Abstract 998: Rise and fall of subclones from diagnosis to relapse in pediatric B-acute lymphoblastic leukemia (B-ALL): A report from the children's oncology group (COG) - Target - St. Jude Pediatric Cancer Genome Project
2014
Proceedings: AACR Annual Meeting 2014; April 5-9, 2014; San Diego, CA
Tumor clonal heterogeneity has been demonstrated in several cancers; however, the rise and fall of subclones under therapy have not been well characterized. To gain insight into the subclonal population architecture of pediatric B-ALL, we analyzed somatic lesions including sequence mutations, structural alterations and DNA copy number abnormalities derived from high coverage (200X) exome sequencing, whole-genome sequencing and SNP arrays of diagnosis (Dx)-remission-relapse DNA from 20 patients (median 7 yrs, range 2 to 19) treated on recent COG B-ALL trials. Cases were selected for analysis based upon occurrence of an early bone marrow relapse (median 19.2 months, range 3.8 to 35.7), which is associated with poor survival. We identified a high frequency of mutations in B-cell development (80%, e.g. PAX5, IKZF1, TCF3, BTG1, TOX, and EBF1 ), RAS signaling (65%, e.g. NRAS, KRAS, PTPN11 and FLT3), TP53 (60%, e.g. CDKN2A, TP53 and RB1), kinase signaling (25%, e.g. JAK2, CRLF2 and SH2B3) and chromatin remodeling (60%, e.g. WHSC1, MLL2, CREBBP, ARID2 and SETD2) pathways. Somatic lesions in these pathways were mostly retained (68%) between Dx and relapse. In contrast, mutations in the purine metabolism genes NT5C2 (45%) and NT5CB1 (15%) were relapse-specific. We constructed clonal lineage of somatic lesions for 15 cases based on subclonal population fractions estimated by a binomial mixture model that adjusts for sequence coverage of mutant alleles. The subclone number (median 3, range 1-5) at Dx was comparable to that at relapse (median=3, range 2-4). Notably, 6 Dx samples were observed with multiple subclones harboring distinct driver mutations in the same oncogene (e.g. NRAS, KRAS and JAK2). In almost all cases only one subclone from Dx arose to be the predominant clone present at relapse. Moreover, the mutation burden in an emergent subclone was comparable to those eradicated by therapy (P=0.43, Wilcoxon rank sum). In 80% of cases, the predominant clone at relapse originated from the smallest Dx subclone; in 45% of those cases, mutant allele present in relapse was detectable in remission DNA (∼0.1%) obtained at the end of the 1st month of therapy. Clonal lineage shows that the earliest acquired mutations at relapse are NT5C2 (n=5), NRAS (n=3), USH2A (n=3), WHSC1 (n=2), TP53 (n=2), IKZF1 (n=1) and CREBBP (n=1). Our study is the first that analyzes the genetic composition and population architecture of subclones that rise or fall after B-ALL therapy. The majority of the “rising” subclones are oligoclonal at Dx, which may explain acquisition of mutations such as those found in NT5C2 to confer growth advantage in relapse. Additional studies of patients who were cured of B-ALL are needed to determine if the presence of mutant clone at end induction might help to predict risk and tempo of relapse.
Citation Format: Xiaotu Ma, Mignon L. Loh, Michael Rusch, Michael Edmonson, Richard C. Harvey, David A. Wheeler, Oliver A. Hampton, John Easton, Donald Yergeau, Bhavin Vadodaria, Gang Wu, William L. Carroll, I-Ming Chen, Daniela S. Gerhard, Julie M. Gastier-Foster, Mary V. Relling, Malcolm A. Smith, Meenakshi Devidas, Jaime M. Guidry Auvil, James R. Downing, Cheryl L. Willman, Charles G. Mullighan, Stephen P. Hunger, Jinghui Zhang. Rise and fall of subclones from diagnosis to relapse in pediatric B-acute lymphoblastic leukemia (B-ALL): A report from the children's oncology group (COG) - Target - St. Jude Pediatric Cancer Genome Project. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 998. doi:10.1158/1538-7445.AM2014-998
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI