A $2 Million Prior Auth Just Hit Your Desk. Here's How to Handle It.

When a multi-million dollar gene therapy prior auth hits your desk, you have to decide three things at once: your documented medical necessity criteria, your exact stop-loss position, and your payment structure. Build all three before the request arrives. The new rules give you as little as seven days to respond.

Picture the call. Your TPA phones on a Tuesday. An employee's child qualifies for a gene therapy, and the price tag is $2.1 million.

Most employers freeze. They don't have a policy. They don't know their stop-loss position, and the clock is already running.

This is happening more often. According to an NBER working paper, annual US gene therapy spending is projected to peak around $25.3 billion in 2026. The confidence intervals are wide, but the direction isn't.

What do the new prior auth rules require?

Faster decisions with written reasons. The CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F) sets new timelines effective January 1, 2026. Urgent requests require a decision within 72 hours. Standard requests get seven calendar days.

Clear denial reasons are required, not optional. That same rule makes impacted payors publish approval and denial rates, appeal outcomes, and turnaround times. The full list of services requiring prior auth must be published too.

Self-funded employers aren't directly regulated by CMS-0057-F. But your TPA likely is, and their timelines become your timelines. If you don't have documented medical necessity criteria for gene therapies today, you're already behind.

What belongs in your gene therapy medical necessity criteria?

Specifics that hold up under review. This isn't like approving an MRI. You're evaluating a one-time, potentially curative treatment with a multi-million dollar price and limited long-term data.

A defensible framework includes at least these elements:

• FDA approval status for the specific indication
• Confirmed diagnosis with documented treatment history
• Evidence of inadequate response to standard-of-care alternatives
• Independent clinical review, not just a TPA recommendation
• Outcomes data from peer-reviewed sources or manufacturer registries

The outcomes data is improving. Vertex presented longer-term results on CASGEVY (exagamglogene autotemcel) at the 2025 European Hematology Association Congress. Sickle cell follow-up now exceeds 5.5 years, with 43 of 45 evaluable patients free of vaso-occlusive crises for at least 12 straight months.

The USC Schaeffer Center's cell and gene therapy policy research tracks this evidence as it develops. Build your criteria around current data. Plan to update them every year.

What should you ask your stop-loss carrier first?

Whether the claim is covered, and how. This isn't a coverage question for your plan. It's a contract question for your carrier.

The ICER and NEWDIGS white paper on gene therapy financing examines how stop-loss and reinsurance can absorb portions of these claims, and where they fall short. Reinsurers can protect against aggregate losses exceeding projections. But only if the claim is structured and communicated correctly from the start.

Ask your stop-loss carrier these questions right away:

• Is this therapy explicitly listed as a covered or excluded claim?
• Does your specific attachment apply to the full claim amount?
• Are installment payment structures recognized under your policy?
• What documentation is required for reimbursement on gene therapy claims?

Some carriers now write gene therapy carve-outs. Others exclude them entirely. You need to know which situation you're in before a claim arrives.

How do outcomes-based contracts and installment payments work?

They tie money to results and time. Paying $2.1 million upfront for a therapy that might not deliver durable results is a hard sell. Outcomes-based contracting changes that math.

Under these models, the manufacturer agrees to rebates or refunds if the therapy underperforms over a defined period. Some contracts tie payment milestones to clinical benchmarks, like transfusion independence for sickle cell patients. Installment models spread the cost over multiple years.

The ICER and NEWDIGS paper also discusses a potential federal carve-out benefit program. That infrastructure isn't finalized, but it signals where this is heading.

CMS is already moving in this direction. Its Cell and Gene Therapy Access Model began in January 2025, with the first state going live on March 1, 2025.

Thirty-three states plus DC and Puerto Rico have signed agreements. Those states represent roughly 84% of Medicaid beneficiaries with sickle cell disease. The model tests outcomes-based payment at scale, and private employers should watch it closely.

What should you do before the phone rings?

Waiting for a $2 million request to build your framework is the most expensive mistake you can make. You'll decide under pressure, with no documented criteria and no payment plan in place.

Talk to your stop-loss carrier now. Ask your TPA if they have a gene therapy management protocol. Find out whether your plan document includes or excludes these therapies by name.

The call is coming. The only question is whether you're ready when it does.

Frequently asked questions

How much can a single gene therapy claim cost?

List prices run into the millions. CASGEVY, for example, carries a list price around $2.2 million per patient. A single claim can exceed an entire mid-market plan's annual stop-loss attachment, which is why choosing the right attachment point matters before one hits.

Does CMS-0057-F apply to self-funded ERISA plans?

Not directly. The rule covers Medicare Advantage, Medicaid and CHIP, and federal Marketplace issuers. But most TPAs administer regulated business too, so their 72-hour and seven-day timelines tend to become your operating standard.

How can stop-loss actually protect against a gene therapy claim?

Stop-loss reimburses claims above your specific or aggregate attachment, but only if the therapy isn't excluded and the claim is documented correctly. Gene therapy carve-outs and lasering are now common at renewal. Our funding models guide explains how stop-loss fits into a self-funded structure.

What durability data supports these treatments?

It's growing. Vertex reported sickle cell follow-up beyond 5.5 years for CASGEVY at EHA 2025, with 43 of 45 evaluable patients free of vaso-occlusive crises for at least a year. Strong durability data is exactly what makes outcomes-based contracts workable.

Where do gene therapy claims fit in my 2026 cost planning?

They belong in your stop-loss and specialty pipeline forecasts, not as a surprise. Rising reinsurance costs already reflect this risk, as covered in our breakdown of 2026 stop-loss rate increases and the 2027 specialty drug pipeline.