Should family physicians test for human papillomavirus infection?: an opposing view.
Several years ago it became possible to isolate HPV from genital tract lesions using molecular hybridization techniques. Initially this process was quite laborious. Today procedures based on molecular hybridization have been modified so that they are easy to do and simple to
Table 1. Summary of Available Human Papillomavirus Detection Tests.
Number of Tests Cost/Test Product Method per Kit ($) Digene In situ 20 15 Enzo In situ 40 10 ONCOR Southern blot 60 7 Virapap Slot blot 50 7 ViraType In situ 20 15
learn. Several kits are now commercially available for hospital laboratories (Table 1). As the clinical manifestations of HPV become publicized, more patients will request the test. In some instances, information about the HPV status of a patient will affect management. Because HPV-related lesions are sexually transmitted, the emotional impact of whether a patient has such an infection may be profound. It is imperative that the physician be as accurate as possible. This discussion will center on the data now available on the clinical utility of testing for HPV. After a brief review of the basics of HPV testing, we will focus on clinical settings where such tests are useful.
HPV Testing: Molecular
Molecular hybridization is the basis for all the important tests available for HPV detection. In brief, the technique involves attaching single-stranded nucleic acid (either DNA or RNA) from a sample, called the target, to a labeled probe. The probe is usually labeled with either radioactive or biotinylated nucleotides that can be detected by autoradiography or standard immunohistochemical metthods.sup.7,8]. The target and probe will attach (hybridize) because of the hydrogen bonds formed between complementary nucleotides (for DNA this would be G-C and A-T). Homology is a reflection of the degree of base-pair matching between the target and probe. If the degree of homology is high (ie, there is substantial G-C and A-T matching), then the target-probe complexes will tent to remain hybridized and resist efforts to separate (denature) it. On the other hand, target-probe complexes with poor homology will disassociate easily. The number and strength of the hydrogen bonds in hybridized DNA with good homology is much greater than for antibody and antigen associations. This translates into a much higher sensitivity and specificity for molecular hybridization when compared with prior methods. The ability of this technique to reliably detect the presence of HPV is exceptional. As few as 10 viruses per 1 million cells of tissue can be detected. There are three tests based on molecular hybridization techniques: filter hybridization, in situ hybridization, and the polymerase chain reaction.
There are two techniques based on filter hybridization (Southern blot and slot blot hybridization). The techniques are called filter hybridization because in each case the target DNa is purified from the sample (the patient's tissue) and then placed on a special type of filter to which it can bind. In slot blot hybridization, the DNA is added directly to the filter. In Southern blot hybridization, electrophoresis of the target DNA precedes transfer of the DNA to the filter. This step is done to remove possible impurities. The filter can then be treated with the labeled probel to determine whether the sample contains the DNA of [interest.sup.9,10]
The major advantage of filter hybridization is its sensitivity. As few as one virus for every 100 cells can be dettected.sup.11]. When HPV infects tissue, it is common to find as many as 10,000 virus particles per cell. Filter hybridization should, therefore, easily identify such an infection. HPV-detection kits that use slot blot and Southern blot hybridization are now commercially available (Table 1).
There are two major disadvantages with filter hybridization. First, the tissue must be fresh or frozen. This criterion can present a problem for clinicians performing colposcopy in out-of-hospital settings. Second, in most cases radioactive probes are needed (ONCOR has recently introduced a nonradioactive filter hybridization kit) (Table 1). A wait of up to 1 week for results is involved along with the attendant problems in disposing of radioactive wastes. Each of these problems can be avoided using in situ hybridization.
With in situ hybridization the proble is applied directly to a tissue section. The probe is a "viral cocktail" containing specific segments o HPV DNA. Attached to each HPV-specific segment is a protein "label" (biotin). In the presence of HPV, the biotinylated DNA will concentrate in the nucleus of an infected cell. The tissue is then treated with a colorizing agent. The HPV-infected nuclei will turn a brilliant blue. Infected cells can be easily recognized under a standard light microscope.
The major advantage of in situ hybridization is that fixed, paraffin-embedded tissue, available from any patient who has had a biopsy, can be used. Testing, therefore, may be considered after the biopsy report is completed. Another important advantage is that once can use nonradioactive (biotin-labeled) [probes.sup.11,12]. Detection of a biotin-labeled target-probe complex is a simple technique available in most hospital laboratories. The test can be completed in a matter of a few hours. In short, in situ hybridization is an easy, rapid test for HPV that can also be used to tell which specific HPV type has infected the tissue. Why, thenm even bother to use filter hybridization? The answer is contained in the detection threshold for in situ hybridization analysis. About 10 to 20 viruses must be present in the cell for the in situ test to be possitive.sup.11,12]. Although there is no problem in low-grade intraepithelial lesions, occult infection by HPV has far fewer particles. In situ hybridization is usually negative in such infections, whereas filter hybridization usually detects the virrus.sup.11]
The Polymerase Chain Reaction
The purpose of the polymerease chain reaction is to amplify the amount of HPV DNA present in a patient's tissue sample by making copies of the HPV DNA. The procedure is done in the following steps: short, HPV-specific fragments are "manufactured" through a process called DNA sequencing. The resulting fragments are called primers. When a cocktail of the primers is added to a tissue sample, the primers will bind to homologous areas of the viral DNA. The result will be a strand of patient HPV DNA with multiple short HPV-manufactured segments attached. Gaps will exist between each of the fragments. An HPV-specific enzyme called Taq polymerase is then added. This enzyme fills in the gaps by synthesizing the missing segment of DNA. Next, the primer-target complex is denatured and the process restarted. With each cycle the amount of HPV DNA doubles. After 30 such cycles there are generally over a million HPV particles, which can then be detected by a characteristic electrophoretic [pattern.sup.13,14].
The major advantage and disadvantage of this test are related. Because viral DNA is amplified, it is possible to detect very small amounts of HPV; as few as 10 viruses per 1 million cells. Even a small amount of contaminating HPV DNA, however, can result in a false-positive result. The technician must take great care that contamination does not occur. another advantage of polymerase chain reaction is that paraffin-embedded tissue can be used. Further, there is no need for radioactive probes. The main advantage of polymarase chain reaction is that if the test is negative, one can be certain that the morphological changes in the tissue in question are not related to HPV infection. Currently the polymerase chain reaction test should be reserved for research laboratories.
When Clinical Testing for HPV is
Determination of HPV Type
A great deal of attention has focused on the different HPV types. It is often stated that HPV types 6 and 11 are "good" (not associated with cervical carcinoma) and that HPV types 16 and 18 are "bad" (associated with an increased cancer [risk).sup.4,15-17]. So should lesiions associated with HPVB 6 and 11 be left alone? We say no. As anyone who regularly treats vulvar or penile low-grade lesions (condlomata) knows, these lesions have a high rate of recurrence. As many as 70% will [recur.sup.18]. They can become large and are often distressful to the patient. The only possible use of specific HPV typing might be to determine which women carry HPV 18. This type, which occurs in less than 1% of the population, is detected in the majority of adenocarcinomas of the cervix. Adenocarcinomas are responsible for 5% to 7% of cervical cancers. As many as 50% of women with adenocarcinomas are missed on routine Papanicolaou smear [testing.sup.19]. Currently there is no information to suggests that a major change in the screening for adenocarcinomas of the cervix is warranted.
Detection of Occult Infection
Several investigators have shown that about 10% of women and mean will have HPV DNA detected from cervical or penile swabs, respectively, but will have either normal Papanicolaou smears or no visible [lesions.sup.20-22]. Should these people undergo more extensive testing? At this stage there is no evidence to suggest that in the absence of a visible lesion (an acetowhite area found during colposcopy), these women are more likely to have cervical intraepithelial lesiions (CIL). Indeed, some evidence suggests that these women may be less likely to develop [CIL.sup.22]. The emotional impact of telling a woman that she has a potentially cancer-causing virus in the cervix even though there is no evidence of disease is of course profound. Follow-up data do not exist at this time to justify alarming 10% of the population. The Papanicolaou smear continues to be the most effective screening tool in the detection of cervical intraepithelial disease. It has yet to be demonstrated whether HPV testing will increase the detection of cervical intraepithelial lesions relative to Papanicolaou smears in women who have normal Papanicolaou smears over time.
When Clinical Testing for
HPV is Indicated
Detection of HPV DNA in Women with an
In certain settings testing for HPV DNA can provide very useful information to determine who is at high risk for CIL or who does not have an HPV-related lesion. A common clinical problem for the colposcopist involves a woman who has an abnormal Papanicolaou smear (often squamous atypia) and an acetowhite lesion on colposcopy which on biopsy is negative for condyloma or CIL. What follow-up is indicated for such a patient? Repeat the colposcopy? Treat the cervix anyway? And what should the patient be told? It has recently been shown that HPV detection in this setting is an excellent way to determine who really is at high risk for CIL and who is not.  We propose HPV testing for all women with atypical Papanicolaou smears who have acetowhite lesions with biopsies nondiagnostic for CIL. Patients who are virus positive should undergo colposcopy again in 4 to 6 months, whereas those who are virus negative may be best followed by repeat Papinicolaou smears.
As an Aid to the Histological Examination of
Although in its classic form the histologica changes of HPV infection are easily diagnosed, it is not infrequent that the histological changes, though suggestive of a lesion, may be equivocal. Pathologists may try to be helpful by signing out such cases as "borderline, early, or equivocal for condyloma," but this may be confusing to the physician and especially the patient, who wonders, "Do I or don't I have this sexually transmitted disease?" Here, HPV testing by in situ hybridization can be very helpful. Over 955 of low-grade vulvar, penile, and cervical intraepithelial lesions will be positive by this methodology, whereas in normal tissue the in situ test is invariably negative.  It has been shown that in lesions clinically suggestive of low-grade infections (condylomata) where the histological changes are equivocal, HPV DNA can be detected by this test at a rate of 2% for the cervix and about 10% for the vulva and penis. HPV-positive cases are best considered low-grade lesions. HPV-negative cases are most likely mimics for which another condition, such as candida infection or chronic irritation, should be explored. These conclusions can be supported by testing with the highly snesitive polymerase chain reaction technique.
Recent advances in the field of molecular biology have expanded the knowledge of HPV manifestations. We have attempted to separate those circumstances when HPV testing may be useful in patient management and when it may not. Routine screening for HPV has not yet been shown to be useful. One must weigh informing 10% of patients that they have a potentially oncogenic virus against the ability of a repeat Papanicolaou smear to detect the lesion. On the other hand, HPV detection in the setting of an abnormal Papanicolaou smear and an acetowhite lesion that on biopsy lacks the histologic features of CIL does predict who is at high risk for CIL. HPV typing should not affect clinical management, and we do not recommend it. HPV detection by in situ hybridization is a very useful adjunct to histological analysis in genital tract lesions that are clinically suggestive of an HPV lesion but where the histological analysis is equivocal. This ability to distinguish true HPV lesions from its mimics is often crucial to the patient, given the implications of having a sexually transmitted disease.
 Oriel JD. Condyloma acuminata as a sexually transmitted disease. Dermatol Clin 1983; 1:93-102.
 Nuovo GJ, Crum CP, Silvestein SJ. Papilloma virus infection of the uterine cervix. Microbiol Pathogenesis 1987; 3:71-8.
 Lorincz AT, Temple GF, Kurman RJ, et al. Oncogenic association of specific human papillomavirus types with cervical neoplasia. J Natl Cancer Inst 1987: 79:671-7.
 Reid R, Greenburg M, Jenson AB, et al. Sexually transmitted papillomavirus infections I. The anatomic distribution and pathologic grade of neoplastic lesions associated with different viral types. Am J Obster Gynecol 1987; 156:212-22.
 LaVecchia C, Franceschi S, DeCarli A, et al. Sexual factors, venereal disease, and the risk of intraepithelial and invasive cervical neoplasia. Cancer 1986; 58:935-41.
 Newkirk GR, Granath BD. Teaching colposcopy and androscopy in family practioce residencies. J Fam Pract 1990; 31:171-8.
 Lorincz AT. Detection of human papillomavirus infection by nucleic acid hybridization. Obstet Gynecol Clin North Am 1987; 14:451-69.
 Nuovo GJ, Richart RM. Human papillomavirus: a review. In: Yearbook of obstetrics and gynecology, 1989. Chicago; Yearbook Medical, 1989.
 Nuovo GJ. A comparison of slot blot, Southern blot and in situ hybridization analyses for human papillomavirus DNA in genital tract lesions. Obstet Gynecol 1989; 74:673-7.
 Southern EM. Detection of specific sequences among DNA fragments selected by gel electrophoresis. J Mol Biol 1975; 98:503-17.
 Nuovo GJ. A comparison of different methodologies (biotin based and 35S based) for the detection of human papillomavirus DNA. Lab Invest 1989; 61:471-6.
 Walboomers JMM, Melchers WJG, Mullink H, et al. Sensitivity of in situ detection with biotinlylated probes of human papillomavirus type 16 DNA in frozen tissue sections of squamous cell carcinoma of the cervix. Am J Pathol 1988; 131:587-94.
 Nuovo GJ. Human papillomavirus DNA in genital tract lesions histologically negative for condylomata; analysis by in situ, Southern blot hybridization and the polymerase chain reaction. Am J Surg Pathol 1990; 14:643-51.
 Shibata D, Fu YS, Gupta JW, et al. Detection of human papillomavirus in normal and dysplastic tissue by the polymerase chain reaction. Lab Invest 1988; 59:555-9.
 Crum CP, Ikenberg H, Richart RM, et al. Human papillomavirus type 16 and early cervical neoplasia. N Engl J Med 1984; 310:880-3.
 Lorincz AT, Temple GF, Kurman RJ, et al. Oncogenic association of specific human papillomavirus types with cervical neoplasia. J Natl Cancer Inst 1987; 79:671-7.
 Nuovo GJ, Friedman D. In situ hybridization analysis of HPV DNA segregation patterns in lesions of the female genital tract. Gynecol Oncol 1990; 36:256-62.
 Ferenczy A, Mitao M, Nagai N, et al. Latent papillomavirus and recurring genital warts. N Engl J Med 1985; 313:784-8.
 Walker J, Bloss JD, Liao S, et al. Human papillomvirus genotype as a prognostic indicator in carcinoma of the uterine cervix. Obstet Gynecol 1989: 74:781-5.
 DeVilliers EM, Schneider A, Miklaw H, et al. Human papilloma-virus infections in women with and without abnormal cervical cytology. Lancet 1987; 1:703-6.
 Grussendorf-Conen EI, DeVilliers EM, Gissman L. Human papillomavirus genomes in penile smears of healthy men. Lancet 1986; 1:1092.
 Nuovo GJ, Cottral S, Richart RM. Occult infection of the uterine cervix by human papillomavirus in postmenopausal women. Am J Obstet Gynecol 1989; 160-340-4.
 Nuovo GJ, Hochman H, Eliezri YD, et al. Detection of human papillomvarus DNA in penile lesions histologically negative for condylomata: analysis by in situ hybridization and the polymerase chain reaction. Am J Surg Pathol 1990; 14:829-36.
|Printer friendly Cite/link Email Feedback|
|Author:||Nuovo, Gerard J.; Nuovo, James|
|Publication:||Journal of Family Practice|
|Date:||Feb 1, 1991|
|Previous Article:||Should family physicians test for human papillomavirus infection?: an affirmative view.|
|Next Article:||Chronic pain in primary care: identification and management of psychosocial factors.|