Presentations

Abstract

Introduction:

The current standard of care for patients with stage I non-small cell lung cancer (NSCLC) is surgery, but patients may undergo stereotactic body radiotherapy (SBRT) if they cannot have surgery due to significant comorbidities or age. However, the recurrence rate for those treated with SBRT is 30-40% compared to a 15-20% rate for those treated with surgery. Clinical management is also challenging due to variations in tumor histopathology and behavior. Computer Aided Nodule Analysis and Risk Yield (CANARY), a quantitative CT analysis tool, allows for non-invasive tumor assessment by grouping tumors into "good", "intermediate", and "poor" risk stratifications. While CANARY has been studied in stage I NSCLC surgically- resected cases, it has not been analyzed in SBRT-treated NSCLC. The aims of this project were to determine the clinical factors associated with recurrence in stage I NSCLC, and to utilize CANARY to analyze SBRT-treated NSCLCs with recurrence.

Methods:

This was a retrospective cohort study of 176 patients who received SBRT at Inova from 2016 to 2022. Inclusion criteria was a diagnosis of stage I NSCLC at the time of SBRT. Exclusion criteria were: incomplete SBRT; lung metastases from other cancers; and/or a diagnosis other than stage I NSCLC. The primary outcome measured was recurrence. Data was collected on demographics, comorbidities, tumor characteristics, SBRT-specific factors, and recurrence features. Statistical analysis was performed via univariate analysis with Chi-square and Kaplan-Meier. CANARY analysis was performed for those who met the inclusion criteria and had available image annotations, grouping the tumors into "good", "intermediate", or "poor" prognostic risk groups.

Results:

Of the 176 patients screened, 92 met the inclusion criteria and were eligible. Of these 92 patients, 27 had recurrent stage I NSCLC (recurrence rate: 29.3%). 84 of the 176 patients were not eligible due to incomplete SBRT (2), lung metastasis from another cancer (48), or not being stage I at the time of treatment (34). Gender was significantly associated with recurrence (p = 0.037), with recurrence among males greater than females. Smoking pack-years did not meet statistical significance (p = 0.084), with a majority of those with recurrence smoking more than 31-60 pack-years. As part of recurrence-free survival (RFS), the median follow-up time was 1.8 years, and three-year RFS was 61.8%. Of the 92 patients, 70 had available CANARY annotations grouped into: good (7, 0 of whom had recurrence); intermediate (9, 1 of whom had recurrence); and poor (54, 9 of whom had recurrence). Three-year RFS was calculated for: full cohort (56.8%); good (100%); intermediate (88.9%); and poor (48.2%). Median follow-up was 1.9 years. While each risk group was not statistically significant from each other (p = 0.1), there was statistical significance (p = 0.04) when grouping good and intermediate risk vs. poor risk.

Conclusions:

Our findings indicate that gender is a clinical factor associated with recurrence, with males having greater incidence of recurrence than females. While the amount of smoking pack-years did not reach statistical significance, it may be an important factor to consider when evaluating recurrence. CANARY analysis did not reveal significant difference in RFS when comparing the good, intermediate, and poor risk groups against each other. However, there was statistical significance (p = 0.04) when comparing RFS of both good and intermediate risk groups against the poor risk group. With these results, the goals for future studies would be to: apply CANARY and other clinico-radiographic factors (i.e. tumor volume, PET maximum standardized uptake value) in prognostication of stage I lung cancers; and to use these factors to update current surveillance protocols. Ultimately, we hope to utilize these results to create a clinico-radiographic model to aid in the current management of lung cancer.