AACR Annual Meeting 2026

April 17–22, 2026 · San Diego · 2418 items

After the meeting★★

Immunotherapy: CAR-T in solid tumors, mRNA vaccines, TCR-T

12 people · 14 posts4 topics

Start here

CAR-T already has seven approvals in blood cancers and has struggled in solid tumors: messy targets, a microenvironment that traps T cells, short persistence. Personalized mRNA vaccines are the other path — a shot built on a patient’s own mutations, teaching the immune system the tumor.

Carl June’s opening talk put solid-tumor CAR-T on the main stage. Two numbers stick: after pancreatic surgery, 7 of 8 vaccine responders were still alive at 4–6 years; CAR-T in smoldering myeloma, before myeloma is myeloma, cleared MRD in every patient. Cell therapy is moving earlier, even into prevention.

Solid-tumor CAR-T is stuck on three walls: target choice (mesothelin, GD2, EGFRvIII, Claudin 18.2), the microenvironment (myeloid suppression, TGF-β, exhaustion), and cells that neither traffic nor persist. KIR-CAR, CAR-NK and TCR-T are different engineering bets around those walls.

Personalized mRNA vaccines are simpler on paper: sequence the tumor, build a shot, add a PD-L1 antibody and chemotherapy. MSK’s autogene cevumeran showed responders living more than three times as long as non-responders. Non-response is information too: the vaccine has to raise immunity before anyone talks about survival.

Analysis

Three things landing together is why this was not another “CAR-T concept talk”: KIR-CAR into humans, vaccine follow-up tied to survival, and CAR-T in people who do not yet have myeloma. The boundary moved from late treatment to early intervention.

Blood-cancer success does not copy into solid tumors. CAR-PRISM’s universal MRD negativity happened in a window of healthier T cells and lower burden; solid tumors do not have that window yet. Whether the 260-patient Phase 2 recreates 7 of 8 is the node this line stands or falls on.

Still open

Whether autogene cevumeran’s Phase 2 recreates the responder survival advantage.
Whether CAR-T in smoldering myeloma has long-term toxicity that justifies treating people who are not yet ill.
Whether any solid-tumor CAR-T target can move from anecdotal responses to something repeatable.

Key readout

autogene cevumeran · 7 of 8 responders alive at 4–6 years; 2 of 8 non-respondersResected PDAC · Phase 1; Phase 2 n=260 enrolling

Records

Drug / trial	Readout	Setting
autogene cevumeran#MRNA #PANCREATICCANCER	7 of 8 responders alive at 4–6 years; 2 of 8 non-responders	Resected PDAC · Phase 1; Phase 2 n=260 enrolling
CAR-PRISM#CART #MMSM	20/20 MRD-negative; no progression at 15.3 mo median follow-up	High-risk smoldering myeloma · concurrent Nature Medicine
EGFRvIII CAR#CART #CARTCELL	13 patients: 2 PR, 7 SD	GBM · June plenary
mRNA-4157#MRNA	Melanoma RFS HR 0.44	Moderna / Merck · KEYNOTE-942 Phase 2b
局部 nivolumabnivolumab	Oral pre-cancer clinical response 85%, 1-year clearance 82%	Systemic exposure ~1/10 of IV · floor disclosure