The $4 Million Problem [Guest Post]

Editor’s Notes: You know us for our Monday to Friday biotech & pharma news compilations, but on Saturdays we post long form content - occasionally written by us, occasionally written by external authors (with permission). The below is in the latter category - your regularly scheduled news compilations are back on Monday!

Chris Abraham works on the Ad Network team at beehiiv and writes Thinking Folds, a newsletter revealing how scientific breakthroughs transform into venture-backed biotech startups. With a background in cell biology, he's drawn to where innovation meets commercial reality—the patterns, people, and decisions that turn discoveries into companies.

Originally published on beehiiv on December 12th, 2025. All the below is Chris Abraham’s writing, and republished here with his permission.

Thirty-four thousand Americans have diseases that can now be cured with a single injection. The problem isn't the science. It's that each cure costs $3-4 million, and nobody knows who should pay for it.

Lenmeldy can prevent a fatal brain disease called Metachromatic Leukodystrophy with one infusion. It costs $4.25 million. Hemgenix can cure Hemophilia B. It costs $3.5 million. These therapies work. Patients walk in sick, walk out cured. But the companies that provide health insurance (often the employers themselves) face a question that breaks the entire system: why would we spend $4 million today to cure someone who might leave the company in two years?

This is the infrastructure crisis at the heart of gene therapy. The science arrived. The payment system didn't. And until someone solves it, most patients with curable diseases won't get cured.

On December 9, 2025, Aradigm Health emerged from stealth with $20 million to build what they call the "infrastructure layer" for cures. Not insurance. Not a pharmacy manager. Something new. And if they're right, they're unlocking a market growing from $6 billion to $51 billion by 2034—a market that literally can't function without infrastructure they're building. The question: can risk pooling and orchestration solve what insurance couldn’t?

The $4 million problem nobody can budget for

Fig. 1 - THE SCIENCE

Gene therapy broke healthcare economics by compressing lifetime costs into a single payment.

A Hemophilia patient on traditional treatment costs roughly $300,000 annually for life. Over 40 years, that's $12 million. Predictable. Budgetable. Spread across decades. Gene therapy collapses that into one $3.5 million event. Same lifetime cost, delivered as a financial shock most employers can't absorb.

Here's the mismatch: the average American changes jobs every 4.1 years. An employer pays $4.25 million in 2026 to cure an employee. That employee leaves for a competitor in 2028. The employer funded a lifetime cure but captured two years of benefit. The competitor gets 40 years of a healthy employee for free.

For a mid-sized company with 500 employees, a single $4 million claim can violate banking agreements or trigger cash flow crises. And the cost varies wildly. The same CAR-T therapy can cost $2 million at one hospital and $4.8 million at another for identical outcomes.

"The arrival of a life-saving therapy is often viewed by payers not as a medical triumph, but as a catastrophic financial 'lightning strike.'"

Traditional insurance has stop-loss coverage for catastrophic claims. But carriers responded to gene therapy by "lasering"—setting deductibles equal to the treatment cost. An employer buys protection for a $3.5 million gene therapy, then discovers the insurance company assigned a "$3.5 million laser" to that exact treatment. It's like buying fire insurance that excludes kitchen fires. They're unprotected anyway.

The numbers:

• 60% of patients abandon prescriptions when costs exceed $500

• 12.5% of Medicaid requests for high-cost therapies get denied

• The US gene therapy market was $6.35 billion in 2024, projected to hit $51 billion by 2034

• By 2030, 340,000 patients annually will need these therapies (a 28-fold increase from 2020)

What started with ultra-rare diseases is expanding to Sickle Cell Disease (100,000 Americans), Duchenne Muscular Dystrophy, and beyond. The pipeline is full. The payment infrastructure doesn't exist.

Even when insurance approves treatment, delivery is a nightmare. Gene therapy requires a multi-month process: cell collection, shipping to manufacturing facilities, weeks of processing, travel to one of 10-20 specialized centers nationwide (often 1,000+ miles away), chemotherapy preparation, infusion, and weeks of monitoring. Miss one step (manufacturing delays, authorization expires, patient can't travel) and the entire process restarts.

This is why Hemgenix, despite FDA approval in 2022, had treated fewer than 50 patients worldwide by mid-2024. Thousands were eligible. The barrier wasn't efficacy. It was infrastructure. But what if you could build that infrastructure from scratch?

Pooling risk and managing the patient journey

Fig. 2 - THE STARTUP

Aradigm attacks both problems: financial risk and delivery logistics.

Instead of each employer bearing a potential $4 million hit, Aradigm pools 15+ large employers into a single fund. Each pays a fixed monthly fee per employee. When a $4.25 million claim hits, it's paid from the pooled fund, not one company's budget. Volatility gets distributed across millions of lives. No single employer gets bankrupted.

The key difference: Aradigm is a Public Benefit Corporation running a "cost-plus" model. Monthly fees cover claims plus a transparent management fee. Any surplus gets returned to employers. Traditional insurance keeps the profit when claims are low. Aradigm gives it back. Zero incentive to deny care.

But financial coverage doesn't solve logistics. This is where Aradigm's model gets defensible.

They build concierge support for the entire patient journey. They coordinate manufacturing timing with biotech companies, arrange flights and housing for families, handle insurance paperwork so nothing expires mid-treatment, and synchronize all parties so cells arrive exactly when the patient is ready.

The patient gets one point of contact instead of navigating 15 phone numbers across 4 organizations. Aradigm handles the bureaucracy. The patient focuses on treatment.

If executed well, this becomes the moat. Providers prefer Aradigm patients because payment is faster and coordination is smoother. Manufacturers prefer Aradigm because patients actually complete treatment instead of abandoning mid-process. Employers prefer Aradigm because the alternative is chaos.

Aradigm's three-part infrastructure: Access, Pay, and Deliver. Source: a16z

Current status: The platform launches April 2026 with 15 of the nation's largest employers and 6 major insurance partners already signed. Zero patients treated yet, but the pre-launch validation matters. They secured Fortune 500 companies before treating anyone. Demand is real.

Healthcare insiders building the missing infrastructure

Fig. 3THE VENTURE

Aradigm's team spent decades inside the healthcare system and concluded it needed rebuilding:

• Dr. Will Shrank, the CEO, designed payment reform models at the Center for Medicare and Medicaid Innovation, led population health at Humana, and ran physician partnerships at CVS Health.

• Spencer Carrucciu, Chief Product Officer, built data systems at Cityblock Health (which manages complex patients) and led product at Remedy Partners (the bundled payment leader). Gene therapy delivery is essentially a bundled episode spanning months and multiple organizations. He's built this before.

• Annie Collins joined from Andreessen Horowitz, where she was an investing partner. Her move from investor to operator signals conviction. a16z sent one of their own to help build it.

• Justus Ruff manages network operations. He built value-based care networks at Pearl Health. His job: construct the national network of treatment centers and negotiate pricing with providers and manufacturers.

"Cures are already here. They just aren't evenly distributed."

— Julie Yoo & Jay Rughani, Andreessen Horowitz

The funding reflects insider credibility. Aradigm raised $5 million from a16z in 2024, then $20 million in December 2025 led by Frist Cressey Ventures (founded by former Senate Majority Leader Bill Frist). Morgan Health (JPMorgan Chase's healthcare arm) participated. JPMorgan is one of America's largest employers. Their investment validates demand.

The a16z thesis: innovation in biology (the science of cures) has outpaced innovation in healthcare payment systems. The US insurance system was built for managing chronic conditions with ongoing treatment. Gene therapies require paying upfront for lifetime cures. That's not incremental. It's structural. Aradigm is infrastructure for an era the current system can't support. The problem: the biggest pharmacy players built their own versions years ago.

Why independence beats the big pharmacy players

Fig. 4WHY THIS MATTERS

The major pharmacy benefit managers launched competing products years ago. Cigna's Evernorth offers employers a monthly fee to absorb gene therapy costs. CVS built a network of 300+ treatment centers bundled into Aetna plans. Both work within their ecosystems.

But they're vertically integrated. Evernorth only works with Cigna's infrastructure. CVS only works if you use Aetna. And neither is transparent about surplus. When claims are lower than premiums collected, they keep the difference as profit.

Aradigm's differentiation is structural. They're carrier-agnostic. An employer using Blue Cross for medical coverage can carve out gene therapy to Aradigm without changing their entire setup. They return surplus to employers instead of keeping it. And they manage the patient journey, not just pay claims.

The question is whether the orchestration layer proves sticky enough to defend against commoditization. If Aradigm's value is just pooling risk, large reinsurers could copy the model. But if providers and manufacturers genuinely prefer working with Aradigm patients because the process is smoother, they've built something defensible.

The data advantage could be significant. As Aradigm manages more cases, they build a registry of real-world outcomes for rare therapies. This matters to manufacturers (who need data for FDA requirements), insurers (who need it for coverage decisions), and researchers (who need long-term efficacy data). More data makes risk models more accurate, which attracts more employers, which generates more data. But network effects only matter if you can scale fast enough to capture them.

Fig. 5WHAT'S NEXT

If gene therapy prices drop from $4 million to $500,000, does the model break? Not necessarily. Lower prices mean more patients become eligible, more therapies get approved, and utilization increases. Aradigm's revenue would shift from risk premiums to orchestration fees. The logistics problem remains even if financial risk shrinks.

Manufacturers likely view Aradigm as a partner. Despite FDA approval, drugs like Hemgenix have sluggish sales because employers won't cover them. Aradigm removes that barrier. They create a fast lane for reimbursement by eliminating the laser risk that makes employers refuse coverage.

For Aradigm to prove the thesis over five years, they need to scale the risk pool to 10-15 million covered lives (roughly 5-7% of the commercial insurance market) for actuarial stability, become the default network for the top 20 emerging gene therapies, and actually return surplus to employers in low-claim years to validate their alignment model.

The cure-based era of medicine is here. The infrastructure to deliver it isn't. Aradigm is building the rails.

If this was useful, forward it to someone building breakthrough therapeutics in biotech. More stories like this every week at Thinking Folds. 

Big thanks again to Chris Abraham for letting us republish this article!