Aditya Juloori, Jacob A. Miller, Shireen Parsai, Rupesh Kotecha, Manmeet S. Ahluwalia, Alireza M. Mohammadi, Erin S. Murphy, John H. Suh, Gene H. Barnett, Jennifer S. Yu, Michael A. Vogelbaum, Brian Rini, Jorge Garcia, Glen H. Stevens, Lilyana Angelov and Samuel T. Chao
The object of this retrospective study was to investigate the impact of targeted therapies on overall survival (OS), distant intracranial failure, local failure, and radiation necrosis among patients treated with radiation therapy for renal cell carcinoma (RCC) metastases to the brain.
All patients diagnosed with RCC brain metastasis (BM) between 1998 and 2015 at a single institution were included in this study. The primary outcome was OS, and secondary outcomes included local failure, distant intracranial failure, and radiation necrosis. The timing of targeted therapies was recorded. Multivariate Cox proportional-hazards regression was used to model OS, while multivariate competing-risks regression was used to model local failure, distant intracranial failure, and radiation necrosis, with death as a competing risk.
Three hundred seventy-six patients presented with 912 RCC BMs. Median OS was 9.7 months. Consistent with the previously validated diagnosis-specific graded prognostic assessment (DS-GPA) for RCC BM, Karnofsky Performance Status (KPS) and number of BMs were the only factors prognostic for OS. One hundred forty-seven patients (39%) received vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors (TKIs). Median OS was significantly greater among patients receiving TKIs (16.8 vs 7.3 months, p < 0.001). Following multivariate analysis, KPS, number of metastases, and TKI use remained significantly associated with OS.
The crude incidence of local failure was 14.9%, with a 12-month cumulative incidence of 13.4%. TKIs did not significantly decrease the 12-month cumulative incidence of local failure (11.4% vs 14.5%, p = 0.11). Following multivariate analysis, age, number of BMs, and lesion size remained associated with local failure. The 12-month cumulative incidence of radiation necrosis was 8.0%. Use of TKIs within 30 days of SRS was associated with a significantly increased 12-month cumulative incidence of radiation necrosis (10.9% vs 6.4%, p = 0.04).
Use of targeted therapies in patients with RCC BM treated with intracranial SRS was associated with improved OS. However, the use of TKIs within 30 days of SRS increases the rate of radiation necrosis without improving local control or reducing distant intracranial failure. Prospective studies are warranted to determine the optimal timing to reduce the rate of necrosis without detracting from survival.
Jacob A. Miller, Ehsan H. Balagamwala, Lilyana Angelov, John H. Suh, Brian Rini, Jorge A. Garcia, Manmeet Ahluwalia and Samuel T. Chao
Systemic control of metastatic renal cell carcinoma (mRCC) has substantially improved with the development of VEGF, mTOR, and checkpoint inhibitors. The current first-line standard of care is a VEGF tyrosine kinase inhibitor (TKI). In preclinical models, TKIs potentiate the response to radiotherapy. Such improved efficacy may prolong the time to salvage therapies, including whole-brain radiotherapy or second-line systemic therapy.
As the prevalence of mRCC has increased, the utilization of spine stereotactic radiosurgery (SRS) has also increased. However, clinical outcomes following concurrent treatment with SRS and TKIs remain largely undefined. The purpose of this investigation was to determine the safety and efficacy of TKIs when delivered concurrently with SRS. The authors hypothesized that first-line TKIs delivered concurrently with SRS significantly increase local control compared with SRS alone or TKIs alone, without increased toxicity.
A retrospective cohort study of patients undergoing spine SRS for mRCC was conducted. Patients undergoing SRS were divided into 4 cohorts: those receiving concurrent first-line TKI therapy (A), systemic therapy–naïve patients (B), and patients who were undergoing SRS with (C) or without (D) concurrent TKI treatment after failure of first-line therapy. A negative control cohort (E) was also included, consisting of patients with spinal metastases managed with TKIs alone. The primary outcome was 12-month local failure, defined as any in-field radiographic progression. Multivariate competing risks regression was used to determine the independent effect of concurrent first-line TKI therapy upon local failure.
One hundred patients who underwent 151 spine SRS treatments (232 vertebral levels) were included. At the time of SRS, 46% were receiving concurrent TKI therapy. In each SRS cohort, the median prescription dose was 16 Gy in 1 fraction. Patients in Cohort A had the highest burden of epidural disease (96%, p < 0.01).
At 12 months, the cumulative incidence of local failure was 4% in Cohort A, compared with 19%–27% in Cohorts B–D and 57% in Cohort E (p < 0.01). Multivariate competing risks regression demonstrated that concurrent first-line TKI treatment (Cohort A) was independently associated with a local control benefit (HR 0.21, p = 0.04). In contrast, patients treated with TKIs alone (Cohort E) experienced an increased rate of local failure (HR 2.43, p = 0.03). No toxicities of Grade 3 or greater occurred following SRS with concurrent TKI treatment, and the incidence of post-SRS vertebral fracture (overall 21%) and pain flare (overall 17%) were similar across cohorts.
The prognosis for patients with mRCC has significantly improved with TKIs. The present investigation suggests a local control benefit with the addition of concurrent first-line TKI therapy to spine SRS. These results have implications in the oligometastatic setting and support a body of preclinical radiobiological research.