Stereotactic radiosurgery trial for brain metastases from NSCLC ‘urgently needed’
The use of stereotactic radiosurgery followed by epidermal growth factor receptor tyrosine kinase inhibitors resulted in longer OS than other strategies for patients with EGFR–mutated non–small cell lung cancer who developed brain metastases, according to a multi-institutional analysis published in Journal of Clinical Oncology.
This approach may allow patients with NSCLC to avoid the potential neurocognitive sequelae of whole-brain radiotherapy (WBRT), according to the researchers.
“A prospective, multi-institutional randomized trial of stereotactic radiosurgery [SRS] followed by EGFR TKI vs. EGFR TKI followed by SRS at intracranial progression is urgently needed and would help answer the question as to why OS was so significantly different between the groups,” Veronica L.S. Chiang, MD, associate professor of neurosurgery and therapeutic radiology at Yale School of Medicine, told HemOnc Today.
In 2015, more than 158,000 people in the United States died of lung cancer, more than from breast, colon and prostate cancers combined. As many as 40% of patients diagnosed with NSCLC will develop brain metastases during the course of their disease, with a greater risk in those who harbor an EGFR mutation.
Numerous phase 2 trials have investigated the efficacy of EGFR TKIs for brain metastases, but none had evaluated EGFR TKIs followed by SRS or WBRT at intracranial progression compared with SRS or WBRT followed by EGFR TKIs.
“With the rapid development of novel drugs that potentially have better brain penetration than prior chemotherapies, it is important for us as physicians to try and work out what the roles of radiation and surgery are in TKI–eligible patients with brain metastases,” Chiang said.
In their study, Chiang and colleagues assigned 351 patients with NSCLC from six institutions into three subgroups: 37% (n = 131; median age at brain metastases, 60 years) received an upfront EGFR TKI followed by SRS or WBRT at intracranial progression; 34% (n = 120; median age, 58 years) underwent upfront WBRT followed by EGFR TKI; and 29% (n = 100; median age, 63 years) underwent upfront SRS followed by EGFR TKI.
OS and intracranial PFS served as the primary endpoints.
Median follow-up was 22 months.
Patients who received upfront EGFR TKIs were less likely to have symptomatic brain metastases (12% vs. WBRT, 51% vs. SRS, 49%; P < .001) and were more likely to have brain metastases smaller than 1 centimeter (66% vs. WBRT, 35% vs. SRS, 44%; P < .001). Further, patients who received upfront WBRT were more likely to have more than 10 brain metastases (37% vs. EGFR TKI, 15% vs. SRS, P < .001). Additionally, a greater proportion of patients treated with upfront EGFR TKIs (91%) or upfront WBRT (92%) were stage IV at diagnosis than patients treated with upfront SRS (80%; P = .014).
Median OS after brain metastases for the entire cohort was 30 months (95% CI, 27-34).
Median OS was 46 months (95% CI, 37-57) for upfront SRS, 30 months (95% CI, 27-38) for upfront WBRT and 25 months (95% CI, 20-28) for upfront EGFR TKI.
The rate of 2-year OS was 78% (95% CI, 66-85) for upfront SRS, 62% (95% CI, 52-70) for upfront WBRT, and 51% (95% CI, 42-60) for upfront EGFR TKI.
After controlling for significant covariables, upfront SRS was independently associated with improved OS relative to EGFR TKI (adjusted HR = 0.39; 95% CI, 0.26-0.58). Upfront WBRT also was associated with improved OS relative to EGFR-TKI (adjusted HR = 0.7; 95% CI, 50-98).
The median time to intracranial progression was 23 months (95% CI, 18-28) for upfront SRS, 24 months (95% CI, 21-30) for upfront WBRT, and 17 months (95% CI, 14-30) for EGFR TKI.
In patients with a less favorable prognosis — defined as a diagnosis-specific graded prognostic assessment of 0 to 1.5 — OS was longer among those who underwent SRS (median, 33 months; 95% CI, 19-44) than upfront WBRT (27 months; 95% CI, 19-30) and EGFR TKI (19 months; 95% CI, 17-25).
“It is not clear exactly why the [SRS followed by EGFR TKI] group had the longest survival,” Chiang said. “While it is possible to postulate a positive interaction between high-dose radiation and TKI therapy, any retrospective study can be biased by patient selection.”
Chiang and colleagues noted that excluding patients with missing covariables could have led to selection bias and that researchers did not account for potential toxicities associated with local therapies and their impact on quality of life.
“What this study does not address well is burden of disease in the body and rapidity of its progression,” Chiang said. “It also is possible then that patients with minimal brain disease and significant body disease were started on TKI first as opposed to those who had little body disease but more significant brain disease who got radiation first. Both SRS and WBRT are very effective treatments of brain disease and, therefore, survival would be determined by amount of disease in the body. In this scenario, patients with least brain and body disease — ie, those getting SRS — would have the longest OS.”
Until a prospective, multi-institutional study is conducted and published, the standard-of-care treatment of newly diagnosed brain metastases should remain SRS followed by systemic therapy, Chiang and colleagues wrote.
A clinical question that remains is whether SRS could be delayed in patients who achieved intracranial response with first-line EGFR TKIs, Lin Zhou, MD, of West China Hospital at Sichuan University in China, and colleagues wrote in an accompanying editorial.
“With the shrinkage and reduction of the number of lesions in the brain and improvement of symptoms by EGFR TKIs, is upfront SRS necessary, or is it an optimal option?” they wrote. “It seems time to weigh the benefits of deferring or withholding WBRT in the era of targeted therapies. In addition, enthusiasm for the benefits of upfront SRS can be appropriately balanced. However, multidisciplinary discussion is always required, and prospective data will be the key solution to these questions.” – by Chuck Gormley
For more information:
Veronica L. Chiang, MD, can be reached at Yale Department of Neurosurgery, PO Box 208082, New Haven, CT 06520; email: email@example.com.
Disclosure: Chiang reports no relevant financial disclosures. Please see the full study for a list of all other researchers’ relevant financial disclosures. Zhou reports no relevant financial disclosures. One editorial author reports consultant/advisory roles with AstraZeneca, Eli Lilly, Elekta, Hoffmann-La Roche, Pfizer and Varian Medical Systems.