User login
Background
Programmed death-ligand 1 (PD-L1) checkpoint inhibitors revolutionized the treatment of advanced non-small cell lung cancer (aNSCLC) by improving overall survival compared to chemotherapy. PD-L1 biomarker testing is paramount for informing treatment decisions in aNSCLC. Real-world data describing patterns of PD-L1 testing within the Veteran Health Administration (VHA) are limited. This retrospective study seeks to evaluate demographic and clinical factors associated with PD-L1 testing in VHA.
Methods
Veterans diagnosed with aNSCLC from 2019-2022 were identified using VHA’s Corporate Data Warehouse. Wilcoxon Rank Sum and Chi- Square tests measured association between receipt of PD-L1 testing and patient demographic and clinical characteristics at aNSCLC diagnosis. Logistic regression assessed predictors of PD-L1 testing, and subgroup analyses were performed for significant interactions.
Results
Our study included 4575 patients with aNSCLC; 57.0% received PD-L1 testing. The likelihood of PD-L1 testing increased among patients diagnosed with aNSCLC after 2019 vs during 2019 (OR≥1.118, p≤0.035) and in Black vs White patients (OR=1.227, p=0.011). However, the following had decreased likelihood of PD-L1 testing: patients with stage IIIB vs IV cancer (OR=0.683, p=0.004); non vs current/former smokers (OR=0.733, p=0.039); squamous (OR=0.863, p=0.030) or NOS (OR=0.695,p=0.013) vs. adenocarcinoma histology. Interactions were observed between patient residential region and residential rurality (p=0.003), and region and receipt of oncology community care consults (OCCC) (p=0.030). Patients in rural Midwest (OR=0.445,p=0.004) and rural South (OR=0.566, p=0.032) were less likely to receive PD-L1 testing than Metropolitan patients. Across patients with OCCC, Western US patients were more likely to receive PD-L1 testing (OR=1.554, p=0.001) than patients in other regions. However, within Midwestern patients, those without a OCCC were more likely to receive PD-L1 testing (OR=1.724, p< 0.001) than those with a OCCC. High comorbidity index (CCI≥3) is associated with an increased likelihood of PD-L1 testing in a univariable model (OR=1.286 vs. CCI=0,p=0.009), but not in the multivariable model (p=0.278).
Conclusions
We identified demographic and clinical factors, including regional differences in rurality and OCCC patterns, associated with PD-L1 testing. These factors can focus ongoing efforts to improve PD-L1 testing and efforts to be more in line with recommended care.
Background
Programmed death-ligand 1 (PD-L1) checkpoint inhibitors revolutionized the treatment of advanced non-small cell lung cancer (aNSCLC) by improving overall survival compared to chemotherapy. PD-L1 biomarker testing is paramount for informing treatment decisions in aNSCLC. Real-world data describing patterns of PD-L1 testing within the Veteran Health Administration (VHA) are limited. This retrospective study seeks to evaluate demographic and clinical factors associated with PD-L1 testing in VHA.
Methods
Veterans diagnosed with aNSCLC from 2019-2022 were identified using VHA’s Corporate Data Warehouse. Wilcoxon Rank Sum and Chi- Square tests measured association between receipt of PD-L1 testing and patient demographic and clinical characteristics at aNSCLC diagnosis. Logistic regression assessed predictors of PD-L1 testing, and subgroup analyses were performed for significant interactions.
Results
Our study included 4575 patients with aNSCLC; 57.0% received PD-L1 testing. The likelihood of PD-L1 testing increased among patients diagnosed with aNSCLC after 2019 vs during 2019 (OR≥1.118, p≤0.035) and in Black vs White patients (OR=1.227, p=0.011). However, the following had decreased likelihood of PD-L1 testing: patients with stage IIIB vs IV cancer (OR=0.683, p=0.004); non vs current/former smokers (OR=0.733, p=0.039); squamous (OR=0.863, p=0.030) or NOS (OR=0.695,p=0.013) vs. adenocarcinoma histology. Interactions were observed between patient residential region and residential rurality (p=0.003), and region and receipt of oncology community care consults (OCCC) (p=0.030). Patients in rural Midwest (OR=0.445,p=0.004) and rural South (OR=0.566, p=0.032) were less likely to receive PD-L1 testing than Metropolitan patients. Across patients with OCCC, Western US patients were more likely to receive PD-L1 testing (OR=1.554, p=0.001) than patients in other regions. However, within Midwestern patients, those without a OCCC were more likely to receive PD-L1 testing (OR=1.724, p< 0.001) than those with a OCCC. High comorbidity index (CCI≥3) is associated with an increased likelihood of PD-L1 testing in a univariable model (OR=1.286 vs. CCI=0,p=0.009), but not in the multivariable model (p=0.278).
Conclusions
We identified demographic and clinical factors, including regional differences in rurality and OCCC patterns, associated with PD-L1 testing. These factors can focus ongoing efforts to improve PD-L1 testing and efforts to be more in line with recommended care.
Background
Programmed death-ligand 1 (PD-L1) checkpoint inhibitors revolutionized the treatment of advanced non-small cell lung cancer (aNSCLC) by improving overall survival compared to chemotherapy. PD-L1 biomarker testing is paramount for informing treatment decisions in aNSCLC. Real-world data describing patterns of PD-L1 testing within the Veteran Health Administration (VHA) are limited. This retrospective study seeks to evaluate demographic and clinical factors associated with PD-L1 testing in VHA.
Methods
Veterans diagnosed with aNSCLC from 2019-2022 were identified using VHA’s Corporate Data Warehouse. Wilcoxon Rank Sum and Chi- Square tests measured association between receipt of PD-L1 testing and patient demographic and clinical characteristics at aNSCLC diagnosis. Logistic regression assessed predictors of PD-L1 testing, and subgroup analyses were performed for significant interactions.
Results
Our study included 4575 patients with aNSCLC; 57.0% received PD-L1 testing. The likelihood of PD-L1 testing increased among patients diagnosed with aNSCLC after 2019 vs during 2019 (OR≥1.118, p≤0.035) and in Black vs White patients (OR=1.227, p=0.011). However, the following had decreased likelihood of PD-L1 testing: patients with stage IIIB vs IV cancer (OR=0.683, p=0.004); non vs current/former smokers (OR=0.733, p=0.039); squamous (OR=0.863, p=0.030) or NOS (OR=0.695,p=0.013) vs. adenocarcinoma histology. Interactions were observed between patient residential region and residential rurality (p=0.003), and region and receipt of oncology community care consults (OCCC) (p=0.030). Patients in rural Midwest (OR=0.445,p=0.004) and rural South (OR=0.566, p=0.032) were less likely to receive PD-L1 testing than Metropolitan patients. Across patients with OCCC, Western US patients were more likely to receive PD-L1 testing (OR=1.554, p=0.001) than patients in other regions. However, within Midwestern patients, those without a OCCC were more likely to receive PD-L1 testing (OR=1.724, p< 0.001) than those with a OCCC. High comorbidity index (CCI≥3) is associated with an increased likelihood of PD-L1 testing in a univariable model (OR=1.286 vs. CCI=0,p=0.009), but not in the multivariable model (p=0.278).
Conclusions
We identified demographic and clinical factors, including regional differences in rurality and OCCC patterns, associated with PD-L1 testing. These factors can focus ongoing efforts to improve PD-L1 testing and efforts to be more in line with recommended care.