“It’s Too Early to Think Commercial”

I hear this phrase from founders constantly. It sounds reasonable every time.

The company is pre-clinical, or in early Phase 1, or working through a platform development phase. The science is not derisked. The indication is not locked. The competitive landscape will look different by the time the product reaches the clinic. Hiring a commercial leader now would be premature. Building access infrastructure now would be wasted spend. The team is small. Burn matters. The board agrees. The investors agree. Everyone agrees.

The phrase is doing something interesting anyway.

The access decisions are being made. Just not by anyone who is thinking about access while they decide. Modality. Route of administration. Dosing schedule. Manufacturing platform. Eligibility design. Trial structure. Each one of those is a commercial decision dressed up as a scientific one. By the time the founder agrees it is time to think commercial, the access architecture has already been built. The commercial team that arrives is operating inside a system that has already determined who the therapy can realistically reach.

The companies that get this right are the ones I keep coming back to.

Vertex received FDA approval for Trikafta in October 2019 for cystic fibrosis patients ages 12 and older with at least one F508del mutation. Across nearly seven years, Vertex pursued sequential label expansions. Ninety-four additional mutations in 2020. Age range extended to 6 and older in 2021. Age range extended to 2 and older in 2024. And a final expansion in April 2026 to include additional variants supported by clinical and in vitro data, including those that result in production of CFTR protein, bringing the evidence base to 521 responsive variants.

The April 2026 expansion brought eligibility to approximately 95 percent of all people with cystic fibrosis in the United States. Vertex is now treating more than 75,000 people across more than 60 countries.

That trajectory was not a launch achievement. It was a development strategy executed over seven years through clinical trials, regulatory submissions, and in vitro variant studies that turned an initial F508del-only label into a near-universal therapy for the disease. Access was designed across the entire development arc.

Krystal Biotech offers a different version of the same discipline. The FDA approved Vyjuvek in May 2023, the first redosable topical gene therapy, for dystrophic epidermolysis bullosa. Vyjuvek is applied topically by a caregiver or clinician. No specialized treatment center. No conditioning. No infusion. The therapy is shipped to where the patient lives. Krystal designed access in at the molecule level. By choosing a topical, redosable format for a chronic genetic skin disease, the company eliminated nearly every infrastructure constraint that limits one-time durable gene therapies. Access for Vyjuvek was a design specification.

The same biology can carry profoundly different access designs.

PCSK9 inhibition for LDL cholesterol reduction has three commercialization systems already on the market. Repatha and Praluent are monoclonal antibodies given every two or four weeks through patient-administered injection. Leqvio is an RNAi therapy given twice yearly in a clinic. Each has a different patient who can stay on it. The biweekly injection asks the patient to manage refills, self-administer, and remain adherent across a chronic regimen. The twice-yearly clinic visit shifts the administration burden to the healthcare system and reduces patient effort to two appointments a year. Same biology. Same payer category. Three different commercialization systems, each built years before launch.

The radiopharmaceutical space shows what happens when design and physics meet.

Novartis received FDA approval for Pluvicto in March 2022 for metastatic castration-resistant prostate cancer. The clinical data was strong. Payer coverage came together quickly. The product hit 179 million dollars in sales in Q4 2022 alone. By every conventional launch measure, the program was working.

The constraint sat upstream of payer coverage. Lutetium-177, the isotope at the heart of the therapy, has a half-life of 6.6 days, and at launch Pluvicto was being manufactured at a single facility in Ivrea, Italy. Each dose had a five day window from production to patient infusion. The combination of isotope physics, single-source supply, and dose-specific scheduling determined how many patients could realistically receive therapy in any given week. Novartis has since brought additional manufacturing capacity online, and supply has stabilized considerably.

The decision that determined who could access Pluvicto at launch was made years before approval, by scientists and supply chain executives who did not think they were deciding access policy. They were.

Casgevy and Lyfgenia carry the same lesson at a different scale. Both received FDA approval in December 2023 for sickle cell disease. Both are genuinely transformative one-time cell-based gene therapies. The launches have been a more complicated story than the science deserved, and the conversation in the field is increasingly about how the next generation of products in this disease will reach more patients than the first generation has. Editas pivoted the entire company toward in vivo gene editing in late 2024. Beam is publishing strong ex vivo base editing data while developing in vivo delivery for the next generation. The first generation answered the question of whether the science could be made to work. The companies coming next are answering the question of what the commercialization system needs to look like.

The pattern across all five examples is the same. Whether access works is decided in development. Not at launch.

Which brings me back to the founders.

The version of this conversation that comes up most is the one where a Series A or Series B biotech founder is building toward a strategic acquisition. The thesis is that big pharma will partner or buy the asset, and the access architecture is something the acquirer is better resourced to build. There is logic to that. The acquirer does have the resources, the experience, and the playbook.

What that thesis misses is that the design choices come with the asset. The modality is fixed. The manufacturing platform is fixed. The trial design is fixed. The eligibility criteria are fixed. The acquirer is inheriting a system that already determines what is reachable. The companies that command the strongest acquisition economics are the ones where access was designed in early enough that the system the acquirer inherits already has reach built into it. The deal closes against the population the science can actually serve, not the population the science could theoretically have served.

“It is too early to think commercial” is the phrase that lets the design choices get made unconsciously. It is also the phrase that determines what the company is worth when the deal eventually comes.

It is not too early to think commercial. It is too late by the time everyone agrees it is the right moment.

The companies that escape Commercial Gravity are not smarter than the ones that do not. They are earlier. And they build the systems before they need them.