1

L1-peptides were more specific than VLPs for detecting HPV-infected females having 1

cervical lesions, attending Engativa Hospital, Colombia 2

3

Mauricio Urquiza1,2*, Ricardo Sánchez1,2, Jairo Amaya2,3, Sandra León1, Jenny Acosta1, Manuel 4

A Patarroyo1.2, Milena Camargo1 and Manuel E Patarroyo1,2 5

6

1Fundation Instituto de Inmunologia de Colombia 7

2Universidad Nacional de Colombia 8

3Hospital de Engativa 9

*Corresponding author 10

Carrera 50 # 26-00 11

Bogota Colombia Tel: +57-1-4815269, 57-1-4815219 Fax: 57-1-3244672 Ext.108 12

e-mail: mauricio_urquiza@fidic.org.co 13

Running Title: L1 peptides detected females with HPV cervical lesion 14

Key words: cervical cancer, HPV, serology test, cervical lesion, VLPs 15

ACCEPTED

Copyright © 2008, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.00551-08 JCM Accepts, published online ahead of print on 17 September 2008

2

Abstract 1

A serological test for detecting HPV-infected females at risk of developing cervical cancer could 2

be based on conserved low antigenic L1-peptides specifically recognized by antibodies from 3

high-risk human papillomavirus (HR-HPV) infected cervical lesion patients. The aim was to 4

assess the ability of L1-peptides 18283, 18294 and 18301 to identify these females compared to 5

VLPs. Interview data, peripheral blood and cervical cells were obtained from HPV 391 volunteer 6

females for detection of anti-HPV antibodies and HPV- DNA; all of them having pap smear test; 7

287 were referred for colposcopy/biopsy according to gynecological criteria. Agreement 8

determined by Lin coefficient (Rho) showed that 75-83% HR-HPV DNA positive cervical lesion 9

females have antibodies recognizing VLPs, peptides 18283, 18294 or 18301 while 15-23% HPV-10

DNA negative normal cytology females have antibodies recognizing these three peptides and 11

45% recognizing VLPs. Agreement between peptides and VLPs was higher in HPV-DNA 12

positive cervical lesion patients. Peptides 18283, 18294 and 18301 showed similar sensitivity for 13

detecting HR-HPV DNA positive cervical lesions and were more specific than VLPs. Peptide 14

18301 might be detecting protective antibodies in HPV-DNA negative females having ASCUS. 15

These peptides could be useful for designing a serology test for detecting HR-HPV-infected 16

females having cervical lesion at risk of cervical cancer. 17

ACCEPTED

3

Introduction 1

More than 100 human papillomavirus types (HPV) infect human tissues; 40 of these viruses 2

(named high-risk HPV) are able to induce carcinomas such as cervical cancer (2, 23). Around 70 3

percent of females become HPV-infected within the next two years following sexual debut (22, 4

38); 90 percent of them having a transient HPV-infection (1, 13). Cervical lesions caused by 5

persistent HPV-infection are detected in around 5 percent; some of them can progress to cervical 6

cancer. 70-90 percent of HPV-infected females become sero-converted 6 to 18 months after 7

HPV-DNA has been detected and that this rarely occurs in females having transient HPV-8

infection (3, 12, 25). 9

There is a relationship between high viral load and the progression of cervical lesion and cervical 10

cancer (13, 27). Anti-HPV antibody response is correlated with viral load (10, 36). Most 11

potentially oncogenic, persistent, long-term HPV-infections elicit an antibody response which 12

could be detected by using virus-like particles (VLPs) (4, 20). In fact, anti-VLP antibody 13

response is stable, related to persistent infection, viral load and neoplasic lesion development and 14

rarely found in transitory HPV-infected patients (9, 16, 25, 29). 15

VLP-Seropositivity seems to be a good indicator of the risk of suffering cervical cancer in spite 16

of controversial results regarding its specificity; in fact, an optimized VLP-based ELISA test has 17

been recently reported presenting 93 percent sensitivity and 98.5 percent specificity for 18

discriminating between positive and negative control sera (35). 19

Some L1-protein linear epitopes, exposed for interacting with antibodies (21) have been 20

identified by using peptides, i.e. the 473GLAKPKFTLGKKATPTTS491 peptide is specifically 21

recognized by serum antibodies from 91 percent of those HPV-16 infected patients, 24 percent of 22

children and 66 percent of HPV-16 negative patients (7). The nona-peptide (IHSMNSTIL) 23

discriminates between HR-HPV and LR-HPV infected females (29); peptides 18283 24

55PNNNKILVPKVSGLQYRVFR74, 18294 275LYIKGSGSTANLASSNYFPT294 and 18301 25

414EDTYRFVTSQAIACQKHTPPA434 are specifically recognized by serum antibodies from 26

cervical lesion patient showing 92-97 percent sensitivity and 89-95 percent specificity in 27

recognizing precancerous and cervical cancer patients (35). Taking into account the difference in 28

amino acid variability among the different HPV-types, it is probable that seroreactivity, 29

sensitivity and specificity should be different between peptides 18283, 18294 and 18301 and 30

VLPs-16. The aim of this study was to evaluate agreement in anti-HPV antibody reactivity by 31

ACCEPTED

4

pairing peptides 18294, 18283, 18301 and these peptides with VLP-16 and their ability to 1

discriminate HR-HPV-infected females having cervical lesion from non-infected females 2

presenting normal cytology in a female population having high HPV-infection frequency. 3

ACCEPTED

5

Materials and Methods 1

2

Peptide synthesis. 20-mer peptides, corresponding to HPV type 16 L1 (32), were synthesized by 3

solid-phase multiple peptide system (14). MBHA resin (0.7 meq/g); t-Boc amino-acid, and low-4

high cleavage techniques were used. Peptides were analyzed by MALDI-TOF mass spectrometry 5

and reverse phase-high performance liquid chromatography (RP-HPLC). 6

7

HPV-16 VLP production. HPV16 virus-like particles (VLP-16) were produced in Sf21 8

according to previously described procedure (34). Recombinant baculoviruses encoding HPV16 9

L1-protein were used for infecting SF21 cells at 20 multiplicity of infection. Cells were 10

harvested 4 days post-infection; cytoplasmic and nuclear fractions were separated by 0.5 percent 11

Nonidet P-40 treatment, followed by spinning (10,000 x g, 15 min). CsCl gradient fractions were 12

collected and densities determined by refractometry. Fractions having a density of around 1,272 13

grams/cm3 were pooled in 1x PBS and ultra-centrifuged (4°C, 1h, 130,400 x g). VLP assembly 14

was verified by electron microscopy. 15

16

Population 17

This study, approved by Review Board, scientific and ethical committees from Engativa hospital 18

and Fundacion Instituto de Inmunologia de Colombia, was conducted according to Declaration 19

of Helsinki. 391 females from the urban area, attending Engativa hospital in Bogota, Colombia, 20

between 2005 and 2006 voluntarily accepted participating, signed the informed-consent form and 21

answered a questionnaire enquiring about socioeconomic status, gynecological and obstetric 22

history, current and past sexual behavior and genital track infections. These females were aged 23

between 15 and 68 (mean-age 33.5±12); most of them housewives from families having incomes 24

between U$200 and U$400 per month. Median number of life-time male sex partners was 1.96 25

and most females reported having had two or fewer sex partners. 73 percent of these females 26

were aged 16-20 when they had their first sexual intercourse; 43 percent reported not using any 27

contraception method, 88 percent had had children and 86 percent reported that they had never 28

smoked. Pap smear was performed for each female; all 264 females having abnormal cytology 29

and 23 of normal cytology suffering from other cervical diseases were referred for colposcopy 30

and/or punch biopsy according to gynecological criteria. Blood was obtained by venopuncture 31

ACCEPTED

6

and cervical samples were collected with a cytobrush and kept in 95 percent ethanol for DNA 1

analysis (8, 18, 26). 2

3

ELISA. 12-well ELISA-plates were coated with 5µg/ml L1-peptide 18301 and VLPs and 1.25 4

µg/ml L1-peptide 18294 and 18283, diluted in PBS at 37°C for 1 ½ h. Plates were washed three 5

times with PBS containing 0.05 percent Tween 20 (PBS-T), blocked with 200 µl PBS-T with 6

0.01 percent BSA for 20 min at 37°C, washed; 100 µl serum diluted 1:200 in blocking buffer 7

were added and incubated for 1 h at 37°C. After washing, 100 µl peroxidase-conjugated rabbit 8

anti-human IgG (F(ab´)2 fragment (VECTOR), diluted 1:5,000 was added. The plates were 9

incubated for 1 h at 37°C; then 100 µl peroxidase substrate (TMB KPL) was added to the plates. 10

The reaction was stopped by adding 50 µl 1N sulphuric-acid and read at 450 nm. The assay was 11

performed by triplicate and the variation was lower than 10 percent. Seroreactivity was 12

calculated: (S-N)/(P-N) where S is optical density (OD) of the tested serum obtained with each 13

antigen, P the OD of each serum obtained with anti-human antibody and N the OD obtained with 14

no antigen. Cut-off point for defining seropositive female was 0.15 for peptides and 0.25 for 15

VLPs, according to control serum reactivity. 16

17

HPV-DNA detection by PCR: 18

Cervical samples were washed with PBS, spinning at 12,000 rpm for 10 min, incubated in 100 19

µL lysis buffer (10 mM Tris–HCl [pH 7.9], 0.45 percent Nonidet P-40, 0.45 percent Tween 20 20

and 60 µg/ml Proteinase-K), first at 60°C for 1 hour and then at 95°C for 10 min (24). All 21

samples were then amplified by PCR using human β-globin primers GH20/PC04 (11, 30, 31) 22

and using two different HPV generic primer sets: GP5+/GP6+ and MY09/MY11 (11, 17, 28). 23

PCR using GP5+/GP6+ primers (11) was carried out in a final 20 µl volume with 1x 24

amplification buffer (Bioline), 100 µM each dNTP, 3 mM MgCl2, 1 U Taq Polymerase and 40 25

pmol of each primer. PCR mix was denatured at 94°C for 10 minutes, followed by 40 26

amplification cycles consisting of 94°C for 1 minute, 40°C for 2 minutes and 72°C for 1.5 27

minutes; a final elongation step at 72°C for 7 minutes (11). PCR using MY09/MY11 primers 28

was carried out in a final 20 µl volume PCR mix was denatured at 94°C for 5 minutes, followed 29

by 40 amplification cycles consisting of 94°C for 30 seconds, 45°C for 1 minute and 72°C for 1 30

minute; a final elongation step at 72°C for 7 minutes (28). DNA from Sf21 cells transfected with 31

ACCEPTED

7

a vector containing the HPV-16 L1 gene was used as positive control while ultrapure distilled 1

water (GIBCO) was used as negative control. All PCR products were run on 2 percent agarose 2

gels stained with SYBR safe (Invitrogen) and scanned using the Molecular Imager Fx (BIO 3

RAD). 4

5

Statistical analysis. Group values were presented as percentages, means and standard deviations 6

(SD). Agreement between pairs of tests was assessed using the Lin concordance correlation 7

coefficient (Rho); this coefficient is calculated by combining a measure of accuracy (bias 8

correction factor) and an estimate of precision (Pearson correlation coefficient) (19). STATA 9 9

software was used for statistical analysis. 10

ACCEPTED

8

Results 1

264 abnormal cytology and 127 normal cytology females (in total 391) were included in this 2

study. The 264 abnormal cytology females were classified according to the Bethesda system as 3

having atypical squamous cells of undetermined significance (ASCUS), atypical glandular cells 4

(AGC), low-grade squamous intraepithelial cervical lesion (cyt-LSIL) or high-grade squamous 5

intraepithelial cervical lesion (cyt-HSIL). All the abnormal cytology patients and 23 having 6

normal cytology but suffering from other cervical diseases were referred for colposcopy/biopsy; 7

104 females remained classified as having normal cytology (Normal cytology-no col). It was 8

found no cervical lesions (negative colposcopy) in 159; LSIL (col-LSIL) in 110; HSIL (col-9

HSIL) in 14 and cervical cancers in 4. 10

HPV-DNA detection was performed in these 307 samples from those 391 female cervical 11

samples that amplified beta-globin housekeeping gene. HPV DNA was found in 49 percent, 73 12

percent, 73 percent, 82 percent and 100 percent of the col-Normal cytology, negative 13

colposcopy, col-LSIL, col-HSIL and cervical cancer patients, respectively. 14

Seroreactivity of the 391 female serum samples to peptides 18283, 18294, 18301 and VLPs was 15

determined. The median optical density (OD) was between 0.22 and 0.25 for peptides and 0.41 16

for VLPs with sera from Normal cytology-no col females; this median OD for these four 17

antigens increased to between 0.25 and 0.45 with sera from HPV-DNA negative cervical lesion 18

females and to between 0.38 and 0.52 for with sera from HPV-DNA positive cervical lesion 19

females (Table 1). Seropositivity ranged from 18 to 31 percent for L1-peptides 18283, 18294, 20

18301 and 43 percent for VLPs in Normal cytology-no col females. It was increased to between 21

64 and 73 percent for these three peptides and VLPs in negative colposcopy patients, in col-LSIL 22

and col-HSIL patients. The four cervical cancer patients were seropositive with these four 23

antigens. Seropositivity was generally higher in HPV-DNA positive than in HPV-DNA negative 24

females, except for females having negative colposcopy, and increased along with lesion 25

severity. Seropositivity was 23 percent for 18283 and 18301, 15 percent for 18294, and 45 26

percent for VLPs in HPV-DNA negative females having Normal cytology-no col; this 27

seropositivity increased to 83 percent for peptide 18301, 75 percent for peptide 18283, 18294 28

and VLPs in HPV-DNA positive cervical lesion females (Table 1). Negative colposcopy females 29

(68 percent previously classified as having ASCUS) displayed a higher seropositive percentage 30

in HPV-DNA-negative than in HPV- DNA positive females. 31

ACCEPTED

9

Agreement analysis was performed with the seroreactivity of these four antigens, in the overall 1

population and females classified by cytology report, colposcopy/biopsy diagnosis, HPV-DNA 2

status and age. There was agreement in seropositivity on the overall population between peptide-3

peptide and peptide-VLP pairs, the highest being 18294-18283 pairing (Rho=0.744) and the 4

lowest 18294-VLP pairing (Rho=0.576) (Table 2). 5

In all pairs analyzed seroreactivity agreement was higher in HPV-DNA positive (Rho between 6

0.610 and 0.773) than in HPV-DNA negative females (Rho between 0.441 and 0.679); it was 7

higher for 18283-VLP (Rho=0.760) and 18294-18283 pairs (Rho=0.773) in the HPV-DNA 8

positive group and lower for 18301-18283 and 18294-VLP pairs in the HPV-DNA-negative 9

group (both Rho=0.441) (Table 3). 10

Seroreactivity agreement was higher in HPV-DNA positive Normal cytology, ranging from 11

Rho=0.723 for 18301-VLP to 0.931 for 18294-18283 pairs than in HPV-DNA negative Normal-12

cytology females, particularly for peptide-VLP pairs (Rho between 0.085 and 0.210) (Table 2). 13

There were no differences in agreement between HPV-DNA positive and HPV-DNA negative in 14

ASCUS for all pairs analyzed and in 18294-VLP, 18283-VLP and 18301-18294 pairs in cyt-15

LSIL females. On the contrary, there was higher agreement in HPV-DNA positive (Rho between 16

0.616 and 0.739) than in HPV-DNA negative cyt-LSIL females (Rho between 0.087 and 0.384) 17

for 18301-18283, 18294-18283 and 18301-VLP pairs. The highest agreement in HPV-DNA 18

positive cyt-HSIL patients was found for 18301-VLPs (Rho=0.802), 18294-VLP (Rho=0.843), 19

18283-VLP (Rho=0.833) and 18301-18294 (Rho=0.937) pairs (Table 2). 20

The agreement in Normal cytology-no col group was lower than in Normal-cytology group for 21

peptide-VLP (Rho between 0.071 and 0.616) and 18301-18283 pairing (Rho=0.121). The 22

agreement in negative colposcopy was higher in HPV-DNA positive than HPV-DNA negative 23

females for 18294-VLPs and 18301-18283 pairing. The lowest agreement in negative 24

colposcopy females was for 18294-VLP (Rho=0.223) and 18294-VLP pairing (Rho=0.065), 25

respectively (Table 3). 26

In females having colposcopy-detected cervical lesion the highest seroreactivity agreement was 27

for VLPs-18283 (Rho=0.782) and 18294-18283 (Rho=0.783) pairs and the lowest for 18301-28

18283 (Rho=0.490) (Table 3). Agreement was higher in HPV-DNA positive than in negative 29

females in col-LSIL patients, particularly for 18301-VLP, 18294-VLP and 18301-18283 pairs. 30

The highest agreement in HPV-DNA positive col-LSIL females was between 18283-VLP 31

ACCEPTED

10

(Rho=0.856) and 18294-18283 pairs (Rho=0.817) and in HPV-DNA-negative col-LSIL females 1

was between 18283-VLP pair (Rho=0.843). The highest agreement in HPV-DNA positive col-2

HSIL was for 18294-VLP pair (Rho=0.775) and the lowest for 18301-18283 pair (Rho=0.361 3

and 0.267 in HPV-DNA positive and negative, respectively) (Table 3). Agreement was not 4

determined on ASC-H, AGC, non-infected col-HSIL or on cervical cancer patients because of 5

the low number of cases. 6

Agreement was high in the 18294-18283 pair in most of the analyzed age-groups, irrespective of 7

whether they were HPV-DNA positive or not, but was higher in HPV-DNA positive than in 8

HPV-DNA negative females, particularly in the 25-34 and 35-44 age-groups in most pairs 9

analyzed. The highest agreement in the HPV-DNA positive 35-44 age-group was for 18301-10

18294 (Rho=0.876) and 18294-18283 (Rho Lin 0.898) pairs. Very low agreement was found in 11

HPV-DNA negative females in the 25-34 age-group for 18301-VLP (Rho=0.084), 18294-VLP 12

(Rho=0.184), 18283-VLP (Rho=0.302) and 18301-18283 (Rho=0.300) pairs and in the >44 age-13

group for 18301-18294 (Rho=0.323) and 18294-VLPs (Rho=0.271) pairs (Table 4). 14

ACCEPTED

11

Discussion 1

A reliable serological VLP-based ELISA test for detecting HPV-DNA positive females at risk of 2

developing cervical cancer present several drawbacks; the most relevant being the high 3

frequency of antibodies reacting with VLPs in females having normal cytology (low specificity) 4

and most detected antibody response seems to be HPV-type specific. In principle, both of them 5

could be resolved identifying specific high risk HPV type B-cell epitopes, having low sequence 6

variability among them; moreover, the immunogenecity of these B-cell epitopes should be 7

associated to viral load. In fact, several reports have shown the use of peptides for identifying 8

persistent HPV- DNA positive females (5-7, 29, 33). This work compared the ability of peptides 9

18283, 18294 18301 and VLPs to identify HPV-DNA positive females having cervical lesions. 10

Seropositivity was similar for the four antigens (54 percent-61 percent), in the general 11

population; higher in HPV-DNA positive females than in HPV-DNA-negative and increased 12

with lesion severity; particularly for peptide 18283 which seemed to be more specific for 13

discriminating HPV-DNA positive from HPV-negative females. Peptide 18283 could contain 14

specific strong B-cell epitopes since antibodies against this peptide were detected before lesions 15

appear. There were a higher seropositivity for VLPs than for peptides in HPV-DNA negative 16

normal cytology females; this could have been due to cross-reactivity of antibodies elicited by 17

some low-risk HPV with VLPs (e.g. HPV-6 and 11) (15, 37-39). 18

The seropositivity for peptides 18294 and 18301 was lower than for VLPs and 18283 in non-19

infected Negative cytology-no col females suggesting that B-cell epitopes displayed on peptides 20

18294 and 18301 could be less immunogenic than epitopes present on 18283 and VLPs; also 21

could be that antibodies detected by these peptides remained for a shorter time than antibodies 22

detected by VLPs or 18283. Moreover, peptides 18294 and 18301 presented similar and higher 23

ability for detecting HPV-DNA positive cervical lesion females than VLPs, respectively 24

suggesting that peptides 18294 and 18301 were more specific than VLPs for discriminating 25

HPV-DNA positive cervical lesion females from HPV-DNA negative Normal cytology-no col 26

(Table 1). 27

Comprehensive analysis of all the data revealed that there was agreement in most of the analyzed 28

pairs in HPV-DNA positive females, suggesting that there were similarities in peptide- and VLP-29

antigenicity during natural HPV-infection. Good agreement was found between VLP-18283 and 30

18283-18294 pairs but lower for VLP-18294 pair, suggesting that there are at least two epitope 31

ACCEPTED

12

types on 18283; one presenting antigenicity similar to 18294 and the other similar to VLPs. On 1

the contrary, low agreement in some of the antigen pairs analyzed (especially VLP-18301) could 2

have been due to differences in their antigenicity; VLPs contain epitopes ranging from poorly to 3

highly antigenic (immunodominant) epitopes, but the peptide epitope repertory is restricted; 4

some peptides mainly contain immunodominant epitopes and others contain poorly antigenic 5

epitopes. This is also supported by the fact that immune pressure on these sequences seems to be 6

different according to differences in amino acid sequence variability; VLP-surface contains 7

highly variable sequence; on the contrary, peptide 18301 contains a highly conserved sequence. 8

The lowest agreement among peptide pairs was found between peptides 18301-18283; especially 9

in HPV-DNA negative normal cytology females and the highest agreement between peptides 10

18294-18283. Peptides 18283 (the most variable of these peptides) and 18294 probably contain 11

VLP-surface exposed dominant B-cell epitopes and seemed to be more antigenic on HPV than 12

the most conserved peptide 18301 sequence. 13

Good agreement was found in 18283-VLP pairing in all colposcopy diagnosed cervical lesion 14

females and was very low in Normal cytology-no col females; also, 18294-VLPs displayed low 15

agreement in HPV-DNA negative Normal cytology-no col females which could have been due to 16

lower frequency of anti-18283 and anti-18294 than anti-VLP antibodies. This suggests that these 17

peptides were more sensitive for detecting HPV-associated cervical lesion than cytology in this 18

setting (sensitivity around 70 percent). Good agreement in 18294-18283 pair and low agreement 19

18301-18283 pair in females classified according to colposcopy results suggests similar 20

antigenicity between 18294-18283 and different between 18301-18283 in natural HPV-infection. 21

Negative colposcopy females (65 percent of whom had been classified as having ASCUS) 22

displayed higher frequency of HPV-DNA positivity and seropositivity than Normal cytology-no 23

col females; these negative colposcopy females, having cell abnormalities detected in the Pap 24

smear but no colposcopy detected, could have endocervical or microscopic lesions or their 25

cervical lesions had disappeared when they were examined by colposcopy. Interestingly, 26

seropositivity in this group was higher in HPV-DNA negative than in HPV-DNA positive 27

females for peptide 18301, suggesting this peptide detected antibodies involved in controlling 28

development of detectable cervical lesions and/or HPV-infection. 29

Antibody immune response during infection is roughly parallel to females’ age, taking into 30

account that most females had their first sexual intercourse at the same age and became HPV-31

ACCEPTED

13

DNA positive in the two years following their sexual debut. There was good agreement in the 1

seroreactivity to the various antigens among the different age-groups (age-groups were between 2

84 and 108 females), being slightly lower in most non-infected females. The antibody repertory 3

during natural HPV-infection depends on different epitope types, ranging from 4

immunodominant-epitopes (rapidly inducing antibodies) to low antigenic epitopes which only 5

induce antibodies after a long time exposition to the immune system. Moreover, the 6

concentration of these antibodies could become slowly or rapidly reduced after HPV-infection 7

disappears. VLP-based ELISA measures all antibodies induced by the different epitopes; On the 8

contrary, peptides could measure some of these antibodies, since they contain fewer epitopes; 9

thus displaying different sensitivity and specificity than VLPs for detecting HPV-DNA positive 10

females having cervical lesion. 11

Agreement in HPV-DNA positive <25 year old females was high for 18283-VLP pairing 12

(Rho=0.798) and then decreased as age increased, supporting the idea that some of the VLPs-13

immunodominant epitopes inducing antibodies shortly after infection are contained on peptide 14

18283 as females aged other B-cell epitopes on HPV became immunodominant. It also seemed 15

that antibodies recognized by peptide 18283 decreased soon after HPV-infection disappeared; 16

since frequency of antibodies detected for this peptide was higher in HPV-DNA positive than in 17

non-infected females. Agreement was higher in HPV-DNA positive females in the 35-44 group 18

than in other age-groups; this is the age range displaying high frequency of HPV-associated 19

cervical lesions, infection and anti-HPV antibodies. Agreement between 18283 and 18294 was 20

high in all analyzed age-groups; these peptide sequences are building the top of spikes on VLP-21

model and parts of their sequences are variable, probably due to immune pressure. On the 22

contrary, concordance was low between 18301 and 18283 peptides in most age-groups in spite of 23

being on VLP-surface. 24

In conclusion, peptides 18283 and 18294 proved to be more specific and peptide 18301 was 25

more sensitive than VLPs for detecting HPV-DNA positive cervical lesions and for 26

discriminating HPV-DNA negative normal cytology females. Peptide 18283 is useful for 27

detecting HPV-positive females. According to the results presented above, a serological test can 28

probably be designed for detecting HR-HPV-positive females at risk of developing cervical 29

cancer. 30

ACCEPTED

14

ACKNOWLEDGMENTS 1

2

We especially thank to Fundación Caja Navarra (CAN) and Investigación Solidaria SADAR 3

from Spain for financing this project. We would like also to express our sincere gratitude to 4

Karen Orjuela for HPV PCR amplification, Claudia Reyes and Magnolia Vanegas for peptide 5

synthesis and Jason Garry for translating the manuscript. 6

ACCEPTED

15

References 1

2

1. Baseman JG, K. L. 2005. The epidemiology of human papillomavirus infections. J Clin 3

Virol. 32:S16-24. 4

2. Brentjens, M., Yeung-Yue KA, Lee PC, Tyring SK. 2002. Human papillomavirus: a 5

review. Dermatol Clin 20(2):315-31. 6

3. Carter JJ, K. L. Wipf GC. Christensen ND. Lee SK. Kuypers J. Kiviat N. Galloway 7

DA. 1996. The natural history of human papillomavirus type 16 capsid antibodies among 8

a cohort of university women. J Infect Dis. 174:927-36. 9

4. Carter JJ, K. L., Hughes JP, Lee SK, Kuypers J, Kiviat N, Galloway DA. 2000. 10

Comparison of human papillomavirus types 16, 18, and 6 capsid antibody responses 11

following incident infection. J Infect Dis 181(6):1911-9. 12

5. Carter JJ, W. G. Benki SF. Christensen ND. Galloway DA. 2003. Identification of a 13

human papillomavirus type 16-specific epitope on the C-terminal arm of the major capsid 14

protein L1. J Virol. 77:11625-32. 15

6. Carter JJ, W. G., Madeleine MM, Schwartz SM, Koutsky LA, Galloway DA. 2006. 16

Identification of human papillomavirus type 16 L1 surface loops required for 17

neutralization by human sera. J Virol 80(10):4664-72. 18

7. Cason J, K. P., Best JM, McCance DJ. 1992. Detection of antibodies to a linear epitope 19

on the major coat protein (L1) of human papillomavirus type-16 (HPV-16) in sera from 20

patients with cervical intraepithelial neoplasia and children. Int J Cancer 50(3):349-55. 21

8. Cervantes J, L. C., Hurtado L.Andrade R, Quiroga G. Garcia G, Torricos L. 22

Zegarra L. Vera V.Panoso W. Arteaga R. Segurondo D. Romero F. Dulon 23

ACCEPTED

16

A.Asturizaga D.Hurtado Gomez L. Sonoda S. 2003. Prevalence of human 1

papillomavirus infection in rural villages of the Bolivian Amazon. Rev Inst Med Trop 2

Sao Paulo. 45:131-5. 3

9. Combita AL, B. M., Touze A, Orozco O, Coursaget P. 2002. Serologic response to 4

human oncogenic papillomavirus types 16, 18, 31, 33, 39, 58 and 59 virus-like particles 5

in colombian women with invasive cervical cancer. Int J Cancer 97(6):796-803. 6

10. de Gruijl TD, B. H. Walboomers JM. Schiller JT. Stukart MJ. Groot BS. Chabaud 7

MM. Remmink AJ. Verheijen RH. Helmerhorst TJ. Meijer CJ. Scheper RJ. 1997. 8

Immunoglobulin G responses against human papillomavirus type 16 virus-like particles 9

in a prospective nonintervention cohort study of women with cervical intraepithelial 10

neoplasia. J Natl Cancer Inst. 89:630-8. 11

11. de Roda Husman AM, W. J. van den Brule AJ. Meijer CJ. Snijders PJ. 1995. The 12

use of general primers GP5 and GP6 elongated at their 3' ends with adjacent highly 13

conserved sequences improves human papillomavirus detection by PCR. J Gen 14

Virol.:1057-62. 15

12. Geijersstam V, K. M. Wang Z. Koskela P. Pukkala E. Schiller J. Lehtinen M. 16

Dillner J. 1998. Stability over time of serum antibody levels to human papillomavirus 17

type 16. J Infect Dis. 177:1710-4. 18

13. Ho GY, B. R., Klein S, Kadish AS, Chang CJ, Palan P, Basu J, Tachezy R, Lewis R, 19

Romney S. 1995. Persistent genital human papillomavirus infection as a risk factor for 20

persistent cervical dysplasia. J Natl Cancer Inst 87(18):1365-71. 21

ACCEPTED

17

14. Houghten RA. 1985. General method for the rapid solid-phase synthesis of large 1

numbers of peptides: specificity of antigen-antibody interaction at the level of individual 2

amino acids. Proc Natl Acad Sci U S A. 82:5131-5. 3

15. Hudelist G, M. M., Pischinger KI, Watkins-Riedel T, Singer CF, Kubista E, 4

Czerwenka KF. 2004. Physical state and expression of HPV DNA in benign and 5

dysplastic cervical tissue: different levels of viral integration are correlated with lesion 6

grade. Gynecol Oncol. 92:873-80. 7

16. Karem KL, P. A., Bierl C, Nisenbaum R, Unger E. 2002. Optimization of a human 8

papillomavirus-specific enzyme-linked immunosorbent assay. Clin Diagn Lab Immunol 9

9(3):577-82. 10

17. Karlsen F, K. M. Jenkins A. Pettersen E. Kristensen G. Holm R. Johansson B. 11

Hagmar B. 1996. Use of multiple PCR primer sets for optimal detection of human 12

papillomavirus. J Clin Microbiol. 34:2095-100. 13

18. Lema CH, H. L. Segurondo D. Romero F. Dulon A. Asturizaga D. Panoso W. Garcia 14

G. Fujiyoshi T. Yashiki S. Li HC. Lou H. Cervantes J. Gomez LH. Sonoda S. 2001. 15

Human Papillomavirus Infection among Bolivian Amazonian Women. Asian Pac J 16

Cancer Prev. 2:135-141. 17

19. Lin LI. 1989. A concordance correlation coefficient to evaluate reproducibility. 18

Biometrics 45:255-68. 19

20. Matsumoto K, Y. T., Oki A, Fujii T, Nagata C, Sekiya S, Hoshiai H, Taketani Y, 20

Kanda T, Kawana T, Yoshikawa H. 2006. IgG antibodies to HPV16, 52, 58 and 6 L1-21

capsids and spontaneous regression of cervical intraepithelial neoplasia. Cancer Lett 22

231(2):309-13. 23

ACCEPTED

18

21. Modis Y, T. B., Harrison SC. 2002. Atomic model of the papillomavirus capsid. EMBO 1

J:4754-62.:21(18). 2

22. Moscicki AB, H. N. Shiboski S. Powell K.Jay N. Hanson E.Miller S. Clayton L. 3

Farhat S. Broering J. Darragh T. Palefsky J. 2005. Impact of HPV infection in 4

adolescent populations. J Adolesc Health. 37:S3-9. 5

23. Munoz N. 2000. Human papillomavirus and cancer: the epidemiological evidence. J Clin 6

Virol. 19:1-5. 7

24. Nelson JH, H. G. Edlund K. Evander M. Kjellberg L. Wadell G. Dillner J. 8

Gerasimova T. Coker AL. Pirisi L. Petereit D. Lambert PF. 2000. A novel and rapid 9

PCR-based method for genotyping human papillomaviruses in clinical samples. J Clin 10

Microbiol. 38:688-95. 11

25. Nonnenmacher B, H. N., Kirnbauer R, Shah KV, Munoz N, Bosch FX, de Sanjose S, 12

Viscidi R, Lowy DR, Schiller JT. 1995. Serologic response to human papillomavirus 13

type 16 (HPV-16) virus-like particles in HPV-16 DNA-positive invasive cervical cancer 14

and cervical intraepithelial neoplasia grade III patients and controls from Colombia and 15

Spain. J Infect Dis 172(1):19-24. 16

26. Peto J, G. C. Deacon J. Taylor C. Evans C. Binns W. Haywood M. Elanko N. 17

Coleman D. Yule R. Desai M. 2004. Cervical HPV infection and neoplasia in a large 18

population-based prospective study: the Manchester cohort. Br J Cancer 91:942-53. 19

27. Rajeevan MS, S. D. Nisenbaum R. Lee DR. Vernon SD. Ruffin MT. Horowitz IR. 20

Flowers LC. Kmak D. Tadros T. Birdsong G. Husain M. Srivastava S. Unger ER. 21

2005. Epidemiologic and viral factors associated with cervical neoplasia in HPV-16-22

positive women. Int J Cancer. 115:114-20. 23

ACCEPTED

19

28. Remmerbach TW, B. U. Hemprich A. Chekol M. Kuhndel K. Liebert UG. 2004. 1

PCR detection of human papillomavirus of the mucosa: comparison between MY09/11 2

and GP5+/6+ primer sets. J Clin Virol. 30:302-8. 3

29. Rocha-Zavaleta L, A. J., Mora-Garcia Mde L, Cruz-Talonia F, Hernandez-Montes 4

J, Weiss-Steider B, Ortiz-Navarrete V, Monroy-Garcia A. 2004. Detection of 5

antibodies against a human papillomavirus (HPV) type 16 peptide that differentiate high-6

risk from low-risk HPV-associated low-grade squamous intraepithelial lesions. J Gen 7

Virol 85(Pt 9):2643-50. 8

30. Saiki RK, G. D. Stoffel S. Scharf SJ. Higuchi R. Horn GT. Mullis KB. Erlich HA. 9

1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA 10

polymerase. Science. 239:487-91. 11

31. Saiki RK, S. S. Faloona F. Mullis KB. Horn GT. Erlich HA. Arnheim N. 1985. 12

Enzymatic amplification of beta-globin genomic sequences and restriction site analysis 13

for diagnosis of sickle cell anemia. Science. 230:1350-4. 14

32. Seedorf K., K. G., Dürst M., Suhai S., Röwekamp WG. 1985. Human papillomavirus 15

type 16 DNA sequence. Virology 145:181-185. 16

33. Sharma BK, R. A., Murthy NS. 1996. Prevalence of serum antibodies to synthetic 17

peptides to HPV16 epitopes among Indian women with cervical neoplasia. Eur J Cancer 18

32A(5):872-6. 19

34. Touze A, E. M. S. Sizaret PY. Mougin C. Munoz N. Coursaget P., and. 1998. The L1 20

major capsid protein of human papillomavirus type 16 variants affects yield of virus-like 21

particles produced in an insect cell expression system. J Clin Microbiol. 36:2046-51. 22

ACCEPTED

20

35. Urquiza, M., Guevara, T., Espejo, F., Bravo, MM., Rivera, Z., Patarroyo, ME. 2005. 1

Two L1-peptides are excellent tools for serological detection of HPV-associated cervical 2

carcinoma lesions. Biochem Biophys Res Commun 332(1):224-32. 3

36. Viscidi RP, K. K., Clayman B, Russ K, Shapiro S, Shah KV. 1997. Prevalence of 4

antibodies to human papillomavirus (HPV) type 16 virus-like particles in relation to 5

cervical HPV infection among college women. Clin Diagn Lab Immunol 4(2):122-6. 6

37. Wiley D, M. E. 2006. Human papillomavirus: the burden of infection. Obstet Gynecol 7

Surv. 61(6 Suppl 1):S3-14. 8

38. Wiley D, M. E. 2006. Human papillomavirus: the burden of infection. Obstet Gynecol 9

Surv. 61:S3-14. 10

39. Zuna RE, W. S. Schiffman M. Solomon D. 2006. Comparison of human papillomavirus 11

distribution in cytologic subgroups of low-grade squamous intraepithelial lesion. Cancer 12

108:288-97. 13

ACCEPTED

21

Table 1. Seropositivity in patients diagnosed by colposcopy/biopsy

* This 104 patients were not referred to colposcopy because they displayed normal cytology ** IQ interquartile

*** HPV-DNA positive plus HPV-DNA negative does not equal to the total because only PCR β-globin positive patients were analyzed for HPV-infection

18301 18294 18283 VLPs Status disease/HR-HPV infection (#) # Positive

(%)

Median of

OD IQ range**

# Positive

(%)

Median of

OD IQ range

# Positive

(%)

Median of

OD IQ range

# Positive

(%)

Median of

OD IQ range

col Normal cytology (104) * 23 (22) 0.22 0.17 - 0.30 19 (18) 0.22 0.18 - 0.29 32 (31) 0.25 0.20 - 0.32 45 (43) 0.41 0.30 - 0.54 col Normal cytology / Neg (48) *** 11 (23) 0.21 0.17 - 0.27 7 (15) 0.22 0.16 - 0.29 11 (23) 0.24 0.20 - 0.32 20 (45) 0.43 0.30 - 0.54 col Normal cytology / Pos (47) *** 9 (19) 0.22 0.18 - 0.32 10 (21) 0.22 0.18 - 0.29 16 (34) 0.25 0.20 - 0.35 21 (45) 0.41 0.29 - 0.55 col Negative colposcopy (159) 109 (69) 0.42 0.30 - 0.64 102 (64) 0.40 0.30 - 0.62 111 (70) 0.41 0.29 - 0.58 105 (66) 0.52 0.39 - 0.66 col Negative colposcopy / Neg (32) *** 26 (81) 0.50 0.35 - 0.71 21 (66) 0.42 0.31 - 0.71 22 (69) 0.40 0.31 - 0.69 22 (69) 0.56 0.44 - 0.71 col Negative colposcopy / Pos (86) *** 56 (65) 0.42 0.30 - 0.59 54 (63) 0.41 0.29 - 0.59 63 (73) 0.45 0.29 - 0.63 60 (70) 0.56 0.42 - 0.69 col LSIL (110) 81 (73) 0.49 0.32 - 0.72 76 (69) 0.41 0.30 - 0.66 76 (69) 0.38 0.27 - 0.54 75 (68) 0.51 0.36 - 0.66 col LSIL / Neg (21) *** 14 (66) 0.45 0.30 - 0.69 15 (71) 0.37 0.29 - 0.64 13 (62) 0.33 0.26 - 0.49 10 (48) 0.41 0.34 - 0.58 col LSIL / Pos (58) *** 47 (81) 0.52 0.35 - 0.72 42 (72) 0.42 0.30 - 0.66 41 (71) 0.38 0.29 - 0.59 41 (71) 0.51 0.38 - 0.65 col HSIL (14) 10 (71) 0.44 0.29 - 0.64 9 (64) 0.38 0.33 - 0.66 9 (64) 0.45 0.27 - 0.56 9 (64) 0.49 0.38 - 0.73 col HSIL / Neg (2) *** 1 (50) 0.33 0.18 - 0.48 1 (50) 0.27 0.21 - 0.34 0 (0) 0.25 0.24 - 0.26 0 (0) 0.31 0.24 - 0.38 col HSIL / Pos (9) *** 8 (89) 0.44 0.35 - 0.64 7 (78) 0.41 0.35 - 0.67 8 (89) 0.49 0.37 - 0.56 8 (89) 0.51 0.46 - 0.73 col cancer (4) 4 (100) 0.82 0.55 - 1.12 4 (100) 0.79 0.63 - 0.96 4 (100) 0.78 0.62 - 1.08 4 (100) 0.85 0.63 - 1.12 col cancer / Pos (4) 4 (100) 0.82 0.55 - 1.12 4 (100) 0.79 0.63 - 0.96 4 (100) 0.78 0.62 - 1.08 4 (100) 0.85 0.63 - 1.12 Overall (391) 227 (58) 0.37 0.24 - 0.63 210 (54) 0.35 0.24 - 0.57 232 (59) 0.35 0.25 - 0.51 238 (61) 0.49 0.34 - 0.64 Overall / Neg (103) *** 52 (51) 0.31 0.21 - 0.56 44 (43) 0.29 0.21 - 0.42 46 (45) 0.30 0.23 - 0.41 52 (51) 0.47 0.32 - 0.58 Overall / Pos (204) *** 124 (61) 0.40 0.26 - 0.63 117 (57) 0.37 0.25 - 0.60 132 (65) 0.38 0.26 - 0.54 134 (66) 0.50 0.38 - 0.66 ACCEPTED

22

Table 2. Concordance correlation coefficients in seroreactivity of patients classified by cytology (Bethesda system)

18301 - VLPs 18294 - VLPs 18283 - VLPs 18301 - 18294 18301 - 18283 18294 - 18283 Pap Smear result/HR-

HPV infection Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 %

Normal cytology 0.619 0.52 - 0.72 0.779 0.72 - 0.84 0.804 0.74 - 0.86 0.711 0.63 - 0.79 0.617 0.51 - 0.72 0.883 0.85 - 0.91 Normal cytology / Neg 0.150 -0.07 - 0.37 0.085 - 0.02 - 0.19 0.210 0.04 - 0.37 0.347 0.21 - 0.47 0.169 -0.04 - 0.38 0.583 0.42 - 0.74 Normal cytology / Pos 0.723 0.61 - 0.83 0.880 0.83 - 0.92 0.893 0.84 - 0.94 0.788 0.71 - 0.86 0.776 0.68 - 0.87 0.931 0.90 - 0.95 Abnormal cytology 0.583 0.50 0.65 0.452 0.36 0.54 0.648 0.58 0.71 0.567 0.48 0.64 0.548 0.47 0.62 0.667 0.60 0.72 ASC-US 0.577 0.47 - 0.68 0.472 0.35 - 0.59 0.630 0.55 - 0.71 0.574 0.47 - 0.68 0.408 0.29 - 0.53 0.591 0.50 - 0.68 ASC-US / Neg 0.660 0.49 - 0.82 0.447 0.22 - 0.66 0.730 0.58 - 0.87 0.480 0.23 - 0.72 0.454 0.22 - 0.68 0.602 0.44 - 0.76 ASC-US / Pos 0.516 0.35 - 0.67 0.477 0.30 - 0.64 0.661 0.55 - 0.76 0.578 0.42 - 0.73 0.406 0.22 - 0.58 0.547 0.40 - 0.69 cyt - LSIL 0.566 0.45 - 0.67 0.391 0.25 - 0.52 0.656 0.55 - 0.75 0.523 0.37 - 0.67 0.672 0.56 - 0.78 0.726 0.63 - 0.82 cyt - LSIL / Neg 0.087 -0.44 - 0.62 0.288 - 0.24 - 0.82 0.750 0.53 - 0.96 0.335 -0.05 - 0.72 0.101 -0.38 - 0.58 0.384 - 0.19 - 0.95 cyt - LSIL / Pos 0.616 0.49 - 0.73 0.379 0.21 - 0.54 0.655 0.53 - 0.77 0.485 0.28 - 0.68 0.732 0.61 - 0.85 0.739 0.62 - 0.85 cyt - HSIL 0.759 0.61 - 0.90 0.816 0.65 - 0.97 0.782 0.60 - 0.96 0.859 0.73 - 0.98 0.504 0.26 - 0.75 0.671 0.44 - 0.89 cyt - HSIL / Pos* 0.802 0.64 - 0.95 0.843 0.66 - 1.02 0.833 0.64 - 1.02 0.937 0.86 - 1.01 0.572 0.28 - 0.85 0.640 0.32 - 0.95 Overall 0.607 0.54 - 0.66 0.576 0.52 - 0.64 0.712 0.66 - 0.76 0.624 0.56 - 0.98 0.584 0.52 - 0.64 0.744 0.71 - 0.78

* The analysis was not performed on HSIL / Neg because of the low number of patients.ACCEPTED

23

Table 3. Concordance correlation coefficients in the seroreactivity of patients diagnosed by colposcopy/biopsy 18301 - VLPs 18294 - VLPs 18283 - VLPs 18301 - 18294 18301 - 18283 18294 - 18283 Disease status/HR-HPV

infection Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 %

Negative lesion (Normal

cytology, neg colposcopy) 0.602 0.53 - 0.97 0.481 0.40 - 0.56 0.648 0.58 - 0.71 0.634 0.56 - 0.70 0.633 0.56 - 0.70 0.715 0.66 - 0.76 Normal cytology-no col** 0.118 -0.03 - 0.26 0.148 0.05 - 0.24 0.208 0.10 - 0.31 0.398 0.27 - 0.52 0.121 0.03 - 0.28 0.403 0.28 - 0.58 Normal cytology-no col / neg 0.134 -0.09 - 0.60 0.071 -0.02 - 0.16 0.187 0.02 - 0.35 0.307 0.18 - 0.43 0.135 -0.08 -0.35 0.557 0.39 - 0.71 Normal cytology-no col / pos 0.095 -0.11 - 0.30 0.616 0.47 - 0.75 0.208 0.05 - 0.36 0.509 0.30 - 0.71 0.103 -0.15 - 0.35 0.358 0.11 - 0.60 Negative colposcopy 0.648 0.56 - 0.72 0.645 0.36 - 0.56 0.681 0.60 -0.76 0.601 0.50 - 0.69 0.631 0.54 - 0.72 0.687 0.61 -.075 Negative colposcopy / neg 0.660 0.48 - 0.83 0.065 -0.03 - 0.16 0.690 0.51 - 0.86 0.443 0.16 - 0.72 0.430 0.16 - 0.69 0.599 0.42 - 0.77 Negative colposcopy / pos 0.669 0.56 - 0.77 0.223 0.05 - 0.38 0.683 0.57 - 0.78 0.650 0.53 - 0.76 0.786 0.70 - 0.86 0.724 0.63 - 0.81 Positive lesions (LSIL, HSIL,

cancer) 0.587 0.47 - 0.70 0.693 0.60 - 0.78 0.782 0.71 - 0.84 0.577 0.46 - 0.69 0.490 0.37 - 0.61 0.783 0.72 - 0.84 col - LSIL 0.579 0.45 - 0.70 0.699 0.61 - 0.79 0.780 0.71 - 0.84 0.555 0.43 - 0.68 0.487 0.35 - 0.62 0.787 0.72 - 0.85 col - LSIL / Neg 0.404 0.08 - 0.72 0.558 0.26 - 0.85 0.843 0.73 - 0.95 0.579 0.30 - 0.85 0.299 -0.04 - 0.64 0.648 0.39 - 0.90 col - LSIL / Pos 0.614 0.45 - 0.77 0.780 0.68 - 0.88 0.856 0.79 - 0.92 0.505 0.32 - 0.69 0.548 0.37 - 0.72 0.817 0.74 - 0.89 col - HSIL 0.675 0.40 - 0.95 0.755 0.53 - 0.98 0.773 0.58 - 0.96 0.743 0.53 - 0.96 0.361 -0.02 - 0.70 0.682 0.39 - 0.97 col - HSIL / Pos 0.628 0.25 - 1.01 0.775 0.49 - 1.05 0.683 0.38 - 0.99 0.680 0.34 - 1.02 0.267 -0.19 - 0.73 0.686 0.35 - 1.02 Overall HR-HPV infection Negative 0.505 0.37 - 0.63 0.441 0.30 - 0.57 0.605 0.49 - 0.71 0.548 0.41 - 0.68 0.441 0.30 - 0.57 0.679 0.59 - 0.76 Positive 0.638 0.55 - 0.71 0.610 0.53 - 0.68 0.760 0.71 - 0.81 0.628 0.54 - 0.71 0.678 0.61 - 0.75 0.773 0.72 - 0.82

* The analysis was not performed on HSIL / Neg and cervical cancer patients because of the low number of patients. ** This 104 patients were not referred to colposcopy because they displayed normal cytologyACCEPTED

24

Table 4. Concordance Correlation coefficients in seroreactivity according to the different age groups 18301 - VLPs 18294 - VLPs 18283 - VLPs 18301 - 18294 18301 - 18283 18294 - 18283 Age groups / HR-HPV

infection Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 % Rho IC 95 %

< 25 0.641 0.53 - 0.75 0.576 0.47 - 0.67 0.798 0.73 - 0.86 0.512 0.39 - 0.63 0.503 0.36 - 0.64 0.743 0.68 - 0.80 < 25 / Neg 0.444 0.11 - 0.77 0.615 0.34 - 0.88 0.587 0.35 - 0.82 0.727 0.52 - 0.92 0.441 0.07 - 0.80 0.768 0.61 - 0.91 < 25 / Pos 0.681 0.55 - 0.80 0.576 0.45 - 0.70 0.848 0.78 - 0.91 0.516 0.38 - 0.64 0.628 0.48 - 0.77 0.753 0.68 - 0.82 25 - 34 0.469 0.34 - 0.59 0.495 0.36 - 0.62 0.600 0.49 - 0.70 0.702 0.61 - 0.78 0.648 0.54 - 0.75 0.781 0.70 - 0.85 25 - 34 / Neg 0.084 -0.25 - 0.41 0.184 -0.07 - 0.44 0.302 0.05 - 0.55 0.519 0.29 - 0.74 0.300 0.01 - 0.59 0.830 0.70 - 0.95 25 - 34 / Pos 0.507 0.34 - 0.67 0.615 0.45 - 0.77 0.688 0.57 - 0.80 0.673 0.54 - 0.80 0.747 0.62 - 0.86 0.809 0.72 - 0.89 35 - 44 0.715 0.61 - 0.82 0.661 0.54 - 0.78 0.716 0.62 - 0.80 0.759 0.67 - 0.84 0.601 0.48 - 0.72 0.737 0.64 - 0.82 35 - 44 / Neg 0.665 0.49 - 0.83 0.494 0.23 - 0.75 0.732 0.55 - 0.90 0.549 0.26 - 0.83 0.414 0.14 - 0.68 0.552 0.31 - 0.78 35 - 44 / Pos 0.745 0.61 - 0.88 0.724 0.59 - 0.85 0.703 0.57 - 0.82 0.876 0..81 - 0.94 0.711 0.58 - 0.84 0.898 0.84 - 0.95 > 44 0.603 0.48 - 0.72 0.532 0.40 - 0.65 0.679 0.57 - 0.78 0.614 0.48 - 0.74 0.536 0.40 - 0.66 0.670 0.58 - 0.75 > 44 / Neg 0.496 0.27 - 0.72 0.271 0.03 - 0.51 0.689 0.52 - 0.85 0.323 0.02 - 0.622 0.502 0.24 - 0.76 0.734 0.61 - 0.85 > 44 / Pos 0.606 0.43 - 0.77 0.586 0.40 - 0.76 0.668 0.52 - 0.81 0.723 0.57 - 0.86 0.509 0.33 - 0.68 0.624 0.48 - 0.76

ACCEPTED