Intratumoral and intracranial hemorrhage associated with MAPK-pathway targeted therapy: a systematic review and mechanistic synthesis.
MAPK-pathway inhibitors including BRAF, MEK, and type II RAF inhibitors are now integral to treatment of pediatric low-grade glioma (pLGG), BRAF V600-mutant glioma, NF1-associated tumors, and melanoma brain metastases. Intratumoral and intracranial hemorrhage has emerged as a clinically relevant safety signal, but drug-specific incidence estimates, phenotypic definitions, and contributing mechanisms remain incompletely characterized. We systematically reviewed the evidence and synthesized contributing mechanisms.
We performed a systematic review of PubMed/MEDLINE and Embase from database inception through May 2026, supplemented by FDA prescribing information, to identify studies reporting intratumoral or intracranial hemorrhage with MAPK-pathway targeted agents across tumor types. Results were synthesized narratively given heterogeneity in reporting definitions and study designs.
CNS hemorrhage signals were identified across agents. Dabrafenib-based regimens in melanoma brain metastases produced ICH rates of 6% (BREAK-MB monotherapy) and a fatal hemorrhage in 0.8% (COMBI-MB combination). In the phase 2 FIREFLY-1 trial of tovorafenib in BRAF-altered pLGG (n = 137), any-site hemorrhage of any grade occurred in 42% of patients with grade 3-4 hemorrhage in 7 patients (5%) and, separately, one fatal grade 5 intratumoral hemorrhage; in the overlapping pooled FDA-label safety population (N = 140), intratumoral hemorrhage specifically was reported in 9%. Combining BRAF inhibitors with stereotactic radiosurgery was associated with 3-fold higher odds of ICH versus radiosurgery alone (OR 3.16; 95% CI 1.43-6.96).
MAPK-pathway inhibitors are associated with a consistent but heterogeneous CNS hemorrhage signal. Reported incidence varies by agent, population, and hemorrhage ascertainment method (active imaging-based surveillance versus clinical reporting), from < 1% symptomatic ICH with dabrafenib in adult melanoma to 9% intratumoral hemorrhage in the pooled pediatric tovorafenib safety population. Standardized definitions, CNS-specific CTCAE capture, and prospective imaging surveillance are needed to define true incidence and risk as MAPK-directed therapy expands in neuro-oncology.
This systematic review was not prospectively registered in PROSPERO.
We performed a systematic review of PubMed/MEDLINE and Embase from database inception through May 2026, supplemented by FDA prescribing information, to identify studies reporting intratumoral or intracranial hemorrhage with MAPK-pathway targeted agents across tumor types. Results were synthesized narratively given heterogeneity in reporting definitions and study designs.
CNS hemorrhage signals were identified across agents. Dabrafenib-based regimens in melanoma brain metastases produced ICH rates of 6% (BREAK-MB monotherapy) and a fatal hemorrhage in 0.8% (COMBI-MB combination). In the phase 2 FIREFLY-1 trial of tovorafenib in BRAF-altered pLGG (n = 137), any-site hemorrhage of any grade occurred in 42% of patients with grade 3-4 hemorrhage in 7 patients (5%) and, separately, one fatal grade 5 intratumoral hemorrhage; in the overlapping pooled FDA-label safety population (N = 140), intratumoral hemorrhage specifically was reported in 9%. Combining BRAF inhibitors with stereotactic radiosurgery was associated with 3-fold higher odds of ICH versus radiosurgery alone (OR 3.16; 95% CI 1.43-6.96).
MAPK-pathway inhibitors are associated with a consistent but heterogeneous CNS hemorrhage signal. Reported incidence varies by agent, population, and hemorrhage ascertainment method (active imaging-based surveillance versus clinical reporting), from < 1% symptomatic ICH with dabrafenib in adult melanoma to 9% intratumoral hemorrhage in the pooled pediatric tovorafenib safety population. Standardized definitions, CNS-specific CTCAE capture, and prospective imaging surveillance are needed to define true incidence and risk as MAPK-directed therapy expands in neuro-oncology.
This systematic review was not prospectively registered in PROSPERO.