Clinical and Public Health Microbiology: A Tsunami of Change. John Besser CDC, Enteric Diseases Laboratory Branch

Transcription

1 Clinical and Public Health Microbiology: A Tsunami of Change John Besser CDC, Enteric Diseases Laboratory Branch

2 Clinical and Public Health Microbiology: A Tsunami of Change Clinical diagnostic trends Public health challenges Public health opportunities

3

4 Rapid Tests and Single-analyte CIDTs

5 Legionella pneumophila Diagnostics Connection between environmental and human isolates Courtesy of Dr. Gayle Langley, CDC

6 Outbreaks Courtesy of Dr. Gayle Langley, CDC

7 STEC Incidence Rates by Various Case Definitions, STEC 3.00 Cases per 100, not fda approved unknown fda approved Best estimate Verified at state Culture 0.00 Cx All Cx All+verified at State Cx All+Best Estimate Cx All +FDA approved All CX+ All CIDT

8 Syndromic CIDT Panels: GI Disease BD Max BioFire FilmArray GenMark Dx esensor XT-8 Cepheid GeneXpert Nanosphere's Verigene Enteric Pathogens (EP) Test Applied Biocode Luminex xtag GPP

9 Syndromic CIDT Panels: Respiratory BD Max BioFire FilmArray GenMark Dx esensor XT-8 Cepheid GeneXpert Nanosphere's Verigene Respiratory panels (3) Luminex xtag GPP

10 Syndromic CIDT Panels: Invasive Disease BD Max BioFire FilmArray (from blood culture) Cepheid GeneXpert Enteroviral meningitis Nanosphere's Verigene (from blood culture)

11 300 Number of Publications (Multiplex + Infectious Diseases)

12 March 12, 2015

13 New Syndrome-based Culture-Independent Test Panels: Advantages Fast Efficient workflow Good performance Wider range of pathogens

14

15 Traditional Culture and Susceptibility Testing Methods are Slow to Produce Results Mixed infections report to doctor Susceptibility test (fastidious) Susceptibility test Prelim report Medical Value Culture Specimen Taken (4 days!!!) Hours Slide courtesy of Dr. Fred Tenover

16 Molecular Methods Positively Impact Patient Outcomes Specimen Taken PCR report Early results optimize antibiotic therapy Medical Value Final report to doctor for mixed infection Susceptibility test (fastidious) Susceptibility test Culture PCR testing 24 Hours Slide courtesy of Dr. Fred Tenover

17 The Cloud of Stool Testing Methods Send outs: Noro PCR O & P Bacterial isolates to State PHL C. difficile toxin PCR Specimen arrives in Lab Order entry and correction Virology shell culture Rotavirus EIA DFA confirmatory Giardia/Crypto Primary Set-Up Enteric culture inoculation Specimen aliquoting Giardia/Crypto EIA Adapted from K. Chapin, 2013 Culture desk (Salmonella, Shigella, Campylobacter, STEC, Yersinia, Vibrio)

18 Syndrome-based Stool Testing Panels Send outs: O & P Bacterial isolates to State PHL Specimen arrives in Lab Order entry and correction PCR multi-agent panel(s) Adapted from K. Chapin, 2013

19 Demise of GC Culture Fast, accurate Urine specimen, vs urethral swab Includes Chlamydia trachomatis No susceptibility data Specimen incompatible with culture Empiric treatment, reliance on guidelines

20 Direction of Clinical Diagnostic Testing for Infections of Public Health Importance Federal laboratories State/local Public Health Laboratories Clinical Laboratories Point-of-care Rare infections Exotic infections TB Diphtheria Rabies Common infections

21

22 Direction of Clinical Diagnostic Testing for Infections of Public Health Importance Federal laboratories State/local Clinical Public Health Laboratories Laboratories Home Testing Rare infections Point-of-care Exotic infections TB Diphtheria Rabies Common infections

23 N Engl J Med Jun 19;370(25):

24 Bottom Line The field of diagnostic microbiology is undergoing its most radical transformation since Pasteur

25 CIDT Challenges Patient care Accurate case counting Maintaining isolate-based surveillance

26 CIDT Challenges Patient care o Interpretation o High specificity* Accurate case counting Maintaining isolate-based surveillance * detects only what you re looking for

27

28 Multicenter Campylobacter Test Evaluation

29 CIDT Challenges Patient care o Interpretation o High specificity* Accurate case counting Maintaining isolate-based surveillance * detects only what you re looking for

30 Accurate Case Counts: Burden, Attribution, Trends

31

32 Public Health CIDT Challenges Patient care Accurate case counting Maintaining isolate-based surveillance o Strain tracking (including WGS) o Susceptibility surveillance o Virulence, vaccine efficacy * detects only what you re looking for

33

34 CIDT A Non-Issue for Event-Driven Outbreaks Best way to discover new PATHOGENS

35 White City, Jackson Co, 1982

36 Nationwide reporting began in 1912 Reported Salmonella infections in the United States, Typhoid Fever Non-typhoid Salmonellosis Incidence per 100,000 population CDC, National surveillance data National salmonella serotype surveillance Years

37 Listeriosis Surveillance System LM case Interview Cases Isolates State/Local Health Agency Case / Food questionnaire Isolates CDC EMBL (Europe) NCBI (GenBank) analysis upload WGS PFGE PulseNet WGS Case-Case Studies DDBJ (Japan) WGS PFGE International Nucleotide Sequence Database Collaboration GenomeTrakR Food / animal, environment sampling FDA USDA

38 Next era?

39 Listeria Cluster Metrics Before and After WGS Pre-WGS (Sept 2012 Aug 2013) WGS Year 1 (Sept 2013 Aug 2014) WGS Year 2 (Sept 2014 Aug 2015) No. of clusters detected N/A 6 6 No. of clusters detected sooner or only by WGS No. of outbreaks solved (food source identified) 3 Median no. of cases per cluster Note that cluster 1508MLGX6-1WGS counted as solved with 24 cases

40 Listeria and Caramel Apples 35 cases 12 states 34 hospitalizations 7 deaths

41

42 Public Health WGS Workflow LIMS Users at CDC and in the States Sequencer Raw sequences External storage NCBI, ENA, BaseSpace PH databases Genus/species Serotype Pathotype Resistance 7-gene MLST rmlst cmlst wgmlst Nomenclature server Allele databases Data pathway Proposed data pathway Analysis request Calculation engine Trimming, mapping, de novo assembly, SNP detection, allele detection

43 Point-and-Click Epidemiology* *case-case study still in development

44 Subtyping Methods: Isolate Dependency Method Isolates Required? PFGE Yes MLVA Yes WGS Yes

45 Nationwide reporting began in 1912 Reported Salmonella infections in the United States, Typhoid Fever Non-typhoid Salmonellosis Incidence per 100,000 population CDC, National surveillance data National salmonella serotype surveillance Years

46 It is ironic that new tests which are likely better for patient care could result in many thousands of additional cases per year if appropriate measures are not taken.

47 Clinical and Public Health Microbiology: A Tsunami of Change Clinical diagnostic trends Public health challenges Public health opportunities

48 Multi-analyte DNA-based Panels Polymicrobic infections New tracking possibilities Luminex xtag GPP panel

49 Multi-analyte DNA-based Panels Multiple agents New tracking possibilities Luminex xtag GPP panel

50 Multi-analyte DNA-based Panels Multiple agents New tracking possibilities Increased case reporting? Luminex xtag GPP panel

51 Strategy to Address Loss of Isolates for Public Health Activities 1. Preserve cultures Surveillance by current methods (serotyping, AST, PFGE, MLVA etc.) Device industry consult Regulatory / legal Reimbursement New isolate recovery methods Sentinel surveillance 2. Prepare for the future working on pure cultures Sequence-based infrastructure large genome databases Surveillance by whole genome sequencing (WGS) 3. Metagenomics No cultures amplicon sequencing (short-term) shotgun metagenomics (longer-term) Surveillance and diagnostics by metagenomics

52 Sentinel Susceptibility Surveillance: Gonorrhea

53 TN Burden 6/15-9/15 Isolates 81% Other 19% EIA + broth 67% Syndrome CIDT 21% HD / o/b 12% Courtesy of Dr. Amy Woron, TN Dept of Health

54 TN Burden 6/15-9/15 Isolates 81% Other 19% EIA + broth 67% Syndrome CIDT 21% HD / o/b 12% 7/1 1/31 70 from + EIA 304 Syndrome CIDT 293 Diatherix 11 Luminex Isolates 68% Other 32% Syndrome CIDT 66% HD / o/b EIA + 19% broth 15% Courtesy of Dr. Amy Woron, TN Dept of Health 54

55 Multi-State Isolate Recovery Project Goal: Provide data for isolate recovery recommendations (by APHL Food Safety CIDT Workgroup)

56 Multi-State Isolate Recovery Project Projects: Media study Specimen study (to study process and biostability) Prospective study

57 Strategy to Address Loss of Isolates for Public Health Activities 1. Preserve cultures Surveillance by current methods (serotyping, AST, PFGE, MLVA etc.) Device industry consult Regulatory / legal Reimbursement New isolate recovery methods Sentinel surveillance 2. Prepare for the future working on pure cultures Sequence-based infrastructure large genome databases Surveillance by whole genome sequencing (WGS) 3. Metagenomics No cultures amplicon sequencing (short-term) shotgun metagenomics (longer-term) Surveillance and diagnostics by metagenomics

58

59 AMD and Influenza Surveillance Expansion of 3C.2a in US and Bangladesh Global expansion of 3C.3a Tree Courtesy of U of Cambridge, UK Adapted from slide provided by John Barnes, CDC

60 Starting Material Sputum Dx Culture Subculture From Dr. James Posey, ICEID 2015

61 Foodborne Bacterial Outbreak Detection: Potential Improvements in Timeline Patient Eats Contaminated Food Stool Sample Collected Public Health Laboratory Receives Sample 1 3 days Contact with health care system: 1 5 days Diagnosis: 1 3 days Shipping: 0 7 days Serotyping & DNA fingerprinting: 2 10 days Patient Becomes Ill Salmonella Identified Case Confirmed as Part of Outbreak

62 Human Feces Science. 336: Microbial genomes o Bacteria o Viruses o Parasites o Fungi Other genomes o Human o Food Animals o Plants

63 Public Health Metagenomics : From Specimen To Answer $ MB Amplicon sequencing TB Shotgun metagenomics $$$$ Single-cell sorting and sequencing

64 CaliciNet : Culture-Independent Since 1968 CaliciNet National Norovirus Outbreak Network

65 Amplicon Sequencing Pathogen Marker Virulence Determinants AR Markers STRAINTYPE Bifidobacterium spp. Bacteroides spp. STEC (STRAIN 1) STEC (STRAIN 2) Non-pathogenic E. coli Enterobacter spp.

66 Amplicon Sequencing PHASING

67

68 O157:H7 Intact Shiga-toxin Converting Phages Lori Gladney Stx1 Phages w/stx1a,1b Stx2 Phages w/stx2a,2b Phages are genetically similar overall with some conserved flanking regions.

69 From Dr. Jo Williams Core gene alignments (no gaps). Choose a pseudorandom subset (40). Create uncorrected pairwise distance matrix. # diffs / alignment length The Wallace coefficient quantifies the likelihood that an isolate appears in the same cluster by two different methods. Provides directional information (i.e. can detect when one method is more discriminatory than the other). Cluster by furthest neighbor (complete linkage). Compare cluster assignments to all core gene classification via Adjusted Wallace Coefficient. Adjusted version takes into account that the agreement between two methods could be due to chance alone. Input Python/Bio BLAST ClustalW Mothur R Comparing Partitions Website

70 Subset of 40 core genes with perfect bidirectional cluster agreement Alignment length 35,785 bp (end gaps stripped) Furthest neighbor (complete linkage) clustering 33 clusters at uncorrected distance of (36 SNPs) Clusters of > 1 genome highlighted by line at bottom Bootstrap support from 100 replicates Note: Relative relationships of clusters differ Bootstrap values generally lower Sal_JPX_DT Sal_JPX_ Sal_JPX_ATCC_ Sal_JPX_T Sal_JPX_LT2 Sal_JPX_U288 Sal_JPX_var_5_CFSAN Sal_JPX_14028S Sal_JPX_08_ Sal_JPX_UK_ Sal_JPX_VNP20009 Sal_JPX_L_3553 Sal_JPX_D Sal_JPX_ST4_74 71 Sal_JPX_SL Sal_JPX_798 Sal_JPX_DT2 33 Sal_JKX_SPB7 23 Sal_JP6_RM Sal_ABO_ Sal_JAP_CFSAN Sal_JJP_USMARC_S3124_1 Sal_JJP_SL Sal_TDF_ Sal_JF6_CFSAN Sal_JF6_ Sal_JF6_CFSAN Sal_JF6_B182 Sal_JF6_SL476 Sal_JAG_ATCC_BAA_ Sal_PAR_RKS4594 Sal_JD6_SC_B Sal_JAB_460004_2_ Sal_JAB_ Sal_JAB_SL483 Sal_JNX_TXSC_TXSC08_19 Sal_JDG_CFSAN Sal_JKP_ATCC_ Sal_JKP_AKU_12601 Sal_JKP_CMCC_50503 Sal_JKP_CMCC_ Sal_JM6_CVM Sal_JBX_CFSAN Sal_JIX_507440_20 Sal_JRE_ATCC_35640 Sal_JGG_CFSAN Sal_JPP_Ty Sal_JPP_CT18 Sal_JPP_Ty21a Sal_JPP_P_stx_ Sal_JEG_EC Sal_JEG_EC Sal_JEG_EC Sal_JEG_EC Sal_JEG_OLF_SE4_0317_8 32 Sal_JEG_SEJ 35 Sal_JEG 44 Sal_JEG_EC Sal_JEG_P Sal_JEG_EC Sal_JEG_OLF_SE9_ Sal_JRA_RKS Sal_JRA_CDC1983_67 Sal_JRB_S Sal_JRA_287_91 Sal_JDX_CT_ Sal_JEG_EC Sal_JEG_EC Sal_JEG_EC Sal_JEG_EC Sal_JR3_62_z4_z23_RSK2980 Sal_JR3_62_z36_RKS Sal_JR5_Sbon_N268_08 Sal_JR5_NCTC_12419

71 Amplicon Sequencing: Phasing Solutions Identify pathogen-specific PCR targets. Sources of variation: o Flanked by heterogeneous regions o Presence/absence Quantitative marker linkage Sequence non-specific markers (e.g. 16S, 18S.amplified metagenomics)

72 Public Health Metagenomics: From Specimen To Answer $ MB Amplicon sequencing TB Shotgun metagenomics $$$$ Single-cell sorting and sequencing

73 Metagenomics Enteric Public Health Applications In situ pathogen characterization Pathogen discovery Food identification (in and outside the body) Root cause analysis Population biology (e.g. dysbiosis) Host factors

74 In-situ Pathogen Characterization Loman et al; JAMA Apr 10;309(14):

75 Shotgun Metagenomics SIGNAL TO NOISE PHASING

76 Human Feces Science. 336: Microbial genomes o Bacteria o Viruses o Parasites o Fungi Other genomes o Human o Food Animals o Plants

77 Metagenomics: Signal-to-Noise (just considering bacteria) 8x10 9 reads (~1Tbase) 1x10 11 organisms/ml For a positive stool specimen containing 1x10 5 CFU/mL STEC, you might expect (at best) <0.15X genome coverage from a full HiSeq* run. *For reference purposes only; does not imply endorsement. Reference: * Lepage et al. Gut 2012.

78 Clutter Mitigation Strategies NUCLEIC ACID EXTRACTION DNA RNA TNA POST-EXTRACTION AND LIBRARY CONSTRUCTION LIBRARY SEQUENCING Differential cell lysis Filtering, Concentration Separation/Pulldown Direct amplification Laser capture, microfluidics Nucleases (RNAse/DNAse) New platforms/approaches cdna conversion Multiplexing and pooling rrna depletion Bioinformatic strategies Bind/degrade CpG methylated DNA Preferential separation (mass, seq, chem) Genome/transcriptome amplification Sequencing platform selection Library method and parameters Size selection Many variables impact sequence yield, quality and bias.

79 New Long-read Sequencing Technology

80 Hi-C Crosslinking Technology

81 Metagenomics: Limiting Factors Factor Likelihood of being resolved Cost Sequencing read length (and error rate) Metagenomic-specific software, pipelines Computing processing power, bandwidth Signal to noise High High High High High

82 Salmonella Heidelberg Outbreaks, 2013 Very short incubation High morbidity/mortality K K K K K K K K K K K K K K K-1038-M K K K K K K T T T T T T-3059-A T T T T K K K K K K K AM K K K K K K FF pattern 122 FF pattern 672 State A gathering pattern 22 FF pattern 22/45 complex MO dinner pattern 80 State B funeral pattern 22 FF pattern 326 TN prison p pattern 80 FF Pattern 41/258 complex 0.02

83 Questions Can we detect Salmonella signal in a metagenomic sample? Can we distinguish the two outbreaks using metagenomic shotgun sequencing? Is another unrecognized pathogen involved?

84

85 Phylogeny of isolates and metagenomes K K K K K K State A Isolate State A Isolate State A Isolate State A metagenome Assembly State A metagenome assembly 2 State A 2013 State Isolate 4 NC K K-0574 State B Isolate 1 (Across 2,1 M sites out of 4.4 M) 5 SNPs per 100k bp State B Isolate 2 25 State B metagenomic assembly 2 0 State B metagenomic assembly 3 State B Isolate 3 State B Isolate 4 12 State B Isolate 5 1 State B Isolate 6 1 State B Isolate 7 7 State B Isolate 8 State B Isolate 9 State B Isolate 10 State B metagenomic assembly 1 State B

86

87 Clinical and Public Health Microbiology: A Tsunami of Change Jbesser@cdc.gov The findings and conclusions in this presentation are those of the author and do not necessarily represent the views of the Centers for Disease Control and Prevention