Stereotactic Body Radiation Therapy for Mediastinal and Hilar Lymph Node Metastases: A Single-Institutional Review of Clinical Outcomes

Abstract

Stereotactic body radiation therapy (SBRT) to metastatic mediastinal and hilar lymphadenopathy (mMHL) is challenging due to the proximity of centrally located, thoracic organs-at-risk. As limited data exist on the safety/efficacy of SBRT for mMHL, a retrospective review of clinical outcomes was conducted. We hypothesize that SBRT to mMHL is well tolerated with high rates of local control (LC). Patients were identified from a prospectively maintained SBRT database at a large tertiary cancer center. At our institution, mMHL SBRT is a strategy to delay the need to start or change systemic therapy (ST), and/or to prevent airway/great vessel compression. Eligible patients received SBRT to mMHL between 2014-2019 for the following indications: oligometastases (OM), oligoprogression (OP), or LC of a dominant area of progression (DAP). The primary endpoint was grade 3 or greater (G3+) toxicity (CTCAEv5.0). All G3+ toxicities were reviewed independently by four radiation oncologists, with final scoring and attribution achieved by consensus. Secondary endpoints were cumulative incidence (CIN) of local failure (LF), progression-free survival (PFS), overall survival (OS), and CIN of starting or changing systemic therapy (SCST). Time zero for all endpoints was SBRT completion date. Fifty-two patients (84 lesions treated) were included. Median follow-up was 19.9 months. Common primary cancer sites were kidney (54%), lung (13%), and breast (8%). Indications for SBRT were OP (n = 35; 67%), OM (n = 10; 19%), or DAP (n = 7; 14%). Thirty-two patients (62%) received ST prior to SBRT, and of these, most (n = 29; 91%) temporarily stopped ST before and during SBRT. The majority (n = 31; 60%) received SBRT to a single lymph node metastasis. Median maximum lesion size was 2.0 cm. Of the 84 metastatic lymph nodes, 53 (63%) were mediastinal and 31 (37%) were hilar. Median SBRT dose was 35 Gy (range: 30-50Gy) with a median BED10 of 59.5 Gy (range: 48-100 Gy). All treatments were delivered in 5 fractions, every other day. Seven G3+ toxicities were experienced by six patients (11.5%) and were mostly transient (5/7; 71%). One (1.9%) probable G5 event was found (radiation pneumonitis) in a patient who underwent synchronous SBRT to a parenchymal lung metastasis. The CIN of LF was 9.0% at 2 years. Median PFS was 4.0 months (95% CI: 2.8-7.3) and median OS was 31.7 months (95% CI: 23.8-87.5). The CIN of SCST was 33.2% and 57.1% at 1- and 2-years, respectively. In one of the largest series of patients undergoing SBRT for mMHL, we found that treatment was generally well tolerated and provided excellent LC. A significant proportion of patients did not require a change in ST strategy following SBRT. Most G3+ toxicities were transient; however, given the potential risk for serious toxicity, prospective evaluation of SBRT for mMHL is warranted.