Introduction
Malignant rhabdoid tumor (MRT) and related INI1-deficient neoplasms are present in the kidney and various extrarenal sites, including the central nervous system (atypical teratoid/rhabdoid tumor). These occurrences predominantly affect infants and young children and are genetically characterized by pathogenic variants of SMARCB1 (>95%), or rarely, of SMARCA4 (<5%).1–7 These genes encode proteins that are components of the chromatin remodeling complex SWI/SNF, a highly conserved transcription regulator that recruits other transcription factors for target genes or alters nucleosome positions to modulate target gene expression.8,9 The multimodal approach to management yielded disappointing outcomes.10,11 The latter experience has motivated the pursuit of studies to better understand this tumor and its microenvironment. Technical advances have provided methods for dissecting the molecular and genomic landscape to gain insight into the events of tumorigenesis, tumor progression, and molecular alterations during treatment.
The molecular heterogeneity of MRT has been documented over the past several years.12 Our hypothesis is that in addition to driver mutations, other genomic alterations in MRT might also contribute to tumorigenesis, tumor progression, and response or lack thereof to management and poor outcome. This study utilized whole exome sequencing to analyze somatic copy number alterations (SCNAs) and somatic variants in MRT patients. The results demonstrated genomic alterations, in addition to driver mutations, providing an additional layer of insight into one of the most aggressive neoplasms of childhood.
Material and methods
The sample selection and experimental studies were performed as previously described.13 All case and case identification numbers were the same as previously described.13
Patient samples
This study was approved by the Washington University Institutional Review Board (IRB # 201102311 “Nextgen Sequencing Approaches” and 201705056 “Expression of Tumor Immunotherapy Related Markers in Pediatric Malignancies”; informed consent was waived). The present study was retrospective and included chart review and residual tissue study following relevant guidelines and regulations. The terms “malignant rhabdoid tumor” and “atypical teratoid/rhabdoid tumor” were searched in the departmental archives of the Lauren V. Ackerman Laboratory of Surgical Pathology, Barnes-Jewish Hospital and St. Louis Children’s Hospital from 1990 to 2017. Hematoxylin and eosin (H&E) and related immunohistochemical staining slides were reviewed, and patients with sufficient tissue (formalin-fixed paraffin-embedded (FFPE) tissue from biopsy/resection specimens) were included.
DNA extraction and whole-exome sequencing
Cases with sufficient amounts of both normal and tumor tissue underwent whole exome sequencing (WES). Genomic DNA was extracted from FFPE tissue from both the normal and neoplastic tissues using standard methods as detailed below.
DNA extraction
Genomic DNA of both normal and neoplastic tissue was extracted from the FFPE tissue at the Washington University Genome Technology Access Center (GTAC), St. Louis, MO, using the AllPrep DNA/RNA FFPE Kit (Qiagen, cat#80234). The extracted DNA was qualified using a Tapestation 4200 (Agilent).
Whole-exome sequencing
Whole-exome sequencing was performed at the Washington University GTAC facility as mentioned earlier. The genomic DNA was sonicated to an average size of 175 bp; the DNA fragments were ligated to the ends of Illumina sequencing adapters. The ligated DNA fragments were amplified for 7 cycles. DNA fragments were then hybridized to biotinylated RNA oligos specific to regions of interest (Clinical Research Exome [Agilent] and selected from the remaining fragments using streptavidin beads. The enriched library was amplified for 14 cycles with primers incorporating a unique indexing sequence tag. The resulting libraries were sequenced using the Illumina HiSeq-3000 platform to obtain 150 bp paired-end reads. Sequencing data revealed 25–30 M reads for normal tissue and 45–50 M reads for neoplastic tissue.
Somatic single nucleotide variant (SNV) determination
The raw sequencing data were processed, including variant score recalibration following the Genome Analysis Toolkit (GATK) v 3.3.0 best practices recommendations. Mutation (variation) analysis and somatic mutation (variation) discovery for SNVs were performed using MuTect v1.1.4. Indel calling and somatic indel identification were performed using the GATK IndelGenotyper tool v2. The SNVs and indels were subsequently annotated using ANNOVAR. The tumor mutational burden was calculated as the total number of missense somatic variants/54 mb, representing the size of the Agilent clinical exome. The detected variants were subsequently evaluated for clinical significance, including pathogenicity, following published guidelines.14
SCNAs
The presence of SCFAs (short-chain fatty acids) was determined by comparing the number of aligned reads per gene obtained by WES in tumors with that in normal tissues (controls) via the cn.mops tool. Gain or loss was defined by abs_CN, where 0 or 1 = deletion and 3 or more = amplification.
Immunohistochemistry (IHC)
Representative sections of tumor tissue from each case (one section per case) were selected. IHC staining with appropriate controls was performed on FFPE tissue samples for the following IHC markers such as PD-1 (2.97 µg/mL, mouse monoclonal, clone NAT105, Ventana, Tucson, AZ, USA), following standard protocols on a Ventana automated stainer (Ventana Medical Systems, Tucson, AZ, USA) in the AMP (Anatomic and Molecular Pathology) Core Lab, Department of Pathology & Immunology, Washington University School of Medicine. IHC staining was evaluated as follows: PD-1 expression was assessed semiquantitatively, and the number of stained cells per high-power field (400×) was calculated based on staining intensity (1+, 2++, and 3+++).
Results
Demographic data of the MRT patients
Five patients with adequate material were subjected to tumor-normal paired whole-exome sequencing (Table 1). All five patients had germline SMARCB1/INI1 gene mutations to corroborate the pathological interpretation.13 Among these patients, two had paired primary (pre-treatment, designated as C) and metastatic (post-treatment, designated as M) tissue samples, two had primary tissue samples (only pre-treatment tissue). Additionally, one had both pre-treatment and post-treatment relapsed tissue samples (designated as R). In total, there were three patients with post-treatment samples.
Table 1Demographic data of the MRT patients
| Case number | Sequencing number | Tumor status | Age at diagnosis | Gender | Location | Germline SMARCB1 variations |
|---|
| Primary, pre-treatment |
| 3 | 40310_C | Primary, before treatment | 9 weeks | F | Kidney | Intronic |
| 6 | 40311_C | Primary, before treatment | 9 months | M | Soft tissue | p.R332W |
| 5 | 40312_C | Primary, before treatment | 6 months | M | Liver | p.R40X, p.P221L |
| 8 | 40307_C | Primary, before treatment | 10 months | M | Kidney | Intronic |
| 13 | 40309_C | Primary, before treatment | 19 months | F | Kidney | pW197X |
| Post-treatment |
| 6 | 40311_R | Post-treatment residual tumor | 11 months | | Soft tissue | |
| 5 | 40312_M | Post-treatment metastatic tumor | 9 months | | Lung | |
| 8 | 40307_M | Post-treatment metastatic tumor | 15 months | | Lung | |
SCNAs
SCNAs were analyzed in both the pre-treatment and post-treatment groups. Twenty-one loci with SCNAs were identified at the whole-exome level in both groups, while the post-treatment group had seven unique foci with SCNAs (Fig. 1a and b, Table 2). Ten foci of SCNAs harbored malignancy-related genes.
Table 2Genes and functions associated with somatic copy number alterations (SCNAs)
| Cytogenetic band | Samples and changes (_C: primary; _M, _R: post-treatment) | Gene and functions | Roles in tumorigenesis |
|---|
| Overall in all samples |
| 2q37.3 | 40309_C (gain), 40312_C (gain),40307_M (gain), 40311_R (gain) | PD-1(programmed death 1) | The programmed death 1 (PD-1) gene encodes for PD-1 (also known as CD279), a negative costimulator in the immune system. PD-1 renders potent inhibitory effects on T and B lymphocytes as well as monocyte responses. Expression of PD-1 ligands by tumor cells has been reported to contribute in immune system evasion. |
| | TWIST2 (twist family BHLH transcription factor 2). This gene is hypermethylated and overexpressed in multiple human cancers, and the encoded protein promotes tumor cell invasion and metastasis, as well as metastatic recurrence. | Overexpression of TWIST1 and TWIST2 proteins are seen in soft tissue sarcomas. |
| 14q32.33 | 40309_C (gain), 40311_R (gain), 40312_C (gain) | MTA1 (metastasis-associated gene 1). Expression of this gene has been correlated with the metastatic potential. | Cancer invasion and metastasis |
| 19p13.2 | 40307_M (gain), 40309_C (loss), 40311_R(gain), 40312_C (loss), 40312_M (gain) | SMARCA4 (SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4) | Bona fide tumor suppressor gene and clearly implicated in cancer development |
| 22q11.23 | 40309_C (loss), 40312_M (loss) | SMARCB1 (SWI/SNF-related matrix-associated actin-dependent regulator of chromatin subfamily B member 1). The protein encoded by this gene is a member of the SWI/SNF family of proteins. Members of this family have helicase and ATPase activities and are thought to regulate transcription of certain genes by altering the chromatin structure around those genes. | Encoding subunit of the SWI/SNF complex and subject to biallelic mutations (germinal and somatic) in rhabdoid tumors |
| Xq25 | 40309_C (loss), 40312_C (loss) | XIAP (X-linked inhibitor of apoptosis). This gene encodes a protein that belongs to a family of apoptotic suppressor proteins. | Seen in variable cancers |
| SCNA in post-treatment samples |
| 11p15.4 | 40307_M (gain), 40311_R (gain), 40312_M (gain) | LMO1 (LIM domain only 1 (rhombotin 1)). LMO1 encodes a protein containing a cysteine-rich LIM domain involved in protein–protein interactions. Recent studies have shown that LMO1 functions as an oncogene in several cancer types. | Neuroblastoma, colorectal cancer, lung cancer, T-ALL |
| | MMP26 (matrix metallopeptidase 26). Proteins of the matrix metalloproteinase (MMP) family are involved in the breakdown of extracellular matrix in normal physiological processes, such as embryonic development, reproduction, and tissue remodeling, as well as in disease processes, such as arthritis and metastasis. | Seen in variable cancers |
| 12q24.31 | 40307_M (gain), 40311_R (gain) | TRIAP1 (TP53 regulated inhibitor of apoptosis 1). Its related pathways are Gene Expression and TP53 Regulates Transcription of Cell Death Genes. | Solid tumors |
| 14q32.32 | 40307_M (gain), 40311_R (gain) | TRAF3 (TNF Receptor Associated Factor 3). This protein is found to be a critical component of the lymphotoxin-beta receptor (LTbetaR) signaling complex, which induces NF-kappaB activation and cell death initiated by LTbeta ligation. | Hodgkin disease, Lymphomas |
| 17q25.1 | 40307_M (gain), 40311_R (gain) | SLC9A3R1 (solute carrier family 9 (sodium/hydrogen exchanger), member 3 regulator 1). The protein also interacts with proteins that function as linkers between integral membrane and cytoskeletal proteins. | There is growing evidence SLC9A3R1 plays an important role in cancer progression. |
| 11q12.2 | 40307_M (gain), 40311_R (gain) | DAGLA (diacylglycerol lipase alpha). Its product is a hydrolytic enzyme that yields 2-AG and free fatty acids. | DAGLA is found in aggrevating malignant phenotypes and cancer progression in hepatocellular carcinoma. |
Several recurrent SCNAs were identified that harbored gene alterations previously detected in malignancies. These alterations included 2q37.3 gain (4/5, 80%, PD-1, TWIST2), 7q32.1 gain (3/5, 60%), 11q12.2 gain (3/5, 60%), 14q32.3 gain (4/5, 80%), 19p13.2 loss harboring SMARCA4 (4/5, 80%), 21q22.3 gain (3/5, 60%), and 22q11.1 loss (2/5, 40%) involving SMARCB1. Alterations more frequently encountered in post-treatment MRTs were 11p15.4 gain (3/3, 100%) and 11q12.2 gain (2/3, 67%).
In the post-treatment group, seven unique foci of SCNAs were detected, including gain of copy number in four genes known to be involved in tumorigenesis: chromosome 11p15.4 harbors LMO1 (LIM domain only 1) and MMP26 (matrix metallopeptidase 26); chromosome 12q24.31 has TRIAP1 (TP53 Regulated Inhibitor of Apoptosis 1); chromosome 14q32.32, TRAF3 (TNF Receptor Associated Factor 3) with amplification; and chromosome 17q25.1 has SLC9A3R1 (solute carrier family 9 (sodium/hydrogen exchanger), member 3 regulator 1).
Somatic variation/mutation landscape
Somatic mutations (variations) are listed in Figure 2. These mutations included nonsense mutations, frame shift insertions, frame shift deletions, frame insertions, frame deletions, splice site deletions, missense mutations, and 5′ flanking regions. A search in ClinVar suggested that most of these variations are more likely benign. Notably, DDX11 (DEAD/H-box helicase 11) R186 W, was found in 2/5 cases (40%) and could possibly be actionable (Table 3).14–17
Table 3Somatic variants found in more than one malignant rhabdoid tumor sample
| Gene name | Samples | Mutations (AA change) | Actionable | GnomAD allele frequency | Functions | Roles in neoplasms |
|---|
| CLCNKA | 40307, 40310 (also found in 40311_R 40307_M?) | A287V | Possibly Benign (1 Submission in ClinVar though) https://www.ncbi.nlm.nih.gov/clinvar/variation/773389/?oq=CLCNKA [gene]+AND+A287V[varname]+&m=NM_004070.4(CLCNKA):c.860C%3ET%20(p.Ala287Val) | Allele Frequency 0.005537 (exomes) https://gnomad.broadinstitute.org/variant/1-16354394-C-T?dataset=gnomad_r2_1 | Chloride voltage-gated channel Ka | RNA expression ratios based on the four-gene panel can accurately classify subtypes of RCC as well as help distinguish some oncocytomas from chromophobe RCC.15 |
| DDX11 | 40310, 40311 | R186 W | Possibly actionable as the mutation is located on a functional Helicase ATP-binding domain. No variant evidence in ClinVar though. | Allele Frequency 0.1208 (Exomes) | DEAD/H-box helicase 11 | DDX11 was significantly upregulated and predicted poor prognosis in lung adenocarcinoma.16 |
| FCGBP | 40307, 40312 | M1617V (40307_T) K3848E (40307_M D3847E (40307_M) R300W (40312_T) | M1617V – Possibly benign as GnomAD exomes allele frequency is high(0.092) and position is not conserved; K3848E – Possibly Benign as GnomAD exomes allele frequency = 0.396 and position is not conserved; D3847E- Possibly benign as not inconserved region. Variant not found in gnomAD exomes; R300W – Uncertain Significance – not conserved but is absent in controls in GnomAD with good coverage in region. Extremely rare? | M1617V - 0.092 (Exomes); K3848E – 0.396 (Exomes); D3847E – Absent in exomes; R300W - Absent in exomes | Fc fragment of IgG binding protein | Differentially expressed in paired tumor-benign tissue samples from patients with stage II CRC.17 |
IHC analysis of PD-1
Gain of chromosome 2q37.3 in both the pre- and post-treatment groups suggested possible amplification of PD-1 and TWIST2 (the twisted family BHLH transcription factor 2, Fig. 3a). To investigate whether PD-1 locus gain correlated with protein expression, IHC was performed on PD-1 (Fig. 3b). The PD-1 immunostaining score was higher in tumors with 2q37.3 gain (Fig. 3c). With an immune score of 100 as the cutoff, a significant difference in the IHC score was observed between those cases with 2q37.3 gain versus no gain (3/3 vs 0/4, P = 0.03; Fisher’s exact test).
Discussion
This study is the first effort to investigate SCNAs and somatic variations in primary, pre-treatment, and post-treatment metastatic or recurrent MRTs by normal-tumor whole-exome sequencing. Several recurrent SCNAs were identified and were shared by more than one patient, some of which harbored genes with alterations known tumorigenic alterations. These alterations included 2q37.3 gain (4/5, 80%; PD-1, TWIST2), 7q32.1 gain (3/5, 60%), 11q12.2 gain (3/5, 60%), 14q32.3 gain (4/5, 80%), 19p13.2 loss harboring SMARCA4 (4/5, 80%), 21q22.3 gain (3/5, 60%), and 22q11.1 loss (2/5, 40%) involving SMARCB1. Alterations more common in post-treatment MRTs included 11p15.4 gain (3/3, 100%) and 11q12.2 gain (2/3, 67%). The recurrent 2q37.3 gain involving the PD-1 gene correlated with PD-1 IHC.
Previous genomic studies of MRT/AT/RT and comparisons with our findings
To date, genetic and genomic studies have investigated somatic variations, SCNAs, gene expression via pathway analysis, microRNAs, and methylation. These studies have revealed the presence of driver mutation(s), epigenetic dysregulations, and dysregulated pathways (Table 4).3,12,18–21 In our study, loss of 22q11.1 was identified in two patients (40309_C and 40310_C), while loss of 22q11.23 was observed in two patients (40309_C and 40312_M) where SMARCB1 was located. These findings are consistent with the findings of earlier studies.3SMARCA4 mutations are present in MRTs. The heterozygous nonsense mutation c.3565C>T (p.Arg1189X) was found in SMARCA4, suggesting that either a severely truncated translation product or nonsense-mediated decay of mRNA were possible consequences.4,18 SMARCA4 is located at 19p13.2 according to Ensembl. In our study, the 19p13.2 deletion was found in two patients (40309_C, 40312_C), and the 19p13.2 gain was observed in three patients (40307_M, 40311_R, 40312_M).
Table 4Literature review related to genomic alterations in malignant rhabdoid tumors
| Chromosome locus | Alterations | Genes | Functions and oncogenesis | Methods | References |
|---|
| 22q11.2 (Cytogenetic band: 22q11.23 by HGNC 22q11.23|22q11 by Entrez Gene 22q11.23 by Ensembl) | Deletion | SMARCB1 | hSNF5/INI1 encodes a member of the SWI/SNF complexes, which are thought to facilitate the transcriptional activation of inducible genes through the remodeling of chromatin | PCR, Southern blot | 3 |
| MRT cell of origin is related to neural crest cells, and that the consequences of SMARCB1 loss in MRT are heterogeneous at the epigenetic and gene expression levels. | SMARCB1 | Same as above. The study revealed gene expression and methylation subgroups and focused on dysregulated pathways, including those involved in neural crest development. | Whole-genome sequencing, whole-genome bisulfite sequencing, whole transcriptome (RNA-seq) and microRNA sequencing (miRNA-seq), and histone modification profiling to characterize extracranial MRTs. | 12 |
| Regulator of NOTCH signaling | ASCL1 | Transcriptional analyses identified two atypical teratoid rhabdoid tumor subgroups with differential enrichment of genetic pathways, and distinct clinicopathological and survival features. Expression of ASCL1, a regulator of NOTCH signaling, correlated with supratentorial location (p = 0.004) and superior 5-year overall survival. | 259 rhabdoid tumors from 37 international institutions and assessed transcriptional profiles in 43 primary tumors and copy number profiles in 38 primary tumors to discover molecular subgroups of atypical teratoid rhabdoid tumors. Gene and pathway enrichment analyses were used to discover group-specific molecular markers and did immunohistochemical analyses on 125 primary tumors to evaluate clinicopathological significance of molecular subgroup and ASCL1-NOTCH signaling. | 20 |
| 21 MRT specimens. In eight samples, uniparental disomy (UPD) of 22q segments caused homozygous mutations/deletions of SMARCB1. Ten samples had homozygous focal deletions commonly involving a 175-kb region (ch22:22,353,181-22,528,353), which exclusively included SMARCB1. | SMARCB1 CCNL1 (3q25.31), POT1 (7q31.33), CNTNAP2, and PRPTD | Same as above | SNP-chip analysis | 19 |
| mutations | SMARCB1 | Alterations affecting the SMARCB1 locus could be demonstrated by MIP SNP in 15 out of 16 evaluable cases (94%). Remarkably, MIB SNP analysis did not yield any further recurrent chromosomal gains, losses, or copy neutral LOH. On MIP SNP screening for somatic mutations, the presence of a SMARCB1 mutation (c.472C>T p.R158X) was confirmed, but no recurrent mutations of other cancer relevant genes could be identified. | molecular inversion probe single-nucleotide polymorphism (MIP SNP) assay (Affymetrix OncoScan formalin-fixed paraffin-embedded express) | 18 |
| N/A | N/A | N/A | N/A | Cytogenetic and molecular studies of an AT/RT on a 15-month-old boy. The tumor showed a complex karyotype with one cell line showing monosomy 22 and another near-tetraploid one with additional chromosomal abnormalities, involving chromosomes 2, 3, 5, 6, and Y. | 21 |
SCNAs in MRT
In addition to the findings of prior studies, the present study revealed several recurrent SCNAs that harbor genes with alterations involved in malignancy. These alterations included 2q37.3 gain (4/5, 80%; PD-1, TWIST2), 7q32.1 gain (3/5, 60%), 11q12.2 gain (3/5, 60%), 14q32.3 gain (4/5, 80%), 19p13.2 loss harboring SMARCA4 (4/5, 80%), 21q22.3 gain (3/5, 60%), and 22q11.1 loss (2/5, 40%) involving SMARCB1.
Alterations more frequently observed in posttreatment MRTs included 11p15.4 gain (3/3, 100%) and 11q12.2 gain (2/3, 67%). 11p15.4 was the only SCNA in two metastatic patients (40307_M, 40312_M), suggesting that a gain of 11p15.4 might indicate the possibility of metastasis.
Potential pathogenic roles of gene amplification in MRT
In our study, SCNAs were also found in regions where several other important tumorigenesis-related genes are located (Table 2). A gain in chromosome 2q37.3 was found in four patients (40307_M, 40309_C, 40311_R, and 40312_C), in which the PD-1 gene and TWIST2 (twist family BHLH transcription factor 2) gene were located. In the CD8+ cytotoxic antitumor response, receptor-ligand interactions between molecules such as programmed cell death 1/programmed cell death ligand 1 (PD-1/PD-L1) suppress the CD8+ cytotoxic response. The PD-1 gene encodes PD-1 (also known as CD279), which is a negative stimulator of the immune system with potent inhibitory effects on T and B lymphocytes as well as the monocyte response.22–25 The expression levels of these genes during persistent antigen exposure are observed in chronic infections and cancer. IHC staining for PD-1 was performed, revealing a significantly higher PD-1 immunostaining score in tumors from patients with 2q37.3 gain.
Overexpression of the TWIST1 and TWIST2 proteins has been reported in other soft tissue sarcomas, with TWIST2 amplification observed in rhabdomyosarcoma, contributing to the repression of myogenesis and promotion of oncogenesis.26 Chromosome rearrangement at 14q32.33 is noted in multiple myeloma with variable partner sites, including 11q13.3, 8q24.1, 18q21.3, and 6p21.1.27 In our study, chromosome 14q32.33 was found in three patients (40309_C, 40311_R, 40312_C), where MTA1 (metastasis-associated gene 1) is located. MTA1 is related to invasion and metastasis.28–32 Chromosome Xq25 is lost in two patients (40309_C and 40312_C), in which the XIAP gene (X-linked inhibitor of apoptosis) resides. The XIAPs regulate cell death signaling pathways through binding and inhibiting caspases and also participate in cancer initiation, promotion, and progression.33–36
SCNA in post-treatment MRT
In our study, several SCNAs were identified in posttreatment tumors with chromosomal gains. Chromosome 11p15.4 was found in three patients (40307_M, 40311_R, 40312_M), in which LMO1 (LIM domain only 1 or rhombotin 1) and MMP26 (matrix metallopeptidase 26) were located. LMO1 functions as a neuroblastoma oncogene and is implicated as an oncogene in colorectal and lung cancer.37,38 Proteins of the matrix metalloproteinase (MMP) family are involved in the breakdown of the extracellular matrix in normal physiological processes, such as embryonic development, reproduction, and tissue remodeling. Additionally, they are involved in disease processes such as arthritis and metastasis and have been reported as biomarkers of various cancers.39–41 Chromosome 12q24.31 is gained in two patients (40307_M and 40311_R), hosting the TP53-regulated inhibitor of apoptosis 1 (TRIAP1) gene. TRIAP1 is a novel apoptosis inhibitor that binds HSP70 in the cytoplasm and inhibits apoptosome and caspase-9 activation. TRIAP1 has been shown to be upregulated in various cancer types.42,43 Chromosome 14q32.32 was found in two patients (40307_M, 40311_R), in which TNF receptor-associated factor 3 (TRAF3) was located. TRAF3 is expressed in Hodgkin disease and lymphomas.44,45 Chromosome 17q25.1 was found in two patients (40307_M, 40311_R), in which solute carrier family 9 (sodium/hydrogen exchanger) was located. There is growing evidence that SLC9A3R1 plays an important role in cancer progression.46,47
Overall, our study revealed that in addition to driver mutations, there are recurrent SCNAs in different cases of MRT. These alterations are associated with malignancy-related genes, including SMARCB1, SMARCA4, PD-1, TWIST2, TRIAP1, MTA1, and XIAP. Notably, specific alterations, including those in LMO1, MMP26, TRIAP1, TRAF3, and SLC9A3R1, are specific for post-treatment MRTs. The identification of 11p15.4 as the only recurrent SCNA in two metastatic patients (40307_M, 40312_M) suggests its potential as an indicator of metastasis. There are various treatments available for patients with MRT, and our study may contribute to the development of treatments based on SCNAs.48 Importantly, our study is the first to report specific SCNAs in post-treatment MRTs.
Somatic mutation landscape
No obvious pathogenic variants were found in the current study, with the exception of DDX11 (DEAD/H-box helicase 11) R186 W, which was found in two of five patients. DDX11 was found to be substantially upregulated in lung adenocarcinoma and to predict poor prognosis.49 This mutation may be actionable because it is located within a functional helicase ATP-binding domain, while there is no variant evidence in ClinVar.
Limitations
Due to the limited tissue available for clinical trials, a major limitation is the small sample size. Two of the three post-treatment samples were derived from metastatic sites; there is a possibility that these differences were due to changes in tumor metastasis rather than to clonal evolution specifically from external therapeutic pressures. Moreover, the possibility that posttreatment alterations may be artifactual due to poor sample quality or necrosis cannot be ruled out.