Closed Testing Procedure Research Articles

Evaluating and Updating the IMPACT Model to Predict Outcomes in Two Contemporary North American Traumatic Brain Injury Cohorts.

The International Mission on Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury (IMPACT) model is a widely recognized prognostic model applied after traumatic brain injury (TBI). However, it was developed with patient cohorts that may not reflect modern practice patterns in North America. We analyzed data from two sources: the placebo arm of the phase II double-blinded, multicenter, randomized controlled trial Prehospital Tranexamic Acid for TBI (TXA) cohort and an observational cohort with similar inclusion/exclusion criteria (Predictors of Low-risk Phenotypes after Traumatic Brain Injury Incorporating Proteomic Biomarker Signatures [PROTIPS] cohort). All three versions of the IMPACT model-core, extended, and laboratory-were evaluated for 6-month mortality (Glasgow Outcome Scale Extended [GOSE] = 1) and unfavorable outcomes (GOSE = 1-4). Calibration (intercept and slope) and discrimination (area under the receiver operating characteristic curve [ROC-AUC]) were used to assess model performance. We then compared three model updating methods-recalibration in the large, logistic recalibration, and coefficient update-with the best update method determined by likelihood ratio tests. In our calibration analysis, recalibration improved both intercepts and slopes, indicating more accurate predicted probabilities when recalibration was done. Discriminative performance of the IMPACT models, measured by AUC, showed mortality prediction ROCs between 0.61 and 0.82 for the TXA cohort, with the coefficient updated Lab model achieving the highest at 0.84. Unfavorable outcomes had lower AUCs, ranging from 0.60 to 0.79. Similarly, in the PROTIPS cohort, AUCs for mortality ranged from 0.75 to 0.82, with the coefficient updated Lab model also showing superior performance (AUC 0.84). Unfavorable outcomes in this cohort presented AUCs from 0.67 to 0.73, consistently lower than mortality predictions. The closed testing procedure using likelihood ratio tests consistently identified the coefficient update model as superior, outperforming the original and recalibrated models across all cohorts. In our comprehensive evaluation of the IMPACT model, the coefficient updated models were the best performing across all cohorts through a structured closed testing procedure. Thus, standardization of model updating procedures is needed to reproducibly determine the best performing versions of IMPACT that reflect the specific characteristics of a dataset.

Combining Partial True Discovery Guarantee Procedures.

Closed testing has recently been shown to be optimal for simultaneous true discovery proportion control. It is, however, challenging to construct true discovery guarantee procedures in such a way that it focuses power on some feature sets chosen by users based on their specific interest or expertise. We propose a procedure that allows users to target power on prespecified feature sets, that is, "focus sets." Still, the method also allows inference for feature sets chosen post hoc, that is, "nonfocus sets," for which we deduce a true discovery lower confidence bound by interpolation. Our procedure is built from partial true discovery guarantee procedures combined with Holm's procedure and is a conservative shortcut to the closed testing procedure. A simulation study confirms that the statistical power of our method is relatively high for focus sets, at the cost of power for nonfocus sets, as desired. In addition, we investigate its power property for sets with specific structures, for example, trees and directed acyclic graphs. We also compare our method with AdaFilter in the context of replicability analysis. The application of our method is illustrated with a gene ontology analysis in gene expressiondata.

Extended-release ketamine tablets for treatment-resistant depression: a randomized placebo-controlled phase 2 trial

Ketamine has rapid-onset antidepressant activity in patients with treatment-resistant major depression (TRD). The safety and tolerability of racemic ketamine may be improved if given orally, as an extended-release tablet (R-107), compared with other routes of administration. In this phase 2 multicenter clinical trial, male and female adult patients with TRD and Montgomery–Asberg Depression Rating Scale (MADRS) scores ≥20 received open-label R-107 tablets 120 mg per day for 5 days and were assessed on day 8 (enrichment phase). On day 8, responders (MADRS scores ≤12 and reduction ≥50%) were randomized on a 1:1:1:1:1 basis to receive double-blind R-107 doses of 30, 60, 120 or 180 mg, or placebo, twice weekly for a further 12 weeks. Nonresponders on day 8 exited the study. The primary endpoint was least square mean change in MADRS for each active treatment compared with placebo at 13 weeks, starting with the 180 mg dose, using a fixed sequence step-down closed test procedure. Between May 2019 and August 2021, 329 individuals were screened for eligibility, 231 entered the open-label enrichment phase (days 1–8) and 168 responders were randomized to double-blind treatment. The primary objective was met; the least square mean difference of MADRS score for the 180 mg tablet group and placebo was −6.1 (95% confidence interval 1.0 to 11.16, P = 0.019) at 13 weeks. Relapse rates during double-blind treatment showed a dose response from 70.6% for placebo to 42.9% for 180 mg. Tolerability was excellent, with no changes in blood pressure, minimal reports of sedation and minimal dissociation. The most common adverse events were headache, dizziness and anxiety. During the randomized phase of the study, most patient dosing occurred at home. R-107 tablets were effective, safe and well tolerated in a patient population with TRD, enriched for initial response to R-107 tablets. ClinicalTrials.gov registration: ACTRN12618001042235.

Assessing the ability of the Cancer and Aging Research Group tool to predict chemotherapy toxicity in older Japanese patients: A prospective observational study

IntroductionThe Cancer and Aging Research Group (CARG) prediction tool was designed in the United States to predict grade ≥ 3 chemotherapy-related adverse events (CRAE) in older patients. However, its usefulness among Japanese people, who have different sensitivities to anticancer drugs and life expectancy, remains unknown. We aimed to prospectively evaluate the utility of the CARG tool for predicting severe CRAE in older Japanese patients with cancer. Material and MethodsPatients with solid tumors aged 65 years and older who commenced anticancer drug regimens from April 2018 to October 2020 were divided into three groups (low, medium, and high-risk) based on their CARG risk scores. Toxicity was prospectively observed by a pharmacist. The primary objective was to evaluate the correlation between the incidence of grade ≥ 3 CRAE and the CARG risk score. The secondary objective was to evaluate hematological and non-hematological toxicities. CRAE incidence was compared among the three groups using a closed testing procedure: (1) Cochran–Armitage test for trend and (2) chi-square test for paired comparison. ResultsThe patients (N = 165) had a median age of 71 years (range: 65–89 years). CRAE in patients divided into low-, medium-, and high-risk groups, based on CARG risk scores, were 39%, 55%, and 82%, respectively (low vs high; p < 0.001, medium vs high; p < 0.01). The incidence of severe hematologic toxicity was 37%, 35%, and 50% in the low-, medium-, and high-risk groups, respectively; the incidence of severe non-hematologic toxicity was 15%, 36%, and 65%, respectively (low vs medium; p < 0.01, low vs high; p < 0.001, and medium vs high; p < 0.01). DiscussionTo our knowledge, this is the first prospective observational study to validate the CARG prediction tool in older Japanese patients with cancer. The CARG risk score may be effective in predicting the development of non-hematologic toxicities. These results should be considered when administering chemotherapy to older Japanese patients with advanced solid tumors.

Abstract PO1-27-05: De-escalated neoadjuvant weekly nab-paclitaxel with trastuzumab and pertuzumab in HER2-positive early breast cancer (HELEN-006):a randomized,phase 3 trial

Abstract Background: Chemotherapy de-escalation is currently under investigation as a treatment approach for HER2-positive early-stage breast cancer(EBC). We conducted a phase 3 clinical trial to evaluate the effectiveness of a weekly regimen of nab-paclitaxel monotherapy in combination with dual HER2 blockade for HER2-positive EBC. Methods: We conducted a multicenter, randomized, open-label Phase 3 trial at six hospitals in China. Eligible participants were women aged 18-70 with locally confirmed, HER2-positive, stage II-III, invasive, operable breast cancer and an Eastern Cooperative Oncology Group performance status of 0 or 1. Participants were randomly assigned (1:1) using central block randomization, with block sizes of four, stratified by hormone receptor status, tumor stage, and node status. They were assigned to receive either intravenous nab-paclitaxel (125mg/m2 on days 1, 8, and 15) for six 3-week cycles (group A), or intravenous docetaxel (75mg/m² on day 1) and carboplatin (area under the concentration-time curve 6mg/mL per min on day 1) for six 3-week cycles (group B). Both groups received concurrent trastuzumab (8mg/m2 loading dose, 6mg/m2 maintenance dose on day 1) and pertuzumab (intravenous 840mg loading dose, 420mg maintenance dose on day 1). The primary endpoint was the proportion of patients achieving a pathological complete response(pCR) in the breast and axilla (ypT0/is ypN0) as determined by a local pathologist after surgery, assessed using a modified intention-to-treat analysis. To test for non-inferiority, we used a closed test procedure, considering the nab-paclitaxel group non-inferior to the docetaxel plus carboplatin group if the lower 95% confidence interval(CI) for the odds ratio (OR) was above 0.802, which accounted for a 5.5% non-inferiority margin (10% of the 55% pCR rate for docetaxel plus carboplatin and dual HER2 blockade regimen). We planned to test for superiority only if the non-inferiority test was positive, using an alpha value of 0.05. Safety was assessed in all patients who received the study drug. The trial is registered with ClinicalTrials.gov, with the registration number NCT04547907. Findings: Between September 10, 2020, and March 1, 2023, we randomly assigned 689 women, of whom 669 started treatment (332 with nab-paclitaxel and 337 with docetaxel plus carboplatin). The nab-paclitaxel group had a higher proportion of patients achieving a pCR (220 [66.3%, 95% CI 61.2-71.4]) compared to the docetaxel plus carboplatin group (194 [57.6%, 52.3-62.9]). The OR for pCR was 1.448 (95% CI 1.058-1.981), indicating a statistically significant difference (unadjusted p=0.021). Overall, 228 (34.1%) patients were noted to have at least one grade 3 or 4 event, 100(30.1%) in the nab-paclitaxel group and 128 (38.0%) in the docetaxel plus carboplatin group (P=0·032). The most common grade 3 or 4 events overall were nausea (98[14.6%] of 669 patients), diarrhea (80 [12.0%]), and neuropathy (51 [7.6%]).However, no treatment-associated deaths were identified. Conclusions: In neoadjuvant treatment of HER2-positive EBC, 18 weeks of nab-paclitaxel monotherapy combined with dual HER2 blockade has shown significantly higher pCR rates and improved tolerability compared to the standard docetaxel plus carboplatin and dual HER2 blockade regimen. These findings could potentially reshape preferences for neoadjuvant therapy in HER2-positive EBC. Patient Characteristics pCR by clinical-pathological variables Abbreviation:Nab-PHP,nanoparticle albumin-bound paclitaxel plus trastuzumab and pertuzumab; TCbHP,docetaxel plus carboplatin ,trastuzumab and pertuzumab; HR,hormone receptor; IHC,immunohistochemistry. *estrogen and/or progesterone receptor positive is classified as hormone receptor positive;estrogen and progesterone receptor negative is classified as hormone receptor negative Adverse events Abbreviation:Nab-PHP,nanoparticle albumin-bound paclitaxel plus trastuzumab and pertuzumab; TCbHP,docetaxel plus carboplatin ,trastuzumab and pertuzumab;ALT,alanine aminotransferase;AST,aspartate transaminase. Citation Format: Zhenzhen Liu, Xiuchun Chen, Jianghua Qiao, Jiao Dechuang, Chengzheng Wang, Xianfu Sun, Zhenduo Lu, Lianfang Li, Chongjian Zhang, Min Yan, Ya Wei, Chen Bo, Yueqing Feng, Miao Deng, Mingde Ma. De-escalated neoadjuvant weekly nab-paclitaxel with trastuzumab and pertuzumab in HER2-positive early breast cancer (HELEN-006):a randomized,phase 3 trial [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-27-05.

Closed testing procedure for comparing sizes of normal means based on ordered statistics

The aim of this study is to propose a stepwise multiple comparison procedure for comparing the sizes of normal means. Our stepwise procedure is constructed by modifying Imada (2022)’s closed testing procedure based on ordered statistics. We give numerical results regarding the critical values and power of the test intended to compare our stepwise procedure with the stepwise procedures proposed by Imada (2022). Furthermore, we illustrate our procedure with an example.

Optimal weighted Bonferroni tests and their graphical extensions.

Regulatory guidelines mandate the strong control of the familywise error rate in confirmatory clinical trials with primary and secondary objectives. Bonferroni tests are one of the popular choices for multiple comparison procedures and are building blocks of more advanced procedures. It is usually of interest to find the optimal weighted Bonferroni split for multiple hypotheses. We consider two popular quantities as the optimization objectives, which are the disjunctive power and the conjunctive power. The former is the probability to reject at least one false hypothesis and the latter is the probability to reject all false hypotheses. We investigate the behavior of each of them as a function of different Bonferroni splits, given assumptions about the alternative hypotheses and correlations between test statistics. Under independent tests, unique optimal Bonferroni weights exist; under dependence, optimal Bonferroni weights may not be unique based on a fine grid search. In general, we propose an optimization algorithm based on constrained nonlinear optimization and multiple starting points. The proposed algorithm efficiently identifies optimal Bonferroni weights to maximize the disjunctive or conjunctive power. In addition, we apply the proposed algorithm to graphical approaches, which include many Bonferroni-based multiple comparison procedures. Utilizing the closed testing principle, we adopt a two-step approach to find optimal graphs using the disjunctive power. We also identify a class of closed test procedures that optimize the conjunctive power. We apply the proposed algorithm to a case study to illustrate the utility of optimal graphical approaches that reflect study objectives.

Cluster extent inference revisited: quantification and localisation of brain activity

Abstract Cluster inference based on spatial extent thresholding is a popular analysis method multiple testing in spatial data, and is frequently used for finding activated brain areas in neuroimaging. However, the method has several well-known issues. While powerful for finding regions with some activation, the method as currently defined does not allow any further quantification or localisation of signal. In this paper, we repair this gap. We show that cluster-extent inference can be used (1) to infer the presence of signal in any region of interest and (2) to quantify the percentage of activation in such regions. These additional inferences come for free, i.e. they do not require any further adjustment of the alpha-level of tests, while retaining full family-wise error control. We achieve this extension of the possibilities of cluster inference by embedding the method into a closed testing procedure, and solving the graph-theoretic k-separator problem that results from this embedding. We demonstrate the usefulness of the improved method in a large-scale application to neuroimaging data from the Neurovault database.

External validation of NTCP-models for radiation pneumonitis in lung cancer patients treated with chemoradiotherapy

PurposeNormal tissue complication probability (NTCP) models can be used to estimate the risk of radiation pneumonitis (RP). The aim of this study was to externally validate the most frequently used prediction models for RP, i.e., the QUANTEC and APPELT models, in a large cohort of lung cancer patients treated with IMRT or VMAT. [1–2] Methods and materialsThis prospective cohort study, included lung cancer patients treated between 2013 and 2018. A closed testing procedure was performed to test the need for model updating. To improve model performance, modification or removal of variables was considered. Performance measures included tests for goodness of fit, discrimination, and calibration. ResultsIn this cohort of 612 patients, the incidence of RP ≥ grade 2 was 14.5%. For the QUANTEC-model, recalibration was recommended which resulted in a revised intercept and adjusted regression coefficient (from 0.126 to 0.224) of the mean lung dose (MLD),. The APPELT-model needed revision including model updating with modification and elimination of variables. After revision, the New RP-model included the following predictors (and regression coefficients): MLD (B = 0.250), age (B = 0.049, and smoking status (B = 0.902). The discrimination of the updated APPELT-model was higher compared to the recalibrated QUANTEC-model (AUC: 0.79 vs. 0.73). ConclusionsThis study demonstrated that both the QUANTEC- and APPELT-model needed revision. Next to changes of the intercept and regression coefficients, the APPELT model improved further by model updating and performed better than the recalibrated QUANTEC model. This New RP-model is widely applicable containing non-tumour site specific variables, which can easily be collected.

Blinded sample size recalculation in multiple composite population designs with normal data and baseline adjustments.

The increasing interest in subpopulation analysis has led to the development of various new trial designs and analysis methods in the fields of personalized medicine and targeted therapies. In this paper, subpopulations are defined in terms of an accumulation of disjoint population subsets and will therefore be called composite populations. The proposed trial design is applicable to any set of composite populations, considering normally distributed endpoints and random baseline covariates. Treatment effects for composite populations are tested by combining p-values, calculated on the subset levels, using the inverse normal combination function to generate test statistics for those composite populations while the closed testing procedure accounts for multiple testing. Critical boundaries for intersection hypothesis tests are derived using multivariate normal distributions, reflecting the joint distribution of composite population test statistics given no treatment effect exists. For sample size calculation and sample size, recalculation multivariate normal distributions are derived which describe the joint distribution of composite population test statistics under an assumed alternative hypothesis. Simulations demonstrate the absence of any practical relevant inflation of the type I error rate. The target power after sample size recalculation is typically met or close to beingmet.

Effects of influential points and sample size on the selection and replicability of multivariable fractional polynomial models

BackgroundThe multivariable fractional polynomial (MFP) approach combines variable selection using backward elimination with a function selection procedure (FSP) for fractional polynomial (FP) functions. It is a relatively simple approach which can be easily understood without advanced training in statistical modeling. For continuous variables, a closed test procedure is used to decide between no effect, linear, FP1, or FP2 functions. Influential points (IPs) and small sample sizes can both have a strong impact on a selected function and MFP model.MethodsWe used simulated data with six continuous and four categorical predictors to illustrate approaches which can help to identify IPs with an influence on function selection and the MFP model. Approaches use leave-one or two-out and two related techniques for a multivariable assessment. In eight subsamples, we also investigated the effects of sample size and model replicability, the latter by using three non-overlapping subsamples with the same sample size. For better illustration, a structured profile was used to provide an overview of all analyses conducted.ResultsThe results showed that one or more IPs can drive the functions and models selected. In addition, with a small sample size, MFP was not able to detect some non-linear functions and the selected model differed substantially from the true underlying model. However, when the sample size was relatively large and regression diagnostics were carefully conducted, MFP selected functions or models that were similar to the underlying true model.ConclusionsFor smaller sample size, IPs and low power are important reasons that the MFP approach may not be able to identify underlying functional relationships for continuous variables and selected models might differ substantially from the true model. However, for larger sample sizes, a carefully conducted MFP analysis is often a suitable way to select a multivariable regression model which includes continuous variables. In such a case, MFP can be the preferred approach to derive a multivariable descriptive model.

Permutation-based true discovery guarantee by sum tests

Abstract Sum-based global tests are highly popular in multiple hypothesis testing. In this paper, we propose a general closed testing procedure for sum tests, which provides lower confidence bounds for the proportion of true discoveries (TDPs), simultaneously over all subsets of hypotheses. These simultaneous inferences come for free, i.e., without any adjustment of the α-level, whenever a global test is used. Our method allows for an exploratory approach, as simultaneity ensures control of the TDP even when the subset of interest is selected post hoc. It adapts to the unknown joint distribution of the data through permutation testing. Any sum test may be employed, depending on the desired power properties. We present an iterative shortcut for the closed testing procedure, based on the branch and bound algorithm, which converges to the full closed testing results, often after few iterations; even if it is stopped early, it controls the TDP. We compare the properties of different choices for the sum test through simulations, then we illustrate the feasibility of the method for high-dimensional data on brain imaging and genomics data.

Homeopathy for seasonal allergic rhinitis: rationale, design and methods of the three-armed randomized controlled HOMEOSAR trial

BackgroundPatients with seasonal allergic rhinitis (SAR) frequently use homeopathic therapy. Although there is some evidence that homeopathy may be effective in treating symptoms of SAR, there is a lack of high-quality clinical trials. Therefore, the aim of the homeopathy for SAR (HOMEOSAR) trial is to determine the efficacy of individualized or standardized homeopathic drug treatment compared to placebo regarding rhinitis-related quality of life in patients with SAR.MethodsThis randomized, placebo-controlled, double-blind, three-armed intervention study will be conducted at two university hospital outpatient clinics for complementary and integrative medicine in Berlin and in 12 office-based practices specializing in homeopathic treatment in Germany. A total of 270 patients with clinical symptoms of SAR and positive allergy test to birch and grass pollen will receive homeopathic anamnesis and subsequently be randomized into (a) standardized homeopathic drug treatment with Galphimia Glauca (potency D6), (b) individualized homeopathic drug treatment (D6), or (c) placebo. All three groups can receive on-demand rescue medication as needed. Treatment will consist of two consultations and daily intake of the study medication for 4 weeks during the pollen season. The primary outcome is the mean overall score of the Rhinitis Quality of Life Questionnaire (RQLQ) in weeks 3 and 4, analyzed using analysis of covariance (adjusted for baseline RQLQ overall score and study center). A closed testing procedure will be used to control the overall type I error comparing the 3 treatment groups. Secondary outcomes include the overall RQLQ and its seven domain scores, responder status (decrease in RQLQ overall score of at least 0.5 points compared to the baseline value), use of rescue medication, intensity of total and individual SAR symptoms based on visual analog scale, generic health-related quality of life, safety, utilization of health care resources and associated costs. In addition, a qualitative data analysis is planned.ConclusionThe results of our study will contribute to clarifying the possible therapeutic effects of homeopathic drug treatment for patients with SAR.Trial registrationThis study has been registered in the German Clinical Trial Registry with trial ID DRKS00018081 on June 09, 2020.

Validation of the 18F-FDG PET Image Biomarker Model Predicting Late Xerostomia after Head and Neck Cancer Radiotherapy

<h3>Purpose/Objective(s)</h3> PET image biomarkers can improve the prediction of radiation-induced late xerostomia in head and neck cancer (HNC) patients [1,2]. A PET image biomarker, the 90^th percentile (<i>P90</i>) of the standardized uptake value was identified as a promising predictor for late xerostomia, where HNC patients with lower metabolic activity in the pre-radiation parotid gland (<i>PG</i>) were more at risk of developing late xerostomia after radiation [2]. Addition of <i>P90</i> improved model performance of the reference model based on baseline xerostomia (<i>Xer-base</i>) and <i>PG</i> mean dose. The aim of the current study was to validate the pre-treatment <i>PET-P90 model</i> in an external HNC validation cohort. <h3>Materials/Methods</h3> HNC patients were treated with primary (chemo-)radiation with curative intent. ¹⁸F-FDG PET/CT images were acquired 2 weeks before radiotherapy. The main endpoint was patient-rated moderate-to-severe xerostomia 12 months after radiotherapy which was assessed via the EORTC QLQ-H&N35 questionnaire. The <i>P90</i> and <i>PG</i> mean dose of contralateral <i>PG</i> were extracted from PET images and radiation dose distribution, respectively. We validated the reference model that was based on <i>Xer-base</i> and <i>PG</i> mean dose only, as well as the PET-P90 model, which had the <i>P90</i> as an additional variable. The closed testing procedure was used to test whether the original coefficients (see [2]) of these models needed to be updated or revised in the current external validation cohort. Finally, model performance (AUC, R²) was evaluated and compared for the reference model and the PET-P90 model. Calibration was tested with the Hosmer-Lemeshow test (HL test). <h3>Results</h3> The current external validation cohort consisted 127 patients of which 40% developed moderate-to-severe xerostomia. The closed testing procedure indicated that all original coefficients of the models did not require any update or revision in the current cohort. Compared to reference model (Table 1. AUC_ref=0.67, R²_ref=0.13), the PET-P90 model showed better predictive performance (AUC_P90=0.71, R²_P90=0.18). Calibration was both good for the reference model and the PET-P90 model in current external cohort (HL test: <i>p</i>_ref=0.09, <i>p</i>_P90=0.09). <h3>Conclusion</h3> The PET-P90 model performed well in this external validation cohort, and outperformed the reference model. This contributes to the hypothesis that lower <i>PG</i> metabolic activity is a pre-treatment risk factor for late xerostomia.

The Dutch Initiative for the Model-Based Selection of Patients for Proton Therapy – A Federated Learning IT Infrastructure

<h3>Purpose/Objective(s)</h3> Model-based selection for proton therapy (PT) patients has been implemented since 2018 in the Netherlands. Normal tissue complication probability (NTCP) models translate dose differences in Organs at Risk (OARs) (Δdose) into differences in NTCP (ΔNTCP). With these NTCP models, patients that will benefit most from PT are selected for proton therapy. These models are based on patients treated with modern photon techniques due to a lack of data for proton therapy to base them on, assuming that their performance is similar among patients treated with PT. Our primary objective was to test the accuracy of one of these NTCP models in a proof-of-concept study with the external validation of the grade 3+ dysphagia NTCP model (six months after the treatment). <h3>Materials/Methods</h3> We used data from 320 head and neck (HNC) patients who were treated between 2019-2021 with (chemo)radiotherapy in one of the Dutch PT centers. Semantic Web technologies and the Findable Accessible Interoperable Reusable (FAIR) principles were applied to the variables needed for the computation of the NTCP grade 3+ dysphagia model formula in order to be structured in a machine-readable format for the Personal Health Train (PHT) infrastructure implementation. The PHT enables the secure exchange of statistical algorithms (i.e., "trains") between different hospitals without the exchange of patient data. Implementing the closed testing procedure methodology, the statistical algorithm for the federated NTCP external validation was built in the RStudio software. This algorithm constitutes the statistical "train" that will be distributed among the different PT Dutch centers. <h3>Results</h3> We successfully completed and implemented the statistical analysis-"train" for the external validation procedure, establishing the end-to-end IT-infrastructure needed in our center according to the requirements of the federated learning ProTRAIT (PROton Therapy ReseArch regIsTry) IT-infrastructure. The performance of the original NTCP model for grade 3+ dysphagia (which was adopted as the updated selected model from the closed testing procedure) had reasonable discriminative power in the validation cohort (AUC=0.84, sensitivity=0.69, specificity=0.75). <h3>Conclusion</h3> With this work we externally validated the existing HNC grade 3+ dysphagia NTCP model in our PT center using the Personal Health Train infrastructure (PHT). The next steps will focus on expanding more applications of the different components of the model-based approach such as the model-based clinical evaluation in the PT centers, including NTCP-model development and validation in a privacy-preserving federated manner with data from the other PT centers.

Stepwise multiple testing procedures for the successive comparison of variances

In this paper, stepwise multiple testing procedures are proposed for comparing successive populations in a sequence of several independent normal populations using index parameter variance. The proposed procedures have advantages over the single-step procedures and closed testing procedures available in the existing literature. The proposed stepwise testing procedures, control the family-wise error rate (FWER) strongly, and dramatically improve in power over the relevant single-step procedures. The closed testing procedure, which is step-down in nature and is developed for a testing problem, is very complex in its implementation, and this complexity increases further as the number of successive comparisons increases. The relevant critical constants have been tabulated to facilitate the implementation of the proposed procedures. We also proposed testing procedures for comparing the successive populations in a sequence of two-parametric exponential populations with regards to their scale parameters. In an effort to discern the efficiency of proposed procedures, simulated power comparisons with relevant existing procedures are presented, and the working of the proposed procedures is exemplified by means of two real-life data sets.

Independent validation of a dysphagia dose response model for the selection of head and neck cancer patients to proton therapy.

Background and purposeThe model based approach involves the use of normal tissue complication models for selection of head and neck cancer patients to proton therapy. Our goal was to validate the clinical utility of the related dysphagia model using an independent patient cohort.Materials and MethodsA dataset of 277 head and neck cancer (pharynx and larynx) patients treated with (chemo)radiotherapy between 2019 and 2021 was acquired. For the evaluation of the model discrimination we used statistical metrics such as the sensitivity, specificity and the area under the receiver operating characteristic curve. After the validation we evaluated if the dysphagia model can be improved using the closed testing procedure, the Brier and the Hosmer-Lemeshow score.ResultsThe performance of the original normal tissue complication probability model for dysphagia grade II-IV at 6 months was good (AUC = 0.80). According to the graphical calibration assessment, the original model showed underestimated dysphagia risk predictions. The closed testing procedure indicated that the model had to be updated and selected a revised model with new predictor coefficients as an optimal model. The revised model had also satisfactory discrimination (AUC = 0.83) with improved calibration.ConclusionThe validation of the normal tissue complication probability model for grade II-IV dysphagia was successful in our independent validation cohort. However, the closed testing procedure indicated that the model should be updated with new coefficients.

Air quality in Mexico City during the COVID-19 lockdown possibly decreased COPD exacerbations.

The reduction of air pollution during the #COVID19 lockdown in Mexico City possibly reduced the exacerbation rate in #COPD patients due to biomass and tobacco despite that the self-isolation was not as strict as expected. https://bit.ly/3Iyv98t

P762: NONLINIEAR IMPACT OF CLINICAL FEATURES ON THE SURVIVAL OUTCOME OF MYELODYSPLASTIC SYNDROMES PATIENTS

Background: Myelodysplastic syndromes (MDS) are a group of hematopoietic stem-cell disorders with heterogenous prognosis. The revised international prognostic scoring system (IPSS-R) (Greenberg et al., Blood, 2012) is the most widely used score for prognostication. However, the allocation of IPSS-R risk categories remains imprecise for some patients leading to unsatisfactory therapy decisions. Newer molecular prognostic models (Nazha et al., JCO, 2021 and Bernard et al., ASH, 2021) have added mutation data to improve survival predictions for MDS patients. Aims: Our aims were to investigate the nonlinear association of normalized clinical data in combination with molecular and cytogenetic data and to improve the prognostication potential of IPSS-R for patients with MDS. Methods: A multivariable fractional polynomials (MFP) algorithm was applied on clinical data of 943 MDS patients and nonlinear associations between continuous covariates and survival outcomes were efficiently modeled within Cox regression. Selected transformed variables were picked and the potential overfit was avoided by a closed test procedure for function selection (Royston et al., John Wiley & Sons, 2008). The performance of the model was evaluated in an independent set of 436 MDS patients and the accuracy was assessed using a concordance (c)index and the area under the Receiver Operating Characteristic curve (AUC) at each time point. Mutation data was obtained in a subset of 695 patients using next-generation sequencing, evaluating a panel of 81 genes at the time of diagnosis. Results: An MFP-model consisting of cytogenetics per IPSS-R and WHO subtypes, continuous forms of age and hemoglobin, and non-linear transformations of platelets and bone marrow blast percentage yielded a c-index performance of 0.82 vs 0.78 for IPSS-R and a significant AUC upgrade at all time points in the training set (Figure a). In comparison with the IPSS-R, absolute neutrophil count was not identified as an independent prognostic covariate and common features were treated in a non-linear fashion. Validation of the model in an independent set of 438 patients yielded a higher c-index (0.72 vs 0.68) with an improved AUC over time compared to the IPSS-R (Figure b). Based on MFP risk score, patients were then clustered into 5 different risk categories and compared with IPSS-R classifications. Our data showed no significant difference in time to death between IPSS-R groups for very low risk and low risk or between low risk and intermediate risk (Figure c); however, MFP classification showed 5 significantly different subgroups in predicting time to death for MDS patients (Figure d). We further investigated the effect of molecular data on disease prognostics in a subset of 695 patients sequenced using an 81 gene panel. In addition to the selected clinical covariates, occurrence of TP53, SRSF2, and ZRSR2 mutations were independently associated with lower survival, while the presence of TET2 mutation showed a depression effect on TP53 mutation towards a better outcome. As SF3B1 was not selected in the final model, we further evaluated its potential interactions and identified its correlation with cytogenetics and platelets, which predominated in the MFP model. This might be due to heterogeneity within the data set (e.g., 25% co-occurrence of SF3B1 with TP53 mutation) and needs to be investigated in larger cohorts. Image:Summary/Conclusion: We have established a nonlinear model that yields improved risk assessment compared to IPSS-R criteria.

A unified framework for weighted parametric group sequential design.

Group sequential design (GSD) is widely used in clinical trials in which correlated tests of multiple hypotheses are used. Multiple primary objectives resulting in tests with known correlations include evaluating (1) multiple experimental treatment arms, (2) multiple populations, (3) the combination of multiple arms and multiple populations, or (4) any asymptotically multivariate normal tests. In this paper, we focus on the first three of these and extend the framework of the weighted parametric multiple test procedure from fixed designs with a single analysis per objective to a GSD setting where different objectives may be assessed at the same or different times, each in a group sequential fashion. Pragmatic methods for design and analysis of weighted parametric group sequential design under closed testing procedures are proposed to maintain the strong control of the family-wise Type I error rate when correlations between tests are incorporated. This results in the ability to relax testing bounds compared to designs not fully adjusting for known correlations, increasing power, or allowing decreased sample size. We illustrate the proposed methods using clinical trial examples and conduct a simulation study to evaluate the operatingcharacteristics.