There’s a better way to screen for cervical cancer

When it comes to cervical cancer, the attention is centred on the vaccine for Human Papillomaviruses (HPV). It’s an incredible development, and I am hopeful that one day the HPV vaccine could render cervical cancer a killer of the past. But there’s another innovation – called the HPV test – that could be harnessed immediately to improve cervical cancer screening.

Before I get into the HPV test, I need to explain what’s wrong with the 70-year-old Pap test, our current cervical cancer screening method. For the Pap test to produce an accurate reading, cervical cells have to be properly spread and stained on a glass slide that’s viewed under a microscope. The sample of exfoliated cells has to be representative, which means the health care provider has to properly collect the sample for an accurate diagnosis to be made. Then, the technologist who is viewing the slide has to interpret it correctly. Looking at samples under a microscope is a monotonous task; it can lead to mental fatigue and errors as a result. Whether because of health worker or technologist error or a mix of both, we know that Pap tests give false negative results – meaning that they don’t detect the presence of precancerous or cancerous cells. Even in Canada, where we have rigorous quality control, a study I conducted found that the false negative rate of Pap tests is 45%. (Don’t be too alarmed by this, cervical cancer develops very slowly and the test is performed frequently, every two to three years depending on the jurisdiction. So even with the high false negative rate, most cervical pre-cancerous cells are detected in time when provincial screening protocols are followed.)

There is another screening option that is increasingly being used in other countries, including the US, UK and Netherlands. It’s called the HPV test and it’s been available since 2003. Economic models show the introduction of this test will make cervical cancer screening more efficient and affordable, in addition to making cervical cancer screening more convenient for the majority of women.

Before we get into why HPV testing (combined with Pap smears) is superior to Pap tests alone, I need to explain some key information. Cervical cancer is caused by some strains of HPV, which are sexually transmitted viruses. Most genital HPV infections in women (and men) are asymptomatic, even if they involve carcinogenic (or cancer-causing) HPV types. Most – even the carcinogenic types – eventually clear and are no longer detectable after one year. But a small proportion of the carcinogenic HPV infections that persist may progress to precancerous lesions of the uterine cervix. If untreated, these lesions may progress to cervical cancer. The entire process from acquisition of an HPV infection until development of cervical cancer may take 15 to 25 years. Although most sexually active women (and men) eventually acquire an HPV infection, the proportion of women that eventually develop cervical cancer over a lifetime in Canada is about 0.5-1%.

HPV screening can detect the strains of the HPV virus that research shows are most likely to cause cervical cancer. In an HPV test, a sample is gathered from the cervix, but unlike in the Pap test, it doesn’t have to be highly representative. And the sample is then extracted, processed, and read by a machine – taking human error out of the interpretation. With HPV tests, the false negative rate is only 5%.

Here’s how HPV testing would work. Up until around age 30 (experts differ slightly as to the appropriate age), women would be screened using Pap tests. That’s because young women tend to have more sexual partners and are exposed to HPV at a high rate. But, as mentioned, the virus doesn’t tend to stick around and cause problems. So a positive HPV test may not mean much.

After 30, however, HPV can be the dominant screening method. If the HPV test is negative, a Pap test wouldn’t be required and the HPV test wouldn’t have to be repeated for another five years.  If an HPV test is positive, that’s concerning, because that suggests the woman has a persistent HPV infection that may have already developed into a precancerous lesion or will in the future. So for these women, we do a Pap test to see if the infection has already become associated with cellular changes in the cervix indicative of developing cancer. If the Pap tests are also found to be positive, women would then be referred for colposcopy for a final diagnostic confirmation and eventually treatment.

For women who are positive with the HPV test but not with the Pap test, they would have to repeat the HPV and/or the Pap tests within one year. If the infection persists or the Pap test became positive then their physician will want to refer them for a colposcopy.

The combination of HPV and Pap test screening would be beneficial for many reasons. Firstly, only those who are found to have infections with carcinogenic HPV types would be referred for Pap testing. Only about 6% to 10% of women above 30 have HPV infections, so more than 90% of women would be able to move from Pap tests every three years to HPV tests every five years. Secondly, Pap tests would only be done when there is a high concern that pre-cancerous lesions could be present. So technologists would be viewing 10-fold less Pap test slides, meaning the interpretation would be less prone to mental fatigue. Because technologists know that HPV has been detected in the samples they view, they would be more vigilant in looking for abnormal cells. Finally, combination testing would be cheaper – because HPV tests are automated and don’t require the expertise of high-salaried technologists, they’re not as expensive as Pap tests.

There is overwhelming evidence for the superiority of HPV testing as the primary technology in cervical cancer screening, all from large-scale randomized controlled trials, including two in Canada (available here and here).

You may be wondering why I’m concentrating on the need to detect HPV viruses when the HPV vaccine is available. Publicly funded, school-based HPV vaccination of teenage girls (and now also boys in some provinces) has been a reality in Canada since 2007. Although HPV vaccination will bring enormous benefits in reducing the incidence of cervical cancer, the vaccines do not prevent an existing infection from progressing to cervical cancer. It will be another 30-40 years before we can even hope to end cervical cancer screening. And in the meantime, as the vaccination reduces the risk of HPV infections, the Pap test could become less and less effective. Why? If a technologist is looking at higher and higher rates of normal smears, there’s a high risk that the technologists could become complacent, and miss the odd abnormal test. In other words, as HPV comes down, HPV testing will become even more important.

So why then haven’t we adopted HPV screening in Canada? Part of the reason comes down to short-term thinking on cost. As of today, HPV tests are more expensive than Pap cytology. However, this is because HPV testing has been used in a limited capacity, such as to provide additional information when a Pap smear isn’t clear, which is not the best use of this technology. In the high-volume scenario of primary screening and with the pricing control measures of centralized procurement, the HPV test will be reduced to be about half of that of Pap tests currently. The extension of screening intervals to every five years that comes from a change to HPV screening will lower the cost of cervical cancer control even further for provinces.

Using the HPV test as the primary method of screening is the best from a clinical, economic and patient convenience perspective. We have the opportunity to make screening better and cheaper right now.

Dr. Eduardo Franco is Professor and Chair of the Department of Oncology at McGill University in Montreal. He is also the Director of that department’s Division of Cancer Epidemiology and Editor-in-Chief of Preventive Medicine. During the last 30 years, much of his research has been on the topic of HPV and cancer.

Dr. Franco’s research into HPV and cervical cancer is publicly funded. He has no financial interest in companies that make the HPV test. He has consulted for biotech companies, for which he has received expenses and honoraria.

Leave a Comment

Enter the debate: reply to an existing comment

  1. Brianne Wood

    As Dr. Franco noted, there is overwhelming evidence to support HPV testing as a primary screening test, and a self-collection option for HPV testing may present another advantage over traditional cytology. Most cervical cancers are diagnosed in women who do not participate in regular screening, meaning that marginalized populations carry a disproportionate burden of this disease, in Canada and globally. Self-collection (i.e., a woman can collect a vaginal sample without the supervision of a health care provider) may overcome some of the barriers that women face with traditional cervical screening, with comparable sensitivity and specificity to physician-collected samples (Arbyn et al, 2014). There is evidence suggesting that women may prefer the self-collection modality to the in-clinic alternative (Racey et al, 2013). Shifting to primary HPV screening offers an opportunity to implement an evidence-based, woman-centered cervical screening program that includes a self-collection option. Most importantly, we need to support women in their decision-making as we transition to a new model of cervical cancer control.

  2. Elizabeth EB

    I’m Australian and have always declined to take part in our OVER-screening pap testing program, far too much risk with screening, I was content with my near zero risk of cervical cancer. (lifetime risk of cc is 0.65% while the lifetime risk of colposcopy/biopsy is a massive and hidden 77%)
    We need to remember that informed consent is a legal and ethical requirement for all cancer screening, some women will choose not to have pap tests and mammograms. So often women are ordered, pressured and even coerced into testing with no real information on the actual benefits and risks with screening.
    (I’ve also declined breast screening: real information is available on the website of the Nordic Cochrane Institute, here we get a screening “story” and celebrity endorsement. Over-diagnosis and over-treatment is a serious issue, I believe the risks of breast screening exceeds any benefit)

    I’ve watched other countries and believe the Finns and Dutch are the ones to watch when it comes to cervical screening, they’ve followed the evidence and put women first, all women, the small number who might benefit from cervical screening and the vast majority who can never benefit but can be harmed by false positives, excess biopsies, and overtreatment.
    Here in Australia with early screening and serious over-screening in place we have huge, and mostly avoidable, referral rates for colposcopy & biopsy and we “treat” more than 10 times the number of women than a country like Finland. (Early screening and over-screening means more false positives for no additional benefit to women)
    Our new program will again side with excess and ignore some of the inconvenient evidence. We’ll likely offer 10 or 11 HPV tests, 5 yearly from 25 until mid 70’s.
    You only have to do basic research to find HPV testing is not recommended before age 30, so we’ll throw our young women under the screening bus once again. No country in the world has shown a benefit doing pap tests on women under 30, the Finns and Dutch have always provided their young women with the best advice, no screening, just see a doctor with any persistent and unusual symptoms. (Young screened women tend to get false negative pap tests and they produce the most false positives of any age group) Our current excessive pap testing program starts at about 18, some women are even younger and calls for 2 yearly screening until age 70.

    The new Dutch program is 5 HPV tests, or HPV self-testing with the Delphi Screener, at ages 30,35,40,50 and 60 and a 5 yearly pap test will ONLY be offered to the roughly 5% who test HPV+
    This will save more lives, takes most women out of pap testing and harms way and see referral rates for colposcopy, biopsy and “treatments” plummet. (Dutch referral rates have always been MUCH lower than Australian rates anyway) It will also save scarce health resources.

    It’s important to keep the influence of vested interests away from these programs, they tend to put their interests first and favour excess. Screening should always follow the evidence, put women first and respect informed consent. It’s concerning that women are often viewed as sheep to be rounded up for screening with little respect for consent or informed consent.

    HPV self-testing will be locked away here until women refuse the invasive HPV test for 6 years, but more women are getting to the evidence and understand they can buy self-testing devices online.
    There was never a need to worry and harm so many to help a few women. Screening has the potential to cause a lot of harm if it’s handled by those who use it to promote their own interests.

  3. James Dickinson

    Vigorous lobbying for HPV tests.

    Dr. Franco argues for using HPV tests for cervical screening, citing this as a better and cheaper test. While I respect and share his passion to minimise cervical cancer, his argument is selective, denigrating the pap test as not sensitive, with a 45% false negative rate. True – for precancer lesions, most of which regress, so do not need to be found – but better for higher-grade disease. The trials of HPV testing, funded by commercial sources, show good results in the short term, but long-term evaluations of using new technologies in real-world practice often reveal unexpected problems, and outcomes not as good as the trials. He glosses over the reality that the current cervical screening program serves us well: cervical cancer has been substantially reduced in Canada, such that the majority of invasive cancer now occurs among those women who do not participate in screening. Therefore any major gains must come from screening the non-participants: and while HPV technology holds out hope that self-testing will improve participation, these women do not readily participate in self-testing either. Meantime, most women in Canada are over-screened: starting too young, and repeating too often. This causes extra false positive results that lead to excess investigation and treatment, which can cause harm, especially to women who have not completed their reproductive plans.

    Replacement by HPV technology requires it to be as good as or better than current pap testing, without adding extra cost. Dr. Franco quotes an economic study that shows HPV testing could cost less: provided that central procurement obtained lower prices, and women and doctors in Canada were willing to extend the interval to five years. Neither seems likely in the short term: Canada seems incapable of bargaining hard with drug and device makers, and despite some years of recommendations to raise the starting age for pap tests, and increase the interval between them, it is taking a long time to retrain both women and physicians out of their old habits of testing too early and too often. HPV testing tends to increase false positive rates, leading to even more referrals for follow-up by colposcopy, with over-treatment on occasion. If testing intervals did not increase, harms to women and monetary costs would rise substantially with almost no extra benefit – except to the marketers of the test.

    The major player in the HPV test market appears to be Quiagen, who are currently funding a public relations campaign across Canada by Santis Health, a lobbying firm, to persuade provincial policy-makers to change to their HPV test. For this purpose, their “Evidence Review and Report” was formally authored by 14 experts from across Canada, including Dr. Franco, whose opinion article summarises that lobbying document. Dr. Franco states that his research is publically funded, but “He has consulted for biotech companies, for which he has received expenses and honoraria.” So have many of his co-authors.

    There is a good case for being cautious about changing from an effective program until we can be confident about the effects and costs of the new. This is not “short term thinking on cost”, but prudent recommendations about public resources, and care for women.

    Dr. James Dickinson is a Professor of Family Medicine and Community Health Sciences at University of Calgary, and a member of the Canadian Task Force on Preventive Health Care, which recommended in 2013 against using HPV testing until better information is available. However, the opinions he expresses in this letter are his own. He has no financial conflicts of interest.

  4. Robert Lotocki

    Without question the Pap test has been instrumental in reducing the incidence of and mortality from cervical cancer. The use of the Pap test has occurred without any randomized clinical trials.
    However, the weaknesses of the Pap test are apparent. It has a low sensitivity. Cytology laboratories have rigorous Quality Assurance programs. Despite this false negatives still occur. The Pap test is extremely effective in identifying Squamous Cancer and its squamous precursors. However, the Pap test does not have the same ability to demonstrate Adenocarcinomas and especially adenocarcinoma precursors. Unfortunately, adenocarcinomas are increasing, occurring in women who have had regular Pap testing. There is increasing evidence that HPV testing can identify these Adenocarcinomas and potentially its precursors.

    In addition, Cytology identifies a large number of low grade squamous lesions that would not progress to cancer. This results in over management and over treatment of low grade lesions that would not progress. HPV Testing, either in a primary testing role or as a secondary triage helps identify more significant low grade lesions.

    Organized cervical screening programs are being developed across Canada despite the recommendations made more than 4 decades ago. Unfortunately, Canada functions as multiple provincial and territorial jurisdictions. Some of these jurisdictions do not have an Organized cervical screening programs.

    Each Canadian jurisdiction needs to assess technology individually. Technology assessment in Canada is not assessed as a pilot and rolled out as a National Program if effective as is the process in other countries. Countries like Australia, the Netherlands, the United Kingdom, etc, have taken this approach. As a result, incorporation of HPV Testing as a Triage method to determine more significant low grade lesions was adopted over a decade ago. These same countries are positioning themselves to use HPV Testing as a Primary screening test. They have demonstrated effectiveness and cost effectiveness but more importantly planning to use a better primary screening test than the Pap test. The Canadian Task Force on Preventive Health Care in 2013 wasted an opportunity to lead Canadian jurisdictions on the values of HPV testing and Organized Cervical Screening Programs.

    Cytology will still have a place. Primary screening with cytology is still necessary in women 21 to 30. A suitable triage test is also necessary for women having positive HPV tests. It would be more advantage to using the Pap test to triage these individuals rather than sending all women with positive HPV tests for colposcopy. We would need then to rethink how cytology is incorporated form a primary screen to a triage use in women who test HPV positive.

    Unfortunately, cervical cancer occurs in women who do not participate in screening or participate infrequently with testing well beyond recommended intervals. Organized Cervical Screening Programs have successfully used novel methods to educate women on the positive sides of screening to improve screening rates. They share these screening initiatives through the Pan-Canadian Cervical Screening Initiative(PCCSI) and the Canadian Partnership Against Cancer (CPAC). HPV technology will help improve participation with self-testing in this unscreened/under screened population.

  5. Eduardo Franco

    Some brief comments in response to Dr Dickinson’s and Dr Lotocki’s remarks:

    To Dr Dickinson: The lobbying is strong indeed, as it should be by those who understand the science and have contributed to it, given the inaction that has persisted despite the overwhelming evidence. Cohorts of vaccinated young women will reach screening age soon. Pap cytology is a technology that is not suitable to screen under conditions of low lesion prevalence. The vast majority of positive results will be false, whereas screening via HPV testing followed by cytology as a triage step has the requisite performance characteristics to serve as the complement to HPV vaccination during the next 2-3 decades. Ultimately, we will abandon cervical screening altogether because when all birth cohorts received the protection from vaccination the harms from screening will exceed the benefits.

    As to the gratuitous and hurtful comments about conflict of interest (COI), I submit to you that experienced scientists should view the opportunity to advise industry as a duty and a privilege that their opinion may come to shape the quality of new technologies to be brought to the public. It is highly improbable that a pharmaceutical or biotechnology company could obtain the same wealth of knowledge and insights from a lay advisory board, and that would be detrimental to the public. Scientists who feel that others perceive their opinions as tainted because of their record as advisors to industry should not fear to reveal to the community that there is duty and distinction in that role and that medicine and public health are all the better for it. On the same value system, it would have been silly of me to claim that Dr Dickinson has a COI because he stands to lose from a future world in which there will be fewer medical procedures stemming from a rational use of screening technologies.

    To Dr Lotocki: There is no proposal for HPV primary screening to exclude a triage step. Pap cytology should serve as that step for the foreseeable future. On another point you made, HPV testing affords us the opportunity to reach women (via self-sampling) who are missed by screening programs or resist attendance because of cultural or religious concerns. On yet another of your points, eventually, HPV primary screening will prove valuable even for women under the age of 30 years after HPV vaccination eliminates the excess HPV infection prevalence that currently exists in the years post-onset of sexual activity. In any case, when we get to this point, screening will have to be redesigned altogether to be done much less frequently than is done now.

    As others have said wisely, all screening causes harm, some bring benefits. We have a few years to implement a rational change in the existing paradigm of placing a subjective, expensive, and poorly reproducible technology as the anchor screening test for cervical cancer. Pap cytology served us well while we did not have anything better. We compensated for its deficiencies (high false negative rate and poor reproducibility) by doing it too early in life and too frequently over a lifetime.

Submit a comment