To investigate the genomic evolution of metastatic pediatric osteosarcoma, we performed whole-genome and targeted deep sequencing on 14 osteosarcoma metastases and two primary tumors from four patients (two to eight samples per patient). All four patients harbored ancestral (truncal) somatic variants resulting in TP53 inactivation and cell-cycle aberrations, followed by divergence into relapse-specific lineages exhibiting a cisplatin-induced mutation signature. In three of the four patients, the cisplatin signature accounted for >40% of mutations detected in the metastatic samples. Mutations potentially acquired during cisplatin treatment included NF1 missense mutations of uncertain significance in two patients and a KIT G565R activating mutation in one patient. Three of four patients demonstrated widespread ploidy differences between samples from the sample patient. Single-cell seeding of metastasis was detected in most metastatic samples. Cross-seeding between metastatic sites was observed in one patient, whereas in another patient a minor clone from the primary tumor seeded both metastases analyzed. These results reveal extensive clonal heterogeneity in metastatic osteosarcoma, much of which is likely cisplatin-induced.

Implications:

The extent and consequences of chemotherapy-induced damage in pediatric cancers is unknown. We found that cisplatin treatment can potentially double the mutational burden in osteosarcoma, which has implications for optimizing therapy for recurrent, chemotherapy-resistant disease.

This article is featured in Highlights of This Issue, p. 837

Increased understanding of intrapatient tumor heterogeneity has fueled progress in many cancers (1). For example, analysis of heterogeneity can reveal subclonal drug resistance mechanisms (2) and early mutation events that can be preferentially targeted (3). However, the clonal heterogeneity of metastatic pediatric solid tumors such as osteosarcoma is not well understood.

Osteosarcoma is a cancer arising from bone (4) and occurs in children and adolescents during active bone growth (5). Current therapy includes chemotherapy with the MAP (methotrexate, doxorubicin, and cisplatin) regimen and surgery (4). Five-year survival from time of diagnosis is 60% to 70%, with little improvement in the last three decades (5–7). Osteosarcoma most frequently metastasizes to the lungs, which accounts for most deaths (4, 5).

To better understand tumor heterogeneity and clonal evolution in osteosarcoma, we analyzed 14 metastatic samples and two primary tumors from four patients as part of the St. Jude/Washington University Pediatric Cancer Genome Project (PCGP), which revealed substantial intrapatient heterogeneity associated with cisplatin treatment. These results demonstrate the impact of chemotherapy on shaping the clonal architecture of metastatic osteosarcoma.

Sample information

Samples were used under institutional review board approval, in accordance with the Declaration of Helsinki, at St. Jude Children's Research Hospital and Washington University in St. Louis, and written informed consent and/or assent from patients and/or guardians was obtained. SJOS001101_M1 was obtained during thoracotomy of the left lung ∼48 weeks postdiagnosis and was a lung slice with two discrete metastatic nodules. SJOS001101 samples M2-M8 were obtained 15 hours postmortem (Supplementary Table S1). Five of the 16 samples were included in a previous study focused on identifying significantly mutated genes in osteosarcoma: SJOS001107_M1, SJOS001107_M2, SJOS001105_D1, SJOS010_D, and SJOS010_M (8).

Whole-genome sequencing and capture validation

Whole-genome sequencing (WGS) was performed as described (8). Capture validation was performed using custom Nimbelgen Seqcap EZ solution bait sets (Roche) and sequencing was performed as described (8). In SJOS001101, 91% of WGS somatic SNVs were analyzed by capture validation; in SJOS010, 84% were analyzed. Thus, in Fig. 1 heatmaps capture validation data were used as it approached whole-genome level. In SJOS001105 and SJOS001107, a lesser percentage of SNVs were analyzed by capture validation (31% and 78%, respectively). Thus, Fig. 1 heatmaps for these patients rely on WGS but later clonal evolution analysis relies on capture validation. Whole-genome coverage was measured for WGS; for capture validation, the number of counts (mutant or wild-type) at the target site was used to quantify coverage. Indels, CNVs, and structural variants (SV) were computed from WGS.

Figure 1.

Osteosarcoma treatment history and evolution. Treatment history and mutational heterogeneity in patients with osteosarcoma SJOS001101 (A), SJOS001105 (B), SJOS001107 (C), and SJOS010 (D). Left side shows platinum treatment time periods and locations and times (in weeks postdiagnosis) of sample acquisition. Red samples indicate samples on which WGS was performed, whereas gray indicates nonsequenced samples. Right shows heatmap of somatic SNVs, with color indicating VAF adjusted for tumor purity. Possible driver SNVs along with indels, SVs, and CNVs are indicated next to SNV cluster; clusters are colored to the right, and match tree branch colors at far right. Evolutionary tree branches are proportional to the number of mutations, with truncal variants at top and private variants at bottom, and driver variants indicated. In SJOS001101 (A), at week 70 autopsy the lung was filled with numerous contiguous bulky metastases which is indicated by diffuse gray color, and sample M1 at week 48 was two adjacent metastatic lesions (dotted lines). Most lesions sampled at autopsy were in the lung except M2 (right lateral chest wall), M5 (superior mediastinal mass), and M8 (diaphragmatic and left lower lobe of lung). SJOS001101 (A) and SJOS010 (D) heatmaps are based on capture sequencing, whereas SJOS001105 (B) and SJOS001107 (C) are based on WGS.

Figure 1.

Osteosarcoma treatment history and evolution. Treatment history and mutational heterogeneity in patients with osteosarcoma SJOS001101 (A), SJOS001105 (B), SJOS001107 (C), and SJOS010 (D). Left side shows platinum treatment time periods and locations and times (in weeks postdiagnosis) of sample acquisition. Red samples indicate samples on which WGS was performed, whereas gray indicates nonsequenced samples. Right shows heatmap of somatic SNVs, with color indicating VAF adjusted for tumor purity. Possible driver SNVs along with indels, SVs, and CNVs are indicated next to SNV cluster; clusters are colored to the right, and match tree branch colors at far right. Evolutionary tree branches are proportional to the number of mutations, with truncal variants at top and private variants at bottom, and driver variants indicated. In SJOS001101 (A), at week 70 autopsy the lung was filled with numerous contiguous bulky metastases which is indicated by diffuse gray color, and sample M1 at week 48 was two adjacent metastatic lesions (dotted lines). Most lesions sampled at autopsy were in the lung except M2 (right lateral chest wall), M5 (superior mediastinal mass), and M8 (diaphragmatic and left lower lobe of lung). SJOS001101 (A) and SJOS010 (D) heatmaps are based on capture sequencing, whereas SJOS001105 (B) and SJOS001107 (C) are based on WGS.

Close modal

SNV and indel variant identification and clustering

WGS reads were aligned to GRCh37-lite with BWA (9) and somatic SNVs/indels were called with Bambino (10) followed by a postprocess which removes paralogous variants and sequencing artifacts caused by poor quality or alignment artifacts (11). Only exonic indels were analyzed. Capture validation data were aligned to GRCh37-lite and mutant and wild-type counts for each SNV were determined using an in-house pipeline which determines mutant and total counts of a pre-identified mutation list, while taking into account read quality (unpublished). Validation rate of WGS SNVs was determined by comparing WGS VAFs with capture validation VAFs for each variant with >15 coverage in both platforms and a positive WGS call in the sample. Variants were validated if Fisher exact test comparing capture validation mutant reads versus nonmutant reads in germline versus tumor yielded P < 0.05 in at least one sample from the patient. We also excluded variants with germline VAF ≥ 0.01, low germline coverage (≤20), or low tumor coverage (≤15). Of the 21,963 high-quality SNVs selected for validation, 21,779 (99.2%) were validated. For SJOS001105 and SJOS001107, heatmaps in Fig. 1 were generated from WGS variants after post-processing while remaining analysis used capture validation sequencing. Kataegis was analyzed as described (8).

VAFs were adjusted for tumor purity as follows. Let p represent tumor purity as a proportion between 0 and 1, and c represent the integer copy number of the region containing the variant (determined by rounding copy number to nearest integer). The proportion of total reads contributed by tumor cells at the mutation site (t) is:

formula

2(1 − p) represents normal contribution and c(p) represents tumor contribution. The adjusted VAF is:

formula

SNV clusters were identified by determining which samples the SNV was present in at adjusted VAF ≥0.05 (12). SNV clusters with <250 SNVs, except private clusters, were excluded from branching evolution analysis, to ignore SNV clusters caused by SNV dropout from private copy losses, but were included in Supplementary Figs. S9 and S11 (“Undefined” SNV cluster).

Copy number variant identification

CNVs were identified using CONSERTING (13) from WGS. Segmented CNV data were adjusted for normal contamination as described (12), which also determined tumor purity. Tumor purity was corroborated by B-allele frequencies. Regions with 1.8 to 2.2 copies were considered diploid and copy-altered otherwise. Cancer Gene Census (14) genes were downloaded from https://cancer.sanger.ac.uk/census. To perform CNV Euclidean distance clustering, we sampled segmented copy data at every 10,000th genomic position and used dist and hclust functions in R. Absolute copies reported for individual genes in manuscript text are rounded to nearest integer.

Mutational signature analysis

Trinucleotide context for SNVs was determined with an in-house script (12). NMF was used to extract mutational signatures from PCGP osteosarcomas (WGS; ref. 8), including the four patients on which this study focuses, using SigProfiler (15). Three signatures was optimal. Signature definitions for 30 COSMIC signatures and our extracted cisplatin signature were used as input to SigProfilerSingleSample to query signature strengths in samples (or SNV clusters), with the following parameters: signatures 1, 3, and 5 were included as signatures “included in all samples regardless of rules or sparsity,” with signature 3 added to the default signatures 1 and 5 due to strong presence in osteosarcoma (16); signatures 2 and 13 (APOBEC-associated) considered “connected”; and an improvement of accuracy of 0.02 (rather than default 0.05) in add_all_single_signatures. For evolutionary tree construction, signature weights in SNV clusters were superimposed on evolutionary trees (17).

Probabilities of being cisplatin-induced for NF1 and KIT variants

To calculate probabilities that NF1 and KIT SNVs were cisplatin-induced, we used an approach described previously (18). SNV clusters containing NF1 and KIT variants were >0.95 explained (cosine similarity) by the signatures queried.

Evolutionary tree construction

Evolutionary trees were constructed using principles described previously (3, 19). Each branch represents an SNV cluster. Branch lengths are proportional to number of SNVs. Trees were organized such that each sample (tree bottom) possesses the SNVs on the clusters attached to it above the sample. For SJOS001101 M4, no private SNVs were identified due to low tumor purity (<20%). For SJOS001101 M1, multiple lineages were present, and the most clearly defined lineage (the M1-M5-M6 cluster) was represented.

Subclone analysis

Subclone analysis was performed using principles described previously (12, 20). SNVs that were diploid (copy number 1.8–2.2) in a single sample (Supplementary Fig. S8) or across all samples when possible (Supplementary Figs. S9, S10, and S11C) were used for clonal analysis. SJOS001107 had <30 such variants and was not included in density analysis, and SNVs in regions with one to four copies were analyzed by two-dimensional and three-sample VAF plots (Supplementary Fig. S11B) whereas in SJOS001105 variants diploid in D1 and D2 and in three-copy regions in R1 were analyzed (Supplementary Fig. S11A), due to widespread copy gains in R1 leading to fewer than 10 pan-diploid SNVs in this patient. Overlapping VAF density peaks indicated co-occurring SNV clusters (Supplementary Fig. S8). Private SNV clusters with two peaks, including a minor peak with VAF <0.4, revealed descendant clone(s). Pairwise sample VAF comparisons clarified minor admixture of clones between sites. For cross-seeding analysis between SJOS001101 samples M5 and M6, the median M5 VAF (of M6-private SNVs detected in M5 at VAF >0, all of which were <0.05 due to our clustering method) was calculated and multiplied by two to determine CCF of the M6 site-specific founder clone in M5. Descendant clone(s) revealed by secondary private VAF density peaks (Supplementary Fig. S8) were shown as a single descendant clone in Fig. 4A by a subcircle, although it is possible that descendants-of-descendants or multiple independent descendants were present. SJOS001107 metastasizing clone CCF was estimated <5% but could not be determined with certainty due to lack of diploid variants.

PyClone (21) and MACHINA (22) were used in SJOS001101. We clustered SNVs diploid in all samples thus: (1) we classified variants based on presence or absence in each sample, and (2) within these groups, clustered on VAF using PyClone. Mutations were “present” if the posterior probability of the variant's presence was ≥0.95. This procedure yielded 35 clusters; we analyzed the 17 clusters with ≥4 mutations.

We then performed clonal analysis using MACHINA in tree inference mode. We calculated 95% confidence bounds on cluster frequency (22), assuming heterozygosity except (1) clusters 1 and 2, which were likely homozygous in M1 and M4-M8 following LOH; (2) cluster 13, which was interpreted as mutations lost in the LOH event.

We used a Bayesian model to determine the posterior probability of a variant's presence. Let X = 1 indicate presence of a variant in and X = 0 indicate absence. The posterior probability of a variant's presence is:

formula

where V is the number of variant reads, and T is the total reads covering the locus. We modeled sequencing as a binomial process with probability of success f = 0 if the mutation is absent and f ∼ Beta(α = 1, β = 1) if present. We used a uniform prior probability Pr(X = x) = 0.5. Thus, we have

formula
formula

For a mutation j, we assigned sample profile $\bar {\rm x}$j = [$\bar {\rm x}$1,j, …, $\bar {\rm x}$m,j] such that for sample i, $\bar {\rm x}$i,j = 1 provided that Pr(Xi,j = x | Vi,j, Ti,j) > 0.95. We ran PyClone using beta-binomial density and default parameters on each set of mutations Vx, such that all mutations iVx have the same sample profile $\bar {\rm x}$i, = x. This yielded a set of cluster assignments Cx for each sample profile x. To obtain input for MACHINA, we took the union of sample profiles C = ∪xCx.

To test whether SJOS001107 cluster M1-M2 variants, which had low VAFs in D1, represented true signal, we used CleanDeepSeq (Ma and colleagues, under review). CleanDeepSeq removes reads which have problematic alignment, discordant bases in regions overlapped by forward and reverse read pairs, or a high percentage (≥5%) of bases with low quality. Overlapping portions of the surviving read pairs are collapsed to prevent read count duplication. M1-M2 SNVs with ≥3 mutant reads in D1 were considered to “bleed” into D1. This was compared with M2 private SNVs (the other nontruncal SNV cluster >100 mutations) with ≥3 mutant reads in D1 or M1, by Fisher exact test.

Data and code availability

WGS is available through St. Jude Cloud (https://stjude.cloud) and EGA (EGAS00001000263). Supplementary Tables S2 to S5 report capture validation. Custom code is available upon request.

Osteosarcoma patient history and samples

We performed WGS on osteosarcoma samples from four patients for which multiple tumor samples were available (Fig. 1; Supplementary Fig. S1). The first three patients, SJOS001101, SJOS001105, and SJOS001107 (Fig. 1A–C) received standard MAP chemotherapy (methotrexate, doxorubicin, and cisplatin) combined with bevacizumab (23) followed by resection of the primary tumor and postoperative MAP. The fourth, SJOS010 (Fig. 1D) was treated with a carboplatin-containing regimen (24) and cisplatin years later.

In SJOS001101, one lung metastasis from 48 weeks postdiagnosis and seven autopsy samples from the lung and adjacent tissues were analyzed, but the primary tumor was unavailable (Fig. 1A). In SJOS001105, the primary tumor (left proximal humerus) along with a femur and a lung metastasis >1 year later were sequenced (Fig. 1B). For SJOS001107, we analyzed the primary tumor (left proximal humerus) along with a lung metastasis also detected at diagnosis, and a recurrent lung metastasis from >1 year later (Fig. 1C). For SJOS010, we sequenced bone and lung metastases which appeared 8 to 9 years after diagnosis (Fig. 1D). Together, this dataset enabled analysis of temporal evolution through drug treatment and spatial evolution associated with metastatic spread (Supplementary Table S1).

Branching evolution of osteosarcoma metastases

To evaluate tumor heterogeneity between sites, we performed WGS on the 16 osteosarcoma samples at 42-62X median genome coverage and 35-44X for germline (Supplementary Fig. S2) to identify single-nucleotide variants (SNV) and indels (Materials and Methods), SVs (25), and copy number variants (CNV; ref. 13). We initially focused on SNVs for analysis of clonal evolution. More than 30,000 somatic SNVs were identified in nonrepetitive genomic regions. To validate these variants and enable robust clonal analysis, we performed deep capture validation sequencing (8) on all samples (Materials and Methods; Supplementary Tables S2–S5) with median coverage of 149-1145X per sample (Supplementary Fig. S3) with an overall validation rate of 99.2% (Materials and Methods).

To determine relationships between metastatic sites, we clustered SNVs based on their presence [defined by variant allele frequency (VAF) ≥0.05 after adjusting for tumor purity] or absence in each sample, excluding clusters with relatively few SNVs (heatmaps in Fig. 1; Materials and Methods). In each patient, variants present in all samples, referred to as truncal variants, accounted for <20% of all somatic SNVs (Fig. 1). Sample-specific private variants were present in all except SJOS001101_M4, which was estimated to have <20% tumor purity (Fig. 1A; Supplementary Table S2). Several samples had relatively high numbers of private SNVs, including M2, M3, and M8 in SJOS001101 (Fig. 1A), D2 and R1 in SJOS001105 (Fig. 1B), M2 in SJOS001107 (Fig. 1C), and both samples in SJOS010 (Fig. 1D). Of the multiregion autopsy samples analyzed for SJOS001101, M2 and M3 clustered separately from other samples, consistent with the anatomical proximity of these two metastases on the right side, whereas most other lesions were on the left (Fig. 1A).

Truncal driver variants included SVs or SNVs causing TP53 inactivation in each patient (Fig. 1), consistent with previous osteosarcoma studies (8, 26). Each patient also harbored at least one cell-cycle–related truncal variant (Supplementary Fig. S4 and S5), including RB1 W78* in SJOS001101 (Fig. 1A), CCND3 and CDK4 amplification in SJOS001105 (Fig. 1B), CCND3 amplification and CDKN2A homozygous deletion in SJOS001107 (Fig. 1C), and CCNE1 amplification in SJOS010 (Fig. 1D). Finally, a 0.5 to 1.0 Mb region on 3q13.31 containing the noncoding genes TUSC7, MIR4447, and LINC00901 underwent deletions in all four patients (27); the minimal overlapping region of homozygous deletion in one sample (SJOS010) included only LINC00901 (Supplementary Fig. S4A), although in some osteosarcomas this gene is not in the minimally deleted region (8). 3q13.31 deletions were truncal in three patients, SJOS001105, SJOS001107, and SJOS010 (Fig. 1B–D) and “shared” (present in some but not all samples) in a fourth, SJOS001101 (Fig. 1A).

Nontruncal variants were acquired later, and they included a shared heterozygous NF1 G722E mutation in SJOS001101 samples M4, M7, and M8, (Fig. 1A); a private NF1 S2684I variant at a splice acceptor site, which may affect splicing, in SJOS001105 sample R1 (Fig. 1B); and a private KIT G565R mutation in SJOS001101 sample M3 (Fig. 1A). Although NF1 variants have been reported in osteosarcoma (26), these two NF1 mutations are variants of uncertain significance due to lack of functional validation data. The KIT G565R mutation was likely activating as it has been reported to confer sensitivity to imatinib in a gastrointestinal stromal tumor (28) and has been found in mucosal melanomas (29) and a gastric adenocarcinoma (30). SJOS010 harbored a private DLG2 deletion, likely an important disease driver in osteosarcoma (8, 31), in sample D (Fig. 1D). Osteosarcoma evolutionary trees displayed short trunks and long branches in each case (Fig. 1, far right), in contrast to some other cancer types such as breast and lung cancer, where in most patients 50% or more of mutations were truncal (3, 12, 17). However, the known driver events in osteosarcoma were usually truncal, as seen in some other cancer types (32), notwithstanding the high level of intrapatient heterogeneity.

Mutational signature analysis reveals cisplatin-induced mutagenesis

We analyzed trinucleotide mutation context signatures to determine potential causes of mutagenesis at different stages of tumor evolution (33). Rather than comparing mutational signatures between samples directly, we analyzed each SNV cluster, representing a distinct branch of evolution (Fig. 1) to determine the mutational processes giving rise to truncal, shared, and private variants (Fig. 2A). This revealed a unique mutational signature in many shared and private SNV clusters but absent in truncal SNVs, implying a late mutational process. This signature was characterized by C[C>T]C, C[C>T]T, C[T>A]N and secondarily C[C>A]T variants (Fig. 2A, arrows), and did not match any COSMIC mutational signature (33, 34). To determine the etiology of this signature, we checked for the signature in WGS data of 29 additional PCGP osteosarcomas (Supplementary Fig. S6A, which includes the four multisample patients for comparison; ref. 8). The signature was detected exclusively in relapsed but not pretreatment samples, suggesting it may be therapy-induced (Supplementary Fig. S6A, arrows). Indeed, of the two patients (SJOS001105 and SJOS001107) with matched pretreatment samples and relapsed samples, the signature could be detected in relapsed samples but not matched pretreatment samples (Fig. 2A; Supplementary Fig. S6A).

Figure 2.

Osteosarcoma branching evolution shaped by cisplatin treatment. A, Mutation trinucleotide context of SNV clusters shown in Fig. 1. Red color indicates the proportion of SNVs falling into the indicated trinucleotide context. Asterisks indicate mutation contexts that the novel signature has a predilection to mutate. “Pre-tx” indicates pretreatment samples; “rel” indicates relapsed samples. B, Cisplatin signature weight of novel cisplatin signature in PCGP osteosarcomas based on SNVs identified from capture validation for SJOS001101 and SJOS010 and WGS for other patients. PCGP sample nomenclature is of the format “SJOS002_D,” where SJOS002 indicates patient ID and D indicates sample ID. C, Branching evolution of osteosarcomas with mutational signature data from Supplementary Fig. S6E, right, superimposed. SJOS001101 M4 private variants were not detected due to low tumor purity and M1 private SNVs were not included due to unresolvable clonal composition. Potentially pathogenic variants are indicated on the appropriate evolutionary branches. Branch length is proportional to number of SNVs in branch as indicated by scale (bottom-right).

Figure 2.

Osteosarcoma branching evolution shaped by cisplatin treatment. A, Mutation trinucleotide context of SNV clusters shown in Fig. 1. Red color indicates the proportion of SNVs falling into the indicated trinucleotide context. Asterisks indicate mutation contexts that the novel signature has a predilection to mutate. “Pre-tx” indicates pretreatment samples; “rel” indicates relapsed samples. B, Cisplatin signature weight of novel cisplatin signature in PCGP osteosarcomas based on SNVs identified from capture validation for SJOS001101 and SJOS010 and WGS for other patients. PCGP sample nomenclature is of the format “SJOS002_D,” where SJOS002 indicates patient ID and D indicates sample ID. C, Branching evolution of osteosarcomas with mutational signature data from Supplementary Fig. S6E, right, superimposed. SJOS001101 M4 private variants were not detected due to low tumor purity and M1 private SNVs were not included due to unresolvable clonal composition. Potentially pathogenic variants are indicated on the appropriate evolutionary branches. Branch length is proportional to number of SNVs in branch as indicated by scale (bottom-right).

Close modal

To quantitatively determine the spectrum of trinucleotide context SNVs in this signature, we performed nonnegative matrix factorization (NMF; refs. 15, 33) on SNVs from the PCGP osteosarcoma cohort, including the four patients on which this study focuses (Supplementary Fig. S6B and S6C). This revealed three mutational signatures, two of which resembled known APOBEC, homologous recombination deficiency, and clock-like signatures (33), whereas the third represented the novel signature described above (Supplementary Fig. S6C). The novel signature showed high cosine similarity (0.940; ref. 15) to a cisplatin signature from Boot and colleagues generated from extended exposure of the MCF 10A breast cell line to cisplatin (35) and also resembled a recently identified cisplatin signature in bladder cancer (cosine similarity 0.801; ref. 36), thus indicating that it is a cisplatin mutational signature (Supplementary Fig. S6C). Further, our novel cisplatin signature was similar to a novel mutational signature of unknown origin in a recent pan-pediatric cancer study that included osteosarcoma (cosine similarity 0.958; Supplementary Fig. S6C; ref. 37). The cisplatin signature showed no evidence of localized hypermutation (kataegis; Supplementary Fig. S7), indicating that cisplatin does not affect specific regions but may mutate globally.

Further, we measured the strength of our cisplatin signature in each PCGP osteosarcoma, including the four multisample patients, alongside reported COSMIC mutational signatures (33). The cisplatin signature was detected in the four patients under study, although not in two additional relapsed cases; all six of these patients had received DNA-damaging (38) platinum treatment, usually in the form of cisplatin (Supplementary Fig. S6D, Fig. 2B; Supplementary Table S1). None of the 30 pretreatment samples possessed the cisplatin signature, confirming its specificity to treated samples (Fig. 2B). In three of six posttherapy osteosarcoma patients (SJOS001101, SJOS001105, and SJO001107), the cisplatin signature accounted for >40% of SNVs (Fig. 2B), and in SJOS001101 >60% in most samples. Among the other three patients, the signature was absent in two patients (SJOS001 and SJOS001112) and detected in one of two lesions from SJOS010 (19% of SNVs in D, and not detected in M). This indicates that platinum therapy's mutational effects are variable. Although these three signature-low patients had all received platinum treatment, SJOS001112 had received carboplatin rather than cisplatin, and SJOS010 had received carboplatin primarily and cisplatin as a secondary treatment (Supplementary Table S1). Whether the use of carboplatin rather than cisplatin accounts for the lower signature strength cannot be determined with this sample size.

Finally, where matched pre- and posttherapy samples were available (SJOS001105 and SJOS001107), the cisplatin signature was exclusive to posttherapy samples, reaffirming its specificity (Fig. 2B). These data indicate that cisplatin causes significant damage to osteosarcoma genomes, in some cases potentially doubling the number of somatic SNVs.

Mutational signature tree reveals cisplatin-associated evolution

To determine the evolutionary changes potentially induced by cisplatin, we next determined the robustness of the cisplatin signature and COSMIC mutational signatures (39) in each patient's evolutionary SNV clusters, such as truncal, shared or private clusters (Supplementary Fig. S6E). This analysis confirmed the cisplatin signature's strong presence in variants of later (private and shared) but not truncal clusters (Supplementary Fig. S6E). We superimposed these mutational signature data onto the previously-constructed evolutionary trees to create mutational signature evolutionary trees (Fig. 2C; ref. 17). Truncal variants were enriched for COSMIC signature 1 (a ubiquitous clock-like signature caused by 5-methylcytosine deamination; ref. 39), signature 3 (indicating homologous recombination deficiency; refs. 16, 40), and signature 5 (another clock-like signature; ref. 39). Shared variants, in contrast, were enriched for the cisplatin signature, suggesting that cisplatin may have induced branching events (Fig. 2C, left, SJOS001101). Private SNVs in some relapse samples (SJOS001101 samples M2, M3, M8; SJOS001105 samples D2 and R1; SJOS001107 sample M2) were also enriched for the cisplatin signature (Fig. 2C). However, mutations on private branches of several relapsed samples were relatively free of the cisplatin signature (SJOS001101 samples M5, M6, M7), suggesting that these mutations were acquired after, or near cessation of, cisplatin treatment (Fig. 2C, left). Regional heterogeneity of the cisplatin signature was found in SJOS010 as the cisplatin signature was present in only one of two posttherapy (post-carboplatin and post-cisplatin) samples—tibia metastasis D had 19% of SNVs platinum-induced and the signature was absent in lung metastasis M (Fig. 2C, right).

Variants potentially induced by cisplatin

The SJOS001101 KIT and NF1 variants, and SJOS001105 NF1 variant, appeared in branches enriched for the cisplatin signature (Fig. 2C), suggesting they may have been cisplatin-induced. To calculate the probability that these mutations were cisplatin-induced, we used an approach described previously (18). This revealed a 100% likelihood that the NF1 G722E variant was cisplatin-induced, consistent with its presence at cisplatin hotspot site C[C>T]C. For KIT G565R, the probability of being cisplatin-induced was 60%. This variant was found at C[C>T]A, which is not a cisplatin hotspot but still a potential target (Supplementary Fig. S6C). SJOS001105 had an NF1 S2684I variant (potentially affecting splicing) with a 70% likelihood of being cisplatin-induced (at A[C>A]C, which is not a strong cisplatin hotspot). No pathogenic SNVs were detected in cisplatin-enriched branches in the two additional patients with the cisplatin signature (SJOS001107 and SJOS010; Fig. 2C). Given the uncertain significance of the two NF1 mutations, and the 60% probability that the KIT variant was cisplatin-induced, further investigation is needed to determine whether cisplatin induces functional, cancer-promoting variants.

Clonal heterogeneity of copy number variation

To understand clonal heterogeneity from the perspective of copy number variation (CNV), we compared intrapatient CNV profiles (Fig. 3). We first clustered CNV profiles in SJOS001101 (Fig. 3A), the patient with the most samples, which mirrored SNV-based clustering as M2 and M3, both collected from the right side (Fig. 1A), clustered separately from other samples. Although M2 and M3 had relatively diploid copy profiles, other samples harbored widespread copy gains (Fig. 3A, bottom). Specifically, although M2 and M3 had copy gains in 38% to 44% of the genome, the remaining samples had gains in 78% to 85% (Fig. 3B, left), indicating that the tumor cells experienced widespread gains of multiple chromosomes.

Figure 3.

Osteosarcoma copy number heterogeneity. A, Top, Euclidean distance clustering of copy number data, inferred from WGS, from each osteosarcoma metastatic sample in SJOS001101. M4 was not included due to very low tumor purity. Possible driver CNVs are indicated in red. Bottom, Circos plots showing copy number of each sample. Chromosome numbers are indicated around outside of plot. Sample names inside circle indicate the CNV tracks from top to bottom at the 12 o'clock position (see key on left plot). Red indicates copy number gain, blue copy number loss. Data are on a log2 scale and each gray line represents a log2 difference of 0.5. Gray dotted line represents the split between two major CNV genetic lineages. B, Left, percent of genome with copy gained, lost, and neutral (2 ± 0.2 copies) regions in each sample. Right, number of copies (linear; non-log2 scale) of indicated genes in each sample. Gray line indicates diploid status (2 copies). C–E, Circos plots as in A and additional graphs as in B for patients SJOS001105, SJOS001107, and SJOS010. SJOS001107 (D) and SJOS010 (E) were male and X chromosome copy scale is half of that shown in legend in A for these patients (black line with no color indicates one copy).

Figure 3.

Osteosarcoma copy number heterogeneity. A, Top, Euclidean distance clustering of copy number data, inferred from WGS, from each osteosarcoma metastatic sample in SJOS001101. M4 was not included due to very low tumor purity. Possible driver CNVs are indicated in red. Bottom, Circos plots showing copy number of each sample. Chromosome numbers are indicated around outside of plot. Sample names inside circle indicate the CNV tracks from top to bottom at the 12 o'clock position (see key on left plot). Red indicates copy number gain, blue copy number loss. Data are on a log2 scale and each gray line represents a log2 difference of 0.5. Gray dotted line represents the split between two major CNV genetic lineages. B, Left, percent of genome with copy gained, lost, and neutral (2 ± 0.2 copies) regions in each sample. Right, number of copies (linear; non-log2 scale) of indicated genes in each sample. Gray line indicates diploid status (2 copies). C–E, Circos plots as in A and additional graphs as in B for patients SJOS001105, SJOS001107, and SJOS010. SJOS001107 (D) and SJOS010 (E) were male and X chromosome copy scale is half of that shown in legend in A for these patients (black line with no color indicates one copy).

Close modal

To identify CNV driver genes, we identified Cancer Gene Census genes (14) with <0.5 copies (homozygous deletions) or >7 copies (gains; Fig. 3B, right). This revealed nonfocal truncal copy gains in MYC (5–9 copies per sample; 13–16 Mb region), AKT1 (5–9 copies; 5 Mb region); branch-specific nonfocal gains in CCND3 (6–10 copies in non-M2-M3 in a 24–27 Mb region, but only 3 copies in M2 and M3), and 3q13.31 deletion (0–1 copies in M2-M3; 0.5–1 Mb region); and a private TERT gain (7 copies in M7; 2 Mb region; Fig. 3B, right). Thus, CNVs show substantial intrapatient heterogeneity in SJOS001101.

We also compared intrapatient CNV profiles in SJOS001105 (Fig. 3C), SJOS001107 (Fig. 3D), and SJOS010 (Fig. 3E). These patients harbored truncal gains in CCND3, CDK4, CCNE1, and MYC, although the number of copies varied. Each patient also harbored truncal homozygous deletions on 3q13.31 (including LINC00901) and one patient had homozygous CDKN2A deletion (Fig. 3C–E; Supplementary Fig. S4; homozygous deletions were confirmed by B-allele frequencies in Supplementary Fig. S5). Like SJOS001101, SJOS001105 (Fig. 3C) and SJOS010 (Fig. 3E) had two lineages with large differences in ploidy. In SJOS001105, D1 and D2 had copy gains in 67% to 69% of the genome, whereas R1 had copy gains in 94% due to widespread copy gains (Fig. 3C, bottom-left). In SJOS010, sample D had copy gains in 69% of the genome whereas M had copy gains in 91% (Fig. 3E, bottom-left). Thus, three of four patients showed ploidy heterogeneity with widespread copy gains in one lineage; such gains are associated with increased metastatic potential and drug resistance (41).

These results indicate that osteosarcomas, known to have highly complex genomes (8), continue to evolve by acquiring additional gross chromosomal alterations after disease initiation. However, most clear driver CNVs, such as those affecting the CDK complex (cyclins, CDKs and CDK inhibitors) and MYC, are likely early events.

Intra- and intersite clonal evolution

To determine the clonal composition of each metastasis, we first determined the cancer cell fraction (CCF; the proportion of cancer cells belonging to a clone) of the site-specific “founder” clone (the clone defined by each sample's private SNVs) in each sample. To do this, we identified SNVs in two-copy regions in each sample and determined representative VAFs (from deep capture sequencing) of each SNV cluster by identifying VAF density peaks (Supplementary Fig. S8). Where VAF density peaks of private SNV clusters were very close to truncal SNV peaks, this indicated a high CCF of the site-specific founder clone.

For example, in five samples from SJOS001101 (M2, M3, M6, M7, M8), the VAF density peak for private variants was indeed close to the VAF density peak of the truncal variants (private/truncal > 0.95), indicating that the unique site-specific founder clone in each of these sites had a CCF >95% (Supplementary Fig. S8A). Thus, the dominant population at each of these sites was the site-specific founder clone of a single unique lineage of tumor cells harboring the truncal, shared, and private variants. The site-specific founder clones also gave rise to descendant clone(s) in M3, M6, and M7, as evidenced by minor private VAF density peaks (a second peak with lower VAF; Supplementary Fig. S8A). The M5 site-specific founder clone had a slightly lower CCF of 92%, indicating admixture of ∼8% of cells from other clone(s) harboring the M5 site-specific founder clone's truncal and shared SNV clusters (Supplementary Fig. S8A, red, orange peaks overlap truncal). Finally, M1 showed high admixture of clones, with the M1 site-specific founder clone(s) having a CCF of only 42% (Supplementary Fig. S8A). This may have been due to the presence of two discrete lung nodules in the M1 lung slice, and consequent sampling of two lineages (Fig. 1A). The low CCF made it impossible to resolve whether the M1 private variants represented one clone or multiple independent clones.

To determine whether cross-seeding (seeding of a single site by two independent lineages) occurred in SJOS001101, we performed pairwise sample comparisons of SNVs that were diploid in all samples (Supplementary Fig. S9). As expected, M2, M3, M6, M7, and M8 showed little evidence of crossover SNVs—private SNVs representing the site-specific founder clone in one sample which were also found at low VAF in another sample, evidencing cross-seeding—consistent with their high CCF for their site-specific founder clones (CCFs of >95%; Fig. 4A; Supplementary Figs. S8A, S9). This confirmed that these samples consisted of tumor cells from a single lineage (with descendant clones in M3, M6, and M7; Fig. 4A). Sample M5, in contrast, showed evidence of ∼4% of cancer cells being from sample M6, a nearby but spatially distinct site, suggesting cross-seeding between these sites (Fig. 4A; Supplementary Fig. S9 enlarged graphs). This is consistent with the estimated CCF of 92% for the M5 site-specific founder clone, which was lower than site-specific founder clones at other sites (Supplementary Fig. S8A). Two samples were not analyzed using this approach: M1 contained multiple distinct metastatic nodules, and M4 had <20% tumor purity.

Figure 4.

Osteosarcoma clonal heterogeneity and evolution. A, The clonal composition of each metastatic site is represented by colored circles. Sites with a >95% site-specific founder clone CCF (private VAF density/truncal VAF density >0.95) are represented by one solid circle, and those with both a site-specific founder clone and its descendant clone(s) are shown by two concentric circles. Cross-seeding is indicated by arrows. Circle area is proportional to CCF. Genetic cluster drivers for right vs. left anatomical side and selected shared and private drivers are indicated. M1 was not included due to containing multiple discrete lung nodules and M4 was not included due to low tumor purity. B, Clones are represented as in A. Metastatic seeding from a minor clone in the SJOS001107 primary tumor (D1) to lung metastases (M1 and M2) is indicated by arrows. Red represents M1 private variants not detectable in D1; yellow represents M2 private variants not detected in D1.

Figure 4.

Osteosarcoma clonal heterogeneity and evolution. A, The clonal composition of each metastatic site is represented by colored circles. Sites with a >95% site-specific founder clone CCF (private VAF density/truncal VAF density >0.95) are represented by one solid circle, and those with both a site-specific founder clone and its descendant clone(s) are shown by two concentric circles. Cross-seeding is indicated by arrows. Circle area is proportional to CCF. Genetic cluster drivers for right vs. left anatomical side and selected shared and private drivers are indicated. M1 was not included due to containing multiple discrete lung nodules and M4 was not included due to low tumor purity. B, Clones are represented as in A. Metastatic seeding from a minor clone in the SJOS001107 primary tumor (D1) to lung metastases (M1 and M2) is indicated by arrows. Red represents M1 private variants not detectable in D1; yellow represents M2 private variants not detected in D1.

Close modal

To validate clonal evolution results in SJOS001101, which has a total of eight samples, we used an automated approach based on PyClone (21) and MACHINA (22). This resulted in two possible clone trees (Supplementary Fig. S10). Both trees recapitulated cross-seeding between M5 and M6, and identified the M3 descendant clone of the M3 site-specific founder clone. The M6 and M7 descendant clones could not be detected with this method because pan-diploid SNVs were used, thus filtering out many single-sample-diploid variants used to detect these clones with density analysis (Supplementary Fig. S8A). Both trees also reaffirmed that most samples (M2, M3, M6, M7, M8) consisted of a site-specific founder clone and possibly its descendants. Clone tree 1 appeared more plausible than tree 2, as tree 2 possessed a branching event in an M5 precursor supported by a small subset of cluster 10 (Supplementary Fig. S10C). MACHINA analysis was consistent with our previous clonal analysis and reinforced our findings with a rigorous mathematical approach.

The clonal evolution of SJOS001101, combining mutation and CNVs, is summarized in Fig. 4A. All metastases shared inactivation of TP53, ATRX, and RB1 and copy gains in MYC and AKT1. We observed two genetic lineages which correlated spatially, with M2 and M3 being relatively close on the anatomical right and most other lesions on the left (Fig. 4A). The non-M2-M3 cluster (genetic cluster 2 in Fig. 4A) developed widespread copy gains with ∼80% of the genome having above two copies, including CCND3 gains; an NF1 variant in a subset of samples (M4, M7, M8; cluster 2b); and TERT copy gain in a single sample (M7). Within the M2-M3 cluster, we observed an activating KIT G565R private SNV in M3. Two samples (M2 and M8) consisted of a single dominant homogeneous clone (Fig. 4A). Three samples (M3, M6, and M7) were composed of tumor cells of a single lineage, which also branched off to descendant clone(s) as evidenced by minor VAF peaks among private variants (Fig. 4A; Supplementary Fig. S8A) and clustering analysis (Supplementary Fig. S10). These descendant clones might have developed after metastatic colonization. Finally, one sample, M5, showed evidence of cross-seeding, where multiple lineages colonize the same metastatic site. It is unclear whether an M5 population moved to M6, vice versa, or some other site seeded both. Some of the branching evolution may have been induced by cisplatin treatment; indeed, mutational signature analysis suggests that 46%–77% of SNVs in this patient were induced by cisplatin.

We also performed clonal evolution analysis on SJOS001105, SJOS001107, and SJOS010 (Supplementary Figs. S8B, S8C, and S11). SJOS001105 did not show evidence of cross-seeding between sites, and the primary tumor (D1) did not show evidence of a minor clone that went on to seed metastases D2 and R1 (Supplementary Figs. S11A). SJOS010 likewise did not show evidence of cross-seeding (Supplementary Fig. S11C). In SJOS001107, by contrast, deep capture sequencing revealed that the M1-M2 cluster of mutations (Fig. 1C), present in lung metastases M1 and M2 but not primary tumor D1 by WGS, was in fact detectable at a low level in D1 upon deeper sequencing (Supplementary Fig. S11B). Indeed, 250 of 458 SNVs (54.6%) in the M1-M2 cluster could be detected in D1 at a median VAF of 0.8%; other clusters did not “bleed” into other samples in this way (e.g., M2 private variants could be detected at low frequency in other samples for only 12 of 1,420 (0.6%) of SNVs, Fisher exact test P = 2 × 10−16, which was likely background noise; see Materials and Methods), indicating this was not artifactual (Supplementary Fig. S11B). This suggests that a minor clone (<5% CCF) detected in the primary tumor may have possessed enhanced metastatic potential, as it gave rise to both lung metastases (Fig. 4B; red indicates M1 private variants not detectable in the primary tumor; yellow indicates M2 private variants not detectable in the primary tumor). The identity of the variants potentially causing this metastatic potential is unclear, as we did not observe clear driver mutations in this mutation cluster. Overall, 6 of 10 metastases across three patients with sufficient two-copy SNVs showed evidence of single-clone seeding (Supplementary Fig. S8; private/truncal > 95%), suggesting this is a common form of metastatic seeding.

Although it has been known for decades that chemotherapy can induce mutations and secondary cancers (42, 43), the extent of cisplatin-induced mutational burden shown here—up to 77% of SNVs—is notable. It is unclear whether some of the nontruncal copy alterations we observed were also induced by cisplatin, but given cisplatin's ability to induce double-stranded DNA breaks (38), this possibility could not be ruled out. The evolutionary pattern of the cisplatin signature was consistent with a transient mutational process induced by treatment, as the signature appears after disease development (post-truncal variants) and disappears after treatment cessation as indicated by the shared and private branches in case SJOS001101 (Fig. 2C).

The cisplatin signature likely applies broadly to multiple cancer types. A recent pan-pediatric cancer study reported a signature similar to our cisplatin signature in one ependymoma, multiple atypical teratoid rhabdoid tumors (ATRT), and one osteosarcoma, though the cause was unknown (37). The ependymoma identified was from the PCGP, and we have confirmed that the specimen was acquired post-cisplatin. The osteosarcoma sample was also a PCGP sample, SJOS010, which we analyzed in this study and had received cisplatin. The ATRT patients' treatment history was unknown, but patients with ATRT commonly receive cisplatin (44). The cisplatin signature has also been found, with (35, 36, 45) or without (12) recognition of the cause, in liver (35), esophageal (35), breast (12), ovarian (45), and bladder cancer (36) after cisplatin treatment.

One weakness of our study is the lack of primary tumors in two patients (SJOS001101 and SJOS010). Therefore, we could not rule out the possibility that some of the “truncal” variants described here may have been acquired later in evolutionary history. Nevertheless, the ubiquitous absence of cisplatin signature in truncal variants and in 30 untreated osteosarcoma samples from PCGP provided strong support for the association between the cisplatin signature and the treatment history of osteosarcoma. Further, we did not observe the signature in 19 untreated osteosarcomas in a recent study (16). Given that our primary analysis only focuses on four patients, future studies are needed to investigate the contribution of cisplatin-induced mutations in osteosarcoma tumor progression, including functional analysis of cisplatin-induced variants.

Our analysis of SJOS001101, the only case with multiregional lung samples, showed that cross-seeding occurred in only one of six metastases (Fig. 4A), which suggests that multiple-seeding may not be common in osteosarcoma (46, 47). Our findings fit the original view that metastasis arises from a single clone in most cases (48). Additional studies are required to determine whether this is a general observation.

Our findings highlight the importance of investigation of mutational signatures in possible chemotherapy-induced secondary malignancies (4, 49). Indeed, patients with osteosarcoma can develop secondary malignancies up to 25 years after treatment (4). Although this may be due to inherited germline mutations in cancer predisposition genes (50), further studies on noninherited cases are needed to investigate whether the cisplatin signature might be associated with the development of the secondary malignancy. More broadly, it will be valuable to determine the point and structural mutation signatures of every DNA-damaging chemotherapy to better understand the implications for both secondary malignancies and the development of drug resistance.

B.J. Raphael is a consultant at and has ownership interest (including stock, patents, etc.) in Medley Genomics. No potential conflicts of interest were disclosed for other authors.

Conception and design: R.K. Wilson, J.R. Downing, J. Zhang

Development of methodology: S.W. Brady, X. Ma, J. Easton, E.R. Mardis, R.K. Wilson, B.J. Raphael, J. Zhang

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Bahrami, H.L. Mulder, J. Easton, E.R. Mardis, R.K. Wilson, A.S. Pappo

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S.W. Brady, X. Ma, A. Bahrami, G. Satas, G. Wu, S. Newman, M. Rusch, D.K. Putnam, M.N. Edmonson, L.B. Alexandrov, X. Chen, A.S. Pappo, B.J. Raphael, J. Zhang

Writing, review, and/or revision of the manuscript: S.W. Brady, X. Ma, H.L. Mulder, E.R. Mardis, R.K. Wilson, A.S. Pappo, M.A. Dyer, J. Zhang

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): D.A. Yergeau, R.K. Wilson, M.A. Dyer, J. Zhang

Study supervision: J. Zhang

The authors thank Timothy Hammond and the Biomedical Communications Department at St. Jude Children's Research Hospital for illustrations. We acknowledge Yu Liu for help with Circos plots, Yongjin Li for computer code, and David Ellison for information on ependymoma treatment history. We thank the Tissue Bank at St. Jude Children's Research Hospital for managing samples. This research was supported by the NCI through Cancer Center Support Grant P30 CA021765 (to J. Zhang) and by the American Lebanese Syrian Associated Charities of St. Jude Children's Research Hospital.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
McGranahan
N
,
Swanton
C
. 
Clonal heterogeneity and tumor evolution: past, present, and the future
.
Cell
2017
;
168
:
613
28
.
2.
Turke
AB
,
Zejnullahu
K
,
Wu
YL
,
Song
Y
,
Dias-Santagata
D
,
Lifshits
E
, et al
Preexistence and clonal selection of MET amplification in EGFR mutant NSCLC
.
Cancer Cell
2010
;
17
:
77
88
.
3.
de Bruin
EC
,
McGranahan
N
,
Mitter
R
,
Salm
M
,
Wedge
DC
,
Yates
L
, et al
Spatial and temporal diversity in genomic instability processes defines lung cancer evolution
.
Science
2014
;
346
:
251
6
.
4.
Luetke
A
,
Meyers
PA
,
Lewis
I
,
Juergens
H
. 
Osteosarcoma treatment - where do we stand? A state of the art review
.
Cancer Treat Rev
2014
;
40
:
523
32
.
5.
Ottaviani
G
,
Jaffe
N
. 
The epidemiology of osteosarcoma
.
In:
Jaffe
N
,
Bruland
OS
,
Bielack
S
,
editors
.
Pediatric and adolescent osteosarcoma
.
Boston
:
Springer
; 
2009
.
p.
3
13
.
6.
Smith
MA
,
Seibel
NL
,
Altekruse
SF
,
Ries
LA
,
Melbert
DL
,
O'Leary
M
, et al
Outcomes for children and adolescents with cancer: challenges for the twenty-first century
.
J Clin Oncol
2010
;
28
:
2625
34
.
7.
Harrison
DJ
,
Geller
DS
,
Gill
JD
,
Lewis
VO
,
Gorlick
R
. 
Current and future therapeutic approaches for osteosarcoma
.
Expert Rev Anticancer Ther
2018
;
18
:
39
50
.
8.
Chen
X
,
Bahrami
A
,
Pappo
A
,
Easton
J
,
Dalton
J
,
Hedlund
E
, et al
Recurrent somatic structural variations contribute to tumorigenesis in pediatric osteosarcoma
.
Cell Rep
2014
;
7
:
104
12
.
9.
Li
H
,
Durbin
R
. 
Fast and accurate short read alignment with Burrows-Wheeler transform
.
Bioinformatics
2009
;
25
:
1754
60
.
10.
Edmonson
MN
,
Zhang
J
,
Yan
C
,
Finney
RP
,
Meerzaman
DM
,
Buetow
KH
. 
Bambino: a variant detector and alignment viewer for next-generation sequencing data in the SAM/BAM format
.
Bioinformatics
2011
;
27
:
865
6
.
11.
Zhang
J
,
Ding
L
,
Holmfeldt
L
,
Wu
G
,
Heatley
SL
,
Payne-Turner
D
, et al
The genetic basis of early T-cell precursor acute lymphoblastic leukaemia
.
Nature
2012
;
481
:
157
63
.
12.
Brady
SW
,
McQuerry
JA
,
Qiao
Y
,
Piccolo
SR
,
Shrestha
G
,
Jenkins
DF
, et al
Combating subclonal evolution of resistant cancer phenotypes
.
Nat Commun
2017
;
8
:
1231
.
13.
Chen
X
,
Gupta
P
,
Wang
J
,
Nakitandwe
J
,
Roberts
K
,
Dalton
JD
, et al
CONSERTING: integrating copy-number analysis with structural-variation detection
.
Nat Methods
2015
;
12
:
527
30
.
14.
Futreal
PA
,
Coin
L
,
Marshall
M
,
Down
T
,
Hubbard
T
,
Wooster
R
, et al
A census of human cancer genes
.
Nat Rev Cancer
2004
;
4
:
177
83
.
15.
Alexandrov
LB
,
Nik-Zainal
S
,
Wedge
DC
,
Campbell
PJ
,
Stratton
MR
. 
Deciphering signatures of mutational processes operative in human cancer
.
Cell Rep
2013
;
3
:
246
59
.
16.
Ma
X
,
Liu
Y
,
Liu
Y
,
Alexandrov
LB
,
Edmonson
MN
,
Gawad
C
, et al
Pan-cancer genome and transcriptome analyses of 1,699 paediatric leukaemias and solid tumours
.
Nature
2018
;
555
:
371
6
.
17.
Yates
LR
,
Knappskog
S
,
Wedge
D
,
Farmery
JHR
,
Gonzalez
S
,
Martincorena
I
, et al
Genomic evolution of breast cancer metastasis and relapse
.
Cancer Cell
2017
;
32
:
169
184
.
18.
Morganella
S
,
Alexandrov
LB
,
Glodzik
D
,
Zou
X
,
Davies
H
,
Staaf
J
, et al
The topography of mutational processes in breast cancer genomes
.
Nat Commun
2016
;
7
:
11383
.
19.
Gerlinger
M
,
Rowan
AJ
,
Horswell
S
,
Larkin
J
,
Endesfelder
D
,
Gronroos
E
, et al
Intratumor heterogeneity and branched evolution revealed by multiregion sequencing
.
N Engl J Med
2012
;
366
:
883
92
.
20.
Ma
X
,
Edmonson
M
,
Yergeau
D
,
Muzny
DM
,
Hampton
OA
,
Rusch
M
, et al
Rise and fall of subclones from diagnosis to relapse in pediatric B-acute lymphoblastic leukaemia
.
Nat Commun
2015
;
6
:
6604
.
21.
Roth
A
,
Khattra
J
,
Yap
D
,
Wan
A
,
Laks
E
,
Biele
J
, et al
PyClone: statistical inference of clonal population structure in cancer
.
Nat Methods
2014
;
11
:
396
8
.
22.
El-Kebir
M
,
Satas
G
,
Raphael
BJ
. 
Inferring parsimonious migration histories for metastatic cancers
.
Nat Genet
2018
;
50
:
718
26
.
23.
Navid
F
,
Santana
VM
,
Neel
M
,
McCarville
MB
,
Shulkin
BL
,
Wu
J
, et al
A phase II trial evaluating the feasibility of adding bevacizumab to standard osteosarcoma therapy
.
Int J Cancer
2017
;
141
:
1469
77
.
24.
Meyer
WH
,
Pratt
CB
,
Poquette
CA
,
Rao
BN
,
Parham
DM
,
Marina
NM
, et al
Carboplatin/ifosfamide window therapy for osteosarcoma: results of the St Jude Children's Research Hospital OS-91 trial
.
J Clin Oncol
2001
;
19
:
171
82
.
25.
Wang
J
,
Mullighan
CG
,
Easton
J
,
Roberts
S
,
Heatley
SL
,
Ma
J
, et al
CREST maps somatic structural variation in cancer genomes with base-pair resolution
.
Nat Methods
2011
;
8
:
652
4
.
26.
Perry
JA
,
Kiezun
A
,
Tonzi
P
,
Van Allen
EM
,
Carter
SL
,
Baca
SC
, et al
Complementary genomic approaches highlight the PI3K/mTOR pathway as a common vulnerability in osteosarcoma
.
Proc Natl Acad Sci U S A
2014
;
111
:
E5564
73
.
27.
Pasic
I
,
Shlien
A
,
Durbin
AD
,
Stavropoulos
DJ
,
Baskin
B
,
Ray
PN
, et al
Recurrent focal copy-number changes and loss of heterozygosity implicate two noncoding RNAs and one tumor suppressor gene at chromosome 3q13.31 in osteosarcoma
.
Cancer Res
2010
;
70
:
160
71
.
28.
Debiec-Rychter
M
,
Cools
J
,
Dumez
H
,
Sciot
R
,
Stul
M
,
Mentens
N
, et al
Mechanisms of resistance to imatinib mesylate in gastrointestinal stromal tumors and activity of the PKC412 inhibitor against imatinib-resistant mutants
.
Gastroenterology
2005
;
128
:
270
9
.
29.
Palacios
GA
,
Molina-Vila
,
Gras-Cabrerizo
JR
,
León
X
,
Mayo
C
,
Servat
JC
, et al
Molecular analysis in serial biopsies in sinonasal mucosal melanoma
.
Ann Oncol
2016
;
27
(suppl_6):1005P
.
30.
Gluszek
S
,
Koziel
D
,
Kowalik
A
,
Zięba
S
,
Urbaniak-Wasik
S
,
Wincewicz
A
, et al
KRAS, KIT and TP53 mutations in mother's and daughter's gastric cardia adenocarcinomas
.
Gastroenterol Rev
2018
;
13
:
76
9
.
31.
Shao
YW
,
Wood
GA
,
Lu
J
,
Tang
QL
,
Liu
J
,
Molyneux
S
, et al
Cross-species genomics identifies DLG2 as a tumor suppressor in osteosarcoma
.
Oncogene
2019
;
38
:
291
8
.
32.
Reiter
JG
,
Makohon-Moore
AP
,
Gerold
JM
,
Heyde
A
,
Attiyeh
MA
,
Kohutek
ZA
, et al
Minimal functional driver gene heterogeneity among untreated metastases
.
Science
2018
;
361
:
1033
7
.
33.
Alexandrov
LB
,
Nik-Zainal
S
,
Wedge
DC
,
Aparicio
SA
,
Behjati
S
,
Biankin
AV
, et al
Signatures of mutational processes in human cancer
.
Nature
2013
;
500
:
415
21
.
34.
Helleday
T
,
Eshtad
S
,
Nik-Zainal
S
. 
Mechanisms underlying mutational signatures in human cancers
.
Nat. Rev. Genet.
2014
;
15
:
585
98
.
35.
Boot
A
,
Huang
MN
,
Ng
AWT
,
Ho
SC
,
Lim
JQ
,
Kawakami
Y
, et al
In-depth characterization of the cisplatin mutational signature in human cell lines and in esophageal and liver tumors
.
Genome Res
2018
;
28
:
654
65
.
36.
Liu
D
,
Abbosh
P
,
Keliher
D
,
Reardon
B
,
Miao
D
,
Mouw
K
, et al
Mutational patterns in chemotherapy resistant muscle-invasive bladder cancer
.
Nat Commun
2017
;
8
:
2193
.
37.
Gröbner
SN
,
Worst
BC
,
Weischenfeldt
J
,
Buchhalter
I
,
Kleinheinz
K
,
Rudneva
VA
, et al
The landscape of genomic alterations across childhood cancers
.
Nature
2018
;
555
:
321
7
.
38.
Huang
X
,
Okafuji
M
,
Traganos
F
,
Luther
E
,
Holden
E
,
Darzynkiewicz
Z
. 
Assessment of histone H2AX phosphorylation induced by DNA topoisomerase I and II inhibitors topotecan and mitoxantrone and by the DNA cross-linking agent cisplatin
.
Cytometry
2004
;
58A
:
99
110
.
39.
Alexandrov
LB
,
Jones
PH
,
Wedge
DC
,
Sale
JE
,
Campbell
PJ
,
Nik-Zainal
S
, et al
Clock-like mutational processes in human somatic cells
.
Nat Genet
2015
;
47
:
1402
7
.
40.
Kovac
M
,
Blattmann
C
,
Ribi
S
,
Smida
J
,
Mueller
NS
,
Engert
F
, et al
Exome sequencing of osteosarcoma reveals mutation signatures reminiscent of BRCA deficiency
.
Nat Commun
2015
;
6
:
8940
.
41.
Bakhoum
SF
,
Cantley
LC
. 
The multifaceted role of chromosomal instability in cancer and its microenvironment
.
Cell
2018
;
174
:
1347
60
.
42.
Ewig
RA
,
Kohn
KW
. 
DNA damage and repair in mouse leukemia L1210 cells treated with nitrogen mustard, 1,3-bis(2-chloroethyl)-1-nitrosourea, and other nitrosoureas
.
Cancer Res
1977
;
37
:
2114
22
.
43.
Harris
CC
. 
A delayed complication of cancer therapy—cancer
.
J Natl Cancer Inst
1979
;
63
:
275
7
.
44.
Zaky
W
,
Dhall
G
,
Ji
L
,
Haley
K
,
Allen
J
,
Atlas
M
, et al
Intensive induction chemotherapy followed by myeloablative chemotherapy with autologous hematopoietic progenitor cell rescue for young children newly-diagnosed with central nervous system atypical teratoid/rhabdoid tumors: The head start III experience
.
Pediatr Blood Cancer
2014
;
61
:
95
101
.
45.
Patch
AM
,
Christie
EL
,
Etemadmoghadam
D
,
Garsed
DW
,
George
J
,
Fereday
S
, et al
Whole-genome characterization of chemoresistant ovarian cancer
.
Nature
2015
;
521
:
489
94
.
46.
Hoadley
KA
,
Siegel
MB
,
Kanchi
KL
,
Miller
CA
,
Ding
L
,
Zhao
W
, et al
Tumor evolution in two patients with basal-like breast cancer: a retrospective genomics study of multiple metastases
.
PLOS Med
2016
;
13
:
e1002174
.
47.
Kim
MY
,
Oskarsson
T
,
Acharyya
S
,
Nguyen
DX
,
Zhang
XH
,
Norton
L
, et al
Tumor self-seeding by circulating cancer cells
.
Cell
2009
;
139
:
1315
26
.
48.
Chaffer
CL
,
Weinberg
RA
. 
A perspective on cancer cell metastasis
.
Science
2011
;
331
:
1559
64
.
49.
Bhatia
S
,
Robison
LL
,
Oberlin
O
,
Greenberg
M
,
Bunin
G
,
Fossati-Bellani
F
, et al
Breast cancer and other second neoplasms after childhood Hodgkin's disease
.
N Engl J Med
1996
;
334
:
745
51
.
50.
Toguchida
J
,
Yamaguchi
T
,
Dayton
SH
,
Beaughamp
RL
,
Herrera
GE
,
Ishizaki
K
, et al
Prevalence and spectrum of germline mutations of the p53 gene among patients with sarcoma
.
N Engl J Med
1992
;
326
:
1301
8
.