Why Rare Disease Launches Lose Patients After Approval — and It Isn't the Market

Roughly 40% of new launches miss their two-year forecasts. In rare disease, the reason is rarely efficacy, awareness, or access. It's the handoffs between teams that no one owns.

The short version: An analysis of medications launched between 2009 and 2017 found that about 40% failed to meet their two-year sales forecasts. In rare disease, the instinct is to blame the usual suspects — small populations, payer hurdles, awareness. But rare disease launches actually tend to outperform on analyst expectations. Which makes the real failure mode clearer, not murkier: patients are lost not in the market, but in the cross-functional handoffs between the teams trying to help them — the seams in the 90% of the journey that no single function owns.

The launch that looked perfect

Picture a biotech celebrating FDA approval for a rare disease therapy. Years of development, hundreds of millions invested, and finally the green light. The commercial team is ready. The hub is staffed. Awareness campaigns are live. Everything looks perfect.

Eighteen months later, only 23% of eligible patients have initiated therapy.

The company scrambles to understand what went wrong. Marketing reach was strong. Physician awareness was high. Payer coverage was secured. Yet hundreds of patients who need this therapy never actually receive it — and the cost runs into hundreds of millions in delayed revenue, shaken investor confidence, and patients left waiting.

This isn't an isolated story. It's a pattern, and it's worth being precise about what kind of pattern, because the data cuts against the easy explanation.

The counterintuitive part: rare disease launches don't usually fail on the market

Here's what makes this failure mode specific rather than generic. When ~40% of launches miss their two-year forecasts, the assumption is that rare disease is especially exposed — small populations, reimbursement friction, diagnostic odyssey. But the evidence points the other way: rare disease launches frequently meet or exceed analyst launch-year expectations, outperforming oncology and chronic-disease areas.

So if the market isn't the problem, and awareness isn't the problem, and coverage isn't the problem — where are the patients going?

They're falling through the space between the teams.

Not the barriers you'd expect. Not affordability, not awareness. The invisible gaps in the handoffs between the functions that each own a piece of the launch:

• A specialty pharmacy that can't coordinate the genetic testing the therapy requires — because that coordination belonged to no one's job.
• Patient navigators overwhelmed in two weeks because case complexity was underestimated by the team that scoped the role.
• Caregivers without the infrastructure for home administration, because the intake form — owned by one function — never asked about the home environment.
• Financial assistance applications taking six-plus weeks, so patients drop out before therapy begins, while every team's own metrics look fine.

Each of those is a handoff failure. And handoff failures are invisible precisely because no single team owns the handoff.

Why patient support can't catch this: it's a collection of teams, not a system

Most companies run patient support as a collection of functions and vendors, each selected to solve a specific problem, each doing its job reasonably well. A hub manages enrollment. A separate team handles copay and financial assistance. Another owns disease education. Advocacy lives in one department, HCP engagement in the commercial team, specialty pharmacy as a hands-off relationship, the digital experience somewhere in IT.

Each function optimizes its own piece. Each produces its own reports. Each has its own portal, its own patient ID, its own timeline. The sum of these competent parts does not equal a functioning whole — and it fails at launch for reasons that are all variations of the same thing: the patient is lost in the seams between functions.

There's no single source of truth. When a patient drops out, no one can definitively say where or why. The hub shows enrollment complete. Financial assistance shows an approved application. The patient never initiated therapy. By the time you reconstruct the story across functions, the patient is gone.

Handoff failures are invisible until too late. A patient gets stuck in the space between two functions, and no one notices because no one owns that space. The hub thinks it handed off to specialty pharmacy. Specialty pharmacy is waiting on information the hub didn't know it needed. The patient hears nothing and assumes they've been forgotten.

The model is reactive by construction. You can only respond to barriers after patients hit them. There's no mechanism to see them coming — so you're forever fixing the problem for the next patient while the current one is already lost.

Each function is context-blind. Everyone can show that their own metrics look good — enrollments up, handle times down, approval rates strong — while patients aren't getting treated and no dashboard reveals why, because the "why" lives between the dashboards.

And the speed problem is brutal at launch. It takes weeks to spot a systemic issue when information flows through disconnected reports. By the time you see the pattern, dozens of patients have hit the same wall; by the time you coordinate a fix across functions, you've lost a quarter of your launch curve.

What this looks like in practice

Three patients, three perfect-on-paper journeys, three failures in the seam:

A patient gets payer approval for a gene therapy. The hub celebrates the enrollment. But the specialty pharmacy needs a specific genetic test result before dispensing — and the hub can't access lab data, because that's not in its scope. The patient waits six weeks for someone to figure out who owns the test coordination. By week eight, the disease has progressed and the physician is reconsidering.

A patient qualifies for a complex infusion therapy. Financial assistance is approved. Home infusion requires caregiver training — but no one identified that need at enrollment, because the intake form (owned by one function) focused on insurance and diagnosis, not the home environment. The therapy sits in the refrigerator for three weeks. The specialty pharmacy charges a restocking fee when it expires.

A patient has prior auth approved, copay covered, specialty pharmacy assigned. But they work two jobs, and the only infusion center within 75 miles runs weekday business hours. From each function's perspective, their part worked perfectly. The barrier never appears in any report — because it lives in the space between all of them.

In each case, every team did its job. The patient was still lost. That's the signature of a cross-functional handoff failure: no individual failure, and a failed outcome.

The fix is upstream, and it's about the seams — not more vendors

The instinct after a launch stumbles is to add more support — another program, another vendor, another touchpoint. But adding capacity to functions that already work doesn't close the gaps between them. You can staff every function excellently and still lose patients in the handoffs, because excellence within a function does nothing for the space between functions.

What actually prevents these failures is doing one thing before launch that the normal process skips: mapping, specifically and across every function, where persistence is likely to break in the handoffs — and fixing those seams proactively, before the first patient arrives, rather than discovering them patient by patient after.

That's a cross-functional question, and it can't be answered from inside any single function, because no function can see the whole journey. It requires a view across patient services, market access, field, hub, and specialty pharmacy at once — reconciled into one picture of where the handoffs will fail.

Producing that view is what our patent-pending diagnostic, the Gap Finder, was built to do — a reconciled, prioritized map of where persistence leaks across the functions, so a launch team can close the highest-impact seams before launch instead of absorbing the cost of them after. Not another vendor in the collection. The view across the collection that no single function has.

The question worth asking before approval

For any team planning a complex or rare disease launch, one question is worth answering before FDA approval, not after:

Do we know, specifically, where patients will fall through the handoffs between our functions — or are we assuming each team doing its own job well will add up to a patient who makes it onto therapy?

If it's the latter, the launch is optimized for a market that probably isn't the problem — and exposed in exactly the place the data says patients are actually lost: the seams no one owns.

Rare disease launches don't usually fail because the market wasn't ready. They fail because the ecosystem of teams wasn't connected. And that's the one failure you can map, and close, before the first patient ever arrives.

Linked Patient Learning helps life science teams see where patient persistence will break across functions before a launch — so the highest-impact handoff gaps get closed before the first patient arrives, not discovered after. If you're planning a 2027-2028 launch for a complex or rare disease therapy, let's talk.