Darolutamide associated with less cognitive change than enzalutamide in prostate cancer

Patients with advanced prostate cancer who received darolutamide (Nubeqa) experienced significantly less change in objectively measured cognitive function at 48 weeks than those treated with enzalutamide (Xtandi), according to findings from the phase 2 ARACOG (AFT-47) trial presented at the 2026 European Society for Medical Oncology Annual Congress in Madrid, Spain.¹

Additionally, the authors reported a non-significant by clinically meaningful difference in perceived cognitive change by week 48 favoring darolutamide. The longer-term analyses, which spanned through week 48, were consistent with previously reported outcomes from the 24-week analysis presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago, Illinois.

Trial design and cognitive assessments

ARACOG (NCT04335682) is a randomized, open-label phase 2 study. Between August 2021 and March 2025, investigators enrolled 111 patients with metastatic hormone-sensitive prostate cancer, metastatic castration-resistant prostate cancer (CRPC), or nonmetastatic CRPC. Participants were randomly assigned 1:1 to receive darolutamide (n = 55) or enzalutamide (n = 56).

Investigators assessed cognition using the Functional Assessment of Cancer Therapy–Cognitive Function (FACT-Cog) test, a patient-reported cognitive measure, and the Cambridge Neuropsychological Test Automated Battery (CANTAB), a computerized assessment covering 5 cognitive domains. Testing was conducted at baseline and weeks 12, 24, and 48.

Crossover was permitted at weeks 12 or 24 if patients met prespecified criteria. Patients who crossed over were evaluated at crossover and analyzed according to their randomized treatment group.

By week 24, 65 participants were eligible to switch treatment, including 32 in the darolutamide group and 33 in the enzalutamide group. In total, 30 patients crossed over by week 24, all from enzalutamide to darolutamide. The most common reasons for crossover included objective cognitive decline (n = 14), subjective cognitive decline (n = 11), and neurologic toxicity (n = 8).

At week 48, the median change in the maximally changed CANTAB domain (MCCD) from baseline was −20.0% (IQR, - 41.7 to -4.4) with darolutamide and −33.3% (IQR, -50.0 to -20.0) with enzalutamide, with a median between-group difference of 12.7 percentage points (95% CI, 0.0 to 25.0; P = .043). The median change across individual CANTAB domains was improved with darolutamide over enzalutamide.

Study author Alicia K. Morgans, MD, discussed this phenomenon in a previous interview with Urology Times®, noting, “It's important to know that the learning effect does not suggest that anybody's cognitive function improved. The learning effect is a well-described finding in cognitive test studies where patients take the same test and seem to have better scores over time. It's simply because they're taking the same tests and they're learning how to take the test. So we expect that when people have stable cognitive function—not improved cognitive function—they will actually have better test scores over time. We did see this with the darolutamide-treated patients, suggesting stable cognitive function. For those patients treated with enzalutamide, they actually had stable cognitive test scores, or slightly decreased cognitive test scores, which means that on repeated testing their scores didn't get better, which then suggests that their cognitive function had some mild degree of decline.”

Looking at patient-reported cognitive outcomes at week 24, the median change in the FACT-Cog total score was −0.5 (IQR, -9 to 5) with darolutamide and −5 (IQR, -25 to 2) with enzalutamide. The between-group difference was 4.4 points (95% CI, −4.4 to 13.2; P = .32). At week 48, the corresponding median changes were −2.0 and −12.0, respectively, with a between-group difference of 10.5 points (95% CI, −0.8 to 21.7; P = .067). Neither comparison reached statistical significance.

________

REFERENCE

1. Kwon DH, Bobek O, Berg SA, et al. Long-term patient-reported and objective cognitive effects of darolutamide vs enzalutamide in ARACOG (AFT-47). Presented at: 2026 European Society for Medical Oncology Congress. October 23-27, 2026. Madrid, Spain. Poster 2280P