CHMP Rare Disease Pipeline and NICE Threshold Implementation

MarcusWelcome to Access Brief, the daily AI podcast on HEOR, HTA, and market access. I'm Marcus, with Sara. Today: CHMP's rare disease approval momentum, NICE threshold implementation details, and JCA capacity hitting hard limits. Let's get into it.

SaraStarting with EMA — the CHMP recommended eight medicines for approval at its May meeting. But look at the pattern here, Marcus. This isn't random volume.

MarcusAgreed. The May session reinforces what we've been tracking — the CHMP is processing a pipeline skewed heavily toward rare, molecularly targeted, and specialty indications. For sponsors in those spaces, the committee has developed institutional fluency with small patient populations, surrogate endpoints, and conditional authorisation pathways.

SaraThat institutional fluency is exactly the problem. We're seeing approval machinery optimised for rare diseases while common conditions get squeezed out. The evidence standards aren't getting more rigorous — they're getting more niche.

MarcusI don't buy that framing, Sara. The CHMP is responding to where innovation actually lives right now. Rare disease sponsors have learned to build dossiers that work within existing frameworks. That's not lowered standards — that's better evidence strategy.

SaraBetter evidence strategy for whom? We're creating a two-tier system where rare diseases get procedural advantages while high-prevalence conditions face the same old evidence bars. That's a policy choice, not a scientific inevitability.

MarcusMoving to NICE — they've finally released implementation details for the threshold increase. All ongoing and new appraisals continue under the existing £20,000-£30,000 range until April 2026. NICE will categorise ongoing topics into three groups, with cost-effective products under current thresholds continuing as normal.

SaraThe transition mechanics matter less than the projected impact. NICE currently recommends 91% of medicines they evaluate — around 70 per year. Their analysis suggests the £25,000-£35,000 threshold will allow them to recommend an additional 3-5 new medicines or indications annually.

MarcusThat's underwhelming. We're talking about a fundamental threshold shift for a marginal gain of 3-5 additional recommendations. Either NICE's current recommendation rate already captures most viable products, or this increase isn't as meaningful as the fanfare suggested.

SaraOr it signals that NICE knows exactly where the real cost-effectiveness cliff sits. They're not opening floodgates — they're making targeted adjustments for products that were borderline under the old thresholds. It's surgical, not transformational.

MarcusThat surgical approach makes sense given JCA constraints intensifying across Europe. The European Commission announced a maximum of 17 slots for scientific advice — 8-12 for medicines, 2-5 for medical devices. That's a hard capacity ceiling.

SaraAnd medical device JCAs are starting in June with approximately 5 joint clinical assessments planned. We're watching HTA infrastructure strain in real time. The question isn't whether JCA will work — it's whether it can scale.

MarcusHere's where I push back on the scaling pessimism. Constrained capacity forces better prioritisation. If we only get 17 scientific advice slots, sponsors will submit higher-quality requests. Artificial scarcity can improve signal-to-noise ratios.

SaraThat assumes rational prioritisation, Marcus. What we'll actually see is political allocation — member states gaming the system for their national champions while genuinely innovative products get squeezed out by procedural bottlenecks.

MarcusISPOR 2026 highlighted a critical shift in how we think about evidence generation. AI topped the trends report, rising from number 3 to number 1, but the real insight came from the mandate change. We're no longer being evaluated on our ability to produce data — we're being judged on our ability to influence decisions under constraint.

SaraThat's the most important development this week. The HEOR field is maturing past data generation into decision architecture. Real-world evidence ranks second in ISPOR's trends, but it's not about having more data — it's about having the right data at the right decision point.

MarcusThe constraint-driven approach forces discipline we've lacked. When HTA capacity is unlimited, you can throw evidence at every problem. When it's constrained, you have to understand which evidence actually moves decisions.

SaraThe danger is that constraint-driven thinking becomes constraint-accepting thinking. We should be building HTA systems that can handle innovation volume, not innovation rationing systems that make artificial scarcity seem efficient.

MarcusWe're seeing HTA bodies acknowledge their limits while the industry adapts to work within them. That's not rationing — that's market clearing at a different equilibrium point.

SaraNext week will show whether these capacity constraints are temporary growing pains or the new normal. The evidence suggests we're designing permanent limitations into European HTA infrastructure.

MarcusBack tomorrow on Access Brief. Show notes at outcomes-analytica.no.