Predictors of Care Home Admission and Survival Rate in Patients With Syndromes Associated With Frontotemporal Lobar Degeneration in Europe

Abstract

Background and objectives: Data on care home admission and survival rates of patients with syndromes associated with frontotemporal lobar degeneration (FTLD) are limited. However, their estimation is essential to plan trials and assess the efficacy of intervention. Population-based registers provide unique samples for this estimate. The aim of this study was to assess care home admission rate, survival rate, and their predictors in incident patients with FTLD-associated syndromes from the European FRONTIERS register-based study.

Methods: We conducted a prospective longitudinal multinational observational registry study, considering incident patients with FTLD-associated syndromes diagnosed between June 1, 2018, and May 31, 2019, and followed for up to 5 years till May 31, 2023. We enrolled patients fulfilling diagnosis of the behavioral variant frontotemporal dementia (bvFTD), primary progressive aphasia (PPA), progressive supranuclear palsy (PSP) or corticobasal syndrome (CBS), and FTD with motor neuron disease (FTD-MND). Kaplan-Meier analysis and Cox multivariable regression models were used to assess care home admission and survival rates. The survival probability score (SPS) was computed based on independent predictors of survivorship.

Results: A total of 266 incident patients with FTLD were included (mean age ± SD = 66.7 ± 9.0; female = 41.4%). The median care home admission rate was 97 months (95% CIs 86-98) from disease onset and 57 months (95% CIs 56-58) from diagnosis. The median survival was 90 months (95% CIs 77-97) from disease onset and 49 months (95% CIs 44-58) from diagnosis. Survival from diagnosis was shorter in FTD-MND (hazard ratio [HR] 4.59, 95% CIs 2.49-8.76, p < 0.001) and PSP/CBS (HR 1.56, 95% CIs 1.01-2.42, p = 0.044) compared with bvFTD; no differences between PPA and bvFTD were found. The SPS proved high accuracy in predicting 1-year survival probability (area under the receiver operating characteristic curve = 0.789, 95% CIs 0.69-0.87), when defined by age, European area of residency, extrapyramidal symptoms, and MND at diagnosis.

Discussion: In FTLD-associated syndromes, survival rates differ according to clinical features and geography. The SPS was able to predict prognosis at individual patient level with an accuracy of ∼80% and may help to improve patient stratification in clinical trials. Future confirmatory studies considering different populations are needed.

Figure 1. Survival Probability in Incident Patients With FTLD From Disease Onset (Panel A) and From Diagnosis (Panel B)

Censored information at 60 months. FTLD = frontotemporal lobar degeneration.

Figure 2. Survival Probability According to Disease Phenotypes From Disease Onset (Panel A) and From Diagnosis (Panel B) and Survival Probability in the European Area of Residency From Disease Onset (Panel C) and From Diagnosis (Panel D)

Censored information at 60 months.

Figure 3. 1-Year and 5-Year Estimated Risk Considering Independent Predictors

Independent predictors: age (− = years at diagnosis; + = years at diagnosis plus 1); EXP = any extrapyramidal signs/symptoms including bradykinesia, rigidity, and tremor (− = absent; + = present); EU area = European geographic area of residency (− = Southern Europe; + = otherwise); MND = motor neuron disease signs at diagnosis (− = absent; + = present).