Medical University of South Carolina Charleston, South Carolina, United States
Disclosure(s):
Kevin Xue Huang, MD: No financial relationships to disclose
Purpose: The landmark CheckMate 816 trial demonstrated a significant event-free survival benefit of pre-operative immunotherapy and chemotherapy versus pre-operative chemotherapy alone in the stage IB to IIIA non-small cell lung cancer (NSCLC) population [1]. We aimed to evaluate the pragmatic implementation of these findings in the stage IB NSCLC population. Methods: Stage IB NSCLC patients who underwent definitive surgery from 2004 to 2020 were isolated in the National Cancer Database and separated into a pre-trial (2004-2016), peri-trial (2017-2019), and post-trial period (2020). This database captures approximately 70% of all new cancer diagnoses in the United States of America [2]. We used the Cochran-Armitage test to determine if there was any significant trend in the use of immunotherapy over time. Chi-square analysis and logistic regression were used to determine if there was any statistically significant association with the receipt of immunotherapy and age, sex, race, comorbidities, insurance, education, income, population size, distance to hospital, hospital type, hospital volume, tumor size, tumor histology, and lympho-vascular invasion. A fixed effects model was built to account for clustering at the facility level. We estimated overall survival using the Kaplan-Meier method and compared the curves using the log-rank test. Results: 34,848 stage IB NSCLC patients underwent definitive surgical treatment during the study period, of which 323 received immunotherapy. There was a statistically significant positive trend in use of immunotherapy in patients undergoing definitive surgical treatment (p < 0.0001). However, immunotherapy was still quite underutilized as only 0.2% (61/25,761) of patients received immunotherapy in the pre-trial era, 2.7% (196/7,245) in the peri-trial era, and 3.6% (66/1,842) in the post-trial era. There was a statistically significant increase in likelihood of immunotherapy administration with increasing distance from hospital (p < 0.01), Academic/Research hospital (p < 0.0004), higher hospital volume (p < 0.03), adenocarcinoma histology (p < 0.02), and presence of lympho-vascular invasion (p < 0.0001) as well as a significant decrease in likelihood with increasing age (p < 0.05). A multivariable logistic regression revealed that age (OR=0.9887 per year, 95%CI=0.9776 to 0.9998, p< 0.05) and histology other than squamous cell carcinoma or adenocarcinoma (OR=0.3834, 95%CI=0.1841 to 0.7984, p< 0.01) were the independently negatively associated with use of immunotherapy. In the fixed effects model, only the facility type and histology were significantly associated with the use of immunotherapy. There was a significant difference in overall survival with the receipt of immunotherapy in addition to surgery as shown in the figure (p=0.038). Conclusion: Overall, immunotherapy appears underused in this cohort, despite the results from Checkmate 816. Increasing age was negatively associated with the use of immunotherapy whereas increasing distance to the hospital, Academic/Research hospital type, higher hospital volume, adenocarcinoma histology, and presence of lympho-vascular invasion were all positively associated. However, after controlling for each facility only facility type and histology were significantly associated with use of immunotherapy suggesting that much of the variation in use of immunotherapy can be attributed to interfacility differences. Further investigation is needed to understand the reasons for poor dissemination of immunotherapy use in the stage IB NSCLC population.
Identify the source of the funding for this research project: There was no funding for this research project
