Brain metastases (BrM) are common among patients with solid malignancies.1 The prognosis of patients with BrM has generally been poor, with population-based studies demonstrating survival times of ≤12 months for patients with BrM at cancer diagnosis and ≤4 months for most elderly patients (>65 years), irrespective of primary site.2,3 However, longer-term survival is increasingly observed among select subsets, largely due to advances in systemic and local, brain-directed treatment.4,5 Here, we characterize factors associated with long-term (≥5 years) survival among a contemporary cohort of patients with brain metastases.
We identified 1281 patients diagnosed with BrM without leptomeningeal disease at Dana-Farber/Brigham and Women’s Cancer Center between January 2008 and December 2015 (most recent date to permit 5-year survival analyses) who either died or survived ≥5 years. Median survival and 5-year survival estimates were calculated with the Kaplan-Meier method. Multivariable logistic regression was used to characterize predictors of 5-year survival after BrM diagnosis. Patients censored (ie, lost to follow-up) before 5 years were excluded from the multivariable logistic regression model (N = 51), leaving 1230 patients to be analyzed.
Among all 1281 patients, median survival after BrM diagnosis was 11.9 months. In total, 123 patients (11.4%) survived ≥5 years. Subgroups displaying high rates of 5-year survival (Table 1) included patients with solitary brain metastases (36.1%), non-small cell lung cancer (NSCLC) with ALK alterations (30.8%) or EGFR mutations (15.1%), and HER2-positive breast cancer (21.0%).
Table 1.
Survival Outcomes of Patients With Brain Metastases by Clinical Subset
| Number of Patientsa | Median Survival in Months From Time of BrM Diagnosis (95% CI) | 5-Year Survival From Time of BrM Diagnosis (%) | |
|---|---|---|---|
| All patients | 1281 | 11.9 (11.0-12.7) | 11.4 |
| Primary tumor site | |||
| Breast | 191 | 17.9 (15.0-22.1) | 12.6 |
| NSCLC | 569 | 12.7 (11.0-14.4) | 11.7 |
| Melanoma | 170 | 10.5 (8.9-12.2) | 15.3 |
| Other | 351 | 9.4 (7.3-11.0) | 8.3 |
| Karnofsky performance status | |||
| 90-100 | 551 | 18.4 (16.0-20.7) | 17.4 |
| 80 or below | 730 | 8.2 (7.0-9.4) | 6.8 |
| Number of brain metastases | |||
| 1 | 507 | 13.5 (11.9-15.7) | 16.8 |
| 2-4 | 464 | 12.5 (10.9-14.0) | 10.1 |
| 5 or more | 310 | 8.6 (6.3-10.4) | 4.7 |
| Charlson comorbidity index | |||
| 0-1 | 1063 | 12.5 (11.7-13.8) | 12.0 |
| 2-6 | 218 | 9.5 (8.2-11.0) | 8.3 |
| Initial brain-directed management strategyb | |||
| Any neurosurgical resection | 339 | 18.3 (15.2-21.3) | 21.3 |
| Stereotactic radiation therapy without surgery | 489 | 11.5 (9.9-13.0) | 12.0 |
| Whole-brain radiation therapy without surgery or stereotactic radiation therapy | 431 | 7.5 (5.8-9.5) | 2.5 |
| Lung | |||
| ALK-rearranged | 26 | 31.8 (11.8-43.8) | 30.8 |
| EGFR-mutant | 97 | 15.8 (11.3-21.5) | 15.1 |
| No EGFR or ALK mutation | 279 | 13.8 (11.9-16.5) | 12.8 |
| By EGFR alteration | |||
| Exon 18 | 4 | 20.3 (5.8-not reached) | 0.0 |
| Exon 19 | 42 | 18.5 (10.3-24.3) | 19.1 |
| Exon 20 S768Ic | 1 | 17.8 | 0.0 |
| Exon 20 (all others) | 8 | 7.2 (2.3-31.8) | 0.0 |
| Exon 21 | 34 | 15.8 (7.8-31.3) | 15.9 |
| Non-18 to 21 Exons | 2 | 16.3 (12.0-not reached) | 0.0 |
| Unspecified | 6 | 11.0 (0.53-not reached) | 33.3 |
| Breast | |||
| HER2-positive | 90 | 27.3 (23.1-32.7) | 21.0 |
| HR-positive/HER2-negative | 59 | 15.9 (9.7-19.0) | 3.6 |
| Triple-negative | 42 | 7.8 (3.4-11.1) | 7.1 |
| Solitary brain metastasisd | 95 | 26.3 (17.2-33.1) | 36.1 |
| Number of prior systemic therapy regimens | |||
| 0 | 721 | 14.5 (12.8-15.9) | 14.8 |
| 1-2 | 388 | 9.6 (8.2-11.4) | 8.6 |
| 3-4 | 116 | 5.9 (4.5-8.3) | 5.2 |
| 5+ | 56 | 8.6 (3.8-12.2) | 0.0 |
| Number of sites of extracranial diseasee | |||
| 0 | 313 | 22.4 (19.6-25.0) | 24.8 |
| 1-2 | 522 | 11.9 (10.2-13.5) | 9.5 |
| 3-4 | 409 | 7.6 (6.5-9.1) | 4.8 |
| 5-6 | 37 | 3.1 (2.1-5.3) | 0.0 |
| Progressive extracranial disease at the time of brain metastasis | |||
| No | 376 | 18.5 (16.0-23.1) | 21.2 |
| Yesf | 905 | 9.5 (8.9-10.3) | 7.3 |
Abbreviations: BrM, brain metastasis; NSCLC, non-small cell lung cancer.
aIncludes patients (N = 51) censored within 5 years due to loss of follow-up.
bDoes not include patients with unknown/other treatment strategy.
cS768I mutation in Exon 20 was analyzed separately from other Exon 20 mutations due to data suggesting that it is more favorable than other Exon 20 mutations.
dPatients with a single brain metastasis, no extracranial sites of disease, and a controlled primary tumor.
eExtracranial sites included: lung, liver, bone, distant lymph node, adrenal, and soft tissue sites.
fIncludes patients with brain metastases at diagnosis of malignancy (ie, including patients with untreated extracranial disease).
On multivariable logistic regression (N = 1230), significant predictors of survival ≥5 years, from diagnosis of intracranial disease, included: breast primaries vs reference of NSCLC (OR: 2.76 [95% CI, 1.38-5.50], P = .004), Karnofsky performance score 90-100 vs <90 (OR: 2.36 [95% CI, 1.50-3.70], P < .001), neurosurgical resection as initial management (OR: 1.87 [95% CI, 1.01-3.46, P = .05), and controlled extracranial disease at BrM diagnosis (OR: 2.30 [95% CI, 1.44-3.66], P < .001). Factors significantly associated with not surviving ≥5 years included: receipt of whole brain radiation therapy vs stereotactic radiation therapy as initial management (OR: 0.29 [95% CI, 0.12-0.68], P = .004), an increasing extracranial disease burden (OR: 0.71 per additional extracranial organ involved [95% CI, 0.57-0.88], P = .002), and an increasing number of prior systemic therapy regimens for metastatic disease (OR: 0.65 per additional regimen [95% CI, 0.50-0.84], P = .001). Nonsignificant co-variates included: year of BrM diagnosis, age, Charlson comorbidity score, number of BrM, and size of largest BrM.
Although the prognosis of patients with BrM has historically been poor, our study highlights that select patients, such as those with controlled extracranial disease, solitary BrM, targetable molecular alterations, and high performance status may display extended survival. These findings can be explained by several factors, including that some patients with solitary brain metastases can be cured with aggressive local therapy.6 In addition, improvements in systemic therapy have explanatory potential, especially in patients with targetable mutations/alterations, such as EGFR/ALK-mutated/rearranged NSCLC or HER2-positive breast cancer.1 Although our cohort was limited to patients diagnosed in/before 2015 (to allow for 5-year survival analyses), advances in systemic therapy since then would strongly suggest the potential for further improvements in 5-year survival among more contemporary cohorts. For example, the FLAURA study demonstrated improved overall survival with osimertinib over older EGFR inhibitors in EGFR-mutant NSCLC. In ALK-rearranged NSCLC, newer targeted agents including alectinib, ceritinib, brigatinib, and lorlatinib have proven oncologically superior to crizotinib. In addition, targetable alterations in NSCLC continue to be discovered, including MET, RET, BRAF, HER2, NTRK, ROS1, and KRAS G12C, such that approximately 40% of patients with lung adenocarcinomas now have a targeted drug option.1 In HER2-positive breast cancer, HER2CLIMB and DESTINY-Breast01 demonstrated the potency of tucatinib and trastuzumab deruxtecan, respectively, to control intracranial and extracranial disease in highly pre-treated populations.1 Most strikingly, dual-agent immunotherapy in metastatic melanoma has transformed a disease with a historically dismal to prognosis to one where a sizeable percentage of patients can be cured.1
Our data provide optimism for patients with brain metastases and favorable disease characteristics, while also highlighting the importance of minimizing longer-term, treatment-related toxicities among select subsets of patients with a favorable prognosis. In addition, while patients with BrM have often been excluded from systemic therapy-based clinical trials, the results of this study suggest that select patients with favorable characteristics could be enrolled on studies of central nervous system-active agents given their potential for long-term survival, particularly if such studies permit salvage brain-directed therapy.
Conflict of interest statement. D.N.C. is a recipient of research support from NH TherAguix. A.A.A. reports research funding from Varian Medical Systems and NH TherAguix and consulting fees from Novartis and Seagen. The remaining authors declare no conflicts of interest.
Authorship statement. Study conception: N.L. and A.A.A.; Data collection: D.N.C. and A.A.A.; Statistical analysis: N.L. and A.A.A.; Drafting of the manuscript: N.L. and A.A.A.; Critical revision of the manuscript: all authors; Supervision: A.A.A.
Funding
No funding was required for this study.
References
- 1. Lamba N, Wen PY, Aizer AA. Epidemiology of brain metastases and leptomeningeal disease. Neuro Oncol. 2021;23(9):1447–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Cagney DN, Martin AM, Catalano PJ, et al. Incidence and prognosis of patients with brain metastases at diagnosis of systemic malignancy: a population-based study. Neuro Oncol. 2017;19(11):1511–1521. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Lamba N, Kearney RB, Catalano PJ, et al. Population-based estimates of survival among elderly patients with brain metastases. Neuro Oncol. 2021;23(4):661–676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sperduto PW, Yang TJ, Beal K, et al. Estimating survival in patients with lung cancer and brain metastases: an update of the graded prognostic assessment for lung cancer using molecular markers (Lung-molGPA). JAMA Oncol. 2017;3(6):827–831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Sperduto PW, Jiang W, Brown PD, et al. Estimating survival in melanoma patients with brain metastases: an update of the graded prognostic assessment for melanoma using molecular markers (Melanoma-molGPA). Int J Radiat Oncol Biol Phys. 2017;99(4):812–816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Lamba N, Cagney DN, Brigell RH, et al. Neurosurgical resection and stereotactic radiation versus stereotactic radiation alone in patients with a single or solitary brain metastasis. World Neurosurg. 2019;122:e1557–e1561. [DOI] [PubMed] [Google Scholar]