← Back to Home

Mixed Phase 3 Results for AriBio’s Oral Alzheimer’s Drug: 31% Slower Decline in Monotherapy Population, No Significant Benefit Overall

AR1001 met the prespecified efficacy threshold in 604 patients not taking certain existing therapies but did not demonstrate a benefit across all 1,535 participants. The results offer clues for oral treatment while putting the eligible patient population and strength of evidence at the center of the next regulatory discussions.

By SURL BioNews

If a daily pill could slow the progression of Alzheimer’s disease, patients and caregivers might gain more time. However, the Phase 3 trial of AR1001 reported by South Korea’s AriBio provides an answer with clear limits: efficacy was observed in the prespecified monotherapy population, while results in the overall trial population did not reach statistical significance. The oral drug’s potential is emerging, but the evidence is not yet sufficient to extend the findings to all patients with early Alzheimer’s disease.

The randomized, double-blind, placebo-controlled trial, named POLARIS-AD, enrolled 1,535 patients with early Alzheimer’s disease to receive once-daily AR1001 or placebo for 52 weeks. AriBio presented preliminary results at the CTAD 2026 conference in Boston on November 16, Eastern Time. The primary assessment tool was the Clinical Dementia Rating–Sum of Boxes (CDR-SB), used to track changes in cognition and daily functioning.

In the analysis of all participants, the difference between AR1001 and placebo was −0.11 points and did not reach statistical significance (p=0.4015). In the 604-patient monotherapy population, however, the difference between the two groups reached −0.44 points (p=0.0266), equivalent to a 31% reduction in clinical decline. This population was defined by the absence of acetylcholinesterase inhibitor or memantine use at the start of the trial; the 31% refers to the reduction in decline relative to placebo, not a 31% recovery in cognitive ability.

The key to interpreting these results is that both analysis populations were prespecified. According to information released by Fosun Pharma, the significance threshold was 0.03 for the monotherapy population and 0.02 for the overall population. AriBio said the prespecified multiple-testing procedure allowed the primary efficacy objective to be met if either population reached its threshold. The monotherapy result therefore differs from a favorable analysis selected only after the trial ended. However, the differing results between the two populations still do not directly establish that existing medications weakened AR1001’s effect; the reasons require further analysis.

AR1001 is a phosphodiesterase type 5 (PDE5) inhibitor intended to work by enhancing intracellular cGMP signaling. Cerebrospinal fluid analyses released by the company showed increased cGMP concentrations in the treatment group, supporting the drug’s intended biological activity in the central nervous system. However, a gap remains between biomarker changes and long-term patient benefit. Results on another cognitive scale, ADAS-Cog 13, also failed to reach statistical significance in the monotherapy population, a reminder that the various measures cannot all be treated as confirmation of efficacy.

Safety also affects the practicality of this long-term treatment. According to the company’s disclosures, serious adverse events occurred in 10.0% of the AR1001 group and 9.8% of the placebo group; treatment discontinuation due to adverse events occurred in 5.2% and 1.7%, respectively. Diarrhea was more common in the treatment group. These differences need to be weighed alongside the potential clinical benefit; a lower treatment burden cannot be assumed solely from the oral formulation.

AriBio plans to discuss the full data with the U.S. Food and Drug Administration; AR1001 has not yet been approved for Alzheimer’s disease. Fosun Pharma, which holds an option to license the drug in certain global markets, also said it would further evaluate the data before deciding whether to exercise that option. The most immediate practical questions are whether the benefit in the monotherapy population can provide sufficient evidence for marketing approval and whether more trials are needed to clarify which patients can benefit from treatment.

References

  1. BioSpace