Both Amyloid and Tau Assessments Are Desired Before Initiating Disease-modifying Therapy for Alzheimer Disease: Three Real-world Cases

To the Editor: We have recently highlighted the potential pitfalls of amyloid-beta (Aβ) PET, suggesting that both Aβ and tau assessments are necessary to relate clinical symptoms to Alzheimer disease (AD) pathology.1 Herein, we would like to present 3 additional cases related to the clinical practice of disease-modifying therapy (DMT) for AD. Lecanemab and donanemab, which are monoclonal antibodies directed against aggregated forms of Aβ, are currently available for clinical use as the DMT under universal health insurance system in Japan. According to the optimal clinical use guidelines for the 2 drugs, tau assessment is not required before initiating either medication, although Aβ assessment is mandatory. One of the reasons is that the Japanese insurance system covers Aβ tests, but not tau tests. The relationship between Aβ and tau assessments is classified into 4 categories depending on the status of each biomarker (Fig. 1): A(+) T(+), A(-) T(-), A(+) T(-), and A(-) T(+). Positive or negative Aβ PET is assessed qualitatively by visual inspection, or quantitatively through SUVR and Centiloid measurements. Tau PET is usually assessed by using PET-based Braak stage.2 Consistent with histopathologic studies,3,4 little change in cognitive assessments is observed from PET-based Braak stages 0–II.2 Therefore, the extent of tau accumulation corresponding to an asymptomatic condition is likely equivalent to PET-based Braak stages 0–II, although this is potentially debatable due to the limited evidence.FIGURE 1: Tau PET using 18F-MK-6240 in patients who received lecanemab or donanemab treatment as a disease-modifying therapy for Alzheimer disease. All patients were confirmed to be A(+). No accumulation of 18F-MK-6240 was identified in Patient 1, corresponding to PET-based Braak stage 0. 18F-MK-6240 binding was confined to the entorhinal area (white arrows) in Patients 2 and 3, corresponding to PET-based Braak stage I/II. Patients 1, 2, and 3 received lecanemab treatment. In addition, representative cases (Patients 4 and 5) of “A(+) T(+)” are shown as positive examples. Patients 4 and 5 showed PET-based Braak stage III/IV and V/VI, respectively. MMSE and CDR global scores were 22 and 0.5 in Patient 4, and 24 and 0.5 in Patient 5, respectively. Patients 4 and 5 received donanemab and lecanemab treatments, respectively.A(+) T(+) and A(-) T(-) indicate symptomatic AD and non-AD, respectively. We have recently reported that cases of A(+) T(-) and A(-) T(+) may carry a risk of mistakenly linking cognitive decline to AD pathology, particularly when Aβ is the only biomarker evaluated.1 Here, we present 3 patients (Patients 1, 2, and 3) with amnestic mild cognitive impairment or mild dementia, who received the DMT. MMSE scores of Patients 1, 2, and 3 were 23, 23, and 24, respectively. The corresponding CDR global scores were 1, 0.5, and 0.5, respectively. All patients were confirmed to be A(+), and subsequently received lecanemab treatment. Later, they underwent tau PET using 18F-MK-6240, and were confirmed to be T(-) (Fig. 1). Patient 1 showed PET-based Braak stage 0 (ie, no accumulation). Patients 2 and 3 showed PET-based Braak stage I/II (ie, confined to the entorhinal area). These results from tau assessments indicate that their symptoms are irrelevant to AD pathology, even if the Aβ biomarker is positive. Meanwhile, based on the phase 3 trial results,5,6 lecanemab and donanemab are currently designed to be used in patients who exhibit mild cognitive impairment or mild dementia due to AD pathology. Therefore, our 3 cases exemplify that evaluating only Aβ pathology may lead to a misdiagnosis of AD as the cause of symptoms, thereby resulting in the unanticipated use of DMT. We strongly suggest that both Aβ and tau assessments are important to relate clinical symptoms to AD pathology, and are desired to ensure the expected use and evaluation of DMT.