The Integration Edge

Why Manufacturing Integration Determines Cell Therapy Winners

Article 2 of 10 in a series on cell therapy partnership strategy.

In 1982, Hewlett-Packard conducted a study that sent shockwaves through Silicon Valley. The company tested 300,000 memory chips from three American and three Japanese suppliers. The results were unambiguous: Japanese chips had failure rates approaching zero. American chips failed at rates up to 6 times higher.

The quality gap wasn't a design problem. American engineers were producing comparable—sometimes superior—chip architectures. The gap was manufacturing. Japanese yields reached 70-80%, while American competitors topped out at 50-60%. The products coming off Japanese lines were both better and cheaper.

By 1985, American semiconductor companies faced an existential crisis. Intel had one of its largest layoffs in history. Texas Instruments saw profits turn negative. National Semiconductor's operating profit of $59 million became a loss of $117 million. By 1989, six of the top ten global chip makers were Japanese.

The lesson took a decade to absorb, but it transformed American manufacturing philosophy. Intel's "copy exact" discipline—replicating every detail of a successful manufacturing process across facilities—emerged directly from this crisis. The companies that survived did so by re-integrating design and production.

Cell therapy is now at a similar inflection point. The partnerships that will define the next decade of the sector are being signed today. The question is whether deal-makers are learning from the semiconductor industry's hard-won lessons about manufacturing integration.

The Separation Problem

The semiconductor crisis of the 1980s stemmed from a structural choice. American companies had separated chip design from manufacturing. Design teams optimized for performance on paper. Manufacturing teams dealt with the consequences—and often couldn't achieve the yields that designs theoretically permitted.

Japanese companies operated differently. Vertically integrated organizations kept design and manufacturing teams in constant communication. Design decisions incorporated manufacturing constraints from the start. The result: designs that were manufacturable at high yields, producing chips that were both higher quality and lower cost.

Cell therapy partnerships often repeat the American semiconductor model. A biotech licenses a CAR construct and a data package to a pharma partner. The pharma partner then attempts to manufacture at scale, often discovering that the process doesn't transfer cleanly. Manufacturing yields disappoint. Vein-to-vein times extend. Quality metrics suffer.

The partnerships that succeed look more like the Japanese model. Legend didn't license a construct to Janssen—they brought an integrated manufacturing operation. AstraZeneca didn't acquire Gracell for a pipeline—they acquired FasT CAR-T, a manufacturing platform that reduces production time from weeks to 22-36 hours.

Why Integration Matters More in Cell Therapy

The manufacturing integration imperative is even stronger in cell therapy than it was in semiconductors. Semiconductors are standardized products manufactured in controlled environments. Cell therapy starts with variable patient material and must produce a living product within tight timelines.

Consider the variables a cell therapy manufacturer must control: starting T-cell quality (which varies dramatically by patient, disease state, and prior treatment), activation kinetics, transduction efficiency, expansion dynamics, and final product phenotype. Each of these variables interacts with design decisions about the CAR construct, the vector, and the manufacturing protocol.

Teams that design constructs without manufacturing expertise optimize for the wrong parameters. They may create CARs with high theoretical potency that require extended expansion times—degrading T-cell quality and extending vein-to-vein timelines. They may select vectors that transduce efficiently in laboratory conditions but perform poorly with patient-derived cells.

Integrated teams make different choices. They design for manufacturability: shorter expansion protocols, simplified activation, reduced reagent dependency. The result is processes that work reliably across the full range of patient starting material.

"The design decisions that optimize manufacturing—shorter expansion, simplified activation, reduced reagent dependency—emerge from teams that see both sides of the process."

The Vein-to-Vein Metric

Manufacturing integration shows up most clearly in vein-to-vein time—the duration from patient apheresis to infused product. This metric captures the entire manufacturing operation's effectiveness.

Standard autologous CAR-T requires 4-6 weeks vein-to-vein. During this period, patients often need bridging therapy to control disease progression. Some patients—estimates range from 10% to 30% depending on indication—never receive their manufactured product due to disease progression, death, or manufacturing failure.

Gracell's FasT CAR-T platform achieves manufacturing in 22-36 hours, with vein-to-vein times of 1-2 weeks. PRIMCAR-type platforms achieve manufacturing in under 48 hours. These aren't incremental improvements—they represent fundamentally different manufacturing approaches that only integrated design-and-production teams can achieve.

The clinical implications are substantial. Faster manufacturing preserves T-cell phenotype. Data from multiple programs demonstrates that shorter production times correlate with higher percentages of naive and stem cell memory T-cells (Tscm)—the populations most strongly associated with durable responses and long-term persistence.

In one analysis of 50 patients, CAR-T persistence was the only factor significantly correlated with long-term event-free survival. Not dose, not tumor burden, not prior treatment history—persistence. And persistence correlates with T-cell phenotype at infusion, which correlates with manufacturing time.

Evaluating Integration in Due Diligence

For acquirers and partners evaluating cell therapy opportunities, manufacturing integration should be a primary diligence criterion. The questions to ask go beyond "does manufacturing work" to "how are design and manufacturing connected."

Organizational structure: Do design and manufacturing teams report through the same leadership? Do they share facilities? How frequently do they collaborate on protocol optimization?

Design-for-manufacturability evidence: Can the company demonstrate design decisions made specifically to improve manufacturing outcomes? Have they iterated on the construct or process based on manufacturing data?

Yield and timeline metrics: What are manufacturing success rates across patient populations? What's the distribution of vein-to-vein times? How do these compare to industry benchmarks?

Technology transfer history: Has the company successfully transferred manufacturing to new facilities or partners? What challenges emerged? How were they resolved?

In my experience advising on cross-border partnerships, the technology transfer question is often the most revealing. Companies with genuine manufacturing integration can articulate exactly why their process works and what parameters must be controlled. Companies that have separated design from manufacturing often discover gaps only when transfer fails.

The Partnership Structure Implication

Manufacturing integration has direct implications for partnership structure. Traditional licensing models—upfront payment, milestones, royalties—assume that manufacturing capability transfers cleanly. When it doesn't, both parties suffer.

The Legend-Janssen structure reflects this reality. Rather than a simple license, the partnership created joint governance with profit-sharing. Legend maintains manufacturing involvement; Janssen provides commercial infrastructure. Neither party can succeed without the other's continued engagement.

For early-stage partnerships, structures that preserve the licensor's manufacturing role often outperform clean transfers. This might mean manufacturing agreements alongside development licenses, joint quality oversight, or ongoing technology collaboration requirements.

The Kite-Daiichi Sankyo partnership for Yescarta in Japan illustrates the challenges of the alternative approach. Despite a $50 million upfront payment and clear commercial intent, manufacturing transfer proved difficult enough that the partnership was ultimately restructured, with Gilead's US facility eventually supplying the Japanese market.

The Integration Advantage Compounds

The semiconductor industry's integration lesson applies beyond current products. Companies that master integrated design-and-manufacturing develop capabilities that extend to next-generation technologies.

The lentiviral vector expertise required for 48-hour autologous CAR-T is directly applicable to in vivo CAR-T delivery. The quality systems that enable bedside production are the foundation for hospital-based manufacturing of future therapies. The process engineering that reduces vein-to-vein time today will drive cost reduction across modalities tomorrow.

Partners who bring manufacturing mastery don't just deliver better current products—they position both parties for the manufacturing challenges ahead. In an industry where cost reduction will eventually determine market access, that positioning may prove more valuable than any single clinical asset.

The semiconductor industry learned this lesson through crisis. Cell therapy deal-makers have the opportunity to learn it through strategy.

Next in series: Three Manufacturing Strategies That Work

Previous: What the Best Cell Therapy Partnerships Have in Common

For advisory on cell therapy partnership strategy, contact Kerlann Advisory.