Prediction of overall survival for patients with metastatic castration-resistant prostate cancer: development of a prognostic model through a crowdsourced challenge with open clinical trial data.

BACKGROUND: Improvements to prognostic models in metastatic castration-resistant prostate cancer have the potential to augment clinical trial design and guide treatment strategies. In partnership with Project Data Sphere, a not-for-profit initiative allowing data from cancer clinical trials to be shared broadly with researchers, we designed an open-data, crowdsourced, DREAM (Dialogue for Reverse Engineering Assessments and Methods) challenge to not only identify a better prognostic model for prediction of survival in patients with metastatic castration-resistant prostate cancer but also engage a community of international data scientists to study this disease. METHODS: Data from the comparator arms of four phase 3 clinical trials in first-line metastatic castration-resistant prostate cancer were obtained from Project Data Sphere, comprising 476 patients treated with docetaxel and prednisone from the ASCENT2 trial, 526 patients treated with docetaxel, prednisone, and placebo in the MAINSAIL trial, 598 patients treated with docetaxel, prednisone or prednisolone, and placebo in the VENICE trial, and 470 patients treated with docetaxel and placebo in the ENTHUSE 33 trial. Datasets consisting of more than 150 clinical variables were curated centrally, including demographics, laboratory values, medical history, lesion sites, and previous treatments. Data from ASCENT2, MAINSAIL, and VENICE were released publicly to be used as training data to predict the outcome of interest-namely, overall survival. Clinical data were also released for ENTHUSE 33, but data for outcome variables (overall survival and event status) were hidden from the challenge participants so that ENTHUSE 33 could be used for independent validation. Methods were evaluated using the integrated time-dependent area under the curve (iAUC). The reference model, based on eight clinical variables and a penalised Cox proportional-hazards model, was used to compare method performance. Further validation was done using data from a fifth trial-ENTHUSE M1-in which 266 patients with metastatic castration-resistant prostate cancer were treated with placebo alone. FINDINGS: 50 independent methods were developed to predict overall survival and were evaluated through the DREAM challenge. The top performer was based on an ensemble of penalised Cox regression models (ePCR), which uniquely identified predictive interaction effects with immune biomarkers and markers of hepatic and renal function. Overall, ePCR outperformed all other methods (iAUC 0·791; Bayes factor >5) and surpassed the reference model (iAUC 0·743; Bayes factor >20). Both the ePCR model and reference models stratified patients in the ENTHUSE 33 trial into high-risk and low-risk groups with significantly different overall survival (ePCR: hazard ratio 3·32, 95% CI 2·39-4·62, p<0·0001; reference model: 2·56, 1·85-3·53, p<0·0001). The new model was validated further on the ENTHUSE M1 cohort with similarly high performance (iAUC 0·768). Meta-analysis across all methods confirmed previously identified predictive clinical variables and revealed aspartate aminotransferase as an important, albeit previously under-reported, prognostic biomarker. INTERPRETATION: Novel prognostic factors were delineated, and the assessment of 50 methods developed by independent international teams establishes a benchmark for development of methods in the future. The results of this effort show that data-sharing, when combined with a crowdsourced challenge, is a robust and powerful framework to develop new prognostic models in advanced prostate cancer. FUNDING: Sanofi US Services, Project Data Sphere.

Molecular markers identify subtypes of stage III colon cancer associated with patient outcomes.

BACKGROUND & AIMS: Categorization of colon cancers into distinct subtypes using a combination of pathway-based biomarkers could provide insight into stage-independent variability in outcomes. METHODS: We used a polymerase chain reaction-based assay to detect mutations in BRAF (V600E) and in KRAS in 2720 stage III cancer samples, collected prospectively from patients participating in an adjuvant chemotherapy trial (NCCTG N0147). Tumors deficient or proficient in DNA mismatch repair (MMR) were identified based on detection of MLH1, MSH2, and MSH6 proteins and methylation of the MLH1 promoter. Findings were validated using tumor samples from a separate set of patients with stage III cancer (n = 783). Association with 5-year disease-free survival was evaluated using Cox proportional hazards models. RESULTS: Tumors were categorized into 5 subtypes based on MMR status and detection of BRAF or KRAS mutations which were mutually exclusive. Three subtypes were MMR proficient: those with mutations in BRAF (6.9% of samples), mutations in KRAS (35%), or tumors lacking either BRAF or KRAS mutations (49%). Two subtypes were MMR deficient: the sporadic type (6.8%) with BRAF mutation and/or or hypermethylation of MLH1 and the familial type (2.6%), which lacked BRAF(V600E) or hypermethylation of MLH1. A higher percentage of MMR-proficient tumors with BRAF(V600E) were proximal (76%), high-grade (44%), N2 stage (59%), and detected in women (59%), compared with MMR-proficient tumors without BRAF(V600E) or KRAS mutations (33%, 19%, 41%, and 42%, respectively; all P < .0001). A significantly lower proportion of patients with MMR-proficient tumors with mutant BRAF (hazard ratio = 1.43; 95% confidence interval: 1.11-1.85; Padjusted = .0065) or mutant KRAS (hazard ratio = 1.48; 95% confidence interval: 1.27-1.74; Padjusted < .0001) survived disease-free for 5 years compared with patients whose MMR-proficient tumors lacked mutations in either gene. Disease-free survival rates of patients with MMR-deficient sporadic or familial subtypes was similar to those of patients with MMR-proficient tumors without BRAF or KRAS mutations. The observed differences in survival rates of patients with different tumor subtypes were validated in an independent cohort. CONCLUSIONS: We identified subtypes of stage III colon cancer, based on detection of mutations in BRAF (V600E) or KRAS, and MMR status that show differences in clinical and pathologic features and disease-free survival. Patients with MMR-proficient tumors and BRAF or KRAS mutations had statistically shorter survival times than patients whose tumors lacked these mutations. The tumor subtype found in nearly half of the study cohort (MMR-proficient without BRAF(V600E) or KRAS mutations) had similar outcomes to those of patients with MMR-deficient cancers.

Summary of the DREAM8 Parameter Estimation Challenge: Toward Parameter Identification for Whole-Cell Models.

Whole-cell models that explicitly represent all cellular components at the molecular level have the potential to predict phenotype from genotype. However, even for simple bacteria, whole-cell models will contain thousands of parameters, many of which are poorly characterized or unknown. New algorithms are needed to estimate these parameters and enable researchers to build increasingly comprehensive models. We organized the Dialogue for Reverse Engineering Assessments and Methods (DREAM) 8 Whole-Cell Parameter Estimation Challenge to develop new parameter estimation algorithms for whole-cell models. We asked participants to identify a subset of parameters of a whole-cell model given the model's structure and in silico "experimental" data. Here we describe the challenge, the best performing methods, and new insights into the identifiability of whole-cell models. We also describe several valuable lessons we learned toward improving future challenges. Going forward, we believe that collaborative efforts supported by inexpensive cloud computing have the potential to solve whole-cell model parameter estimation.

Somatic expression of ENRAGE is associated with obesity status among patients with clear cell renal cell carcinoma.

An association between obesity and development of clear cell renal cell carcinoma (ccRCC) has been established in the literature; however, there are limited data regarding the molecular mechanisms that underlie this association. Therefore, we used a multistage design to identify and validate genes that are associated with obesity-related ccRCC. We conducted a microarray study and compared gene expression between obese and non-obese subjects in ccRCC tumors and patient-matched normal kidney tissues. Analyses were stratified by smoking status and subsequently performed on the combined cohort. The primary objective was to identify genes where the fold change of ccRCC tumor expression between obese and non-obese subjects was different than the fold change in the patient-matched normal kidney tissue. Thus, we utilized a mixed model and evaluated the tissue type-by-obesity status interaction term. Targeted validation was performed using reverse transcription-polymerase chain reaction (RT-PCR) on an independent cohort. ENRAGE was identified in the microarray study and subsequently validated using RT-PCR to have a statistically significant tissue type-by-obesity status interaction. Specifically, although ENRAGE is similarly expressed across obese and non-obese subjects in normal tissue, it is upregulated in the patient-matched ccRCC tumor tissue. Additionally, ENRAGE is upregulated in tumors that are wild-type for the von Hippel Lindau gene and in tumors for subjects with poorer overall survival. In summary, we provide evidence that overexpression of ENRAGE in ccRCC tumor tissue is an obesity-associated somatic alteration. Upregulation of ENRAGE could lead to local, autocrine stimulation of the RAGE receptor and thus support cancer progression.

ANKS1B is a smoking-related molecular alteration in clear cell renal cell carcinoma.

BACKGROUND: An association between cigarette smoking and increased risk of clear cell renal cell carcinoma (ccRCC) has been established; however, there are limited data regarding the molecular mechanisms that underlie this association. We used a multi-stage design to identify and validate genes that are associated with smoking-related ccRCC. METHODS: We first conducted a microarray study to compare gene expression patterns in patient-matched ccRCC and normal kidney tissues between patients with (n = 23) and without (n = 42) a history of smoking. Analyses were first stratified on obesity status (the other primary risk factor for ccRCC) and then combined and analyzed together. To identify genes where the fold change in smokers relative to non-smokers was different in tumor tissues in comparison to patient-matched normal kidney tissues, we identified Affymetrix probesets that had a significant tissue type-by-smoking status interaction pvalue. We then performed RT-PCR validation on the top eight candidate genes in an independent sample of 28 smokers and 54 non-smokers. RESULTS: We identified 15 probesets that mapped to eight genes that had candidate associations with smoking-related ccRCC: ANKS1B, ACOT6, PPWD1, EYS, LIMCH1, CHRNA6, MT1G, and ZNF600. Using RT-PCR, we validated that expression of ANKS1B is preferentially down-regulated in smoking-related ccRCC. CONCLUSION: We provide the first evidence that ANKS1B expression is down regulated in ccRCC tumors relative to patient-matched normal kidney tissue in smokers. Thus, ANKS1B should be explored further as a novel avenue for early detection as well as prevention of ccRCC in smokers.

Technical and biological variance structure in mRNA-Seq data: life in the real world.

BACKGROUND: mRNA expression data from next generation sequencing platforms is obtained in the form of counts per gene or exon. Counts have classically been assumed to follow a Poisson distribution in which the variance is equal to the mean. The Negative Binomial distribution which allows for over-dispersion, i.e., for the variance to be greater than the mean, is commonly used to model count data as well. RESULTS: In mRNA-Seq data from 25 subjects, we found technical variation to generally follow a Poisson distribution as has been reported previously and biological variability was over-dispersed relative to the Poisson model. The mean-variance relationship across all genes was quadratic, in keeping with a Negative Binomial (NB) distribution. Over-dispersed Poisson and NB distributional assumptions demonstrated marked improvements in goodness-of-fit (GOF) over the standard Poisson model assumptions, but with evidence of over-fitting in some genes. Modeling of experimental effects improved GOF for high variance genes but increased the over-fitting problem. CONCLUSIONS: These conclusions will guide development of analytical strategies for accurate modeling of variance structure in these data and sample size determination which in turn will aid in the identification of true biological signals that inform our understanding of biological systems.

Disentangling personalized treatment effects from "time-of-the-day" confounding in mobile health studies.

Ideally, a patient's response to medication can be monitored by measuring changes in performance of some activity. In observational studies, however, any detected association between treatment ("on-medication" vs "off-medication") and the outcome (performance in the activity) might be due to confounders. In particular, causal inferences at the personalized level are especially vulnerable to confounding effects that arise in a cyclic fashion. For quick acting medications, effects can be confounded by circadian rhythms and daily routines. Using the time-of-the-day as a surrogate for these confounders and the performance measurements as captured on a smartphone, we propose a personalized statistical approach to disentangle putative treatment and "time-of-the-day" effects, that leverages conditional independence relations spanned by causal graphical models involving the treatment, time-of-the-day, and outcome variables. Our approach is based on conditional independence tests implemented via standard and temporal linear regression models. Using synthetic data, we investigate when and how residual autocorrelation can affect the standard tests, and how time series modeling (namely, ARIMA and robust regression via HAC covariance matrix estimators) can remedy these issues. In particular, our simulations illustrate that when patients perform their activities in a paired fashion, positive autocorrelation can lead to conservative results for the standard regression approach (i.e., lead to deflated true positive detection), whereas negative autocorrelation can lead to anticonservative behavior (i.e., lead to inflated false positive detection). The adoption of time series methods, on the other hand, leads to well controlled type I error rates. We illustrate the application of our methodology with data from a Parkinson's disease mobile health study.

Remote smartphone monitoring of Parkinson's disease and individual response to therapy.

Remote health assessments that gather real-world data (RWD) outside clinic settings require a clear understanding of appropriate methods for data collection, quality assessment, analysis and interpretation. Here we examine the performance and limitations of smartphones in collecting RWD in the remote mPower observational study of Parkinson's disease (PD). Within the first 6 months of study commencement, 960 participants had enrolled and performed at least five self-administered active PD symptom assessments (speeded tapping, gait/balance, phonation or memory). Task performance, especially speeded tapping, was predictive of self-reported PD status (area under the receiver operating characteristic curve (AUC) = 0.8) and correlated with in-clinic evaluation of disease severity (r = 0.71; P < 1.8 × 10-6) when compared with motor Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS). Although remote assessment requires careful consideration for accurate interpretation of RWD, our results support the use of smartphones and wearables in objective and personalized disease assessments.

Evaluation of guanylyl cyclase C lymph node status for colon cancer staging and prognosis.

PURPOSE: The prognostic significance of guanylyl cyclase C (GCC) gene expression in lymph nodes (LNs) was evaluated in patients with stage II colon cancer who were not treated with adjuvant chemotherapy. We report a planned analysis performed on 241 patients. METHODS: GCC mRNA was quantified by RT-qPCR using formalin-fixed LN tissues from patients with untreated stage II colon cancer who were diagnosed from 1999-2006 with at least ten LNs examined and blinded to clinical outcomes. Lymph node ratio (LNR) is the number of GCC-positive nodes divided by total number of informative LNs. Risk categories of low (0-0.1) and high (>0.1) for LNR were chosen by significance using Cox regression models. The data were tested for association with time to recurrence. RESULTS: Twenty-nine patients (12%) had a disease recurrence or cancer death. The LNR significantly predicted higher recurrence risk for 84 patients (34.9%) classified as high risk (hazard ratio (HR), 2.38; P=0.02). The estimated 5-year recurrence rates were 10% and 27% for the low- and high-risk groups, respectively. After adjusting for age, T stage, number of nodes assessed, and MMR status, a significant association remained (HR, 2.61; P=0.02). In a subset of patients (n=181) with T3 tumor, ≥12 nodes examined and negative margins, a significant association between the GCC LNR and recurrence risk also was observed (HR, 5.06; P=0.003). CONCLUSIONS: Our preliminary results suggest that detection of GCC mRNA in LNs is associated with risk of disease recurrence in patients with untreated stage II colon cancer. A larger validation study is ongoing.

Assessment of the All of Us research program's informed consent process.

Informed consent is the gateway to research participation. We report on the results of the formative evaluation that follows the electronic informed consent process for the All of Us Research Program. Of the nearly 250,000 participants included in this analysis, more than 95% could correctly answer questions distinguishing the program from medical care, the voluntary nature of participation, and the right to withdraw; comparatively, participants were less sure of privacy risk of the program. We also report on a small mixed-methods study of the experience of persons of very low health literacy with All of Us informed consent materials. Of note, many of the words commonly employed in the consent process were unfamiliar to or differently defined by informants. In combination, these analyses may inform participant-centered development and highlight areas for refinement of informed consent materials for the All of Us Research Program and similar studies.

Detecting the impact of subject characteristics on machine learning-based diagnostic applications.

Collection of high-dimensional, longitudinal digital health data has the potential to support a wide-variety of research and clinical applications including diagnostics and longitudinal health tracking. Algorithms that process these data and inform digital diagnostics are typically developed using training and test sets generated from multiple repeated measures collected across a set of individuals. However, the inclusion of repeated measurements is not always appropriately taken into account in the analytical evaluations of predictive performance. The assignment of repeated measurements from each individual to both the training and the test sets ("record-wise" data split) is a common practice and can lead to massive underestimation of the prediction error due to the presence of "identity confounding." In essence, these models learn to identify subjects, in addition to diagnostic signal. Here, we present a method that can be used to effectively calculate the amount of identity confounding learned by classifiers developed using a record-wise data split. By applying this method to several real datasets, we demonstrate that identity confounding is a serious issue in digital health studies and that record-wise data splits for machine learning- based applications need to be avoided.

Physical activity, sleep and cardiovascular health data for 50,000 individuals from the MyHeart Counts Study.

Studies have established the importance of physical activity and fitness for long-term cardiovascular health, yet limited data exist on the association between objective, real-world large-scale physical activity patterns, fitness, sleep, and cardiovascular health primarily due to difficulties in collecting such datasets. We present data from the MyHeart Counts Cardiovascular Health Study, wherein participants contributed data via an iPhone application built using Apple's ResearchKit framework and consented to make this data available freely for further research applications. In this smartphone-based study of cardiovascular health, participants recorded daily physical activity, completed health questionnaires, and performed a 6-minute walk fitness test. Data from English-speaking participants aged 18 years or older with a US-registered iPhone who agreed to share their data broadly and who enrolled between the study's launch and the time of the data freeze for this data release (March 10 2015-October 28 2015) are now available for further research. It is anticipated that releasing this large-scale collection of real-world physical activity, fitness, sleep, and cardiovascular health data will enable the research community to work collaboratively towards improving our understanding of the relationship between cardiovascular indicators, lifestyle, and overall health, as well as inform mobile health research best practices.

The asthma mobile health study, smartphone data collected using ResearchKit.

Widespread adoption of smart mobile platforms coupled with a growing ecosystem of sensors including passive location tracking and the ability to leverage external data sources create an opportunity to generate an unprecedented depth of data on individuals. Mobile health technologies could be utilized for chronic disease management as well as research to advance our understanding of common diseases, such as asthma. We conducted a prospective observational asthma study to assess the feasibility of this type of approach, clinical characteristics of cohorts recruited via a mobile platform, the validity of data collected, user retention patterns, and user data sharing preferences. We describe data and descriptive statistics from the Asthma Mobile Health Study, whereby participants engaged with an iPhone application built using Apple's ResearchKit framework. Data from 6346 U.S. participants, who agreed to share their data broadly, have been made available for further research. These resources have the potential to enable the research community to work collaboratively towards improving our understanding of asthma as well as mobile health research best practices.

The Healthy Pregnancy Research Program: transforming pregnancy research through a ResearchKit app.

Although maternal morbidity and mortality in the US is among the worst of developed countries, pregnant women have been under-represented in research studies, resulting in deficiencies in evidence-based guidance for treatment. There are over two billion smartphone users worldwide, enabling researchers to easily and cheaply conduct extremely large-scale research studies through smartphone apps, especially among pregnant women in whom app use is exceptionally high, predominantly as an information conduit. We developed the first pregnancy research app that is embedded within an existing, popular pregnancy app for self-management and education of expectant mothers. Through the large-scale and simplified collection of survey and sensor generated data via the app, we aim to improve our understanding of factors that promote a healthy pregnancy for both the mother and developing fetus. From the launch of this cohort study on 16 March 2017 through 17 December 2017, we have enrolled 2058 pregnant women from all 50 states. Our study population is diverse geographically and demographically, and fairly representative of US population averages. We have collected 14,045 individual surveys and 107,102 total daily measurements of sleep, activity, blood pressure, and heart rate during this time. On average, women stayed engaged in the study for 59 days and 45 percent who reached their due date filled out the final outcome survey. During the first 9 months, we demonstrated the potential for a smartphone-based research platform to capture an ever-expanding array of longitudinal, objective, and subjective participant-generated data from a continuously growing and diverse population of pregnant women.

Phase I North Central Cancer Treatment Group Trial-N9923 of escalating doses of twice-daily thoracic radiation therapy with amifostine and with alternating chemotherapy in limited stage small-cell lung cancer.

PURPOSE: The primary goal was to identify the maximum tolerable dose (MTD) of thoracic radiation therapy (TRT) that can be given with chemotherapy and amifostine for patients with limited-stage small-cell lung cancer (LSCLC). METHODS AND MATERIALS: Treatment began with two cycles of topotecan (1 mg/m(2)) Days 1 to 5 and paclitaxel (175 mg/m(2)) Day 5 (every 3 weeks) given before and after TRT. The TRT began at 6 weeks. The TRT was given in 120 cGy fractions b.i.d. and the dose escalation (from 4,800 cGy, dose level 1, to 6,600 cGy, dose level 4) followed the standard "cohorts of 3" design. The etoposide (E) (50 mg/day) and cisplatin (C) (3 mg/m(2)) were given i.v. before the morning TRT and amifostine (500 mg/day) was given before the afternoon RT. This was followed by prophylactic cranial irradiation (PCI). The dose-limiting toxicities (DLTs) were defined as Grade > or =4 hematologic, febrile neutropenia, esophagitis, or other nonhematologic toxicity, Grade > or =3 dyspnea, or Grade > or =2 pneumonitis. RESULTS: Fifteen patients were evaluable for the Phase I portion of the trial. No DLTs were seen at dose levels 1 and 2. Two patients on dose level 4 experienced DLTs: 1 patient had a Grade 4 pneumonitis, dyspnea, fatigue, hypokalemia, and anorexia, and 1 patient had a Grade 5 hypoxia attributable to TRT. One of 6 patients on dose level 3 had a DLT, Grade 3 esophagitis. The Grade > or =3 toxicities seen in at least 10% of patients during TRT were esophagitis (53%), leukopenia (33%), dehydration (20%), neutropenia (13%), and fatigue (13%). The median survival was 14.5 months. CONCLUSION: The MTD of b.i.d. TRT was 6000 cGy (120 cGy b.i.d.) with EP and amifostine.

Formative Evaluation of Participant Experience With Mobile eConsent in the App-Mediated Parkinson mPower Study: A Mixed Methods Study.

BACKGROUND: To fully capitalize on the promise of mobile technology to enable scalable, participant-centered research, we must develop companion self-administered electronic informed consent (eConsent) processes. As we do so, we have an ethical obligation to ensure that core tenants of informed consent-informedness, comprehension, and voluntariness-are upheld. Furthermore, we should be wary of recapitulating the pitfalls of "traditional" informed consent processes. OBJECTIVE: Our objective was to describe the essential qualities of participant experience, including delineation of common and novel themes relating to informed consent, with a self-administered, smartphone-based eConsent process. We sought to identify participant responses related to informedness, comprehension, and voluntariness as well as to capture any emergent themes relating to the informed consent process in an app-mediated research study. METHODS: We performed qualitative thematic analysis of participant responses to a daily general prompt collected over a 6-month period within the Parkinson mPower app. We employed a combination of a priori and emergent codes for our analysis. A priori codes focused on the core concepts of informed consent; emergent codes were derived to capture additional themes relating to self-administered consent processes. We used self-reported demographic information from the study's baseline survey to characterize study participants and respondents. RESULTS: During the study period, 9846 people completed the eConsent process and enrolled in the Parkinson mPower study. In total, 2758 participants submitted 7483 comments; initial categorization identified a subset of 3875 germane responses submitted by 1678 distinct participants. Respondents were more likely to self-report a Parkinson disease diagnosis (30.21% vs 11.10%), be female (28.26% vs 20.18%), be older (42.89 years vs 34.47 years), and have completed more formal education (66.23% with a 4-year college degree or more education vs 55.77%) than all the mPower participants (P<.001 for all values). Within our qualitative analysis, 3 conceptual domains emerged. First, consistent with fully facilitated in-person informed consent settings, we observed a broad spectrum of comprehension of core research concepts following eConsent. Second, we identified new consent themes born out of the remote mobile research setting, for example the impact of the study design on the engagement of controls and the misconstruction of the open response field as a method for responsive communication with researchers, that bear consideration for inclusion within self-administered eConsent. Finally, our findings highlighted participants' desire to be empowered as partners. CONCLUSIONS: Our study serves as a formative evaluation of participant experience with a self-administered informed consent process via a mobile app. Areas for future investigation include direct comparison of the efficacy of self-administered eConsent with facilitated informed consent processes, exploring the potential benefits and pitfalls of smartphone user behavioral habits on participant engagement in research, and developing best practices to increase informedness, comprehension, and voluntariness via participant coengagement in the research endeavor.

The Mole Mapper Study, mobile phone skin imaging and melanoma risk data collected using ResearchKit.

Sensor-embedded phones are an emerging facilitator for participant-driven research studies. Skin cancer research is particularly amenable to this approach, as phone cameras enable self-examination and documentation of mole abnormalities that may signal a progression towards melanoma. Aggregation and open sharing of this participant-collected data can be foundational for research and the development of early cancer detection tools. Here we describe data from Mole Mapper, an iPhone-based observational study built using the Apple ResearchKit framework. The Mole Mapper app was designed to collect participant-provided images and measurements of moles, together with demographic and behavioral information relating to melanoma risk. The study cohort includes 2,069 participants who contributed 1,920 demographic surveys, 3,274 mole measurements, and 2,422 curated mole images. Survey data recapitulates associations between melanoma and known demographic risks, with red hair as the most significant factor in this cohort. Participant-provided mole measurements indicate an average mole size of 3.95 mm. These data have been made available to engage researchers in a collaborative, multidisciplinary effort to better understand and prevent melanoma.

PERSONALIZED HYPOTHESIS TESTS FOR DETECTING MEDICATION RESPONSE IN PARKINSON DISEASE PATIENTS USING iPHONE SENSOR DATA.

We propose hypothesis tests for detecting dopaminergic medication response in Parkinson disease patients, using longitudinal sensor data collected by smartphones. The processed data is composed of multiple features extracted from active tapping tasks performed by the participant on a daily basis, before and after medication, over several months. Each extracted feature corresponds to a time series of measurements annotated according to whether the measurement was taken before or after the patient has taken his/her medication. Even though the data is longitudinal in nature, we show that simple hypothesis tests for detecting medication response, which ignore the serial correlation structure of the data, are still statistically valid, showing type I error rates at the nominal level. We propose two distinct personalized testing approaches. In the first, we combine multiple feature-specific tests into a single union-intersection test. In the second, we construct personalized classifiers of the before/after medication labels using all the extracted features of a given participant, and test the null hypothesis that the area under the receiver operating characteristic curve of the classifier is equal to 1/2. We compare the statistical power of the personalized classifier tests and personalized union-intersection tests in a simulation study, and illustrate the performance of the proposed tests using data from mPower Parkinsons disease study, recently launched as part of Apples ResearchKit mobile platform. Our results suggest that the personalized tests, which ignore the longitudinal aspect of the data, can perform well in real data analyses, suggesting they might be used as a sound baseline approach, to which more sophisticated methods can be compared to.

Mobile health: the power of wearables, sensors, and apps to transform clinical trials.

Mobile technology has become a ubiquitous part of everyday life, and the practical utility of mobile devices for improving human health is only now being realized. Wireless medical sensors, or mobile biosensors, are one such technology that is allowing the accumulation of real-time biometric data that may hold valuable clues for treating even some of the most devastating human diseases. From wearable gadgets to sophisticated implantable medical devices, the information retrieved from mobile technology has the potential to revolutionize how clinical research is conducted and how disease therapies are delivered in the coming years. Encompassing the fields of science and engineering, analytics, health care, business, and government, this report explores the promise that wearable biosensors, along with integrated mobile apps, hold for improving the quality of patient care and clinical outcomes. The discussion focuses on groundbreaking device innovation, data optimization and validation, commercial platform integration, clinical implementation and regulation, and the broad societal implications of using mobile health technologies.

The mPower study, Parkinson disease mobile data collected using ResearchKit.

Current measures of health and disease are often insensitive, episodic, and subjective. Further, these measures generally are not designed to provide meaningful feedback to individuals. The impact of high-resolution activity data collected from mobile phones is only beginning to be explored. Here we present data from mPower, a clinical observational study about Parkinson disease conducted purely through an iPhone app interface. The study interrogated aspects of this movement disorder through surveys and frequent sensor-based recordings from participants with and without Parkinson disease. Benefitting from large enrollment and repeated measurements on many individuals, these data may help establish baseline variability of real-world activity measurement collected via mobile phones, and ultimately may lead to quantification of the ebbs-and-flows of Parkinson symptoms. App source code for these data collection modules are available through an open source license for use in studies of other conditions. We hope that releasing data contributed by engaged research participants will seed a new community of analysts working collaboratively on understanding mobile health data to advance human health.