic acid, xanthurenic acid, 3-hydroxyanthranilic acid, and quinolinic acid.
The team stained 5 µm FFPE tissue microarray sections from colorectal cancer and breast cancer cohorts, using two independent cores per case. L-kynurenine IHC was scored semi-quantitatively from 0 to 3, with positivity defined as ≥1 in colorectal cancer and ≥1.5 in breast cancer. L-kynurenine was detected in 14 of 69 colorectal cancer samples (20.3%) and 18 of 69 breast cancer samples (26.1%).
In positive tumors, L-kynurenine accumulated mainly in the cytoplasm of tumor cells, consistent with intracellular production through the kynurenine pathway. Occasional stromal staining was also observed in the tumor microenvironment. The paper notes that L-kynurenine staining was heterogeneous across tumor regions, reinforcing why two cores per case were evaluated.
The same colorectal cancer cohort was stained for IDO1 by IHC. IDO1 was positive in 9 of 69 samples (13%), but its expression did not strictly match L-kynurenine production. Three IDO1-positive tumors lacked detectable L-kynurenine, while some L-kynurenine-positive tumors were IDO1-negative, suggesting that IDO2, TDO2, or other local biology may contribute to metabolite accumulation.
The team assessed whether L-kynurenine or IDO1 immunoscores associated with clinical and histopathological variables, including tumor grade, tumor size, lymph node invasion, and metastases. No statistically significant association was identified in either tumor type. The main signal was therefore biological and biomarker-oriented: enzyme expression alone was not enough to infer local L-kynurenine production.
The paper reframed tryptophan-catabolism biomarker strategy around direct metabolite detection in tissue. For IDO1, IDO2, and TDO2 programs, the key issue is not only whether an enzyme is present, but whether the tumor site accumulates immunosuppressive kynurenine.
For drug developers working on IDO1, IDO2, TDO2, AhR, or broader kynurenine-axis modulation, this paper argues for biomarker strategies that measure pathway output at the tumor site. A tissue-compatible L-kynurenine readout could help translational teams separate enzyme expression from functional metabolite accumulation, improving patient stratification logic for immunometabolism programs.
