LAUSR

Brain metastases (BM) affect nearly half of NSCLC patients, yet these patients have historically been excluded from clinical trials due to decreased life expectancy and unclear evidence of blood brain barrier (BBB) penetration of novel agents. Immunotherapy offers an alternative approach by not necessarily requiring the passage of a drug across the BBB but rather the passage of immune cells. Recent mouse models have demonstrated the presence of functional lymphatic vessels lining the dural sinuses that are connected to the deep cervical lymph nodes. BM that demonstrated immune cell infiltration and higher densities of tumor-infiltrating lymphocytes positively correlated to better patient survival [1]. Despite wide adoption of immune checkpoint inhibitors (ICI) into the treatment of NSCLC, there remains limited prospective data regarding intracranial activity of these agents. Pivotal trials leading to the approval of ICI excluded patients with untreated BM or leptomeningeal disease (LMD) and did not include CNS endpoints in analysis [2–4]. Retrospective studies suggest that comparable response rates can be achieved from ICI in intracranial and systemic disease [5]. Additionally, encouraging data from an ongoing trial demonstrated intracranial responses in PDL1 positive NSCLC tumors [6]. Interestingly, there can be discordance between the CNS and systemic response in some patients, which has also been noted between primary and metastatic disease in other studies [7]. Further, early results from a phase II study of pembrolizumab in LMD (multiple histologies—breast, lung, gastric) demonstrated a 3-month overall survival rate of 44% [8]. Mature results from these studies are awaited. To better understand patterns of inclusion and exclusion criteria of immunotherapy trials for NSCLC patients, we reviewed clinical trials from the clinicaltrials.gov registry between October 10, 2017 and March 31, 2018. Search parameters are summarized in Fig. 1. Of the 281 immunotherapy-based trials in this analysis, 53 trials (19%) excluded patients with BM, and 40 trials (14%) did not mention BM in the enrollment criteria. That resulted in 188 trials (67%) which incorporated patients with BM in some capacity. However, of those 188 trials, only 7 trials (4%) included patients with active BM, which was inconsistently defined, but typically meant progressive metastases. 52 trials (28%) allowed enrollment of patients with untreated asymptomatic BM, 84 trials (45%) allowed enrollment of patients with treated BM with documented stability, and 45 trials (24%) allowed enrollment of either treated or untreated asymptomatic BM. Only 7 trials of the 281 included a CNS endpoint. All 7 of the trials that enrolled patients with active BM included intervention with ICI, and 3 trials involved concomitant intracranial radiation through either whole brain radiation or stereotactic radiosurgery. There is current interest regarding the combination of ICI with radiation therapy to create a synergistic effect on both local and distant disease [9]. Patients with LMD are even more frequently excluded from clinical trials, with 46% of trials in our review strictly excluding LMD and only 3% of studies incorporating patients with LMD in any capacity. Consensus groups such as the Response Assessment in Neuro-Oncology Brain Metastases (RANO-BM) Working Group as well as the American Society of Clinical Oncology–Friends of Cancer Research Brain Metastases Working Group have developed guidelines for including BM into clinical trial design, however not specific to immunotherapy [10, 11]. These groups call for clear definitions of endpoints * Saba S. Shaikh [email protected]