Abstract
Sporadic reports have described cancer cases in which multiple driver mutations (MMs) occur in the same oncogene1,2. However, the overall landscape and relevance of MMs remain elusive. Here we carried out a pan-cancer analysis of 60,954 cancer samples, and identified 14 pan-cancer and 6 cancer-type-specific oncogenes in which MMs occur more frequently than expected: 9% of samples with at least one mutation in these genes harboured MMs. In various oncogenes, MMs are preferentially present in cis and show markedly different mutational patterns compared with single mutations in terms of type (missense mutations versus in-frame indels), position and amino-acid substitution, suggesting a cis-acting effect on mutational selection. MMs show an overrepresentation of functionally weak, infrequent mutations, which confer enhanced oncogenicity in combination. Cells with MMs in the PIK3CA and NOTCH1 genes exhibit stronger dependencies on the mutated genes themselves, enhanced downstream signalling activation and/or greater sensitivity to inhibitory drugs than those with single mutations. Together oncogenic MMs are a relatively common driver event, providing the underlying mechanism for clonal selection of suboptimal mutations that are individually rare but collectively account for a substantial proportion of oncogenic mutations.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Our findings are supported by data that are available from public online repositories, or data that are publicly available upon request from the data provider. See the ‘Data set preparation’ section above for details. Long-read WGS data for cell lines have been deposited in the European Genome-phenome Archive (EGA; https://www.ebi.ac.uk/ega/) under accession number EGAS00001003763. Data generated here are available as Source Data files accompanying Fig. 4 and Extended Data Figs. 10, 11.
Code availability
Source codes for CisChecker, DNVChecker and the permutation test are available at https://github.com/nccmo/CisChecker, https://github.com/nccmo/DNVChecker and https://github.com/nccmo/Permutation-test, respectively.
References
Kohsaka, S. et al. A method of high-throughput functional evaluation of EGFR gene variants of unknown significance in cancer. Sci. Transl. Med. 9, eaan6566 (2017).
Madsen, R. R. et al. Oncogenic PIK3CA promotes cellular stemness in an allele dose-dependent manner. Proc. Natl Acad. Sci. USA 116, 8380–8389 (2019).
Garraway, L. A. & Lander, E. S. Lessons from the cancer genome. Cell 153, 17–37 (2013).
Vogelstein, B. et al. Cancer genome landscapes. Science 339, 1546–1558 (2013).
Heinrich, M. C. et al. Molecular correlates of imatinib resistance in gastrointestinal stromal tumors. J. Clin. Oncol. 24, 4764–4774 (2006).
Kobayashi, S. et al. EGFR mutation and resistance of non-small-cell lung cancer to gefitinib. N. Engl. J. Med. 352, 786–792 (2005).
Pao, W. et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain. PLoS Med. 2, e73 (2005).
Soh, J. et al. Oncogene mutations, copy number gains and mutant allele specific imbalance (MASI) frequently occur together in tumor cells. PLoS One 4, e7464 (2009).
Lawrence, M. S. et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499, 214–218 (2013).
Martincorena, I. et al. Universal patterns of selection in cancer and somatic tissues. Cell 171, 1029–1041 (2017).
Weng, A. P. et al. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 306, 269–271 (2004).
Dogruluk, T. et al. Identification of variant-specific functions of PIK3CA by rapid phenotyping of rare mutations. Cancer Res. 75, 5341–5354 (2015).
Ng, P. K. et al. Systematic functional annotation of somatic mutations in cancer. Cancer Cell 33, 450–462 (2018).
Ghandi, M. et al. Next-generation characterization of the Cancer Cell Line Encyclopedia. Nature 569, 503–508 (2019).
Gkeka, P. et al. Investigating the structure and dynamics of the PIK3CA wild-type and H1047R oncogenic mutant. PLOS Comput. Biol. 10, e1003895 (2014).
Zhang, M., Jang, H. & Nussinov, R. The mechanism of PI3Kα activation at the atomic level. Chem. Sci. 10, 3671–3680 (2019).
Ding, L. et al. Perspective on oncogenic processes at the end of the beginning of cancer genomics. Cell 173, 305–320 (2018).
Sanchez-Vega, F. et al. Oncogenic signaling pathways in The Cancer Genome Atlas. Cell 173, 321–337 (2018).
Bailey, M. H. et al. Comprehensive characterization of cancer driver genes and mutations. Cell 173, 371–385 (2018).
Ma, X. et al. Pan-cancer genome and transcriptome analyses of 1,699 paediatric leukaemias and solid tumours. Nature 555, 371–376 (2018).
Mullighan, C. G. et al. Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. N. Engl. J. Med. 360, 470–480 (2009).
Pugh, T. J. et al. The genetic landscape of high-risk neuroblastoma. Nat. Genet. 45, 279–284 (2013).
Tyner, J. W. et al. Functional genomic landscape of acute myeloid leukaemia. Nature 562, 526–531 (2018).
Chapuy, B. et al. Molecular subtypes of diffuse large B cell lymphoma are associated with distinct pathogenic mechanisms and outcomes. Nat. Med. 24, 679–690 (2018).
AACR Project GENIE Consortium. AACR Project GENIE: powering precision medicine through an international consortium. Cancer Discov. 7, 818–831 (2017).
Hartmaier, R. J. et al. High-throughput genomic profiling of adult solid tumors reveals novel insights into cancer pathogenesis. Cancer Res. 77, 2464–2475 (2017).
Kataoka, K. et al. Integrated molecular analysis of adult T cell leukemia/lymphoma. Nat. Genet. 47, 1304–1315 (2015).
Campbell, B. B. et al. Comprehensive analysis of hypermutation in human cancer. Cell 171, 1042–1056 (2017).
Papaemmanuil, E. et al. Genomic classification and prognosis in acute myeloid leukemia. N. Engl. J. Med. 374, 2209–2221 (2016).
Gu, Z. et al. PAX5-driven subtypes of B-progenitor acute lymphoblastic leukemia. Nat. Genet. 51, 296–307 (2019).
Li, J. F. et al. Transcriptional landscape of B cell precursor acute lymphoblastic leukemia based on an international study of 1,223 cases. Proc. Natl Acad. Sci. USA 115, E11711–E11720 (2018).
Steeghs, E. M. P. et al. JAK2 aberrations in childhood B-cell precursor acute lymphoblastic leukemia. Oncotarget 8, 89923–89938 (2017).
Forero-Castro, M. et al. Mutations in TP53 and JAK2 are independent prognostic biomarkers in B-cell precursor acute lymphoblastic leukaemia. Br. J. Cancer 117, 256–265 (2017).
Puente, X. S. et al. Non-coding recurrent mutations in chronic lymphocytic leukaemia. Nature 526, 519–524 (2015).
Bielski, C. M. et al. Widespread selection for oncogenic mutant allele imbalance in cancer. Cancer Cell 34, 852–862 (2018).
Buisson, R. et al. Passenger hotspot mutations in cancer driven by APOBEC3A and mesoscale genomic features. Science 364, eaaw2872 (2019).
Chakravarty, D. et al. OncoKB: a precision oncology knowledge base. JCO Precis. Oncol. 1, 1–16 (2017).
Huang, L. et al. The cancer precision medicine knowledge base for structured clinical-grade mutations and interpretations. J. Am. Med. Inform. Assoc. 24, 513–519 (2017).
Tamborero, D. et al. Cancer Genome Interpreter annotates the biological and clinical relevance of tumor alterations. Genome Med. 10, 25 (2018).
Griffith, M. et al. CIViC is a community knowledgebase for expert crowdsourcing the clinical interpretation of variants in cancer. Nat. Genet. 49, 170–174 (2017).
Gao, J. et al. 3D clusters of somatic mutations in cancer reveal numerous rare mutations as functional targets. Genome Med. 9, 4 (2017).
Chang, M. T. et al. Identifying recurrent mutations in cancer reveals widespread lineage diversity and mutational specificity. Nat. Biotechnol. 34, 155–163 (2016).
Chang, M. T. et al. Accelerating discovery of functional mutant alleles in cancer. Cancer Discov. 8, 174–183 (2018).
Tokheim, C. et al. Exome-scale discovery of hotspot mutation regions in human cancer using 3D protein structure. Cancer Res. 76, 3719–3731 (2016).
Tokheim, C. & Karchin, R. CHASMplus reveals the scope of somatic missense mutations driving human cancers. Cell Syst. 9, 9–23 (2019).
Acknowledgements
We acknowledge support from the Japan Society for the Promotion of Science (JSPS) KAKENHI (grant numbers 17K19592,18K06594 and 15H05912) and National Cancer Center Research and Development Funds (30-A-1), together with many other funding bodies, organizations and individuals (see Supplementary Note).
Author information
Authors and Affiliations
Contributions
Y. Saito and K.K. designed the study. Y. Saito, Y.K. and K.K. analysed sequencing data. M.B.M., H.T., T.K., S. Miyano and Y. Shiraishi assisted with the collection and analysis of sequencing data. S.S. and M.T. performed sequencing experiments. J.K. and K.Y. performed immunoblots, capillary-based immunoassays, cell proliferation assays and in vivo xenograft assays. J.K. and Y.K. performed luciferase assays. M.A., S. Matsumoto, Y.I. and Y.O. performed molecular-dynamics simulations. Y. Saito and K.K. generated figures and tables and wrote the manuscript. K.K. led the entire project. All authors participated in discussions and interpretation of the data and results.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature thanks Mark Lackner, Inigo Martincorena and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Workflow showing discovery and assessment of oncogenic MMs.
a, Data set preparation, identification of MM+ oncogenes, and analysis of genetic features, lineage specificity and functional role. Distribution of cancer types (Supplementary Table 1) for the total cohort (n = 60,954). Cancer types with less than 1% frequency were combined as ‘OTHERS’. AF, allele frequency; NCCRCC, non-clear cell renal-cell carcinoma; NSCLCNOS, NSCLC not otherwise specified. b, Classification of hotspot/functional mutations according to mutation type, position and amino-acid change.
Extended Data Fig. 2 Data set preparation for the TCGA, TARGET, HM, FM and GENIE cohorts.
a, Steps involved in reannotation, sample exclusion and mutation curation in each cohort. b, Number of mutated samples (right) and proportion of samples in each cohort (left) for 45 cancer types (with 200 or more samples). ACC, adrenocortical carcinoma; APAD, appendiceal adenocarcinoma; GBC, gallbladder cancer; GINET, gastrointestinal neuroendocrine tumours; MNGT, meningothelial tumour; NBL, neuroblastoma; OS, osteosarcoma; PANET, pancreatic neuroendocrine tumour; PCPG, pheochromocytoma and paraganglioma; PLMESO, pleural mesothelioma; RCCNOS, renal-cell carcinoma not otherwise specified; TET, thymic epithelial tumour; UCP, undifferentiated carcinoma of the pancreas, USARC, uterine sarcoma/mesenchymal (see also Supplementary Table 1). c, DNVs and TNVs arising in the same codon (considered as a single genomic event) in representative cases. The TCGA identification code for the sample is shown below each chart. d, Frequency of MM+ samples with or without DNV and TNV curation for 30 oncogenes (as in Fig. 1a) in primary samples from TCGA. e, Removed SNVs around indels. In this representative case, SNVs (T-to-C) and frameshift deletions (AGTGGTTTCTT-to-C) were independently miscalled, but considered to be a single genomic event. f, Distribution of cancer types for primary samples included in the TCGA, TARGET, HM, FM and GENIE cohorts. Cancer types with less than 1% frequency were combined as ‘OTHERS’ in each cohort. WT, Wilms’ tumour.
Extended Data Fig. 3 Identification of MM+ oncogenes by a permutation test across cancers.
a, Number of mutated samples and frequency of MM+ samples for 30 oncogenes (as in Fig. 1a) after excluding microsatellite instability (MSI)-high or POLE/POLD1-mutated samples in the discovery cohort. b, Frequency of MM+ samples in PIK3CA, EGFR, CTNNB1, ERBB2 and ERBB3 according to tumour purity in primary samples from the TCGA cohort (n = 8,699). c, Number of mutated samples and frequency of MM+ samples for 30 oncogenes (as in Fig. 1a) in the TCGA, TARGET, HM, FM and GENIE cohorts. d, Representation of the permutation-based framework. In the standard approach, to identify genes significantly affected by driver mutations, the expected number of samples with mutations in gene X (the gene of interest; green) is estimated by permuting all coding mutations randomly across the coding region in all samples (for example, samples A–F). Statistical significance is determined by comparing the observed number of samples with nonsynonymous mutations and the expected distribution in gene X. In our approach, to identify genes significantly affected by putative driver–driver MMs, the expected number of samples with MMs in gene X (red) is estimated by permuting all coding mutations other than gene X mutations in samples harbouring gene X mutations (samples C–F). Statistical significance is determined by comparing the observed number of samples with MMs and the expected distribution in gene X. e, In the random-choice model, mutations are moved to another position with equal probability (P), whereas in the weighted-choice model, mutations are moved with unequal probability (P´), reflecting expression and DNA replication time. f, Synonymous mutation frequency per megabase according to expression and DNA replication time. g, Pathways related to 60 oncogenes analysed here. MM+ oncogenes identified in pan-cancer and cancer type-specific analyses are indicated in red (q < 0.001)/pink (q < 0.01) and green, respectively. h, Proportion of synonymous to total mutations according to MM status in TSGs and NFGs. i, Proportion of synonymous to total mutations according to MM status in MM− oncogenes under high and low selective pressure (that is, oncogenes in which the proportion of synonymous to total mutations is less than and more than 15% in samples with single mutations, respectively). The proportion of synonymous mutations was substantially increased in MM+ samples, even in MM− oncogenes under high selective pressure. b, h, i, Two-sided Fisher’s exact test. The numbers examined are shown in parentheses.
Extended Data Fig. 4 Allelic configuration (cis versus trans) of oncogenic MMs.
a, Top, sequencing reads encompassing MMs were classified into those containing both mutant alleles (cis reads), one mutant and one reference allele (trans reads), and both reference alleles (reference reads). Bottom, one-sided permutation test (n = 10,000) for the allelic configuration (cis versus trans) of MMs. In this representative case, the observed numbers of cis (left) and trans (right) reads (red) were significantly higher and lower, respectively, than the expected distribution (black); thus, this example is considered to be cis. b, Difference in mutant allele frequency (AF) between MMs across 60 oncogenes and 35 TSGs in the discovery cohort. Each dot represents an MM, coloured by cancer type. Two-sided Brunner–Munzel test. Box plots show medians (lines), interquartile ranges (IQRs; boxes) and ±1.5 × IQRs (whiskers). c, Proportion of MMs (combinations of missense mutations only) showing concordant or discordant allele frequencies in MM+ oncogenes in primary samples from the total cohort. d, Fraction of PIK3CA, EGFR, ERBB2 and PDGFRA copy-number (CN) alterations according to MM status in recurrently mutated cancer types (defined as those with 20 or more hotspot/functional mutations) in primary samples from TCGA. e, Proportion of MMs in cis and trans (with distances between mutations of 25 bp or more) by phasing from RNA-seq or WES/WGS in MM+ oncogenes with and without concurrent CN amplification of the mutated gene. f, Allelic configuration (cis versus trans) assessed by cDNA amplicon sequencing for PIK3CA P539R–H1047R, E545D–E970K and E545K–D549N mutations in BT-20, SUP-T1 and HRT-18 cell lines, respectively. Proportions of mutant and reference alleles are shown. b–f, Examined numbers are shown in parentheses. g, Density plot illustrating the distribution of read length and average read quality for each of three long-read WGS samples. h, Percentage of bases covered by at least ×2, ×5, ×10, ×20 and ×30 sequencing reads for three long-read WGS samples. i, Validation rate of SNV calling from long-read WGS according to base quality and/or flanking indels. Examined read numbers are shown in parentheses. d, e, i, Two-sided Fisher’s exact test. j, Density plot showing the correlation between variant allele frequencies in short-read and long-read WGS in positions with coverage of at least ×40 and at least ×20, in short-read and long-read WGS, respectively. Two-sided Pearson’s correlation test. k, Phasing of MMs using long-read WGS reads. Positions of MMs (red) and in-between SNVs (blue) according to their genomic position (top) and long-read WGS reads between them (bottom). Reads supporting both mutant alleles and both reference alleles present in cis are shown in orange and black, respectively; discordant reads are shown in green.
Extended Data Fig. 5 Comparison of PIK3CA MMs among cohorts.
a, Proportion of PIK3CA mutations according to type, position and amino-acid change in recurrently mutated cancer types (defined as those with 20 or more hotspot/functional mutations) in primary samples from the TCGA, FM and GENIE cohorts. b, Distribution of mutations and fraction of MM+ mutations for each hotspot/functional position. Asterisks indicate major positions (in which 10% or more of mutations were present in any of the recurrently mutated cancer types). The horizontal blue dotted lines represent the mean values of major positions. c, Number of MMs according to mutational combinations (n = 407). Red and black circles indicate, respectively, observed and expected (median with 95% confidence intervals) values. Representative combinations and observed numbers are also shown. d, Significant pairwise associations (q < 0.01) with observed/expected ratios among functional domains (n = 471). Orange and blue colours depict co-occurring (observed number of MMs significantly higher than expected) and mutually exclusive (lower than expected) associations. c, d, Two-sided simulation test (n = 10,000) with Benjamini–Hochberg correction. e, Major–major combinations on different alleles within close proximity (E542–E545). f, Proportion of MMs showing concordant or discordant allele frequencies according to mutational combinations. g, Order of major and minor hotspot mutations in major–minor combinations showing discordant allele frequencies (n = 54). Two-sided binomial test. h, Proportion of mutations with APOBEC signature (C-to-G/T at TpCpX trinucleotides) according to hotspot/functional position in PIK3CA in the discovery cohort. i, Proportion of mutations with APOBEC signature according to MM status in all coding (left) and PIK3CA (right) mutations in the discovery cohort. b, f, h, i, Two-sided Fisher’s exact test. Examined numbers are shown in parentheses.
Extended Data Fig. 6 Genetic features of MMs in a variety of oncogenes.
a, Proportion of mutations according to type, position and amino-acid change across MM+ oncogenes in recurrently mutated cancer types (defined as those with 20 or more hotspot/functional mutations) in primary samples from the total cohort. b, Distribution of mutations and fraction of MM+ mutations for each hotspot/functional position in EGFR, KRAS, ERBB2 and CTNNB1. Positions showing five mutations or more (and within the top 20 for EGFR) are shown in the bar plots. Asterisks indicate major positions (in which 10% or more of mutations were present in any of the recurrently mutated cancer types). The horizontal blue dotted lines represent the mean values of major positions. c, Number of MMs according to mutational combinations in EGFR (n = 185), KRAS (n = 39), ERBB2 (n = 24) and CTNNB1 (n = 19). Red and black circles indicate, respectively, observed and expected (median with 95% confidence intervals) values. Two-sided simulation test (n = 10,000) with Benjamini–Hochberg correction. d, Major–major combinations on different alleles within close proximity (NRAS G12–G13). e, Proportion of MMs showing concordant or discordant allele frequencies (with distance between mutations of less than 25 bp) according to allelic configuration (cis versus trans). b, e, Two-sided Fisher’s exact test. f, Average mutant allele frequency of MMs showing concordant allele frequencies present in copy-number-neutral region according to allelic configuration (cis versus trans). Box plots show medians (lines), interquartile ranges (IQRs; boxes) and 1.5 × IQRs (whiskers). Two-sided Welch’s t-test. Examined numbers are shown in parentheses.
Extended Data Fig. 7 Similarities and differences of oncogenic MMs across cancer types.
a, Correlation between somatic mutation frequency per megabase (median value shown) and fraction of MM+ samples according to MM+ oncogenes in primary samples from the total cohort. Each dot represents a cancer type, coloured by gene. b, Distribution of mutations and/or fraction of MM+ mutations for each hotspot/functional position in PDGFRA for GBLGG and GIST in primary samples from the total cohort. Asterisks indicate major positions (in which 10% or more of mutations were present in any of the recurrently mutated cancer types (defined as those with 20 or more hotspot/functional mutations)). c, Distribution of mutations and fraction of MM+ mutations for each hotspot/functional position in PIK3CA for BRCA, COADREAD and UCEC in primary samples from the total cohort. Asterisks indicate major positions. The horizontal blue dotted lines represent the mean values of major positions. d, e, Distribution of hotspot/functional mutations (d) and fraction of MM+ mutations occurring at major positions and minor positions in the ECD I–II and kinase (exon 18–21) domains (e) in EGFR for NSCLC and GBLGG in primary samples from the total cohort. b, c, e, Two-sided Fisher’s exact test. Examined numbers are shown in parentheses.
Extended Data Fig. 8 Cancer-type-specific oncogenic MMs.
a, Distribution of FGFR3 mutations in samples with single and multiple mutations in recurrently mutated cancer types (defined as those with 20 or more hotspot/functional mutations) in primary samples from the total cohort. Mutations specific to MM+ samples are indicated in red. b, Cancer-type-specific analysis of MM+ oncogenes identified by pan-cancer analysis. Shown are numbers of MM+ samples for MM+ oncogenes in the corresponding cancer type from the discovery cohort (with q < 0.01 and three or more MMs). Red and black circles indicate, respectively, observed and expected (median with 95% confidence intervals) values. One-sided permutation test (n = 10,000) with Benjamini–Hochberg correction. c, Proportion of mutations according to type, position and amino-acid change, in six cancer-type-specific MM+ oncogenes in the corresponding cancer type. Asterisk, AML, and dagger, BALL were analysed in primary samples from the total cohort and (an) independent set(s) of patients29,30,31,32,33. d, e, Distribution of hotspot/functional mutations (d) and frequency of MM+ samples (e) in NOTCH1 for TALL and CLL in primary samples from the total cohort. An independent cohort34 was also analysed for CLL. f, Significant pairwise associations (q < 0.01) with observed/expected ratios among functional domains (n = 65). Orange and blue colours depict co-occurring (observed number of MMs significantly higher than expected) and mutually exclusive (lower than expected) associations. Two-sided simulation test (n = 10,000) with Benjamini–Hochberg correction. g, Fraction of MM+ mutations for missense mutations and in-frame indels (consisting mainly of internal tandem duplications, ITDs) as well as distribution of mutations and fraction of MM+ mutations for each hotspot/functional position (the top ten are shown) in FLT3 for AML in primary samples from the total cohort and an independent set of patients. h, Distribution of mutations and fraction of MM+ mutations for each hotspot/functional position in JAK2 for BALL in primary samples from the total cohort and independent sets of patients. g, h, Asterisks indicate major positions (in which 10% or more of all mutations were present). e, g, h, Two-sided Fisher’s exact test. Examined numbers are shown in parentheses.
Extended Data Fig. 9 Comparison of oncogenic MMs between primary and metastatic samples across cancers.
a, Distribution of cancer types for metastatic samples included in the total, TCGA and TARGET, FM and GENIE cohorts. Cancer types with less than 1% frequency were combined as ‘OTHERS’ in each cohort. Cancer-type abbreviations are listed in Supplementary Table 1. b, Comparison of the frequency of MM+ samples in primary and metastatic samples across 14 MM+ oncogenes in the total cohort. c, Distribution of EGFR and KIT hotspot/functional mutations in primary and metastatic samples in NSCLC and GIST, respectively. Acquired TKI-resistant mutations are indicated in red. d, Frequency of MM+ samples for EGFR and KIT in primary and metastatic samples. e, f, Fraction of MM+ mutations for missense mutations and in-frame indels (e) and major and minor hotspots (f) in EGFR and KIT in recurrently mutated cancer types in metastatic samples. b, d–f, Two-sided Fisher’s exact test. g, Number of MMs according to mutational combinations in EGFR in recurrently mutated cancer types in metastatic samples (n = 81). Red and black circles indicate, respectively, observed and expected (median with 95% confidence intervals) values. Two-sided simulation test (n = 10,000) with Benjamini–Hochberg correction. d–g, After exclusion of acquired TKI-resistant mutations, including EGFR T790M in NSCLC and KIT V654, T670, C809, D816, D820, N822 and Y823 missense mutations in GIST. Examined numbers are shown in parentheses.
Extended Data Fig. 10 Biological relevance of PIK3CA MMs.
a, Viability of Ba/F3 cells expressing mock control or wild-type, single-mutant or double-mutant PIK3CA without WEHI-3-conditioned medium for three days (n = 3). Data represent means + s.d. Two-sided Welch’s t-test. b, Dependency of 196 CCLE cell lines on AKT1, with cell lines coloured by PIK3CA MM status, in the DepMap CRISPR–Cas9 knockout data (see Methods). Five cancer types for which there are cell lines harbouring PIK3CA MMs were analysed by two-sided Mann–Whitney U-test. c, Box plots showing sensitivity to six PI3K inhibitors (Z-scores) for 417 CCLE cell lines, according to MM status. Five cancer types for which there are cell lines harbouring PIK3CA MMs were analysed. d, Immunoblot analysis of AKT, pS473 AKT, PIK3CA and β-actin in Ba/F3 cells expressing mock control or wild-type, single-mutant or double-mutant PIK3CA in the presence of WEHI-3-conditioned medium. Representative of three independent experiments. e, Capillary-based immunoassay of AKT, pS473 AKT, mTOR, pS2448 mTOR, p70 S6K, pT389 p70 S6K, PRAS40, pT246 PRAS40, GSK-3β, pS9 GSK-3β, PIK3CA and β-actin in Ba/F3 cells expressing mock control or wild-type, single-mutant or double-mutant PIK3CA without WEHI-3-conditioned medium. Representative of two independent experiments. d, e, See Supplementary Fig. 1 for source images. f, g, Box plots showing levels of phosphorylated S473 (pS473; left) and T308 (pT308; right) in RPPA data relative to total AKT protein expression for 316 CCLE cell lines (with cancer types for which there are cell lines harbouring PIK3CA MMs) (f) and in 3,164 TCGA patients (with purity of 50% or more, in recurrently mutated cancer types) (g) according to MM status. Each dot represents a sample, coloured by cancer type. c, f, g, Single mutations were classified into missense mutations in major and minor hotspot positions, in-frame indels, and non-hotspot/functional mutations, some of which are not shown owing to small numbers. Box plots show medians (lines), interquartile ranges (IQRs; boxes) and 1.5 × IQRs (whiskers). GLM, generalized linear model (see Methods). Examined numbers are shown in parentheses.
Extended Data Fig. 11 Molecular mechanism underlying the enhanced functional activity of oncogenic MMs.
a, Fraction of PIK3CA copy-number (CN) alterations according to PIK3CA MM status for 385 UCEC patients in TCGA. b, Fraction of PIK3R1 and PTEN mutations according to PIK3CA MM status for 1,339 UCEC patients in the total cohort. a, b, Two-sided Fisher’s exact test. Single mutations were classified into missense mutations in major and minor hotspot positions, in-frame indels, and non-hotspot/functional mutations, some of which are not shown owing to small numbers. Examined numbers are shown in parentheses. c, The distance between key residues of the p110α/p85α complex16. Data represent means + s.d. (250–500 ns). d, Distance between R88 and D746 atoms. e, Distance between R38 and D743 atoms. d, e, Data represent means ± s.d. (250–500 ns). c–e, Two-sided Welch’s t-test. d, e, Different colours show independent simulations. f, Overall structure of wild-type PIK3CA (left) and a close-up view of the mean structure at the ABD–kinase interface (right). g, Mean backbone structures of R88Q, H1047R and R88Q–H1047R mutants, showing the orientation of the ABD, kinase and iSH2 (p85α) domains, superimposed on those of the wild type. Movements of the ABD and iSH2 domains induced by each mutant are indicated by arrows, along with the backbone root mean square deviation (r.m.s.d.) of the ABD (residues 16–105; magenta) and iSH2 (490–540; green) domains from the wild-type structure. c–g, Molecular-dynamics simulations for wild-type and each mutant PIK3CA (combined results from three independent 500-ns simulations are shown). h, Dependency of 60 CCLE cell lines from haematopoietic and lymphoid tissue on NOTCH1, with cell lines coloured by MM status, in the DepMap CRISPR–Cas9 knockout data. i, Luciferase assays of TP-1 reporter activity in 293T cells transfected with mock control or the indicated amounts of wild-type, single-mutant or double-mutant NOTCH1 vectors (n = 5). Data represent means + s.d. Two-sided paired t-test on log-transformed values.
Supplementary information
Supplementary Information
This file contains Supplementary Methods, Supplementary Discussion, Supplementary Table Guide, Supplementary Figure 1, Supplementary Note (acknowledgements) and Supplementary References.
Supplementary Tables
This file contains Supplementary Tables 1-17. See Supplementary Information document for a full guide.
Rights and permissions
About this article
Cite this article
Saito, Y., Koya, J., Araki, M. et al. Landscape and function of multiple mutations within individual oncogenes. Nature 582, 95–99 (2020). https://doi.org/10.1038/s41586-020-2175-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-020-2175-2
This article is cited by
-
Characterization of pancreatic cancer with ultra-low tumor mutational burden
Scientific Reports (2023)
-
Pan-cancer clinical impact of latent drivers from double mutations
Communications Biology (2023)
-
Oncogenic RAS promotes leukemic transformation of CUX1-deficient cells
Oncogene (2023)
-
Subclonal accumulation of immune escape mechanisms in microsatellite instability-high colorectal cancers
British Journal of Cancer (2023)
-
Overview and clinical significance of multiple mutations in individual genes in hepatocellular carcinoma
BMC Cancer (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
