Secondary Logo

Journal Logo

Three-Year Risk of Cervical Precancer and Cancer After the Detection of Low-Risk Human Papillomavirus Genotypes Targeted by a Commercial Test

Castle, Philip E. PhD, MPH; Hunt, William C. MA; Langsfeld, Erika BS; Wheeler, Cosette M. PhD

doi: 10.1097/AOG.0000000000000013
Contents: Original Research
Free

OBJECTIVE: To investigate the risk of cervical precancer and cancer associated with detection of human papillomavirus (HPV) 6, 11, and 42.

METHODS: We used data from the New Mexico Human Papillomavirus Pap Registry. A stratified sample of 59,644 residual cervical cytology specimens from a population of 379,000 underwent HPV genotyping. We measured the 3-year cumulative incidence of cervical intraepithelial neoplasia grade 2 or more severe (CIN 2+) and grade 3 or more severe (CIN 3+) after detection of single HPV 6, 11, or 42 infections or single or multiple infections of HPV 6, 11, or 42 (“HPV 6, 11, 42, or combinations”; n=581).

RESULTS: The overall prevalence of a single infection of HPV 6, 11, or 42 was 0.8% (95% confidence interval [CI] 0.7–0.9%). The 3-year risks of CIN 2+ and CIN 3+ after HPV 6, 11, 42, or combinations infections (n=581) were 0.4% (CI 0.1–0.7%) for CIN 2+ and 0.0% for CIN 3+ (nota bene, no CI was calculable because no events occurred), respectively. By comparison, the 3-year risks of CIN 2+ and CIN 3+ after a negative HPV result (n=27,522) were 0.2% (95% CI 0.1–0.2%) and 0.1% (95% CI 0.0–0.1%), respectively.

CONCLUSION: Detection of HPV 6, 11, 42, or combinations in the absence of high-risk HPV types does not identify women at increased 3-year risk for cervical precancer. Testing for HPV 6, 11, 42, or combinations of those types should be discontinued because it has no proven benefit to patients.

LEVEL OF EVIDENCE: II

Human papillomavirus 6, 11, and 42 testing does not meaningfully predict the development of cervical precancer and therefore should not be included in cervical cancer screening.

Albert Einstein College of Medicine, New York, New York; and the University of New Mexico Health Sciences Center, Departments of Pathology and Obstetrics and Gynecology, House of Prevention Epidemiology, Albuquerque, New Mexico.

Corresponding author: Cosette M. Wheeler, PhD, Department of Pathology and Department of Obstetrics and Gynecology, House of Prevention Epidemiology, MSC 02-1670, 1816 Sigma Chi Road NE, Albuquerque, NM 87131; e-mail: cwheeler@salud.unm.edu.

* For a list of members of The New Mexico HPV Pap Registry Steering Committee, see Appendix 1 online at http://links.lww.com/AOG/A451.

Philip E. Castle is a Visiting Professor at the Albert Einstein College of Medicine, New York, New York.

Supported by the National Institute of Allergy And Infectious Diseases of the National Institutes of Health under Award Number U19AI084081 to Cosette M. Wheeler.

Financial Disclosure Dr. Castle has received human papillomavirus (HPV) tests and testing for research at a reduced or no cost from Qiagen and Roche. He has been personally compensated for serving as a member on a Data and Safety Monitoring Board for Merck and has been compensated as a consultant for BD, Cepheid, Roche, and Gen-Probe. Dr. Wheeler has received funds through the University of New Mexico for grants and cooperative agreements from the National Institutes of Health related to cervical screening, funds from GSK for HPV vaccine studies, and reimbursements for travel related to publication activities and equipment and reagents from Roche Molecular Systems for HPV genotyping. The other authors did not report any potential conflicts of interest.

Roche Molecular Systems, Inc provided HPV LINEAR ARRAY Genotyping Test and equipment to automate HPV genotyping assays.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

High-risk human papillomavirus (HPV) DNA testing has been increasingly integrated into cervical cancer screening and management. Current screening guidelines recommend HPV cotesting every 5 years in women aged 30–65 years as the preferred alternative to cytology every 3 years1,2 and high-risk HPV DNA testing is recommended for the management of atypical squamous cells of undetermined significance (ASC-US) cytology (“reflex” testing), for follow-up of women who have undergone colposcopy, and for surveillance after treatment of precancerous lesions.3,4

Despite national guidelines for the use of high-risk HPV DNA testing and warnings against inappropriate use, there is evidence that clinicians are ordering HPV testing in excess of its recommended use.5,6 This includes ordering cotesting more frequently than recommended6 for patients who already have cytologic evidence of serious abnormalities and should be referred to colposcopy regardless of the HPV test result8 and testing for low-risk HPV types 6, 11, 42, 43, and 44 using Probe A.5 The latter potentially adds cost and identifies more women as HPV-positive while apparently offering little benefit because these types rarely or never cause cervical cancer.

However, it is theoretically possible that the detection of these types could identify women at higher risk of subsequently acquiring high-risk HPV genotypes that do cause cervical cancer because they are all transmitted identically through sexual contact.7 We therefore conducted a population-based analysis of the 3-year risks for women to develop either cervical intraepithelial neoplasia (CIN) 2+ or CIN 3+ associated with infection with HPV 6, 11, and 42. HPV 43 and 44 results were not available.

Back to Top | Article Outline

MATERIALS AND METHODS

The New Mexico HPV Pap Registry (NMHPVPR) is located at the University of New Mexico and acts as a designee of the New Mexico Department of Health. The NMHPVPR operates under New Mexico Administrative Code 7.4.3.12, which specifies the list of Notifiable Diseases and Conditions for the state of New Mexico. In 2006, with the intention of monitoring the effect of HPV vaccination, New Mexico Administrative Code 7.4.3.12 specified that laboratories must report to the New Mexico HPV Pap Registry all cervical cytology, cervical pathology, and HPV tests performed on New Mexico residents. NMAC 7.4.3.12 was updated in 2009 to include vulvar and vaginal pathology (http://nmhealth.org/ERD/healthdata/documents/NotifiableDiseasesConditions022912final.pdf).

During the 17-month period of December 2007 through April 2009, approximately 379,000 cervical cytology tests were reported to the New Mexico HPV Pap Registry by nine in-state and seven out-of-state clinical laboratories.8 All available liquid cervical cytology specimens were collected from seven of the nine in-state laboratories, which accounted for 79% of all cervical cytology tests done during this period. Specimens were randomly selected from each of these seven laboratories for HPV genotyping by Linear Array (LA; Roche Molecular Systems, Pleasanton, CA, USA) within four strata defined by the age of the woman (30 years or younger compared with older than 30 years) and by the cytologic result on the laboratory report (negative compared with abnormal). Randomization was carried out by using a random number generator to select, without replacement, a target sampling proportion of all cytology samples within each stratum. Target sampling proportions varied by strata: 45% of negative specimens in women 30 years of or younger, 8% of negative specimens in women older than 30 years, and 100% of abnormal specimens regardless of the woman's age. Ultimately, a total of 59,644 specimens on 54,848 women were retrieved and successfully genotyped for HPV. Before HPV genotyping, specimens were deidentified by the use of randomly assigned study-specific identifiers. The University of New Mexico human research review committee approved this study.

HPV genotyping was done as previously described.8–10 Briefly, DNA was purified from 500-microliter aliquots from vigorously mixed residual liquid cytologic specimens of SurePath or ThinPrep using a Cobas X421 robot. Fifty microliters of purified DNA was transferred to a tube with 50 microliters of HPV LINEAR ARRAY Genotyping test mastermix, and the mixture was amplified by polymerase chain reaction as specified by the manufacturer. Controls for contamination and assay sensitivity were included in each 96-well assay.

The HPV genotyping by LINEAR ARRAY is based on PGMY 09/11 consensus polymerase chain reaction primers and a prototype Line Blot assay, which have been previously reported in detail.9–11 Using the Roche HPV LINEAR ARRAY Genotyping Test detection kit, hybridizations were automated using Tecan ProfiBlot-48 robots. The Roche HPV LINEAR ARRAY Genotyping Test detects 13 high-risk (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) and 24 low-risk 6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66–73, 81, 82, 82v, 83, 84, and 89) HPV types. The presence of HPV 52 is not determined directly by a type-specific probe but through inference as previously described.14 Two independent readers interpreted the presence of HPV genotypes using a reference template provided by the manufacturer. Any discrepancies were identified by a custom computer application applied to the data input and were adjudicated by a third review.

Cytologic results were classified according to the 2001 Bethesda System12: high-grade squamous intraepithelial lesion (HSIL), atypical squamous cells cannot rule out HSIL (ASC-H), atypical glandular cells (AGC), low-grade squamous intraepithelial lesion (LSIL), ASC-US, and negative for intraepithelial lesion or malignancy. There were a few examples of use of non-Bethesda System terminology, which were classified as follows8: 1) LSIL cannot rule out HSIL as ASC-H; 2) CIN grade 1 as LSIL; CIN 2, CIN 3, carcinoma in situ (CIS), or possible carcinoma as HSIL; and 4) atypical squamous and glandular cells of undetermined significance as AGC. Cytologic results were generally dichotomized as normal (negative) and abnormal (ASC-US or more severe).

Data on cytologic interpretations and histologic diagnoses were collected from medical records. Both were based on local readings and no attempt was made to review them centrally or undertake quality assurance activities.

Diagnoses for cervical precancer (CIN 2, CIN 3) and cancer were based on the results of all cervical biopsies, including excisional biopsies and all endocervical curettages from the date of the index cytology for a period of 3 years. An outcome of CIN 2+ was defined as a result of CIN 2, HSIL, CIN 2–3, CIN 3, CIS, adenocarcinoma in situ (AIS), or carcinoma. An outcome of CIN 3+ was defined as a result of CIN 2–3, CIN 3, CIS, AIS, or carcinoma.

For women with more than one cytology specimen, only the chronologically earliest was considered. In addition, only specimens from screening tests were used, which was defined as a cervical cytology test with no previous cytology collected within 300 days (n=47,541). Estimation of the 3-year risk of CIN 2+ and CIN 3+ was restricted to the 27,522 HPV-negative specimens, and specimens positive for a single infection by HPV 6 (n=235), by HPV 11 (n=29), or by HPV 42 (n=307) (ie, 571 single HPV genotype infections by 6, 11, or 42) and specimens positive for one (n=571) or more (n=10) of HPV 6, 11, or 42 and no other HPV type (“HPV 6, 11, 42, or combinations”; n=581). Additional analyses considered the risk of CIN 2+ and CIN 3+ for women positive for one or more high-risk HPV types (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) with or without HPV 6, 11, 42, or combinations. All CIs and P values are exact except for those reported for “All Ages,” “All Cytology,” and “All,” which were computed using sample survey techniques appropriate for a stratified random sample with unequal sampling fractions. SAS 9.3 procedures FREQ and SURVEYFREQ were used to compute exact and weighted proportions and tests of significance, respectively. Sample weights were computed as the inverse of the sampling fraction and variances were computed by the Taylor series linearization method. Weighted CIs were based on normal approximations.

Back to Top | Article Outline

RESULTS

Overall, 2.7% of women tested positive for an infection by HPV 6, 11, 42, or combinations (95% CI 2.6–2.9%). The majority of these infections were concurrent with infections of other HPV types and single HPV genotype infections by HPV 6, 11, or 42 were found in only 0.8% (95% CI 0.7–0.9%) of women. This analysis focused on the subgroup of women infected with HPV 6, 11, 42, or combinations of these types to estimate the excess risk of cervical precancer and cancer attributable to HPV 6, 11, or 42 infections above testing HPV-negative.

The prevalence of single infections of HPV 6, 11, or 42 was less than 2% in all sampling subgroups defined by cytologic status (negative compared with abnormal) and age (30 years or younger compared with older than 30 years) (Table 1). The prevalence was significantly lower in women aged older than 30 years with negative cytology compared with all other subgroups (P<.001). The prevalence among those with abnormal cytology in either age group was only slightly greater than the prevalence among those aged 30 years or younger with negative cytology.

Table 1

Table 1

Table 2 shows the distribution of cytologic results concurrent with infections by HPV 6, 11, 42, or combinations compared with those with no HPV detected. HPV 6, 11, 42, positives were associated with significantly more abnormal squamous cell cytology (P<.001), specifically ASC-US and LSIL.

Table 2

Table 2

The 3-year risks of CIN 2+ and CIN 3+ after single HPV 6, 11, or 42 infections and HPV 6, 11, 42, or combination infection compared with HPV-negative results are shown in Table 3. No women with single HPV 6, 11, or 42 infections and negative cytology were diagnosed with CIN 2+ or CIN 3+ and very few were after a negative HPV result. A small number of women with HPV 6, 11, 42, or combinations and abnormal cytology was diagnosed with CIN 2+ but none were diagnosed with CIN 3+. Accounting for sampling fractions, the overall 3-year risks after HPV 6, 11, 42, or combinations infections (n=581) were 0.4% (95% CI 0.1–0.7) for CIN 2+ and 0.0% for CIN 3+ (nota bene, no CI was calculable because no events occurred). By comparison, the 3-year risks after a negative HPV result (n=27,522) were 0.2% (95% CI 0.1–0.2) for CIN 2+ and 0.1% (95% CI 0.0–0.1) for CIN 3+. Overall, the difference in risk of CIN 2+ between HPV-negative and women positive for HPV 6, 11, 42, or combinations was small but statistically significant (P=.02). Among women with abnormal cytology, the risk for CIN 2+ but not CIN 3+ was greater for HPV 6, 11, or 42–positive women than HPV-negative women (3.1% compared with 0.6%, respectively; P<.001), most notably in women older than 30 years of age (4.5% compared with 0.4%, respectively; P<.001).

Table 3

Table 3

As an ancillary analysis, we examined whether detection of HPV 6, 11, 42, or combinations increased the risk of CIN 2+ and CIN 3+ among those with a concurrent high-risk HPV infection (Table 4). In most strata, testing positive for these low-risk HPV types did not increase risk of CIN 2+ or CIN 3+. Among women with abnormal cytology, detection of HPV 6, 11, 42, or combinations was associated with a significant decrease in risk of CIN 2+ and CIN 3+.

Table 4

Table 4

Back to Top | Article Outline

DISCUSSION

Although CIN 2 is the threshold for excisional treatment, the increased risk associated with HPV 6, 11, 42, or combinations is minor on an absolute risk scale and is not actionable, ie, a risk that is sufficiently greater to take more aggressive clinical action such as referring women positive for any of these HPV types to colposcopy.13 In addition, it is well known that CIN 2 is an equivocal, regressive high-grade diagnosis and is a diagnosis that sometimes can be caused by noncarcinogenic HPV genotypes with no risk of progression to cancer.14–17 These additional cases of CIN 2 possibly attributable to HPV 6, 11, 42, or combinations of infections may not represent real cancer risk but would be treated by an excisional procedure in routine clinical practice as standard of care. Notably, these cases were found without the use of low-risk HPV testing. The introduction of p16INK4A immunohistochemistry to clarify which CIN 2 are truly precancer should reduce the treatment of CIN 2 caused by low-risk HPV types like HPV 6, 11, and 42 (as well as HPV 43 and 44) that is not destined to become cancer.18

Another limitation is that there was only 3 years of follow-up for this cohort and small sampling fractions of the Pap-negative population, which likely resulted in a small number of measurable events occurring in these low-risk populations and lead to imprecise estimates of risk. Moreover, because an abnormal Pap test result was required for referral to colposcopy and biopsy, and women positive for these HPV types were more likely to have an abnormal Pap test result than the HPV-negative women, there is potentially a bias in the disease ascertainment that would lead to higher risk estimates in the HPV 6, 11, 42-positive women. However, the percentage of women with at least one follow-up cytology, cervical biopsy, or endocervical curettage in the 3-year period was not significantly different between those who were HPV negative and those who tested positive for a single infection of HPV 6, 11, or 42 (70.7% compared with 72.3%; P=.6). Nevertheless, longer-term follow-up will address both of these limitations.

Because little disease occurs in HPV-negative women, where the necessary cause of cervical cancer and its immediate precursors are absent,19 and because HPV 6, 11, 42, or combinations contribute very little to the risk of CIN 2 and not at all to that of CIN 3, we had insufficient power to detect small risk differences between the two groups. A post hoc power calculation found that there was 80% power to detect a risk difference between HPV 6, 11, 42, or combination positive and HPV-negative results of 1.7, 4.4, 2.1, and 2.6 cases of CIN 2+ per 100 women in the 30 years of age or younger and negative cytology, 30 years of age or younger and abnormal cytology, older than 30 years of age and negative cytology, and older than 30 years of age and abnormal cytology groups, respectively. Thus, smaller but clinically unimportant differences would likely have been missed in this analysis.

The risks were likely underestimated because some women in this cohort moved out of state during follow-up and then developed cervical disease. Although we have no way of knowing the outmigration specifically for women in this cohort, the estimates of outmigration from the state of New Mexico decreased with increasing age and varied by observational year of the cohort but were never greater than 8% for a given year and age group (data not shown). In an analysis comparing all women with HPV 6, 11, 42, or combinations infections to those with adequate follow-up (ie, adequate follow-up was defined for high-grade Pap samples as cervical biopsy or endocervical curettage, for LSIL or high-risk HPV-positive ASC-US Pap samples as biopsy, or endocervical curettage or repeat negative cytology, ie, negative for high-risk HPV, and for all other Pap samples as biopsy or endocervical curettage or repeat negative cytology) (Appendix 2, available online at http://links.lww.com/AOG/A452), there was no meaningful difference in the risk estimates. Because HPV infection is asymptomatic, we do not anticipate a differential bias in our estimates of risk as a result of outmigration and therefore our conclusions regarding similar risk between HPV negative and HPV 6, 11, or 42-positive populations are likely correct.

We conclude that risk of cervical precancer within 3 years of detection of HPV 6, 11, 42, or combinations is low and not appreciably different than for HPV-negative women, and therefore testing for these types should not be included in cervical cancer prevention programs. Testing for HPV 6, 11, or 42 or combinations of these types may result in some women being labeled as HPV-positive, which can have a psychosocial effect such as stigmatization from testing positive for a sexually transmitted infection or anxiety from being “screen positive.” Importantly, testing for HPV 6, 11, or 42 could lead to the interpretation of a positive “screening” result that might result in referral to colposcopy and a diagnosis of CIN 2, which is the threshold of treatment in the United States.20 It is now recognized that excisional treatment, currently the predominant method for treating CIN 2 and CIN 3, may increase the risk of negative reproductive outcomes, including preterm delivery and perinatal mortality.21 Importantly, HPV prevalence is greatest in young reproductive-aged women, those who might experience the greatest harms of low-risk HPV testing. Therefore, testing for low-risk HPV 6, 11, 42, or combinations of these types (or related types such as HPV 43 and 44) should be strongly discouraged to reduce the potential harms resulting from unnecessary testing and to conserve health care resources needed for other services that do benefit to patients. Commercial tests that target these HPV genotypes should be discontinued.

Back to Top | Article Outline

REFERENCES

1. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al.. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin 2012;62:147–72.
2. Moyer VA. Screening for cervical cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2012;156:880–91.
3. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D. 2006 consensus guidelines for the management of women with cervical intraepithelial neoplasia or adenocarcinoma in situ. J Low Genit Tract Dis 2007;11:223–39.
4. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D. 2006 consensus guidelines for the management of women with abnormal cervical screening tests (published erratum appears in J Low Genit Tract Dis 2008;12:255). J Low Genit Tract Dis 2007;11:201–22.
5. Lee JW, Berkowitz Z, Saraiya M. Low-risk human papillomavirus testing and other nonrecommended human papillomavirus testing practices among U.S. health care providers. Obstet Gynecol 2011;118:4–13.
6. Saraiya M, Berkowitz Z, Yabroff KR, Wideroff L, Kobrin S, Benard V. Cervical cancer screening with both human papillomavirus and Papanicolaou testing vs Papanicolaou testing alone: what screening intervals are physicians recommending? Arch Intern Med 2010;170:977–85.
7. Schiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S. Human papillomavirus and cervical cancer. Lancet 2007;370:890–907.
8. Wheeler CM, Hunt WC, Cuzick J, Langsfeld E, Pearse A, Montoya GD, et al.. A population-based study of human papillomavirus genotype prevalence in the United States: baseline measures prior to mass human papillomavirus vaccination. Int J Cancer 2012;132:198–207.
9. Peyton CL, Gravitt PE, Hunt WC, Hundley RS, Zhao M, Apple RJ, et al.. Determinants of genital human papillomavirus detection in a US population. J Infect Dis 2001;183:1554–64.
10. Gravitt PE, Peyton CL, Alessi TQ, Wheeler CM, Coutlee F, Hildesheim A, et al.. Improved amplification of genital human papillomaviruses. J Clin Microbiol 2000;38:357–61.
11. Castle PE, Gravitt PE, Solomon D, Wheeler CM, Schiffman M. Comparison of linear array and line blot assay for detection of human papillomavirus and diagnosis of cervical precancer and cancer in the atypical squamous cell of undetermined significance and low-grade squamous intraepithelial lesion triage study. J Clin Microbiol 2008;46:109–17.
12. Solomon D, Davey D, Kurman R, Moriarty A, O'Connor D, Prey M, et al.. The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA 2002;287:2114–9.
13. Castle PE, Sideri M, Jeronimo J, Solomon D, Schiffman M. Risk assessment to guide the prevention of cervical cancer. J Low Genit Tract Dis 2008;12:1–7.
14. Castle PE, Stoler MH, Solomon D, Schiffman M. The relationship of community biopsy-diagnosed cervical intraepithelial neoplasia grade 2 to the quality control pathology-reviewed diagnoses: an ALTS report. Am J Clin Pathol 2007;127:805–15.
15. Trimble CL, Piantadosi S, Gravitt P, Ronnett B, Pizer E, Elko A, et al.. Spontaneous regression of high-grade cervical dysplasia: effects of human papillomavirus type and HLA phenotype. Clin Cancer Res 2005;11:4717–23.
16. Castle PE, Schiffman M, Wheeler CM, Solomon D. Evidence for frequent regression of cervical intraepithelial neoplasia-grade 2. Obstet Gynecol 2009;113:18–25.
17. Belinson S, Smith JS, Myers E, Olshan A, Belinson J, Pretorius R, et al.. Descriptive evidence that risk profiles for cervical intraepithelial neoplasia 1, 2, and 3 are unique. Cancer Epidemiol Biomarkers Prev 2008;17:2350–5.
18. Darragh TM, Colgan TJ, Cox JT, Heller DS, Henry MR, Luff RD, et al.. The lower anogenital squamous terminology standardization project for HPV-associated lesions: background and consensus recommendations from the College of American Pathologists and the American Society for colposcopy and cervical pathology. Arch Pathol Lab Med 2012;136:1266–97.
19. Dillner J, Rebolj M, Birembaut P, Petry KU, Szarewski A, Munk C, et al.. Long term predictive values of cytology and human papillomavirus testing in cervical cancer screening: joint European cohort study. BMJ 2008;337:a1754.
20. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D; 2006 American Society for Colposcopy and Cervical Pathology-sponsored Consensus Conference. 2006 consensus guidelines for the management of women with cervical intraepithelial neoplasia or adenocarcinoma in situ. Am J Obstet Gynecol 2007;197:340–5.
21. Kyrgiou M, Koliopoulos G, Martin-Hirsch P, Arbyn M, Prendiville W, Paraskevaidis E. Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and meta-analysis. Lancet 2006;367:489–98.
Figure

Figure

Supplemental Digital Content

Back to Top | Article Outline
© 2014 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.