Intracranial primary collision tumours: A comprehensive systematic review on preoperative radiological accuracy and neuro-oncological insights.

Background: Intracranial collision tumours, characterized by the coexistence of two histologically distinct neoplasms within the same anatomical region without histological transition or metastatic interaction, are rare in neuro-oncology. Their atypical imaging appearance often mimics solitary lesions, posing diagnostic challenges. Research question: How accurately can preoperative neuroimaging identify both components of intracranial collision tumours, and what factors influence detection? Methods: A systematic review was conducted following PRISMA guidelines and registered with PROSPERO (CRD420251008646). Included studies were adult case reports and series, including histologically confirmed intracranial collision lesions with preoperative neuroimaging. Tumour-to-tumour metastasis, synchronous, composite, or recurrent tumours were excluded. Results: A total of 67 published cases were analysed, with a male-to-female ratio of 28:38 and a mean age of 52.4 years (SD = 15.95), ranging from 18 to 87 years. Meningioma was the most prevalent tumour type (65.7 %), commonly paired with glioblastoma (26.9 %). The sellar region was the most frequent location (34.3 %), followed by the two frontal lobes (31.3 %). Preoperative diagnosis correctly identified both lesions in only 26.9 % of cases. Detection rates varied by anatomical location (p = 0.0095), whereas no clear association was observed with tumour pair type (p = 0.1351). Surgical resection was the primary treatment, frequently combined with chemo-radiotherapy. Recurrence occurred in 17.9 %, especially in high-grade tumour components such as glioblastoma. Mean survival was 8.6 months, with 11.9 % mortality. No statistically significant survival differences were observed between tumour pair types (p = 0.149). Conclusion: Intracranial collision tumours remain diagnostically challenging. Improved neuroimaging and molecular understanding are crucial to enhance early diagnosis and optimize clinical management.