Resident New York University New York, New York, United States
Introduction: Traditional prognostication of metastatic cancer is based on static measures, such as age, performance status, and current disease burden. But cancer is a heterogeneous disease driven by a complex interplay of histology, mutations, and unique patient factors. Dynamic measures of tumor changes over time may offer a better window to characterize an individual’s disease.
Methods: Using NYUMets, our large, open dataset comprised of 8366 brain scans with associated clinical data, 550 brain metastases patients treated with serial stereotactic radiosurgery (1572 interventions) were evaluated to assess rates of influx of tumors and tumor volume changes over time. These dynamic characteristics were evaluated with other patient and tumor variables using univariable and multivariable Cox proportional hazards models.
Results: For the shortest time interval available (median 5.4 months, IQR 2.5-10.8 months), the median rate of influx of new tumors was 0.43 tumors per month (IQR 0.16-1.7) and median rate of change in volume per tumor per month was 0.06 cc (IQR 0.01-0.30 cc). While several variables demonstrated significant association with survival on univariable analysis, including initial number of tumors, initial total volume, age, and KPS, dynamic measures had the largest hazard effects, with HR 1.27 per doubling rate of tumor influx (95% CI 1.21-1.33, p< 0.001) and HR 1.17 per doubling rate of volume per tumor (95% CI 1.12-1.22, p< 0.001). On multivariable analysis, besides the effect from specific primary histologies (breast, GI), only the rate of tumor influx had a significant effect on survival (HR 1.27 per doubling rate, 95% CI 1.18-1.38, p< 0.001).
Conclusion : Dynamic measures of metastatic disease offer a robust window into an individual’s cancer and could provide an improved method for monitoring, evaluating treatments, and prognosticating. The rate of influx of new intracranial tumors especially has a strong association with survival, possibly representing a most sensitive measure of active systemic disease.
