Abstract
PURPOSE/OBJECTIVE(S)
Immunotherapy (IT) and lymph node irradiation (LNI) are essential cancer treatment modalities. In combination, the optimal sequence to enhance tumor response is poorly understood. While LNI depletes radiosensitive IT-related immune cells, LNI also provides an immunostimulatory effect by shedding tumor neoantigens in the microenvironment. Here we aim to 1) assess IT and LNI sequence patterns among patients with common cancers, 2) evaluate propensity score (PS) matched overall survival (OS) between IT first and LNI first treatments, and 3) compare OS by cancer site.
MATERIALS/METHODS
A PS matched retrospective cohort study was conducted using the National Cancer Database. The primary endpoint was OS (time from diagnosis to death). Patients age ≥18 years with breast, GI (pancreas, colorectal, liver), GU (prostate, kidney, bladder), lung, lymphoma, melanoma, and oral cavity cancers were included. All patients underwent both IT and LNI and those with incomplete treatment timing were excluded. IT followed by LNI (IT first) was compared to LNI followed by IT (LNI first). PS were performed with 1:1 matching and a standard mean difference cutoff of 0.1 for covariate balancing. PS matched age, stage, comorbidity index, facility type, and upfront lymph node surgery. Unadjusted Kaplan-Meier (KM) estimates with log-rank tests assessed OS. Multivariable Cox proportional hazard (CPH) models adjusted for patient demographics compared IT first versus LNI first with adjusted hazard ratios (aHR) and 95% confidence intervals (95% CI). Models were stratified by cancer site.
RESULTS
A total of 23,238 patients treated with IT and LNI were included, 88% of patients underwent IT first. Median (interquartile range [IQR]) age and follow-up were 57 (48-66) years and 39 (27-53) months, respectively. Cancers included 74% breast, 8% oral cavity, 6% GI, 5% lymphoma, 4% lung, 2% GU, and 1% melanoma. Median (IQR) weeks to treatment were 7 (4-13) for IT first and 8 (4-17) for LNI first. On unadjusted analysis, OS was significantly inferior with LNI first for overall (p<.0001), breast (p<.0001), GI (p = .004), lymphoma (p = .0003), and oral cavity cancer (p = .005). There were no significant differences in OS for GU, lung, and melanoma. On PS matched adjusted CPH analysis, LNI first had significantly higher risk of death overall (aHR = 3.2, 95% CI = 3.0-3.4), compared to IT first. PS matched cancer stratified analyses found OS was significantly inferior with LNI first for breast (aHR = 1.5, 95% CI = 1.2-1.8), GI (aHR = 1.2, 95% CI = 1.1-1.4), GU (aHR = 1.3, 95% CI = 1.0-1.8), lymphoma (aHR = 1.8, 95CI = 1.3-2.6), and oral cavity cancer (aOR = 1.3, 95% CI = 1.1-1.5).
CONCLUSION
PS matched analyses revealed superior OS for patients receiving IT first then LNI for breast, GI, GU, lymphoma, and oral cavity cancers. These findings suggest the importance of an intact immune system prior to IT. Future prospective studies are warranted to validate these findings.
Collapse