Survival benefits of primary tumor surgery for synchronous brain metastases: A SEER-based population study with propensity-matched comparative analysis

Abstract

Background: Evidence about the prognostic value of primary tumor surgery (PTS) in patients with brain metastatic malignancies is ambiguous and controversial. This study assessed the survival benefits of primary tumor surgery in patients with brain metastases (BMs).

Methods: Adults patients with BMs that originated from lung, breast, kidney, skin, colon, and liver diagnosed between 2010 and 2018 were derived from the Surveillance, Epidemiology, and End Results database (SEER). Propensity score matching (PSM) was used to balance the bias between patients with or without PTS. Then the prognostic value of PTS was estimated by Kaplan-Meier analysis and Cox proportional hazard regression models.

Results: A total of 32,760 patients with BMs secondary to non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), breast cancer, renal cancer, melanoma, colorectal cancer, and liver cancer were identified from the database. After PSM at 1:1 ratio, PTS appeared to significantly prolong cause-specific survival (CSS) time for patients with BMs secondary to NSCLC, breast cancer, renal cancer, and colorectal cancer (hazard ratio [HR] = 0.60 [0.53-0.68], 0.56 [0.43-0.73], 0.47 [0.37-0.60], and 0.59 [0.37-0.95], respectively, all p < 0.05). Patients with earlier T and N classifications, no extracranial metastasis, and cancer-specific subtypes (adenocarcinoma in NSCLC, hormone receptor-negative breast cancer) may derive more survival benefits from PTS when suffering from BMs.

Conclusion: This population-based study supported PTS could provide survival benefits for patients with BMs secondary to NSCLC, breast cancer, renal cancer, and colorectal cancer. More emphasis should be put on PTS of selected patients with BMs.

Keywords: SEER; brain metastases; primary tumor surgery; survival benefits.

FIGURE 1

The treatment trends of patients with brain metastases from six primary tumor types between 2010 and 2018. NSCLC, non‐small cell lung cancer; PTS, primary tumor surgery; SCLC, small cell lung cancer.

FIGURE 2

Kaplan–Meier curves of cause‐specific survival by primary tumor surgery after propensity score matching depending on the primary tumor types. Individual surgeon selection bias was not excluded (A) or excluded (B). The p values were calculated by log‐rank tests. NSCLC, non‐small cell lung cancer; PTS, primary tumor surgery; recomm, recommended; SCLC, small cell lung cancer; w/o, without.

FIGURE 3

Kaplan–Meier curves of cause‐specific survival by primary tumor surgical procedures after propensity score matching depending on the primary tumor types. The p values were calculated by log‐rank tests. NSCLC, non‐small cell lung cancer.

FIGURE 4

Kaplan–Meier curves for cause‐specific survival of patients only with metastases in brain by surgical sites depending on the primary tumor types. The p values were calculated by log‐rank tests. MS, metastatic surgery; NSCLC, non‐small cell lung cancer; PTS, primary tumor surgery.

FIGURE 5

Kaplan–Meier curves of cause‐specific survival by treatment after propensity score matching depending on the primary tumor types. The p values were calculated by log‐rank tests. NSCLC, non‐small cell lung cancer; PTS, primary tumor surgery.

FIGURE 6

Subgroup analyses estimating the prognostic value of primary tumor surgery grouped by patients with different demographic and clinical characteristics after propensity score matching depending on the primary tumor types. Cox proportional hazard regression for cause‐specific survival was presented as the hazard ratio and 95% confidential interval. CI, confidential interval; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; HR, hormone receptor; NSCLC, non‐small cell lung cancer.