SurvdigitizeR: an algorithm for automated survival curve digitization.

BACKGROUND: Decision analytic models and meta-analyses often rely on survival probabilities that are digitized from published Kaplan-Meier (KM) curves. However, manually extracting these probabilities from KM curves is time-consuming, expensive, and error-prone. We developed an efficient and accurate algorithm that automates extraction of survival probabilities from KM curves. METHODS: The automated digitization algorithm processes images from a JPG or PNG format, converts them in their hue, saturation, and lightness scale and uses optical character recognition to detect axis location and labels. It also uses a k-medoids clustering algorithm to separate multiple overlapping curves on the same figure. To validate performance, we generated survival plots form random time-to-event data from a sample size of 25, 50, 150, and 250, 1000 individuals split into 1,2, or 3 treatment arms. We assumed an exponential distribution and applied random censoring. We compared automated digitization and manual digitization performed by well-trained researchers. We calculated the root mean squared error (RMSE) at 100-time points for both methods. The algorithm's performance was also evaluated by Bland-Altman analysis for the agreement between automated and manual digitization on a real-world set of published KM curves. RESULTS: The automated digitizer accurately identified survival probabilities over time in the simulated KM curves. The average RMSE for automated digitization was 0.012, while manual digitization had an average RMSE of 0.014. Its performance was negatively correlated with the number of curves in a figure and the presence of censoring markers. In real-world scenarios, automated digitization and manual digitization showed very close agreement. CONCLUSIONS: The algorithm streamlines the digitization process and requires minimal user input. It effectively digitized KM curves in simulated and real-world scenarios, demonstrating accuracy comparable to conventional manual digitization. The algorithm has been developed as an open-source R package and as a Shiny application and is available on GitHub: https://github.com/Pechli-Lab/SurvdigitizeR and https://pechlilab.shinyapps.io/SurvdigitizeR/ .

Cost-effectiveness of the McGill interactive pediatric oncogenetic guidelines in identifying Li-Fraumeni syndrome in female patients with osteosarcoma.

BACKGROUND: Li-Fraumeni syndrome (LFS) is a penetrant cancer predisposition syndrome (CPS) associated with the development of many tumor types in young people including osteosarcoma and breast cancer (BC). The McGill Interactive Pediatric OncoGenetic Guidelines (MIPOGG) decision-support tool provides a standardized approach to identify patients at risk of CPSs. METHODS: We conducted a cost-utility analysis, from the healthcare payer perspective, to compare MIPOGG-guided, physician-guided, and universal genetic testing strategies to detect LFS in female patients diagnosed at an age of less than 18 years with osteosarcoma. We developed a decision tree and discrete-event simulation model to simulate the clinical and cost outcomes of the three genetic referral strategies on a cohort of female children diagnosed with osteosarcoma, especially focused on BC as subsequent cancer. Outcomes included BC incidence, quality-adjusted life-years (QALYs), healthcare costs, and incremental cost-utility ratios (ICURs). We conducted probabilistic and scenario analyses to assess the uncertainty surrounding model parameters. RESULTS: Compared to the physician-guided testing, the MIPOGG-guided strategy was marginally more expensive by $105 (-$516; $743), but slightly more effective by 0.003 (-0.04; 0.045) QALYs. Compared to MIPOGG, the universal testing strategy was $1333 ($732; $1953) more costly and associated with 0.011 (-0.043; 0.064) additional QALYs. The ICUR for the MIPOGG strategy was $33,947/QALY when compared to the physician strategy; the ICUR for universal testing strategy was $118,631/QALY when compared to the MIPOGG strategy. DISCUSSION: This study provides evidence for clinical and policy decision-making on the cost-effectiveness of genetic referral strategies to identify LFS in the setting of osteosarcoma. MIPOGG-guided strategy was most likely to be cost-effective at a willingness-to-pay threshold value of $50,000/QALY.

Simplified Methods for Modelling Dependent Parameters in Health Economic Evaluations: A Tutorial.

BACKGROUND: In health economic evaluations, model parameters are often dependent on other model parameters. Although methods exist to simulate multivariate normal (MVN) distribution data and estimate transition probabilities in Markov models while considering competing risks, they are technically challenging for health economic modellers to implement. This tutorial introduces easily implementable applications for handling dependent parameters in modelling. METHODS: Analytical proofs and proposed simplified methods for handling dependent parameters in typical health economic modelling scenarios are provided, and implementation of these methods are illustrated in seven examples along with the SAS and R code. RESULTS: Methods to quantify the covariance and correlation coefficients of correlated variables based on published summary statistics and generation of MVN distribution data are demonstrated using examples of physician visits data and cost component data. The use of univariate normal distribution data instead of MVN distribution data to capture population heterogeneity is illustrated based on the results from multiple regression models with linear predictors, and two examples are provided (linear fixed-effects model and Cox proportional hazards model). A conditional probability method is introduced to handle two or more state transitions in a single Markov model cycle and applied in examples of one- and two-way state transitions. CONCLUSIONS: This tutorial proposes an extension of routinely used methods along with several examples. These simplified methods may be easily applied by health economic modellers with varied statistical backgrounds.

Clinical and economic impact of molecular testing for BRAF fusion in pediatric low-grade Glioma.

BACKGROUND: Treatment personalization via tumor molecular testing holds promise for improving outcomes for patients with pediatric low-grade glioma (PLGG). We evaluate the health economic impact of employing tumor molecular testing to guide treatment for patients diagnosed with PLGG, particularly the avoidance of radiation therapy (RT) for patients with BRAF-fusion. METHODS: We performed a model-based cost-utility analysis comparing two strategies: molecular testing to determine BRAF fusion status at diagnosis against no molecular testing. We developed a microsimulation to model the lifetime health and cost outcomes (in quality-adjusted life years (QALYs) and 2018 CAD, respectively) for a simulated cohort of 100,000 patients newly diagnosed with PLGG after their initial surgery. RESULTS: The life expectancy after diagnosis for individuals who did not receive molecular testing was 39.01 (95% Confidence Intervals (CI): 32.94;44.38) years and 40.08 (95% CI: 33.19;45.76) years for those who received testing. Our findings indicate that patients who received molecular testing at diagnosis experienced a 0.38 (95% CI: 0.08;0.77) gain in QALYs and $1384 (95% CI: $-3486; $1204) reduction in costs over their lifetime. Cost and QALY benefits were driven primarily by the avoidance of long-term adverse events (stroke, secondary neoplasms) associated with unnecessary use of radiation. CONCLUSIONS: We demonstrate the clinical benefit and cost-effectiveness of molecular testing in guiding the decision to provide RT in PLGG. While our results do not consider the impact of targeted therapies, this work is an example of the value of simulation modeling in assessing the long-term costs and benefits of precision oncology interventions for childhood cancer, which can aid decision-making about health system reimbursement.

The intranasal dexmedetomidine plus ketamine for procedural sedation in children, adaptive randomized controlled non-inferiority multicenter trial (Ketodex): a statistical analysis plan.

BACKGROUND: Procedural sedation and analgesia (PSA) is frequently required to perform closed reductions for fractures and dislocations in children. Intravenous (IV) ketamine is the most commonly used sedative agent for closed reductions. However, as children find IV insertion a distressing and painful procedure, there is need to identify a feasible alternative route of administration. There is evidence that a combination of dexmedetomidine and ketamine (ketodex), administered intranasally (IN), could provide adequate sedation for closed reductions while avoiding the need for IV insertion. However, there is uncertainty about the optimal combination dose for the two agents and whether it can provide adequate sedation for closed reductions. The Intranasal Dexmedetomidine Plus Ketamine for Procedural Sedation (Ketodex) study is a Bayesian phase II/III, non-inferiority trial in children undergoing PSA for closed reductions that aims to address both these research questions. This article presents in detail the statistical analysis plan for the Ketodex trial and was submitted before the outcomes of the trial were available for analysis. METHODS/DESIGN: The Ketodex trial is a multicenter, four-armed, randomized, double-dummy controlled, Bayesian response adaptive dose finding, non-inferiority, phase II/III trial designed to determine (i) whether IN ketodex is non-inferior to IV ketamine for adequate sedation in children undergoing a closed reduction of a fracture or dislocation in a pediatric emergency department and (ii) the combination dose for IN ketodex that provides optimal sedation. Adequate sedation will be primarily measured using the Pediatric Sedation State Scale. As secondary outcomes, the Ketodex trial will compare the length of stay in the emergency department, time to wakening, and adverse events between study arms. DISCUSSION: The Ketodex trial will provide evidence on the optimal dose for, and effectiveness of, IN ketodex as an alternative to IV ketamine providing sedation for patients undergoing a closed reduction. The data from the Ketodex trial will be analyzed from a Bayesian perspective according to this statistical analysis plan. This will reduce the risk of producing data-driven results introducing bias in our reported outcomes. TRIAL REGISTRATION: ClinicalTrials.gov NCT04195256 . Registered on December 11, 2019.

Non-steroidal or opioid analgesia use for children with musculoskeletal injuries (the No OUCH study): statistical analysis plan.

BACKGROUND: Pediatric musculoskeletal injuries cause moderate to severe pain, which should ideally be addressed upon arrival to the emergency department (ED). Despite extensive research in ED-based pediatric pain treatment, recent studies confirm that pain management in this setting remains suboptimal. The No OUCH study consist of two complementary, randomized, placebo-controlled trials that will run simultaneously for patients presenting to the ED with an acute limb injury and a self-reported pain score of at least 5/10, measured via a verbal numerical rating scale (vNRS). Caregiver/parent choice will determine whether patients are randomized to the two-arm or three-arm trial. In the two-arm trial, patients will be randomized to receive either ibuprofen alone or ibuprofen in combination with acetaminophen. In the three-arm trial, patients can also be randomized to a third arm where they would receive ibuprofen in combination with hydromorphone. This article details the statistical analysis plan for the No OUCH study and was submitted before the trial outcomes were available for analysis. METHODS/DESIGN: The primary endpoint of the No OUCH study is self-reported pain at 60 min, recorded using a vNRS. The principal safety outcome is the presence of any adverse event related to study drug administration. Secondary effectiveness endpoints include pain measurements using the Faces Pain Scale-Revised and the visual analog scale, time to effective analgesia, requirement of a rescue analgesic, missed fractures, and observed pain reduction using different definitions of successful analgesia. Secondary safety outcomes include sedation measured using the Ramsay Sedation Score and serious adverse events. Finally, the No OUCH study investigates the reasons given by the caregiver for selecting the two-arm (Non-Opioid) or three-arm (Opioid) trial, caregiver satisfaction, physician preferences for analgesics, and caregiver comfort with at-home pain management. DISCUSSION: The No OUCH study will inform the relative effectiveness of acetaminophen and hydromorphone, in combination with ibuprofen, and ibuprofen alone as analgesic agents for patients presenting to the ED with an acute musculoskeletal injury. The data from these trials will be analyzed in accordance with this statistical analysis plan. This will reduce the risk of producing data-driven results and bias in our reported outcomes. TRIAL REGISTRATION: ClinicalTrials.gov NCT03767933 . Registered on December 7, 2018.

A pragmatic randomized controlled trial of multi-dose oral ondansetron for pediatric gastroenteritis (the DOSE-AGE study): statistical analysis plan.

BACKGROUND: Acute gastroenteritis is a leading cause of emergency department visits and hospitalizations among children in North America. Oral-rehydration therapy is recommended for children with mild-to-moderate dehydration, but children who present with vomiting are frequently offered intravenous rehydration in the emergency department (ED). Recent studies have demonstrated that the anti-emetic ondansetron can reduce vomiting, intravenous rehydration, and hospitalization when administered in the ED to children with dehydration. However, there is little evidence of additional benefit from prescribing ondansetron beyond the initial ED dose. Moreover, repeat dosing may increase the frequency of diarrhea. Despite the lack of evidence and potential adverse side effects, many physicians across North America provide multiple doses of ondansetron to be taken following ED disposition. Thus, the Multi-Dose Oral Ondansetron for Pediatric Gastroenteritis (DOSE-AGE) trial will evaluate the effectiveness of prescribing multiple doses of ondansetron to treat acute gastroenteritis-associated vomiting. This article specifies the statistical analysis plan (SAP) for the DOSE-AGE trial and was submitted before the outcomes of the study were available for analysis. METHODS/DESIGN: The DOSE-AGE study is a phase III, 6-center, placebo-controlled, double-blind, parallel design randomized controlled trial designed to determine whether participants who are prescribed multiple doses of oral ondansetron to administer, as needed, following their ED visit have a lower incidence of experiencing moderate-to-severe gastroenteritis, as measured by the Modified Vesikari Scale score, compared with a placebo. To assess safety, the DOSE-AGE trial will investigate the frequency and maximum number of diarrheal episodes following ED disposition, and the occurrence of palpitations, pre-syncope/syncope, chest pain, arrhythmias, and serious adverse events. For the secondary outcomes, the DOSE-AGE trial will investigate the individual elements of the Modified Vesikari Scale score and caregiver satisfaction with the therapy. DISCUSSION: The DOSE-AGE trial will provide evidence on the effectiveness of multiple doses of oral ondansetron, taken as needed, following an initial ED dose in children with acute gastroenteritis-associated vomiting. The data from the DOSE-AGE trial will be analyzed using this SAP. This will reduce the risk of producing data-driven results and bias in our reported outcomes. The DOSE-AGE study was registered on ClinicalTrials.gov on February 22, 2019. TRIAL REGISTRATION: ClinicalTrials.gov NCT03851835 . Registered on 22 February 2019.

Determining a Bayesian predictive power stopping rule for futility in a non-inferiority trial with binary outcomes.

BACKGROUND/AIMS: Non-inferiority trials investigate whether a novel intervention, which typically has other benefits (i.e., cheaper or safer), has similar clinical effectiveness to currently available treatments. In situations where interim evidence in a non-inferiority trial suggests that the novel treatment is truly inferior, ethical concerns with continuing randomisation to the "inferior" intervention are raised. Thus, if interim data indicate that concluding non-inferiority at the end of the trial is unlikely, stopping for futility should be considered. To date, limited examples are available to guide the development of stopping rules for non-inferiority trials. METHODS: We used a Bayesian predictive power approach to develop a stopping rule for futility for a trial collecting binary outcomes. We evaluated the frequentist operating characteristics of the stopping rule to ensure control of the Type I and Type II error. Our case study is the Intranasal Ketamine for Procedural Sedation trial (INK trial), a non-inferiority trial designed to assess the sedative properties of ketamine administered using two alternative routes. RESULTS: We considered implementing our stopping rule after the INK trial enrols 140 patients out of 560. The trial would be stopped if 12 more patients experience a failure on the novel treatment compared to standard care. This trial has a type I error rate of 2.2% and a power of 80%. CONCLUSIONS: Stopping for futility in non-inferiority trials reduces exposure to ineffective treatments and preserves resources for alternative research questions. Futility stopping rules based on Bayesian predictive power are easy to implement and align with trial aims. TRIAL REGISTRATION: ClinicalTrials.gov NCT02828566 July 11, 2016.