scholarly journals Electronic patient-reported outcomes and machine learning in predicting immune-related adverse events of immune checkpoint inhibitor therapies

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Sanna Iivanainen ◽  
Jussi Ekstrom ◽  
Henri Virtanen ◽  
Vesa V. Kataja ◽  
Jussi P. Koivunen
Keyword(s):  
Machine Learning ◽  
Early Detection ◽  
Adverse Events ◽  
Cancer Patients ◽  
Immune Checkpoint ◽  
Patient Reported Outcomes ◽  
Prediction Models ◽  
Accuracy Score ◽  
Patient Reported ◽  
Auc Value

Abstract Background Immune-checkpoint inhibitors (ICIs) have introduced novel immune-related adverse events (irAEs), arising from various organ systems without strong timely dependency on therapy dosing. Early detection of irAEs could result in improved toxicity profile and quality of life. Symptom data collected by electronic (e) patient-reported outcomes (PRO) could be used as an input for machine learning (ML) based prediction models for the early detection of irAEs. Methods The utilized dataset consisted of two data sources. The first dataset consisted of 820 completed symptom questionnaires from 34 ICI treated advanced cancer patients, including 18 monitored symptoms collected using the Kaiku Health digital platform. The second dataset included prospectively collected irAE data, Common Terminology Criteria for Adverse Events (CTCAE) class, and the severity of 26 irAEs. The ML models were built using extreme gradient boosting algorithms. The first model was trained to detect the presence and the second the onset of irAEs. Results The model trained to predict the presence of irAEs had an excellent performance based on four metrics: accuracy score 0.97, Area Under the Curve (AUC) value 0.99, F1-score 0.94 and Matthew’s correlation coefficient (MCC) 0.92. The prediction of the irAE onset was more difficult with accuracy score 0.96, AUC value 0.93, F1-score 0.66 and MCC 0.64 but the model performance was still at a good level. Conclusion The current study suggests that ML based prediction models, using ePRO data as an input, can predict the presence and onset of irAEs with a high accuracy, indicating that ePRO follow-up with ML algorithms could facilitate the detection of irAEs in ICI-treated cancer patients. The results should be validated with a larger dataset. Trial registration Clinical Trials Register (NCT3928938), registration date the 26th of April, 2019

scholarly journals Electronic Patient-Reported Outcomes and Machine Learning in Predicting Immune-Related Adverse Events of Immune Checkpoint Inhibitor Therapies

2021 ◽  
Author(s):  
Sanna Iivanainen ◽  
Jussi Ekstrom ◽  
Henri Virtanen ◽  
Vesa Kataja ◽  
Jussi Koivunen
Keyword(s):  
Machine Learning ◽  
Early Detection ◽  
Adverse Events ◽  
Cancer Patients ◽  
Immune Checkpoint ◽  
Patient Reported Outcomes ◽  
Prediction Models ◽  
Accuracy Score ◽  
Patient Reported ◽  
Auc Value

Abstract Background: Immune-checkpoint inhibitors (ICIs) have introduced novel immune-related adverse events (irAEs), arising from various organ systems without strong timely dependency on therapy dosing. Early detection of irAEs could result in improved toxicity profile and quality of life (QoL). Symptom data collected by electronic (e) patient-reported outcomes (PRO) could be used as an input for machine learning (ML) based prediction models for early detection of irAEs.Methods: The utilized dataset consisted of two data sources. The first dataset (n=16540) comprised symptoms from 34 ICI-treated cancer patients, including 18 monitored symptoms collected using the Kaiku Health digital platform. The second dataset included prospectively collected irAE data, Common Terminology Criteria for Adverse Events (CTCAE) class, and severity of 26 irAEs. The ML models were built using extreme gradient boosting algorithms. The first model was trained to detect the presence and the second model the onset of irAEs. Results: The model trained to predict the presence of irAEs had an excellent performance based on four metrics: accuracy score 0.97, Area Under the Curve (AUC) value 0.99, F1-score 0.94 and Matthew´s correlation coefficient (MCC) 0.92. The prediction of the irAE onset was more difficult with accuracy score 0.96, AUC value 0.93, F1-score 0.66 and MCC 0.64 but the model performance was still at a good level.Conclusion: Current study suggests that ML based prediction models, using ePRO data as an input, can predict the presence and onset of irAEs with high accuracy, indicating that ePRO follow-up with ML algorithms could facilitate the detection of irAEs in ICI-treated cancer patients. The results should be validated with a larger dataset.

scholarly journals Linguistic validation of the simplified Chinese version of the US National Cancer Institute’s patient-reported outcomes version of the common terminology criteria for adverse events (PRO-CTCAE™)

BMC Cancer ◽  
2020 ◽  
Vol 20 (1) ◽  
Author(s):  
Cheng KKF ◽  
S. A. Mitchell ◽  
N. Chan ◽  
E. Ang ◽  
W. Tam ◽  
...  
Keyword(s):  
Adverse Events ◽  
Cancer Patients ◽  
Patient Reported Outcomes ◽  
A Priori ◽  
Cognitive Interviews ◽  
Simplified Chinese ◽  
Patient Reported ◽  
The Us ◽  
The Common ◽  
Response Choices

Abstract Background The aim of this study was to translate and linguistically validate the U.S. National Cancer Institute’s Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE™) into Simplified Chinese for use in Singapore. Methods All 124 items of the English source PRO-CTCAE item library were translated into Simplified Chinese using internationally established translation procedures. Two rounds of cognitive interviews were conducted with 96 cancer patients undergoing adjuvant treatment to determine if the translations adequately captured the PRO-CTCAE source concepts, and to evaluate comprehension, clarity and ease of judgement. Interview probes addressed the 78 PRO-CTCAE symptom terms (e.g. fatigue), as well as the attributes (e.g. severity), response choices, and phrasing of ‘at its worst’. Items that met the a priori threshold of ≥20% of participants with comprehension difficulties were considered for rephrasing and retesting. Items where < 20% of the sample experienced comprehension difficulties were also considered for rephrasing if better phrasing options were available. Results A majority of PRO-CTCAE-Simplified Chinese items were well comprehended by participants in Round 1. One item posed difficulties in ≥20% and was revised. Two items presented difficulties in < 20% but were revised as there were preferred alternative phrasings. Twenty-four items presented difficulties in < 10% of respondents. Of these, eleven items were revised to an alternative preferred phrasing, four items were revised to include synonyms. Revised items were tested in Round 2 and demonstrated satisfactory comprehension. Conclusions PRO-CTCAE-Simplified Chinese has been successfully developed and linguistically validated in a sample of cancer patients residing in Singapore.

Efficacy and feasibility of OncoSmart, a tool for remote detection and monitoring chemoterapy-related adverse events in metastatic colorectal cancer patients: a pilot retrospective study. (Preprint)

2020 ◽  
Author(s):  
Teresa Troiani ◽  
Stefania Napolitano ◽  
Marinella Terminiello ◽  
Pietro Paolo Vitiello ◽  
Fortunato Ciardiello ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Adverse Events ◽  
Metastatic Colorectal Cancer ◽  
Cancer Patients ◽  
Patient Reported Outcomes ◽  
Wearable Device ◽  
Strong Impact ◽  
Care Providers ◽  
Symptom Intensity ◽  
Patient Reported

BACKGROUND In metastatic colorectal cancer (mCRC) treatment-related health symptoms may have a strong impact on patient’s quality of life (QoL). It has been shown that a considerable number of health care providers underestimates symptom intensity. In this context, the systematic collection of electronic patient-reported outcomes (ePROs) has been demonstrated to be a valid, reliable, feasible and precise approach to tabulating symptomatic toxicities and to detect symptoms missed by clinicians. OBJECTIVE We aimed to evaluate feasibility as well as patients’ acceptance of remote technology system to detect and monitoring chemotherapy-related adverse events in metastatic colorectal cancer outpatients. METHODS We enrolled 8 mCRC outpatients who received an oncological treatment. A wearable device (smart watch) allowing automatic vitals measurement (blood pressure, heart rate, oxygen saturation, respiratory rate, pedometer and sleeping monitor) has been provided to all patients. Moreover, two mobile applications have been developed: the first one to monitor vital measurements recorded by the wearable device, the second one to identify treatment-related toxicities and QoL parameters using a 30-items questionnaire (some taken from EORTCQLQ-C30 and others composed by the investigators). Clinicians filled the electronic health records (EHR) at each visit with symptoms reported by patients, physical examination and any treatment modifications. RESULTS a total of 8 patients were enrolled, 2 women (25%) and 6 men (75%); median age was 54 years (range 35-69). Compliance was 77%. Overall concordance between ePRO and symptoms detected by clinicians was 80%; in 15% of cases of electronic patient-reported outcomes (ePROs) included symptoms missed during the visit, while in 5% of cases clinicians reported toxicities not recorded by patients. Regarding the symptoms that led to treatment modifications and/or suspension, the concordance between ePROs and clinician’s evaluation during the visit was 100%. CONCLUSIONS In our pilot experience this type of ePROs is feasible and well tolerated, showing high compliance (80%), and allowing identification of toxicities missed by clinicians in 15% of cases. These data suggest that the integration of ePROs with EHR may improve the management of cancer patients. These strategies should be prioritized to optimize active oncological treatments and supportive care in order to improve patient’s QoL and reduce inappropriate hospitalization.

scholarly journals The Effect of Collaborative Reviews of Electronic Patient-Reported Outcomes on the Congruence of Patient- and Clinician-Reported Toxicity in Cancer Patients Receiving Systemic Therapy: Prospective, Multicenter, Observational Clinical Trial (Preprint)

2021 ◽  
Author(s):  
Andreas Trojan ◽  
Nicolas Leuthold ◽  
Christoph Thomssen ◽  
Achim Rody ◽  
Thomas Winder ◽  
...  
Keyword(s):  
Clinical Trial ◽  
Clinical Practice ◽  
Adverse Events ◽  
Cancer Patients ◽  
Systemic Therapy ◽  
Patient Reported Outcomes ◽  
Patient Reported Outcome ◽  
Primary Objective ◽  
Patient Reported ◽  
Level Of Agreement

BACKGROUND Electronic patient-reported outcomes (ePRO) are a relatively novel form of data and have the potential to improve clinical practice for cancer patients. In this prospective, multicenter, observational clinical trial, efforts were made to demonstrate the reliability of patient-reported symptoms. OBJECTIVE The primary objective of this study was to assess the level of agreement κ between symptom ratings by physicians and patients via a shared review process in order to determine the future reliability and utility of self-reported electronic symptom monitoring. METHODS Patients receiving systemic therapy in a (neo-)adjuvant or noncurative intention setting captured ePRO for 52 symptoms over an observational period of 90 days. At 3-week intervals, randomly selected symptoms were reviewed between the patient and physician for congruency on severity of the grading of adverse events according to the Common Terminology Criteria of Adverse Events (CTCAE). The patient-physician agreement for the symptom review was assessed via Cohen kappa (κ), through which the interrater reliability was calculated. Chi-square tests were used to determine whether the patient-reported outcome was different among symptoms, types of cancer, demographics, and physicians’ experience. RESULTS Among the 181 patients (158 women and 23 men; median age 54.4 years), there was a fair scoring agreement (κ=0.24; 95% CI 0.16-0.33) for symptoms that were entered 2 to 4 weeks before the intended review (first rating) and a moderate agreement (κ=0.41; 95% CI 0.34-0.48) for symptoms that were entered within 1 week of the intended review (second rating). However, the level of agreement increased from moderate (first rating, κ=0.43) to substantial (second rating, κ=0.68) for common symptoms of pain, fever, diarrhea, obstipation, nausea, vomiting, and stomatitis. Similar congruency levels of ratings were found for the most frequently entered symptoms (first rating: κ=0.42; second rating: κ=0.65). The symptom with the lowest agreement was hair loss (κ=–0.05). With regard to the latency of symptom entry into the review, hardly any difference was demonstrated between symptoms that were entered from days 1 to 3 and from days 4 to 7 before the intended review (κ=0.40 vs κ=0.39, respectively). In contrast, for symptoms that were entered 15 to 21 days before the intended review, no congruency was demonstrated (κ=–0.15). Congruency levels seemed to be unrelated to the type of cancer, demographics, and physicians’ review experience. CONCLUSIONS The shared monitoring and review of symptoms between patients and clinicians has the potential to improve the understanding of patient self-reporting. Our data indicate that the integration of ePRO into oncological clinical research and continuous clinical practice provides reliable information for self-empowerment and the timely intervention of symptoms. CLINICALTRIAL ClinicalTrials.gov NCT03578731; https://clinicaltrials.gov/ct2/show/NCT03578731

Electronic patient-reported outcomes (ePROs) and machine learning (ML) in predicting the presence and onset of immune-related adverse events (irAEs) of immune checkpoint inhibitor (ICI) therapies.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e14058-e14058
Author(s):  
Sanna Iivanainen ◽  
Jussi Ekström ◽  
Vesa V Kataja ◽  
Henri Virtanen ◽  
Jussi Koivunen
Keyword(s):  
Early Detection ◽  
Cancer Patients ◽  
Performance Metrics ◽  
Prediction Models ◽  
Model Performance ◽  
Gradient Boosting ◽  
Random Allocation ◽  
Organ Systems ◽  
Extreme Gradient Boosting ◽  
Patient Reported

e14058 Background: ICIs have introduced novel irAEs, arising from various organ systems without strong timely dependency on initiation and discontinuation of the therapy. Early detection of the irAEs could result in improved safety profile of the treatment and better quality of life for patients. Symptom data collected by ePROs could be used as an input for ML based prediction models for early detection of irAEs. Methods: The utilized dataset consisted of two data sources. The first dataset consisted of 16 540 reported symptoms from 33 ICI-treated cancer patients, including 18 monitored symptoms collected using Kaiku Health digital platform. The second dataset included prospectively collected irAE data, including initiation and end dates, CTCAE class, and severity of 26 irAEs (the longest irAE lasted 799 days, and the shortest two days while median duration was 61 days). Two ML models were built using extreme gradient boosting, a well-known classification algorithm. Using the ePRO data, the first model was trained to detect the presence and the second model to detect the onset (0-21 days prior to diagnosis) of irAEs. The dataset was split into training (70 % of the data) and test sets (30 % of the data) by random allocation. The test set was left out from the model training and tuning, and was used only to evaluate the model performance. Results: The model trained to predict the presence of irAEs had an excellent performance with the test dataset. The prediction of the irAE onset was more difficult, but the model performance was still at a very good level. The performance metrics for the ML models are presented in Table. Conclusions: Current study suggests that ML based prediction models, using ePRO data as input for the models, can predict the presence and onset of irAEs with high accuracy. Thus, it indicates that digital symptom monitoring combined with ML could enable the detection of irAEs in ICI-treated cancer patients. The results should be validated with a larger dataset from prospective clinical trials. Clinical trial information: NCT03928938. [Table: see text]

Sign in / Sign up

Export Citation Format

Share Document