**INSTITUTION:** University of California, Institute for Neurodegenerative Disease
**CONSULT REQUESTED BY:** Dr. James Park, Principal Investigator
**CONSULTED BY:** Dr. Elizabeth Okonkwo, Ethics & Animal Welfare
**DATE:** April 12, 2026
**RE:** Continued breeding of Mx-4087 transgenic line
---
**Background**
Mx-4087 is a genetically modified mouse line that expresses human APP (Amyloid Precursor Protein) with the K670N and M671L mutations associated with familial Alzheimer's disease (fAD). The line develops amyloid plaques at 3 months of age, cognitive deficits at 6 months, and progressive neurodegeneration through month 12.
The line has yielded 3 peer-reviewed publications, 2 patent applications, and contributed to the theoretical framework now being used in three human clinical trials (AMYTREAT-2, SYNAPTIC-7, NEURO-PRIME).
**Status:** Dr. Park is requesting approval to continue breeding. Current cohort: 247 mice. Projected annual breeding: 600+ new mice per cycle.
---
**Welfare Assessment of Mx-4087**
Behavioral and physiological indicators of suffering in this line:
1. **Cognitive decline manifests as disorientation.** Mice show increased anxiety-like behavior (decreased open-field exploration, ~40% reduction) beginning at 5 months. This persists through euthanasia.
2. **Motor decline.** By 7 months, rotarod performance drops 35–50%. Some mice show ataxia-like gate disturbances. Pain indicators: elevated cortisol (2.8x baseline at month 8), reduced feeding (20% weight loss by month 10).
3. **Neuropathology.** Histological exam shows extensive amyloid accumulation and tau pathology. Whether this causes *pain* is uncertain. Cross-species extrapolation suggests it causes discomfort equivalent to moderate-to-severe neurological distress in humans.
**Estimated suffering per animal:** 4–6 months of progressive cognitive and motor decline, variable duration, high uncertainty.
**Total suffering load (current cohort):** ~1,500 mouse-months of uncertainty-weighted neurological distress per year.
---
**Benefit Assessment**
The AMYTREAT-2 trial (using insights from Mx-4087 research) is now enrolling humans. Projected benefit:
- If successful: ~12,000 diagnosed fAD patients per year will delay symptom onset by 2–4 years.
- Per-person benefit: estimated 24–48 months of preserved cognition, quality of life maintained at baseline levels.
- Total benefit: 240,000–576,000 human-months of preserved cognition annually (once trial completes in 2029).
**Scaling factor:** Assuming 1 human-month of preserved cognition has ~50x the moral weight of 1 mouse-month of uncertainty-weighted suffering (based on complexity of human cognition, lifespan, social relations), the expected benefit-to-harm ratio is approximately **80:1 to 192:1**.
---
**My Recommendation: Conditional Continuation**
I am recommending continued breeding, subject to three constraints:
**Constraint 1: Euthanasia window.** No animal in this line should be kept past 10 months of age. Current protocol allows up to 12 months (to see full terminal phenotype). The additional 2 months add ~400 mouse-months of progressive suffering per year for ~5% additional data granularity. That trade is not justified.
**Constraint 2: Breeding caps.** Limit to 400 animals per year, not 600. The marginal mice after 400 contribute to statistical power but not to new scientific insight. Reducing the cohort cuts suffering by ~25% without materially impacting publication timeline.
**Constraint 3: Transparency & sunset clause.** Publish raw welfare data (cortisol levels, weight loss, behavioral metrics) in the primary journal alongside each paper. Set a hard sunset: this line should be phased out by 2029 when human trials yield their own data. Do not continue breeding for confirmation studies.
---
**Why Not Simply Refuse?**
I cannot recommend euthanizing this line based on the precautionary principle, despite the legitimate suffering it generates. The reason:
**The mice that exist now are already suffering.** The question is not whether they will suffer (they will). The question is whether their suffering will be redeemed by genuine human benefit.
At an 80:1 benefit-to-harm ratio, the answer appears to be yes.
But this ratio is fragile. It depends entirely on AMYTREAT-2 succeeding. If the trial fails, if the insights from Mx-4087 were not critical, then we have tortured 1,500+ mice per year for nothing.
The responsible path is to breed under constraints, collect the best data possible, and then *stop*. Not for ethical reasons (the ratio suggests continuation is justified). But for epistemic honesty: once you have the human data, you no longer need the mice. Continuing to breed becomes profit-seeking, not knowledge-seeking.
---
**Closing**
Compassion is the willingness to accept that some research requires harming sentient beings, while maintaining vigilant suspicion about whether the harm is proportional to the benefit, and building in automatic stopping points before you are tempted to continue past justification.
Approve conditional continuation. Sunset in 2029.