Sangini S. Sheth, MD, MPH

Excellent. As we begin to wrap up this conversation, does anybody have any additional comments they'd like to make? Dr Sheth? Dr Hill?

 I ust wanted to comment that we didn't really get a chance to talk about this, but there are, I think, also some studies looking at artificial intelligence and deep learning algorithms to implement some of these technologies that we have for screening; triaging women with HPV related issues is also a kind of an exciting development. Again, we don't have a lot of time to go into it, but a lot of interesting studies are supporting that as a new technology that we can use for helping managing HPV related diseases.

Essentially using visual methods with artificial intelligence trained algorithms?

 Yes, my final thought is we focused a lot today on screening, of course we talked about vaccination, and the final piece is that women who have or people who have abnormal results, a key component of that is the linkage to care to follow up on abnormal results, and I think ultimately the goal of this conversation is to close the gap on cervical cancer or even working towards elimination. The last piece of that is that linkage of care of abnormal results, whatever form of screening, we may be using.

Right, very important points, clearly. Dr Stoler?

 I'd emphasize that last point that the whole purpose of screening is to find the precancer, and if you can't treat the precancer, you're not doing anything with screening. That being said, life is getting a little more complex. We talked about extended genotyping, dual stain, different algorithms. How is the clinician and/or patient supposed to keep this all straight? I think more important than artificial intelligence apply to Pap smears [Papanicolaou tests] as a way of improving the performance of pap smears is the development of information systems and apps like the ASCCP [American Society of Colposcopy and Cervical Pathology] has developed that will essentially allow the app or the program to acquire the data from the patient record. What's the Pap history? What's the HPV test? What's the type? Was dual stain done to then give the clinician more direct guidance? You don't have to remember what the HPV type is. You just need to know whether the patient is safe to get retested in 5 years vs they have to get a colposcopy vs their risk is so high that no matter what the colposcopy shows you might end up treating them anyhow. It's that synthesis of clinical data that we now have so much good data that really says we can more precisely at the individual level synthesize these data and give you a risk that clearly tells the clinician what should happen to the patient. I think we're going to see more and more of that as we go forward because clinicians don't want to really know which test and they don't have time to become necessarily as expert in all the nuances of laboratory testing and data analysis that the data really support.

Excellent. I'd like to thank all of our panel members for a really informative discussion, and I'd like to thank the viewers for watching this 

 viewpoint. If you enjoyed the content please subscribe for the newsletters, E-newsletters, to receive upcoming viewpoints and other great content right into your inbox. Thank you all so much, have a good day.

Videos

Drs Emily K. Hill, Sangini S. Sheth, and Mark H. Stoler share practice pearls for HPV and cervical cancer screening and express their hope for emerging testing options.

Key Takeaways for HPV and Cervical Cancer Screening

The other approach to triaging HPV [human papillomavirus] positive women is to do genotyping. I think everyone is aware of the importance of knowing whether a woman is HPV 16 or 18 positive. We already alluded in this conversation about the high rate of cervical cancer precursors in women are HPV 16 positive. It's about 25 of those women will become CIN2 or 3 positive within 3 years of being diagnosed as having HPV 16. What about other genotypes? Some people are discussing extended genotyping which means you would look for other HPV genotypes such as 31 or 35. Dr Hill, are you aware of that data and have any thoughts about it?

 No, I'm sorry I don't know that I have enough expertise to speak on that. I might have to defer to the other panelists.

 I don't know the specific data except to know that there are increasing commercial tests available. I think I just saw something earlier this week that 31 was added to one of the assay journals and I often teach that this is an area that we're going to see change over the course of our clinical careers, the more specific genotyping.

Okay, fair enough. Dr Stoler, you were involved in some of those studies.

 Yes. I think there's more than 1 way to skin this cat, if you want to use that old phrase. The ASCCP [American Society of Colposcopy and Cervical Pathology] has promoted for many years in their guidelines equal management for equal risk. When HPV primary screening was FDA approved, it incorporated partial genotyping, namely HPV 16 and 18 genotyping because of that risk that I've already talked about with particularly HPV 16. If you're HPV 16 positive, you need to get a colposcopy. I don't care what your Pap smear [Papanicolaou test] shows even though Pap smear abnormality will stratify that risk even higher. You have ASC-H and HPV 16. Your risk is so high that you might consider treating the patient without biopsy. What we found out in epidemiological studies over the last 10 years is that the 12 other types that are masked in most of the prior FDA approved assays have differential risks. Just like 16 and 18 have the highest risk, we now know that 31 and 33 in the New Mexico study, (52 if you're in Asia, etc.) have risks that approach the risk of HPV 18. If you're going to take an HPV 18 positive person to colposcopy, why shouldn't you take the HPV 31 positive person to colposcopy? There is an assay on the market today, the Onclarity assay, that has gotten an FDA approval to reveal extended genotyping because it has the capability of unmasking the data that were already there. Other assays don't have that capability, but you can use dual stain to perform equivalently. I think what you're going to see going forward from the NCI [National Cancer Institute] and during guidelines project that's working with practices and manufacturers is that there'll be different ways depending upon which HPV assay is offered in the laboratory to get the right people the colposcopy in their triage step. You might do extended genotyping. You might do dual stain. In some places you might do even a combination, but the point is that not all HPV types are of equal virulence, if you will, in terms of promoting the development of precancer or the rate of transformation to invasive cancer and that we now have that data, so why not use it. I think you're going to see in some practices, HPV extended genotyping where they take into account the other genotypes as a way of improving triage specificity.

Experts in obstetric and gynecologic management provide insight on the role of HPV genotypes in cervical cancer risk.

Role of HPV Genotypes in Cervical Cancer 

As we move to HPV primary screening, the big issue is going to be triage of HPV-positive women. Currently, when we cotest, we use cytology as a triage. If you've got an abnormal cytology, you're managed differently than if you have a negative cytology, depending on whether or not you've got HPV 16 or 18. There are other triage methods which are currently available and under development. One of those is CINtec PLUS which is testing for p16 and Ki-67. P16 is a small molecular way marker for cell cycle progression as is Ki-67. Normal cells will not coexpress p16 and Ki-67 whereas neoplastic cells do. We can use that for cytology as a way of determining whether or not there is a CIN really present or not. Dr Stoler, you were involved in a large study which got FDA approval of CINtec. Can you talk about that?

 The IMPACT trial which led to the FDA approval as well as data retrospectively that you and I have published from the ATHENA study basically looked in the primary HPV screening algorithm at the performance of triage test. Either cytology is a triage at various cut points, ASCUS [squamous cells of undetermined significance] or above, LSIL [low-grade squamous intraepithelial lesion] or above vs dual stain cytology which is performed on the same kind of liquid-based cytology slide but is an immunohistochemical test. What we found in those studies repeatedly is that CINtec PLUS or p16 and Ki-67 dual stain as a scientific name for the procedure outperformed cytology by a significant fraction. Remarkably, it does something that's very rare in laboratory testing. It was 25% to 30% more sensitive meaning it found the CIN 3 in the HPV positive population or the precancers of the HPV positive population more effectively than cytology even at the lowest threshold of cytologic abnormality; it was simultaneously more specific at least in some of the analyses. You had the best of both worlds in terms of triage and the implication of these studies and why it was FDA approved is that in a primary screening algorithm you greatly improved the safety and efficacy of the triage procedure. You get the right patients to go to colposcopy and you don't colposcope or biopsy nearly as many women as you would with a cytology test as triage.

Mark H. Stoler, MD, FASCP, provides insight on the role of dual testing in the management of HPV positive women.

Role of Dual Testing in HPV Positive Triage Management 

Let's move on now to recent advances in screening and looking ahead. How do you all foresee cervical cancer screening guidelines changing as more HPV-vaccinated individuals reach the screening age? Do you think this is just going to get more as time goes on as our vaccination rates increase? Dr Sheth?

I think we may see more interest in the adoption of primary HPV screening so a change in the modality of screening. Two, as we have increased screening rates, we could see the intervals shift and we could see the age change of initiation of screening. It's hard to predict right now because the vaccine has been available to us for over 15 years, and we're still so far from that optimal vaccination rate that it's hard to know when we're going to get there.

What is your feeling about why we're so far from the optimal vaccination rate?

 I think there are a few factors at play. The vaccine has some bad press that is not founded in any science or evidence in terms of side effects. Then there's just a lack of understanding that it is a vaccine and therefore it should be offered and recommended to everyone, so there's often this [attitude of], “Well, who needs it.” Trying to assess risk where, like any preventative measure, the point is for everyone to get it. I think those 2 issues combined have made uptake challenging.

Not surprising. We have vaccine hesitancy in the United States given what we've seen with COVID. Dr Stoler, would you address this issue of the performance of HPV testing vs cytology in vaccinated populations?

Sure, but I want to make a quick comment to what was just said. There's a study that was just published. It was in many of the newspapers last week out of Virginia, where they found that 25% of the population haven't even heard of the HPV vaccine. This is 15 years after FDA approval that every pediatrician and family practice doc should be vaccinating their teenagers and 25% of adults have never heard of it.

Sure, Mark. There's been a lot of discussion about the performance of cytology vs HPV as we get more people vaccinated.

 As I said, screening performance is driven by prevalence. HPV vaccinated individuals with Gardasil 9 have a 90% decrease in the rates of infection for the 7 types that cause 90% of all cervical cancers. In those individuals, the only thing that will work is the more sensitive HPV test. I think in this conversation, whether you're doing cotesting because that's what's available or primary HPV, the important thing is if you're going to screen, you have to do HPV testing. Now, as the vaccination goes through the population, the prevalence will drop particularly the prevalence of the most important HPV types. What you'll see is decreasing rates of HPV in vaccinated individuals and decrease in screen positives, and the first things that will drop out are the most important drivers of cancer screening, which are the most high-risk type 16 and 18 that account for 70% of cervical cancer. Lower sensitivity tests like a Pap [Papanicolaou test] will just perform less and less. It's not only less sensitive. It becomes increasingly less specific. We know that most abnormalities on Pap are already nonspecific ASCUS [squamous cells of undetermined significance] cases. When we take ASCUS out, the performance of Pap testing is quite poor. Yet the meaning of those ASCUS cases for the types that are not included in the vaccine becomes less and less specific. Vaccinated populations will be ahead of the game in that their risk of precancer is decreased by almost 90% or more. They still need some screening for a while until the overall population prevalence drops and that screening has to be molecularly based with HPV.

Drs Sangini S. Sheth and Mark H. Stoler share their thoughts on the adoption of primary HPV screening and how they foresee changes to cervical cancer screening as more individuals get vaccinated.

Addressing the Impact of HPV Vaccination Uptake on Cervical Cancer Screening

There's lots of data to show that in the first few months of COVID depending on geographically where the surge was, as clinics and offices shut down doors, screening and preventative care took a huge hit, and that includes cervical cancer screening. It also includes vaccination. It seems that as surges moved on, screening rates did start to increase, but as we continue to go through these waves, I think most places have not resumed to the full rate of screening that everyone was seeing prior to the pandemic.

 Unfortunately, I would say we have problems with screening in Iowa at baseline even before COVID in terms of the number of cervical cancer patients that I see with invasive cancer; it is quite shocking, actually. I think there's definitely national data that screening rates for cervical cancer during the pandemic went down on the order of 80% to 90%, so there was a significant drop. Then I think you also have the backlog of people. You've got people who are aware that they need to do that screening, trying to catch up, and having trouble getting in or again forgetting about the need for that because they put it off for 2 years; then it sort of slips your mind entirely, so getting some of those women that would have come for screening back into our fold is a challenge.

 Certainly, with the older women who are, I would assume, more hesitant to come into the office than the younger women, you probably still have quite a backlog?

 We do. I think I can say that, at least anecdotally, we certainly saw some women with locally advanced or metastatic cancers who we might not have seen if we didn't have the pandemic to contend with in really all of the cancers I take care of including cervical cancer.

 Certainly, in the New York area, the rate of screening, since I read the Pap smears, I do 80 a day, typically abnormal, perhaps now and before the pandemic. During the height of the pandemic through about August, it was down to 2 or 3. People just were not getting screened.

Thomas C. Wright Jr., MD; Emily K Hill, MD; and Sangini S. Sheth, MD, MPH, FACOG, comment on the impact COVID-19 had on HPV and cervical cancer screening in their respective practices.

COVID-19 and Cervical Cancer Screening Rates

 One of the things I'd just like to add to this discussion is that in the ATHENA trial, which was the first HPV [human papillomavirus] primary screening trial run in the United States, we found a lot of high-grade disease in women 25 to 29 years of age. In fact, there were worse CIN 2 and CIN 3 lesions in that age group than there were in women over the age of 40 combined, and half of the precancers were missed by cytology in the 25 to 29 age group. There was a lot of discussion at the FDA panel meeting about whether we should accept that you can use HPV for primary screening in that age group. The panel decided that due to the large burden of disease and the poor performance of cytology in that age group, it would be an acceptable approach. I certainly understand all the issues about complexity and high rates of HPV positivity in the younger age group. Dr Sheth, what would you add to this discussion?

 I think it's just to point out that the primary HPV screening algorithm does incorporate triage of a positive HPV result. Currently, that algorithm, the triage is actually with cytology. We may discuss it later. There are other options on the table. Also, cytology triage as well as genotyping triage partial genotyping. That can help improve the specificity as Dr Hill was pointing out the specificity concerns with primary HPV screening. Otherwise, I think that we just have to work through the implementation issues, and we need our various institutions whether it's insurance or a lot of the quality metrics also need to change how they measure that cervical cancer screening has been performed to include primary HPV screening as an acceptable modality. As those institutions change, then we will see I hope in this country, primary HPV screening also change.

 Certainly, in the New York area ever since the American Cancer Society guidelines came out, what we've seen is that people have begun cotesting at age 25. They heard, “Do HPV screening beginning at age 25?” They said, “Well, I'm not going to do primary screening right now, but they have adopted cotesting, which may be the worst of all worlds.” Dr Stoler, do you want to comment?

 I do want to comment. I have a couple of comments sort of overall on the discussion. The first was just alluded to. When we talk about primary HPV and the concerns about specificity, particularly in young women, besides the data that you brought up about ATHENA and the poor performance of cytology in women 25 to 29, we should really emphasize to the listener or the viewer that primary HPV does not mean just HPV. Primary HPV means do the test and then apply an appropriate triage test to the positives. It's that triage test, as I'm sure we're going to talk about, that brings the specificity and the performance that you're looking for so that you're not sending every woman in her 20s to colposcopy; you're sending the right women in her 20s to get a coloscopy. It's the data that has now been published in several clinical trials in the United States as well as around the world that show that doing the cotest doesn't get you much compared to primary HPV; this means, do a good HPV test, and then a very good triage test. It's not just testing for HPV. The other thing is the meaning of an HPV test. These HPV tests are not screening for HPV alone. They're designed to say, if you have this HPV, you have this risk of having precancer or cancer. For instance, the big surprise from ATHENA and subsequently confirmed in other studies is women with a normal Pap smear who are high-risk HPV16+ have a 10% to 20% chance of having CIN 3 even though their Pap smear is normal. It's that kind of power that really is pushing the adoption of primary HPV.

Drs Emily K. Hill, Sangini S. Sheth, and Mark H. Stoler, comment on the implementation considerations surrounding HPV and cervical cancer screening. 