ZKZbigniew Kadziola

Zbigniew Kadziola

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Zbigniew Kadziola has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Next-Generation Sequencing–Based Testing Among Patients With Advanced or Metastatic Nonsquamous Non–Small Cell Lung Cancer in the United States: Predictive Modeling Using Machine Learning Methods

    by Alan James Michael Brnabic, Ilya Lipkovich, Zbigniew Kadziola, Dan He, Peter M Krein, Lisa M Hess

    Background: Next-generation sequencing (NGS) has become a cornerstone of treatment for lung cancer and is recommended in current treatment guidelines for patients with advanced or metastatic disease. Objective: This study was designed to use machine learning methods to determine demographic and clinical characteristics of patients with advanced or metastatic non–small cell lung cancer (NSCLC) that may predict likelihood of receiving NGS-based testing (ever vs never NGS-tested) as well as likelihood of timing of testing (early vs late NGS-tested). Methods: Deidentified patient-level data were analyzed in this study from a real-world cohort of patients with advanced or metastatic NSCLC in the United States. Patients with nonsquamous disease, who received systemic therapy for NSCLC, and had at least 3 months of follow-up data for analysis were included in this study. Three strategies, logistic regression models, penalized logistic regression using least absolute shrinkage and selection operator penalty, and extreme gradient boosting with classification trees as base learners, were used to identify predictors of ever versus never and early versus late NGS testing. Data were split into D1 (training+validation; 80%) and D2 (testing; 20%) sets; the 3 strategies were evaluated by comparing their performance on multiple m=1000 splits in the training (70%) and validation data (30%) within the D1 set. The final model was selected by evaluating performance using the area under the receiver operating curve while taking into account considerations of simplicity and clinical interpretability. Performance was re-estimated using the test data D2. Results: A total of 13,425 met the criteria for the ever NGS-tested, and 17,982 were included in the never NGS-tested group. Performance metrics showed the area under the receiver operating curve evaluated from validation data was similar across all models (77%-84%). Among those in the ever NGS-tested group, 84.08% (n=11,289) were early NGS-tested, and 15.91% (n=2136) late NGS-tested. Factors associated with both ever having NGS testing as well as early NGS testing included later year of NSCLC diagnosis, no smoking history, and evidence of programmed death ligand 1 testing (all P<.05). Factors associated with a greater chance of never receiving NGS testing included older age, lower performance status, Black race, higher number of single-gene tests, public insurance, and treatment in a geography with Molecular Diagnostics Services Program adoption (all P<.05). Conclusions: Predictors of ever versus never as well as early versus late NGS testing in the setting of advanced or metastatic NSCLC were consistent across machine learning methods in this study, demonstrating the ability of these models to identify factors that may predict NGS-based testing. There is a need to ensure that patients regardless of age, race, insurance status, and geography (factors associated with lower odds of receiving NGS testing in this study) are provided with equitable access to NGS-based testing.


  • Open Access

    Next-Generation Sequencing–Based Testing Among Patients With Advanced or Metastatic Nonsquamous Non–Small Cell Lung Cancer in the United States: Predictive Modeling Using Machine Learning Methods (Preprint)

    by Alan James Michael Brnabic, Ilya Lipkovich, Zbigniew Kadziola, Dan He, Peter M Krein, Lisa M Hess

    BACKGROUND Next-generation sequencing (NGS) has become a cornerstone of treatment for lung cancer and is recommended in current treatment guidelines for patients with advanced or metastatic disease. OBJECTIVE This study was designed to use machine learning methods to determine demographic and clinical characteristics of patients with advanced or metastatic non–small cell lung cancer (NSCLC) that may predict likelihood of receiving NGS-based testing (ever vs never NGS-tested) as well as likelihood of timing of testing (early vs late NGS-tested). METHODS Deidentified patient-level data were analyzed in this study from a real-world cohort of patients with advanced or metastatic NSCLC in the United States. Patients with nonsquamous disease, who received systemic therapy for NSCLC, and had at least 3 months of follow-up data for analysis were included in this study. Three strategies, logistic regression models, penalized logistic regression using least absolute shrinkage and selection operator penalty, and extreme gradient boosting with classification trees as base learners, were used to identify predictors of ever versus never and early versus late NGS testing. Data were split into D1 (training+validation; 80%) and D2 (testing; 20%) sets; the 3 strategies were evaluated by comparing their performance on multiple m=1000 splits in the training (70%) and validation data (30%) within the D1 set. The final model was selected by evaluating performance using the area under the receiver operating curve while taking into account considerations of simplicity and clinical interpretability. Performance was re-estimated using the test data D2. RESULTS A total of 13,425 met the criteria for the ever NGS-tested, and 17,982 were included in the never NGS-tested group. Performance metrics showed the area under the receiver operating curve evaluated from validation data was similar across all models (77%-84%). Among those in the ever NGS-tested group, 84.08% (n=11,289) were early NGS-tested, and 15.91% (n=2136) late NGS-tested. Factors associated with both ever having NGS testing as well as early NGS testing included later year of NSCLC diagnosis, no smoking history, and evidence of programmed death ligand 1 testing (all P<.05). Factors associated with a greater chance of never receiving NGS testing included older age, lower performance status, Black race, higher number of single-gene tests, public insurance, and treatment in a geography with Molecular Diagnostics Services Program adoption (all P<.05). CONCLUSIONS Predictors of ever versus never as well as early versus late NGS testing in the setting of advanced or metastatic NSCLC were consistent across machine learning methods in this study, demonstrating the ability of these models to identify factors that may predict NGS-based testing. There is a need to ensure that patients regardless of age, race, insurance status, and geography (factors associated with lower odds of receiving NGS testing in this study) are provided with equitable access to NGS-based testing.

  • Open Access

    Predicting optimal treatment regimens for patients with HR+/HER2- breast cancer using machine learning based on electronic health records

    by Zhanglin Lin Cui, Zbigniew Kadziola, Ilya Lipkovich, Douglas E Faries, Kristin M Sheffield, Gebra Cuyun Carter

    Aim: To predict optimal treatments maximizing overall survival (OS) and time to treatment discontinuation (TTD) for patients with metastatic breast cancer (MBC) using machine learning methods on electronic health records. Patients/methods: Adult females with HR+/HER2- MBC on first- or second-line systemic therapy were eligible. Random survival forest (RSF) models were used to predict optimal regimen classes for individual patients and each line of therapy based on baseline characteristics. Results: RSF models suggested greater use of CDK4 & 6 inhibitor-based therapies may maximize OS and TTD. RSF-predicted optimal treatments demonstrated longer OS and TTD compared with nonoptimal treatments across line of therapy (hazard ratios = 0.44∼0.79). Conclusion: RSF may help inform optimal treatment choices and improve outcomes for patients with HR+/HER2- MBC.


  • Open Access

    Does participation in a weight control program also improve clinical and functional outcomes for Chinese patients with schizophrenia treated with olanzapine?

    by William Montgomery, Tamas Treuer, Wenyu Ye, Hai Bo Xue, Sheng Hu Wu, Li Liu, Zbigniew Kadziola, Michael D Stensland, Haya Ascher-Svanum

    Objectives: This study examined whether participation in a weight control program (WCP) by patients with schizophrenia treated with olanzapine was also associated with improvements in clinical and functional outcomes. Methods: A post-hoc analysis was conducted using data from the Chinese subgroup (n=330) of a multi-country, 6-month, prospective, observational study of outpatients with schizophrenia who initiated or switched to oral olanzapine. At study entry and monthly visits, participants were assessed with the Clinical Global Impression of Severity, and measures of patient insight, social activities, and work impairment. The primary comparison was between the 153 patients who participated in a WCP at study entry (n=93) or during the study (n=60) and the 177 patients who did not participate in a weight control program (non-WCP). Mixed Models for Repeated Measures with baseline covariates were used to compare outcomes over time. Kaplan–Meier survival analysis was used to assess time to response. Results: Participants had a mean age of 29.0 years and 29.3 years, and 51.0% and 57.6% were female for WCP and non-WCP groups, respectively. Average initiated daily dose for olanzapine was 9.5±5.4 mg. WCP participants gained less weight than non-participants (3.9 kg vs 4.9 kg, P=0.03) and showed statistically significant better clinical and functional outcomes: greater improvement in illness severity (–2.8 vs –2.1, PPPPP Conclusion: Participation in a WCP may not only lower the risk of clinically significant weight gain in olanzapine-treated patients, but may also be associated with additional clinical and functional benefits.