KYNU Reconsidered: A Missing Piece in the IDO1 Inhibitor Story?

📍Key Takeaways
1. CD8 T cells can’t degrade kynurenine — KYNU is missing. That’s why T cell suppression persists even when IDO1 is blocked.

2. G-APCs express 32× more KYNU than M-APCs. The APC subtype in the tumor microenvironment may determine whether kynurenine-induced suppression happens at all.

3. KYNU-expressing G-APCs restore CD8 T cell function — without direct cell contact. Metabolic context, not just signaling, drives immune outcomes in the TME.

The Hypothesis: Block IDO1, Restore Immunity

For years, IDO1 inhibition was considered one of the most promising strategies in tumor immunotherapy. The logic was straightforward: cancer cells overexpress IDO1, an enzyme that converts tryptophan into kynurenine. As kynurenine accumulates in the tumor microenvironment, it enters CD8 T cells — the immune cells that directly kill cancer cells — and shuts them down. It does this by activating AHR (aryl hydrocarbon receptor), which suppresses IFN-γ production and disables cytotoxic T cell function.

Block IDO1, reduce kynurenine, rescue T cells, attack the tumor. The reasoning was clean and compelling.

What Actually Happened

In 2019, that hypothesis met a phase 3 clinical trial. The combination of the IDO1 inhibitor epacadostat and the PD-1 inhibitor pembrolizumab failed to improve progression-free survival or overall survival in patients with unresectable or metastatic melanoma (ECHO-301/KEYNOTE-252). Multiple IDO1 inhibitor programs were subsequently scaled back or discontinued.

The question that remained: why didn’t it work?

The Missing Piece: Who Degrades the Kynurenine?

A study published in Cell Reports in April 2026 by Giacomantonio et al. offers a compelling answer. The argument is this: even if IDO1 is inhibited and kynurenine production is reduced, T cell suppression won’t resolve if there’s no cell in the environment capable of degrading the kynurenine that’s already there.

The core problem is that CD8 T cells take up kynurenine efficiently but cannot break it down. They lack appreciable expression of KYNU (kynureninase), the enzyme responsible for kynurenine degradation. Once kynurenine enters the cell, it continues activating AHR unchecked, keeping T cells in a suppressed state.

So where is KYNU expressed?

KYNU Is Selectively Expressed in Specific Antigen-Presenting Cells

The research team compared two antigen-presenting cell (APC) models — immune cells that present antigens to T cells and shape their responses.

G-APCs (differentiated with GM-CSF, comprising macrophages and dendritic cells) express 32-fold more KYNU protein than M-APCs. When kynurenine is added to their culture environment, G-APCs take it up and break it down into downstream metabolites including anthranilic acid, picolinic acid, and quinolinic acid.

M-APCs (differentiated with M-CSF) express negligible KYNU and cannot degrade kynurenine, which accumulates in the culture medium unchanged.

Two additional findings are worth noting. First, KYNU activity in G-APCs operates independently of IDO1 expression. Second, kynurenine catabolism in these cells does not contribute to de novo NAD⁺ synthesis. This means G-APCs are not degrading kynurenine for energy metabolism — the function appears to be specifically immunoregulatory. The same pattern was confirmed in alveolar macrophages collected in situ: high KYNU, low IDO1, functionally analogous to G-APCs.

Confirmed Experimentally, Step by Step

The team validated the causal chain through conditioned media and co-culture experiments.

CD8 T cells treated with conditioned media from wild-type G-APCs showed significantly reduced kynurenine-induced AHR activation (Cyp1a1 and Cyp1b1 expression). This effect disappeared when using media from KYNU-knockout G-APCs (KYNUem1), and was entirely absent in AHR-deficient CD8 T cells. The pathway — KYNU degrades kynurenine → AHR activation is suppressed → T cell function is restored — was confirmed at each step.

Going further, when IDO1-overexpressing TC1 lung cancer cells were co-cultured with G-APCs, the G-APCs degraded cancer cell-produced kynurenine and restored IFN-γ secretion in CD8 T cells. The same result held in transwell experiments, where cells were physically separated. Direct cell contact was not required — lowering ambient kynurenine concentration was sufficient.

What This Means Going Forward

The study opens three directions worth watching.

Reinterpreting IDO1 inhibitor failures. In tumor microenvironments where G-APCs are scarce or KYNU expression is low, IDO1 inhibition alone may be insufficient to relieve kynurenine-mediated T cell suppression. APC subtype composition could be a critical biomarker for predicting IDO1 inhibitor response in future trials.

Targeting KYNU directly. PEGylated KYNU enzyme therapy and KYNU-expressing CAR-T cells have already shown preclinical promise. This study adds another avenue: leveraging endogenous G-APCs in the tumor microenvironment as a therapeutic strategy.

Lung cancer implications. Alveolar macrophages share functional characteristics with G-APCs, and lung tissue was found to contain abundant kynurenine. How the KYNU expression status of alveolar macrophages influences immune function in lung cancer is an open question for follow-up research.

The authors themselves note the limitations: efficacy of KYNU-expressing APCs in in vivo tumor models, effects on other immune cell types, and clinical significance remain to be established.

---

Blocking IDO1 alone was not enough. Without a cell capable of actually degrading kynurenine, T cell suppression persists. This study is the first to clearly identify which cells play that degradation role — and why it matters for tumor immunotherapy.

---

---

References

Giacomantonio MA, et al. Subversion of kynurenine-induced AHR activation in CD8 T cells by kynureninase-expressing antigen-presenting cells. Cell Reports. 2026. https://doi.org/10.1016/j.celrep.2026.117149

Long GV, et al. Epacadostat plus pembrolizumab versus placebo plus pembrolizumab in patients with unresectable or metastatic melanoma (ECHO-301/KEYNOTE-252). Lancet Oncol. 2019;20:1083–1097.