← Science Nature Medicine · Nov 3, 2026

Regulatable IL-15 engineered TIL therapy in ICI-resistant melanoma: RP2D ORR 66.7% (10/15)

Amsokigene autoleucel (OBX-115) expresses acetazolamide-regulatable membrane-bound IL-15, removing the need for exogenous IL-2. In pooled phase 1 and 1/2 data in immune checkpoint inhibitor (ICI)-resistant advanced melanoma, 10 of 15 patients at the recommended phase 2 dose (RP2D) achieved objective responses (66.7%; 95% CI 38–88%), including two complete responses. No dose-limiting toxicities, ICU admissions, or treatment-related deaths occurred.

Mechanism figure
Engineered TILs expressing regulatable membrane-bound IL-15 expand without exogenous IL-2. Illustration generated with AI; not a figure from the original paper.

Background and open questions

Tumor-infiltrating lymphocyte (TIL) therapy has shown clinical activity in advanced melanoma, but current regimens require high-dose exogenous IL-2 to support T-cell expansion and persistence, which causes substantial toxicity including capillary leak syndrome. Whether a pharmacologically controllable cytokine signal built into the cell product could support TIL function without systemic IL-2 remained untested in humans.

Study design

The investigators conducted two single-arm early-phase trials (phase 1 and phase 1/2) in patients with checkpoint-inhibitor-resistant advanced melanoma. Amsokigene autoleucel (OBX-115) is an autologous TIL product engineered to express a membrane-bound form of IL-15 whose activity is switched on by the small-molecule drug acetazolamide. Patients received lymphodepleting chemotherapy followed by TIL infusion without exogenous IL-2. The pooled analysis evaluated 37 treated patients, with 15 at the recommended phase 2 dose (RP2D).

Key results

  • At the RP2D (Key Efficacy Set: phase 1 DL3 plus phase 2, 15 patients), 10 achieved objective responses (66.7%; 95% CI 38–88%), including 2 complete responses; in the phase 2 cohort alone (9 patients), ORR was 6/9 (66.7%).
  • The first patient experienced rapid lymphocyte expansion and edema around a previously irradiated brain metastasis (not meeting DLT criteria), prompting protocol modifications: maximum cell dose reduced 80%, acetazolamide dose reduced 75%.
  • Common grade ≥3 treatment-emergent adverse events (TEAEs) included neutropenia (64.9%), leukopenia (56.8%), and lymphopenia (56.8%), primarily attributable to lymphodepleting chemotherapy.
  • No dose-limiting toxicities, ICU admissions, or treatment-related deaths occurred; serious TEAEs occurred in 10/37 patients.
  • Median follow-up was 8.3 months; median PFS was 8.1 months.

Mechanistic interpretation

The data provide human proof-of-concept that a membrane-bound cytokine signal under small-molecule control can support TIL expansion and antitumor activity without the systemic toxicity of high-dose IL-2. The acetazolamide-dependent switch offers a potential safety lever if uncontrolled expansion occurs.

Limitations

  • The dataset is small and nonrandomized; the pooled analysis (post hoc) does not establish comparative efficacy versus approved TIL products or other melanoma therapies.
  • These are early-phase data; durability, long-term safety, and efficacy in larger populations remain to be determined.
  • The dose-adjustment history after the first patient's adverse event (which did not meet DLT criteria) underscores that the optimal regimen is still being refined.

Authors and source

Nature Medicine (2026). Published online 3 November 2026. https://doi.org/10.1038/s41591-026-04743-6.

Numbers are as reported by the authors. Corrections: contact@inlightbio.com.

Related science

Get new science briefs every week

One email, with links to every paper. Reports and custom landscapes: contact@inlightbio.com.

Opens your email app with a pre-filled request; we add you by hand. Free, unsubscribe any time.