Andrew Maltez Thomas IRCBG20081 Metagenomica e culturomica aspetti complementari della microbiologia omica Colorectal cancer metagenomics in diagnostic pratice ID: 1048188
Download Presentation The PPT/PDF document "Trieste, 11 dicembre 2020" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. Trieste, 11 dicembre 2020Andrew Maltez ThomasIRCBG_20081“Metagenomica e culturomica: aspetti complementari della "microbiologia omica"Colorectal cancer: metagenomics in diagnostic pratice
2. What is the microbiome?Berg et al. Microbiome. 20192
3. “Configurations” of our microbiome than can influence cancer in several different waysAs a causal agent in the development of tumorsEx: by producing toxins and/or metabolitesModulating our response to treatmentAffecting our response to chemotherapy and immunotherapyInfluencing cancer associated comorbiditiesEx: Cancer cachexia, obesity3What is the oncomicrobiome?
4. Use of antibiotics and cancer riskBased on:125,441 cases and 490,510 matched controlsBoursi et al . Eur J Cancer. 20164Epidemiological evidence of the oncomicrobiome
5. Experimental evidence of the oncomicrobiomeGavage of an oncogenic stool microbiomeGavage of stool from responder patients improves response to immunotherapyWong et al. Gastroenterology. 2016Routy et al. Science. 20185
6. Mechanisms of microbial carcinogenesis in colorectal cancerSchwabe RF & Jobin C. Nature Rev Cancer. 2013Microbial metabolitesGenotoxinsChanges in the microbiome6
7. The diagnostic potential of the gut microbiome in CRC7
8. The diagnostic potential of the gut microbiome in CRC8Machine learning:- High prediction of CRC using the gut microbiome- Increased sensitivity (45%) when combined with FOBTIs this diagnostic potential also seen across cohorts?Zeller et al. Mol Systems Bio. 2014
9. DatasetTotalControlsAdenomasCRCNo. of reads (×109)CountryCM_Cohort1802427298.2ItalyCM_Cohort26028-325.1ItalyZellerG_20141566142539.4FranceYuJ_201512854-747.2ChinaFengQ_20151546147468.3AustriaVogtmannE_201610452-526.9USAHanniganGD_2017822827270.5USA, CanadaValidation Cohort112565-606.8GermanyValidation Cohort28040-403.6JapanTotal96941314341356--Thomas, Manghi et al. Nat Med. 20199In collaboration with: CM_Cohort 1: Dr. Alessio Naccarati, Italian Institute for Genomic Medicine, Turin CM_Cohort2: Prof. Maria Rescigno, Humanitas Research Hospital, Milan Dr. Sara Gandini & Dr. Davide Serrano, IEO, MilanCohorts
10. AverageThomas, Manghi et al. Nat Med. 201910Can the gut microbiome detect CRC across datasets?Yes, but some datasets predict CRC better than others
11. AverageAverageThomas, Manghi et al. Nat Med. 201911Can the gut microbiome detect CRC across datasets?There is a poor transportability of one dataset to another
12. Thomas, Manghi et al. Nat Med. 201912Combining datasets improves predictionCan the gut microbiome detect CRC across datasets?
13. 13Partial overlap in samplesDifferent taxonomic profilingDifferent machine learning models (LASSO models)Wirbel et al. Nat Med 2019Combining datasets improves predictionCan the gut microbiome detect CRC across datasets?
14. 16 species achieved cross-validation AUC >0.8 for most datasets, with little improvement from using all remaining species (2% average improvement in AUC value).Thomas, Manghi et al. Nat Med. 201914Features can be minimized
15. LODO OR Fecal occult blood test (FOBOT) improves the sensitivity/specificity trade-off at high specificity levels16 features generally improved the results Thomas, Manghi et al. Nat Med. 2019Relationship to other non-invasive clinical tests
16. Combining –omics to detect CRCTarallo et al. mSystems. 2019MetagenomicsSmall bacterial RNAsMetagenomics + Small bacterial RNAs16Metagenomics + Small bacterial RNAs+ Small human RNAs
17. Fusobacterium nucleatumParvimonas micraParvimonas sppGemella morbillorumPeptostreptococcus stomatisSolobacterium mooreiClostridium symbiosumPorphyromonas asaccharolyticaAnaerococcus vaginalisPorphyromonas someraePrevotella intermediaPorphyromonas uenoisBacteroides fragilisAnaerococcus obesiensisPeptostreptococcus anaerobiusAnaerotruncus sppRuminococcus torquesGranulicatella adiancensStreptococcus thermophilusBifidobacterium catenulatumThomas, Manghi et al. Nat Med. 201917Reproducible biomarkers
18. Schmidt et al. elife. 201918Out of 125 species, 77% showed evidence of oral-fecal transmission: Streptococcus, Veillonella, Actinomyces and HaemophilusN = 470310 speciesOral fecal-transmission
19. Schmidt et al. elife. 201919Thomas, Manghi et al. Nat Med. 2019CarcinomasControlsOral fecal-transmission
20. 20Flynn et al. mSphere 2016. Oral species, biofilms and CRC formation
21. 21Willis, Gabaldon. Microorganisms. 2020The oral microbiome and disease
22. Thomas, Manghi et al. Nat Med. 2019Controls: diet associated changesCRC: Putrefaction, gluconeogenesis22Reproducible biomarkers
23. Presented by: Andrew M. Thomas – AC Camargo Cancer Center23TMA production by the gut bacteria
24. TMA production by the gut bacteria
25. Increased abundance of choline TMA-lyases in 4/7 datasets and in meta-analysis (p = 0.002, I2 = 0%, Q-test = 0.7) Thomas, Manghi et al. Nat Med. 2019Thomas, Manghi et al. Nat Med. 201925Increased abundance and transcription by qPCRMicrobial TMA producing genes and CRC
26. Odds Ratio 1.8595% CI [1.13, 3]Odds Ratio 2.1495% CI [1.29, 3.56]Odds Ratio 0.3895% CI [0.25, 0.57]Different sequence subtypes correlate with disease statusThomas, Manghi et al. Nat Med. 201926Microbial TMA producing genes and CRC
27. Validation in additional datasetsIncreased cutC abundanceHigh predictions using LOSO modelsCRC model remained highly predictive and specific (AUC ≥ 0.80).
28. In SummaryDiagnostic potential of the gut microbiome for CRC detection across cohortsCRC-specific detection of the microbiomeUse of a minimal microbial signature for CRC detectionReproducible taxonomic/functional microbial biomarkers across datasetsCholine degrading microbial enzymes may be relevant in CRC
29. The Computational Metagenomics Laboratory PI Nicola Segata@cibiocmAdrian TettFederica PintoFederica ArmaniniFrancesco AsnicarGiulia MasettiFabio CumboMireia Valles-ColomerAitor BlancoSerena ManaraPaolo GhensiMoreno ZolfoKun HuangFrancesco Beghini*Paolo ManghiNicolai KarcherLeonard DuboisEleonora NigroChiara MazzoniGianmarco PiccinnoGermana BaldiIGMAlessio NaccaritiBarbara PardiniSonia TaralloAntonio FrancavillaIEODavide SerranoSara GandiniChiara PozziUniversity of NaplesEdoardo PasolliClinica Santa RitaGaetano GalloMario TrompettoUniversity of TurinFrancesca CorderoGiulio FerreroHumanitasMaria RescignoCUNYLevi Waldronhttp://segatalab.cibio.unitn.itGeorg ZellerPeer BorkManimozhiyan ArumugamEmmanuel Dias-NetoJoão Carlos Setubal