Spatial and temporal evolution of distal 10q deletion, a prognostically unfavorable event in diffuse low-grade gliomas

Abstract

Background: The disease course of patients with diffuse low-grade glioma is notoriously unpredictable. Temporal and spatially distinct samples may provide insight into the evolution of clinically relevant copy number aberrations (CNAs). The purpose of this study is to identify CNAs that are indicative of aggressive tumor behavior and can thereby complement the prognostically favorable 1p/19q co-deletion.

Results: Genome-wide, 50 base pair single-end sequencing was performed to detect CNAs in a clinically well-characterized cohort of 98 formalin-fixed paraffin-embedded low-grade gliomas. CNAs are correlated with overall survival as an endpoint. Seventy-five additional samples from spatially distinct regions and paired recurrent tumors of the discovery cohort were analyzed to interrogate the intratumoral heterogeneity and spatial evolution. Loss of 10q25.2-qter is a frequent subclonal event and significantly correlates with an unfavorable prognosis. A significant correlation is furthermore observed in a validation set of 126 and confirmation set of 184 patients. Loss of 10q25.2-qter arises in a longitudinal manner in paired recurrent tumor specimens, whereas the prognostically favorable 1p/19q co-deletion is the only CNA that is stable across spatial regions and recurrent tumors.

Conclusions: CNAs in low-grade gliomas display extensive intratumoral heterogeneity. Distal loss of 10q is a late onset event and a marker for reduced overall survival in low-grade glioma patients. Intratumoral heterogeneity and higher frequencies of distal 10q loss in recurrences suggest this event is involved in outgrowth to the recurrent tumor.

Figure 1

Accrual of samples and clinical data of the discovery cohort. Samples were selected based on criteria listed in the boxes connected with vertical arrows. Reasons for exclusion of samples are listed in boxes in the right panel, connected with horizontal arrows. The number below each box represents the number of patients. Boxes below the dotted line list criteria for collection of recurrent tumors and spatially distinct regions of LGGs. AII, astrocytoma; H&E, hematoxylin and eosin; OII, oligodendroglioma; OAII, oligoastrocytoma.

Figure 2

Unsupervised clustering of CNAs in the discovery cohort. Histological subtypes and patients are color-coded on the x-axis and chromosomes are ordered on the y-axis, 1 to 22 from bottom to top. Shades of green enable visualization of individual chromosomal arms, their size varying by the number of regions. Hence, a chromosomal arm with many breakpoints based on CNAs is depicted as larger compared with one with fewer breakpoints. Red, copy number loss; blue, copy number gain; black, no CNA. OII, oligodendroglioma; OAII, oligoastrocytoma; AII, astrocytoma.

Figure 3

Kaplan Meier plots for distal 10q loss and 1p/19q co-deletion in (A) discovery, (B) validation and (C) confirmation cohorts. The dark blue line indicates loss of distal 10q without 1p/19q co-deletion versus the rest of the cohort (n =15, P-value =0.001 in (A), n =8, P-value =0.018 in (B), and n =14, P-value =0.0018 in (C). The green line indicates 1p/19q co-deletion without distal loss of 10q (n =38, P-value =0.0001 in (A), n =41, P-value =0.0005 in (B), and n =47, P-value =0.74 in (C). The light blue line indicates 10q loss and 1p/19q co-deletion (n =3, P-value =0.39 in (A), n =4, P-value =0.94 in (B), and n =0 in (C). The grey line indicates neither 10q deletion nor 1p/19q co-deletion (n =42 in (A), n =73 in (B), and n =123 in (C). The y-axis represents the fraction of patients alive, cumulative survival (Cum. Surv.), and the x-axis time in months. Censored patients are indicated with a vertical bar.

Figure 4

Chromosomal copy number profiles for sample 240 demonstrating intratumoral copy number heterogeneity. (A) CNA profile of initial tumor, clonal 1p/19q co-deletion, loss of chromosome 4 and intermediate level of loss of chromosome 10. (B) Hematoxylin and eosin stained slide showing regions used for DNA isolation: the red dotted line corresponds to the region used for chromosomal profile of 4A and regions outlined with a solid black line (labeled i to iv) were used for the chromosomal profiles of 4C. (C) CNA profiles from four non-overlapping regions. Insets at the top right corner of each profile show histological features representative for individual regions. In all regions the histopathological diagnosis was LGG, although within a tumor the regions analyzed for spatial heterogeneity often showed some variation in microscopic features, such as cellularity and nuclear size and shape. (i) Clonal 1p/19q co-deletion and loss of chromosome 4; (ii) clonal 1p/19q co-deletion, loss of chromosome 4 and intermediate loss of chromosome 10; (iii) clonal 1p/19q co-deletion, loss of chromosome 4, intermediate loss of chromosome 10 and intermediate gain of chromosome 11; (iv) clonal 1p/19q co-deletion, intermediate loss of chromosomes 4, 10q, 13, 15 and 18. The y-axis represents normalized log2 sequence read counts per bin, and the x-axis represents 15 kb bins ordered by genomic position from chromosomes 1 to 22.

Figure 5

Spatial and temporal evolution of CNAs in LGGs and paired recurrent tumors. (A) CNAs in spatially distinct regions of LGGs of 17 patients. CNAs are categorized by detection in all regions (left panel), more than one region but not all regions (middle), or one region (right). Patients are ordered by the number of regions analyzed of each LGG from high to low. (B) Summary of prognostically relevant CNAs in spatially distinct regions and histology. No intratumoral heterogeneity was observed for 1p/19q co-deletion in any of the tumors, while distal 10q loss was often only detected in subclones. OII, oligodendroglioma; AII, astrocytoma; OAII, oligoastrocytoma. (C) CNAs in initial and paired recurrent tumors of 20 patients. CNAs are categorized by detection in initial tumor only (left panel), both initial and recurrence (middle) or detection uniquely in the recurrence (right). Patients are ordered by the histological malignancy grade of the recurrent tumor. (D) Summary of prognostically relevant CNAs in paired initial and recurrent tumors. 1p/19q co-deletion is stable over time, while distal 10q loss surfaces in recurrences, including two with a higher malignancy grade than LGG. Patients are color-coded on the x-axis and chromosomes are ordered on the y-axis, 1 to 22 from bottom to top. Shades of green enable visualization of individual chromosomal arms, their size varying by the number of regions. Hence, a chromosomal arm with many breakpoints based on CNAs is depicted as larger compared with one with fewer breakpoints. CNAs smaller than 5 Mbp were excluded from this figure. Red, copy number loss; blue, copy number gain; black, no CNA. The arrowhead indicates patient 240, the black squares three LGGs analyzed for both spatial and temporal evolution.

Figure 6

CNAs in initial and recurrent tumors. (A) Gains; (B) losses. The top of each graph shows the initial tumors, and the bottom the recurrences. Partial loss of chromosomal arm 4q, 9p and 10q were more frequently detected in recurrences. Bins are ordered by genomic position and from chromosomes 1 to 22 on the x-axis; percentages of cases showing CNAs are depicted on the y-axis.