Germline TP53 mutations is the genetic basis of Li-Fraumeni syndrome (LFS), a highly penetrant cancer predisposition condition. Recent reports of germline TP53 variants in children with hypodiploid acute lymphoblastic leukemia (ALL) suggest that childhood leukemia is another manifestation of LFS. However, the pattern, prevalence, and clinical relevance of pathogenic variants in TP53 in childhood ALL remain unknown.To comprehensively characterize genetic variants in TP53 in childhood ALL, we performed targeted sequencing of TP53 coding regions in 3858 children from the Children's Oncology Group front-line ALL clinical trials (COG AALL0232 and P9900). In total, 49 different non-silent coding rare TP53 variants were identifiedin 77 (2.0%) patients sequenced, including 40 missense, 1 nonsense, 6 frameshift, and 2 inframe deletion variants. To determine which TP53 variants are potentially pathogenic and related to ALL risk, we examined their experimentally-validated p53 transcriptional activity, bioinformatically-predicted damaging effects on TP53 function, and also the variant frequency in non-ALL populations. Twenty-two TP53 variants were classified as ALL risk-related (VAR), including 7 protein-truncating and 15 missense variants, all of which were either absent or exceedingly rare in non-ALL populations (i.e., allele frequency < 0.006% in the ExAC dataset). Fourteen missense VAR reside in the p53 core DNA-binding domain, with two affecting residues essential for DNA contact and seven located at residues that are important for the overall architecture of the DNA-binding surface. The 27 remaining TP53 variants were classified as “variants of unknown significance” (VUS). Together, of 3858 children in this ALL cohort, 26 (0.67%) patients had a predicted VAR and 51 (1.3%) had a VUS in the TP53 gene.We next evaluated the association of germline TP53 variants with patient demographics and ALL tumor genetic characteristics. Children with a TP53 VAR were significantly older at diagnosis (P < 0.0001), and had significantly lower leukocyte count at presentation than those with a VUS or wild-type (WT) genotype (P= 0.006). Of 26 patients carrying a germline TP53 VAR, 17 (65.4%) exhibited hypodiploidy in ALL blasts (mostly low hypoploidy or masked hypoploidy). In contrast, hypodiploid ALL was only present in 3.9% and 1.2% of children with TP53 VUS or WT genotype, respectively. The prevalence of TP53 VAR did not differ by genetic ancestry (P= 0.9). We also examined the relationship between germline TP53 variants and treatment outcomes of ALL therapy. In the COG AALL0232 cohort, the presence of a TP53 VAR was associated with a significantly lower event free survival (EFS) and overall survival (OS) compared to patients without a risk variant (hazard ratio [HR] of 2.84, P= 0.0007 for EFS; and HR of 3.13, P= 0.003 for OS). In multivariate analyses, TP53 VAR remained prognostic after adjusting for ancestry, age, leukocyte count at diagnosis, and minimal residual disease at the end of remission induction therapy (HR of 3.36, P= 0.0002 for EFS; HR of 2.88, P= 0.007 for OS). Importantly, there was a markedly higher risk of second cancers among patients with TP53 VARs compared to those with TP53 VUS or WT TP53: 5-year cumulative incidence of 25.1% vs. 0.7%, P= 5.4 × 10-11. In the COG P9900 cohort, only 4 patients harbored TP53 VAR, of whom 3 cases experienced adverse events (one patient each with ALL relapse, second cancer, or death in remission), and both EFS and OS were significantly worse than patients with TP53 VUS or WT genotype (HR of 7.14, P= 0.0008 for EFS; HR of 14.19, P= 1.1 x 10-5 for OS). Taken together, 14 of 26 patients with a TP53 VAR experienced adverse events, with 5 ALL relapses (36% of all events) and 5 second cancers (36%). This pattern of adverse events was dramatically different from those in patients with WT TP53 or VUS, for whom ALL relapse accounted for 75% of all events with only 4% as second cancers (P= 1.2 × 10-7). In fact, even within hypodiploid ALL patients who experienced adverse events, the frequency of second cancer was significantly higher in those with a TP53 VAR than those without (50% vs 5%, P= 0.01), further suggesting that germline TP53 variation was the underlying cause of second cancers in these cases.In conclusion, our findings identified germline TP53 variants as an important determinant of ALL predisposition and adverse treatment outcomes, particularly the risk of second malignant neoplasms. DisclosuresBorowitz:Beckman Coulter: Honoraria; Becton-Dickinson Biosciences: Research Funding; HTG Molecular: Honoraria. Mullighan:Loxo Oncology: Research Funding; Amgen: Consultancy. Hunger:Jazz Pharmaceuticals: Honoraria; Novartis: Consultancy; Erytech Pharmaceuticals: Consultancy; Amgen: Consultancy, Equity Ownership.
Read full abstract