Cervical cancer in young women in Taiwan: prognosis is independent of papillomavirus or tumor cell type

The objective of this study was to address the hypotheses that younger patients with cervical cancer have a uniquely worse clinical outcome and/or are more likely to have adverse tumor cell types or specific human papillomaviruses (HPV). Cases of stage Ib-IIa cervical cancer among women 35 years of age or younger (82) and over 35 (54) were analyzed and compared with respect to the following: (1) histologic type (squamous vs nonsquamous), (2) human papillomavirus (HPV) type via polymerase chain reaction, and (3) clinical parameters, including tumor size, nodal metastases, and recurrence/persistence. Patients 35 years of age or younger had a survival similar (71.2% vs 72.4%) to that of older women from the same institution. In the younger group, outcome was not correlated with the presence or absence of HPV or HPV type. Nonsquamous carcinomas, including adenocarcinoma and small cell carcinoma, were strongly associated with HPV18, were more prevalent in the younger group, and had a slightly higher risk of recurrence/persistence; however, these differences were not significant and 71% of the recurrences were squamous cell carcinomas. Thus, in young Taiwanese women with stage Ib-IIa cervical cancer, the majority of deaths cannot be attributed to a specific HPV type or unique tumor morphology.

Human papillomavirus-associated cervical intraepithelial neoplasia following lesbian sex

BACKGROUND

Less than 3% of lesbians develop cervical dysplasia, with increasing risk correlating with previous heterosexual activity. Because they are not currently sexually active with men, many lesbians do not perceive themselves to be at risk for developing dysplasia and do not obtain regular Papanicolaou smears. There are no standard recommendations for Papanicolaou smear intervals for lesbians.

CASE

A 36-year-old, nonsmoking woman had a Papanicolaou smear history of a high-grade squamous intraepithelial lesion of the cervix, which was confirmed by biopsy and successfully treated by laser ablation. Human papillomavirus type 16 was identified in the cervical biopsy by polymerase chain reaction amplification and restriction fragment polymorphism analysis. The patient gave a clear history of having had sexual activity only with women.

CONCLUSION

Regular Papanicolaou testing should be recommended to all lesbians, regardless of type of sexual activity. Papanicolaou testing intervals should be determined using standards similar to those used for heterosexual women: annually after onset of sexual activity or after age 18, and possibly less often after three normal smears at her physician's discretion. An extensive number of sexual partners, current smoking, and prior dysplasia may influence the physician to advise continued yearly Papanicolaou testing for lesbians, similar to advice given to heterosexual patients.

A comparison of early (E7) and late (L1) primer-mediated amplification of papillomaviral DNA in cervical neoplasia

Studies have demonstrated that in 50-90% of cervical carcinomas, human papillomavirus (HPV) DNA sequences are covalently bound (integrated) to the chromosomal DNA. All evidence shows that when integration takes place disruption of the viral genome occurs downstream to the E7 open reading frame, which is invariably retained in functional form. Theoretically, this phenomenon could result in loss of HPV sequences (L1) not critical to the presumed tumourigenic functions and if so, could influence primer selection for HPV DNA detection in these tumours. A series of cervical carcinomas (CA, n = 133), adenocarcinomas in situ (ACIS, n = 28) and high grade squamous intraepithelial lesions (HSIL, n = 30) were analysed for HPV nucleic acids using primers designed to amplify the E7 and L1 regions. Primer sizes and sensitivities were adjusted to produce equivalent amplification efficiency. Of 191 cases studied, 134 (70%) scored positive for HPV16 or 18 with either the E7 or L1 primer set. Of these, 116 (87%) were positive with both primer pairs. There were no significant differences in proportions of HPV 16/18 positives or lesion types scoring positive exclusively with the E7 vs the L1 primer sets. However, HPV18 associated, E7 positive carcinomas were slightly less likely than HPV16 associated carcinomas to be L1 positive (P = 0.07). Although a high proportion of HPV16 and particularly HPV18 positive carcinomas have been associated with exclusively integrated HPV DNA, there is little evidence that this influences detection sensitivity with E7 vs L1 primers. The combination of E7 and L1 primers provided the maximum sensitivity in this study, with 18 of 134 cases scoring positive with only one primer set.

Association between cervical neoplasia and apoptosis as detected by in situ nuclear labeling

Invasive cervical cancer is thought to arise from the progression of precancerous lesions. How these lesions proceed from precancers to cancer remains unknown. Data regarding other tissues indicate that altered programmed cell death (apoptosis), in addition to cellular proliferation, is associated with the development of neoplasia. Therefore, in order to better understand the development of cervical neoplasia, we investigated the rate of apoptosis in cervical precancer and cancer. Archival cervical samples from normal epithelium (n = 11), low-grade squamous intraepithelial lesions (LSIL, n = 11), high-grade squamous intraepithelial lesions (HSIL, n = 10), and squamous cancers (n = 10) were evaluated for chromatin cleavage, a hallmark of programmed cell death. We used in situ end-labeling of DNA strand breaks by terminal deoxynucleotidyltransferase incorporation of biotinylated deoxyuridine to 3'-OH ends of DNA, identified by a nickel-avidin-peroxidase. The apoptotic index (sum of apoptotic bodies divided by the total nuclei times 100) significantly decreased (P < 0.001) as the degree of neoplasia increased: 3.5% (+/-0.4) in normal cervical epithelium (4.8 +/- 0.4) in LSIL, 1.4% (+/-0.4) in HSIL, and 0.4% (+/-0.1) in squamous cancers. Compared to normal epithelium, the total cell number per 10 mm2 increased significantly (P < 0.001): 124 (+/-12) in normal epithelium, 162 (+/-9.7) in LSIL, 315 (+/-31) in HSIL, and 413 (+/-32) in squamous cancers. We conclude that increasing cervical atypicality is associated with a decrease in apoptosis. We hypothesize from our data that one factor involved in the progression of neoplasia in the uterine cervix is a decrease in the rate of normal cellular deletion.

Absence of papillomavirus DNA in normal tissue adjacent to most cervical intraepithelial neoplasms

OBJECTIVE

To determine if human papillomavirus (HPV) DNA is present in the normal mucosa adjacent to cervical intraepithelial neoplasia (CIN).

METHODS

Serial sections of 28 CIN lesions were studied. Lesional and normal epithelia and stroma were microdissected; the DNA was extracted and amplified by polymerase chain reaction (PCR) using primers designed to amplify both HPV late (L1) and human beta-globin sequences. Human papillomavirus was typed by restriction fragment length polymorphism analysis following digestion of PCR products.

RESULTS

Twenty-five of 28 (89%) lesional epithelia scored positive for HPV nucleic acids. In' four of 25 (16%) HPV-positives, the normal squamous epithelium scored positive for HPV nucleic acids, two of which (8%) also scored positive in the stroma. Repeat microdissection and PCR analysis of three of these cases was performed and all were negative in both normal epithelium and stroma, suggesting laboratory contamination.

CONCLUSION

Human papillomavirus nucleic acids are present uncommonly in normal-appearing squamous epithelium adjacent to CIN. This does not exclude occult infection in the natural history of CIN but indicates that when lesions develop, occult infection is not normally maintained in the normal mucosa. This is consistent with the low recurrence rates following ablation as well as low indices of HPV positivity in normal cervices during follow-up. This finding should be taken into consideration when counseling patients and is relevant to the concept of HPV testing during follow-up after cone biopsy.

Bethesda classification of cervicovaginal smears: reproducibility and viral correlates

Fifty-five cervicovaginal smears from women with squamous intraepithelial lesions (SILs) were independently evaluated on two separate occasions by four cytopathologists using a binary classification system (the Bethesda system). Smears were categorized as low-grade (LSIL) or high-grade (HSIL) using previously published criteria. All women had subsequent cervical biopsies containing human papillomavirus (HPV) DNA amplified with the polymerase chain reaction and typed by restriction fragment polymorphism analysis. Three or more observers agreed on classification in 49 of 55 cases (87%); unanimous diagnoses were rendered in 31 cases (56%). Interobserver and intraobserver reproducibility ranged from fair to near-excellent (kappa values 0.40 to 0.63; 0.63 to 0.74, respectively). HPV types included HPV 16 (27%), 18 (7%), 31 (9%), 35 (4%), 39 (4%), 6 (10%), 11 (2%), novel types (30%), and multiple types (4%). High-risk HPV types (16, 18, 31, 35, and 39) were significantly associated (P = .03) with consensus HSIL diagnoses (agreement of three or more observers). This was primarily because of the strong association of HPV 16 with HSIL (P = .001). After excluding HPV 16, the other high-risk HPV types (18, 31, 35, and 39) were no longer significantly associated with consensus HSIL diagnoses (P > .5). Conversely, LSIL diagnoses were significantly associated with non-high-risk HPV types (all HPV types except 16, 18, 31, 35, and 39; P = .006). Binary cytological classification of cervicovaginal SILs is reproducible among cytopathologists. Such classification correlates well with most low-risk HPV types and with the prototypic high-risk HPV 16 but not with other high-risk HPV types.

Pseudobowenoid change of the vulva: a histologic variant of untreated condylata acuminatum

Vulvar condylomata are distinguished from vulvar intraepithelial neoplasia by orderly maturation, a low mitotic index, and the absence of nuclear atypia in the lower epithelial layers. Podophyllin therapy might produce changes in condylomata mimicking vulvar intraepithelial neoplasia but can be distinguished, particularly with a recent history of podophyllin use. This report describes five patients with vulvar condylomata that morphologically resemble both podophyllin effect and vulvar intraepithelial neoplasia. The mean age was 30.5 years. All lesions exhibited superficial keratinocytes in various stages of apoptosis, with a spectrum of nuclear degenerative changes characterized by absence of the nuclear membrane and dispersed nuclear chromatin. The characteristic finding was a perichromatin halo delineated by a rim of dense cytoplasm, beyond which a second zone of uniform clearing was observed, similar to changes noted in vulvar intraepithelial neoplasia. There was an absence, however, of either nuclear atypia in the lower cell layers or apoptosis and mitotic arrest observed in vulvar intraepithelial neoplasia and podophyllin therapy. Two (50%) of four lesions tested positive for human papillomavirus nuclei acids, one by polymerase chain reaction and the other by DNA-DNA in situ hybridization. No common human papillomavirus types were found. Pseudobowenoid change might represents a distinct type of vulvar condyloma, perhaps related to an unusual human papillomavirus.

Viral and histopathologic correlates of MN and MIB-1 expression in cervical intraepithelial neoplasia

A recently studied tumor antigen, MN, has been associated with cervical carcinomas and cervical intraepithelial neoplasms (CIN), suggesting that it may serve as a marker for cervical cancer or cancer risk. To determine if expression of the MN antigen paralleled parameters reflecting viral or biological events in precursor epithelium, MN expression was correlated with MIB-1 expression, morphological phenotype, and human papillomavirus (HPV) distribution and type in a series of CINs. Seventy-three percent, 62% and 83% of CIN I, II, and III, respectively, were MN antigen positive. The proportion of neoplastic cells immunoreactive for MN did not correlate with the CIN grade or with HPV types stratified by their association with cancer. Evaluation of serial sections showed no correlation between the frequency of MN antigen staining, the proportion of MIB-1 immunoreactive cells, or the proportion of HPV positive cells detected by in situ hybridization (ISH). CINs associated with prototypical high risk (HPV 16) types exhibited increased immunostaining for the MIB-1 antigen and were more often classified as HSIL in contrast to the other types. Thus, although MN expression previously has been associated strongly with squamous carcinoma, it did not emerge as a specific marker for either cancer-associated HPV types or high grade CIN. CIN I lesions associated with low and high risk HPV types were not distinguished by MIB-1 expression and viral replication. This emphasizes the interrelationship between vegetative viral functions (including viral replication) and morphological phenotype, irrespective of HPV type.

Prevalence of human papillomaviruses (HPV) in benign and malignant tumors of the upper respiratory tract

The role for human papillomavirus (HPV) in head and neck tumors is not established. To evaluate the possible role of HPV in head and neck neoplasms, 22 cases of laryngeal squamous papilloma (LP), 32 cases of laryngeal squamous cell carcinoma (LSCC), 40 cases of nasal inverted papilloma (NIP), 14 cases of nasal squamous cell carcinoma (NSCC), and 40 cases of nasopharyngeal carcinoma (NPC) were evaluated for the presence of HPV DNA using the polymerase chain reaction (PCR) with two sets of primers in separate reactions: HPV-L1 consensus primers, HPV16 and HPV18 E7 primers for HPV nucleic acid detection. Restriction fragment length polymorphism of L1 PCR product was used for typing of HPV. Overall, HPV DNA was detected in 18 of 22 cases (81.8%) of LP, 3 of 32 cases (9.4%) of LSCC, 17 of 40 cases (42.5%) of NIP, 3 of 14 cases of NSCC (21.4%), and none of 40 cases of NPC. HPV6 was found more frequently in LP and HPV11 in NIP (P < 0.001). In the three HPV positive LSCCs, two cases had previous LP and were HPV6 and HPV11 positive, as were the prior papillomas. One other case was HPV18 positive. Only HPV16 was found in the NSCC patients. There was no significant difference in the index of HPV positivity between the NSCC cases associated with (16.7%) and without NIP (25.6%). Our results suggest that HPV plays a role in the development of both LP and NIP, and that similar viral types (i.e., HPV6 and HPV11) may exhibit relative differences in their tissue specificity. HPV appears to be of limited importance as a co-factor in LSCC and NPC lesions, indicating differences in the pathogenesis between papillomas and carcinomas in the upper respiratory tract.

Twin pregnancies with complete hydatidiform mole and coexisting fetus: use of fluorescent in situ hybridization to evaluate placental X- and Y-chromosomal content

Twin pregnancies with a complete hydatidiform mole (CHM) and a coexisting fetus have an aggressive postevacuation behavior; it is, therefore, important to differentiate these cases from partial hydatidiform moles that rarely require treatment for late sequelae. It has been presumed that twin pregnancies with a CHM and a coexistent fetus are dizygotic gestations, but this has not been confirmed in most cases. The authors investigated the sex chromosomal constitution of paraffin-embedded, formalin-fixed placental tissues in nine pregnancies histopathologically diagnosed as twin gestations with CHM and coexisting fetus, using fluorescent in situ hybridization (FISH) with X- and Y-chromosomal probes. Normal placental tissues showed an even sex distribution--four cases: X signal only, presumably female; four cases: X and Y signals, presumably male. In contrast, all molar tissues of these same pregnancies hybridized with the X-chromosomal probe only. Thus, in four of nine cases, gender differences (ie, different sex chromosome content) in molar villi (X chromosome only, cytogenetic female) versus normal villi (both sex chromosomes, cytogenetic male) confirmed the histopathological diagnosis of dizygotic twinning; a strict relationship between villous morphology (molar vs normal) and chromosomal gender was observed in each instance. This study illustrates that use of FISH on paraffin-embedded tissues can retrospectively establish dizygotic twinning in this unusual type of molar gestation.

Prevalence and histologic significance of cervical human papillomavirus DNA detected in women at low and high risk for cervical neoplasia

OBJECTIVES

To determine and compare the prevalence and histologic significance of human papillomavirus (HPV) nucleic acids in cervical specimens from women at low (routine hysterectomy) and high (suspicion of cervical neoplasia) risk for cervical neoplasia.

METHODS

Cervical brushings were taken from the cervices of hysterectomy and conization or loop electrical excision specimens and analyzed for HPV nucleic acids by polymerase chain reaction and restriction fragment length polymorphism analysis. Histopathology was confirmed by review of reports or, for HPV-positive results, re-review of the histopathology. Statistical analysis used Student t test or Fisher exact test.

RESULTS

Four hundred seventeen and 43 low- and high-risk cervices, respectively, were studied. Statistically significant differences were observed in the index of HPV positivity between the low- and high-risk groups (1.7 versus 42%, P < .001) and the proportion of HPV being cancer-associated HPV types (14 versus 78%, P = .005). None of the 417 cervices from low-risk women contained HPV 16. In the high-risk group, histologically confirmed cervical intraepithelial neoplasia lesion was statistically more likely to be associated with HPV (59 versus 13%, P = .005).

CONCLUSION

Cervices from routine, low-risk hysterectomies in predominately middle-aged women have an extremely low index of cancer-associated HPVs. Considering the strong association of HPV with histologically proven disease, prospective studies exploring the relationship of cancer-associated HPVs to neoplasia in middle-aged women merit consideration.

Postmenopausal squamous atypia: a spectrum including "pseudo-koilocytosis"

In addition to typical atrophy, the postmenopausal cervix may exhibit a spectrum of epithelial and cellular alterations, including prominent perinuclear halos, nuclear hyperchromasia, variation in nuclear size, and multinucleation. It has not been determined whether such changes, termed postmenopausal squamous atypia (PSA), represent age-related epithelial disturbances or human papillomavirus (HPV)-related low-grade squamous intraepithelial lesions (condyloma). We surveyed 30 cervical biopsies from 26 women over the age of 50 that contained cytoplasmic halos and a spectrum of nuclear alterations, either alone or in association with atrophy. Twenty-three exhibited epithelium with 2- to 3-fold nuclear enlargement, and 18 had moderate or marked nuclear staining intensity. Eleven had a maximum of one or more multinucleated cells in a high-power field. Despite the nuclear alterations, none of the biopsies were positive for HPV by PCR analysis. This is in contrast to 104 of 141 low- and high-grade squamous intraepithelial lesions from a wide age range of women analyzed in the same manner (P = .000006). Features distinguishing PSA from HPV-associated low-grade squamous intraepithelial lesions (condyloma) included less variation in nuclear size and staining intensity, more finely and evenly distributed nuclear chromatin, and greater uniformity of perinuclear halos in PSA. In menopausal or postmenopausal women, PSA should be excluded when considering the diagnosis of a low-grade squamous intraepithelial lesion, specifically if the diagnosis rests on the interpretation of koilocytotic atypia.

Vulvar intraepithelial neoplasia: age, morphological phenotype, papillomavirus DNA, and coexisting invasive carcinoma

Recent studies suggest that subsets of vulvar intraepithelial neoplasia (VIN) may be distinguished based on morphological presentation, the presence or absence of human papillomavirus (HPV) nucleic acids, and patient age. We analyzed 65 VIN lesions, including 15 with associated squamous cell carcinoma, to determine the relationship between pathological parameters associated with common types of VIN (multinucleation, koilocytosis, verruco-papillary morphology, diffuse atypia), rarer variants (differentiation, basal atypia), patient age, and papillomavirus nucleic acids. For all lesions higher mean ages were observed in patients with lesions that were associated with cancer and with well differentiated VIN variants with basal atypia only. A strong negative correlation with HPV nucleic acids was observed for differentiated variants with basal atypia (P = .002). In the common or "classic" VIN group patients with lesions with koilocytotic atypia, multinucleation, and verruco-papillary morphology were generally younger. However, no parameter or group of parameters defined a subset of patients with a significantly lower mean age or lesions with a higher index of HPV nucleic acids. Three of six lesions of lichen sclerosus (LS)-associated VIN, including one involving invasive carcinoma in elderly women, contained HPV nucleic acids; all three lesions exhibited the features of classic VIN. The finding of HPV across a broad age range suggests that this virus may play a role in vulvar neoplasia at any point in life. The direct demonstration of HPV nucleic acids within three LS-associated VINs is intriguing because it links two distinct risk factors to the same neoplasm.

Cytologic/histologic correlation for quality control in cervicovaginal cytology. Experience with 1,582 paired cases

For quality-control purposes, federal regulations require cytology laboratories to compare Papanicolaou smear and cervical biopsy reports, if available, and determine the cause of any discrepancies. The authors reviewed 56,497 cervicovaginal smears, of which 1,582 (2.8%) had a subsequent cervical biopsy or endocervical curettage within 2 months. A total of 175 discrepancies (11%) were identifed, and biopsies and smears from these cases were reviewed at a weekly conference. In the majority of cases, the diagnosis of the smear and biopsy was confirmed on review, and the discrepancy was attributed to sampling error (n = 159; 91%). Six cases (3.4% of all discrepant cases) were errors in cytologic diagnosis. Five of these were interpretation errors, and one case was a screening error. There were 10 errors in the evaluation of cervical biopsies (5.7% of all discrepant cases): five biopsies were undercalled, and five were overdiagnosed as a squamous intraepithelial lesion. The results of testing for human papillomavirus DNA by in situ hybridization were helpful in arbitrating some diagnositic disagreements.

Cervical intraepithelial neoplasia

Cervical intraepithelial neoplasia (CIN) has been traditionally defined as a continuum of intraepithelial squamous abnormalities which exhibit nuclear atypia in all epithelial layers and possess some potential for progression to invasive carcinoma if not removed. Efforts to subdivide this spectrum into categories of low and high cancer risk have been based previously on the strong association between CIN III (carcinoma in situ) and subsequent invasive carcinoma. However, in practice, this distinction has been discouraged because CIN I and II may be associated with CIN III and a small proportion may progress to invasive carcinoma. As human papillomaviruses (HPV) have emerged as potential markers for subdividing precursor lesions, so-called "high-risk" HPV types have been associated with all grades of CIN, whereas "low-risk" HPV types have segregated primarily in lesions closely resembling condylomata. The place of condyloma in the spectrum of CIN, as well as the precise definition of CIN I, has been controversial. Some authors distinguish condyloma from CIN I and other use similar criteria for the diagnosis of both. Currently, the trend among pathologists and cytopathologists is to classify CIN I as a process either identical to or closely resembling condyloma (low-grade), and CIN II and III as lesions falling within the spectrum of CIN as classically described (high-grade). As new etiologic perspectives (HPV), classifications (Bethesda) and outpatient managements (LEEP) evolve, morphologic definitions of CIN will remain important to patient care, particularly if management decisions are based on nuances of histologic or cytologic grade. When using cervical lesion morphology as an endpoint in chemoprevention studies, investigators must understand that "morphologic progression" of CIN may not be synonymous with biologic progression, that discrepancies between HPV type and morphology exist, and that cytology and histology provide variable, and at times conflicting, information.

Multinucleated atypia of the vulva. Report of a distinct entity not associated with human papillomavirus

The vulvar mucosa often demonstrates epithelial nuclear atypia in association with reactive and inflammatory conditions. These nuclear changes are usually mild and can be readily distinguished from vulvar intraepithelial neoplasia (VIN) and human papillomavirus (HPV)-related lesions. In a recent survey of vulvar biopsies in reproductive-aged women, we identified 12 cases of an unusual pattern of atypia associated with multinucleated epithelial cells but lacking the usual stigmata of reactive changes, condyloma, or VIN. The average age of the patients with multinucleated atypia of the vulva (MAV) was 37 years. The multinucleated cells were commonly in the lower to middle epithelial layers and contained between two and 10 nuclei, often with prominent nucleoli. In contrast to condyloma and VIN, there was no surface atypia, and the multinucleated cells lacked hyperchromasia, irregularity, or variation in nuclear size. No significant inflammation or identifiable infectious process was present, and none of the patients had received any topical treatment other than mild corticosteroids. Two of the patients had a history of VIN at a noncontiguous site. None of the 12 cases contained HPV DNA by either in situ hybridization or polymerase chain reaction (PCR) restriction fragment length polymorphism (RFLP) analysis. This is in contrast to 49 of 65 women with VIN and 21 of 26 with condyloma who had HPV demonstrable by the PCR method (p < 0.00001). Immunoperoxidase stains for herpes types I and II were also negative in all the cases. Thus, MAV appears to be a distinct entity occurring in relatively young women; when it is not associated with condyloma or VIN, MAV is not related to HPV. As the morphologic features may overlap with both condyloma and VIN, it is important that MAV not be confused with these lesions or vice versa. It is not known whether MAV is a risk factor for VIN, represents an exaggerated reactive response, or is an entity with a distinct origin.

Nondiagnostic squamous atypia of the cervix (atypical squamous epithelium of undetermined significance): histologic and molecular correlates

The histologic criteria for reactive/reparative lesions and precursor lesions (squamous intraepithelial lesions; SIL) of the cervix are established, but a proportion of cervical biopsies contains squamous epithelial alterations that do not fall into either category (nondiagnostic squamous atypia or atypical squamous epithelium of undetermined significance). This study examined (a) the degree to which this diagnosis could be made consistently between two experienced pathologists and (b) its relationship to papillomavirus nucleic acids. One author selected 37 cervical biopsies with a spectrum of cellular changes occurring in the context of inflammatory or reactive epithelial changes, and each case was reviewed by two pathologists independently and classified as reactive, nondiagnostic atypia, and diagnostic (SIL). Concordance between two observers was highest for a diagnosis of SIL (kappa = 0.68), followed by reparative/reactive (kappa = 0.48). Concordance for a designation as nondiagnostic squamous atypia was fair (kappa = 0.39). Seven of 37 (19%) cases were HPV positive, including 2/7, 5/19, and 0/11 cases designated by at least one observer as SIL, nondiagnostic squamous atypia, or neither (i.e., both reactive), respectively. The findings support a category of nondiagnostic squamous atypia but emphasize the value of a second observer and possibly, HPV DNA testing, in resolving such cases. Similarities in HPV positivity between nondiagnostic atypia and SIL in this study may reflect problems of both criteria selection and consistent application of these criteria in a setting of reactive/reparative changes.

Histologic correlates of vulvar human papillomavirus infection in children and young adults

Two clinically important issues in the treatment of vulvar wartlike lesions are the histologic criteria for the recognition of human papilloma virus (HPV)-related lesions and the exclusion of lesions derived from cutaneous rather than genital HPV types. We analyzed a series of 70 biopsies from the vulva or distal vagina of 57 children and 13 young adults for HPV nucleic acids by polymerase chain reaction (PCR) amplification and typed the isolates following isotope labeling and restriction digestion (restriction fragment length polymorphism [RFLP] analysis). Lesions were classified as condyloma, suggestive of condyloma (papillary/verrucous architecture without koilocytotic atypia), or nonspecific epithelial alterations. Three observers independently agreed on the presence of papillary/verrucous architecture and koilocytotic atypia with a high degree of concordance (kappa = 0.74 and 0.71, respectively). By RFLP analysis, 77% of the lesions diagnosed as condyloma and 68% of those diagnosed as suggestive of condyloma contained HPV nucleic acids versus 9% of the nonspecific group. The HPV types identified were HPV 6 (67%), HPV 11 (17%), HPV 16 (3%), and unknown types (14%). No cutaneous HPV types were identified. Three patients with unknown HPV types had a history of sexual abuse, implying a genital source. These findings indicate that verrucopapillary external genital lesions, as defined in this report, are likely to be associated with HPV and that the vast majority contain genital HPV types irrespective of histologic presentation.

Analysis of HPV-positive and -negative vulvar carcinomas for alterations in c-myc, Ha-, Ki-, and N-ras genes

Mutation or overexpression of certain host genes, including c-myc, Ha-ras, and Ki-ras, have been associated with genital squamous neoplasia, specifically in the cervix, and have been implicated in the natural history of these tumors. The relationship of these host gene alterations to vulvar squamous cell carcinomas has not been previously studied. We analyzed archival material from 13 human papillomavirus-positive and -negative vulvar squamous cell carcinomas for mutations in Ha-, Ki-, and N-ras genes, and a smaller number of fresh samples for c-myc amplification, using PCR-based assays. For comparison, eight cervical squamous cell carcinomas (three fixed and five fresh) were also analyzed. Analysis for ras mutations revealed a previously reported silent allelic variant at nucleotide 1744 in the Ha-ras gene, but no mutations in codons 12, 13, or 61. Similarly, genomic amplification of c-myc beyond a maximum of three haploid copies was not identified in the cases. These findings indicate that alterations in myc or ras sequences are not linked to vulvar squamous cell carcinomas or to the presence or absence of HPV nucleic acids. Moreover, they apparently will not distinguish vulvar from cervical carcinomas, both groups appearing to be unlikely to harbor these sequence alterations.

Restriction fragment length polymorphism analysis of isotype-labeled polymerase chain reaction-amplified human papillomavirus DNA combines sensitivity with built-in contaminant detection

The detection of low copy number DNA in archival tissue has been revolutionized by polymerase chain reaction (PCR), but widespread acceptance of this technique for diagnostic purposes has been hampered by problems with contamination. There is a technique, restriction fragment length polymorphism analysis for characterizing PCR-amplified human Papillomavirus (HPV) nucleic acids, that reportedly provides for increased sensitivity in detection of both target and contaminating sequences. This technique is a modification of standard restriction fragment length polymorphism and involves radiolabeled nucleotide incorporation during PCR and restriction enzyme digestion of the products followed by high resolution polyacrylamide gel electrophoresis. The amount of initial target DNA amplified by PCR that can be detected by isotope labeling or ethidium staining was compared, as was the sensitivity of the two methods for analyzing a PCR product of fixed amount. The sensitivity of ethidium staining or autoradiography was comparable for detection of initial target HPV DNA by PCR. However, for subsequent typing of amplified HPV DNA by restriction fragment length polymorphism, isotope-labeled products provided an approximately 125-fold increase in sensitivity over ethidium staining, with a maximum of 625-fold greater sensitivity with a 3-day exposure. The detection of low levels of potential contaminants in PCR-amplified HPV DNA was determined in a serial analysis of cervical biopsies. Contaminating DNA was identified in two gels where multiple samples yielded the same restriction pattern. On re-isolation of genomic DNA, these products of contaminating DNA were not seen. Extrapolating from the above experiments, the contamination of many samples would have escaped detection by ethidium staining alone.(ABSTRACT TRUNCATED AT 250 WORDS)

Human papillomavirus type 18 and intraepithelial lesions of the cervix

The conventional perception of HPV type 18 is that it is associated principally with invasive cancer of the cervix. However, in precursor lesions it is frequently identified in lesions of lower grade morphology, in contrast to the typical high grade lesions associated with HPV 16. To better characterize this uncommon relationship of low grade morphology and high risk virus, we studied four low grade and two high grade intraepithelial lesions of the cervix which were shown to contain HPV 18 by polymerase chain reaction/restriction fragment length polymorphism analysis. Immunohistochemistry and in situ hybridization identified capsid proteins and viral DNA in an epithelial distribution characteristic of episomal replication and late gene expression. Sequencing of the E2, E6, E7, and upstream regulatory regions revealed four silent mutations within these precursor lesions. Two of these sequence alterations were also noted in three of four HPV 18 positive squamous carcinomas, suggesting a viral subtype which was not unique to cervical precursors. The bland morphology of many HPV 18 related precursors contrasts with the high grade morphology of HPV 18 associated cancers. However, this diversity cannot be linked to functional differences in sequences encoding in vitro transforming potential, transcriptional regulation, or transactivation functions. Whether these differences in lesion phenotype relate to unique host variables remains to be determined.

A binary (Bethesda) system for classifying cervical cancer precursors: criteria, reproducibility, and viral correlates

This study of cervical squamous precursors addressed the consistency with which pathologists could agree on diagnosis using a Bethesda system and the degree to which the classification system discriminated "high-risk" human papillomavirus (HPV) types. Four pathologists independently assessed biopsies of 75 squamous lesions; all contained HPV DNA amplified from archival fixed tissue with polymerase chain reaction (PCR) and typed by restriction digestion of the PCR product. Lesions were categorized as low or high grade using published criteria. In independently performed histologic evaluations a majority (three or more) of observers agreed on the classification of 63 of the 75 cases (84%) with good to very good interobserver (kappa values, 0.43 to 0.63), and fair to excellent intraobserver (kappa values, 0.32 to 0.83) agreement. A majority of the observers classified as high grade 15 of 17 (88%) HPV 16-positive lesions (P < .002), but only 15 of 21 (71%) lesions associated with other high-risk HPV types 18, 31, 35, and 39 (P = .089). Concurrence among observers also varied with HPV type; majority agreement between three or more observers was present for 100% and 94%, respectively, for lesions associated with HPV 6/11 and HPV 16 versus 82% and 76% for lesions associated with HPV 18/31/35/39 and other HPV types. A binary system for grading cervical precursor lesions was applied with good reproducibility among pathologists, and segregated as high-grade virtually all lesions associated with the prototype high-risk HPV (HPV 16). Conversely, other presumed high-risk HPV types, particularly type HPV 18, were not distinguished by this grading scheme and were segregated frequently with low-grade lesions. This finding suggests that variables other than HPV type alone will influence lesion grade. Resolution of these variables will determine whether lesion grade is a more potent biologic parameter than HPV type.

The binucleate squamous cell: histologic spectrum and relationship to low-grade squamous intraepithelial lesions

The consistent histologic distinction of HPV-related precursor lesions from nonspecific epithelial changes is hampered by the fact that the scoring of nuclear atypia is subjective and nuclear alterations or cytoplasmic halos may occur with both HPV and non-HPV-related changes. To determine the potential role of another parameter--bi- or multinucleated cells--in the diagnosis of HPV-related lesions, we analyzed a series of epithelial alterations for anisokaryosis, hyperchromasia, and cytoplasmic halos and correlated their presence with the maximum of binucleate cells per high-power field (hpf) and the detection of HPV DNA. A positive correlation was seen between the presence of both anisokaryosis and hyperchromasia and the number of binucleate cells/hpf (p = 0.011) and the presence of HPV nucleic acids by in situ hybridization (p = 0.005). Only one of 23 (5.3%) known HPV-positive lesions did not exhibit any binucleate cells. These findings indicate that, although binucleation may be associated with a spectrum of both HPV and non-HPV related changes, it is most conspicuous and virtually always present in association with HPV-positive low-grade precursor lesions. Thus, this parameter may be useful to confirm the presence when other parameters suggest the diagnosis of a low-grade squamous intraepithelial lesion.