External Validation of the Machine Learning-Based Thermographic Indices for Rheumatoid Arthritis: A Prospective Longitudinal Study.

External validation is crucial in developing reliable machine learning models. This study aimed to validate three novel indices-Thermographic Joint Inflammation Score (ThermoJIS), Thermographic Disease Activity Index (ThermoDAI), and Thermographic Disease Activity Index-C-reactive protein (ThermoDAI-CRP)-based on hand thermography and machine learning to assess joint inflammation and disease activity in rheumatoid arthritis (RA) patients. A 12-week prospective observational study was conducted with 77 RA patients recruited from rheumatology departments of three hospitals. During routine care visits, indices were obtained at baseline and week 12 visits using a pre-trained machine learning model. The performance of these indices was assessed cross-sectionally and longitudinally using correlation coefficients, the area under the receiver operating curve (AUROC), sensitivity, specificity, and positive and negative predictive values. ThermoDAI and ThermoDAI-CRP correlated with CDAI, SDAI, and DAS28-CRP cross-sectionally (ρ = 0.81; ρ = 0.83; ρ = 0.78) and longitudinally (ρ = 0.55; ρ = 0.61; ρ = 0.60), all p < 0.001. ThermoDAI and ThermoDAI-CRP also outperformed Patient Global Assessment (PGA) and PGA + C-reactive protein (CRP) in detecting changes in 28-swollen joint counts (SJC28). ThermoJIS had an AUROC of 0.67 (95% CI, 0.58 to 0.76) for detecting patients with swollen joints and effectively identified patients transitioning from SJC28 > 1 at baseline visit to SJC28 ≤ 1 at week 12 visit. These results support the effectiveness of ThermoJIS in assessing joint inflammation, as well as ThermoDAI and ThermoDAI-CRP in evaluating disease activity in RA patients.

Using Clinical Scales and Digital Measures to Explore Falls in Patients with Lewy Body Dementia.

Introduction: PRESENCE was a phase 2 clinical trial assessing the efficacy of mevidalen, a D1 receptor positive allosteric modulator, for symptomatic treatment of Lewy body dementia (LBD). Mevidalen demonstrated improvements in motor and non-motor features of LBD, global functioning, and actigraphy-measured activity and daytime sleep. Adverse events (AEs) of fall were numerically increased in mevidalen-treated participants. Methods: A subset of PRESENCE participants wore a wrist actigraphy device for 2-week periods pre-, during, and posttreatment. Actigraphy sleep and activity measures were derived per period and analyzed to assess for their association with participants' reports of an AE of fall. Prespecified baseline and treatment-emergent clinical characteristics were also included in the retrospective analysis of falls. Independent-samples t test and χ2 test were performed to compare the means and proportions between individuals with/without falls. Results: A trend toward more falls was observed with mevidalen treatment (31/258 mevidalen-treated vs. 4/86 in placebo-treated participants: p = 0.12). Higher body mass index (BMI) (p < 0.05), more severe disease measured by baseline Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) part II (p < 0.05), and a trend toward improved Alzheimer's Disease Assessment Scale-Cognitive Subscale 13 (ADAS-Cog13) (p = 0.06) were associated with individuals with falls. No statistically significant associations with falls and treatment-emergent changes were observed. Conclusion: The association of falls with worse baseline disease severity and higher BMI and overall trend toward improvements on cognitive and motor scales suggest that falls in PRESENCE may be related to increased activity in mevidalen-treated participants at greater risk for falling. Future studies to confirm this hypothesis using fall diaries and digital assessments are necessary.

Wrist-worn sensor-based measurements for drug effect detection with small samples in people with Lewy Body Dementia.

INTRODUCTION: Few late-stage clinical trials in Parkinson's disease (PD) have produced evidence on the clinical validity of sensor-based digital measurements of daily life activities to detect responses to treatment. The objective of this study was to assess whether digital measures from patients with mild-to-moderate Lewy Body Dementia demonstrate treatment effects during a randomized Phase 2 trial. METHODS: Substudy within a 12-week trial of mevidalen (placebo vs 10, 30, or 75 mg), where 70/344 patients (comparable to the overall population) wore a wrist-worn multi-sensor device. RESULTS: Treatment effects were statistically significant by conventional clinical assessments (Movement Disorder Society-Unified Parkinson's Disease Rating Scale [MDS-UPDRS] sum of Parts I-III and Alzheimer's Disease Cooperative Study-Clinical Global Impression of Change [ADCS-CGIC] scores) in the full study cohort at Week 12, but not in the substudy. However, digital measurements detected significant effects in the substudy cohort at week 6, persisting to week 12. CONCLUSIONS: Digital measurements detected treatment effects in a smaller cohort over a shorter period than conventional clinical assessments. TRIAL REGISTRATION: clinicaltrials.gov, NCT03305809.

Evaluating the Use of Digital Biomarkers to Test Treatment Effects on Cognition and Movement in Patients with Lewy Body Dementia.

BACKGROUND: PRESENCE was a Phase 2 trial assessing mevidalen for symptomatic treatment of Lewy body dementia (LBD). Participants received daily doses (10, 30, or 75 mg) of mevidalen (LY3154207) or placebo for 12 weeks. OBJECTIVE: To evaluate if frequent cognitive and motor tests using an iPad app and wrist-worn actigraphy to track activity and sleep could detect mevidalen treatment effects in LBD. METHODS: Of 340 participants enrolled in PRESENCE, 238 wore actigraphy for three 2-week periods: pre-, during, and post-intervention. A subset of participants (n = 160) enrolled in a sub-study using an iPad trial app with 3 tests: digital symbol substitution (DSST), spatial working memory (SWM), and finger-tapping. Compliance was defined as daily test completion or watch-wearing ≥23 h/day. Change from baseline to week 12 (app) or week 8 (actigraphy) was used to assess treatment effects using Mixed Model Repeated Measures analysis. Pearson correlations between sensor-derived features and clinical endpoints were assessed. RESULTS: Actigraphy and trial app compliance was > 90% and > 60%, respectively. At baseline, daytime sleep positively correlated with Epworth Sleepiness Scale score (p < 0.01). Physical activity correlated with improvement on Movement Disorder Society -Unified Parkinson Disease Rating Scale (MDS-UPDRS) part II (p < 0.001). Better scores of DSST and SWM correlated with lower Alzheimer Disease Assessment Scale -Cognitive 13-Item Scale (ADAS-Cog13) (p < 0.001). Mevidalen treatment (30 mg) improved SWM (p < 0.01), while dose-dependent decreases in daytime sleep (10 mg: p < 0.01, 30 mg: p < 0.05, 75 mg: p < 0.001), and an increase in walking minutes (75 mg dose: p < 0.001) were observed, returning to baseline post-intervention. CONCLUSION: Devices used in the LBD population achieved adequate compliance and digital metrics detected statistically significant treatment effects.

Restricted mean survival time for the analysis of cardiovascular outcome trials assessing non-inferiority: Case studies from antihyperglycemic drug development.

Cardiovascular outcome trials (CVOTs) have been employed in multiple therapeutic areas to explore whether a noncardiovascular drug increases the risk for cardiovascular events. These studies are now a central part of drug development programs for antihyperglycemic drugs. These programs are expected to demonstrate that new antihyperglycemic drugs for patients with Type 2 diabetes do not have unacceptable cardiovascular risk. The hazard ratio, which is usually provided as evidence that patients receiving the investigational treatment are not at statistically significantly greater cardiovascular risk than patients on the control treatment, can be difficult to interpret for various reasons. Therefore, an alternative approach known as the Restricted Mean Survival Time (RMST) or τ-year mean survival time is presented, and its ability to overcome interpretation challenges with the hazard ratio discussed. The RMST approach is applied to five completed CVOTs and is compared with the corresponding hazard ratios. Additionally, detailed considerations are given on how to design a non-inferiority CVOT using the RMST approach. The RMST methodology is shown to be a practical alternative to the hazard ratio methodology for designing a non-inferiority CVOT.

Coding variants in PNPLA3 and TM6SF2 are risk factors for hepatic steatosis and elevated serum alanine aminotransferases caused by a glucagon receptor antagonist.

LY2409021 is a glucagon receptor antagonist that was associated with hepatic steatosis and elevated aminotransferases in phase 2 diabetes studies. We investigated the relationship between selected genetic variants and hepatic steatosis and elevated alanine aminotransferases (ALTs) associated with LY2409021. Patients participated in a 6-week placebo-controlled trial (I1R-MC-GLDI [GLDI], n = 246) and a 52-week placebo- and active comparator-controlled trial (I1R-MC-GLDJ [GLDJ], n = 158). GLDJ had endpoints at 6 months, including measures of hepatic fat fraction (HFF) by magnetic resonance imaging. The five genes tested were patatin-like phospholipase domain containing 3 (PNPLA3) (rs738409 and rs738491), transmembrane 6 superfamily member 2 (TM6SF2) (rs58542926), peroxisome proliferative activated receptor gamma coactivator 1 alpha (PPARGC1A) (rs4361373, rs3774921, rs2970849), adenylate cyclase 3 (ADCY3) (rs713586), and insulin-like growth factor 1 (IGF-1) (rs1520220). In GLDI, PNPLA3 I148M (P = 0.001) and TM6SF2 E167K (P = 0.001) were significantly associated with an increase in ALT at 6 weeks for LY2409021 but not for placebo. In GLDJ, PNPLA3 I148M showed the same effect (P = 0.007) on ALT at 6 months but the placebo or sitagliptin did not. In GLDJ, both PNPLA3 and TM6SF2 risk-allele carriers showed increases in HFF that were numerically greater but not statistically significant. The carriers of PNPLA3 and/or TM6SF2 risk alleles showed significantly increased ALT (GLDI, +13.28 U/L in carriers versus +4.84 U/L in noncarriers, P = 4 × 10-5; GLDJ, +14.6 U/L in carriers versus +1.7 in noncarriers, P = 0.0018) and HFF (GLDJ, +5.35% in carriers versus 2.38% in noncarriers, P = 0.048). Elevation of transaminase and HFF were also noted in the noncarriers but at a significantly lower degree. Conclusion: The carriers of PNPLA3 and/or TM6SF2 variant alleles are at risk for hepatic steatosis and elevated ALT levels caused by LY2409021, a glucagon receptor antagonist. More studies are needed to investigate if our observations are generalizable to hepatic steatosis caused by other medications. (Hepatology Communications 2018;2:561-570).

Post hoc subgroup analysis of the HEART2D trial demonstrates lower cardiovascular risk in older patients targeting postprandial versus fasting/premeal glycemia.

OBJECTIVE: To identify the Hyperglycemia and Its Effect After Acute Myocardial Infarction on Cardiovascular Outcomes in Patients With Type 2 Diabetes Mellitus (HEART2D) trial subgroups with treatment difference. RESEARCH DESIGN AND METHODS: In 1,115 type 2 diabetic patients who had suffered from an acute myocardial infarction (AMI), the HEART2D trial compared two insulin strategies targeting postprandial or fasting/premeal glycemia on time until first cardiovascular event (cardiovascular death, nonfatal MI, nonfatal stroke, coronary revascularization, or hospitalization for acute coronary syndrome). The HEART2D trial ended prematurely for futility. We used the classification and regression tree (CART) to identify baseline subgroups with potential treatment differences. RESULTS: CART estimated the age of >65.7 years to best predict the difference in time to first event. In the subgroup aged>65.7 years (prandial, n=189; basal, n=210), prandial patients had a significantly longer time to first event and a lower proportion experienced a first event (n=56 [29.6%] vs. n=85 [40.5%]; hazard ratio 0.69 [95% CI 0.49-0.96]; P=0.029), despite similar A1C levels. CONCLUSIONS: Older type 2 diabetic AMI survivors may have a lower risk for a subsequent cardiovascular event with insulin targeting postprandial versus fasting/premeal glycemia.