A web-based tool for personalized prediction of long-term disease course in patients with multiple sclerosis

Prognostic factors for disease activity in newly diagnosed teriflunomide-treated patients with multiple sclerosis: a nationwide Danish study

BACKGROUND

Clinicians frequently rely on relapse counts, T2 MRI lesion load (T2L) and Expanded Disability Status Scale (EDSS) scores to guide treatment decisions for individuals diagnosed with multiple sclerosis (MS). This study evaluates how these factors, along with age and sex, influence prognosis during treatment with teriflunomide (TFL).

METHODS

We conducted a nationwide cohort study using data from the Danish Multiple Sclerosis Registry.Eligible participants had relapsing-remitting MS or clinically isolated syndrome and initiated TFL as their first treatment between 2013 and 2019. The effect of age, pretreatment relapses, T2L and EDSS scores on the risk of disease activity on TFL were stratified by sex.

RESULTS

In total, 784 individuals were included (57.4% females). A high number of pretreatment relapses (≥2) was associated with an increased risk of disease activity in females only (OR and (95% CI): 1.76 (1.11 to 2.81)). Age group 50+ was associated with a lower risk of disease activity in both sexes (OR females=0.28 (0.14 to 0.56); OR males=0.22 (0.09 to 0.55)), while age 35-49 showed a different impact in males and females (OR females=0.79 (0.50 to 1.23); OR males=0.42 (0.24 to 0.72)). EDSS scores and T2L did not show any consistent associations.

CONCLUSION

A high number of pretreatment relapses was only associated with an increased risk of disease activity in females, while age had a differential impact on the risk of disease activity according to sex. Clinicians may consider age, sex and relapses when deciding on TFL treatment.

Discussing the potential for progression with patients newly diagnosed with multiple sclerosis: When, how, and why?

Despite convergent evidence that upwards of 50% of patients with MS transition from a relapsing to progressive phase within 20 years of disease onset, and the recent acknowledgement of the commonality of progression independent of relapses, there remains no consensus regarding the nature and timing of a discussion about the possibility of a secondary progressive phase with relapsing-remitting MS patients. Some neurologists prefer to conduct this at the inaugural visit to provide more information about disease behaviour and potential planning that might entail, while others may defer any discussion about this phase, as there is no clear consensus for it and it can be a sensitive topic, with concern that too early a discussion could worsen anxiety and discourage or delay decisions regarding disease modifying treatments. Furthermore, it is unknown at onset which patients will transition to a progressive phenotype. This review and opinion paper will outline some of the opportunities and challenges associated with such a disclosure, and attempt to provide a balanced, patient-centred approach to address this delicate topic.

Development of Registry Data to Create Interactive Doctor-Patient Platforms for Personalized Patient Care, Taking the Example of the DESTINY System

“Real-world evidence (RWE)” is becoming increasingly important in order to integrate the results of randomized studies into everyday clinical practice. The data collection of RWE is usually derived from large-scale national and international registries, often driven by academic centers. We have developed a digitalized doctor–patient platform called DESTINY (DatabasE-assiSted Therapy decIsioN support sYstem) that is utilized by NeuroTransData (NTD), a network of neurologists and psychiatrists throughout Germany. This platform can be integrated into everyday practice and, as well as being used for scientific evaluations in healthcare research, can also serve as an individual, personalized treatment application. Its various modules allow for a timely identification of side-effects or interactions of treatments, can involve patients via the “My NTC Health Guide” portal, and can collect data of individual disease histories that are integrated into innovative algorithms, e.g., for the prediction of treatment response [currently available for multiple sclerosis (MS), with other indications in the pipeline]. Here, we describe the doctor–patient platform DESTINY for outpatient neurological practices and its contribution to improved treatment success as well as reduction of healthcare costs. Platforms like DESTINY may facilitate the goal of personalized healthcare.

Machine Learning Use for Prognostic Purposes in Multiple Sclerosis

The course of multiple sclerosis begins with a relapsing-remitting phase, which evolves into a secondarily progressive form over an extremely variable period, depending on many factors, each with a subtle influence. To date, no prognostic factors or risk score have been validated to predict disease course in single individuals. This is increasingly frustrating, since several treatments can prevent relapses and slow progression, even for a long time, although the possible adverse effects are relevant, in particular for the more effective drugs. An early prediction of disease course would allow differentiation of the treatment based on the expected aggressiveness of the disease, reserving high-impact therapies for patients at greater risk. To increase prognostic capacity, approaches based on machine learning (ML) algorithms are being attempted, given the failure of other approaches. Here we review recent studies that have used clinical data, alone or with other types of data, to derive prognostic models. Several algorithms that have been used and compared are described. Although no study has proposed a clinically usable model, knowledge is building up and in the future strong tools are likely to emerge.

Long-term prognostic counselling in people with multiple sclerosis using an online analytical processing tool

BACKGROUND

Prognostic counselling is a sensitive issue in medicine and especially so in MS due to the highly heterogeneous disease course. However, people with MS (pwMS) seek prognostic information. The web-based 'Evidence-Based Decision Support Tool in Multiple Sclerosis' (EBDiMS) uses data of 717 patients from the London/Ontario cohort to calculate personalized long-term prognostic information.

OBJECTIVE

The aim of this study was to investigate the feasibility and effect of long-term prognostic counselling in pwMS using EBDiMS.

METHODS

Ninety consecutive pwMS were provided with personalized estimations of expected time to reach Expanded Disability Status Scale (EDSS) scores of 6 and 8 and time to conversion to secondary-progressive MS. Participants gave estimates on their own putative prognosis and rated the tool's acceptability on six-step Likert-type scales.

RESULTS

Participants rated EBDiMS as highly understandable, interesting and relevant for patient-physician encounters, coping and therapy decisions. Although it provoked a certain degree of worry in some participants, 95% would recommend using the tool. Participants' own prognosis estimates did not change significantly following EBDiMS.

CONCLUSION

Long-term prognostic counselling using an online tool has been shown to be feasible in a clinical setting. EBDiMS provides pwMS with relevant, easy-to-understand, long-term prognostic information without causing relevant anxiety.

Considering patient clinical history impacts performance of machine learning models in predicting course of multiple sclerosis

Multiple Sclerosis (MS) progresses at an unpredictable rate, but predictions on the disease course in each patient would be extremely useful to tailor therapy to the individual needs. We explore different machine learning (ML) approaches to predict whether a patient will shift from the initial Relapsing-Remitting (RR) to the Secondary Progressive (SP) form of the disease, using only "real world" data available in clinical routine. The clinical records of 1624 outpatients (207 in the SP phase) attending the MS service of Sant'Andrea hospital, Rome, Italy, were used. Predictions at 180, 360 or 720 days from the last visit were obtained considering either the data of the last available visit (Visit-Oriented setting), comparing four classical ML methods (Random Forest, Support Vector Machine, K-Nearest Neighbours and AdaBoost) or the whole clinical history of each patient (History-Oriented setting), using a Recurrent Neural Network model, specifically designed for historical data. Missing values were handled by removing either all clinical records presenting at least one missing parameter (Feature-saving approach) or the 3 clinical parameters which contained missing values (Record-saving approach). The performances of the classifiers were rated using common indicators, such as Recall (or Sensitivity) and Precision (or Positive predictive value). In the visit-oriented setting, the Record-saving approach yielded Recall values from 70% to 100%, but low Precision (5% to 10%), which however increased to 50% when considering only predictions for which the model returned a probability above a given "confidence threshold". For the History-oriented setting, both indicators increased as prediction time lengthened, reaching values of 67% (Recall) and 42% (Precision) at 720 days. We show how "real world" data can be effectively used to forecast the evolution of MS, leading to high Recall values and propose innovative approaches to improve Precision towards clinically useful values.

Reaching an evidence-based prognosis for personalized treatment of multiple sclerosis

Personalized treatment is ideal for multiple sclerosis (MS) owing to the heterogeneity of clinical features, but current knowledge gaps, including validation of biomarkers and treatment algorithms, limit practical implementation. The contemporary approach to personalized MS therapy depends on evidence-based prognostication, an initial treatment choice and evaluation of early treatment responses to identify the need to switch therapy. Prognostication is directed by baseline clinical, environmental and demographic factors, MRI measures and biomarkers that correlate with long-term disability measures. The initial treatment choice should be a shared decision between the patient and physician. In addition to prognosis, this choice must account for patient-related factors, including comorbidities, pregnancy planning, preferences of the patients and their comfort with risk, and drug-related factors, including safety, cost and implications for treatment sequencing. Treatment response has traditionally been assessed on the basis of relapse rate, MRI lesions and disability progression. Larger longitudinal data sets have enabled development of composite outcome measures and more stringent standards for disease control. Biomarkers, including neurofilament light chain, have potential as early surrogate markers of prognosis and treatment response but require further validation. Overall, attainment of personalized treatment for MS is complex but will be refined as new data become available.

Intelligent Computer Systems for Multiple Sclerosis Diagnosis: a Systematic Review of Reasoning Techniques and Methods

OBJECTIVE

Intelligent computer systems are used in diagnosing Multiple Sclerosis and help physicians in the accurate and timely diagnosis of the disease. This study focuses on a review of different reasoning techniques and methods used in intelligent systems to diagnose MS and analyze the application and efficiency of different reasoning methods in order to find the most efficient and applicable methods and techniques for MS diagnosis.

METHODS

A complete research was carried out on articles in various electronic databases based on Mesh vocabulary. 85 articles out of 614 articles published in English between 2000 to 2018 were analyzed, 30 of which have been selected based on inclusion criteria such as system scope and domain, full description of reasoning method and system evaluation.

RESULTS

Results indicate that different reasoning methods are used unintelligent systems of MS diagnosis. In 27% of the studies, the rule-based method was used, in 20% the fuzzy logic method, in 18%the artificial neural network method, and in 35% other reasoning methods were used. The average sensitivity, specificity and accuracy of reasoning methods were0.91, 0.77, and 0.86, respectively.

CONCLUSIONS

Rule-based, fuzzy-logic and artificial neural network methods have had more applications in intelligent systems for the diagnosis of MS, respectively. The highest rate of sensitivity and accuracy indexes is associated to the neural network reasoning method at 0.97 and 0.99, respectively .In the fuzzy logic method, the Kappa rate has been reported as one, which shows full conformity between software diagnosis and the physician's decision .In some articles, in order to remove the limitations of the methods and enhance their efficiency, combinations of different methods are used.

Do people with multiple sclerosis want to know their prognosis? A UK nationwide study

Background

Multiple sclerosis (MS) has a varied and uncertain trajectory. The recent development of analytical processing tools that draw on large longitudinal patient databases facilitates personalised long-term prognosis estimates. This has the potential to improve both shared treatment decision-making and psychological adjustment. However, there is limited research on how people with MS feel about prognosis communication and forecasting. This study investigated the prognosis communication experiences and preferences of people with MS and explored whether clinical, demographic and psychological factors are associated with prognosis information preferences.

Methods

3175 UK MS Register members (59% of those with active accounts) completed an online survey containing 17 questions about prognosis communication experiences, attitudes and preferences. Participants also completed validated questionnaires measuring coping strategies, tendencies to seek out (‘monitor’) or avoid (‘blunt’) information in threatening situations, and MS risk perceptions and reported their clinical and sociodemographic characteristics. Data already held on the MS Register about participants’ quality of life, anxiety and depression symptoms and MS impact were obtained and linked to the survey data.

Results

53.1% of participants had never discussed long-term prognosis with healthcare professionals. 54.2% lacked clarity about their long-term prognosis. 76% had strong preferences for receiving long-term prognosis information. 92.8% were interested in using tools that generate personalised predictions. Most participants considered prognostication useful for decision-making. Participants were more receptive to receiving prognosis information at later time-points, versus at diagnosis. A comprehensive set of sociodemographic, clinical and psychological variables predicted only 7.9% variance in prognosis information preferences.

Conclusions

People with MS have an appetite for individualised long-term prognosis forecasting and their need for information is frequently unmet. Clinical studies deploying and evaluating interventions to support prognostication in MS are now needed. This study indicates suitable contexts and patient preferences for initial trials of long-term prognosis tools in clinical settings.

Personalization and Patient Involvement in Decision Support Systems: Current Trends

OBJECTIVES

This survey aims at highlighting the latest trends (2012-2014) on the development, use, and evaluation of Information and Communication Technologies (ICT) based decision support systems (DSSs) in medicine, with a particular focus on patient-centered and personalized care.

METHODS

We considered papers published on scientific journals, by querying PubMed and Web of ScienceTM. Included studies focused on the implementation or evaluation of ICT-based tools used in clinical practice. A separate search was performed on computerized physician order entry systems (CPOEs), since they are increasingly embedding patient-tailored decision support.

RESULTS

We found 73 papers on DSSs (53 on specific ICT tools) and 72 papers on CPOEs. Although decision support through the delivery of recommendations is frequent (28/53 papers), our review highlighted also DSSs only based on efficient information presentation (25/53). Patient participation in making decisions is still limited (9/53), and mostly focused on risk communication. The most represented medical area is cancer (12%). Policy makers are beginning to be included among stakeholders (6/73), but integration with hospital information systems is still low. Concerning knowledge representation/management issues, we identified a trend towards building inference engines on top of standard data models. Most of the tools (57%) underwent a formal assessment study, even if half of them aimed at evaluating usability and not effectiveness.

CONCLUSIONS

Overall, we have noticed interesting evolutions of medical DSSs to improve communication with the patient, consider the economic and organizational impact, and use standard models for knowledge representation. However, systems focusing on patient-centered care still do not seem to be available at large.

Exploration of machine learning techniques in predicting multiple sclerosis disease course

Objective

To explore the value of machine learning methods for predicting multiple sclerosis disease course.

Methods

1693 CLIMB study patients were classified as increased EDSS≥1.5 (worsening) or not (non-worsening) at up to five years after baseline visit. Support vector machines (SVM) were used to build the classifier, and compared to logistic regression (LR) using demographic, clinical and MRI data obtained at years one and two to predict EDSS at five years follow-up.

Results

Baseline data alone provided little predictive value. Clinical observation for one year improved overall SVM sensitivity to 62% and specificity to 65% in predicting worsening cases. The addition of one year MRI data improved sensitivity to 71% and specificity to 68%. Use of non-uniform misclassification costs in the SVM model, weighting towards increased sensitivity, improved predictions (up to 86%). Sensitivity, specificity, and overall accuracy improved minimally with additional follow-up data. Predictions improved within specific groups defined by baseline EDSS. LR performed more poorly than SVM in most cases. Race, family history of MS, and brain parenchymal fraction, ranked highly as predictors of the non-worsening group. Brain T2 lesion volume ranked highly as predictive of the worsening group.

Interpretation

SVM incorporating short-term clinical and brain MRI data, class imbalance corrective measures, and classification costs may be a promising means to predict MS disease course, and for selection of patients suitable for more aggressive treatment regimens.

How Do People with Multiple Sclerosis Experience Prognostic Uncertainty and Prognosis Communication? A Qualitative Study

Background

Disease progression in multiple sclerosis (MS) is highly variable and predicting prognosis is notoriously challenging. Patients’ prognosis beliefs, responses to prognostic uncertainty and experiences of prognosis-related communication with healthcare professionals (HCPs) have received little study. These issues have implications for patients’ psychological adjustment and are important in the context of the recent development of personalised prognosis forecasting tools. This study explored patient perspectives on the experience of prognostic uncertainty, the formation of expectations about personal prognosis and the nature of received and desired prognosis communication.

Methods

15 MS patients participated in in-depth semi-structured interviews which were analysed using inductive thematic analysis.

Results

Six themes captured key aspects of the data: Experiencing unsatisfactory communication with HCPs, Appreciating and accepting prognostic uncertainty, Trying to stay present-focused, Forming and editing personal prognosis beliefs, Ambivalence towards forecasting the future, and Prognosis information delivery. MS patients report having minimal communication with HCPs about prognosis. Over time MS patients appear to develop expectations about their disease trajectories, but do so with minimal HCP input. Provision of prognosis information by HCPs seems to run counter to patients’ attempts to remain present-focused. Patients are often ambivalent about prognosis forecasting and consider it emotionally dangerous and of circumscribed usefulness.

Conclusions

HCPs must carefully consider whether, when and how to share prognosis information with patients; specific training may be beneficial. Future research should confirm findings about limited HCP-patient communication, distinguish predictors of patients’ attitudes towards prognostication and identify circumstances under which prognostic forecasting benefits patients.

A web-based decision support tool for prognosis simulation in multiple sclerosis

A multiplicity of natural history studies of multiple sclerosis provides valuable knowledge of the disease progression but individualized prognosis remains elusive. A few decision support tools that assist the clinician in such task have emerged but have not received proper attention from clinicians and patients. The objective of the current work is to implement a web-based tool, conveying decision relevant prognostic scientific evidence, which will help clinicians discuss prognosis with individual patients. Data were extracted from a set of reference studies, especially those dealing with the natural history of multiple sclerosis. The web-based decision support tool for individualized prognosis simulation was implemented with NetLogo, a program environment suited for the development of complex adaptive systems. Its prototype has been launched online; it enables clinicians to predict both the likelihood of CIS to CDMS conversion, and the long-term prognosis of disability level and SPMS conversion, as well as assess and monitor the effects of treatment. More robust decision support tools, which convey scientific evidence and satisfy the needs of clinical practice by helping clinicians discuss prognosis expectations with individual patients, are required. The web-based simulation model herein introduced proposes to be a step forward toward this purpose.

Continuous prediction of secondary progression in the individual course of multiple sclerosis

BACKGROUND

Prediction of the course of multiple sclerosis (MS) was traditionally based on features close to onset.

OBJECTIVE

To evaluate predictors of the individual risk of secondary progression (SP) identified at any time during relapsing-remitting MS.

METHODS

We analysed a database comprising an untreated MS incidence cohort (n=306) with five decades of follow-up. Data regarding predictors of all attacks (n=749) and demographics from patients (n=157) with at least one distinct second attack were included as covariates in a Poisson regression analysis with SP as outcome.

RESULTS

The average hazard function of transition to SPMS was 0.046 events per patient year, showing a maximum at age 33. Three covariates were significant predictors: age, a descriptor of the most recent relapse, and the interaction between the descriptor and time since the relapse. A hazard function termed "prediction score" estimated the risk of SP as number of transition events per patient year (range <0.01 to >0.15).

CONCLUSIONS

The insights gained from this study are that the risk of transition to SP varies over time in individual patients, that the risk of SP is linked to previous relapses, that predictors in the later stages of the course are more effective than the traditional onset predictors, and that the number of potential predictors can be reduced to a few (three in this study) essential items. This advanced simplification facilitates adaption of the "prediction score" to other (more recent, benign or treated) materials, and allows for compact web-based applications (http://msprediction.com).