Cerebral cavernous malformations (CCMs), or cavernomas, are abnormal vascular growths that arise in the central nervous system and have no approved medical treatment. Considered a monogenic disease caused by loss of function mutations in the CCM genes, the vast majority of CCMs arise in adults, often presenting as a single fast-growing lesion that culminates in hemorrhage or seizure. Though environmental differences explain the heterogeneity between individuals, why a single cavernoma suddenly becomes aggressive is unknown. Our study reveals a synergistic interaction that underlies a significant proportion of aggressive human CCMs and answers this long-standing question in the field. Using mouse genetic models, we show that cavernoma growth requires two distinct inputs, one through loss of the CCM protein complex as previously described, and another through gain of PI3K signaling, which can be provided either by endogenous angiogenic signals in the neonatal mouse model or by activating mutations in PIK3CA commonly found in cancer. We generated a novel adult mouse model that confirms neither CCM loss nor PI3K gain alone are sufficient, but together, lead to rapid growth of a cavernoma. We identify somatic mutations in PIK3CA in 71% of resected human cavernomas and via single-nucleus DNA sequencing, reveal that PIK3CA mutations arise in the same cells as CCM gene mutations. Considering that asymptomatic CCMs are not resected, this suggests that PIK3CA mutations play a driving role in clinically symptomatic lesions. Our study establishes a “three-hit” mechanism analogous to cancer in which aggressive vascular malformations arise through the loss of vascular “suppressor genes” required to constrain vessel growth and gain of a vascular “oncogene” that stimulates excess vessel growth. To our knowledge, this is the first description of a compound genetic mechanism in vascular malformation pathogenesis. Consistent with these findings, the mTORC1 inhibitor Rapamycin effectively blocks CCM formation in both neonatal and adult mouse models. These studies suggest that clinically approved mTORC1 inhibitors are a promising therapeutic avenue for CCM disease, which at present can only be treated with neurosurgical resection.