Conference · for clinicians

SITC 2026 (Phoenix, 4–8 Nov): three early-phase immunotherapy signals worth knowing

11/9/2026 · 4 min read · updated 9 Nov 2026

Bottom line

Hypothesis-generating only. None changes practice today; each now moves to a registration-enabling or randomised trial.

The Society for Immunotherapy of Cancer met in Phoenix on 4–8 November 2026 (41st Annual Meeting, ~1,200 abstracts). Most of the meeting is translational. Three clinical late-breakers stand out for practising oncologists. All are phase 1 or 1/2, single-arm and small. Figures below are from company releases and the meeting abstracts.

1. Amsoki-cel: TIL therapy without high-dose IL-2 (melanoma)

What: amsokigene autoleucel (formerly OBX-115), an engineered TIL product carrying drug-regulatable membrane-bound IL-15, so no IL-2 is given after infusion. Low-dose lymphodepletion only; tumour can be harvested by core needle biopsy.

Who: advanced melanoma progressing after immune checkpoint inhibitors; 93% had prior ipilimumab + nivolumab.

Results (Agni-01, recommended dose, n=15; cutoff 25 Sep 2026):

ORR 67% (2 CR, 8 PR); median PFS 8.3 months; median DoR not reached at 14.3 months’ follow-up.
33% (5/15) still in response 12 months after a single infusion.
Pooled safety (n=37): no DLTs, treatment-related deaths, ICANS or ICU transfers; one grade 3 CRS.
Published in Nature Medicine on 3 Nov 2026. Registration-enabling cohort topline expected by end-2027.

Why it matters: the IL-2 phase is what makes current TIL therapy (lifileucel) an ICU-level commitment. Removing it could widen who can receive TIL.

Amsoki-cel at recommended dose (n=15)

Objective response67%

Still responding at 12 months33%

Obsidian Therapeutics, SITC 2026. Single-arm phase 1/2; small numbers.

2. Two T-cell engagers together: brenetafusp + tebentafusp (uveal melanoma)

What: tebentafusp (gp100 × CD3, current standard for HLA-A*02:01-positive metastatic uveal melanoma) plus brenetafusp (PRAME × CD3). Both need HLA-A*02:01, so one blood test selects for either.

Results (first-line combination, n=17 evaluable; June 2026 cutoff):

ORR 29%; median DoR 24.1 months; DCR 71%; median PFS 8.1 months.
OS 88% at 1 year and 70% at 2 years; median OS not reached at 29 months’ median follow-up.
ctDNA: 8/9 evaluable had a molecular response; 4/9 cleared ctDNA.
CRS in 77%, no grade 3/4 CRS; grade 3–4 TRAEs 38% (mostly transient lymphopenia); no discontinuations or deaths from TRAEs.
Brenetafusp alone, second line or later (n=16): ORR 19%, median OS 20.3 months.

Next: randomised phase 2/3 DUET-UVEAL vs tebentafusp alone. As with tebentafusp, RECIST response understates benefit; OS and ctDNA are the more meaningful signals.

3. Cesalatamig: PD-1 × VEGF bispecific alone after immunotherapy (NSCLC)

What: cesalatamig (AI-081), a tetravalent, Fc-silenced PD-1 × VEGF bispecific, the same class as ivonescimab.

Who: NSCLC resistant or refractory to PD-(L)1 therapy; median 2 prior lines (42% had ≥3). Safety population n=103 (BiPAVE-001 US and China).

Results (n=55 evaluable, investigator-assessed):

Dose (every 3 weeks)	n	ORR	DCR
1 mg/kg	10	10%	40%
3 mg/kg	24	21%	67%
10 mg/kg	21	33%	90%

Clear dose–response; responses across PD-L1 levels. 10 mg/kg chosen for a planned pivotal study. FDA Fast Track (Sep 2026).
Caveat: investigator-assessed, unblinded, small dose cohorts.

Context: after IO and chemotherapy, docetaxel-based therapy still gives modest results. The PD-1 × VEGF class is also awaiting an FDA decision for ivonescimab in EGFR-mutant NSCLC on 14 Nov (see our post-osimertinib note).

Bottom line

Engineered TIL without IL-2, dual T-cell engagers and PD-1 × VEGF bispecifics are the three themes to follow into 2027.
None should change treatment now; each needs randomised data.

Sources

Educational summary, not medical advice. Opinions are labelled as such and are the author's own. Approvals and availability in India (CDSCO) may differ from the US/EU.

← Back to blog