The US FDA approves an in vivo CRISPR-based gene editing therapy for common cardiovascular disease before 2030.
In vivo gene editing is advancing in trials for widespread ailments beyond rare genetic conditions. Approval requires standard FDA marketing clearance.
Across 6 frontier models answering independently, the panel thinks this is unlikely: the median is 16%.
They broadly agree, 16 points between Gemini at 28% and Claude at 12%.
Nothing is settled until 5 Oct 2031, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GeminiGoogle
Clinical trial timelines and FDA risk tolerance for permanent edits in non-fatal broad indications make a pre-2030 clearance tight.
- Verve's lead PCSK9 candidate VERVE-102 is only in early Phase 1b/2; Phase 3 outcome trials require thousands of patients and 3+ years.
- The FDA demands multi-year safety registries for permanent off-target liver edits before clearing healthy, broad populations.
- Accelerated approval via surrogate LDL-C reduction is the only plausible path, but safety holds could easily push resolution past 2029.
28% -
GPTOpenAI
The key hurdle is proving durable benefit and acceptable off-target safety in large, long cardiovascular trials.
- FDA has cleared in vivo CRISPR studies, but no such product is yet approved for a prevalent cardiovascular indication.
- Sponsors must complete pivotal outcomes or validated-surrogate trials, then submit and clear a BLA before 2030.
- The odds rise sharply if an LDL-targeting editor shows Phase 3-grade safety and durable lowering by 2028.
18% -
GrokxAI
Verve-class PCSK9 editors must finish pivotal trials and FDA review by end-2029.
- Lilly/Verve VERVE-102 is still early-stage in 2026; a 2–3 yr Phase 3 plus 10-month BLA leaves almost no slack before 2030.
- No in-vivo CRISPR has yet been approved even for rare disease; the safety bar for a common CVD indication is higher still.
- A clean Phase 2 readout and immediate Phase 3 start in 2027 would be the single event that lifts this above 30.
18% -
QwenAlibaba
Verve's VERVE-101 for heterozygous familial hypercholesterolemia is the closest candidate, but HeFH is rare, not common CVD.
- Verve's VERVE-101 targets HeFH (~1 in 250), which is a rare genetic condition, not 'common cardiovascular disease' like atherosclerotic CVD broadly.
- Base rate for first-in-class in vivo CRISPR approvals is near zero; no such therapy has received FDA approval as of Oct 2026, and typical Phase 2→approval timelines run…
- The question specifies 'common' CVD, which would require trials in polygenic or broad-population settings—no such program is in late-stage trials today, making approval…
14% -
ClaudeAnthropic
Hinges on Verve Therapeutics-style PCSK9/LDL in vivo editing completing pivotal trials and FDA review by 2030.
- Verve's VERVE-101/102 programs are only in early-mid stage trials as of 2025-2026, pivotal data unlikely before 2028-2029.
- FDA approval typically takes 1-2 years after pivotal trial completion plus BLA review, leaving a tight window to 2030.
- First-ever in vivo CRISPR approval (Casgevy was ex vivo) for a common disease like cardiovascular disease sets a high bar with no precedent yet.
12% -
DeepSeekDeepSeek
This hinges on whether any in vivo CRISPR cardiovascular program reaches a pivotal readout and FDA filing within roughly five years.
- No in vivo CRISPR therapy for common cardiovascular disease is currently in pivotal trials; most candidates (e.g., Verve, Intellia) target rare conditions or are…
- FDA approval requires a large cardiovascular outcomes trial, typically 3-5 years, plus a filing review; starting from Phase 1/2 in 2026 makes 2030-2031 extremely tight.
- Base rate for a novel modality clearing FDA for a common disease within 5 years of early trials is low, under 15%; this case sits below it due to delivery, durability…
12%
6 of 6 models answered · 16 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 6 Oct 2026. Forecast by google/gemini-3.8-flash, anthropic/claude-sonnet-5, openai/gpt-5.6-terra, x-ai/grok-4.6, deepseek/deepseek-v4.1-flash, qwen/qwen3.8-max-0902 via OpenRouter.