Abstract A045: Cooperation between germline and somatic drivers in childhood cancer: insights from the ZERO cohort

Background. Tumorigenesis is a multistep process involving germline and somatic alterations, also known as "drivers." Understanding how these alterations cooperate is crucial for assessing cancer risk and developing strategies to intercept malignant transformation. Objective. To investigate the interplay between germline pathogenic variants (GPVs) and somatic drivers in childhood cancers, and to compare tumor molecular profiles based on the origin of initiating alterations. Methods. We analyzed whole-genome (paired tumor-germline) and transcriptomic data from 2,397 children enrolled in the ZERO Childhood Cancer Program (2018–2025). A subset of 199 participants with autosomal dominant cancer predisposition syndromes consistent with their tumor type was selected for detailed analysis. Results. GPVs were distributed across multiple cancer predisposition genes (CPGs), with TP53, NF1, SMARCB1, and DICER1 being the most prevalent. A second somatic hit in the same CPG was identified in 83% of cases (166/199), 56% resulting from loss of heterozygosity. RNA sequencing revealed additional variants (10% of cases) not detected in the DNA data. Additional somatic drivers were identified in 149 tumors (75%), 82 of which exhibited recurrent drivers for their cancer type (e.g., CDKN2A in NF1-driven malignant peripheral nerve sheath tumor). In 33% of them (27/82), no second hit was present in the CPG (e.g., NF1 GPV in low-grade glioma with somatic FGFR1 mutation), suggesting the GPV may act as a cooperating enabler. Of interest, 67 tumors harbored additional drivers atypical for their cancer type, with most (62/67) showing a second hit in the same CPG, suggesting a distinct molecular profile in this group of GPV-driven tumors (e.g. GPV in FBXW7 and Wilms tumor with somatic CDKN2A loss). In 49 cases, the only somatic event was a somatic second hit in the CPG, indicating that some GPVs may independently initiate tumorigenesis (e.g., SMARCB1, DICER1). This supports the hypothesis that GPVs can enable and promote tumour progression without the need for cooperation with other genomic drivers. Conversely, 25 cases lacked a second hit in the CPG but harbored other somatic drivers, including unexpected tumor alterations (e.g., GPV in RECQL4 and osteosarcoma without TP53 alteration), suggesting possible GPV haploinsufficiency in driving tumorigenesis through an alternative molecular pathway. Conclusions. Our study highlights the multifactorial nature of tumour initiation and promotion in GPV-driven tumorigenesis involving direct initiation and cooperation with somatic drivers. Genes like SMARCB1 can drive cancer alone, while others like NF1 need additional somatic alterations for malignancy. Notably, biallelic inactivation of CPG rarely leads directly to high-grade tumors, emphasizing gene and cancer-type specificity of this process. Future research, such as the study of induced pluripotent cancer stem cells, can help clarify the role of GPV and the sequence of multi-step events. This may guide strategies for preventing malignant transformation and enabling earlier detection. Citation Format: Noemi A Fuentes-Bolanos, Dianne Sylvester, Eliza Courtney, Chelsea Mayoh, Meera Warby, Dong Anh Khuong Quang, Paulette Barahona, Ashleigh Sullivan, Ann Altekoester, Megan Rumford, Hetal Dholaria, Marie Wong, Vaness Tyrrell, Michelle Haber, Katherine Tucker, Mark J Cowley, David S Ziegler, Mark Pinese, Loretta M S Lau. Cooperation between germline and somatic drivers in childhood cancer: insights from the ZERO cohort [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Discovery and Innovation in Pediatric Cancer— From Biology to Breakthrough Therapies; 2025 Sep 25-28; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85(18_Suppl_2):Abstract nr A045.