Thread by @Prof_Oak_

Have been eagerly anticipating the $BLUE / $TSVT Abecma launch as the first glimpse into the asset-level profitability of autologous CAR-T, as the CD19s have all been buried in big pharma income statements

In 3Q21, Abecma generated $67M in US revenue. Product-level profit is split 50 / 50 with $BLUE / $TSVT

$BLUE / $TSVT recognized $13M in collaboration profit in 3Q21, implying a product-level operating margin of ~39%

Such an analysis ignores certain variables (e.g. significant overhead expense) and lacks visibility on others (e.g. S&M allocation within $BMY, which commands a huge hematology sales force), but in any case, helps give some visibility for autologous commercial stories like $LEGN

My view: COGS is likely the bulk of the expense - I'd estimate ~25-30% of ~$450K list. While margins will expand as revenue grows, the COGS structure of autologous will put a low ceiling on product-level profitability because variable costs are high

Let's think about that in the context of $LEGN, a $7B company with $2B consensus sales for cilta cel in 2025. If we apply a 50% operating margin on that topline, $LEGN's recognized profit share income will be $500M, before overhead, capex, R&D, etc.

Even if we very conservatively assume $LEGN can squeak out a 70% operating margin (or $150M in expenses vs. ~$350M last year), that's a $350M profit on a consensus expectation of $2B. A ~20% operating margin on the best-selling cell therapy of all time

This is merely illustrative, but it demonstrates the challenges of commercializing autologous cell therapy products. The business model is simply constrained - distribution is complex, unit economics are poor, and economies of scale are absent

The auto v. allo debate is far from over with respect to clinical utility, but in my view, only allogeneic offers the operating leverage that will be required to support standalone, blockbuster cell therapies, should they ever exist