Vol. 14 No 1 | Autumn 2012
Under the microscope: cervical screening

This article is 12 years old and may no longer reflect current clinical practice.

Cervical screening in the era of human papilloma virus vaccination: is the Pap smear test still the best technique available? In 2006, the Gardasil® human papilloma virus (HPV) vaccine became available in Australia. In 2007, the government program for free vaccination for young women was commenced. Initially, there was a catch-up program for young women up to 26 years of age and there is the ongoing program for girls aged 12–13 years. Overall, there has been good take up of the vaccine. It protects against infection with HPV 16 and 18, which in Australia represents more than 70 per cent of cervical cancers and approximately 50 per cent of high-grade squamous intraepithelial lesion (HSIL) abnormalities. There is also some cross protection against some other high-risk (HR) HPV types.

We expect that we will see a drop in the numbers of high-grade abnormal cytology within a fairly short time frame.1 In low-prevalence populations cervical cytology does not perform as well as in populations with high levels of cervical disease.2 The principle strength of cervical cytology is, while it may have low sensitivity, it generally has good specificity. HPV testing, on the other hand, has excellent sensitivity with poorer specificity. For many women, HPV carriage is transient and it is only persistent carriage that results in intraepithelial lesions developing in the cervix. The specificity of HPV testing, however, does improve in women over the age of 30 years.

There have been many trials comparing HPV testing with cervical cytology either as a standalone cervical screening method or as a triage tool where cytology is inconclusive, including:

  • New Technologies in Cervical Cancer (NTCC)3;
  • ARTISTIC trial4;
  • Sweedscreen6; and
  • Finnish trial7.

These trials have confirmed that HPV testing is more sensitive than cytology in detecting HSIL lesions and the negative predictive value of an HR HPV test is about six per cent higher than that of a negative cytology test. The improved negative predictive values should allow for longer screening intervals. In the studies that followed women over two screening rounds, an average reduction of about 50 per cent in the HPV arm was shown in the second screening round. The NTCC trial was sufficiently powered to show a significantly better prevention of cervical cancer in the HPV arm.

So the Pap smear, which has been the screening tool for cervical cancer since cytology-based programs were introduced in the mid-20th century and has resulted in a significant reduction in cervical cancer in screened women, has come under threat as unequivocally the best tool for primary screening. In 2011, the Netherlands Health Council advised the Ministry of Health in the Netherlands to implement HR HPV testing as the primary test in cervical cancer screening for women aged 30–60 years, with cytology triage for those testing positive. It has suggested that screening be performed at 30, 35, 40, 50 and 60 years of age. With this screening program, the Netherlands will become the first country with an HR HPV-based screening program and triage with cytology that involves only five lifetime screening rounds.8

HPV tests suitable for clinical use became available in the 1990s. Some countries (such as New Zealand, the UK and USA) incorporate their use in triage for colposcopy in women with inconclusive or low-grade changes on cytology. In Australia, in 2006, the National Health and Medical Research Council (NHMRC) recommended HPV testing only for test of cure after treatment for high-grade abnormalities.

In New Zealand, the National Cervical Screening Programme guidelines recommend HR HPV testing in three clinical situations:

  1. women 30 years and older with ASC-US or low-grade changes, to help assist risk of progression (as they use liquid based cytology in New Zealand, this is performed reflexly by the laboratory);
  2. women (all ages) treated for a high-grade lesion, to help assess whether the lesion has been completely resolved; or
  3. where colposcopy has shown discordant results from cytology, to help interpret these results.

The last few years have seen an increase in the number of alternative tests for detection of HR HPV DNA. It has become apparent that various HPV detection methods vary in their clinical performance, with specificity varying substantially.

There have been a number of methods by which the specificity of the HPV test can be improved. One is genotyping. For the ALTS trial it was recognised the HPV carriage with types 16 and 18 carried far greater significance than HPV carriage with one of the other high-risk types.9 Some of the newer tests do allow for identification of HPV 16 and 18.

Viral load is known to be significant. It is important that HPV tests are not so sensitive that they pick up clinically unimportant infections with low viral loads that are generally not associated with HSIL.10 However, higher viral loads are known to be associated with a higher risk of HSIL.

Another marker of significant disease has been over expression of p16INK4a. In the NTCC if only women with over expression of p16 were referred for colposcopy, there was no increase in referral for colposcopy, but HPV testing still gave an improved sensitivity of 1.53 over cytology.3

Testing RNA of E6 and E7 has also shown promise, as levels increase with lesion severity. There is available a nucleic acid sequence-based amplification (NASBA) based assay detecting E6/E7 transcripts from the five most common HR HPV types (PreTect, HPV-Proofer, NorChip AS are some of these). Gen-Probe is currently developing the APTIMA® HPV Assay targeting E6/E7 mRNA from 14 HR HPV types.11

There are three principle technologies used for HR HPV DNA detection assays:

  1. hybridisation followed by signal amplification hybrid capture 2: RNA probe cocktail (HC2; Qiagen) invader technology: third-wave invader HPV test (Cervista);
  2. broad spectrum PCR DNA amplification with consensus primers or multiplex format (some labs have in-house PCRs; Abbott – Real time HPV test, Amplicor HPV & Linear Array HPV Genotyping); and
  3.  in situ hybridisation (ISH) used for years for research into HPV (Ventana Inform the only commercially available and recommended for histological material only).13

The following HPV tests are available in Australia:

  • Digene or Hybrid Capture® 2 (HC2) – used in the ALTS trial, there is a lack of internal control for input DNA and it does have some cross reactivity with low-risk types. It will not identify specific HPV types.
  • In house PCR – this is the most sensitive technique, it allows for testing on samples with fewer cells. Contamination can be an issue. It is capable of type distinction. There is no external validation.
  • Cervista (Hologic) FDA approved (invader technology) can distinguish HPV 16 and 18 from other high-risk types. There is internal control for input DNA,
  • Abbott Real time HPV test (PCR) provides the most accurate type-specific measure of viral load,
  • Roche Cobas 4800 HPV test – used in the Athena trial, it has an internal control and can identify HPV 16 and 18 infections.

The cervical screening program in Australia has been hugely successful; leading to the country having one of the lowest rates and mortality for cervical cancer in the world. However, our understanding of the disease, and the technologies available, has increased dramatically since the introduction of the screening program. We have fallen behind the rest of the world in embracing new technologies. Fortunately, a review of the entire cervical screening program has begun and it is likely that there will be great changes.

I believe that, ultimately, HPV testing will become part of the routine cervical screening program in Australia, as it is in New Zealand, but there are many developments and advances in this testing. The positive predictive value of HPV testing will also decline in the vaccinated population due to removal of all the HPV 16 and 18 lesions, which have much higher association with significant lesions than the other non-vaccine high-risk types. Thus it will become important to use and expand the other methods of improving specificity of HPV testing if it is to be used as a cervical screening tool.


  1. Smith CA, Cranfell K, Brotherton JM et al. Int J Cancer 2008; 123(8):1854-63.
  2. Ratnam S, Franco EL, Ferenzy A. Cancer Epidemiol Biomarkers 2000;9(9):945-51.
  3. Ronco G et al. Lancet Oncol 2010; 11(3):249-57.
  4. Kitchener HC et al. Lancet Oncol 2009; 10:672-82.
  5. Bulkmans NW et al. Lancet 2007; 370 (9601):1764-72.
  6. Naucler P et al. N Engl J Med 2007; 357(16):1589-97.
  7. Leinonen et al. J Natl Cancer Inst 2009; 101(23):1612-23.
  8. Meijer JLM. HPV Today 2011; 24:4-5.
  9. Khan et al. J Natl Cancer Inst 2005; 97(4):1072-9.
  10. Snijders PJF. HPV Today 2011; 24:1-3.
  11. Andersson E et al. J. Clin Microbiol 2011; 49(11):3794-9.
  12. Nishimo HT et al. Cancer Cytopath 2011; 119(4):219-27.

Leave a Reply

Your email address will not be published. Required fields are marked *