Intraclonal diversity in follicular lymphoma analyzed by quantitative ultradeep sequencing of noncoding regions

Abstract

Cancers are characterized by genomic instability, and the resulting intraclonal diversity is a prerequisite for tumor evolution. Therefore, metrics of tumor heterogeneity may prove to be clinically meaningful. Intraclonal heterogeneity in follicular lymphoma (FL) is apparent from studies of somatic hypermutation (SHM) caused by activation-induced deaminase (AID) in IGH. Aberrant SHM (aSHM), defined as AID activity outside of the IG loci, predominantly targets noncoding regions causing numerous "passenger" mutations, but it has the potential to generate rare significant "driver" mutations. The quantitative relationship between SHM and aSHM has not been defined. To measure SHM and aSHM, ultradeep sequencing (>20,000-fold coverage) was performed on IGH (~1650 nt) and nine other noncoding regions potentially targeted by AID (combined 9411 nt), including the 5' untranslated region of BCL2. Single-nucleotide variants (SNVs) were found in 12/12 FL specimens (median 136 SHMs and 53 aSHMs). The aSHM SNVs were associated with AID motifs (p < 0.0001). The number of SNVs at BCL2 varied widely among specimens and correlated with the number of SNVs at eight other potential aSHM sites. In contrast, SHM at IGH was not predictive of aSHM. Tumor heterogeneity is apparent from SNVs at low variant allele frequencies; the relative number of SNVs with variable allele frequency < 5% varied with clinical grade, indicating that tumor heterogeneity based on aSHM reflects a clinically meaningful parameter. These data suggest that genome-wide aSHM may be estimated from aSHM of BCL2 but not SHM of IGH. The results demonstrate a practical approach to the quantification of intratumoral genetic heterogeneity for clinical specimens.

Figure 1

SNVs in the non-coding region of BCL2 are attributable to AID. AID preferentially targets C residues in the motif 5′-WRCY (W is A or T; Y is C or T; R is A or G). The number of SNVs at C within WRCY compared to SNVs at C not in the motif provides an indicator of AID activity (labels on each bar: number within WRCY / number NOT within WRCY). If mutation of C residues were random, they would be expected to occur at the ratio of 0.17, reflecting the number C within WRCY and outside of WRCY. For the first bar, the numbers show the total count of C residues within (588) and outside WRCY (3529). The numbers of observed mutations at C are shown in the remaining bars, demonstrating a profound skewing of C mutations to occur within the WRCY motif. The skewing is apparent for SNVs regardless of VAF, indicating that the low frequency SNVs identified by DDiMAP are valid. (p<0.0001 in each subgroup; chi-square for 2×2 based on each group versus expected). Similar results were obtained for the WA motif associated with POLH repair of AID-induced damage (data not shown). Data are for the 850 nt 5′ UTR of BCL2 from all 12 FL specimens combined.

Figure 2

There is wide variation in the number SNVs detected in the 5′non-coding region of BCL2 and in IGH with no correlation between these parameters (Pearson's test r=0.14). Cases are ordered by number of SNVs in BCL2. (BCL2, filled bars, left-hand axis; IGH, open bars, right-hand axis). The sequenced region in BCL2 is 850 nt in all 12 cases. The amount of IGH region sequenced (coding and non-coding up to the 3′ end of JH6) varied among specimens depending on the junctional JH used in the productively rearranged IGH allele. Even with correction for this variation, there was no significant correlation between the number of SNV in the 5′non-coding region of BCL2 and in IGH (data not shown; Pearson's r=0.26).

Figure 3

Aberrant Somatic Hypermutation at 8 non-IGH sites correlates with aSHM of BCL2 (A) but not SHM of IGH (B). The same data used in Fig. 1 are the abscissa (BCL2 in A; IGH in B) and the aSHM measured for 8 non-IGH putative aSHM sites (BCL6, PIM1, PAX5, RHOH, CD83, MYC (2 transcription start sites), and SYK) are the ordinate. Pearson's r correlation is 0.92 for panel A; exclusion of the single most extreme datum in panel A (farthest right-hand point) decreases r to 0.74, still a highly significant correlation. No correlation was detected between genome-wide aSHM and SHM of the IGH (panel B).

Figure 4

A representative plot of the cumulative SNVs ordered by Variant Allele Frequency (VAF) demonstrates the preponderance of SNVs are at low VAF. Each SNV in a single case is represented by a single point. These data are for patient 6 in Fig 2. Panel A shows that these SNVs are scattered throughout the IGH coding (black points) and adjacent non-coding region (gray points). Furthermore, the vast majority of data is derived from the non-coding portion of IGH, a region in which cells carrying SNVs are presumably not subject to the selective pressures. Panel B shows the count of these SNVs from lowest VAF to highest; the steepest portions of the cumulative SNV plot reflect the range of VAFs that are most highly represented in a specific specimen.

Figure 5

Variants of the IGH locus were detected in all 12 follicular lymphoma specimens (ordered as in Figure 2). The total number of SNVs is shown for each patient. In general, the SNVs with VAFs under 0.05 were most numerous and there was a monotonic decrease in the abundance of SNVs as the VAF increased. (Fig. 3B is displayed as patient 6 for comparison.) SNVs for the IGH are defined as relative to the productively rearranged IGH sequence obtained from the bulk tumor. Therefore, the SNVs for IGH reflect the current variation around the dominant population.

Figure 6

Variants in BCL2 were detected in all 12 follicular lymphoma specimens and the cumulative SNV curve allows distinction of SNVs present in founder (or the currently most dominant clone) and those reflecting newly derived subpopulations within the dominant clone (ordered as in Fig.2). The total number of SNVs detected is shown for each patient. The vertical component of the cumulative curves at high VAF is termed the “clonal” phase and the SNVs with lower VAF, reflecting the variation in the population, is termed the “sub-clonal” phase; the numbers of SNVs in the clonal (C) and sub-clonal (S) phases are shown. In contrast to the IGH data, SNVs for BCL2 are defined as relative to germline sequence. Therefore, the SNVs for BCL2 reflect all variation present in the founder cell for the neoplasm and those that have subsequently arisen during evolution of the neoplasm. To distinguish which specimens showed significant variation from the behavior of the aggregate of specimens, the aggregate number of SNVs in the clonal and sub-clonal phases for all specimens was determined (131 and 235; p values by binomial distribution are bold). Specimens #5 and #7, the only grade 3 specimens in the collection, showed highly significant skewing from the aggregate (p=2×10^-7 and 3.4×10^-4, respectively).