Deep learning to predict rapid progression of Alzheimer's disease from pooled clinical trials: A retrospective study.

The rate of progression of Alzheimer's disease (AD) differs dramatically between patients. Identifying the most is critical because when their numbers differ between treated and control groups, it distorts the outcome, making it impossible to tell whether the treatment was beneficial. Much recent effort, then, has gone into identifying RPs. We pooled de-identified placebo-arm data of three randomized controlled trials (RCTs), EXPEDITION, EXPEDITION 2, and EXPEDITION 3, provided by Eli Lilly and Company. After processing, the data included 1603 mild-to-moderate AD patients with 80 weeks of longitudinal observations on neurocognitive health, brain volumes, and amyloid-beta (Aß) levels. RPs were defined by changes in four neurocognitive/functional health measures. We built deep learning models using recurrent neural networks with attention mechanisms to predict RPs by week 80 based on varying observation periods from baseline (e.g., 12, 28 weeks). Feature importance scores for RP prediction were computed and temporal feature trajectories were compared between RPs and non-RPs. Our evaluation and analysis focused on models trained with 28 weeks of observation. The models achieved robust internal validation area under the receiver operating characteristic (AUROCs) ranging from 0.80 (95% CI 0.79-0.82) to 0.82 (0.81-0.83), and the area under the precision-recall curve (AUPRCs) from 0.34 (0.32-0.36) to 0.46 (0.44-0.49). External validation AUROCs ranged from 0.75 (0.70-0.81) to 0.83 (0.82-0.84) and AUPRCs from 0.27 (0.25-0.29) to 0.45 (0.43-0.48). Aß plasma levels, regional brain volumetry, and neurocognitive health emerged as important factors for the model prediction. In addition, the trajectories were stratified between predicted RPs and non-RPs based on factors such as ventricular volumes and neurocognitive domains. Our findings will greatly aid clinical trialists in designing tests for new medications, representing a key step toward identifying effective new AD therapies.

Morbidity and Mortality Associated With Heart Failure in Acute Coronary Syndrome: A Pooled Analysis of 4 Clinical Trials.

BACKGROUND: Heart failure (HF) may complicate acute coronary syndrome (ACS) and is associated with a high burden of short- and long-term morbidity and mortality. Only limited data regarding future ischemic events and rehospitalization are available for patients who suffer HF before or during ACS. METHODS: A secondary analysis of 4 large ACS trials (PLATO, APPRAISE-2, TRACER, and TRILOGY ACS) using Cox proportional hazards models was performed to investigate the association of HF status (no HF, chronic HF, de novo HF) at presentation for ACS with all-cause and cardiovascular death, major adverse cardiovascular event (MACE ), myocardial infarction, stroke, and hospitalization for heart failure (HHF) by 1 year. Cumulative incidence plots are presented at 30 days and 1 year. RESULTS: A total of 11.1% of the 47,474 patients presenting with ACS presented with evidence of acute HF, 55.0% of whom presented with de novo HF. Patients with chronic HF presented with evidence of acute HF at a higher rate than those with no previous HF (40.3% vs 6.9%). Compared to those without HF, those with chronic and de novo HF had higher rates of all-cause mortality (adjusted hazard ratio [aHR] 2.01, 95% confidence interval [CI] 1.72-2.34 and aHR 1.47, 95% CI1.15-1.88, respectively), MACE (aHR 1.47, 95% CI1.31-1-.66 and aHR 1.38, 95% CI1.12-1.69), and HHF (aHR 2.29, 95% CI2.02-2.61 and aHR 1.48, 95% CI 1.20-1.82) at 1 year. CONCLUSION: In this large cohort of patients with ACS, both prior and de novo HF complicating ACS were associated with significantly higher risk-adjusted rates of death, ischemic events and HHF at 30 days and 1 year. Further studies examining the association between HF and outcomes in this high-risk population are warranted, especially given the advent of more contemporary HF therapies.

Population-Based Mini-Mental State Examination Norms in Adults of Mexican Heritage in the Cameron County Hispanic Cohort.

BACKGROUND: Accurately identifying cognitive changes in Mexican American (MA) adults using the Mini-Mental State Examination (MMSE) requires knowledge of population-based norms for the MMSE, a scale which has widespread use in research settings. OBJECTIVE: To describe the distribution of MMSE scores in a large cohort of MA adults, assess the impact of MMSE requirements on their clinical trial eligibility, and explore which factors are most strongly associated with their MMSE scores. METHODS: Visits between 2004-2021 in the Cameron County Hispanic Cohort were analyzed. Eligible participants were ≥18 years old and of Mexican descent. MMSE distributions before and after stratification by age and years of education (YOE) were assessed, as was the proportion of trial-aged (50-85- year-old) participants with MMSE <24, a minimum MMSE cutoff most frequently used in Alzheimer's disease (AD) clinical trials. As a secondary analysis, random forest models were constructed to estimate the relative association of the MMSE with potentially relevant variables. RESULTS: The mean age of the sample set (n = 3,404) was 44.4 (SD, 16.0) years old and 64.5% female. Median MMSE was 28 (IQR, 28-29). The percentage of trial-aged participants (n = 1,267) with MMSE <24 was 18.6% overall and 54.3% among the subset with 0-4 YOE (n = 230). The five variables most associated with the MMSE in the study sample were education, age, exercise, C-reactive protein, and anxiety. CONCLUSION: The minimum MMSE cutoffs in most phase III prodromal-to-mild AD trials would exclude a significant proportion of trial-aged participants in this MA cohort, including over half of those with 0-4 YOE.

Local Therapy Decisional Regret in Older Women With Breast Cancer: A Population-Based Study.

PURPOSE: Older women with nonmetastatic breast cancer can often choose from several surgery and radiation treatment options. Little is known regarding how these choices contribute to decisional regret, which is a negative emotion reflecting the idea that another surgery or radiation decision might have been preferable. We sought to characterize the burden of and examine potential risk factors for local therapy decisional regret among a population-based cohort of older breast cancer survivors. METHODS AND MATERIALS: National Medicare claims for age ≥67 female breast cancer incident in 2009 identified patients treated with lumpectomy plus whole-breast irradiation, brachytherapy, or endocrine therapy or mastectomy with or without radiation. We sampled 330 patients per treatment group (N = 1650), of whom 1253 agreed to receive a paper survey including the Decisional Regret Scale and EQ-5D-3L Health-Utility Scale. Local therapy regret was defined as neutral or worse response to questions regarding surgery- or radiation-related decisional regret. Local therapy regret risk factors were evaluated using a multivariable generalized linear model. Association of local therapy regret with health utility was modeled using multivariable linear regression. RESULTS: The response rate was 30.2% (n = 498 of 1650); 421 surveys were included in this analysis. Median diagnosis age was 72 years, and surveys were completed 6 years after diagnosis. Overall, 23.8% of respondents (n = 100) reported experiencing local therapy decisional regret. Type of local therapy was not associated with local therapy regret. Predictors of increased regret included black race (risk ratio [RR], 2.09; 95% confidence interval [CI], 1.33-3.29), high school education or less (RR, 1.87; 95% CI, 1.27-2.75), and axillary nodal dissection (RR, 2.13; 95% CI, 1.33-3.41). Local therapy regret was not associated with health utility (P = .37). CONCLUSIONS: Local therapy regret afflicts nearly one quarter of our cohort of older breast cancer survivors, and it is associated with black race, less education, and more extensive nodal dissection, but not breast surgery. Regret is distinct from health utility, suggesting that it is a unique psychosocial construct that merits further study and mitigation strategies.

Long-term Patient-Reported Outcomes in Older Breast Cancer Survivors: A Population-Based Survey Study.

PURPOSE: For older women with breast cancer, local therapy options may include lumpectomy plus whole-breast irradiation (Lump + WBI), lumpectomy plus brachytherapy (Lump + Brachy), lumpectomy alone (Lump alone), mastectomy without radiation therapy (Mast alone), and mastectomy plus radiation therapy (Mast + RT). We surveyed a population-based cohort of older breast cancer survivors to assess the association of local therapy with long-term quality-of-life outcomes. METHODS AND MATERIALS: We used nationally comprehensive Medicare claims to identify women aged ≥67 years in whom nonmetastatic breast cancer was diagnosed in 2009, who were treated with 1 of the 5 aforementioned treatment options, and who were still alive in 2015. From this cohort, 1650 patients (330 patients per treatment) were randomly selected. A survey that included the CanSORT (Cancer Surveillance and Outcomes Research Team) Satisfaction with Breast Cosmetic Outcome, BREAST-Q, Decisional Regret Scale, and EQ-5D-3L was mailed to potential participants. We used multivariable linear regression to assess associations between local therapy and outcomes after adjusting for patient, disease, and treatment covariates. RESULTS: Among the 489 women who returned the surveys (30% response rate), the median age at diagnosis was 72 years (range, 67-87 years). The interval from diagnosis to survey completion was approximately 6 years for all patients. Compared with Lump + WBI (adjusted score, 3.40), the CanSORT cosmetic satisfaction scores were higher for Lump + Brachy (score, 3.77; P = .007) and Lump alone (score, 3.80; P = .04) and lower for Mast + RT (score, 3.01; P = .006). Similar trends were seen for BREAST-Q cosmetic satisfaction. BREAST-Q psychosocial, sexual, and physical well-being and EQ-5D-3L global health status tended to be better in patients treated with less irradiation and less surgery. BREAST-Q adverse radiation effects were worse for Lump + WBI compared with Lump + Brachy. Decisional regret regarding surgery and radiation therapy did not differ across groups. Compared with patients treated with Lump + WBI, patients treated with Lump + Brachy and Lump alone reported slightly higher rates of in-breast recurrence (excess risk of 5.8% and 6.4%, respectively; P = .01). CONCLUSIONS: In this nationally diverse cohort, less irradiation and less surgery were associated with better long-term quality-of-life outcomes. However, patient regret regarding surgery and radiation therapy was similar across all groups.