Radiosurgery for GI Cancer Brain Metastases: Does It Help?

TOPLINE: A study at two US medical centers found that patients with brain metastases from gastrointestinal (GI) cancers had significantly worse overall survival and intracranial progression-free survival (PFS) after stereotactic radiosurgery than those with brain metastases from non-GI cancers. METHODOLOGY: Brain metastases from GI cancers are relatively uncommon and are associated with a poor prognosis

TOPLINE:

A study at two US medical centers found that patients with brain metastases from gastrointestinal (GI) cancers had significantly worse overall survival and intracranial progression-free survival (PFS) after stereotactic radiosurgery than those with brain metastases from non-GI cancers.

METHODOLOGY:

  • Brain metastases from GI cancers are relatively uncommon and are associated with a poor prognosis, but data are limited on patient outcomes following stereotactic radiosurgery in this population.
  • Researchers conducted a retrospective cohort study involving 1383 patients with GI and non-GI cancers who completed initial stereotactic radiosurgery between January 2src15 and December 2src2src.
  • Overall, 1src2 patients had brain metastases from primary GI cancer, including colorectal (45.1%), esophageal (33.3%), and other types (21.6%). Of 1281 patients with brain metastases from non-GI cancers, most (54.7%) had lung cancer, followed by breast (14.7%) and genitourinary (7.9%) cancers.
  • Patients with GI cancer were younger (mean age, 59.1 vs 63.5 years), predominantly male (56.9% vs 44.3%), and more likely to have received systemic therapy (73.5% vs 63.9%) and undergone resection for brain metastases (45.1% vs 25.src%) before stereotactic radiosurgery.
  • Primary outcomes were overall survival and intracranial PFS. The median follow-up duration was 9.93 months.

TAKEAWAY:

  • Median overall survival was significantly shorter for patients with GI cancers (5.4 months) than for those with non-GI cancers (1src.6 months), with a 12-month overall survival of 22.6% vs 46.2%. After adjustment for age, sex, prior therapy, metastatic burden, and other factors, patients with GI cancer had significantly worse overall survival (hazard ratio [HR], 1.92).
  • Patients with GI cancers also had shorter intracranial PFS (6.2 months) than those with other cancer types (12.3 months), with a 12-month intracranial PFS of 18.5% in the GI cancer group and 42.src% in the non-GI cancer group. After adjustment, GI patients had significantly worse intracranial PFS (HR, 1.6src).
  • Intracranial progression was mostly distant in the brain. Only two patients (2%) with GI cancer had progression of the treated lesion compared with 141 (11%) of those with other primary cancers (odds ratio, src.162; P <.srcsrc2).
  • There were no statistically significant differences in intracranial PFS between patients with brain metastases from colorectal or esophageal cancers and those with brain metastases from other GI cancers. However, the authors reported a trend toward longer PFS in patients with colorectal cancer (P=.src82) and toward shorter PFS in patients with esophageal cancer.

IN PRACTICE:

Patients with brain metastases from GI cancer are at a “higher risk for poor outcomes after [stereotactic radiosurgery] and may benefit from enhanced surveillance and oncologic care,” the authors concluded.

SOURCE:

The study, led by Jamiluddin J. Qazi, MD, Duke University Medical Center in Durham, North Carolina, was published online in Advances in Radiation Oncology.

LIMITATIONS:

Most of the non-GI patients had primary lung cancer, where newer targeted therapies may have improved outcomes. Patients with GI cancer had more extracranial disease and prior systemic therapy, which could have affected survival. Treating death as a censored event might have inflated PFS in patients with GI cancers.

DISCLOSURES:

This study did not receive any specific funding. One author disclosed receiving funding from Roche, unrelated to this study.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

Read More

About Author