論文ID: CR-25-0181
Background: The optimal period for discontinuing tafamidis in wild-type transthyretin amyloid cardiomyopathy (ATTRwt-CM) patients based on cost-effectiveness is unknown.
Methods and Results: A retrospective analysis of 80 tafamidis-treated ATTRwt-CM patients was conducted. The median follow up was 26.4 months; 17 patients permanently discontinued tafamidis, and 12 patients died after a median of 1.0 month following tafamidis discontinuation. Discontinuation was mainly due to non-cardiovascular hospitalization; deaths were mostly non-cardiovascular and occurred early after discontinuation.
Conclusions: The period from tafamidis discontinuation to death was shorter in ATTRwt-CM patients hospitalized for non-cardiovascular diseases. Discussions of the right period for discontinuation are needed.

Awareness of wild-type transthyretin amyloid cardiomyopathy (ATTRwt-CM) has increased greatly over the past 10 years due to the establishment of non-invasive diagnostic algorithms and the advent of disease-specific drugs. In clinical trials, tafamidis has been shown to contribute to improved survival rates and decreased cardiovascular hospitalization. From a medical economic perspective, tafamidis discontinuation should be considered in patients unlikely to derive sufficient clinical benefit; however, data on discontinuation in practice are limited. We analyzed the clinical data of patients in our hospital who permanently discontinued tafamidis.
A retrospective study of ATTRwt-CM patients who received tafamidis from April 2019 to August 2024 at Kochi Medical School Hospital was conducted. The cause of death was classified as cardiovascular death, non-cardiovascular death, and unknown death on the basis of the International Classification of Diseases, 10th revision (ICD-10) classification and prior research.1
Eighty patients (median age 79.4 years; interquartile range 74.2–83.1 years; males 86.3%) were included in the study. The median follow-up period was 25.4 (16.7–35.6) months, during which time 17 (21.3%) patients discontinued tafamidis permanently. The median period from the start of tafamidis administration to discontinuation was 21.7 (15.0–31.8) months. The median follow-up period after tafamidis discontinuation was 2.5 (0.9–14.1) months, during which time 12 (15.0%) patients died. In the 5 surviving discontinuation patients, the observation period from discontinuation was 14.1 (2.5–23.4) months, and the cause of discontinuation was cardiovascular hospitalization in 3 patients. For the patients who died, the median period from discontinuation to death was 1.0 (0.3–4.3) months. The most common cause of discontinuation was non-cardiovascular hospitalization, and of the causes of death, non-cardiovascular causes including infectious disease and malignant tumor were more common than cardiovascular causes (Figure). The interval between tafamidis discontinuation and death was shorter for patients who discontinued due to non-cardiovascular causes (0.7 [0.1–1.0] months) than for those who discontinued due to cardiovascular causes (4.9 [3.0–12.6] months; P=0.008).

Clinical course of wild-type transthyretin amyloid cardiomyopathy (ATTRwt-CM) patients who discontinued tafamidis. The gray bar shows the period from starting tafamidis to discontinuation, and the black bar shows the period from tafamidis discontinuation until censoring or death. ‘Age’ indicates the patient’s age at the time of initiation of tafamidis. COVID-19, coronavirus disease 2019; CV, cardiovascular; F, female; GI, gastrointestinal; LOS, low cardiac output syndrome; M, male.
When tafamidis was approved, its high cost attracted considerable controversy; for example, Kazi et al. pointed out that a 92.6% price reduction could be justified.2 The Japanese Circulation Society guidelines state that tafamidis should not be readily administered to patients who are expected to live less than 1 year.3 However, although the economic aspects have been widely scrutinized, there has been almost no discussion of the right time for discontinuation. Furthermore, the effects of discontinuing tafamidis during the terminal stage have not been fully clarified.
In the present cohort, tafamidis was discontinued during hospitalization in most patients. This probably relates to the way drug costs during hospitalization are paid for inclusively under the Japanese system of payment for medical services, so that doctors are reluctant to continue tafamidis due to its high price.
Several risk scores predict the mid- to long-term prognosis of ATTRwt-CM patients,4 but none accurately predict the short-term prognosis in the terminal stage. Development of a more precise prognostic tool and discussion of disease-modifying drug discontinuation, considering both clinical efficacy and efficient resource use, are needed.
The period from tafamidis discontinuation to death was shorter in ATTRwt-CM patients hospitalized for non-cardiovascular diseases. There is a need for discussing the right time to discontinue disease-modifying drugs from a medical economic perspective.
Y.B. has received a research grant from ITO EN, Ltd. T.K. and H.K. have received remuneration for lectures from Pfizer Japan Inc. The other authors declare no conflicts of interest.
This investigation was performed according to the Declaration of Helsinki and was approved by the Ethics Committee on Medical Research of Kochi Medical School (approval number: ERB-109221).
The deidentified participant data will not be shared.