Estimating individual treatment effect on disability progression in multiple sclerosis using deep learning.

Disability progression in multiple sclerosis remains resistant to treatment. The absence of a suitable biomarker to allow for phase 2 clinical trials presents a high barrier for drug development. We propose to enable short proof-of-concept trials by increasing statistical power using a deep-learning predictive enrichment strategy. Specifically, a multi-headed multilayer perceptron is used to estimate the conditional average treatment effect (CATE) using baseline clinical and imaging features, and patients predicted to be most responsive are preferentially randomized into a trial. Leveraging data from six randomized clinical trials (n = 3,830), we first pre-trained the model on the subset of relapsing-remitting MS patients (n = 2,520), then fine-tuned it on a subset of primary progressive MS (PPMS) patients (n = 695). In a separate held-out test set of PPMS patients randomized to anti-CD20 antibodies or placebo (n = 297), the average treatment effect was larger for the 50% (HR, 0.492; 95% CI, 0.266-0.912; p = 0.0218) and 30% (HR, 0.361; 95% CI, 0.165-0.79; p = 0.008) predicted to be most responsive, compared to 0.743 (95% CI, 0.482-1.15; p = 0.179) for the entire group. The same model could also identify responders to laquinimod in another held-out test set of PPMS patients (n = 318). Finally, we show that using this model for predictive enrichment results in important increases in power.

Identification of a Prognostic Signature Based on the Expression of Genes Related to the Insulin Pathway in Early Breast Cancer.

INTRODUCTION: Insulin and the insulin-like growth factor (IGF) family play a key role in breast cancer (BC). OBJECTIVE: In this study, we evaluated on a genomic scale the potential prognostic value of insulin signaling in early BC. METHODS: Candidate genes were selected from the published literature and gene expression profiling experiments. Three publicly available BC datasets, containing gene expression data on 502 cases, were used to test the prognostic ability of the score. The gene signature was developed on GSE1456, containing microarray data from 159 patients, split into a training set (102 breast tumors) and a validation set (n = 57). GSE3494 and GSE2990 (350 patients) were used for external validation. Univariate Mann-Whitney test was used to identify genes differentially expressed between relapsed and nonrelapsed patients. Expression of genes significantly correlated with relapse was combined in a linear score. Patients were classified as low or high risk with respect to the median value. RESULTS: On the training set, 15 genes turned out to be differentially expressed: 8-year disease-free survival (DFS) was 51 and 91% in the high- and low-risk group (p < 0.001), respectively. In the validation set, DFS was 97 and 54% (p = 0.009), respectively. External validation: 8-year DFS was 72 and 61%, respectively, in GSE3494 (p = 0.03) and 74 and 55% in GSE2990 (p = 0.03). By multivariate analyses, the insulin signature was significantly associated with DFS, independently of age, hormone receptor status, nodal status, and grade. CONCLUSIONS: Our findings indicate that the insulin pathway is involved in BC prognosis at a genomic level and provide a window of selectivity for preventive and treatment strategies targeting the insulin/IGF pathway in BC patients.

Tailoring B cell depletion therapy in MS according to memory B cell monitoring.

OBJECTIVE: We wanted to evaluate efficacy on inflammatory parameters of rituximab (RTX)-personalized reinfusion scheme using a memory B cell-based treatment regimen. METHODS: This is a prospective, uncontrolled, open-label study including patients with MS treated with RTX in 2 Italian MS units. All patients were treated with RTX induction, followed by maintenance infusion at the dosage of 375 mg/m2, according to memory B cell repopulation (0.05% of peripheral-blood mononuclear cells [PBMCs] for the first 2 years, 0.1% of PBMC for the third year). MS activity was assessed as clinical or MRI activity. RESULTS: One hundred two patients were included in the analysis. Mean follow-up was 2.40 years (range 0.57-7.15 years). The annualized relapse rate (ARR) was 0.67 in the year before RTX start and decreased to 0.01 in the 3 years after RTX initiation (global ARR). The proportion of patient with MS activity (i.e., relapse or MRI activity) was 63.16% in the year before RTX start and decreased to 8.7% (0-6 months), 1.3% (6-12 months), 0% (12-24 months), and 0% (24-36 months). Annualized RTX infusion rates were 1.67 (95% confidence interval [CI]: 1.43-1.94), 0.76 (95% CI: 0.58-0.98), and 0.78 (95% CI: 0.52-1.12) for the first 3 years after RTX initiation, respectively. Patients were reinfused with a mean infusion interval of 367 days (range 181-839 days). CONCLUSION: The results of this study show that the memory B cell-based RTX reinfusion protocol is able to reduce the mean number of RTX reinfusions with persistent reduction of disease activity. CLASSIFICATION OF EVIDENCE: This study provides Class IV evidence that for patients with MS, a memory B cell-based RTX reinfusion protocol can reduce the mean number of RTX reinfusions with persistent reduction of disease activity.

Clinical trials of disease-modifying agents in pediatric MS: Opportunities, challenges, and recommendations from the IPMSSG.

OBJECTIVE: The impetus for this consensus discussion was to recommend clinical trial designs that can deliver high-quality data for effective therapies for pediatric patients, in a reasonable timeframe, with a key focus on short- and long-term safety. METHODS: The International Pediatric Multiple Sclerosis Study Group convened a meeting of experts to review the advances in the understanding of pediatric-onset multiple sclerosis (MS) and the advent of clinical trials for this population. RESULTS: In the last few years, convincing evidence has emerged that the biological processes involved in MS are largely shared across the age span. As such, treatments proven efficacious for the care of adults with MS have a biological rationale for use in pediatric MS given the relapsing-remitting course at onset and high relapse frequency. There are also ethical considerations on conducting clinical trials in this age group including the use of placebo owing to highly active disease. It is imperative to reconsider study design and implementation based on what information is needed. Are studies needed for efficacy or should safety be the primary goal? Further, there have been major recruitment challenges in recently completed and ongoing pediatric MS trials. Phase 3 trials for every newly approved therapy for adult MS in the pediatric MS population are simply not feasible. CONCLUSIONS: A primary goal is to ensure high-quality evidence-based treatment for children and adolescents with MS, which will improve our understanding of the safety of these agents and remove regulatory or insurance-based limitations in access to treatment.

Reliability of the Performance of Upper Limb assessment in Duchenne muscular dystrophy.

The Performance of Upper Limb was specifically designed to assess upper limb function in Duchenne muscular dystrophy. The aim of this study was to assess (1) a cohort of typically developing children from the age of 3years onwards in order to identify the age when the activities assessed in the individual items are consistently achieved, and (2) a cohort of 322 Duchenne children and young adults to establish the range of findings at different ages. We collected normative data for the scale validation on 277 typically developing subjects from 3 to 25years old. A full score was consistently achieved by the age of 5years. In the Duchenne cohort there was early involvement of the proximal muscles and a proximal to distal progressive involvement. The scale was capable of measuring small distal movements, related to activities of daily living, even in the oldest and weakest patients. Our data suggest that the assessment can be reliably used in both ambulant and non ambulant Duchenne patients in a multicentric setting and could therefore be considered as an outcome measure for future trials.

Quantitative muscle strength assessment in duchenne muscular dystrophy: longitudinal study and correlation with functional measures.

BACKGROUND: The aim of this study was to perform a longitudinal assessment using Quantitative Muscle Testing (QMT) in a cohort of ambulant boys affected by Duchenne muscular dystrophy (DMD) and to correlate the results of QMT with functional measures. This study is to date the most thorough long-term evaluation of QMT in a cohort of DMD patients correlated with other measures, such as the North Star Ambulatory Assessment (NSAA) or three 6-min walk test (6MWT). METHODS: This is a single centre, prospective, non-randomised, study assessing QMT using the Kin Com(®) 125 machine in a study cohort of 28 ambulant DMD boys, aged 5 to 12 years. This cohort was assessed longitudinally over a 12 months period of time with 3 monthly assessments for QMT and with assessment of functional abilities, using the NSAA and the 6MWT at baseline and at 12 months only. QMT was also used in a control group of 13 healthy age-matched boys examined at baseline and at 12 months. RESULTS: There was an increase in QMT over 12 months in boys below the age of 7.5 years while in boys above the age of 7.5 years, QMT showed a significant decrease. All the average one-year changes were significantly different than those experienced by healthy controls. We also found a good correlation between quantitative tests and the other measures that was more obvious in the stronger children. CONCLUSION: Our longitudinal data using QMT in a cohort of DMD patients suggest that this could be used as an additional tool to monitor changes, providing additional information on segmental strength.

Modelling the distribution of cortical lesions in multiple sclerosis.

Recent studies have shown the relevance of the cerebral grey matter involvement in multiple sclerosis (MS). Cortical lesions (CLs), detected by specific MRI sequences, are likely to become a new research outcome in MS studies. The aim of this study was to infer the optimal statistical model describing the distribution of CLs in patients with relapsing-remitting (RR)MS. The negative binomial model gave the best fit to the observed distribution of CLs in a group of 44 RRMS patients with one MRI scan of the brain. This observation has important implications for the statistical analysis of CLs in MS studies.

Duration of chemotherapy for metastatic breast cancer: a systematic review and meta-analysis of randomized clinical trials.

PURPOSE: To evaluate the effect of different first-line chemotherapy durations in patients with metastatic breast cancer (MBC) on overall survival (OS) and progression-free survival (PFS). METHODS: We searched literature databases to identify randomized controlled trials that compared different chemotherapy durations in the first-line treatment of MBC. Only trials with unconfounded comparisons of additional cycles of chemotherapy were included. The main outcome measures for this analysis were OS and PFS. Published data from retrieved studies were analyzed according to standard meta-analytic techniques. RESULTS: We found 11 randomized clinical trials including 2,269 patients. Longer first-line chemotherapy duration resulted into a significantly improved OS (hazard ratio [HR], 0.91; 95% CI, 0.84 to 0.99; P = .046) and PFS (HR, 0.64; 95% CI, 0.55 to 0.76; P < .001). There were no differences in effects on either OS or PFS between subgroups defined by time of random assignment, study design, number of chemotherapy cycles in the control arm or concomitant endocrine therapy. CONCLUSION: Longer first-line chemotherapy duration is associated with marginally longer OS and a substantially longer PFS.