Expert Guidance Issued for Real-World Use of Donanemab

An expert workgroup has developed the first appropriate use recommendations (AURs) for real-world use of the antiamyloid donanemab (Kisunla).

“The role of appropriate use recommendations are to bridge the gap between clinical trials and clinical practice,” said guideline author Gil Rabinovici, MD, with the Department of Neurology, University of California, San Francisco. 

“We know that patients that are seen in clinical trials in general are not representative of real-world patients. In the real world, patients are older, less well educated, have more comorbidities, [and] are more diverse, and so when new drugs are translated into clinical practice, there’s a little bit of a gap,” Rabinovici added. 

He also noted that clinicians may not have as much familiarity with the drug as clinical trial site investigators, and they don’t have the advantage of central review and monitoring. AURs therefore “complete the landscape” between clinical trial publications, prescribing information, and marketing information by providing an “independent academic perspective on how the drugs can be used in the real world.”

The donanemab AURs were presented at the 17th Clinical Trials on Alzheimer’s Disease (CTAD) conference.

The Appropriate Patient

The US Food and Drug Administration approved donanemab on July 2 for adults with early symptomatic Alzheimer’s disease, which includes mild cognitive impairment or mild dementia stage of disease with confirmed amyloid pathology. 

Approval was based on positive data from the phase 3 TRAILBLAZER-ALZ 2 trial, which showed that donanemab significantly reduced brain amyloid plaque burden and significantly slowed cognitive and functional decline compared with placebo in early symptomatic AD. 

The appropriate patient for treatment with donanemab has mild cognitive impairment or mild dementia (clinical stage 3-4; Mini-Mental State Examination score of 20-30) and confirmation of amyloid-beta pathology by PET or cerebrospinal fluid (CSF) analysis. 

“In the trial, this required quantitative amyloid PET, but in the real world, we’re recommending clinical visual reads of amyloid PET, or CSF, most reliably interpreted with ratios such as Aβ42/40 or p-tau181/Aβ42,” Rabinovici noted. 

He said there was a lot of discussion about whether blood-based biomarkers, in and of themselves, would be adequate for confirmation of pathology, and at the end of the day, the group decided that the timing was not quite right. 

“There’s still too much variability in the performance of some of the commercially available blood biomarkers. There isn’t a lot of experience in real-world clinical practice, but we do acknowledge that blood biomarkers may be sufficient for treatment eligibility in the near future,” Rabinovici told conference attendees. 

Appropriate patients for donanemab should have no clinical or MRI findings of a primary non–Alzheimer’s disease cause of impairment; should have no other condition that compromises their understanding of treatment goals or adherence; must have support systems in place, including a care partner, to make sure that they can engage in shared decision-making and are able to adhere with safety monitoring. 

Imaging, Genotyping, Monitoring 

In the clinical trial, tau PET was very useful in predicting clinical response, Rabinovici noted, with patients who started with lower baseline tau, in general, showing a greater degree of clinical response. However, clinical tau PET is not really accessible in most regions in the United States at this time. 

Therefore, the recommendation at this time is that tau PET is not required for inclusion, but if available, it can be used to individualize the estimate of clinical response when discussing risks and benefits with patients.

The prescribing information for donanemab includes a boxed warning for amyloid-related imaging abnormalities (ARIA). Patients who are APOE epsilon 4 homozygotes have a higher incidence of ARIA, including symptomatic and serious ARIA, compared with heterozygotes and noncarriers.

“The recommendation of the committee is that APOE genotyping should be performed prior to treatment informed discussions about ARIA risk, but the AUR does not exclude anyone based on APOE genotype, and they can be treated, provided they understand the risks and are interested in proceeding with treatment,” Rabinovici said. 

“It’s extremely important for patients and emergency services serving those patients to understand that ARIA can present with focal findings that mimic an acute stroke,” Rabinovici noted. 

Intravenous thrombolysis is contraindicated in patients receiving donanemab on the basis of multiple reported fatal cases of ARIA-effusion/hemorrhage in patients treated with thrombolytics on a background of antiamyloid therapy, the workgroup advised. 

A pretreatment MRI should be obtained within 1 year of initiating treatment and patients showing severe white-matter disease or evidence of significant cerebral amyloid angiopathy (eg, four or more microbleeds, macrohemorrhage > 1 cm) should be excluded. 

Although patients taking anticoagulants were not at increased risk for ARIA in the TRAILBLAZER trials, the data are limited, and the AURs “conservatively” advise against donanemab treatment for patients on anticoagulants, especially given the high morbidity and mortality associated with intracerebral macrohemorrhages, Rabinovici said. There is no restriction on antiplatelet use.

Clinical Judgment Paramount 

Donanemab is administered at 700 mg intravenously every 4 weeks for the first 3 doses, followed by 1400 mg intravenously every 4 weeks. Surveillance MRI should be performed before the second, third, and fourth infusions, and at any time that symptoms suspicious for ARIA occur, the workgroup advised. 

Clinicians can consider stopping therapy if follow-up amyloid PET (typically 12-18 months after initiating treatment) is read as negative. 

Rabinovici emphasized that AURs “are not guidelines or criteria, and as always, when treating an individual patient, clinical judgment is paramount.”

He also noted that there is a crucial need for real-world data on longer term efficacy and safety of novel Alzheimer disease therapeutics that are collected via registry, such as ALZNET, and he encouraged clinicians treating patients with donanemab to consider joining the registry.

The manuscript describing the donanemab AURs is being submitted for peer review and will hopefully be published soon, he said. 

This research had no commercial funding. Rabinovici has received research support from Avid Radiopharmaceuticals, Eli Lilly, Genentech, GE Healthcare, and Life Molecular Imaging. He has served as a paid scientific advisor for Alector, Avid Radiopharmaceuticals, C2N, Eli Lilly, GE Healthcare, Johnson & Johnson, Merck, Novo Nordisk, and Roche. He is an associate editor for JAMA Neurology.