Autologous tumor-infiltrating lymphocytes (GC101 TIL) in advanced melanoma: Extended follow-up of a randomized, open-label, pivotal phase II trial

MIZAR-003

Patient / Population Intervention / Exposure Comparison Outcome

In patients with anti-PD-1-resistant advanced melanoma, GC101 TIL with sintilimab improved independently assessed progression-free survival compared with investigator-choice chemotherapy.

N

101patients

Design

Multicentre, randomised 1:1, open-label phase II trial at 25 Chinese centres in patients with advanced melanoma refractory to at least two prior systemic regimens including anti-PD-1, comparing GC101 TIL plus sintilimab with investigator-choice chemotherapy (NCT06703398).

Endpoint

Progression-free survival assessed by independent review committee per RECIST v1.1.

ResultMedian IRC-assessed PFS 4.37 months (95% CI 2.99-5.55) with GC101 TIL vs 1.61 months (1.28-2.76) with investigator-choice chemotherapy (HR = 0.44, 95% CI 0.28-0.70; P = 0.0004). Median OS not reached vs 12.55 months (8.74-NR). ORR 44.0% vs 8.2% and DCR 68.0% vs 40.8%. In acral melanoma median PFS 4.37 vs 1.38 months (HR = 0.41, 95% CI 0.22-0.77; P = 0.0034) and in mucosal melanoma 4.81 vs 1.45 months (HR = 0.23, 95% CI 0.06-0.88; P = 0.0215). Grade ≥3 treatment-related adverse events occurred in 21/50 (42.0%) vs 15/49 (30.6%).

Si L, Chen Y, Zhang W, et al. Presented at ESMO Congress 2026, abstract LBA55, 23 October. Not yet published.

Discussion & critique

The extended analysis confirms a clear PFS advantage for GC101 TIL over chemotherapy in heavily pretreated, anti-PD-1-refractory melanoma, with assessment by an independent review committee partly offsetting the open-label design. With a median follow-up of 8.3 months OS is immature, so the benefit rests on a surrogate endpoint, and the TIL arm also received sintilimab, which means the contribution of the cell product cannot be separated from renewed PD-1 blockade. Chemotherapy is a modest comparator in this setting, and grade ≥3 treatment-related adverse events were more frequent with TIL. The subtype analyses are small, especially mucosal melanoma with 6 vs 10 patients, and the abstract does not report durability of response or how many randomised patients actually received the infusion.