Neurology · Medical Literature
Clinical Meaningfulness of Donanemab in Early Symptomatic Alzheimer Disease: Data From the Randomized Phase 3 TRAILBLAZER-ALZ 2 Trial.
Abstract
Understanding the meaningfulness of clinical trial outcomes is essential for people living with Alzheimer disease (AD) and their clinicians to make evidence-based shared treatment decisions in real-world clinical care. Donanemab, a monoclonal antibody targeting the insoluble form of β-amyloid found in plaques, significantly slows cognitive and functional decline of AD in participants with mild cognitive impairment (MCI) or mild AD-related dementia. This analysis reviews the efficacy of donanemab across various clinical outcome assessments, using both published data and new complementary analyses to provide context on its potential benefits for patients and caregivers. We present findings from prespecified and post hoc analyses from the TRAILBLAZER-ALZ 2 trial. Clinical outcomes assessed were Integrated AD Rating Scale (iADRS), comprising the 13-item AD Assessment Scale-Cognitive Subscale (ADAS-Cog13) and AD Cooperative Study-Instrumental Activities of Daily Living (ADCS-iADL); Clinical Dementia Rating (CDR)-Sum of Boxes (CDR-SB) for clinical severity and individual cognitive and functional domains; CDR-Global for clinical severity stage progression; meaningful within-patient change (MWPC); and ADCS-Activities of Daily Living dependence score. Donanemab reduced the risk of progression from MCI to mild AD by 33% (hazard ratio [HR] = 0.67; 95% CI 0.52-0.87; p = 0.003) and from mild to moderate AD by 50% (HR = 0.50; 95% CI 0.33-0.78; p = 0.002). In addition, donanemab reduced MWPC risk over 76 weeks by 38% for CDR-SB (HR = 0.62; 95% CI 0.52-0.75; p < 0.001) and 30% for iADRS (HR = 0.70; 95% CI 0.58-0.84; p < 0.001). Donanemab-treated participants exhibited significant slowing of clinical progression across multiple ADAS-Cog13 and ADCS-iADL items and all CDR-SB cognitive and functional domains. Donanemab also slowed progression of dependence least-squares mean change difference, -0.14 [95% CI -0.24 to -0.04; p = 0.007]), representing 23% slowing of progression (95% CI 6.17%-40.32%), and reduced risk of progression to requiring in-home support by 27% (HR = 0.74; 95% CI 0.59-0.91; p = 0.005). These results add to the evidence and further support clinically meaningful donanemab-mediated effects on cognition and function for patients and their caregivers and may aid communication of realistic treatment expectations and informed decision-making. ClinicalTrials.gov NCT04437511. Submitted: June 17, 2020; First patient enrolled: June 19, 2020. clinicaltrials.gov/study/NCT04437511 EudraCT Number 2020-000077-25. Start date of recruitment: June 19, 2020. clinicaltrialsregister.eu/ctr-search/trial/2020-000077-25/results.
Clinical relevance
Does donanemab treatment produce clinically meaningful improvements—across multiple cognitive, functional, stage-progression, and within-patient change metrics—for patients with early symptomatic Alzheimer disease (mild cognitive impairment or mild dementia due to AD)? __SNAPSHOT__ {"studyDesign":"Secondary analysis (prespecified and post hoc analyses) of a Phase 3 randomized, double-blind, placebo-controlled, parallel-group, multicenter trial (TRAILBLAZER-ALZ 2)","setting":"277 medical research centers and hospitals across 8 countries (international, multinational); conducted June 2020 to April 2023","sampleSize":"n=1,736 total (donanemab: n=860, placebo: n=876); primary efficacy analysis conducted in the low/medium tau subpopulation (n=1,182) and the combined population (n=1,736)","population":"Adults aged 60–85 years with early symptomatic Alzheimer disease (MCI or mild AD dementia), confirmed amyloid and tau pathology on PET imaging, and a ≥6-month history of gradual and progressive memory change. Stratified by tau burden: low/medium tau (n=1,182; primary target population) and high tau (n=552; more advanced disease).","intervention":"Donanemab (700 mg IV every 4 weeks for the first 3 doses, then 1,400 mg IV every 4 weeks) for up to 72 weeks; participants switched to blinded placebo upon meeting amyloid clearance criteria","comparator":"Placebo administered intravenously every 4 weeks for 76 weeks","primaryOutcome":"Change from baseline in the Integrated Alzheimer Disease Rating Scale (iADRS; range 0–144, lower scores = greater impairment) at 76 weeks in the low/medium tau population. Secondary and complementary endpoints included CDR-Sum of Boxes (CDR-SB), CDR-Global stage progression, individual ADAS-Cog13 and ADCS-iADL item-level analyses, meaningful within-patient change (MWPC) thresholds, and ADCS-Activities of Daily Living dependence score.","followUp":"76 weeks (approximately 18 months) for the primary placebo-controlled period","fundingSource":"Industry (Eli Lilly and Company); trial designed, funded, and data analyzed by Eli Lilly"}
Key findings
In the low/medium tau population (primary analysis), donanemab slowed iADRS decline by 35.1% (95% CI 19.90–50.23) versus placebo at 76 weeks; in the combined population (all tau levels), slowing was 22.3% (95% CI 11.38–33.15; p<0.001). For CDR-SB in the low/medium tau population, between-group difference was –0.67 (95% CI –0.95 to –0.40), representing 36% slowing of clinical progression. Donanemab reduced the risk of CDR-Global stage progression by 38.6% in the low/medium tau population (HR=0.614; 95% CI 0.471–0.800; p<0.001). Stage-specific progression risk was also reduced: MCI to mild AD by 33% (HR=0.67; 95% CI 0.52–0.87; p=0.003) and mild to moderate AD by 50% (HR=0.50; 95% CI 0.33–0.78; p=0.002). Meaningful within-patient change (MWPC) risk over 76 weeks was reduced by 38% for CDR-SB (HR=0.62; 95% CI 0.52–0.75; p<0.001) and 30% for iADRS (HR=0.70; 95% CI 0.58–0.84; p<0.001). Donanemab also slowed progression of functional dependence (least-squares mean difference –0.14; 95% CI –0.24 to –0.04; p=0.007), representing 23% slowing (95% CI 6.17%–40.32%), and reduced the risk of requiring in-home support by 27% (HR=0.74; 95% CI 0.59–0.91; p=0.005). Significant slowing of clinical progression was observed across multiple individual ADAS-Cog13 and ADCS-iADL items and all CDR-SB cognitive and functional domains. Across statistical approaches, treatment was associated with 4.4–7.5 months saved over 18 months in the low/medium tau population. An estimated 47% of donanemab-treated participants had no CDR-SB change at 1 year, versus 29% on placebo. Regarding safety: ARIA-edema/effusions (ARIA-E) occurred in approximately 24% of donanemab-treated participants versus 1.9% on placebo; ARIA-microhemorrhages and hemosiderin deposition (ARIA-H) in 31.4% versus 13.6%. Serious ARIA-E was reported in 1.5–1.6% of donanemab-treated participants; three deaths in the trial were attributed to ARIA. ARIA events were mostly mild-to-moderate, predominantly asymptomatic, and clustered early (58.3% of first ARIA-E events by the third infusion). APOE ε4 carrier status was strongly associated with increased ARIA-E risk (homozygotes OR=4.57; 95% CI 3.27–6.39; p<0.001; heterozygotes OR=2.03; 95% CI 1.55–2.67; p<0.001). Infusion-related reactions occurred in 8.7% of participants, most mild-to-moderate. NNT to prevent one meaningful within-patient change in iADRS score: approximately 11 (calculated, per independent pharmacist analysis of trial data). High tau participants (n=552) showed no statistically significant benefit on the primary iADRS outcome or most secondary clinical outcomes, with the exception of CDR-SB.
Practice implications
This secondary analysis of TRAILBLAZER-ALZ 2 establishes that donanemab produces statistically significant and, by prespecified and post hoc criteria, clinically meaningful slowing of cognitive and functional progression across multiple validated outcome measures in patients with early symptomatic AD and low/medium tau burden. The breadth of consistency—across composite scales, individual domain items, stage-transition risk, functional dependence metrics, and MWPC analyses—strengthens the internal validity of the efficacy signal. The most clinically interpretable evidence is the reduction in CDR-Global stage progression (38.6% risk reduction) and stage-specific transition risk, as any change on CDR-Global represents a discrete, inherently meaningful shift in disease severity. The absolute benefit in months of disease slowing (4.4–7.5 months over 18 months) is modest but may have meaningful downstream implications given the accelerating trajectory of decline at later stages. Findings apply most robustly to patients who mirror the trial population: age 60–85, early symptomatic AD (MCI or mild dementia), confirmed amyloid and low/medium tau pathology on PET, and baseline clinical characteristics consistent with the TRAILBLAZER-ALZ 2 cohort. Patients with high tau burden (more advanced disease) did not demonstrate significant benefit on the primary outcome and most secondary clinical endpoints and are not adequately represented as treatment-responsive. Populations not studied or underrepresented include: patients without PET-confirmed pathology, those >85 years, racial/ethnic minorities (trial demographics not fully detailed in abstract), patients with significant comorbidities excluded from the trial, and those ineligible for ARIA-related MRI monitoring. The CDR-SB between-group differences did not consistently meet pre-established minimal clinically important difference (MCID) thresholds when applied at the group level, and interpreting group-level CDR-SB differences using MCID values designed for individual patients is methodologically inappropriate. Critically, the paper is authored almost entirely by Eli Lilly employees or industry-affiliated investigators, and the primary assessment instrument (iADRS) was developed with Lilly involvement; independent replication and real-world effectiveness data are needed. ARIA risk is clinically significant: 24% ARIA-E and 31% ARIA-H incidence, with three treatment-attributed deaths, mandates formal pre-treatment APOE genotyping and serial MRI monitoring, with particular vigilance in APOE ε4 homozygotes. A significant proportion of patients discontinued the study (26.9% donanemab, 19.7% placebo), which may affect generalizability. Practice should not change based on this paper alone, but the cumulative evidence from TRAILBLAZER-ALZ 2 (of which this is a secondary analysis) supports use in carefully selected patients when the risk-benefit profile is systematically assessed with shared decision-making. This publication is based on abstract-only data as supplied; the full manuscript should be reviewed for complete methodological detail, especially regarding post hoc analysis multiplicity controls. Note: only the abstract was provided for this summary; conclusions should be confirmed against the full published manuscript. __LIMITATIONS__ Substantial industry conflict of interest: study designed, funded, and analyzed by Eli Lilly; most authors are Lilly employees or affiliates; the primary outcome instrument (iADRS) was developed with Lilly involvement.,Mixture of prespecified and post hoc analyses without explicit multiplicity correction for the many endpoints reported; risk of type I error inflation.,High discontinuation rates: 26.9% in the donanemab group and 19.7% in placebo, potentially biasing efficacy estimates.,Tau-stratified analysis design means high tau patients (n=552, more advanced disease) were underpowered for the primary endpoint and showed no significant benefit, limiting generalizability to this subgroup.,Trial required PET-confirmed amyloid and tau pathology for enrollment; real-world diagnostic infrastructure for PET-based eligibility screening is limited in most clinical settings.,Short follow-up (76 weeks/18 months) relative to the decades-long course of AD; long-term durability of effect and cumulative ARIA risk beyond 18 months not fully characterized within this study.,Variable treatment duration due to individualized amyloid clearance-based dose stopping creates heterogeneity in exposure that complicates safety and efficacy interpretation.,CDR-SB between-group differences did not consistently meet established MCID thresholds at the group level; cross-applying individual-level MCID thresholds to group-level differences is methodologically problematic.,Limited demographic diversity; representativeness for non-White, non-European populations not established from available abstract data.,ARIA-related deaths (n=3) and high ARIA incidence (ARIA-E ~24%, ARIA-H ~31%) represent a significant safety burden not fully mitigated by real-world monitoring protocols, which may be less intensive than the trial's protocol.,Only abstract was provided for this summary; full manuscript review is required for complete assessment of statistical methodology, covariate adjustment, and handling of missing data. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In patients aged 60–85 with early symptomatic Alzheimer disease and PET-confirmed low/medium tau and amyloid pathology, donanemab significantly slows cognitive and functional decline across multiple validated endpoints, reduces risk of CDR-Global stage progression by ~39%, and lowers risk of clinically meaningful within-patient worsening—with 4.4–7.5 months of disease slowing over 18 months. These benefits must be weighed against a substantial ARIA burden (ARIA-E ~24%, ARIA-H ~31%, three ARIA-attributed deaths), exclusive industry sponsorship, and a high-tau subgroup showing no significant primary outcome benefit. Donanemab is a potentially practice-informing but risk-bearing therapy requiring rigorous patient selection, APOE genotyping, and serial MRI monitoring within a structured shared decision-making framework.