- Email: [email protected]
Improving Mutation Screening in Patients with Colorectal Cancer Predisposition Using Next-Generation Sequencing
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
The Journal of Molecular Diagnostics, Vol. -, No. -, - 2017
jmd.amjpathol.org
Improving Mutation Screening in Patients with Colorectal Cancer Predisposition Using Next-Generation Sequencing Q29
Jean-Marc Rey,* Vincent Ducros,* Pascal Pujol,y Qing Wang,z Marie-Pierre Buisine,x Hanaa Aissaoui,{ Thierry Maudelonde,*k and Sylviane Olschwang**
Q2 Q3 From the Laboratoire de Biopathologie Cellulaire et Tissulaire des Tumeurs* and the Service de Génétique Médicale,y Département d’Oncogénétique, Hôpital Q4 Q5 z
Arnaud de Villeneuve, Montpellier; the Laboratoire de Génétique Constitutionnelle des Cancers Fréquents, Center Léon Bérard, Lyon; the Laboratoire de Biochimie et Biologie Moléculaire,x Oncologie et Génétique Moléculaire, Center de Biologie Pathologie, CHU Lille; the Infinity Biomarkers, SA,{ Ecully; the Université de Montpellier: EA2415,k IURC; and the INSERM UMR_S910,** Aix-Marseille Université, RGDS Hôpital Clairval, Marseille, France Accepted for publication April 10, 2017.
Q10
Q12
Address correspondence to Jean-Marc Rey, Laboratoire de Biopathologie Cellulaire et Tissulaire des Tumeurs, Hôpital Arnaud de Villeneuve, 371 av, du Doyen Gaston Giraud, 34295 Montpellier Cedex 5, France. E-mail: [email protected].
Identification of genetic alterations is important for family risk assessment in colorectal cancers. Next-generation sequencing (NGS) technologies provide useful tools for single-nucleotide and copy number variation (CNV) identification in many genes and samples simultaneously. Herein, we present the validation of current Multiplicom MASTR designs of mismatch repair combined to familial adenomatous polyposis genes in a single PCR reamplification test for eight DNA samples simultaneously on a MiSeq apparatus. Blood samples obtained from 224 patients were analyzed. We correctly identified the 97 mutations selected among 48 samples tested in a validation cohort. PMS2 NGS analysis of the eight positive controls identified single-nucleotide variations not detected with targeted referent methods. As NGS method could not discriminate if some of them were assigned to PMS2 or pseudogenes, only CNV analysis with multiplex ligand probe-dependent amplification confirmation was retained for clinical use. Twenty-seven new variants of unknown significance, 21 disease-causing variants, and two CNVs were detected among the 176 patient samples analyzed in diagnosis routine. MUTYH disease-causing mutations were identified in two patient samples assessed for mismatch repair testing, confirming that this method facilitates accurate and rapid individual risk assessments. In one sample, the MUTYH mutation was associated with a MSH6 disease-causing mutation, suggesting that this method is helpful to identify additional cancer risk modifiers and provides a useful tool to optimize clinical issues. (J Mol Diagn 2017, -: 1e13; http://dx.doi.org/10.1016/j.jmoldx.2017.04.005)
Colorectal cancer is the third most common cancer, accounting for 10% of all cancers.1 Hereditary nonpolyposis colorectal cancer (HNPCC; alias Lynch syndrome) is an autosomal dominantly inherited disorder defined by the presence of a germline mutation in one of the mismatch repair genes (MMR; namely, MLH1, MSH2, MSH6, and PMS2).2 The complete inactivation of the corresponding MMR gene in the tumor causes a marked reduction in MMR function, which results in microsatellite instability.3 Familial adenomatous polyposis (FAP) is defined by the presence of germline mutations in adenomatous polyposis coli gene APC or in the base excision repair gene MutY homologue MUTYH.4e6 Most of APC alterations can lead to the
development of hundreds to thousands of polyps present mainly in the colon and developing during the teenage years. Untreated, the polyps progress to colorectal cancer at the average age of 39 years, with nearly 100% penetrance.7 Patients with APC mutations located in the exons 3, 4, in the alternative part of exon 9, or downstream codon 1450 in
Supported by the Center Hospitalier Universitaire of Montpellier grant and Laboratoire de Biostatistique, Epidémiologie et Santé Publique, EA2415, IURC, Montpellier, grant. Disclosures: H.A. is the Projects and Products Manager for genetic test development for Multiplicom-France. Current address of H.A., Multiplicom SAS, Lyon, France.
Copyright ª 2017 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmoldx.2017.04.005
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 Q11 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 Q6 Q7 119 120 Q8 Q9 121 122 123 124
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
Rey et al
Q13
exon 15 may develop an attenuated form of polyposis, showing <100 adenomas and later onset of the disease.8,9 Mutations in the MUTYH gene are associated with an autosomal recessive form of polyposis characterized by the presence of few colorectal adenomas and a high risk of colorectal cancer. A substantial proportion of patients (20%) with multiple adenomas like an attenuated form of polyposis harbor germline MUTYH alteration. Moreover, histological type and distribution of polyps might be not specific of Lynch syndrome or familial polyposis requiring both MMR and FAP gene analysis. Molecular genetic testing typically consists of both Sanger sequencing to identify single-nucleotide variations (SNVs) encompassing one or few nucleotides and deletions/ duplications screening. Sanger sequencing is based on the electrophoretic separation of chain-termination products produced in individual sequencing reactions. This method is the gold standard for mutation analysis in cancer diagnostics. Multiplex ligand probe-dependent amplification (MLPA; MRC Holland, Amsterdam, the Netherlands) is commonly used for detection of exon(s) deletions and duplications corresponding to copy number variations (CNVs). This method is based on specific hybridization probes composed of two adjacent oligonucleotides for each exon of target genes. Ligated probes are amplified, and fragment analyses generate ratios by comparison of peak heights of each amplified probe to those of a referent normal sample. Sanger sequencing and MLPA are low throughput and relative low sensitivity, long turnaround time, and overall expensive methods that have called for new paradigms. Next-generation sequencing (NGS) technologies provide useful tools for numerous applications by their capacity to generate several gigabases of sequence data in a single experiment and offer benefits relating to lower cost, increased workflow speed, and enhanced sensitivity in mutation detection.10 As whole genome sequencing may be unaffordable for routine diagnosis, the sequencing targeted on coding regions of selected genes is an attractive option. The use of NGS would allow the sequencing analysis of more than one gene and one sample, as well as the generation of other valuable information for diagnostic purposes as the identification of CNV. Massively parallel NGS sequencing needs target enrichment using PCRreamplification or amplification of capture products.11 Multiplicom MASTR kits (Lyon, France) are multiplex PCR-based amplification kits of genomic DNA targets used to generate libraries, followed by PCR-based incorporation of bare codes to pool libraries obtained from different DNA samples. Illumina (Paris, France) technology is based on bridge amplification generating amplified clusters to be sequenced using colored fluorescent reversible dye terminators. Sequencing by synthesis can be used to identify genetic variations as SNV and rearrangements as CNV.12 The same genomic information gained through these tests can be obtained faster and cost-effectively using NGS technology as compared to referent methods. To improve
molecular testing efficiency of colorectal cancer predispositions, we carefully selected the most frequently involved genes to define a panel and get a relevant tool for a rapid and accurate analysis of several patient samples in a single assay. Herein, we present the validation of current Multiplicom MAST designs of MMR and FAP genes using multiplex PCR, library bare coding, and sample pooling for MiSeq (Illumina) NGS analysis, followed by SeqNext (JSI Medical Systems, Ettenheim, Germany) bioinformatics analytical variables exploration in a daily diagnosis procedure.
2
jmd.amjpathol.org
Materials and Methods Patients and Samples Blood samples from 224 patients were received for MMR and/or FAP gene molecular testing. Informed consents were obtained from each patient tested. For the validation step, 48 samples were retained as positive controls. They were all previously analyzed as part of the genetic testing process. For these samples, variants of unknown significance (VUSs) and disease-causing variants were previously identified using Sanger sequencing or MLPA analysis. Among them, 32 were tested for MMR genes (MLH1, MSH2, and MSH6 for SNV screening or MLH1, MSH2, MSH6, and PMS2 for CNV screening) using HNPCC MASTR kit (runs 1 and 2) or HNPCC Plus MASTR kit combined to FAP MASTR kit (runs 3 and 4) (Figure 1). They were addressed by the ½F1 genetic counseling of the Genetic Department at Hôpital Arnaud de Villeneuve (Montpellier, France). DNA samples from eight other patients were received for APC and MUTYH testing using HNPCC Plus MASTR kit combined to FAP MASTR kit (run 5). They were sent as positive controls from the Laboratoire de Biochimie et Biologie Moléculaire, Center de Biologie-Pathologie, CHU Lille (France). DNA samples from eight other patients were Q14 received for PMS2 SNV screening using HNPCC Plus MASTR kit combined to FAP MASTR kit (run 6). They were obtained as positive controls from the Laboratoire de Génétique Constitutionnelle des Cancers Fréquents, Center Léon Bérard (Lyon, France). One hundred seventy-six samples were then analyzed in diagnosis routine using HNPCC Plus MASTR kit combined to FAP MASTR kit. The first 32 were tested using both referent and NGS methods simultaneously (runs 7 to 10). For DNA samples analyzed in the runs 11 to 28 only, VUSs and disease-causing mutations were confirmed using referent methods (Figure 1). One hundred twelve peripheral blood samples were obtained from patients with a clinical history suggestive of Lynch13,14 or polyposis syndrome.15 Two of them harbored microsatellite instable tumors. Fifty-four samples were obtained from patients with an incomplete family history and 10 from patients with sporadic colorectal cancer at a young age (<40 years old).
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
900 800 700 600 Coverage
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
Improving Mutation Screening Using NGS Q1
500 400 300 200 100 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
32 control samples for MMR SNV and CNV testing
32 samples tested with NGS and Sanger/MLPA
Figure 1 Depth of coverage comparison between overlapping (gray curve) and nonoverlapping (black curve) sequencing regions. The average of amplicon coverage was calculated after each run for one representative sample tested in the 28 runs. Runs 1 to 2 tested HNPCC MASTR Kit, generating 82 amplicons. Runs 3 to 28 tested HNPCC MASTR Plus Kit combined with FAP MASTR Kit, generating 136 amplicons.
Diagnosis routine (176) 8 control samples for PMS2 SNV testing 8 control samples for FAP SNV and CNV testing HNPCC MASTR kit
HNPCC Plus combined to FAP MASTR kits
Nucleic Acid Isolation and Quantitation Genomic DNA was isolated from 1 mL total blood using MagNA Pure Compact Nucleic Acid Isolation Kit I-Large Volume (Roche Diagnostics, Meylan, France) on a MagNA Pure Compact apparatus (Roche Diagnostics), according to the manufacturer’s protocol. Genomic DNA quantitation was made using Qubit dsDNA HS Assay KIT on a Qubit 2.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Waltham, MA), according to the manufacturer’s protocol.
Multiplex PCR-Based Library Constitution Libraries were made to analyze eight patient samples simultaneously according to the flow of patient samples received from the genetic counseling department. Target amplification of coding regions and flanking intronic sequences of selected genes were achieved using MASTR kits (Multiplicom). MMR genes were tested using HNPCC MASTR kits (Multiplicom) in five multiplex PCRs. The HNPCC MASTR MIX kit generated 82 amplicons to analyze the four MMR and exons 8 and 9 of TACSTD1 genes, excluding exons 1 of MSH2 and MSH6. The HNPCC Plus MASTR MIX kit (Multiplicom) generated 84 amplicons to analyze the whole four MMR and exons 8 and 9 of TACSTD1 genes. The whole coding regions of FAP genes (APC and MUTYH ) were amplified in three multiplex PCRs
The Journal of Molecular Diagnostics
-
with FAP MASTR MIX kit (Multiplicom). The HNPCC Plus and FAP MASTR Kits combination generated 136 amplicons and screened 36,223 bp. Briefly, multiplex PCR was performed using 50 ng of genomic DNA per reaction. For each sample tested, mixing of the five multiplex PCR products obtained from HNPCC MASTR KIT or HNPCC Plus MASTR KIT and the three multiplex PCR products obtained from the FAP MASTR KIT was performed according to a predefined mixing scheme. Amplicon libraries were purified with AMPure XP (Beckman Coulter, Brea, CA), followed by a two-step dilution in 1/1000ème to be Q15 used as the matrix for next amplification. Amplicons were tagged in a second universal PCR with the molecular identifiers for Illumina MiSeq (Multiplicom) MID 1-48. PCRs were performed with 20 mL of AR1 reagent when HNPCC MASTR MIX kit or FAP MASTR MIX kit was used or 20 mL of AR2 reagent when HNPCC Plus MASTR MIX kit was used. PCR products were purified with AMPure XP (Beckman Coulter), and quantitation was performed on a Qubit instrument with a Qubit dsDNA HS Assay kit (Invitrogen, Thermofisher Scientific), according to the manufacturer’s protocol. Final libraries were obtained by equimolary pooling barcoded PCR products before denaturation and dilution at a 2 nmol/L concentration. Libraries were combined with PhiX control before loading onto a MiSeq instrument (Illumina). NGS sequencing was performed using Nano flow cells and MiSeq Reagent Nano Kit Q16 v2 with 2 251 bp (Illumina).
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
3
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372
373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
Rey et al
Data Analysis Q17
Q18
Nucleotides were aligned using the MiSeq Reporter software v2.5.1 (Illumina) generating *.bam files. *.bam files were used to identify SNVs and CNVs with the SeqNext module of SeqPilot software v4.1.2 (JSI Medical Systems). Regions of interest and amplicon positions were defined for the SeqNext software using the Multiplicom *.manifest files for MMR and FAP gene analyses using HNPCC and FAP MASTR kits. Primers were excluded from the analyses. Nucleotides with quality scores <30 were filtered out. Settings were selected to analyze 500-bp fragments with a threshold of 20 reads per direction to validate region analyses. A low-coverage warning was defined at 100 reads to target amplicons showing low depth of coverage. A depth of at least 50 reads was defined as the cutoff value for amplicon analysis validation. Mutation sorting in distinct section was defined at 20% of variant nucleotide per direction. CNV mode analysis was also performed using the SeqNext module (JSI Medical Systems). For each plex, relative PCR depth in coverage ratio was calculated, taking into account the variations between each amplicon of a target gene and amplicons of the other genes analyzed in the same plex (control amplicons). Each sample was compared to all other samples included in the same run. As DNA libraries were performed using two different MASTR Kits, MMR and FAP CNV analyses were performed separately. Thresholds were set to 0.7 for deletions and 1.3 for duplications.
sequences) [GenBank, https://www.ncbi.nlm.nih.gov/gene; accession numbers NM_000249.3 (MLH1), NM_000251.2 (MSH2), NM_000179.2 (MSH6), NM_000535.5 (PMS2), NM_000038.5 (APC ), NM_001044817 (MUTYH )].
Criteria for Variant Classification VUSs and newly identified VUSs were classified in five categories, according to chemical, biological, in silico, and clinical features. The amino acid polarity and overall dimension changes, phylogenic conservation, SIFT score, Q20 phenotypic features as microsatellite and immunohistochemical analyses, expression studies to test splicing variants as ex vivo (minigene) and in vivo (RT-PCR) studies, functional data obtained in yeast, trans or cis position of the VUSs when a deleterious mutation is identified in the same sample, and segregation studies were taken into consideration when available. VUS classification was assessed after consensus decision of the Groupe Génétique et Cancer either class 1 (benign variant), class 2 (probably benign, 95% to 99%), class 3 (VUSs, 5% to 94.9% to be disease causing), class 4 (probably disease-causing variant, 95% to 99%), and class 5 (disease-causing variant >99% to be deleterious). At least two major criteria are needed to classify a VUS as disease causing (trans position and informative segregation study). The classification of each variant is upgraded at regular intervals in the Universal Mutation Database to have a reliable and homogeneous interpretation for diagnosis reports.
Results Variant Confirmation and Annotation
Q19
In the validation cohort (48 samples), VUSs and diseasecausing variants were previously identified by referent methods: Sanger sequencing or MLPA using SALSA MLPA P003-C1 MLH1/MSH2, P072-C1 MSH6, P008-B2 PMS2, P043-B1 APC, and P378C1 MUTYH kits (MRC Holland), according to the manufacturer protocols, then confirmed with NGS method. Exon amplification and sequencing of PMS2 were performed as previously described.15 For mutation analysis in PMS2 exon 11, forward primer in exon 10 and reverse primer in exon 11 were used.16 In the clinical cohort (176 samples), only VUSs and disease-causing variants detected in MLH1, MSH2, MSH6, APC, and MUTYH using NGS method were confirmed with Sanger method, even though CNVs detected in all MMR and FAP genes were confirmed with MLPA. For PMS2, only CNV analysis was used for diagnostic reports. Sequencing and MLPA products were analyzed on an ABI PRISM 3130xl instrument (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA). Sequence variations were annotated according to the recommendations of the Human Genome Variation Society (ie, nucleotide numbering starting at position þ1 corresponds for the A of the ATG translation initiation codon in coding cDNA reference
4
In the first validation phase of the study, we assessed the completeness of the MMR and FAP gene coding regions and control splicing sites in the design of HNPCC MASTR Plus and FAP MASTR kits as well as the ability of SeqNext software (JSI Medical Systems) to detect all kinds of SNVs and CNVs. We compared the results displayed using several bioinformatics parameters to fix the analysis settings. In the second validation step, the robustness of the workflow was evaluated in routine diagnostic conditions, and we focused on the false-positive rate of CNV detection.
Method Validation Nucleotide Analysis Runs were validated according to quality parameters. Median values across the 28 runs were 96.2% (92% to 99.5%) Q21 for Q30 value, 96.8% (92.6% to 98.3%) for the average rate Q22 Q23 of identified reads passing filter, and 481 clusters/mm2 (271 Q24 to 672 clusters/mm2) for cluster density. Lower cluster densities were linked to better Q30 and passing filter values (data not shown). There were no differences in cluster density, Q30 values, and identified reads passing filter between runs testing HNPCC MASTR Kit only or HNPCC Plus MASTR Kit combined with FAP MASTR Kit.
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Improving Mutation Screening Using NGS Amplicon Analysis Coverage analysis indicated that 100% of the MMR and FAP regions of interest were covered using the combination of HNPCC MASTR Plus and FAP MASTR kits (Supplemental Appendix S1). Regions of interest overlap the entire coding regions and few nucleotides in the 50 , 30 intronic, 50 untranslated, and 30 untranlsated regions. Intronic regions, 50 untranslated region, and 30 untranslated region were generally covered at least on 22 nucleotides (22 to 236 nucleotides can be analyzed). Five intronic regions could be analyzed on <20 nucleotides. The 50 regions of MLH1, MSH2, and APC introns 13 were covered until 9, 9, and 6 nucleotides, respectively, and the 50 intron 9 and 30 intron 14 regions of PMS2 were covered until 3 and 16 nucleotides, respectively (data not shown). Validation of each amplicon analysis was allowed with a depth of coverage of at least 50 reads. As average amplicon sizes obtained after universal PCR using HNPCC Plus MASTR Q25 Kit and FAP MASTR Kit are 469 and 498 pb, respectively, we chose to perform runs up to 2 251 sequencing cycles. To balance some low PCR efficiency and as the regions spreading over 250 pb of large amplicons were sequenced only in one direction, MASTR primer design was optimized to generate overlapping fragments leading to depth-ofcoverage variations between regions in each gene tested ½F2 (Figure 2). These variations were reproducible across all samples tested in each run (data not shown). Overall, the mean depth across all tested regions was of 225 to 557 (Figure 1). All amplicons showed a depth coverage of at least 50 reads (Figure 2).
Mutation Analysis and Variant Detection SNV Detection Eighty patient samples were entirely tested using referent and NGS methods. For proof of principle and looking for potential false negatives, 48 patient samples (runs 1 to 6) Q26 were previously tested using referent methods: Sanger sequencing for SNV analysis and MLPA (MRC Holland) for CNV analysis. Sixteen patient samples were first analyzed using HNPCC MASTR Kit running on two nano flow cells (Illumina). Thirty-two patient samples were then analyzed using HNPCC Plus MASTR Kit combined with the FAP MASTR MIX Kit (Multiplicom) on four nano flow cells (Illumina). These 48 samples were selected to detect the larger panel of SNV and CNV before starting molecular diagnostics. Seventy-nine SNVs (59 substitutions, 13 deletions, 5 duplications, and 2 deletions/insertions) spanning entire amplicons were tested with particular attention to ½T1 identify those located at the boundaries (Table 1). Twentytwo silent variants were tested at heterozygous or homozygous status. MMR mutations were selected among those identified in our laboratory and tested on four nano flow cells. One was dedicated to identify FAP gene mutations (nine SNVs and one CNV), and another one to identify nine
The Journal of Molecular Diagnostics
-
PMS2 SNVs (Table 1). We correctly identified all of the 79 known SNVs in control samples. The intronic variant affecting MSH2 intron 5 splicing junction (c.942þ3A>T) was detected using the two kits combination. It was present on 35% of 583 reads (Figure 3). ½F3 Even though this SNV was easily identifiable on SeqNext (JSI Medical Systems) electrophoregrams, it was not pointed out in the distinct section of SeqNext Sequence window. Other polynucleotide stretches in the 30 regions of MLH1 introns 11 (polyT21), MSH2 intron 1 (polyT13), and MSH6 introns 7 [polyT13 and 9 (polyT18)] were systematically focused. NGS electrophoregrams were in accord with those obtained with the Sanger method (data not shown). Eight patient samples harboring nine SNVs in PMS2 were analyzed (Table 1). All these variants were detected but present on only 23% to 26% of total reads on amplicons sequenced with a depth of coverage of 283 to 1141 reads (data not shown). Thirty-eight VUSs not previously identified with the referent method were detected. They were present on 21% to 70% of total reads and mostly spanning from exons 1 to 5, exon 9, and exons 11 to 15 (data not shown). A DNA sample showing a variant previously identified as c.2186_2187del in PMS2 exon 13 using Sanger Q27 sequencing was tested as control for the method validation phase. Using the NGS method, the variant was assigned as c.2182_2184delinsG (Figure 4). The depth of coverage was ½F4 834 reads, and this mutation was present on 26% of the 417 forward and reverse reads. A patient sample harboring a disease-causing variant in MSH2 exon 11 (c.1705_1706del) was explored. This 2-bp deletion is located on a probe hybridization site so that MLPA analysis showed a false-positive deletion of exon 11. NGS analysis identified this SNV even though CNV mode analysis was mutation free (Figure 5). ½F5 CNV Detection CNV detection was run for every sample tested. Seventeen of the 22 MMR CNVs identified in our laboratory and one APC CNV provided by the CHRU of Lille were tested as positive controls (Table 1). For samples analyzed using CNV mode, all runs included eight samples, seven as controls. It was important to test whether two samples showing CNV located in the same gene could decrease detection sensitivity when analyzed in the same run. So distribution of positive controls among the different runs was optimized to test the method sensitivity. Taken into consideration the CNV frequency in the global population, two samples harboring CNVs identified in two different genes were included in runs 1 to 3, and three samples harboring CNVs, two in MSH2 and/or TACSTD1 (exons 8 to 9), and one in other genes were included in runs 4 to 6. All positive controls were detected and correctly assigned, confirming that this method is able to detect at least two samples harboring CNVs located in the same gene. To use this NGS method in routine diagnosis, results had to be confirmed. To look for potential false negatives and/or positives, 32 patient samples were simultaneously analyzed (runs 7 to 10) using NGS and referent methods. Patient
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
5
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620
Coverage
MLH1 3000 2500 2000 1500 1000 500 0
Coverage
MSH2 3000 2500 2000 1500 1000 500 0
Coverage
MSH6 3000 2500 2000 1500 1000 500 0
Coverage
APC 3000 2500 2000 1500 1000 500 0
MUTYH
Coverage
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
Rey et al
3000 2500 2000 1500 1000 500 0
Figure 2
Depth of coverage distribution across each targeted exon obtained from a representative patient sample. Horizontal black bars represent exons with their flanking intronic regions analyzed. Vertical solid bars represent the different depth of coverage obtained from overlapping or nonoverlapping amplicon sequencing. Values >500 and >1000 reflect overlapping sequencing regions for MMR and FAP genes, respectively.
One hundred seventy-six patient samples provided by genetic counselors were tested according to routine procedure in 22 nano flow cells (runs 7 to 28). Sanger sequencing or MLPA method confirmed all variants of unknown significance and
disease-causing mutations. Two SNVs located in MLH1 intron 1 at positions c.116þ7A>T and c.116þ11A>G were present in 25% to 45% of total reads. These SNVs were recurrently identified in 29 samples analyzed in runs showing cluster density >580/mm2. As they were also not confirmed by Sanger sequencing, they were classified as false positives. In toto, during the validation phase and clinical runs, 118 CNVs observed with NGS analysis (13%) were not confirmed using MLPA and were thus labeled as false positives. These CNVs showed values slightly beyond from the normal thresholds (0.65 to 0.7 for deletions and 1.30 to 1.35 for duplications), exons deleted and/or duplicated were not
6
jmd.amjpathol.org
samples were provided by genetic counselors and results were reported accordingly. All variants (SNVs and CNVs) identified with the NGS method were confirmed by Sanger sequencing or MLPA. All samples with normal CNV profiles were checked using MLPA without any false negative.
NGS in Routine Diagnosis
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
Improving Mutation Screening Using NGS Table 1
Variants Tested during the NGS Method Validation
Gene
NV at heterozygous state
MLH1
NV at heterozygous or homozygous state c.655A>G c.1668-19A>G
MSH2-TACSTD1
c.984C>T c.1760-62G>A
c.1077-80G>A c.1077-10T>C c.1511-9A>T c.1661þ12G>A c.2006-6T>C c.1-118T>Cy c.211þ9C>Gy
MSH6
c.261-36A>G c.458-52G>T c.628-56C>T c.3557-4del c.3801þ54C>G c.4002-10del
c.116A>Gy c.186C>Ay c.260þ22C>Gy c.276A>G c.540T>C c.642C>T c.3173-101G>C c.3438þ14A>T c.3556þ15T>A c.642C>T c.3173-101G>C c.3438þ14A>T c.3556þ15T>A c.3646þ29_3646þ32del c.3647-51_3647-35del c.3802-40C>G c.3802-43del
PMS2
APC
MUTYH
c.645þ32C>T c.1458T>C c.1635G>A c.4479G>A c.5034G>A c.5268T>C c.5465T>A c.5880G>A c.64G>A c.462þ35A>G c.972G>C c.1405A>C c.1435-40C>G
VUS
DM
CNV
c.1331A>G c.1505A>G* c.1682A>G c.974C>T
c.1489dup c.1852_1853del c.1888_1892del c.1147C>T c.1277-2A>G c.689_691delinsTT c.1705_1706del c.942þ3A>T
E1-19del
c.3801þ53dup
c.3476dup c.742C>T c.1572C>G c.2804_2805del
c.1531A>G c.1789A>T c.1866G>A c.2541G>T
c.2341C>T c.1687C>T c.1746del c.2182_2184delinsG c.2341dup c.1700del c.4970dup c.2547_2550del
c.1398A>T
TACSTD1 del TACSTD1-MSH2-E1-2del TACSTD1-MSH2-E1-5del E1-2del E5-6del E8del E8dup E8-10del E11del E8-16del E9-16del E15-16del E11-16dup E5-6del
E6-11del E10del
E4del
c.494A>G c.[1145G>A(þ)1395_1397del] c.[494A>G(þ)1145G>A] c.[439G>C(þ)494A>G]
CNV, copy number variation; DM, disease-causing mutation; NGS, next-generation sequencing; NV, neutral variant; VUS, variant of unknown significance. *VUS not described in the Universal Mutation Database found associated with the biallelic MUTYH inactivation c.[1145G>A(þ)1395_1397del]. y Variants only detected with HNPCC Plus MASTR Kit.
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
7
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
Rey et al
print & web 4C=FPO
869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 Figure 3 c.942þ3A>T in MSH2 intron 5 detected in one control sample and two patient samples with diagnostic report. 892 893 894 adjacent, and often showed exon duplications and deletions in and MLPA approaches, based on individual testing, to 895 the same gene probably reflecting bad DNA quality. Two of NGS-based analysis. In this study, we validated a method to 896 them showing either one exon deletion or duplication were test a targeted panel of six genes predisposing to familial 897 tested after a new DNA extraction and showed a normal CNV colorectal cancer and Lynch syndrome using a combination 898 profile (data not shown). of two libraries obtained from Multiplicom MASTR kits. 899 Twenty-seven new variants of unknown significance and NGS analysis was performed on a MiSeq apparatus (Illu900 23 disease-causing or probably disease-causing variants (21 mina) and *.bam files were obtained using MiSeq reporter 901 SNVs and 2 CNVs) were identified. The disease-causing (Illumina) processing feature of computationally pipeline 902 903 ½T2 mutation detection rate was 13.1% (Table 2). The MSH2 conversion of images to sequence reads (base calling). 904 intronic variant c.942þ3A>T was identified in two patient Alignment and assembly of data were optimized using the 905 samples. It was detected on 25% and 27% of 221 and 274 SeqNext software (JSI Medical Systems) to identify SNV 906 reads, respectively, and was unambiguous on SeqNext elecand CNV using the same NGS data. Our current study 907 trophoregrams. As the control sample tested in the validation confirmed NGS using Multiplicom Kits, MiSeq (Illumina) 908 phase, this SNV was not pointed out in the distinct section of technology, and SeqNext software analysis as a reliable 909 the Sequence SeqNext window (Figure 3). method to detect mutations with good sensitivity and 910 Not surprisingly, we identified SNVs considered as disspecificity. A few points remain to be improved. 911 ease causing in the MUTYH coding sequence in patient The detection accuracy of Lynch syndromeeassociated 912 samples assessed by genetic counselors for MMR testing SNVs in PMS2 is complicated with the presence of pseu913 only. Two mutations were identified in a patient sample as dogenes. In the current study, all heterozygous SNVs pre914 915 c.[1105del(þ)1145G>A]. Located on two different alleles, viously identified16,17 in the eight control samples were 916 these mutations led to a biallelic inactivation of the MUTYH present only at low percentage of total reads. SNVs not 917 ½F6 gene, where no MMR alteration was detected (Figure 6). A previously detected with referent methods were observed in 918 the same samples. These SNVs were mostly detected in disease-causing mutation was identified in MSH6 at position 919 exons showing sequence similarities with pseudogenes and c.3261dup (p.Phe1088Leufs*5) in another sample. Per920 present in 21% to 70% of total reads. This NGS technology forming NGS sequencing, analysis detected a monoallelic 921 is sensitive enough to detect PMS2 mutations with a diluted mutation of MUTYH (c.494A>G) (data not shown). 922 signal in mixed homologous sequences but cannot 923 discriminate if some of them are part of PMS2 or pseudo924 genes. The identification of SNVs located only on pseudoDiscussion 925 genes is probably reflecting sequence similarities 926 927 coamplification. Although further optimizations are needed An increasing number of genes involved in cancer pre928 dispositions and patient samples provided by genetic to improve PMS2 analysis with this method to provide 929 counseling clearly require a shift from Sanger sequencing clinically relevant results like primer design might be 930
8
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
Q30
Q28
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
print & web 4C=FPO
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
Improving Mutation Screening Using NGS
Figure 4
Mutation identified in a control sample in PMS2 exon 13 as c.2182_2184delinsG using NGS method (A) or c. 2186_2187del using Sanger method (B). The insG signal diluted in the background is pointed out by an arrow.
improved to upgrade PMS2 specific amplification and reduce pseudogene selection, an upgraded software version able to detect haplotypes with many rare SNVs in several samples analyzed in the same run might help to define pseudogene sequences to avoid and select true PMS2 SNVs. Despite these improvements, this method lacks accuracy in assigning SNVs to PMS2. Unequivocal differentiation of PMS2 from its pseudogenes, especially PMS2CL, which shares 98% sequence with the 30 region or the active gene, should use MLPA and long-range PCR/NGS and Sanger combined analysis, as described by Li et al.18 One control sample was tested for a disease-causing mutation located in PMS2 exon 13. The mutation was identified as c.2186_2187del or c.2182_2184delinsG using Sanger or NGS approaches, respectively. Accurate analysis of Sanger results revealed A/G nucleotide superposition with a diluted G signal at position c.2182. Taking this observation into consideration, the light G signal was concordant with NGS assignment: the background observed using Sanger method
The Journal of Molecular Diagnostics
-
herein evidences that complex SNVs as delins and nucleotide variations generating low signal are more accurately assigned using the NGS method as compared to the conventional sequencing method. HNPCC Plus MASTR Kit was developed to improve the sensitivity of SNV detection, particularly those located in GC-rich regions or in nucleotide stretches. SNVs located in GC-rich MSH2 and MSH6 exons 1 were easily identified. The variant at position c.942þ3A>T in intron 5 of MSH2, being the first nucleotide of a 27-A stretch, is known to be critical to detect using the referent sequencing method. In our study, it was detected in one sample tested to validate the method, then in two patient samples assessed by the genetic counseling. Even though this variant was easily identified on SeqNext electrophoregrams, it was not reported in the distinct mutation sorting section of the software, implying systematic focusing on this MSH2 region to complete sample screening for SNV mode analysis.
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
9
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
print & web 4C=FPO
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
Rey et al
Figure 5 False deletion of MSH2 exon 11 using the MLPA method. In the upper part of the figure, MLPA analysis shows exon 11 deletion (red arrow) and Sanger analysis identifies a 2-bp deletion on MLPA probe hybridization site. In the lower part of the figure, a single run analysis identifies the point mutation, whereas CNV mode analysis shows a normal profile.
All colorectal cancer syndromes caused by known highpenetrance gene alterations collectively account for 2% to 6% of all colorectal cancers. We identified 23 deleterious mutations among the 176 patient samples tested, leading to a 13.1% detection rate, rendering this new method at least as sensitive as the referent one. Genomic rearrangements (CNVs) account for 5%, 7%, 20%, 20%, and 100% of mutations in MLH1, MSH6, MSH2, PMS2, and TACSTD1, respectively.19e21 Identification of CNVs is part of colorectal cancer genetic testing. Two DNA samples obtained from disease-free members of the same family were assessed by the genetic counseling to test MMR and FAP genes as probands. Heterozygous MSH2 deletion of exons 9 to 16 was identified in both samples, leading to CNVs’ rate detection close to that of the general population.22 These samples were obtained from a mother and her daughter, confirming the robustness of this approach and showing that this method is able to detect a same CNV present in two different samples analyzed in a single run; nevertheless, in routine diagnostics, it would be advisable to test samples obtained from more than two members of the same family on separate runs. For diagnostic testing, wherein the CNV status of samples was a priori not known, if more than two CNVs located in the same gene were detected in samples analyzed in a single run (when a common
CNV founder mutation is shared by more than two patients), CNV analysis might be performed, excluding one or two samples showing abnormal values. To upgrade the method sensitivity, we have tested and validated this technique to analyze 16 samples in a single run (data not shown). Increasing batch size has permitted us to mitigate the risk that true-positive samples might be not detected if more than two samples harboring the same CNV are analyzed in one run. A patient sample showing a mutation at nucleotides c.1705_1706del on MSH2 was tested. The two-step process analysis using referent methods generated a deletion of exon 11 using MLPA, whereas Sanger analysis revealed a 2-bp deletion located on the MLPA hybridization site. The NGS method merges SNV and CNV mode analysis in the same test and generates information of diagnostic relevance available in a one-step process. A current challenge for molecular testing colorectal cancers is to identify the genetic cause when the colorectal cancer syndrome is associated with low-penetrance alleles, leading to clinical heterogeneity and gene-overlapping syndromes. In some cases, clinical characterization cannot distinguish between Lynch syndrome and attenuated adenomatous polyposis. One of 176 patient samples was first assessed for MMR screening. This patient developed an
10
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302
Improving Mutation Screening Using NGS Table 2
VUSs and Disease-Causing Mutations Identified Using NGS Method in 176 Proband Samples Analyzed with Diagnostic Reports
Gene
Newly identified VUS
Disease-causing mutations
MLH1
c.116þ132G>A c.1039-31A>T
MSH2-TACSTD1
c.645þ175C>G c.1759þ57G>T c.2281G>C c.2281A>G (two probands) c.3467T>C
c.676C>T c.731G>A c.767del c.1178T>C c.1943C>T* c.942þ3A>T (two probands) c.2131C>T
MSH6
APC
MUTYH
c.136-53T>C c.388A>G c.645þ33G>A c.645þ129A>C c.730-160A>G c.1585C>T c.3270A>G c.5009C>T c.5140G>A c.5670A>G c.7481C>T c.7888G>A c.270C>T c.306þ33G>A c.565C>T c.883C>T c.891þ24G>A c.1398A>T c.1476þ73C>T
CNV
E 9-16 dely
c.5C>A c.1344dup c.3261dup c.3477C>A c.3775_3776del c.4001G>A c.694C>T
c.494A>Gz c.1105del c.1145G>Ay
CNV, copy number variation; NGS, next-generation sequencing; VUS, variant of unknown significance. *Variant class 4 in the Universal Mutation Database. y Mutations identified in two different samples. z Mutation identified in three different samples.
adenocarcinoma of the ascending colon at the age of 45 years and two third-degree relatives also exhibited colorectal cancers at the ages of 71 and 75 years. He carried a biallelic germline mutation of MUTYH. Screening of MMR and FAP genes simultaneously facilitates accurate and rapid individual risk assessments, enabling personalized family surveillance and preventive measures. The identification of individuals at high risk of cancer allows prevention and/or early cancer detection, resulting in decreasing disease-specific mortality. The heritable nature of colorectal cancer might be associated with the coheritance of multiple variants. The risk associated with each variant might not be individually systematically high, but the combined effect could significantly contribute to disease burden.23 In this study, we identified two heterozygous causal SNVs assigned as c.3261dup and c.494A>G on MSH6 and MUTYH, respectively, in a patient sample assessed for MMR testing. This patient developed endometrium and breast malignancies at the
The Journal of Molecular Diagnostics
-
ages of 66 and 44 years, respectively. The family history showed endometrium cancers developed in mother, aunt, cousin, and grand-mother, plus three colorectal, one pancreatic, and one brain carcinomas, all present in the maternal branch of the pedigree. The risk of developing malignancies associated with this MSH6 mutation might be increased in the presence of the MUTYH mutation, suggesting that additional lowpenetrance alleles may explain the increased risk of cancer in families with mutations associated with Lynch syndrome. This NGS method is helpful to identify additional cancer risk modifiers in the same analysis and provides a useful tool to optimize clinical issues as preventive measures and genetic counseling. Faithful DNA replication is essential to maintain genomic stability and to prevent mutagenesis and tumor development. Eukaryotic DNA replication accuracy results from a combination of the high fidelity of DNA polymerases and post-replication surveillance by the MMR apparatus.24 Recently, it has been shown that deficient proofreading
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
11
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426
Rey et al
I
1
2
II
4
3
5
IV
7
2
1
III
6
1
2
1
2
Figure 6 Pedigree of a patient assessed for MMR gene screening by the genetic counseling and showing MUTYH mutations (proband III-1, pointed out by an arrow). Siblings I-1 and I-6 developed colon cancer.
activity of DNA polymerases d or ε is associated with cancer progression.25 Germline mutations of POLD1 and POLE encoding the 30 to 50 exonuclease domain components of DNA polymerase d and ε, respectively, have been associated with colorectal and/or endometrial cancer predisposition in families without MMR defect,26,27 and POLE mutations have been identified in patients with adenomatous polyposis and early-onset colorectal cancer.28 As POLD1 or POLE genetic alterations contribute to Lynch-like or FAP syndromes, respectively, molecular testing of these two genes might be included for diagnostic purposes. Other genes already included in large panels to test familial colon cancer predisposition, such as PTEN, SMAD4, STK11, or BMPR1A, have been excluded from ours because their alteration usually contributes to confirm a clinical diagnosis and they are useful for genetic counseling only. In our study, NGS screening using the six-genes panel allowed a time and cost reduction as compared to the referent methods and facilitates the identification of gene mutation carriers by laboratories checking many samples to be rapidly analyzed. This method might improve the effectiveness of preventive measures.
Acknowledgments We thank Isabelle Coupier, Carole Corsini, and Jacqueline Duffour for providing patient samples and Marie-Christine Lecque for preparing genomic DNA.
Supplemental Data Supplemental material for this article can be found at http://dx.doi.org/10.1016/j.jmoldx.2017.04.005.
References 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM: Estimates of worldwide burden of cancer in 2008: GLOBO-CAN 2008. Int J Cancer 2008, 127:2893e2917
12
2. Boland CR: Evolution of the nomenclature for the hereditary colorectal cancer syndromes. Fam Cancer 2005, 4:211e218 3. Aaltonen LA, Peltomäki P, Mecklin JP, Järvinen H, Jass JR, Green JS, Lynch HT, Watson P, Tallqvist G, Juhola M: Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res 1994, 54:1645e1648 4. Sieber OM, Tomlinson IP, Lamlum H: The adenomatous polyposis coli (APC) tumor suppressor genetics, function and disease. Mol Med Today 2000, 6:462e469 5. Jones S, Emmerson P, Maynard J, Best JM, Jordan S, Williams GT, Sampson JR, Cheadle JP: Biallelic germile mutations in MYH predispose to multiple colorectal adenoma and somatic G:C? T:A mutations. Hum Mol Genet 2002, 11:2961e2967 6. Sieber OM, Lipton L, Crabtree M, Heinimann K, Fiadalgo P, Philips RK, Bisgaard ML, Orintof TM, Aaltonin LA, Hodgeson SV, Thomas HJ, Tomlinson IP: Multiple colorectal adenomas, classic adenomatous polyposis, and germline mutations in MYH. N Engl J Med 2003, 348:791e799 7. Bisgaard ML, Fenger K, Bulow S, Neiburh E, Mohr J: Familial adenomatous polyposis (FAP): frequency, penetrance, and mutation rate. Hum Mutat 1994, 3:121e125 8. Soravia C, Berk T, Madlensky L, Mitri A, Cheng H, Gallenger S, Cohen Z, Bapat B: Genotype-phenotype correlations in attenuated adenomatous polyposis coli. Am J Hum Genet 1998, 62:1290e1301 9. Knudsen AL, Bisgaard ML, Bülow S: Attenuated familial adenomatous polyposis (AFAP): a review of the literature. Fam Cancer 2003, 2:43e55 10. Sabatini LM, Mathews C, Ptak D, Doshi S, Tynan K, Hegde MR, Burke TL, Bossler AD: Genomic sequencing procedure microcosting analysis and health economic cost-impact analysis. J Mol Diagn 2016, 18:319e328 11. Pritchard CC, Smith C, Salipante SJ, Lee MK, Thornton AM, Nord AS, Gulden C, Kupfer SS, Swisher EM, Bennett RL, Novetsky AP, Jarvik GP, Olopade OI, Goodfellow PJ, King MC, Tait JF, Walsh T: ColoSeq provides comprehensive lynch and polyposis syndrome mutational analysis using massively parallel sequencing. J Mol Diagn 2012, 14:357e366 12. Campbell PJ, Stephens PJ, Pleasance ED, O’Meara S, Li H, Santarius T: Identification of somatically acquired rearrangements in cancer using genome-wide massively parallel paired-end sequencing. Nat Genet 2008, 40:722e729 13. Vasen HF, Watson P, Mecklin JP, Lynch HT: New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on HNPCC. Gastroenterology 1999, 116:1453e1456 14. Piñol V, Castells A, Andreu M, Castellví-Bel S, Alenda C, Llor X, Xicola RM, Rodríguez-Moranta F, Payá A, Jover R, Bessa X; Gastrointestinal Oncology Group of the Spanish Gastroenterological Association: Accuracy of revised Bethesda guidelines, microsatellite instability, and immunohistochemistry for the identification of patients with hereditary nonpolyposis colorectal cancer. JAMA 2005, 293:1986e1994 15. Luk GD: Diagnosis and therapy of hereditary polyposis syndromes. Gastroenterologist 1995, 3:153e167 16. Nakagawa H, Lockman JC, Frankel WL, Hampel H, Steenblock K, Burgart LJ, Thibodeau SN, de la Chapelle A: Mismatch repair gene PMS2: disease-causing germline mutations are frequent in patients whose tumors stain negative for PMS2 protein, but paralogous genes obscure mutation detection and interpretation. Cancer Res 2004, 64: 4721e4727 17. Auclair J, Leroux D, Desseigne F, Lasset C, Saurin JC, Joly MO, Pinson S, Xu XL, Montmain G, Ruano E, Navarro C, Puisieux A, Wang Q: Novel biallelic mutations in MSH6 and PMS2 genes: gene conversion as a likely cause of PMS2 gene inactivation. Hum Mutat 2007, 28:1084e1090 18. Li J, Dai H, Feng Y, Tang J, Chen S, Tian X, Gorman E, Schmitt ER, Hansen TAA, Wang J, Plon SE, Zhang VW, Wong LJC: A comprehensive strategy for accurate mutation detection of the highly homologous PMS2. J Mol Diagn 2015, 17:545e553
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488
1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
Improving Mutation Screening Using NGS 19. Wijnen J, van der Klift H, Vasen H, Khan PM, Menko F, Tops C, Meijers Heijboer H, Lindhout D, Moller P, Fodde R: MSH2 genomic deletions are a frequent cause of HNPCC. Nat Genet 1998, 20:326e328 20. Plaschke J, Krüger S, Dietmaier W, Gebert J, Sutter C, Mangold E, Pagenstecher C, Holinski-Feder E, Schulmann K, Möslein G, Rüschoff J, Engel C, Evans G, Schackert HK: Eight novel MSH6 germile mutations in patient with familial and nonfamilial colorectal cancer selected by loss of protein expression in tumor tissue. Hum Mutat 2004, 23:285 21. Senter L, Clendenning M, Sotamaa K, Hamplel H, Green J, Potter JD, Lindblom A, Lagerstedt K, Thibondeau SN, Lindor MN, Young J, Winship I, Dowty JG, White DM, Hopper JL, Baglietto L, Jenkins MA, de a Chapelle A: The clinical phenotype of Lynch syndrome due to germline PMS2 mutations. Gastroenterology 2008, 135:419e428 22. Wang Y, Friedl C, Lambert C, Jungck M, Mathiak M, Pagenstecker C, Propping P, Mangold E: Hereditary non polyposis colorectal cancer: frequent occurrence of large genomic deletions in MSH2 and MLH1 genes. Int J Cancer 2003, 103:636e641 23. Theodoratou E, Montazeri Z, Hawken S, Allum GC, Gong J, Tait V, Kirac I, Tazari M, Farrington SM, Demarsh A, Zgaga L, Landry D,
The Journal of Molecular Diagnostics
-
24. 25. 26.
27.
28.
Benson HE, Read SH, Rudan I, Tenesa A, Dunlop MG, Campbell H, Little J: Systematic meta-analysis and field synopsis of genetic association studies in colorectal cancer. J Natl Cancer Inst 2012, 104: 1433e1457 Kunkel TA: DNA replication fidelity. J Biol Chem 2004, 279: 16895e16898 Lange S, Takata K, Wood R: DNA polymerases and cancer. Nat Rev Cancer 2011, 11:96e110 Church DN, Briggs SEW, Palles C, Domingo E, Kearsey SJ, Grimes JM, Gorman M, Martin L, Howarth KM, Hodgson SV; NSECG Collaborators, Kaur K, Taylor J, Tomlinso PM: DNA polymerase e and d exonuclease domain mutations in endometrial cancer. Hum Mol Genet 2013, 22:2820e2828 Palles C, Cazier JB, Howarth KM, Doming E, Jones AM, Broderick P, et al: Germline mutations affecting the proof-reading domains of POLE and POLD1 predispose to colorectal adenomas and carcinomas. Nat Genet 2013, 45:136e144 Briggs S, Tomlinson I: Germline and somatic polymerase epsilon and delta mutations define a new class of hypermutated colorectal and endometrial cancers. J Pathol 2013, 230:148e153
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:46 am EO: 2017_25
13
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528
The Journal of Molecular Diagnostics, Vol. -, No. -, - 2017
jmd.amjpathol.org
Improving Mutation Screening in Patients with Colorectal Cancer Predisposition Using Next-Generation Sequencing Q29
Jean-Marc Rey,* Vincent Ducros,* Pascal Pujol,y Qing Wang,z Marie-Pierre Buisine,x Hanaa Aissaoui,{ Thierry Maudelonde,*k and Sylviane Olschwang**
Q2 Q3 From the Laboratoire de Biopathologie Cellulaire et Tissulaire des Tumeurs* and the Service de Génétique Médicale,y Département d’Oncogénétique, Hôpital Q4 Q5 z
Arnaud de Villeneuve, Montpellier; the Laboratoire de Génétique Constitutionnelle des Cancers Fréquents, Center Léon Bérard, Lyon; the Laboratoire de Biochimie et Biologie Moléculaire,x Oncologie et Génétique Moléculaire, Center de Biologie Pathologie, CHU Lille; the Infinity Biomarkers, SA,{ Ecully; the Université de Montpellier: EA2415,k IURC; and the INSERM UMR_S910,** Aix-Marseille Université, RGDS Hôpital Clairval, Marseille, France Accepted for publication April 10, 2017.
Q10
Q12
Address correspondence to Jean-Marc Rey, Laboratoire de Biopathologie Cellulaire et Tissulaire des Tumeurs, Hôpital Arnaud de Villeneuve, 371 av, du Doyen Gaston Giraud, 34295 Montpellier Cedex 5, France. E-mail: [email protected].
Identification of genetic alterations is important for family risk assessment in colorectal cancers. Next-generation sequencing (NGS) technologies provide useful tools for single-nucleotide and copy number variation (CNV) identification in many genes and samples simultaneously. Herein, we present the validation of current Multiplicom MASTR designs of mismatch repair combined to familial adenomatous polyposis genes in a single PCR reamplification test for eight DNA samples simultaneously on a MiSeq apparatus. Blood samples obtained from 224 patients were analyzed. We correctly identified the 97 mutations selected among 48 samples tested in a validation cohort. PMS2 NGS analysis of the eight positive controls identified single-nucleotide variations not detected with targeted referent methods. As NGS method could not discriminate if some of them were assigned to PMS2 or pseudogenes, only CNV analysis with multiplex ligand probe-dependent amplification confirmation was retained for clinical use. Twenty-seven new variants of unknown significance, 21 disease-causing variants, and two CNVs were detected among the 176 patient samples analyzed in diagnosis routine. MUTYH disease-causing mutations were identified in two patient samples assessed for mismatch repair testing, confirming that this method facilitates accurate and rapid individual risk assessments. In one sample, the MUTYH mutation was associated with a MSH6 disease-causing mutation, suggesting that this method is helpful to identify additional cancer risk modifiers and provides a useful tool to optimize clinical issues. (J Mol Diagn 2017, -: 1e13; http://dx.doi.org/10.1016/j.jmoldx.2017.04.005)
Colorectal cancer is the third most common cancer, accounting for 10% of all cancers.1 Hereditary nonpolyposis colorectal cancer (HNPCC; alias Lynch syndrome) is an autosomal dominantly inherited disorder defined by the presence of a germline mutation in one of the mismatch repair genes (MMR; namely, MLH1, MSH2, MSH6, and PMS2).2 The complete inactivation of the corresponding MMR gene in the tumor causes a marked reduction in MMR function, which results in microsatellite instability.3 Familial adenomatous polyposis (FAP) is defined by the presence of germline mutations in adenomatous polyposis coli gene APC or in the base excision repair gene MutY homologue MUTYH.4e6 Most of APC alterations can lead to the
development of hundreds to thousands of polyps present mainly in the colon and developing during the teenage years. Untreated, the polyps progress to colorectal cancer at the average age of 39 years, with nearly 100% penetrance.7 Patients with APC mutations located in the exons 3, 4, in the alternative part of exon 9, or downstream codon 1450 in
Supported by the Center Hospitalier Universitaire of Montpellier grant and Laboratoire de Biostatistique, Epidémiologie et Santé Publique, EA2415, IURC, Montpellier, grant. Disclosures: H.A. is the Projects and Products Manager for genetic test development for Multiplicom-France. Current address of H.A., Multiplicom SAS, Lyon, France.
Copyright ª 2017 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmoldx.2017.04.005
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 Q11 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 Q6 Q7 119 120 Q8 Q9 121 122 123 124
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
Rey et al
Q13
exon 15 may develop an attenuated form of polyposis, showing <100 adenomas and later onset of the disease.8,9 Mutations in the MUTYH gene are associated with an autosomal recessive form of polyposis characterized by the presence of few colorectal adenomas and a high risk of colorectal cancer. A substantial proportion of patients (20%) with multiple adenomas like an attenuated form of polyposis harbor germline MUTYH alteration. Moreover, histological type and distribution of polyps might be not specific of Lynch syndrome or familial polyposis requiring both MMR and FAP gene analysis. Molecular genetic testing typically consists of both Sanger sequencing to identify single-nucleotide variations (SNVs) encompassing one or few nucleotides and deletions/ duplications screening. Sanger sequencing is based on the electrophoretic separation of chain-termination products produced in individual sequencing reactions. This method is the gold standard for mutation analysis in cancer diagnostics. Multiplex ligand probe-dependent amplification (MLPA; MRC Holland, Amsterdam, the Netherlands) is commonly used for detection of exon(s) deletions and duplications corresponding to copy number variations (CNVs). This method is based on specific hybridization probes composed of two adjacent oligonucleotides for each exon of target genes. Ligated probes are amplified, and fragment analyses generate ratios by comparison of peak heights of each amplified probe to those of a referent normal sample. Sanger sequencing and MLPA are low throughput and relative low sensitivity, long turnaround time, and overall expensive methods that have called for new paradigms. Next-generation sequencing (NGS) technologies provide useful tools for numerous applications by their capacity to generate several gigabases of sequence data in a single experiment and offer benefits relating to lower cost, increased workflow speed, and enhanced sensitivity in mutation detection.10 As whole genome sequencing may be unaffordable for routine diagnosis, the sequencing targeted on coding regions of selected genes is an attractive option. The use of NGS would allow the sequencing analysis of more than one gene and one sample, as well as the generation of other valuable information for diagnostic purposes as the identification of CNV. Massively parallel NGS sequencing needs target enrichment using PCRreamplification or amplification of capture products.11 Multiplicom MASTR kits (Lyon, France) are multiplex PCR-based amplification kits of genomic DNA targets used to generate libraries, followed by PCR-based incorporation of bare codes to pool libraries obtained from different DNA samples. Illumina (Paris, France) technology is based on bridge amplification generating amplified clusters to be sequenced using colored fluorescent reversible dye terminators. Sequencing by synthesis can be used to identify genetic variations as SNV and rearrangements as CNV.12 The same genomic information gained through these tests can be obtained faster and cost-effectively using NGS technology as compared to referent methods. To improve
molecular testing efficiency of colorectal cancer predispositions, we carefully selected the most frequently involved genes to define a panel and get a relevant tool for a rapid and accurate analysis of several patient samples in a single assay. Herein, we present the validation of current Multiplicom MAST designs of MMR and FAP genes using multiplex PCR, library bare coding, and sample pooling for MiSeq (Illumina) NGS analysis, followed by SeqNext (JSI Medical Systems, Ettenheim, Germany) bioinformatics analytical variables exploration in a daily diagnosis procedure.
2
jmd.amjpathol.org
Materials and Methods Patients and Samples Blood samples from 224 patients were received for MMR and/or FAP gene molecular testing. Informed consents were obtained from each patient tested. For the validation step, 48 samples were retained as positive controls. They were all previously analyzed as part of the genetic testing process. For these samples, variants of unknown significance (VUSs) and disease-causing variants were previously identified using Sanger sequencing or MLPA analysis. Among them, 32 were tested for MMR genes (MLH1, MSH2, and MSH6 for SNV screening or MLH1, MSH2, MSH6, and PMS2 for CNV screening) using HNPCC MASTR kit (runs 1 and 2) or HNPCC Plus MASTR kit combined to FAP MASTR kit (runs 3 and 4) (Figure 1). They were addressed by the ½F1 genetic counseling of the Genetic Department at Hôpital Arnaud de Villeneuve (Montpellier, France). DNA samples from eight other patients were received for APC and MUTYH testing using HNPCC Plus MASTR kit combined to FAP MASTR kit (run 5). They were sent as positive controls from the Laboratoire de Biochimie et Biologie Moléculaire, Center de Biologie-Pathologie, CHU Lille (France). DNA samples from eight other patients were Q14 received for PMS2 SNV screening using HNPCC Plus MASTR kit combined to FAP MASTR kit (run 6). They were obtained as positive controls from the Laboratoire de Génétique Constitutionnelle des Cancers Fréquents, Center Léon Bérard (Lyon, France). One hundred seventy-six samples were then analyzed in diagnosis routine using HNPCC Plus MASTR kit combined to FAP MASTR kit. The first 32 were tested using both referent and NGS methods simultaneously (runs 7 to 10). For DNA samples analyzed in the runs 11 to 28 only, VUSs and disease-causing mutations were confirmed using referent methods (Figure 1). One hundred twelve peripheral blood samples were obtained from patients with a clinical history suggestive of Lynch13,14 or polyposis syndrome.15 Two of them harbored microsatellite instable tumors. Fifty-four samples were obtained from patients with an incomplete family history and 10 from patients with sporadic colorectal cancer at a young age (<40 years old).
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
900 800 700 600 Coverage
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
Improving Mutation Screening Using NGS Q1
500 400 300 200 100 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
32 control samples for MMR SNV and CNV testing
32 samples tested with NGS and Sanger/MLPA
Figure 1 Depth of coverage comparison between overlapping (gray curve) and nonoverlapping (black curve) sequencing regions. The average of amplicon coverage was calculated after each run for one representative sample tested in the 28 runs. Runs 1 to 2 tested HNPCC MASTR Kit, generating 82 amplicons. Runs 3 to 28 tested HNPCC MASTR Plus Kit combined with FAP MASTR Kit, generating 136 amplicons.
Diagnosis routine (176) 8 control samples for PMS2 SNV testing 8 control samples for FAP SNV and CNV testing HNPCC MASTR kit
HNPCC Plus combined to FAP MASTR kits
Nucleic Acid Isolation and Quantitation Genomic DNA was isolated from 1 mL total blood using MagNA Pure Compact Nucleic Acid Isolation Kit I-Large Volume (Roche Diagnostics, Meylan, France) on a MagNA Pure Compact apparatus (Roche Diagnostics), according to the manufacturer’s protocol. Genomic DNA quantitation was made using Qubit dsDNA HS Assay KIT on a Qubit 2.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Waltham, MA), according to the manufacturer’s protocol.
Multiplex PCR-Based Library Constitution Libraries were made to analyze eight patient samples simultaneously according to the flow of patient samples received from the genetic counseling department. Target amplification of coding regions and flanking intronic sequences of selected genes were achieved using MASTR kits (Multiplicom). MMR genes were tested using HNPCC MASTR kits (Multiplicom) in five multiplex PCRs. The HNPCC MASTR MIX kit generated 82 amplicons to analyze the four MMR and exons 8 and 9 of TACSTD1 genes, excluding exons 1 of MSH2 and MSH6. The HNPCC Plus MASTR MIX kit (Multiplicom) generated 84 amplicons to analyze the whole four MMR and exons 8 and 9 of TACSTD1 genes. The whole coding regions of FAP genes (APC and MUTYH ) were amplified in three multiplex PCRs
The Journal of Molecular Diagnostics
-
with FAP MASTR MIX kit (Multiplicom). The HNPCC Plus and FAP MASTR Kits combination generated 136 amplicons and screened 36,223 bp. Briefly, multiplex PCR was performed using 50 ng of genomic DNA per reaction. For each sample tested, mixing of the five multiplex PCR products obtained from HNPCC MASTR KIT or HNPCC Plus MASTR KIT and the three multiplex PCR products obtained from the FAP MASTR KIT was performed according to a predefined mixing scheme. Amplicon libraries were purified with AMPure XP (Beckman Coulter, Brea, CA), followed by a two-step dilution in 1/1000ème to be Q15 used as the matrix for next amplification. Amplicons were tagged in a second universal PCR with the molecular identifiers for Illumina MiSeq (Multiplicom) MID 1-48. PCRs were performed with 20 mL of AR1 reagent when HNPCC MASTR MIX kit or FAP MASTR MIX kit was used or 20 mL of AR2 reagent when HNPCC Plus MASTR MIX kit was used. PCR products were purified with AMPure XP (Beckman Coulter), and quantitation was performed on a Qubit instrument with a Qubit dsDNA HS Assay kit (Invitrogen, Thermofisher Scientific), according to the manufacturer’s protocol. Final libraries were obtained by equimolary pooling barcoded PCR products before denaturation and dilution at a 2 nmol/L concentration. Libraries were combined with PhiX control before loading onto a MiSeq instrument (Illumina). NGS sequencing was performed using Nano flow cells and MiSeq Reagent Nano Kit Q16 v2 with 2 251 bp (Illumina).
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
3
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372
373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
Rey et al
Data Analysis Q17
Q18
Nucleotides were aligned using the MiSeq Reporter software v2.5.1 (Illumina) generating *.bam files. *.bam files were used to identify SNVs and CNVs with the SeqNext module of SeqPilot software v4.1.2 (JSI Medical Systems). Regions of interest and amplicon positions were defined for the SeqNext software using the Multiplicom *.manifest files for MMR and FAP gene analyses using HNPCC and FAP MASTR kits. Primers were excluded from the analyses. Nucleotides with quality scores <30 were filtered out. Settings were selected to analyze 500-bp fragments with a threshold of 20 reads per direction to validate region analyses. A low-coverage warning was defined at 100 reads to target amplicons showing low depth of coverage. A depth of at least 50 reads was defined as the cutoff value for amplicon analysis validation. Mutation sorting in distinct section was defined at 20% of variant nucleotide per direction. CNV mode analysis was also performed using the SeqNext module (JSI Medical Systems). For each plex, relative PCR depth in coverage ratio was calculated, taking into account the variations between each amplicon of a target gene and amplicons of the other genes analyzed in the same plex (control amplicons). Each sample was compared to all other samples included in the same run. As DNA libraries were performed using two different MASTR Kits, MMR and FAP CNV analyses were performed separately. Thresholds were set to 0.7 for deletions and 1.3 for duplications.
sequences) [GenBank, https://www.ncbi.nlm.nih.gov/gene; accession numbers NM_000249.3 (MLH1), NM_000251.2 (MSH2), NM_000179.2 (MSH6), NM_000535.5 (PMS2), NM_000038.5 (APC ), NM_001044817 (MUTYH )].
Criteria for Variant Classification VUSs and newly identified VUSs were classified in five categories, according to chemical, biological, in silico, and clinical features. The amino acid polarity and overall dimension changes, phylogenic conservation, SIFT score, Q20 phenotypic features as microsatellite and immunohistochemical analyses, expression studies to test splicing variants as ex vivo (minigene) and in vivo (RT-PCR) studies, functional data obtained in yeast, trans or cis position of the VUSs when a deleterious mutation is identified in the same sample, and segregation studies were taken into consideration when available. VUS classification was assessed after consensus decision of the Groupe Génétique et Cancer either class 1 (benign variant), class 2 (probably benign, 95% to 99%), class 3 (VUSs, 5% to 94.9% to be disease causing), class 4 (probably disease-causing variant, 95% to 99%), and class 5 (disease-causing variant >99% to be deleterious). At least two major criteria are needed to classify a VUS as disease causing (trans position and informative segregation study). The classification of each variant is upgraded at regular intervals in the Universal Mutation Database to have a reliable and homogeneous interpretation for diagnosis reports.
Results Variant Confirmation and Annotation
Q19
In the validation cohort (48 samples), VUSs and diseasecausing variants were previously identified by referent methods: Sanger sequencing or MLPA using SALSA MLPA P003-C1 MLH1/MSH2, P072-C1 MSH6, P008-B2 PMS2, P043-B1 APC, and P378C1 MUTYH kits (MRC Holland), according to the manufacturer protocols, then confirmed with NGS method. Exon amplification and sequencing of PMS2 were performed as previously described.15 For mutation analysis in PMS2 exon 11, forward primer in exon 10 and reverse primer in exon 11 were used.16 In the clinical cohort (176 samples), only VUSs and disease-causing variants detected in MLH1, MSH2, MSH6, APC, and MUTYH using NGS method were confirmed with Sanger method, even though CNVs detected in all MMR and FAP genes were confirmed with MLPA. For PMS2, only CNV analysis was used for diagnostic reports. Sequencing and MLPA products were analyzed on an ABI PRISM 3130xl instrument (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA). Sequence variations were annotated according to the recommendations of the Human Genome Variation Society (ie, nucleotide numbering starting at position þ1 corresponds for the A of the ATG translation initiation codon in coding cDNA reference
4
In the first validation phase of the study, we assessed the completeness of the MMR and FAP gene coding regions and control splicing sites in the design of HNPCC MASTR Plus and FAP MASTR kits as well as the ability of SeqNext software (JSI Medical Systems) to detect all kinds of SNVs and CNVs. We compared the results displayed using several bioinformatics parameters to fix the analysis settings. In the second validation step, the robustness of the workflow was evaluated in routine diagnostic conditions, and we focused on the false-positive rate of CNV detection.
Method Validation Nucleotide Analysis Runs were validated according to quality parameters. Median values across the 28 runs were 96.2% (92% to 99.5%) Q21 for Q30 value, 96.8% (92.6% to 98.3%) for the average rate Q22 Q23 of identified reads passing filter, and 481 clusters/mm2 (271 Q24 to 672 clusters/mm2) for cluster density. Lower cluster densities were linked to better Q30 and passing filter values (data not shown). There were no differences in cluster density, Q30 values, and identified reads passing filter between runs testing HNPCC MASTR Kit only or HNPCC Plus MASTR Kit combined with FAP MASTR Kit.
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Improving Mutation Screening Using NGS Amplicon Analysis Coverage analysis indicated that 100% of the MMR and FAP regions of interest were covered using the combination of HNPCC MASTR Plus and FAP MASTR kits (Supplemental Appendix S1). Regions of interest overlap the entire coding regions and few nucleotides in the 50 , 30 intronic, 50 untranslated, and 30 untranlsated regions. Intronic regions, 50 untranslated region, and 30 untranslated region were generally covered at least on 22 nucleotides (22 to 236 nucleotides can be analyzed). Five intronic regions could be analyzed on <20 nucleotides. The 50 regions of MLH1, MSH2, and APC introns 13 were covered until 9, 9, and 6 nucleotides, respectively, and the 50 intron 9 and 30 intron 14 regions of PMS2 were covered until 3 and 16 nucleotides, respectively (data not shown). Validation of each amplicon analysis was allowed with a depth of coverage of at least 50 reads. As average amplicon sizes obtained after universal PCR using HNPCC Plus MASTR Q25 Kit and FAP MASTR Kit are 469 and 498 pb, respectively, we chose to perform runs up to 2 251 sequencing cycles. To balance some low PCR efficiency and as the regions spreading over 250 pb of large amplicons were sequenced only in one direction, MASTR primer design was optimized to generate overlapping fragments leading to depth-ofcoverage variations between regions in each gene tested ½F2 (Figure 2). These variations were reproducible across all samples tested in each run (data not shown). Overall, the mean depth across all tested regions was of 225 to 557 (Figure 1). All amplicons showed a depth coverage of at least 50 reads (Figure 2).
Mutation Analysis and Variant Detection SNV Detection Eighty patient samples were entirely tested using referent and NGS methods. For proof of principle and looking for potential false negatives, 48 patient samples (runs 1 to 6) Q26 were previously tested using referent methods: Sanger sequencing for SNV analysis and MLPA (MRC Holland) for CNV analysis. Sixteen patient samples were first analyzed using HNPCC MASTR Kit running on two nano flow cells (Illumina). Thirty-two patient samples were then analyzed using HNPCC Plus MASTR Kit combined with the FAP MASTR MIX Kit (Multiplicom) on four nano flow cells (Illumina). These 48 samples were selected to detect the larger panel of SNV and CNV before starting molecular diagnostics. Seventy-nine SNVs (59 substitutions, 13 deletions, 5 duplications, and 2 deletions/insertions) spanning entire amplicons were tested with particular attention to ½T1 identify those located at the boundaries (Table 1). Twentytwo silent variants were tested at heterozygous or homozygous status. MMR mutations were selected among those identified in our laboratory and tested on four nano flow cells. One was dedicated to identify FAP gene mutations (nine SNVs and one CNV), and another one to identify nine
The Journal of Molecular Diagnostics
-
PMS2 SNVs (Table 1). We correctly identified all of the 79 known SNVs in control samples. The intronic variant affecting MSH2 intron 5 splicing junction (c.942þ3A>T) was detected using the two kits combination. It was present on 35% of 583 reads (Figure 3). ½F3 Even though this SNV was easily identifiable on SeqNext (JSI Medical Systems) electrophoregrams, it was not pointed out in the distinct section of SeqNext Sequence window. Other polynucleotide stretches in the 30 regions of MLH1 introns 11 (polyT21), MSH2 intron 1 (polyT13), and MSH6 introns 7 [polyT13 and 9 (polyT18)] were systematically focused. NGS electrophoregrams were in accord with those obtained with the Sanger method (data not shown). Eight patient samples harboring nine SNVs in PMS2 were analyzed (Table 1). All these variants were detected but present on only 23% to 26% of total reads on amplicons sequenced with a depth of coverage of 283 to 1141 reads (data not shown). Thirty-eight VUSs not previously identified with the referent method were detected. They were present on 21% to 70% of total reads and mostly spanning from exons 1 to 5, exon 9, and exons 11 to 15 (data not shown). A DNA sample showing a variant previously identified as c.2186_2187del in PMS2 exon 13 using Sanger Q27 sequencing was tested as control for the method validation phase. Using the NGS method, the variant was assigned as c.2182_2184delinsG (Figure 4). The depth of coverage was ½F4 834 reads, and this mutation was present on 26% of the 417 forward and reverse reads. A patient sample harboring a disease-causing variant in MSH2 exon 11 (c.1705_1706del) was explored. This 2-bp deletion is located on a probe hybridization site so that MLPA analysis showed a false-positive deletion of exon 11. NGS analysis identified this SNV even though CNV mode analysis was mutation free (Figure 5). ½F5 CNV Detection CNV detection was run for every sample tested. Seventeen of the 22 MMR CNVs identified in our laboratory and one APC CNV provided by the CHRU of Lille were tested as positive controls (Table 1). For samples analyzed using CNV mode, all runs included eight samples, seven as controls. It was important to test whether two samples showing CNV located in the same gene could decrease detection sensitivity when analyzed in the same run. So distribution of positive controls among the different runs was optimized to test the method sensitivity. Taken into consideration the CNV frequency in the global population, two samples harboring CNVs identified in two different genes were included in runs 1 to 3, and three samples harboring CNVs, two in MSH2 and/or TACSTD1 (exons 8 to 9), and one in other genes were included in runs 4 to 6. All positive controls were detected and correctly assigned, confirming that this method is able to detect at least two samples harboring CNVs located in the same gene. To use this NGS method in routine diagnosis, results had to be confirmed. To look for potential false negatives and/or positives, 32 patient samples were simultaneously analyzed (runs 7 to 10) using NGS and referent methods. Patient
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
5
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620
Coverage
MLH1 3000 2500 2000 1500 1000 500 0
Coverage
MSH2 3000 2500 2000 1500 1000 500 0
Coverage
MSH6 3000 2500 2000 1500 1000 500 0
Coverage
APC 3000 2500 2000 1500 1000 500 0
MUTYH
Coverage
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
Rey et al
3000 2500 2000 1500 1000 500 0
Figure 2
Depth of coverage distribution across each targeted exon obtained from a representative patient sample. Horizontal black bars represent exons with their flanking intronic regions analyzed. Vertical solid bars represent the different depth of coverage obtained from overlapping or nonoverlapping amplicon sequencing. Values >500 and >1000 reflect overlapping sequencing regions for MMR and FAP genes, respectively.
One hundred seventy-six patient samples provided by genetic counselors were tested according to routine procedure in 22 nano flow cells (runs 7 to 28). Sanger sequencing or MLPA method confirmed all variants of unknown significance and
disease-causing mutations. Two SNVs located in MLH1 intron 1 at positions c.116þ7A>T and c.116þ11A>G were present in 25% to 45% of total reads. These SNVs were recurrently identified in 29 samples analyzed in runs showing cluster density >580/mm2. As they were also not confirmed by Sanger sequencing, they were classified as false positives. In toto, during the validation phase and clinical runs, 118 CNVs observed with NGS analysis (13%) were not confirmed using MLPA and were thus labeled as false positives. These CNVs showed values slightly beyond from the normal thresholds (0.65 to 0.7 for deletions and 1.30 to 1.35 for duplications), exons deleted and/or duplicated were not
6
jmd.amjpathol.org
samples were provided by genetic counselors and results were reported accordingly. All variants (SNVs and CNVs) identified with the NGS method were confirmed by Sanger sequencing or MLPA. All samples with normal CNV profiles were checked using MLPA without any false negative.
NGS in Routine Diagnosis
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
Improving Mutation Screening Using NGS Table 1
Variants Tested during the NGS Method Validation
Gene
NV at heterozygous state
MLH1
NV at heterozygous or homozygous state c.655A>G c.1668-19A>G
MSH2-TACSTD1
c.984C>T c.1760-62G>A
c.1077-80G>A c.1077-10T>C c.1511-9A>T c.1661þ12G>A c.2006-6T>C c.1-118T>Cy c.211þ9C>Gy
MSH6
c.261-36A>G c.458-52G>T c.628-56C>T c.3557-4del c.3801þ54C>G c.4002-10del
c.116A>Gy c.186C>Ay c.260þ22C>Gy c.276A>G c.540T>C c.642C>T c.3173-101G>C c.3438þ14A>T c.3556þ15T>A c.642C>T c.3173-101G>C c.3438þ14A>T c.3556þ15T>A c.3646þ29_3646þ32del c.3647-51_3647-35del c.3802-40C>G c.3802-43del
PMS2
APC
MUTYH
c.645þ32C>T c.1458T>C c.1635G>A c.4479G>A c.5034G>A c.5268T>C c.5465T>A c.5880G>A c.64G>A c.462þ35A>G c.972G>C c.1405A>C c.1435-40C>G
VUS
DM
CNV
c.1331A>G c.1505A>G* c.1682A>G c.974C>T
c.1489dup c.1852_1853del c.1888_1892del c.1147C>T c.1277-2A>G c.689_691delinsTT c.1705_1706del c.942þ3A>T
E1-19del
c.3801þ53dup
c.3476dup c.742C>T c.1572C>G c.2804_2805del
c.1531A>G c.1789A>T c.1866G>A c.2541G>T
c.2341C>T c.1687C>T c.1746del c.2182_2184delinsG c.2341dup c.1700del c.4970dup c.2547_2550del
c.1398A>T
TACSTD1 del TACSTD1-MSH2-E1-2del TACSTD1-MSH2-E1-5del E1-2del E5-6del E8del E8dup E8-10del E11del E8-16del E9-16del E15-16del E11-16dup E5-6del
E6-11del E10del
E4del
c.494A>G c.[1145G>A(þ)1395_1397del] c.[494A>G(þ)1145G>A] c.[439G>C(þ)494A>G]
CNV, copy number variation; DM, disease-causing mutation; NGS, next-generation sequencing; NV, neutral variant; VUS, variant of unknown significance. *VUS not described in the Universal Mutation Database found associated with the biallelic MUTYH inactivation c.[1145G>A(þ)1395_1397del]. y Variants only detected with HNPCC Plus MASTR Kit.
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
7
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
Rey et al
print & web 4C=FPO
869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 Figure 3 c.942þ3A>T in MSH2 intron 5 detected in one control sample and two patient samples with diagnostic report. 892 893 894 adjacent, and often showed exon duplications and deletions in and MLPA approaches, based on individual testing, to 895 the same gene probably reflecting bad DNA quality. Two of NGS-based analysis. In this study, we validated a method to 896 them showing either one exon deletion or duplication were test a targeted panel of six genes predisposing to familial 897 tested after a new DNA extraction and showed a normal CNV colorectal cancer and Lynch syndrome using a combination 898 profile (data not shown). of two libraries obtained from Multiplicom MASTR kits. 899 Twenty-seven new variants of unknown significance and NGS analysis was performed on a MiSeq apparatus (Illu900 23 disease-causing or probably disease-causing variants (21 mina) and *.bam files were obtained using MiSeq reporter 901 SNVs and 2 CNVs) were identified. The disease-causing (Illumina) processing feature of computationally pipeline 902 903 ½T2 mutation detection rate was 13.1% (Table 2). The MSH2 conversion of images to sequence reads (base calling). 904 intronic variant c.942þ3A>T was identified in two patient Alignment and assembly of data were optimized using the 905 samples. It was detected on 25% and 27% of 221 and 274 SeqNext software (JSI Medical Systems) to identify SNV 906 reads, respectively, and was unambiguous on SeqNext elecand CNV using the same NGS data. Our current study 907 trophoregrams. As the control sample tested in the validation confirmed NGS using Multiplicom Kits, MiSeq (Illumina) 908 phase, this SNV was not pointed out in the distinct section of technology, and SeqNext software analysis as a reliable 909 the Sequence SeqNext window (Figure 3). method to detect mutations with good sensitivity and 910 Not surprisingly, we identified SNVs considered as disspecificity. A few points remain to be improved. 911 ease causing in the MUTYH coding sequence in patient The detection accuracy of Lynch syndromeeassociated 912 samples assessed by genetic counselors for MMR testing SNVs in PMS2 is complicated with the presence of pseu913 only. Two mutations were identified in a patient sample as dogenes. In the current study, all heterozygous SNVs pre914 915 c.[1105del(þ)1145G>A]. Located on two different alleles, viously identified16,17 in the eight control samples were 916 these mutations led to a biallelic inactivation of the MUTYH present only at low percentage of total reads. SNVs not 917 ½F6 gene, where no MMR alteration was detected (Figure 6). A previously detected with referent methods were observed in 918 the same samples. These SNVs were mostly detected in disease-causing mutation was identified in MSH6 at position 919 exons showing sequence similarities with pseudogenes and c.3261dup (p.Phe1088Leufs*5) in another sample. Per920 present in 21% to 70% of total reads. This NGS technology forming NGS sequencing, analysis detected a monoallelic 921 is sensitive enough to detect PMS2 mutations with a diluted mutation of MUTYH (c.494A>G) (data not shown). 922 signal in mixed homologous sequences but cannot 923 discriminate if some of them are part of PMS2 or pseudo924 genes. The identification of SNVs located only on pseudoDiscussion 925 genes is probably reflecting sequence similarities 926 927 coamplification. Although further optimizations are needed An increasing number of genes involved in cancer pre928 dispositions and patient samples provided by genetic to improve PMS2 analysis with this method to provide 929 counseling clearly require a shift from Sanger sequencing clinically relevant results like primer design might be 930
8
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
Q30
Q28
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
print & web 4C=FPO
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
Improving Mutation Screening Using NGS
Figure 4
Mutation identified in a control sample in PMS2 exon 13 as c.2182_2184delinsG using NGS method (A) or c. 2186_2187del using Sanger method (B). The insG signal diluted in the background is pointed out by an arrow.
improved to upgrade PMS2 specific amplification and reduce pseudogene selection, an upgraded software version able to detect haplotypes with many rare SNVs in several samples analyzed in the same run might help to define pseudogene sequences to avoid and select true PMS2 SNVs. Despite these improvements, this method lacks accuracy in assigning SNVs to PMS2. Unequivocal differentiation of PMS2 from its pseudogenes, especially PMS2CL, which shares 98% sequence with the 30 region or the active gene, should use MLPA and long-range PCR/NGS and Sanger combined analysis, as described by Li et al.18 One control sample was tested for a disease-causing mutation located in PMS2 exon 13. The mutation was identified as c.2186_2187del or c.2182_2184delinsG using Sanger or NGS approaches, respectively. Accurate analysis of Sanger results revealed A/G nucleotide superposition with a diluted G signal at position c.2182. Taking this observation into consideration, the light G signal was concordant with NGS assignment: the background observed using Sanger method
The Journal of Molecular Diagnostics
-
herein evidences that complex SNVs as delins and nucleotide variations generating low signal are more accurately assigned using the NGS method as compared to the conventional sequencing method. HNPCC Plus MASTR Kit was developed to improve the sensitivity of SNV detection, particularly those located in GC-rich regions or in nucleotide stretches. SNVs located in GC-rich MSH2 and MSH6 exons 1 were easily identified. The variant at position c.942þ3A>T in intron 5 of MSH2, being the first nucleotide of a 27-A stretch, is known to be critical to detect using the referent sequencing method. In our study, it was detected in one sample tested to validate the method, then in two patient samples assessed by the genetic counseling. Even though this variant was easily identified on SeqNext electrophoregrams, it was not reported in the distinct mutation sorting section of the software, implying systematic focusing on this MSH2 region to complete sample screening for SNV mode analysis.
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
9
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
print & web 4C=FPO
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178
Rey et al
Figure 5 False deletion of MSH2 exon 11 using the MLPA method. In the upper part of the figure, MLPA analysis shows exon 11 deletion (red arrow) and Sanger analysis identifies a 2-bp deletion on MLPA probe hybridization site. In the lower part of the figure, a single run analysis identifies the point mutation, whereas CNV mode analysis shows a normal profile.
All colorectal cancer syndromes caused by known highpenetrance gene alterations collectively account for 2% to 6% of all colorectal cancers. We identified 23 deleterious mutations among the 176 patient samples tested, leading to a 13.1% detection rate, rendering this new method at least as sensitive as the referent one. Genomic rearrangements (CNVs) account for 5%, 7%, 20%, 20%, and 100% of mutations in MLH1, MSH6, MSH2, PMS2, and TACSTD1, respectively.19e21 Identification of CNVs is part of colorectal cancer genetic testing. Two DNA samples obtained from disease-free members of the same family were assessed by the genetic counseling to test MMR and FAP genes as probands. Heterozygous MSH2 deletion of exons 9 to 16 was identified in both samples, leading to CNVs’ rate detection close to that of the general population.22 These samples were obtained from a mother and her daughter, confirming the robustness of this approach and showing that this method is able to detect a same CNV present in two different samples analyzed in a single run; nevertheless, in routine diagnostics, it would be advisable to test samples obtained from more than two members of the same family on separate runs. For diagnostic testing, wherein the CNV status of samples was a priori not known, if more than two CNVs located in the same gene were detected in samples analyzed in a single run (when a common
CNV founder mutation is shared by more than two patients), CNV analysis might be performed, excluding one or two samples showing abnormal values. To upgrade the method sensitivity, we have tested and validated this technique to analyze 16 samples in a single run (data not shown). Increasing batch size has permitted us to mitigate the risk that true-positive samples might be not detected if more than two samples harboring the same CNV are analyzed in one run. A patient sample showing a mutation at nucleotides c.1705_1706del on MSH2 was tested. The two-step process analysis using referent methods generated a deletion of exon 11 using MLPA, whereas Sanger analysis revealed a 2-bp deletion located on the MLPA hybridization site. The NGS method merges SNV and CNV mode analysis in the same test and generates information of diagnostic relevance available in a one-step process. A current challenge for molecular testing colorectal cancers is to identify the genetic cause when the colorectal cancer syndrome is associated with low-penetrance alleles, leading to clinical heterogeneity and gene-overlapping syndromes. In some cases, clinical characterization cannot distinguish between Lynch syndrome and attenuated adenomatous polyposis. One of 176 patient samples was first assessed for MMR screening. This patient developed an
10
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302
Improving Mutation Screening Using NGS Table 2
VUSs and Disease-Causing Mutations Identified Using NGS Method in 176 Proband Samples Analyzed with Diagnostic Reports
Gene
Newly identified VUS
Disease-causing mutations
MLH1
c.116þ132G>A c.1039-31A>T
MSH2-TACSTD1
c.645þ175C>G c.1759þ57G>T c.2281G>C c.2281A>G (two probands) c.3467T>C
c.676C>T c.731G>A c.767del c.1178T>C c.1943C>T* c.942þ3A>T (two probands) c.2131C>T
MSH6
APC
MUTYH
c.136-53T>C c.388A>G c.645þ33G>A c.645þ129A>C c.730-160A>G c.1585C>T c.3270A>G c.5009C>T c.5140G>A c.5670A>G c.7481C>T c.7888G>A c.270C>T c.306þ33G>A c.565C>T c.883C>T c.891þ24G>A c.1398A>T c.1476þ73C>T
CNV
E 9-16 dely
c.5C>A c.1344dup c.3261dup c.3477C>A c.3775_3776del c.4001G>A c.694C>T
c.494A>Gz c.1105del c.1145G>Ay
CNV, copy number variation; NGS, next-generation sequencing; VUS, variant of unknown significance. *Variant class 4 in the Universal Mutation Database. y Mutations identified in two different samples. z Mutation identified in three different samples.
adenocarcinoma of the ascending colon at the age of 45 years and two third-degree relatives also exhibited colorectal cancers at the ages of 71 and 75 years. He carried a biallelic germline mutation of MUTYH. Screening of MMR and FAP genes simultaneously facilitates accurate and rapid individual risk assessments, enabling personalized family surveillance and preventive measures. The identification of individuals at high risk of cancer allows prevention and/or early cancer detection, resulting in decreasing disease-specific mortality. The heritable nature of colorectal cancer might be associated with the coheritance of multiple variants. The risk associated with each variant might not be individually systematically high, but the combined effect could significantly contribute to disease burden.23 In this study, we identified two heterozygous causal SNVs assigned as c.3261dup and c.494A>G on MSH6 and MUTYH, respectively, in a patient sample assessed for MMR testing. This patient developed endometrium and breast malignancies at the
The Journal of Molecular Diagnostics
-
ages of 66 and 44 years, respectively. The family history showed endometrium cancers developed in mother, aunt, cousin, and grand-mother, plus three colorectal, one pancreatic, and one brain carcinomas, all present in the maternal branch of the pedigree. The risk of developing malignancies associated with this MSH6 mutation might be increased in the presence of the MUTYH mutation, suggesting that additional lowpenetrance alleles may explain the increased risk of cancer in families with mutations associated with Lynch syndrome. This NGS method is helpful to identify additional cancer risk modifiers in the same analysis and provides a useful tool to optimize clinical issues as preventive measures and genetic counseling. Faithful DNA replication is essential to maintain genomic stability and to prevent mutagenesis and tumor development. Eukaryotic DNA replication accuracy results from a combination of the high fidelity of DNA polymerases and post-replication surveillance by the MMR apparatus.24 Recently, it has been shown that deficient proofreading
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
11
1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364
1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426
Rey et al
I
1
2
II
4
3
5
IV
7
2
1
III
6
1
2
1
2
Figure 6 Pedigree of a patient assessed for MMR gene screening by the genetic counseling and showing MUTYH mutations (proband III-1, pointed out by an arrow). Siblings I-1 and I-6 developed colon cancer.
activity of DNA polymerases d or ε is associated with cancer progression.25 Germline mutations of POLD1 and POLE encoding the 30 to 50 exonuclease domain components of DNA polymerase d and ε, respectively, have been associated with colorectal and/or endometrial cancer predisposition in families without MMR defect,26,27 and POLE mutations have been identified in patients with adenomatous polyposis and early-onset colorectal cancer.28 As POLD1 or POLE genetic alterations contribute to Lynch-like or FAP syndromes, respectively, molecular testing of these two genes might be included for diagnostic purposes. Other genes already included in large panels to test familial colon cancer predisposition, such as PTEN, SMAD4, STK11, or BMPR1A, have been excluded from ours because their alteration usually contributes to confirm a clinical diagnosis and they are useful for genetic counseling only. In our study, NGS screening using the six-genes panel allowed a time and cost reduction as compared to the referent methods and facilitates the identification of gene mutation carriers by laboratories checking many samples to be rapidly analyzed. This method might improve the effectiveness of preventive measures.
Acknowledgments We thank Isabelle Coupier, Carole Corsini, and Jacqueline Duffour for providing patient samples and Marie-Christine Lecque for preparing genomic DNA.
Supplemental Data Supplemental material for this article can be found at http://dx.doi.org/10.1016/j.jmoldx.2017.04.005.
References 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM: Estimates of worldwide burden of cancer in 2008: GLOBO-CAN 2008. Int J Cancer 2008, 127:2893e2917
12
2. Boland CR: Evolution of the nomenclature for the hereditary colorectal cancer syndromes. Fam Cancer 2005, 4:211e218 3. Aaltonen LA, Peltomäki P, Mecklin JP, Järvinen H, Jass JR, Green JS, Lynch HT, Watson P, Tallqvist G, Juhola M: Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res 1994, 54:1645e1648 4. Sieber OM, Tomlinson IP, Lamlum H: The adenomatous polyposis coli (APC) tumor suppressor genetics, function and disease. Mol Med Today 2000, 6:462e469 5. Jones S, Emmerson P, Maynard J, Best JM, Jordan S, Williams GT, Sampson JR, Cheadle JP: Biallelic germile mutations in MYH predispose to multiple colorectal adenoma and somatic G:C? T:A mutations. Hum Mol Genet 2002, 11:2961e2967 6. Sieber OM, Lipton L, Crabtree M, Heinimann K, Fiadalgo P, Philips RK, Bisgaard ML, Orintof TM, Aaltonin LA, Hodgeson SV, Thomas HJ, Tomlinson IP: Multiple colorectal adenomas, classic adenomatous polyposis, and germline mutations in MYH. N Engl J Med 2003, 348:791e799 7. Bisgaard ML, Fenger K, Bulow S, Neiburh E, Mohr J: Familial adenomatous polyposis (FAP): frequency, penetrance, and mutation rate. Hum Mutat 1994, 3:121e125 8. Soravia C, Berk T, Madlensky L, Mitri A, Cheng H, Gallenger S, Cohen Z, Bapat B: Genotype-phenotype correlations in attenuated adenomatous polyposis coli. Am J Hum Genet 1998, 62:1290e1301 9. Knudsen AL, Bisgaard ML, Bülow S: Attenuated familial adenomatous polyposis (AFAP): a review of the literature. Fam Cancer 2003, 2:43e55 10. Sabatini LM, Mathews C, Ptak D, Doshi S, Tynan K, Hegde MR, Burke TL, Bossler AD: Genomic sequencing procedure microcosting analysis and health economic cost-impact analysis. J Mol Diagn 2016, 18:319e328 11. Pritchard CC, Smith C, Salipante SJ, Lee MK, Thornton AM, Nord AS, Gulden C, Kupfer SS, Swisher EM, Bennett RL, Novetsky AP, Jarvik GP, Olopade OI, Goodfellow PJ, King MC, Tait JF, Walsh T: ColoSeq provides comprehensive lynch and polyposis syndrome mutational analysis using massively parallel sequencing. J Mol Diagn 2012, 14:357e366 12. Campbell PJ, Stephens PJ, Pleasance ED, O’Meara S, Li H, Santarius T: Identification of somatically acquired rearrangements in cancer using genome-wide massively parallel paired-end sequencing. Nat Genet 2008, 40:722e729 13. Vasen HF, Watson P, Mecklin JP, Lynch HT: New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on HNPCC. Gastroenterology 1999, 116:1453e1456 14. Piñol V, Castells A, Andreu M, Castellví-Bel S, Alenda C, Llor X, Xicola RM, Rodríguez-Moranta F, Payá A, Jover R, Bessa X; Gastrointestinal Oncology Group of the Spanish Gastroenterological Association: Accuracy of revised Bethesda guidelines, microsatellite instability, and immunohistochemistry for the identification of patients with hereditary nonpolyposis colorectal cancer. JAMA 2005, 293:1986e1994 15. Luk GD: Diagnosis and therapy of hereditary polyposis syndromes. Gastroenterologist 1995, 3:153e167 16. Nakagawa H, Lockman JC, Frankel WL, Hampel H, Steenblock K, Burgart LJ, Thibodeau SN, de la Chapelle A: Mismatch repair gene PMS2: disease-causing germline mutations are frequent in patients whose tumors stain negative for PMS2 protein, but paralogous genes obscure mutation detection and interpretation. Cancer Res 2004, 64: 4721e4727 17. Auclair J, Leroux D, Desseigne F, Lasset C, Saurin JC, Joly MO, Pinson S, Xu XL, Montmain G, Ruano E, Navarro C, Puisieux A, Wang Q: Novel biallelic mutations in MSH6 and PMS2 genes: gene conversion as a likely cause of PMS2 gene inactivation. Hum Mutat 2007, 28:1084e1090 18. Li J, Dai H, Feng Y, Tang J, Chen S, Tian X, Gorman E, Schmitt ER, Hansen TAA, Wang J, Plon SE, Zhang VW, Wong LJC: A comprehensive strategy for accurate mutation detection of the highly homologous PMS2. J Mol Diagn 2015, 17:545e553
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:45 am EO: 2017_25
1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488
1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
Improving Mutation Screening Using NGS 19. Wijnen J, van der Klift H, Vasen H, Khan PM, Menko F, Tops C, Meijers Heijboer H, Lindhout D, Moller P, Fodde R: MSH2 genomic deletions are a frequent cause of HNPCC. Nat Genet 1998, 20:326e328 20. Plaschke J, Krüger S, Dietmaier W, Gebert J, Sutter C, Mangold E, Pagenstecher C, Holinski-Feder E, Schulmann K, Möslein G, Rüschoff J, Engel C, Evans G, Schackert HK: Eight novel MSH6 germile mutations in patient with familial and nonfamilial colorectal cancer selected by loss of protein expression in tumor tissue. Hum Mutat 2004, 23:285 21. Senter L, Clendenning M, Sotamaa K, Hamplel H, Green J, Potter JD, Lindblom A, Lagerstedt K, Thibondeau SN, Lindor MN, Young J, Winship I, Dowty JG, White DM, Hopper JL, Baglietto L, Jenkins MA, de a Chapelle A: The clinical phenotype of Lynch syndrome due to germline PMS2 mutations. Gastroenterology 2008, 135:419e428 22. Wang Y, Friedl C, Lambert C, Jungck M, Mathiak M, Pagenstecker C, Propping P, Mangold E: Hereditary non polyposis colorectal cancer: frequent occurrence of large genomic deletions in MSH2 and MLH1 genes. Int J Cancer 2003, 103:636e641 23. Theodoratou E, Montazeri Z, Hawken S, Allum GC, Gong J, Tait V, Kirac I, Tazari M, Farrington SM, Demarsh A, Zgaga L, Landry D,
The Journal of Molecular Diagnostics
-
24. 25. 26.
27.
28.
Benson HE, Read SH, Rudan I, Tenesa A, Dunlop MG, Campbell H, Little J: Systematic meta-analysis and field synopsis of genetic association studies in colorectal cancer. J Natl Cancer Inst 2012, 104: 1433e1457 Kunkel TA: DNA replication fidelity. J Biol Chem 2004, 279: 16895e16898 Lange S, Takata K, Wood R: DNA polymerases and cancer. Nat Rev Cancer 2011, 11:96e110 Church DN, Briggs SEW, Palles C, Domingo E, Kearsey SJ, Grimes JM, Gorman M, Martin L, Howarth KM, Hodgson SV; NSECG Collaborators, Kaur K, Taylor J, Tomlinso PM: DNA polymerase e and d exonuclease domain mutations in endometrial cancer. Hum Mol Genet 2013, 22:2820e2828 Palles C, Cazier JB, Howarth KM, Doming E, Jones AM, Broderick P, et al: Germline mutations affecting the proof-reading domains of POLE and POLD1 predispose to colorectal adenomas and carcinomas. Nat Genet 2013, 45:136e144 Briggs S, Tomlinson I: Germline and somatic polymerase epsilon and delta mutations define a new class of hypermutated colorectal and endometrial cancers. J Pathol 2013, 230:148e153
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI597_proof 11 May 2017 4:46 am EO: 2017_25
13
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528