Brain Imaging Surveillance for Patients With MetastaticBreast Cancer: A Randomized Clinical Trial Is Requiredto Guide Practice

The topic of brain imaging surveillance among patients with metastatic breast cancer (MBC) has been a matter of debate over the past two decades, without the benefit of high-quality clinical evidence to provide resolution. Accordingly, international guidelines provide conflicting recommendations. For example, as summarized in Table 1, the advanced breast cancer (ABC) guidelines version 6 and 71 and current National Comprehensive Cancer Network Breast Cancer guidelines2 do not recommend surveillance to detect asymptomatic brain metastases, whereas the European Society of Medical Oncology (ESMO) guidelines indicate that subtype-oriented surveillance may be considered for certain subtypes based on a high incidence of brain metastases among patients with human epidermal growth factor receptor 2–positive (HER21) and triple-negative breast cancer (TNBC),3 and the ASCO guidelines indicate that screening can be considered based on shared decision making with patients.4
Among patients with metastatic lung cancer and melanoma, for whom the lifetime incidence of brain metastases is up to 40% and 80%, respectively, brain imaging surveillance has been adopted into clinical practice guidelines in the absence of randomized clinical trial data. Among patients with metastatic HER21 or TNBC, 33%-50% will develop brain metastases during their lifetime5
; this is much higher than the approximately 15% lifetime risk of brain metastases generally reported for patients with hormone receptor–positive (HR1)/HER2-negative MBC.5 The question remains—is a high incidence of brain metastases on its own enough to warrant a generalized brain metastasis surveillance program for patients with MBC, or would one first need to demonstrate net clinical benefit from this practice?
In the past, there was little rationale for considering brain imaging surveillance for patients with MBC because toxicities associated with available local therapies, including surgical resection and/or whole-brain radiotherapy (WBRT), were not felt to justify the possible benefit. Furthermore, brain metastases were felt to presage death, and therefore, earlier detection and treatment would not affect prognosis and only increase patient anxiety and suffering. However, in the era of focal treatment with stereotactic radiosurgery (SRS) and a growing repertoire of CNS-active systemic therapies, the balance between risk versus benefit of surveillance might have changed.6 It is logical to hypothesize that early detection and intervention for asymptomatic brain metastases may be advantageous if modern well-tolerated SRS procedures and/or CNS-penetrant systemic therapies could prevent the development of potentially debilitating and often irreversible neurologic symptoms associated with progression. Indeed, in a retrospective cohort study of patients with breast cancer and non–small cell lung cancer (NSCLC) diagnosed with brain metastases between January 2000 and December 2015, those with MBC (n 5 349) who did not typically undergo brain imaging surveillance had larger and more numerous brain metastases than patients (n 5 659) with NSCLC (who did typically undergo brain imaging surveillance).7 Patients with MBC and brain metastases were also more likely to have neurologic symptoms (75.9% v 60.5%; P < .001) than those with NSCLC.7 Importantly, neurologic death rates were higher among patients with MBC (37.3% v 19.9%, P < .001) with the caveat of similar overall survival rates (1.45 v 1.09 years, respectively, P 5 .06).7 These data support the potential for early diagnosis of brain metastases to positively affect quality of life, especially with the advent of CNS-active agents that not only delay progression of existing metastases but also reduce the incidence of new metastases.8 These agents may even avoid or at least delay the need for local CNS therapies, particularly for patients with HER21