BB, you are right. DCs are adherent. Ori cannot handle them. EDEN was purpose-built for the barnacle cells. Linda Powers and Annalisa Jenkins together are exactly the team this platform needs. No argument.
But go one step further. The DC is not the final product. The DC is the instructor that manufactures the final product.
Kalinski is a paid consultant for Northwest Bio (confirmed in a conflict of interest disclosure filed May 2, 2026). His lab page lists adoptive T cell therapy with DC-instructed polyclonal CTLs as a current project. The patent families Linda Powers licensed from Roswell Park in June 2024 include claims on CXCR4+ CD8+ T cells and type-1 CD4+ T cells as drug products. These are not DC vaccine claims. They are T cell product claims. On April 29, Kalinski himself said publicly: this may be better than TIL and may be better than CAR-T. No gene transfer. No genetic engineering. Young CTLs, not pre-exhausted. Two-hundred-fold enhanced recognition through dual TCR/NKR engagement.
A single-cell paper posted May 2 on bioRxiv, with Kalinski as co-author, just showed why this matters for manufacturing. The alphaDC1 preparation everyone has treated as homogeneous is actually three subpopulations. Only 0.5-2% produce IL-12p70 (Cluster E, IRF8-high). One subpopulation (Cluster D) actively recruits Tregs against the therapeutic intent. They identified surface markers to sort these populations at baseline using existing GMP protocols.
This is where the Ori question dies. With an unsorted preparation, 98% of DCs are non-producing or counter-instructing, so you need massive T cell expansion to compensate for instructional inefficiency. That is the scenario where a second suspension bioreactor makes sense. Sort for Cluster E first, and every T cell in the co-culture receives correct instruction. The yield of properly educated cells goes from a diluted fraction to the entire output. When every cell carries Kalinski’s two-hundred-fold enhanced dual-key recognition, you need fewer cells because each one is more potent. Published data show 1,000-fold expansion is achievable within EDEN. Billions of T cells from a single blood draw. No transfer step. No second bioreactor. Kozbial’s EDEN handles the full transition from adherent DC culture to suspension T cell harvest in one closed system.
Now add the macrophage arm. Sharadhi Siri spent 3.5 years at BobcatBio manufacturing SIRPalpha-low M1 macrophages, the product that demonstrated complete responses in CAR-T-refractory DLBCL. She joined NWBO as Senior Cell Specialist. Macrophages are also adherent, also barnacle cells, also manufactured from the same monocyte starting material through a different cytokine program. In desmoplastic solid tumors, the fibrotic stroma physically excludes T cells, NK cells, and endogenous DCs. Only the macrophage can breach that barrier, liberate antigen, broadcast CXCL9, and die. No CAR-T platform has a barrier clearance arm. No TIL platform has one. I should note that no formal NWBO-BobcatBio collaboration has been publicly announced. Siri’s move is an operational fact. The CKM-mediated macrophage reprogramming is Kalinski’s published work under NWBO’s license. But a dedicated macrophage cell product integrated with the DC and T cell arms remains a thesis supported by the evidence, not a disclosed corporate plan.
CKM conditioning protects all three arms simultaneously. Kalinski explained on April 29 why it can be given IV without the toxicity that killed systemic IL-12 in the 1990s: rintatolimod engages only endosomal TLR3 through TRIF/IRF3, avoiding the cytosolic NF-kappaB pathway that drives suppressor recruitment. The signal amplifies selectively in tumors. Healthy tissues do not respond. Clinical data: 78% favorable pathologic response in neoadjuvant TNBC associated with lack of recurrence. Timing data show same-day delivery is required for full efficacy.
Against CAR-T: polyclonal multi-antigen targeting instead of single-antigen escape. Natural dual-key recognition instead of synthetic CAR. No genetic modification. No viral vectors. No lymphodepletion. No leukapheresis. No CRS. Biological off-switch through DC apoptosis. A macrophage barrier arm no other platform possesses. Cluster E sorting for manufacturing consistency. The entire lifecycle from one blood draw.
BB, you called it a perfect biotech diamond. I agree. Linda Powers assembled the IP estate and the team: Kalinski for the science, Annalisa Jenkins for regulatory and commercial execution, Siri carrying macrophage manufacturing knowledge into the DC facility, Kozbial’s EDEN handling all three product pathways in one closed system. But the diamond has more facets than the barnacle cells alone. The macrophage opens the door. The DC writes the instruction. The T cell carries it everywhere and remembers.
Cheers.
Market Data 
Markets 
AI
