CDKN2A/B homozygous deletion

What is measured: biallelic loss of the CDKN2A and CDKN2B tumour suppressors at 9p21, which encode p16, p14ARF and p15. How: FISH with a 9p21 probe (also on effusion cytology in mesothelioma), copy-number from methylation or SNP arrays, next-generation sequencing copy-number calls, or MTAP immunohistochemistry as a surrogate because MTAP is co-deleted in 80 to 90 percent; p16 immunohistochemistry loss is less specific. Frequencies: 15 to 20 percent of IDH-mutant astrocytomas (grade 4 by definition in WHO 2021 whatever the histology), a defining feature of grade 3 meningioma, about 45 percent of mesotheliomas, most pancreatic cancers, many melanomas and sarcomas; germline CDKN2A causes familial melanoma. What a positive result changes: in glioma it assigns grade 4 and full radiotherapy with temozolomide, and excludes the patient from de-escalation; in meningioma it assigns grade 3 and radiotherapy; in mesothelioma it separates malignancy from reactive mesothelial proliferation; MTAP co-deletion opens trials of PRMT5 inhibitors (MRTX1719, AMG 193) and MAT2A inhibitors; CDK4/6 inhibitor trials have shown only modest activity. Where it matters: IDH-mutant astrocytoma, oligodendroglioma, glioblastoma, meningioma, paediatric gliomas, mesothelioma, HPV-negative head and neck cancer, MPNST, sarcomas, papillary RCC and Richter transformation.