Prognostic Value of Time To Progression vs. Tumor Volume Doubling Time in Patients with Adenoid Cystic Carcinoma Metastatic To The Lung

Purpose/Objective(s) Adenoid cystic carcinoma (ACC) tends to recur despite aggressive local therapy, with the lung being the most common site of distant metastasis. The optimal timing for initiating systemic therapy in patients with metastatic ACC remains a matter of debate. The objective of this study was to retrospectively assess the natural history of ACC lung metastasis and evaluate tumor dynamics using RECIST criteria and tumor volume doubling time (TVDT). Materials/Methods Patients with histologically proven metastatic ACC with at least one pulmonary metastasis ≥5 mm and ≥ two chest CT scans were eligible for this study. Radiology assessment was performed with commercially available software (v6.9.8) from the first scan where the metastasis was detected until the initiation of any intervention or death. Time to progression (TTP) was assessed per RECIST 1.1. To assess whether the tumor growth rate (TGR) was constant, we calculated the correlation coefficient (R) and the coefficient of determination (R2) for all measured lung nodules. TVDT was calculated using the Schwartz formula. Results Out of 581 ACC patients in our database, 75 met eligible criteria. Most had non-solid histology (63%) and had lung as the sole site of metastases upon initial CT scan (89%). TGR was predominantly constant (median R2=0.974). The median TTP was 11.2 months (mos), and the median TVDT was 7.5 mos. Poor survival outcomes were associated with solid histology (2.2 vs. 8.5 years; p=0.002), presence of synchronous extrapulmonary metastases (7.9 vs. 2.7 years; p=0.009), and development of metastasis within the first year after diagnosis for M0 (5.6 vs. 11.4 years; p<0.0001). Given the common eligibility criterion of disease progression within 6 mos for ACC patient enrollment in clinical trials, we established TTP and TVDT cutoffs for assessing both variables as prognostic factors. A shorter TVDT (<6 mos) was significantly linked to poorer OS (HR=0.48; p=0.037), while there was no statistically significant correlation between TTP and OS (HR=1.02; p=0.96). A Cox-regression analysis indicated that TVDT, but not TTP, significantly correlated with OS, suggesting TVDT superior prognostic value in this population. Considering tumor segmentation is not feasible in the clinical practice, we explored whether estimating tumor volume using only one measurement in up to two lung lesions would suffice for calculating TVDT. We found that TVDT calculated using estimated tumor volume significantly correlated with TVDT based on tumor segmentation (p<0.0001). In a multivariate analysis that included this new variable, estimated TVDT emerged as a predictor of survival in lung metastatic ACC (p=0.048). Conclusion Most ACC lung metastases exhibit a constant growth rate. TVDT may serve as a more robust prognostic indicator than TTP per RECIST for lung metastatic ACC. TVDT can be estimated through a single longitudinal measurement in clinical practice.