Screening of Ovarian Cancer - HD

Introduction

Hi, I am Debbie Levine.

I'm from Beth Israel Deaconess Medical Center in Boston and Harvard Medical School.

And today I'm gonna be talking about screening for ovarian cancer.

My talk today is on screening for ovarian cancer and the purpose is to discuss the findings of early ovarian cancer on ultrasound to describe the use of ultrasound and ovarian cancer screening.

Discuss how tumors progress or how they do not from benign to malignant variants and discuss how to diagnoses that do not require follow-up so that some of the expense and morbidity of screening can be avoided.

Role of CT and MR in Early Diagnosis

Now I wanna be very clear that CT and MR do not play a role in early diagnosis of ovarian cancer.

CT actually plays no role in early ovarian cancer, diagnosis.

If they're small cysts, that we would need to find, if we wanna find early cancer, we're not gonna be adequately able to evaluate them on ct.

And MR is really used to differentiate lesions where we need more information, but it's not a screening test.

It's too expensive and it's not sufficiently widely available.

So we use it for problem solving to avoid surgery for benign lesions.

So here I've got some images of a normal ovary on mr.

I've got an endometrioma very bright on T one weighted images, and I've got some classic appearances of dermoids, where we've got the fat going from bright, to dark on this fat saturated sequence here.

But basically Mrs for problem solving.

Ovarian Cancer Statistics

So let's now think about ovarian cancer and the statistics.

This is back in 2008 and you can see that ovarian cancer is a little bit down on the list of female cancers, but it's still relatively common.

And it's also a relatively common cause, of death in women.

When we look at new cases and deaths.

They have been slowly going down over time, but the five year survival rate is still only about 44.6%.

The problems with ovarian cancer is that women don't present with symptoms until quite late.

There's really a lack of early symptoms.

And so when women present, they're at a more advanced stage when the prognosis is poor.

And so at diagnosis, only 20% of women have disease that's still confined to their ovary.

Now, when the disease is confined to the ovary, the survival is of course much better than when it's more extensive and there's a 92% five year survival for this limited confined ovarian cancer.

It's been estimated that if we were able to screen for ovarian cancer, if 25% of women currently diagnosed with stage one disease were increased to 75%, we could decrease the mortality of ovarian cancer by 50%.

And that's why screening is such an interesting concept if we're able to do it well.

Available Screening Tests

Now, screening tests that are available for ovarian cancer include a physical exam, which I'm sure you know is not very reliable.

It's very difficult to adequately feel the ovaries ca 1 25 a serum test, which is also not very reliable.

In transvaginal ultrasound, and you'll notice I didn't say ultrasound, but I did say transvaginal ultrasound.

'cause in order to do a good job of screening, you really need to get up close to the ovaries to be able to see the internal characteristics well, and that's gonna be with a transvaginal approach.

Normal Appearance of Ovaries

So we need to then review the normal appearance of the ovaries.

And the normal ovary in women of reproductive age has a varying appearance throughout the menstrual cycle where you'll have lots of developing follicles in the first week and a dominant follicle, in the second week.

A rule of thumb is up to a centimeter in the first week and up to two centimeters in the second week.

But of course there's a lot of variability, particularly once the corpus lutetium develops.

And a lot of times we'll see more than one corpus lutetium in the ovary.

That could be because the woman has ovulated twice, or it could be that one corpus lium is developing while another one is regressing from a prior cycle.

Now, some people call, simple cysts that are small within the ovaries follicles, and some people call them simple cysts.

The important thing is to know that in order to say that it is completely benign and that it's simple, it has a thin wall, the wall is smooth, it's either round or oval, it's completely anti coic.

And that's why we need our transvaginal transducer.

There's no flow within the cyst and it's small, less than or equal to three centimeters.

That's the classic appearance of a follicle in a normal ovary in a woman of reproductive age.

Benign Cysts

Now benign cysts.

And here we're using cysts to say, that they're larger than what we would expect for a regular follicle, again, can be described as simple cysts, and they have the exact same sonographic characteristics, thin wall koic, and through transmission.

And here's one that's 6.4 centimeters, and the size really doesn't matter when we're talking about the internal characteristics.

What happens though, as a cyst gets larger and larger is that it can be more difficult for us to see the entire wall with the transvaginal transducer.

And that's why when a cyst gets very large, you might actually need transabdominal ultrasound to evaluate the entire cyst.

And why we end up recommending Mr for very large cysts in order to see the entire wall of the cysts.

That might be out of the range of our transvaginal transducer.

Complex Cysts

Well, let's now move on from simple cysts to complex cysts.

And I've put on this slide a variety of appearances that we commonly see when we do pelvic ultrasound.

We've got a hemorrhagic cyst where we've got these very acute angles of the clot as it retracts a dermoid, a very echogenic mass with shadowing hydro cell.

Pinks a long tubular structure.

And it would be great if every time one of these lesions occurred, we'd be able to clearly say what they are.

But they all have variety of different appearances that people who do a lot of ultrasound can just say that's what this is.

This for example, with, the layering debris, you might think about an endometrioma, you might think about a dermoid.

Here we've got a cobweb appearance here.

We've got the retractile clot and no central flow.

So for the corpus lutetium, very, very common appearance is the thick walled cyst with a granulated appearance and flow around the wall.

You might see internal echoes within the cyst because of clot, hemorrhagic cysts, reticular pattern, internal echoes, and sometimes there'll be a little area that you might think is solid.

You can jiggle it with the transvaginal transducer and see that it moves kind of like jelly.

You can also show that there's no flow and you need to be clear that if you see a little spec of flow, you need to put spectral doppler on it to make sure that that flow is real.

So you don't wanna say that there's something solid within something that looks like a classic hemorrhagic cyst like this one with a cobweb appearance.

Importance of Ovarian Cysts in Cancer Screening

So the question arises, why do we care about ovarian cysts?

And the answer is because greater than 85% of ovarian cancers are cystic.

The next question that arises, do malignant tumors always have a complex appearance or do they sometimes begin a simple cysts?

Because if they began a simple cyst, we'd have a much easier time of screening for them.

So here's a simple cyst, we're not worried about it.

And here's a very complex lesion, lots of solid ular elements and lots of blood flow, clearly a malignancy.

And it turns out, that tumors of low malignant potential do go through this dysplasia to carcinoma sequence where you eventually get gross invasion, but there are precursor lesions that you can see.

But the ovarian cancer, that's the bad actor, the one that really leads to death.

We haven't seen this very easy classic dysplasia of carcinoma sequence, and that's what makes it so hard to screen for.

In fact, many people now believe it's not actually ovarian cancer per se, but that it actually begins in the fallopian tube.

I so cystadenomas and cyst adeno fibromas.

The benign lesions that we see in the ovary may be precursor lesions to low malignant potential tumors and low grade carcinoma, but the rate of transformation is exceedingly low, and that's why we can follow cysts that have a benign appearance.

Measuring Ovarian Cysts and SRU Consensus Recommendations

So what do we do when we're looking at ovarian cysts?

Well, we wanna make sure that we measure them in three planes and that we make sure that the pressure from the vaginal probe is not too hard.

Because if you think about a cyst that's round like this and think about what you do with a vaginal transducer, if you press hard, you can make it look ovoid.

And since when we're talking about ovarian cysts, we give the largest measurement.

You can make a cyst look larger if you press too hard.

So you want to measure when you've got the cyst looking as round as possible and use the maximal diameter of the cyst in the recommendations.

And what recommendations am I talking about?

Well, the SRU put together a consensus conference on a NAL cysts.

That's the Society of Radiologists and Ultrasound.

And we got together a lot of experts in gynecology, including gynecologic ultrasound and, reproductive endocrinology and gynecologic oncology and minimally invasive surgery.

We had radiologists, we had epidemiologists.

We talked about how we should follow a lot of these cysts because the really benign cysts, the ones that are never going to become cancer, it would be wonderful if we didn't have to follow them so often.

Whereas the bad acting cysts, the ones that, probably are malignant, those are the ones that we would like to recommend for surgery as early as possible.

So it's really important to understand when you're thinking about screening for ovarian cancer is that the vast majority of cystic aneal lesions are benign.

Many, many lesions.

Many, many cysts have a typical sonographic appearance that allows you to make a confident determination that this is a benign lesion.

And most of these do not require follow up.

Now, there are of course, some lesions that have indeterminate sonographic features that don't allow a confident diagnosis to be made.

And those are the ones that we might need to get follow up on, or we might need to recommend a different imaging modality.

Ultrasound Technique and Doppler Use

Now it's important with ultrasound that we do the best job that we can.

So there's some technical factors we need to discuss.

And the first one is, as I said before, that we need to start with vaginal scanning.

If you just look at this lesion here, transabdominal, here's our full bladder, you can see that it's measuring around five centimeters and it's really difficult to tell if it's cystic or solid.

And when you look with a vaginal transducer, you can then tell that this has a very heterogeneous course solid, appearance.

So we do a transvaginal scan for the internal echo texture.

And again, we might need a transabdominal study, to look at the entire appearance of large cysts.

What about doppler?

Well, color or power doppler ultrasound is definitely needed for the evaluation of most cysts, but we're not looking at the resistive index.

What we're doing instead is looking for solid elements with flow.

So we're using doppler to figure out if a lesion is completely cystic, and if we see something that looks solid, either a solid area or a septation looking for flow within it.

So here we have a transabdominal image of a lesion.

It even has through transmission.

You can definitely see some internal echoes within it.

Difficult to tell on this image if this is a complex cyst or a solid lesion.

But once you turn on color, you can clearly see branching vessels.

Clearly a solid lesion.

Why do we then use spectral doppler?

Well, it's to make sure that what we're seeing within the cyst is real and not an artifact.

So resistive index and pulsatile index RI and PI aren't sufficiently better than gray scale morphology assessment of tumors to assess if something's malignant or benign.

But the presence of flow in a solid element is the most important doppler feature.

So here we've got a very, very small cyst with a tiny nodule with just a little bit of flow.

And this is what's going to make you think that this is, potentially a malignant lesion.

And indeed, this was a very early stage ovarian cancer.

So this spectral waveform analysis, we do place spectral doppler on a small element like this to make sure that the flow is real.

But the RI and pi, are not as important.

Patient Age and Solid Elements

What about patient age?

Here we've got a relatively, small lesion.

It was about 2.8 centimeters in a 33-year-old.

It had a septation and most importantly, it had this heart shaped solid element within it.

And we really don't care if this woman is 33 years old or 93 years old, once there's this solid element that's real.

That's not just the granulated wall from, hemorrhagic corpus lium.

This is, neoplasm until proven otherwise.

Common Neoplastic Lesions: Serous and Mucinous Tumors

So speaking of neoplasms, let's then discuss the common appearance of the most common cyst adenomas that we'll see, which are the serous and mucinous tumors.

The serous tumors are thin walled cysts.

They're usually ocular, but they may have, some septations within them and they may or may not have papillary projections.

And the larger the papillary projection is, the more blood flow there is, the more we're going to worry about malignancy.

Mucinous tumors are also thin walled cysts.

They can be multilocular, they can have thin septations, and the components can differ in echogenicity so that when you see course internal echoes like this and multiple septations think about a mucinous tumor.

And then here are lesions, that you're not quite sure what the underlying etiology is, but you've got solid elements with flow, solid element with flow, and that's going to make you think of, malignant process.

I also really like color to look at vessels.

If you see something like this, the tangle of vessels, and here's the cyst that we're actually evaluating, it's very, very helpful to show what's a cyst and what's actually vascular.

So here's that early, cancer that I showed you before, 2.2 centimeter cyst.

The answer is yes, we can find early cancer, but what we don't wanna do is call every single cystic lesion that we see within the ovary.

Worrisome for neoplasm.

IOTA Simple Rules for Ovarian Assessment

Now, there's a very active group, in Europe, the IOTA group that's developed some simple rules for when assessing the ovaries.

And those are, that there are benign features that we should think about a ocular cyst.

There can be a solid component, but it needs to be small, less than seven millimeters.

Acoustic shadows are a benign feature, a smooth multilocular tumor, even if it's less than 10 centimeters and no intratumoral blood flow.

So although there's a solid component here, it won't have blood flow.

And that's because, the, the cyst adeno fibromas are incredibly common and those are benign lesions.

Now, malignant features, irregular solid tumor ascites at least four papillary projections, irregular, multilocular solid tumor greater than 10 centimeters, and very strong intratumoral blood flow.

So here's a study looking at the strategies to diagnose ovarian cancer and actually evaluating these rules from the IOTA group.

They had over, 2,400 patients, slightly over half of them had benign masses.

41% had malignant masses.

And they did have, a nice number of, primary invasive stage one cancers.

5% of the lesions, were stage one cancers.

So how did these simple rules do?

These are the three different evaluation methods, and I won't go into each of them, but it turns out when you look at the sensitivity and specificity of the simple rules here, that they actually did very, very well.

So they looked at the area under the curve of how different rules did, and I'm, pretty much ignoring, a lot of the details, but basically, simple rules here, and simple rules in post-menopausal patients.

And they found out that the simple rules did very, very well.

So benign lesions, unilocular cyst, very, very common, less common in malignant lesions.

But it turned out that they did have some malignant lesions that they had characterized as unilocular cyst.

And you have to figure that they probably missed the small solid component.

What about that rule?

The largest diameter of the solid component being less than seven millimeters.

Again, it was a pretty rare finding.

3% of the benign lesions, 0.2% of the malignant lesions, acoustic shadows, 19% of the benign lesions, 3% of the malignant lesions.

What about a smooth multilocular tumor?

Less than 10 centimeters, 16% of the benign lesions and 1% of the malignant lesions, 10 centimeters, that's a very, very large tumor.

And yet what you have to remember is that it's more likely to be benign than malignant.

No intraoral blood flow, at 40% and 3% respectively.

Pitfalls in Diagnosis: Corpus Luteum

So what do we do when we see a lesion like this Transvaginal scan?

We're getting a good look at the cyst and it really doesn't look like a corpus luteum because it doesn't have a thick wall all the way around.

You're gonna be worried about a cyst like this turn flow on.

You get blood flow in these small solid elements, but also blood flow around the wall of the cyst.

You look in a different imaging plane, and here it's looking more like a corpus lutetium.

And this is what we need to be very, very careful of, is to not forget that the corpus lutetium can have an appearance that looks like ovarian cancer.

And so what you need to do when you see a lesion that you're not sure about, the woman is of menstrual age.

Look back.

If she had no lesion two weeks ago, for example, you're not gonna worry about it.

Now, ovarian cancer, even if it grows fast a day a two or two or a week, is not gonna make a difference.

Sending somebody to surgery for a corpus lium, of course will.

And the nice thing about a corpus lium is that it changes very rapidly.

So because we were worried about this patient, we brought her back very, very short interval.

And less than one week later, this is what that cyst looked like, much more benign appearance.

And since it had changed so incredibly rapidly, clearly a physiologic cyst.

So solid elements with flow are the key keynote to diagnosing cancer.

But you just need to be aware of the corpus lium and make sure that you don't make a mistake.

Short interval follow up can be your friend.

Now you might say, why do we then usually recommend six week follow up for cyst that we're not sure about?

That's a cyst that we really think is most likely benign, and we wanna give it time to resolve.

We want to have it in a different phase of the menstrual cycle.

And that's why six weeks is typically the timing interval for look following up a cyst.

But if I'm very worried about a cyst like this one, I'll do a very short interval follow up.

Screening Programs and Studies

Well, let's now look at the, what other screening programs have done and how successful they've been.

And another big group is that, of the University of Kentucky, and they've done some very, very large, studies looking at screening for ovarian cancer of, 25,000 women.

There's another study, the PLCO trial in the United States, over 78,000 women, postmenopausal women in this case getting annual screens with transvaginal sonography and CA 1 25.

What they considered as abnormal was ovarian volume greater than 10 ccs, an ovarian cyst volume, of greater than 10 ccs, any papillary, projection extending into the cystic cavity of any size and any mixed cystic or solid component within a cystic ovarian tumor.

How did this study do?

Well, they had, 28,800 women undergoing the baseline screening.

4.7 uh, percent had an abnormal vaginal sonogram.

1.4% had abnormal ca 1 25 and in 0.1% were both abnormal and they were able to detect 29 malignancies, 20 were invasive cancer.

But a large number of women, 570 women underwent a surgical procedure as follow up.

So that meant that 541 women underwent surgery but did not have cancer.

And, and this is the problem that we have in screening for ovarian cancer.

We're taking out a lot of benign disease.

So the predictive value of these screening tests were 3.7% for an abnormal ca 1 25 and 1% for an abnormal transvaginal ultrasound.

And really for screening tests to be useful and clinical clinically applicable, we'd like, that value to be more like 10%.

Now, the um, 23.5% was the predictive value of both tests were abnormal, but having an abnormality in both tests, while it had a fairly high predictive value, only nine of the 29 tumors or 31%, would've been found.

So if we look at a lot of these different, screening studies, there's also a multicenter study in Japan, that I didn't uh, mention.

They're looking at different tests, mostly ultrasound, some add ca 1 25.

They have nice large numbers of patients that they're following.

They have a reasonable number of invasive cancer and a reasonable number of early stage cancer and their survival benefit is coming out soon.

And that's what's so incredibly important.

We really need to show that we're doing, more benefit than harm and that's why these long-term survival benefit outcome studies are so incredibly important.

Screening in Postmenopausal Women

So let's move back now to talk about screening, in postmenopausal women because I've spent quite a bit of time talking about premenopausal women who can be very, very complicated to diagnose because they have the normal physiologic changing appearance in the ovaries and they have all of those complex appearing cysts due to the hemorrhagic corpus lutetium and many other benign lesions.

Postmenopausal women are different.

They're, they're harder to scan, because they might have more abdominal and pelvic fat, they have atrophic vaginitis, which might make it difficult to insert the vaginal probe or might make the patient or even the sonographer less comfortable with doing the exam.

And because the ovaries lack follicles, they can be difficult to visualize.

So here's classic postmenopausal ovaries.

They're hypo coic, they're almond shaped and the volume really varies depending on what um, study you look at.

And these are really old studies.

They used a mixture of transabdominal and transvaginal techniques and they came up with a huge range of normal ovarian volume, from less than a centimeter up to 10 centimeters cubed.

And it's really, really difficult.

Therefore to know when you've got an ovary that's an abnormal size.

And a rule of thumb, is to compare it to the opposite side.

Another rule of thumb is when you see an ovary that looks like it's too big, make sure you're really dealing with the ovary because a lot of times I think what people are measuring is the ovary, is actually a loop of bowel.

And when you look at some of these studies and look at the percent of ovaries visualized, you can see one of these studies had ovaries visualized 99% of the time and actually had volume.

That's scary up to 14 ccs.

And I think that's because they probably had, bowel loops visualized rather than actually ovaries.

So when we see a post-menopausal, a NAL mass, that's an koic cyst just like any other cyst, then while the koic very smooth through transmission, most likely it's benign.

Now if it's large, it could be neoplastic, probably a benign serous tumor.

But again, it's probably benign if there's a complex cyst, unlike the premenopausal woman, postmenopausal complex cyst is not gonna be a corpus lium unless a patient is just in that perimenopausal or early postmenopausal period and she's recently ovulated.

If a woman is in that perimenopausal or early postmenopausal period, you might wanna give her a short interval follow-up to see if a complex cyst resolves.

But just like premenopausal women nodularity in septations in pre the likelihood of malignancy, if there's a solid mass, most commonly it's gonna be a fibroma, the coma or metastatic disease, less commonly undifferentiated carcinoma.

But the fibroma, the coma are going to be the most common solid masses in postmenopausal women.

It turns out that these postmenopausal cy are incredibly common.

They become less frequently observed as a patient progresses through the menopause transition.

And there's no doubt that some cysts in early menopause reflect an ovulatory event.

There's also no doubt that many of these cysts are para ovarian, are actually tubal in etiology and even late menopause when ovulation is unlikely to occur.

Small simple cysts up to a centimeter are frequently seen in up to about 21% of women.

Here's a study looking at over 3,500 postmenopausal women.

They found, simple cyst in 6.6%.

They operated on 18 of them.

And while most commonly they found sero cyst adenomas, they also found cyst adeno fibroma, which means they probably missed the small fibrous portion, a dermoid that typically would not be seen as a simple cyst, two normal ovaries and a fibrotic coma, adeno fibroma.

These are both solid lesions.

So these are examples of lesions that were misdiagnosed.

Granted, this is a a very old study, but here's a more recent autopsy study from 2003 where they scanned ovaries that had been removed at autopsy and they found simple cysts and postmenopausal in, women that were of an age to be postmenopausal in 54%.

They found solid benign lesions in 12%.

They then asked the pathologist how many more lesions were present that were not visible on ultrasound, and the pathologist found an additional solid lesion and an additional eight cysts that varied between one and eight millimeters.

The conclusion of that study was that small benign and Neal cysts and small benign solid tumors are so common in postmenopausal women that their presence may be regarded as normal.

And this is a very, very interesting statement and I'm more than willing to agree that when we see small benign appearing simple cys and postmenopausal women, we can probably either ignore them or not recommend follow-up for them.

The small solid tumors, it's different when you look at them directly at ovaries that have already been removed.

And so we're probably seeing lesions that are a little bit larger when we can see them with our transvaginal scanners.

So I think it's still worth, following these, but do realize that a lot of them will be these benign lesions, the fibromas and the Theas.

Consensus Guidelines for Management

So to get back to the sonographic appearance, simple cysts, whether they're ovarian or extra ovarian, regardless of age, are almost certainly benign.

At that consensus conference I mentioned earlier, we decided that seven millimeters was a reasonable cutoff where if you're not going to do surgery to remove this, if you wanna follow it instead you might wanna do another imaging test to make sure you haven't missed any nodularity.

And that other imaging test would be Mr, why do we talk about size?

Well, there's no doubt that cancer does increase as the size of the lesion lesion increases.

And when you have a lesion greater than 10 centimeters, a rule of thumb is about 13% chance of being malignant.

But you can have a cyst this large more than 20 centimeters and have a be benign.

This was a benign ser cystadenoma and you can have a lesion this small and there's a small solid nodule here.

This is that 2.2 centimeter early ovarian cancer.

So really the size can be helpful, but it should not completely guide what you think about the cyst as far as it being benign or malignant.

So the recommendations that we came up with at our consensus conference was that in a woman of reproductive age, because physiologic cyst are so incredibly benign, if they're less than five centimeters, they don't need follow up.

We do recommend describing them in the report if they're between three and five centimeters.

If they're greater than five centimeters and less than seven centimeters, we actually recommended yearly follow up.

And if they're greater than seven centimeters, you can suggest that further imaging might be helpful.

Post-menopausal women, we had a lot of trouble figuring out at what size should assist be followed and at what size do you need to continue to follow one after one fo uh, follow up shows no change.

We all agreed that less than a centimeter are incredibly common, and so we, set that threshold for recommending follow up at one centimeter.

And there's no doubt that in the future that size might even increase.

We might be able to not recommend follow up for smaller cysts.

So we ended up recommending yearly follow up if some of these cysts were between one and seven centimeters, potentially less frequent follow up after initial stability is documented.

And then again, once a lesion got larger, greater than seven centimeters, to consider getting MR or surgical evaluation for indeterminate features multiple thin septations.

Again, if they're thin, it's probably gonna be a benign neoplasm, probably a cystadenoma if there's a nodule that's not hyper coic but does not have flow, that's suggests a benign knee neoplasm such as a cysa no fibroma.

And so you can have a small solid element like this, what to do for these.

There's no doubt that many, gynecologists are going to feel uncomfortable.

Many patients are gonna feel uncomfortable if they know there's a solid element and so many of these women will go to surgery.

We're hopeful that as we gain more knowledge of these, that we can actually follow them.

You do wanna carefully look for blood flow within that solid nodule malignant features, thick septations greater than three millimeters, nodules with flow, focal wall thickening, any of these lesions you should think about, being malignant and recommend surgical consultation.

It turns out that a solid element with flow is the most worrisome finding.

It has the highest likelihood of being malignant.

And so if you have a cyst like this thin wall, solid element clearly has flow.

And again, I would document that with spectral doppler as well.

You don't need to recommend any follow up, for this at all.

These patients can go straight to surgery.

Now, other signs of malignancy and metastatic disease would be a large amount of ascites and extra ovarian extension.

Principles of Ovarian Cancer Screening

So let's now go back to the screening principles.

When you want to identify disease, when you want to set up a screening program that's going to have an impact, you need to make sure that there's serious consequences of disease, which clearly there are ovarian cancer.

You wanna make sure that treatment is more effective when the disease is in the earlier presymptomatic phase.

And there's no doubt at all that ovarian cancer can be more likely cured when it's at an early stage.

Stage one in particular, you wanna have a detectable preclinical phase that's long enough to allow for screening intervals.

And this implies a consistent progression from stage one to stage four in the invasive characteristics.

And this is where we have a lot of trouble with ovarian cancer because as I said before, the very bad acting ovarian cancer, we don't know how to find that preclinical phase very well with ultrasound and it definitely doesn't last very long to allow us to screen at yearly intervals.

The preclinical phase should have a high enough prevalence to justify the expensive screening.

So when we get into actual numbers, if we assume 50 cases per a hundred thousand and an incidence of 13.8, in a hundred thousand women over the age of 40 a test with 99% specificity and a hundred percent sensitivity, both of which, are over estimates, for ovarian cancer would only give us a one in 21 positive predictive value.

So you can see we need exceptionally high specificity to have ovarian cancer screening work because any decrease, in specificity is going to be associated with a large drop in this positive predictive value.

Challenges and Future of Screening

So the, since we have so many problems with ovarian cancer screening, what we need to figure out is how to treat benign ovarian lesions.

Because if we continue to surgically remove them, a lot of women will undergo unnecessary surgery if we continue to watch them all their ethical, psychological, and no doubt cost implications as well.

So for the future, we need to have a test, a screening test that's of lower cost, both in regards to how expensive it is to do transvaginal ultrasound, how and how expensive it's to follow these, patients.

We need to have large scale studies that follow patients over decades to compare screened and unscreened population.

And it's been estimated that these trials need a hundred thousand women in study and control groups and follow them for 10 to 20 years.

Well, you'll notice many of my slides, gave dates that started 10 to 20 years ago.

And it turns out for a lot of these, ovarian cancer screening trials, the future is now.

So the UK collaborative trial, 200,000 women, is at seven years of follow up that, national Cancer Institute PLCO trial, 74,000 women, 2014 is the 23rd year of this trial.

And so we're definitely getting these results out soon.

Conclusion

So as I summarize this talk, ovarian cancer screening in a low risk population right now is not cost effective.

And that's due to the low incidence of disease, due to the expense of the studies and the high false positive rate with all of these benign lesions that we pick up with screening, it's associated with significant morbidity due to surgery for lesions and, even in patients with a positive family history.

At this point, it has no proven benefit.

The issue is that we can find early cancer, we know we can find early cancer and this is what we would like to do.

So if you're doing ultrasound, yes, please do notice these early cancers.

Please do intervene, but we're going to have to wait just a little bit more time for these large studies and for the cost effectiveness analysis of screening, to know if this is worthwhile to set up on a large, scale basis.

So thank you very much.