Human Papillomavirus Reporting: Impact on Bethesda Cytology Reports

The Bethesda System for Papanicolaou (Pap) test reporting is an example of standardized reporting.1,2 Standardized reports are designed to improve clinical understanding of complex pathology data.2–6 This understanding theoretically provides guidance for clinical management. The first Bethesda Workshop was held in 1988 and helped reduce clinical confusion created by the use of multiple classification systems.7 A second Bethesda Workshop was held in 1991 and modified the terminology for Pap test reporting; modifications were based on laboratory and clinical experience of using the Bethesda terms.8 It has been estimated that more than 90% of American laboratories use some form of Bethesda terminology.9 

A great deal of research regarding the Bethesda System terminology has been conducted since 1991. In addition, new technologies increasingly are being used in cervical cancer screening. Both these research data and changes in the use of technologies led to the Bethesda Committee meeting again in 2001 to reevaluate the 1991 Bethesda System terminology.1 The 2001 process consisted of establishing forum groups designed to draft recommendations for changing problematic terminology. One of the 2001 Bethesda forums was the Ancillary Testing Forum.1 Although ancillary tests occasionally have been used in conjunction with Pap tests, recent data have indicated that ancillary tests will be used with increased frequency.

Guidelines developed by the American Society for Colposcopy and Cervical Pathology (ASCCP) Consensus Conference recommended that human papillomavirus (HPV) testing is a management option for women with a Pap test interpretation of atypical squamous cells-undetermined significance (ASC-US),10 and consequently, the HPV test is currently the main ancillary test to be used in conjunction with Pap tests. It is possible that other ancillary tests may replace or be used in addition to HPV testing. The 2001 Bethesda Committee thought that it was important to establish terminology for ancillary testing reports as it has established terminology for cervical cytology reports. The Ancillary Testing Forum used HPV testing as a model for ancillary test reporting. Although some of the Bethesda System terminology recommendations are specific for HPV tests, the modeling for test reporting theoretically may be used for any ancillary test. This article discusses the Bethesda System terminology for HPV tests and provides several examples.

In an opening preamble, the Ancillary Testing Forum wrote that the use of any ancillary test, including the HPV test, was not endorsed or supported by the 2001 Bethesda Committee. However, specific ancillary tests may be used to complement the Pap test. Currently, some laboratories perform HPV DNA testing as an adjunct to the Pap test. Other ancillary tests may become available in the future. For these reasons, it was important that the Bethesda Committee provide recommendations regarding how ancillary tests may be reported in conjunction with the Pap test. The development of practice guidelines addressing clinical indications for testing was outside the purview of the Bethesda conference.

Current data indicate that testing for “high-risk” HPV DNA may be useful for the colposcopic triage of women who have a diagnosis of ASC-US.11–14 Cost-effectiveness analysis of HPV testing for triage of this patient group is ongoing, including relative comparisons with repeat cytology follow-up. More limited data suggest the possible utility of HPV testing in other clinical situations, such as the management of women who have a diagnosis of atypical glandular cells,15 primary screening in older women, quality assurance for cytology, and monitoring posttreatment cure or recurrence of cervical disease.1 Additional studies are needed to investigate the utility of HPV testing in these scenarios. Human papillomavirus testing has been shown to lack utility for the colposcopic triage of women who have low-grade squamous intraepithelial lesion (LSIL)11 or high-grade squamous intraepithelial lesion (HSIL) diagnoses.1 However, HPV testing in women who have LSIL or HSIL may be advocated in individual circumstances (eg, a woman who has HSIL and negative biopsies).1 Current data show that screening for “low-risk” HPV types is of little utility.

The Ancillary Testing Forum addressed 3 specific issues: (1) reporting of HPV test methods; (2) simultaneous reporting of Pap and HPV test results; and (3) reporting options for HPV test results.

Several methods for performing HPV tests are currently available (or will become available). These molecular methods include Hybrid Capture II, in situ hybridization, Southern blot hybridization, ligase chain reaction, branch signal DNA, and polymerase chain reaction. The Ancillary Testing Forum recommended that the pathology report contain information regarding the test method and the performance characteristics of that test. For example, if a laboratory performs HPV testing using the Hybrid Capture 2 method, then this should be stated in the report. Providing the specific test method is important because each method has different performance characteristics that are unique for that test. Commonly used performance characteristics include the metrics of sensitivity, specificity, negative predictive value, and positive predictive value. These data may be derived from the literature, manufacturer, or individual laboratories, and these performance characteristics provide for comparisons among different test methods. For a particular test method, the performance characteristics vary according to the threshold set for a positive result.

In summary, the Ancillary Testing Forum recommended that for HPV testing, the molecular results should be reported as positive or negative for HPV DNA of a certain type or class, and the laboratory method should be indicated. The specific HPV types included in the assay may be listed.1 Several examples follow. Particular manufacturer names specifically have been omitted.

In situ hybridization assay for high-risk HPV types is negative.

The (probe name) HPV (manufacturer name, city, state) probes are analyte-specific reagents and are used to detect HPV DNA by in situ hybridization. The (probe name) HPV high-risk probe targets DNA of HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 68, and 70.

Human papillomavirus Hybrid Capture assay is positive for high-risk types of HPV.

The HPV assay was performed using the (manufacturer name, city, state) (assay name) panels. The (assay name) high-risk panels test for 13 HPV types: 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68. A positive test indicates the presence of 1 or more HPV types in the panel.

Human papillomavirus polymerase chain reaction assay is positive for the following high-risk type(s) of HPV: 16.

The HPV assay was performed using the (manufacturer name, city, state) (assay name) panels. The (assay name) high-risk panels test for 10 HPV types: 16, 18, 31, 33, 35, 39, 45, 51, 56, and 68.

The second Ancillary Testing Forum issue dealt with the integration of the Pap test result with the HPV test result. Ideally, the Pap test result should be reported with the HPV test result, but this may not be possible in all laboratories. Integrated cytologic and molecular reporting improves communication and record keeping, facilitates pathology education, and provides ongoing quality assurance for both the cytology and molecular tests.1 The Bethesda System Committee endorsed integrated reporting when possible. Some laboratories may send out HPV tests to other laboratories and simply receive the results from that other laboratory. If integrated reporting is not possible, then the individual Pap and molecular test results should refer to each other.1 For example, a cytology laboratory could make an interpretation of ASC-US and then make a statement that a specimen has been sent for HPV testing at another laboratory. The following examples show different means to integrate the Pap and HPV test results.

Atypical squamous cells-undetermined significance.

Human papillomavirus testing is being performed and the results will be issued as an addendum. (This report is an example of when the Pap and HPV tests are performed in the same laboratory.)

Atypical squamous cells-undetermined significance.

Human papillomavirus testing is being performed at (laboratory name, city, state) and a separate report will be issued. (This report is an example of when the Pap and HPV tests are performed in different laboratories; this report is issued from the laboratory interpreting the Pap test.)

Human papillomavirus hybrid capture assay is positive for high-risk types of HPV.

The Pap test was interpreted at (laboratory name, city, state). Please see the separate cytologic interpretation report issued by this laboratory. (This report is an example of when the Pap and HPV tests are performed in different laboratories; this report is issued from the laboratory interpreting the HPV test.)

This Ancillary Testing Forum issue was one of the more controversial issues at the 2002 Bethesda System Workshop. This issue deals with different reporting formats and the recommendations for clinical practice. (This issue essentially assumes that the Pap test and HPV test results may be reported simultaneously.) If clinical management recommendations are made, these should be appended to reports of ancillary/cytology test results.

At the 2001 Bethesda System Workshop, several different options for reporting HPV test results were discussed: (1) as a result only; (2) as a result associated with clinical management recommendations; (3) as a result plus a probabilistic statement regarding the risk of an underlying CIN 2 or 3; and (4) as a final definitive interpretation that reflects both the cytomorphology and the HPV status. Examples of the different options follow.

Atypical squamous cells-undetermined significance with detection of high-risk oncogenic HPV DNA.

Atypical squamous cells-undetermined significance with detection of high-risk oncogenic HPV DNA.

The ASCCP has recommended that this group of women undergo colposcopic examination.10 

Atypical squamous cells-undetermined significance with detection of high-risk oncogenic HPV DNA.

These findings are associated with a 10% to 20% risk of underlying CIN 2 or higher lesion.12 

Low-grade squamous intraepithelial lesion.

The preliminary cytologic findings are ASC-US. High-risk oncogenic DNA has been detected. In combination, these results are most consistent with a diagnosis of LSIL.

Option 3 has been referred to as the probabilistic reporting model and option 4 has been referred to as the interpretive reporting model. At the 2001 Bethesda System Workshop, the probabilistic reporting model was favored over the interpretive reporting model, although each model has its advantages and disadvantages.1 Options 1 through 3 do not affect the cytologic interpretation of ASC-US, which remains as the final interpretation. The additional information about the HPV test and its significance is reported as a note or recommendation. In the interpretive model, the final diagnosis is SIL (HPV DNA detected) or negative for intraepithelial lesion or malignancy (NIL) (HPV DNA not detected) in most cases. The interpretive model eliminates ASC-US in all but a few cases in which review of the slide warrants retention of ASC-US, despite the HPV test result. In the interpretive model, the cytologic slides may be reviewed in conjunction with the HPV test result.

Laboratories should be aware that there are different models of reporting the ancillary test data, and different recommendations may be preferred by clinicians and other interested personnel. These models may have different medicolegal implications. To make recommendations for clinical practice, the ancillary test must be available to the cytopathology laboratory; therefore, not all laboratories will be able to make recommendations. Even if HPV test data are available, not all laboratories will choose to make recommendations for clinical practice.

In the Bethesda System Workshop no specific reporting model was recommended, and additional data were deemed necessary before an optimal reporting model(s) could be advocated.

The following reporting examples incorporate the 3 issues discussed:

Atypical squamous cells-undetermined significance with detection of high-risk oncogenic HPV DNA.

These findings are associated with a 10% to 20% risk of underlying CIN 2 or higher lesion.12 

Human papillomavirus Hybrid Capture assay is positive for high-risk types of HPV.

The HPV assay was performed using the (manufacturer name, city, state) (assay name) panels. The (assay name) high-risk panels test for 13 HPV types: 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68. A positive test indicates the presence of 1 or more HPV types in the panel. (This report is an example of when the Pap and HPV tests are performed in the same laboratory.)

Atypical squamous cells-undetermined significance with detection of high-risk oncogenic HPV DNA.

In situ hybridization assay for HPV was performed in the Molecular Laboratory at (hospital name).

Negative for HPV high-risk types.

The (probe name) HPV (manufacturer name, city, state) probes are analyte-specific reagents and are used to detect HPV DNA by in situ hybridization. The (probe name) HPV high-risk probe targets DNA of HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 68, and 70. (This report is an example of when the Pap and HPV tests are performed in different laboratories).

The examples presented here are suggestions for use, and laboratories will use different reporting formats. However, the ultimate goal is to attain uniformity and consistency of included data that are relevant to the clinical management of patients.3 As Powsner et al reported,2 few data exist on the degree to which clinicians comprehend pathology reports. The use of ancillary test reports similar to the formats presented here will encourage explicit reporting and will require less interpretation by the clinician. The examples presented here have been constructed to be as simple as possible, but this construction leads to the limiting of the presented information. A limitation of these models is that they were essentially constructed before the widespread use of ancillary tests in Pap test reporting, so there is an absence of data on how these reports actually affect patient care and how these reports are accepted by clinicians. Thus, these and other ancillary testing models most likely will be evaluated in the years ahead.