Intratumor heterogeneity and branched evolution revealed by multiregion sequencing.
Author(s): Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, Martinez P, Matthews N, Stewart A, Tarpey P, Varela I, Phillimore B, Begum S, McDonald NQ, Butler A, Jones D, Raine K, Latimer C, Santos CR, Nohadani M, Eklund AC, Spencer-Dene B, Clark G, Pickering L, Stamp G, Gore M, Szallasi Z, Downward J, Futreal PA, Swanton C
Publication: N Engl J Med, 2012, Vol. 366, Page 883-92
PubMed ID: 22397650 PubMed Review Paper? No
Purpose of Paper
This paper investigated the effects of tumor heterogeneity and everolimus treatment on identification of mutations and ploidy determination by next generation sequencing (NGS) analysis of renal cell carcinoma (RCC) specimens. Select mutations were confirmed by Sanger sequencing and loss of function of known RCC genes was confirmed through immunohistochemical (IHC) staining.
Conclusion of Paper
Only 34- 37% of mutations identified were found in all regions of the tumor and phylogenetic clustering revealed branching rather than simple tumor progression. Importantly, increasing the sequencing depth only revealed a few additional mutations, indicating that the difference was true tumor heterogeneity. Mutations identified included those in genes previously implicated in RCC including those in VHL, KDM5C, SETD2, and MTOR and 88-95% were confirmed by Sanger sequencing. Ploidy analysis revealed some tumor region-specific changes in ploidy in 3 of 4 patients. Allelic imbalances differed between tumor regions from the same patient but occurred in all regions of chromosome 3p in all patients, chromosome 10q in Patient 2 and in chromosome 5q and 6q in Patient 4. Everolimus treatment did not affect mutational load or clonality. Importantly, when a gene signature was applied to the tumor regions from Patient 1, one region was classified as RCC subgroup A, while the remaining regions were classified as subgroup B.
This study investigated the effects of tumor heterogeneity and everolimus treatment on identification of mutations and ploidy determination by NGS of RCC specimens. Select mutations were confirmed by Sanger sequencing and loss of function of known RCC genes was confirmed through IHC staining. This study included biopsy and nephrectomy specimens from 4 patients with renal cell carcinoma as well as matched blood specimens. Specimens included biopsies of the tumor and metastasis before treatment (Patient 1), tumor (Patients 1, 2, 3, and 4), and perinephric metastasis (Patient 1) obtained by nephrectomy after 6 weeks of treatment with everolimus and a 1-week washout period, excised chest-wall metastasis obtained after 6 weeks of another cycle of everolimus treatment (Patient 1), and a liver metastasis after 9 weeks of additional everolimus (patient 2). Specimens were microdissected to decrease stromal contamination, split in half, and either snap-frozen in liquid nitrogen or formalin-fixed within 45 min. Tumor content was >60% for all specimens from Patient 1 and averaged 52% for specimens from Patient 2. DNA and RNA were purified from blood and biopsy specimens using the Qiagen DNeasy and the RNeasy extraction kits following the manufacturer’s instructions. SNPs were evaluated by whole exome DNA sequencing on a Illumina Omni2.5 and by mRNA profiling using Affymetrix Gene 1.0 arrays.
Summary of Findings:
NGS identified 101 nonsynonymous point mutations and 32 indels in specimens from Patient 1 and 100 nonsynonymous point mutations and 19 indels in specimens from Patient 2. Only 34% and 37% of the mutations identified in Patient 1 and Patient 2; respectively, were found in all regions of the tumor with 46% (59 of 128) of the mutations in Patient 1 found in multiple tumor regions and 23% (29 of 128) found in only a single region. When the sequencing depth for two regions was increased to 262x and 255x, only two additional mutations were identified, indicating that tumor heterogeneity cannot be compensated for by increased sequencing depth. For both patients, phylogenetic clustering revealed branching rather than simple tumor progression with one tumor region from Patient 1 demonstrating the presence of two subclones. Among the mutations identified were those in genes implicated previously in RCC including VHL, KDM5C, SETD2, and MTOR. Most of the mutations were found in only some of the regions of a tumor but the mutations identified in VHL were found in all tumor regions from Patient 1. IHC-staining confirmed that the mutations in SETD2, PTEN, and MTOR resulted in loss of function. Importantly, when a gene signature was applied to the tumor regions from Patient 1, one region was classified as RCC subgroup A while the remaining regions were classified as subgroup B.
Ploidy analysis in specimens revealed two distinct subtetraploid peaks in the chest metastasis from Patient 1, single subtetraploid peaks in tumor regions from Patient 4 and Patient 1, and tetraploid regions in specimens from Patient 4. Additional peaks in the ploidy profile were also observed in some tumor regions from patient 2. Allelic imbalance occurred in all tumor regions of chormosome 3p in all patients, in 10q in Patient 2, and in 5q and 6q in Patient 4.
Sanger sequencing confirmed 37 of 42 (88%) mutations identified in Patient 1 and 14 of 15 (95%) mutations in Patient 2. Specimens obtained before and after everolimus had comparable mutation loads and patterns indicating that everolimus did not affect mutational load or clonality.
- Neoplastic - Carcinoma
Analyte Technology Platform DNA SNP assay DNA Next generation sequencing Protein Immunohistochemistry RNA DNA microarray DNA DNA sequencing
Classification Pre-analytical Factor Value(s) Biospecimen Aliquots and Components Biospecimen heterogeneity Intratumoral sampling (exact positions not specified)
Next generation sequencing Specific Data handling Normal sequencing (average 74 X)
Deep sequenced (262x and 255x)
Next generation sequencing Specific Technology platform Sanger sequencing
Preaquisition Other drugs Before everolimus