| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Molecular Diagnostics and Genetics |
1 WHO HPV LabNet Global Reference Laboratory and Departments of 2 Medical Microbiology and 3 Clinical Chemistry, Lund University, Malmö University Hospital, Malmö, Sweden.
aAddress correspondence to this author at: Department of Clinical Chemistry, Lund University, Malmö University Hospital, SE-205 02 Malmö, Sweden. Fax 46-40-336286; e-mail joyce.carlson{at}med.lu.se.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used the Sequenom MassARRAY platform to develop a high-throughput mass spectrometric (MS) method for detecting 14 specific oncogenic HPV genotypes in multiplex PCR products. We compared results from 532 cervical cell samples to the comparison method, reverse dot blot hybridization (RDBH).
Results: The MS method detected all samples found positive by RDBH. In addition, the MS method identified 5 cases of cervical disease (cervical intraepithelial neoplasia of grade I or higher) that RDBH analysis had missed. Discrepancies in specific genotypes were noted in 20 samples, all positive by MS, with an overall concordance of 
= 0.945.
Conclusions: The MS high-throughput method, with a processing capacity of 10 x 384 samples within 2 working days and at a consumables cost of about US$2 per sample, performed as well as or better than the comparison method.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although several primarily PCR-based methods have been used to type HPV (9), the massive scale of the monitoring and primary-screening efforts likely to be launched in the near future will require more efficient methods. In the present study, we developed a high-throughput multiplex analysis of 14 distinct HPV high-risk genotypes on the Sequenom MassARRAY matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) system. We compared its performance with that of another method for HPV genotyping, reverse dot blot hybridization (RDBH) of PCR amplimers, as well as with histopathologic diagnoses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CIN I, biopsy results CIN I, or abnormal colposcopy results) were treated with loop or laser conization (n = 199). All patients were invited to follow-up visits at 4–6 months and 16–18 months after treatment, whether they received treatment or not, and 261 new samples of cervical cells were obtained for cytologic and HPV testing, for a total of 532 samples. The study was approved by the ethics review board of Västerås Hospital.
sample preparation
Cervical cells were suspended in 1 mL 154 mmol/L NaCl and stored at –20 °C. After thawing, the cells were centrifuged at 3 000g for 10 min, and pellets were resuspended in 1 mL 10 mmol/L Tris-HCl, pH 7.4. Before RDBH analysis, 100-µL aliquots were freeze-thawed and boiled for 10 min to release the DNA.
After storage at –80 °C for 3–6 years, 40-µL aliquots of the same 532 DNA extracts were dried in 96-well plates at 37 °C overnight and resuspended in 8 µL sterile water; 2 µL of this suspension was used as template in the primary PCR reaction of the MS method.
To evaluate the sensitivities of the 2 methods, we used standard 10-fold dilution series of 1–1 000 plasmids/reaction with type-specific inserts of each target HPV type and 1–1 000 copies/reaction of DNA extracted from SiHa cells (a cervical cancer cell line containing 1 copy of HPV 16 DNA per cell; American Type Culture Collection).
comparison method
GP5+/6+ consensus primers (11) and BGPCO5/ BGPCO3 primers (12) were used to amplify HPV DNA and the human β-globin gene (HBB), respectively, in separate 50-µL PCR reaction mixes, each containing 10 µL DNA solution as template. The presence of any of the 14 target HPV types or HBB DNA in 5-µL aliquots of PCR product was detected by enzyme immunoassay (13). Samples positive for HPV (n = 260) in the enzyme immunoassay were further tested by RDBH (10)(14) for the specific HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68.
mass spectrometry
The MS method involves a multiplex primary PCR, in this case with several HPV primers, followed by a homogeneous mass extend (hME) reaction with a single primer of distinct mass that is specific for each genotype. Subsequently, unextended primers demonstrate the absence and specifically extended primers verify the presence and identity of each specific genotype. After laser deionization, the time of flight, which increases with the m/z, is detected for each hME primer. Unless otherwise specified, all procedures were performed with protocols and materials from Sequenom.
primer design
For the primary PCR, we used Oligo 6.0 software (Cambio) to modify the design of GP5+/6+ consensus primers to provide optimal annealing to the 14 target HPV types (Table 1
). We produced new forward primers FA (all 14 HPV types), FB (HPV 16, 39, 45, 51, 56, 59, 66), FC (HPV 18, 31, 33, 52, 58), and FD (HPV 35, 68) and reverse primers RG (HPV 16, 18, 45, 56, 66, 68), RH (HPV 51, 39), RI (HPV 31, 52, 59), and RJ (HPV 33, 35, 58). The number of mismatches was minimized, and A-C mismatches were avoided. All primers have the 5' 10-base extension recommended by Sequenom to improve annealing stability in multiplex PCR reactions (see Table 1
). PCR programs were optimized, and products were viewed after agarose gel electrophoresis. The 14 unique HPV type–specific hME primers are shown in Table 2
.
|
|
primary pcr
PCR reaction mixes (6 µL) contained 2 µL DNA template and a final concentration of 0.3 µmol/L of each outer primer, 200 µmol/L of each deoxynucleoside triphosphate, 10 mmol/L Tris HCl, pH 8.3, 50 mmol/L KCl, 2.5 mmol/L MgCl2, and 0.2 U AmpliTaq Gold DNA polymerase (Roche Molecular Systems). The mixes were robotically pipetted with disposable tips and amplified in 384-well plates. Separate dilutions of plasmids with HPV type–specific inserts, 1 000 copies/µL of each HPV type, and a pool of all HPV plasmids with 1 000 copies of each type/µL were used as positive controls. Negative controls were sterile water and 10 ng/µL of human DNA extracted from pooled human peripheral blood leukocytes. We performed all primary PCR amplifications with a GeneAmp PCR System 9700 (Applied Biosystems) as follows: denaturation at 95 °C for 10 min; 5 cycles of 95 °C for 30 s, 42 °C for 30 s, and 72 °C for 45 s; 45 cycles of 95 °C for 30 s, 64 °C for 30 s, and 72 °C for 45 s; and a final step at 72 °C for 10 min. Primary PCR reaction mixes were dephosphorylated with shrimp alkaline phosphatase (Sequenom).
Hme reaction
The hME reaction mix was added to the dephosphorylated primary PCR reaction mix and included 1 µmol/L of each hME primer (Table 2
), 0.229 µL terminator mix (Sequenom) containing equal amounts of dATP, dCTP, ddGTP, and ddTTP, and 1.25 U Thermo Sequenase (Sequenom) in a final volume of 10 µl. The PCR program was 94 °C for 2 min, followed by 99 cycles of 94 °C for 5 s, 42 °C for 10 s, and 72 °C for 5 s. After desalting by the addition of 6 mg Clean Resin (item #08040, Sequenom) to each 384-well plate, we applied 15 nL of each hME product to a 384-spot SpectroChip (Sequenom). MS analysis was performed and interpreted with MassARRAY Typer software (Sequenom).
discrepancy between methods
Seventy-eight samples showed genotypic discrepancies between the RDBH and MS results. For these samples, we purified DNA from 100-µL aliquots of the original cell suspension in 10 mmol/L Tris-HCl by proteinase K digestion and DNA precipitation (15). The dried pellets were dissolved in 20 µL sterile water. The presence of human DNA in these samples was demonstrated by real-time PCR analysis on a 7900HT instrument (Applied Biosystems) with primers and probes for the human F22
gene [coagulation factor II (thrombin)] (C 8726802; Applied Biosystems). Samples testing negative for human DNA (n = 26) were considered not evaluable and were excluded. All 52 samples positive for human DNA were reanalyzed at least once by the MS method. Fluctuating results for specific genotypes in a few samples were apparently due to viral dosage effects. Of the 52 samples, 39 had 1 or more HPV genotypes detected with the RDBH method than were detected with the MS method, 8 had 1 or more HPV genotypes detected with the MS method than were detected with the RDBH method, and 5 had 1 discrepant HPV genotype detected with each method before the MS reanalysis. These 52 samples with persisting discrepancies between the RDBH and MS results were reanalyzed by RDBH with the GP5+/6+ primers.
After reanalysis with MS and RDBH, discrepant results remained for 24 samples, of which 10 had been positive for a single HPV type. After a new PCR reaction, these 10 samples were sequenced with the same mix of forward primers as in the primary PCR (Table 1
). The presence of HPV found only in the MS method for 6 samples was verified by DNA sequencing. In 4 other cases [none with the diagnosis CIN I or higher (CIN I+)], DNA sequencing was inconclusive. These 4 samples were considered not evaluable and were excluded, for a final total of 502 samples.
statistical methods
Sensitivity, specificity, and positive predictive value were calculated for each method separately regarding the detection of HPV in comparison with histopathologic diagnosis. The concordance between the RDBH and MS methods for the detection of each of the 14 HPV types in all 502 evaluable samples was calculated by means of statistics (16).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
concordance of methods
The degree of overall concordance () between the MS method and the RDBH method was 0.945. Both methods detected multiple HPV types in the same 40 samples and single HPV types in the same 175 samples. In 2 samples, the MS method detected multiple types, whereas the RDBH method detected single types. In another 3 samples, the RDBH method detected multiple types, whereas the MS method detected single types. In another 8 samples, the MS method detected single types, whereas the RDBH method detected no HPV.
We compared the efficiencies of type-specific HPV detection by the 2 methods for all samples. Table 4
shows an overview of the number of type-specific infections detected with the 2 methods. For all 502 samples (with 2 x 14 x 502 results), 28 results (0.2%) in 20 samples did not agree between the 2 methods. Fifteen of the 20 discrepant samples were derived from patients with the diagnosis CIN I+. The HPV types missed by RDBH were among 10 patients with a histopathologic or cytologic diagnosis of CIN I+ and among 5 patients with no CIN, whereas the HPV types missed by MS were among 9 patients with CIN I+ (Table 4
). In samples with multiple HPV types present, the 2 methods may have had slightly different abilities to detect certain HPV types (Table 4
). The MS method alone detected all cases of cancer (Tables 3
and 4
) and detected HPV 68 in 4 samples. The latter result is in contrast to RDBH, which consistently failed to identify this genotype (Table 4
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our evaluation was based on a large clinical sample consecutively enrolled from a population-based screening program, suggesting that our results are generally valid with respect to the method’s usefulness in a screening setting. An additional strength of the study is that we used the detection of HBB or F2 human DNA in each available aliquot to demonstrate the presence of amplifiable DNA. Because the clinical priority to be able to review cytology prevents centrifugation of cervical samples and lysis of all cells to provide a homogeneous DNA solution, the risk for false-negative HPV results due to nonrepresentative aliquots increases with each successive use of a sample.
Although there were discrepancies in the genotyping results, all but 8 discrepant samples (all MS positive and RDBH negative) had at least 1 concordant HPV type and therefore would not have been misclassified as HPV negative in a screening situation. The causes for discrepancy include sampling error, DNA degradation during storage, and potential interference or false-positive reporting of weak RDBH dot blots.
Plasmid dilution series generally demonstrated lower detection limits for MS than for the comparison method, with the most notable difference being the greater ability of the MS method to detect HPV type 68. According to a pooled international analysis of HPV type–specific prevalence in 3 085 cervical cancer cases (17), HPV 68 is the 14th most common type in cervical cancer, more common than HPV 66.
The numbers of type-specific HPV infections detected with the 2 methods were almost equal, but the results of the genotyping differed slightly between methods. HPV 33, 52, and 58 (detected more frequently by MS) and HPV 45 (detected more by RDBH) are among the 8 most common types in cervical cancer (18). HPV types 16 and 18 account for about 70% of cervical cancers worldwide (18). In samples positive for multiple HPV types, these genotypes were each detected weakly in 2 samples by RDBH, whereas MS detected other HPV types, suggesting possible cross-hybridization. For one of the samples positive for HPV 18 by RDBH, there was a weak HPV 18 peak (below cutoff) in the MS analysis, suggesting a low viral load in the sample. Eight additional results were also both MS negative and RDBH positive, and all originated from samples that were concordant with respect to at least 1 HPV type. All samples that were negative for all 14 genotypes by MS were also negative by RDBH.
In a WHO international collaborative study for the assessment of the performance of various HPV-detection assays (19), a panel of HPV types was analyzed in a number of laboratories, of which half failed to detect 10 000 copies of HPV 31 plasmids. Some laboratories did not detect the same copy numbers of HPV 35 and HPV 52. In the present study, the MS analysis detected 100 copies of HPV 31, 10 copies of HPV 35, and a single copy of HPV 52. The primary PCR primers of the MS method were designed to match some target HPV types better than others. A few primers match their templates perfectly, whereas others contain a minimal number of non–C-A mismatches. Although the detection limit for HPV 31 is slightly higher than for all the other types (except HPV 39) with the MS method, the MS method was generally more sensitive for HPV 31, 35, and 52 than the methods in the WHO study. In general, the number of mismatches is slightly lower with the MS primers than with GP5+/6+ primers, but, more importantly, the mismatches are preferentially G-T to optimize thermodynamic stability.
In summary, we have designed a highly automated, potentially high-throughput MS method for the specific detection of 1–100 copies of 14 different oncogenic HPVs in samples of cervical cells. The performance compares well with the comparison method and with other methods reported in the literature. With access to robotic pipetting and the Sequenom system, the MS method allows large-scale HPV genotyping at a low cost, and this method may be useful for cervical screening and to evaluate the effectiveness of HPV vaccination programs via monitoring the circulation of HPV types in vaccinated populations.
|
Acknowledgments |
|---|
Financial Disclosures: None declared.
Acknowledgments: We thank Maria Sterner and Liselott Hall for technical assistance and Per Rymark for the enrollment of patients.
|
Footnotes |
|---|
2 Human genes: F2, coagulation factor II (thrombin); and HBB, (hemoglobin, beta).
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Soderlund-Strand, J. Carlson, and J. Dillner Modified General Primer PCR System for Sensitive Detection of Multiple Types of Oncogenic Human Papillomavirus J. Clin. Microbiol., March 1, 2009; 47(3): 541 - 546. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
