Introduction
Methods
Results
Prognosis of BCBM in dependence of biological subtype
Study | Design | Aim | Study collective | Results |
---|---|---|---|---|
Incidence | ||||
Kuksis et al. 2021 [7] | Systematic Review | Incidence of BCBM among metastatic BC stratified by molecular subtype | Number of BC Patients: HR+/HER2−: 14,656 HER2+: 5971 TNBC: 4102 | Cumulative incidence of BMBC: HR+/HER2−: 15% HER2+: 31% TNBC: 32% |
Prognosis | ||||
Sperduto et al. 2020 [2] | Multicenter retrospective analysis | Breast prognostic index in survival in breast cancer brain metastases | Patients with BCBM: Luminal A: 772 Luminal B: 527 (REF) TNBC: 595 HER2+: 421 | Worse survival associated with luminal A (HR 2.0) and TNBC (HR 2.8) |
Kim et al. 2020 [20] | Multicenter retrospective analysis | Prognostic factors regarding overall survival in BCBM | 730 patients with BCBM | Median FU: 11.9 months HR- and HER2- were associated with inferior OS in multivariate analyses |
Kim et al. 2021 [14] | Multicenter retrospective analysis | Impact of molecular subtype on distant intracranial control after WBRT, SRS, or SRT for BCBM | Patients with BCBM HR+/HER2−: 136 HER2+: 253 TNBC: 149 | 1‑year intracranial distant control: HR+/HER2−: 48.1% HER2+: 56.0% TNBC: 30.4% Anti-HER2 therapy and initial use of WBRT significantly lowered rate of new BCBM |
Cagney et al. 2019 [17] | Monocentric retrospective analysis | Impact of molecular subtype on intracranial control after WBRT, SRS or SRT | Patients with BCBM: HR+/HER2−: 116 HER2+: 164 TNBC: 69 | 2‑year intracranial local control: HR+/HER2−: 82.2% HER2+: 79.1% TNBC: 50.8% Intracranial distant control: Compared to HR+/HER2−: HER2+: HR 1.01 (p = 0.97) TNBC: HR: 3.2 (p < 0.001) |
Chong et al. 2015 [18] | Multicenter retrospective analysis | Impact of molecular subtype on treatment outcome (surgery/SRS + WBRT vs. WBRT alone) | Patients with BCBM HR+/HER2−: 29 HER2+: 47 TNBC: 40 | Median FU: 50.9 months HR+/HER2− had the best, TNBC the worst OS and local control rates. Local treatment (SRS/Surgery) in addition to WBRT improved intracranial control rates in HER2+ and TNBC subtype |
Radiotherapy in BCBM
Study | Design | Study collective/proportion of BCBM | Results |
---|---|---|---|
SRS/SRT (+/− WBRT) for intact BM | |||
Redmond et al. 2021 [22] (HyTEC) | TCP models based on pooled dosimetric and clinical data from 56 studies | Total number of BM 13,929 BCBM among the 4 most frequent entities | SRS for BM ≤ 20 mm with 18 and 24 Gy corresponded to > 85% and 95% 1‑year LC rates, respectively. SRT (3–5 fx) in the range of 27 to 35 Gy, 80% 1‑year LC was achieved for tumors of 21 to 40 mm in diameter |
Andrews et al. 2004 [23] (RTOG 9508) | Randomized multicenter trial Arms: 1) WBRT vs. 2) WBRT + SRS | 333 patients with 1–3 BM 34 patients (10.2%) with BCBM | Survival benefit in 1 BM (median survival 6.5 vs. 4.9 months) for the SRS + WBRT arm |
Kocher et al. 2011 [24] (EORTC 22952-26001) | Randomized phase III multicenter trial Arms: SRS or resection + 1) WBRT vs. 2) Obs | 359 patients with 1–3 BM 42 patients (12%) with BCBM | No difference in median OS (10.9 vs. 10.7 months) WBRT improved both 2‑year intracranial local control (69% vs. 81%) and 2‑year intracranial distant control (52% vs. 67%) |
Churilla et al. 2019 [13] | Unplanned exploratory analysis of EORTC 22952–26001: comparison of intracranial local control of SRS vs. resection | 268 patients with 1–2 BM 28 patients (10.4%) with BCBM | Intracranial local control after 1 year: 82% resection vs. 86.8% SRS. HR 1.15 SRS associated with improved early local control compared with surgery. → No differences between SRS and surgery as primary treatment |
Brown et al. 2016 [25] | Randomized multicenter phase III trial Arms: SRS +1) WBRT vs. 2) obs | 213 patients with 1–3 BM 18 patients (8.4%) with BCBM | Less cognitive deterioration and improved quality of life with SRS alone compared to SRS + WBRT 1‑year intracranial local control: 90.4% (SRS + WBRT) vs. 72.6% (SRS) (p < 0.05) Intracranial distant control: 92.5% (SRS + WBRT) vs. 70% (SRS) (p < 0.05) Median OS 10.4 (SRS) vs. 7.4 months (SRS + WBRT; HR 1.02; p = 0.92) |
Aoyama 2006 [12] | Randomized controlled multicenter trial Arms: 1) WBRT + SRS vs. 2) SRS | 132 patients with 1–4 BM, 9 patients (6.8%) with BCBM | No difference in median OS (7.5 vs. 8.0 months) Additional WBRT improved intracranial distant control after 1 year (recurrence rate 46.8% for SRS + WBRT vs. for 76.4% SRS) |
Wijetunga 2020 [26] | Randomized controlled trial. Arms: 1) SRS + WBRT vs. 2) SRS | 58 patients with 1–3 BM, 8 patients (13.8%) with BCBM | Intracranial distant control at 1 year 73% (SRS + WBRT) vs. 27% (SRS) Mean probability of cognitive dysfunction at 4 months in 52% (SRS + WBRT) vs. 24% (SRS) |
Postoperative SRS for resected BM | |||
[27] | Single center trial Postoperative SRS | 79 patients with 1–4 BM | Median PFS 10.0 months, median OS 14.3 months. 1‑year local intracranial control: 89.8% |
Kayama 2018 [28] | Noninferiority randomized controlled phase III trial Arms: postoperative WBRT vs. salvage SRS to cavity | 271 patients with 1–4 BM, 53 patients (19.5%) with BCBM | No difference in median OS (15.6 months). Improved intracranial PFS after WBRT (10.4 months) vs. salvage SRS (4.0 months) with increased toxicity (grade II–IV cognitive dysfunction in 16.4% vs. 7.7%) |
Mahajan et al. 2017 [29] | Randomized controlled trial Arms: 1) Observation vs. 2) postoperative SRS | 132 patients with 1–3 BM, 23 patients (17.4%) with BCBM | 12-month local tumor control in BCBM 50% SRS with a HR of 0.5 of local control (p < 0.05) |
Brown 2017 [30] | Randomized controlled phase III trial Arms: 1) postoperative WBRT vs. 2) postoperative SRS | 194 patients with 1 resected BM; < 5.0 cm | Significantly worse local and distant brain control in SRS group compared to WBRT. No difference in OS |
Eitz 2020 [31] | Retrospective multicenter cohort study of postoperative SRT | 558 patients with resected BM. 101 patients with BCBM (17.6%) | Local control was 94% at 5 months, 84% at 1 year, 75% at 2 years, and 71% at 3 years (median was not reached) |
Study | Design | Study collective/proportion of BCBM | Results |
---|---|---|---|
SRS in multiple BM | |||
Chang et al. 2010 [32] | Single-center retrospective analysis | 323 patients with BM: Group 1: 215 patients with 1–5 BM Group 2: 58 patients with 6–10 BM Group 3: 17 patients with 11–15 BM Group 4: 33 patients with > 15 BM | No significant differences in intracranial local control rates, however, group 4 showed the worst intracranial distant control No significant differences in median OS: Group 1: 10 months Group 2: 10 months Group 3: 13 moths Group 4: 8 months |
Yamamoto 2014 [33] (JLGK0901) | Multicenter prospective observational study | 1194 patients with 1-10 BM, 455 patients with 1 BM, 531 patients with 2-4 BM, 208 patients with 5-10 BM. 123 patients (10%) with BCBM | No significant difference in OS for patients with 5–10 BM compared to patients with 2–4 BM Median OS: 1 BM: 13.9 months 2–4 BM: 10.8 months 5–10 BM: 10.8 months No significant differences in local recurrence during follow-up |
Wilson et al. 2020 [34] | Single-center retrospective analysis | 91 patients with 1–21 BCBM receiving SRS | Median OS 15.7 months Subgroups: ER+/HER2− 13.8 months ER+/HER2+ 21.4 months ER−/HER2+ 20.4 months TNBC 8.5 months Metastatic volumes > 10 cm3 were associated with significantly worse OS |
SRS in comparison to WBRT for multiple BM | |||
El Shafie 2020 [35] | Retrospective matched-pair analysis | 128 patients with 3–16 BM, 64 patients treated with SRS, 64 matched patients treated with WBRT. 24 patients (18.8%) with BCBM | Prolonged median OS (15.7 months vs. 8.0 months) and 1‑year intracranial local control 91.7% after SRS. Prolonged median intracranial (local and distant) control after WBRT (8.6 months SRS vs. 22.4 months WBRT) |
Hartgerink, Bruynzeel et al. 2021 [36] | Phase III randomized multicenter trial Arms: 1) WBRT vs 2) SRS/SRT | 29 patients with 4–10 BM, 1 patient (WBRT, 3.4%) with BCBM | No statistically significant differences due to premature closure of trial. 1‑year OS: 31% (WBRT) vs. 57% (SRS). QoL decrease in WBRT worse compared to SRS |
J. Li 2020 [37] | Phase III randomized trial Arms: SRS vs. WBRT | 72 patients with 4–15 BM, 36 patients treated with SRS, 36 patients treated with WBRT | SRS with reduced risk of neurocognitive deterioration. Improved median OS after SRS (10.4 vs. 8.4 months). Preliminary analysis of intracranial local control at 4 months LC 100% (SRS) vs. 95.5% (WBRT) |
Kim, Kim et al. 2023 [38] | Multicenter retrospective analysis | 471 patients with 1–10 BCBM 1–4 BCBM: n = 337 5–10 BCBM: n = 134 | 79.9 % of patients with ≥5 BCBM were treated with WBRT and only 9.0 with SRS In patients with ≤10 BCBM the number of BCBM and WBRT did not affect OS.TNBC/extracranial disease decreased OS TNBC/extracranial disease decreased OS |
Limited brain metastases in breast cancer (≤ 4)
Local therapy of multiple (≥ 5) brain metastases
Focal radiotherapy and craniospinal axis irradiation in leptomeningeal carcinomatosis
Study | Design | Study collective/proportion of BCBM | Results |
---|---|---|---|
RT in LC | |||
Rudnicka 2007 [73] | Single-center retrospective study on efficacy of multimodal treatment with IT chemotherapy, systemic therapy, and radiotherapy (WBRT/CSI) | 67 patients with LC in breast cancer | Clinical response in 76% Median OS 16 weeks. In multivariate analysis, survival benefit for systemic therapy and IT chemotherapy (not for WBRT) |
Yang 2022 [75] | Single-center phase II randomized trial Arms: proton CSI (pCSI) vs. involved-field RT (IFRT) | 63 patients with LC, 42 patients treated with pCSI, 21 patients treated with IFRT | Significant benefit in CNS PFS with pCSI (7.5 months vs. 2.3 months) and OS (9.9 months vs. 6.0 months) at planned interim analysis. No difference in grade III/IV toxicity |
Irradiation techniques and follow-up
Study | Design | Study collective/proportion of BCBM | Results |
---|---|---|---|
Donepezil/memantine | |||
Rapp, Case et al. 2015 [83] | Phase III randomized trial “Brain tumor survivors” after WBRT Arms: 1) donepezil, 2) placebo | 66% with primary brain tumors 27% with brain metastases | Significant differences favoring donepezil for memory, motor speed, and dexterity |
Brown et al. 2013 [84] | Multicenter randomized, double-blind controlled study Arms: 1) WBRT + memantine, 2) WBRT + Placebo | 554 patients with BM, 75 patients (13.5%) with BCBM 256 patients received memantine, 252 patients received placebo | No significant differences regarding cognitive decline in patients treated with memantine at 24 weeks, but significant differences favoring memantine in MMSE (Mini Mental State Examination) at 24 weeks |
Hippocampal avoidance during WBRT | |||
Gondi et al. 2014 [80] (RTOG 0933) | Multicenter single-arm phase II study on HA-WBRT—cognitive function with comparison of historical control group | 42 patients with BM treated with HA-WBRT | Significantly lower decline in HVLT‑R DR (Hopkins Verbal Learning Test-Revised delayed recall) at 4 months after HA-WBRT compared to historical control. No decline in QoL. 2 patients with grade III toxicity, no grade IV/V toxicity |
Brownet al. NRG-CC001 [85] | Randomized phase III Arms: 1) WBRT + memantine, 2) HA-WBRT + memantine | 518 patients with brain metastases 18.5% BCBM | Median FU 7.9 months Significantly lower cognitive failure after HA-WBRT plus memantine vs. WBRT + memantine At 4 months: deterioration in executive function: 23.3% (HA-WBRT + Memantine) vs. 40.4% (WBRT + Mematine) At 6 months: Learning and memory: 11.5% (HA-WBRT + Mematine) vs. 24.7% (WBRT + Memantine) |
Rodriguez et al. 2021 [81] | Randomized phase III trial Arms: 1) standard PCI, 2) HA-PCI | 150 patients with brain metastases in SCLC | Median FU 40.4 months Significantly lower decline in delayed free recall in HA-PCI (23.5% vs. 5.8%) at 3 months and in total recall (47.6% vs. 14.2%) at 24 months |
Yang et al. [86] | Single-blind randomized phase II trial Arms: 1) conformal WBRT, 2) HA-WBRT | 65 patients with brain metastases 3.1% BCBM | Median FU 12.4 months No differences in baseline neurocognitive function. Significantly better preservation of HVLT‑R recognition–discrimination index and memory score in HA-WBRT at 6 months No differences in other cognitive tests No differences in OS or PFS |
Belderbos et al. 2021 [79] | Multicenter randomized phase III trial Arms: 1) standard PCI, 2) HA-PCI | 168 patients with brain metastases in SCLC | Median FU 26.6 months No significant difference in HVLT‑R total recall at 4 months (29% vs. 28%) or other cognitive tests Cumulative incidence of brain metastases at 2 years 20% vs. 16% |
Combination of radiotherapy with systemic therapy
1) Are there concerns regarding increased toxicity when combining these treatments?
2) Can local therapy be postponed when systemic therapy with relevant intracranial response rates is given?
Study | Design | Study collective/proportion of BCBM | Results |
---|---|---|---|
Systemic therapies in HER2+ metastatic BC | |||
EMILIA (Krop, Lin et al. 2015 [97]) | Retrospective, exploratory analysis of the EMILIA trial Arms: T‑DM1 vs. XL after trastuzumab therapy | 95 patients with treated, asymptomatic HER2+ BCBM. 45 patients received T‑DM1, 50 patients received XL | Improved median OS with T‑DM1 in patients with BCBM at baseline (26.8 vs. 12.9 months), similar PFS in both arms (5.9 vs. 5.7 months) There was less grade ≥ III toxicity with T‑DM1 compared to XL |
HER2Climb (Lin, Murthy et al. 2023 [99]) | Preplanned subgroup analysis of the HER2CLIMB trial Arms: tucatinib + trastuzumab/capecitabine vs. placebo + trastuzumab/capecitabine | 291 patients with HER2+ BCBM, 66 patients with untreated BCBM | Prolonged median OS with Tucatinib (21.6 vs. 12.5 months) and intracranial PFS (13.9 vs. 5.6 months) Also improved intracranial overall response rate (ORR) with tucatinib (47.3% vs. 20.0%) |
Destiny 03 (Hurvitz, Hegg et al. 2023 [100]) | Subgroup analysis of a randomized phase III study Arms: T‑DM1 vs. T‑DXd | 82 patients with treated, asymptomatic BCBM at baseline (43 patients treated with T‑DXd and 39 patients treated with T‑DM1) | Consistent OS benefit in favor of T‑DXd across subgroup analyses, including those with BM |
TUXEDO‑1 (Bartsch, Berghoff et al. 2022 [101]) | Single-center prospective single-arm phase II study on T‑DXd | 15 patients with HER2+ BCBM, in 40% BCBM were previously untreated | ORR of 73.3%, median PFS of 14.0 months |
KAMILLA (Montemurro, Delaloge et al. 2020 [102]) | Multicenter prospective single-arm study on T‑DM1 | 398 patients (19.9%) with BCBM, 56.8% of whom had received RT previously | Median PFS/OS was 5.5/18.9 months in patients with baseline BM. Best intracranial ORR was 21.4% |
LANDSCAPE (Bachelot, Romieu et al. 2013 [103]) | Multicenter prospective single-arm study on lapatinib + capecitabine | 45 patients with HER2+ previously untreated BCBM | Median OS 17.0 months. Intracranial ORR 66%. 49% with grade III/IV treatment-related toxicity |
NALA (Saura, Oliveira et al. 2020 [104]) | Subgroup analysis of a multicenter randomized controlled phase III study Arms: neratinib + capecitabine (N + C) vs. lapatinib + capecitabine (L + C) | 101 patients with HER2+ BCBM (stable and asymptomatic at baseline) | No improvement in PFS and OS. Significant improvement of intracranial ORR with N + C (26.3% vs. 15.4%) |
TBCRC-022 (Freedman, Gelman et al. 2016 [105]) | Multicenter single-arm phase II study on neratinib | 40 patients with HER2+ BCBM after WBRT/SRS/surgery | Median OS was 8.7 months, median PFS was 1.9 months. Intracranial ORR was 8% |
PATRICIA (Lin, Pegram et al. 2021 [106]) | Multicenter single-arm phase II study on neratinib/capecitabine | 49 patients with HER2+ BCBM | Outcome in lapatinib-naïve patients: median OS was 13.3 months, median PFS was 5.5 months. Intracranial ORR was 49% |
NEfERT‑T (Awada, Colomer et al. 2016 [107]) | Multicenter single-arm phase II study on pertuzumab + high-dose trastuzumab | 39 patients with HER2+ BCBM previously treated with WBRT/SRS | Intracranial ORR was 11%, 6 months clinical benefit rate (CR/PR/SD) 51%, grade ≥ III toxicity in 44% of patients, no grade V toxicity |
Recommendations
-
The biological subtype and breast cancer-specific GPAs should be considered for evaluation of prognosis in patients with BCBM.
-
Limited brain metastases (n = ≤ 4):
-
Local therapy including SRS/SRT is generally recommended irrespective of molecular subtype and systemic therapy.
-
In case of limited intact BCBM (n = ≤ 4), SRS/SRT should be used.
-
After resection with a limited number of remaining BCBM (n = ≤ 4), SRS/SRT to the resection cavity should be used as postoperative treatment with additional SRS/SRT of the intact BCBM.
-
-
Multiple brain metastases:
-
SRS should be considered in case of n = 5–10 intact BCBM (cumulative volume < 15 ml); alternatively, WBRT can be applied.
-
After resection of BCBM and limited further BCBM (n = 5–10 and < 15 ml), SRS/SRT to the resection cavity and remaining intact BCBM is a possible option. Alternatively, WBRT can be applied.
-
In disseminated brain metastases (n = > 10), WBRT is generally recommended.
-
After interdisciplinary discussion, in cases of asymptomatic disseminated brain metastases (n = > 10) or in multiple BCBM if SRS/SRT is not feasible, WBRT can be postponed with early reassessment and reevaluation of local treatment options (8–12 weeks) if HER2-targeted systemic therapy with significant response rates in the CNS (tucatinib/trastuzumab/capecitabine, trastuzumab deruxtecan) is being used.
-
-
Leptomeningeal carcinomatosis:
-
In symptomatic leptomeningeal carcinomatosis, local radiotherapy (WBRT/involved-field SRS/SRT or local spinal irradiation) should be administered to symptomatic lesions in addition to systemic therapy.
-
In case of patients with disseminated leptomeningeal carcinomatosis in good clinical condition and with limited, stable extra-CNS disease, CSI may be considered.
-
-
Technique:
-
The decision on the optimal fractionation regimen is dependent on the size and location of the metastases/resection cavity.
-
In case of WBRT, hippocampal avoidance should be considered, especially for patients with a good prognosis according to GPA score.
-
Administration of memantine or donepezil may be considered with WBRT (off-label use in Germany).
-
-
Concurrent systemic therapy:
-
There is a general lack of data regarding the combination of systemic therapy and SRS/SRT for brain metastases.
-
Each case should be discussed individually in an interdisciplinary setting based on the type of systemic therapy, size and location of the metastases, as well as planned dose and fractionation.
-
Particular caution should be taken when administering SRS/SRT concurrently (≤ 7 days before or ≤ 21 days after) with antibody drug–conjugates.
-