Novel Associations Between Mutations, Prognostic and Clinical Parameters in Untreated Progressive CLL: Data from E1912, a Randomized Phase III Study of the ECOG-ACRIN Cancer Research Group

Sequencing studies have significantly improved our understanding of the CLL genetic landscape and allowed the identification of the first associations between specific mutations and clinical outcome. Hence, mutations in TP53, NOTCH1, SF3B1, BIRC3 and MYD88 are known to be associated with differential prognosis and have been incorporated in CLL risk stratification models. Despite the work performed on these 5 genes, several additional putative tumor-driver genes are recurrently mutated in CLL, but their clinical impact remains unclear.

In order to gain further understanding, we performed targeted sequencing in baseline samples of 308 CLL patients (pts) participating in the North American Intergroup study (NAIG) designated as E1912 and led by ECOG. This trial is a randomized Phase III of Ibrutinib (Ibr) and Rituximab (R) vs FCR in previously untreated CLL. Besides the 5 genes aforementioned, we screened another 19 genes recurrently mutated in CLL, and 2 genes (BTK and PLCy2) recently associated with Ibr resistance. The CD5+/CD19+ B lymphocyte population of these pts was flow-sorted. All coding regions were amplified using customized oligos. Libraries were templated and enriched using IonOneTouch2 and IonOneTouch ES systems, respectively. Samples were run on an IonTorrent sequencer using 318^TM chips. Variants were called using IonTorrent Somatic VariantCaller. VCF files were annotated using BioR, which provides annotation from NCBI/Ensembl, UCSC, dbSNP, HapMap, 1000Genomes, COSMIC, SIFT, and PolyPhen-2. Somatic variants with a Mapping Quality <20 or read depth <5X were removed. Variants of interest were inspected using Integrative Genomics Viewer. The mutational status of genes and pathways was analyzed for association with chromosomal abnormalities detected by FISH, expression of CD38, CD49d and ZAP70, Rai stage, age and gender. We were not able to evaluate associations to clinical responses for the different arms as this trial is still underway and remains blinded for clinical outcome.

We found 511 SNVs (442 missense, 45 nonsense and 24 splice sites) and 115 indels. An average coverage of 770X depth was obtained, allowing the detection of minor subclonal mutations (<10% of reads) in almost 35% (218/626 mutations). Overall, 215 pts (70%) carried at least one mutated gene, 102 (33%) carried 2 or more and 32 (10%) 3 or more mutated genes. The most commonly mutated gene was SF3B1 (22%), followed by NOTCH1 (19%), ATM (12%), XPO1 (11%), BIRC3 (9%), KRAS (7%), BRAF, DDX3X, EGR2, POT1 and TP53 (6% each). No BTK or PLCy2 mutations were found in progressive but untreated CLLs. In contrast to a recent report (Nature 2015;526:525-30), we found BRAF mutations located in the canonical hotspot p.V600E (25% of total mutations). Another 13% were located in amino acid K601 (p.K601E/N) and the remaining in the kinase domain (p.G469A, p.D594G, p.F595L, p.G596R, p.L597Q).Nearly 27% of cases with activating mutations in the RAS pathway (BRAF, KRAS, NRAS), showed >1 mutation affecting one or more genes of the pathway. Similar observations were found in mutational hotspots of BIRC3 (30% of mutated cases showed >1 mutation), MED12 (18%), SF3B1 (7%), XPO1 (6%), and NOTCH1 (5%), suggesting the existence of genetic subclones showing convergent evolution in a substantial number of CLL. We found multiple associations between mutations and known prognostic and clinical parameters. These included mutations in NOTCH1 associated with +12 (p=0.005), and expression of ZAP70 (p=0.01), CD38 (p=0.001) and CD49d (p=0.004). Mutant BIRC3 was associated with +12 (p<0.001) and CD38 expression (p=0.01). Mutations in RAS pathway, with +12 (<0.001), and expression of CD38 (p=0.002) and CD49d (p=0.05). ATM (p=0.007) and EGR2 (p=0.002) with -11q; POT1 with CD49d expression (p=0.01), and MYD88 with -13q (p=0.04) and lack of CD38 (p=0.01) and CD49d (p=0.01) expression.

This study, performed on a large cohort of untreated but progressive CLLs, provides novel data about the mutation status of driver genes and their associations with clinical and prognostic parameters. Our data indicate that convergent evolution is a common event in CLL, affecting several driver genes. Furthermore, it reinforces that BTK and PLCy2 mutations are not found in Ibr-naive pts. Finally, this study provides the genetic groundwork for subsequent prospective comparative analyses of the clinical outcome for CLL cohorts treated with Ibr-R vs FCR in this phase 3 NAIG trial.