Ovarian Cancer - Jeannine Villella, D.O., FACOG, FACS

Ovarian Cancer - Jeannine Villella, D.O., FACOG, FACS

Endofound Medical Conference 2017
"Breast, Ovary and Endometriosis"
October 28, 2017 - Lotte New York Palace Hotel

Ovarian Cancer

Jeannine Villella, D.O., FACOG, FACS
Chair Gynecologic Oncology, Dept of OBGYN, Lenox Hill

Good afternoon. I'd like to thank Dr. Seckin and Endometriosis Foundation for having me. This is a wonderful event and very educational, and I hope I know how to use this. Okay, I think I do.

Okay, so my talk today is going to discuss mostly endometriosis and endometriosis associated with ovarian cancer, and how maybe they're not so different. So, the objectives are to explain the characteristics in ovarian cancers associated with endometriosis, review the literature relating to endometriosis and ovarian cancer. Another objective is to discuss surgical conundrums and challenges and to explore our future directions.

So, endometriosis is a highly prevalent disease in reproductive aged women. It's about 10 to 15% of reproductive aged women, 20 to 30% of women with infertility, 40 to 60% of women with chronic pelvic pain, but the true prevalence is unknown because the way to make this diagnosis is actually from a surgical biopsy. 1.6% risk of developing malignant transformation of endometriosis, so it means you have about a 0.72% chance of an endometrioma becoming malignant. And you have a three fold increased risk of developing clear cell carcinoma if you develop ovarian cancer that arises in endometriosis.

So, with ovarian cancer, actually it's not a common disease. It's the sixth most common cancer that effects women. And clear cell ovarian cancer, which is the one that's most associated with endometriosis is a very rare cell type. It's actually only 5 to 25%. The most common cell type in ovarian cancer is high grade serous carcinomas. So, clear cell carcinoma has a strong association with endometriosis. They tend to be diagnosed in an earlier stage than serous. 47 to 81%, stage one and two, which high grade serous is more like 75% advanced stage and only 25% early stage.

So, we know that clear cell ovarian cancer has a very poor prognosis in advanced stages, that it tends to be very chemo resistant. And not all clear cell carcinomas have evidence of endometriosis, so they can arise de novo. So, really, could there be a distinct clinical entity of clear cell carcinoma associated with endometriosis that develops through different pathologic pathways?

So, we know that endometriosis and ovarian cancer ... clearly we've been talking about how they have some similar characteristics where they can both have metastasized for local and distant invasion, so endometriosis can stay in the ovaries, it can go to the peritoneum, it can go to the bowel. We've seen endometriosis in the omentum, which is in the upper abdomen on the diaphragm surfaces. But, it also damages the tissue. It causes an extensive fibrotic reaction, as does malignant transformation of tissue. And we know that this happens in both endometriosis as well as ovarian cancer due to genetic and immunologic factors. We know that there's also inflammatory mechanisms that occur in endometriosis as well as in ovarian cancer that causes changes in tissues that makes surgery very challenging, but also it's an interesting point of view to see how similar these diseases actually may be.

We also know that for years, ovarian cancer does not originate in the ovaries. It's becoming more apparent that ovarian cancers or ovarian type cancers, which also include fallopian tube cancers and diseases that start de novo in the peritoneum, that those cancers don't start actually in the ovary. That most commonly, they start in the fallopian tube. Then they also hormonal etiology, and they share predisposing factors. Endometriosis and ovarian cancer increase in risk with nulliparity, early menarche, and infertility.

The different characteristics is that although endometriosis can be a really horrible disease, especially intraoperatively and symptomologically, but endometriosis is not catabolic as ovarian cancer is, and it is also rarely fatal, although it can cause pretty significant clinical sequelae.

So, what is the potential link between endometriosis and ovarian cancer? Well, despite numerous studies published to date assessing the potential link between endometriosis and ovarian cancer, a firm causal relationship has not been established. Such an association could lead to the reappraisal of current management of endometriosis, so we know there's been a lot of studies that have shown there is some association between the two, so what has not happened or has happened that could make us change our management of the disease?

So, I'm just going to do a brief review of the literature. There is a lot of systematic review of epidemiologic studies on ovarian cancer and women with endometriosis, and endometriosis and ovarian cancer patients. These review looked at all papers every published before 2014. So, these were done with pub line med searches for key word searches for endometriosis, ovarian cancer, neoplasm, malignancy, and tumor.

So, case control studies, these several have shown ... Ness et al. looked at patients with ovarian cancer and self reported endometriosis, and they found that patients had an increased risk of developing ovarian cancer if they self reported endometriosis. The next study by Ness showed that patients with infertility was associated with ... endometriosis was associated with endometrioid, adenocarcinoma and clear cell ovarian cancer. Borgfeldt and Andolf, patients that were discharged from the Swedish hospitals, they used the Swedish hospital discharge register, and they looked at all diagnoses of patients with endometriosis, and they found that they also had an increased risk of developing ovarian cancer. Patients with ovarian cancer were looked at by Moduguno et al. and found that patients with self reported endometriosis had a higher incidence of ovarian cancer.

And the last study is a study from Ovarian Cancer Association Consortium, which is a very large epidemiologic group. They looked at all patients that had self reported endometriosis, and they found that patients with endometriosis had a high incidence of clear cell carcinoma and endometrioid adenocarcinoma of the ovary. So, here, these large studies, large case control studies, so a positive association with endometriosis and ovarian cancer development.

So these cohort studies are studies where they looked at people with the disease and then they looked back to see what in fact these patients had or if they developed malignancy in a prospective way. So, most of these are retrospective cohort studies, but these are very large studies. The first one, by Brinton et al. showed patients that were hospitalized with endometriosis, they looked at over 20,000 patients from the Swedish inpatient register and national Swedish cancer registry, and they matched to see who developed ovarian cancer, and they found that with a long history of endometriosis, the patients had a very high risk of developing ovarian cancer.

In the next study, again ... That was a very small study, and they did not find much endometriosis. This was the only study that I found that did not show a supported risk of ovarian cancer developing in patients that had self reported history of endometriosis. There's a retrospective study done in five very large infertility groups that showed all patients with infertility, and they had an increased risk of developing ovarian cancer if they had endometriosis. But all of those patients also had infertility.

The next study by Brinton et al. showed a retrospective cohort study that looked at patients that were diagnosed with endometriosis on an in and out patient hospital registry. They were associated with an increased risk of developing ovarian cancer, highest in the patients with a long term diagnosis of endometriosis. And, again, Swedish inpatient register and national Swedish cancer registry, high associated in long term endometriosis patients, and if a patient had a hysterectomy, but did report a history of endometriosis, they did not have an increased risk of developing ovarian cancer.

Another one out of Sweden, very large study, 63,000 women. Swedish inpatient registry and national Swedish cancer registry found 134 cases of ovarian cancer followed over 13.4 years. And those patients with endometriosis had an increased risk of developing ovarian cancer. In Kobyashi et al., that was a prospective and a retrospective study that looked at patients that had ovarian endometriomas, and those patients had an increased risk of developing ovarian cancer, especially if they were diagnosed with endometriomas over the age of 50. And the rest of the studies also were self reported history of surgical endometriosis. Those patients in Gemelli et al., they had asked the patients themselves in they had a surgical biopsy consistent with endometriosis, and they had an increased risk of developing ovarian cancer. Infertility patients in Australia and nulliparity and endometriosis were highly associated with developing ovarian cancer, and also infertility patients in the Netherlands, this was a Dutch pathology database and Netherlands cancer registry. It looked at patients with endometriosis, and they had a higher risk of developing ovarian cancer.

So, clearly, these studies show a higher risk in developing ovarian cancer with endometriosis. But, there are some strengths to systematic review. These are very large numbers of patients, so you're able to look back at a lot of patients. However, many of these have a retrospective design, and they have consistent findings that show that endometriosis is associated with a slight increase in ovarian cancer, but there's a lot of confounding factors in this. However, there is still a correlation with clear cell and endometrioid ovarian cancer in these patient population.

The limitations are that in many of these studies, it was self reported, so we didn't have a diagnosis or histological diagnosis of endometriosis. And, there is some lack of information on confounding factors. You know, these are large databases that are looked back at, and we may not be able to be comparing apples to apples in some of these circumstances, and self reporting of endometriosis is also not optimal. And in case control studies, they have a selection bias, detection bias, and a recall bias. May of these are questionnaire studies. And the cohort studies show ... few are very prospective, so many of them are just retrospective, so they may have inherent biases in just being a retrospective study. Hospital cases may have only been for severe cases. Many of these are patients who were brought to the hospital for endometriosis. So, that doesn't lend itself to a generalizable result in patients that might have mild endometriosis, and a selection and detection bias as well.

So, if you think about endometriosis and ovarian cancer together, we can talk about molecular basis, so in ovarian cancer, we went to treating the disease as a disease site disease, so if a patient has ovarian cancer, they get treated the same way no matter what, and how we're coming to, in 2017, at this point in time, we know that we can't do that. And in fact, that doesn't work. All tumors are not the same. They have inherent biological changes, molecular changes, an din fact, each tumor in a patient may also have different genetic makeup and different drivers, so now we're looking really at a pathway analysis to determine which patients would benefit from what treatment, and this has been slowly evolving, but now it's been pretty standard of care in ovarian cancer.

So, in endometriosis, it's kind of the same idea. Recently, in the past few years, there's been a lot of examination of molecular studies that have come to the forefront. We're evolving on our way to analyze tissues and this has definitely been shown in endometriosis. So, I'm not going to ... because we've had such a beautiful presentation by Dr. Bloon and others, I'm just going to touch briefly on this, but really, how endometriosis molecular changes also have some similarities with ovarian cancer.

So, EGFR and HER2 are the members of the of HER2 family whose signaling pathways that is crucial for tumor infiltration, progression, and metastasis. This is a crucial pathway in many tumors. As a matter of fact, this is a very big target for ovarian cancer treatment. And, EGFR and HER2 are more frequently expressed in clear cell carcinomas than in patients with the most common ovarian cancer, which is high grade serous. Gene amplification is also observed and related to endometriosis, and you can look at patients that have EGFR and clear cell carcinoma, 36% have the gene amplification, and in high grade serous, it's only 27%. And HER2 gene amplification and clear cell carcinoma as opposed to the more common type is high grade serous carcinoma.

So, 44% of clear cell and endometriosis is expressed in high EGFR versus 4% in endometriosis with high grade serous carcinoma. So, we think that EGFR amplification is higher in early stage disease and may be important in the earlier progression of clear cell carcinoma. We know that patients that develop clear cell carcinoma as we discussed before tend to develop more of an earlier stage disease, at time of diagnosis, so maybe this is ... these pathways are up regulated early on in the disease pathogenesis.

Another important study actually is the mTOR pathway for ovarian cancer. It happens to be a desirable target, especially for clear cell carcinoma of the ovary. The PI3Ks are lipid kinases that regulate signaling pathways and are vital of neoplasia proliferation, adhesion, and survival and motility. This gene is mutated in 33 to 40% of clear cell carcinomas, and it is associated with endometriosis. And these mutations infer an improved prognosis, and so the interesting thing is, if you have these mutations in your early stage disease, you tend to have a favorable prognosis. But, in general, clear cell carcinoma diagnosed in later staged disease, which is often not associated with endometriosis, has a very poor prognosis due to chemo resistance. So, really maybe the molecular pathways of endometriosis and ovarian cancer may not actually be that different, especially in the driver pathways of the diseases.

So, I think Stephanie said it before, I'm a surgeon, I have to talk about surgery. But, it's true that endometriosis surgery brings a whole another conundrum when relating to ovarian cancer. These patients tend to be young, and they have fertility sparing as priority. They are crunched for time because they want to have some ovarian preservation in terms of egg retrieval, so it becomes a time crunch and multiple anxieties and issues to deal with at a time when this patient has never been ill in her life, and now she's facing a serious problem.

And, it sets patients up for multiple surgeries. We could see with Dr. Tomer Singer, with the beautiful work that he does, and the extensive operations that he performs, and anyone who operates on patients with endometriosis. These are not one time operations. These patients have multiple surgical procedures because it's fertility sparing, so there tends to be recurrence and back to the operating room. And every time you go back to the operating room, there's more scar tissue and more risk of multiple organs involved. We also know that endometriosis patients have rectal involvement, as I said before, or pelvic tissues, and again, you're trying to preserve organ sparing surgery in these patients.

It's kind of a joke in GYN Oncology, because we can handle any cancer, but if you want to scare us, just give us an endometriosis case. These cases are very difficult. We're not really scared of them, but I'm just saying, just to make the point of the fact that they're extensively difficult surgeries. The tissues are very fibrotic. There's often no planes. These are multi-specialty surgical procedures. You know, Dr. Setchkin is a master at this, and oftentimes, needs other team members to assist, because of the multiple organs involved, which is the same as ovarian cancer. We tend to have to operate on GI, GU, as well as GYN organs.

So, this is a picture of an endometrioma that is malignant, and this patient had pretty extensive adhesions in the pelvis, so this was kind of after we were able to get the tumor out. It's extensively adherent to the bowel and the bladder, and on imaging, this could like an endometrioma, it could look like a cancer, and these women are young, and many times, they are not diagnosed with a malignancy before they go to the operating room, because most endometriomas are not malignant, but this one is and this is a devastating and difficult operation for the patient and a difficult diagnosis.

But, the difference between this tumor and early endometriosis associated ovarian cancer is very different from advanced ovarian cancer, when patients present with large volume [inaudible 00:21:33], carcinomatosis, omental involvement, and bilateral ovarian involvement, and it's just a different operation, so if you think about it, even though the histology may be clear cell, the pathology may be different. Why would one spread so much earlier, or just be compliant to the pelvic organs.

Well, maybe it's a distinct clinical entity. Maybe these two histologic clear cell carcinomas, one that arises in endometriosis and one that is de novo clear cell carcinoma and is more advanced, actually have different driver pathways. Maybe one uses the mTOR pathway and the ... EGFR pathway more strongly than the other. So, it's just interesting. Instead of treating all clear cells the same, we need to start thinking of them being different, especially in this clinical scenario.

Patients tend to be younger and nulliparous. They're diagnosed in earlier stages, they have very low incidence of lymph node metastasies, which mostly early stage, high grade serous tumors have about a 16 to 30% of having positive lymph node metastasies. And these patients have generally none or very low, not none, but low. Some say they have a better survival and some have a worse survival. That's kind of based on stage. Mostly clear cell and endometrioid types, and they're usually pure histologies. You know, most ovarian cancers tend to be mixed. When you have a clear cell, serous, and sometimes you have an endometrioid. With endometriosis, it's typically one histology, which is also quite interesting.

So, why are these patients getting diagnosed early? Well, I said once before, maybe it's the difference in pathway driving, but really the other thing is patients that are younger are screened more. They have pelvic pain, they already have endometriosis, they're being screened more by the gynecologist, and they are associated with more symptoms. We know endometriosis, they have a tremendous amount of adhesions, pain, that may lead to a diagnosis that is sooner as opposed to ovarian cancer, that de novo and high grade serous tumors, that's a silent disease, or it's noted with very non specific symptomotology. It's possibly because patients with a diagnosis of endometriosis are screened, as I said, and earlier ovarian cancer stages at younger ages suggest that endometriosis is associated with ovarian cancer may be completely a different disease, than advanced disease.

So, what are the current challenges? Well, despite these findings, we haven't found any reason to change clinical practice, or it's not really justified at this time, but I always have an issue with the say we've practiced medicine in the past, because we've practiced medicine in disease site silos, so we study ovarian cancer on our own, we would never talk to anybody that had endometriosis research, and I think that's changing, and I think that's a good thing, which is one of the great reasons why this conference is put on today, because we probably need to think of things and thoughts that go across disease sites to make progress.

The problem is now, that we're all doing these small studies, so people are doing small studies on their own rather than larger trials where we may have better numbers and larger sample size to make better conclusions. In GYN Oncology, our cooperative groups study the rare tumors, which clear cell carcinoma is one, and we struggle to get numbers, so studies take 20 years to complete, and maybe if we really looked at endometriosis and we partnered with foundations such as this, that we could identify patients that we can watch long term. Other thing is that this may be a good opportunity to understand molecular mechanisms.

So, I really think that there should be further collaboration between ovarian cancer groups and groups such as this. For surgical diagnosis confirmation, I think that's crucial. Especially if you're going to be looking at molecular staging or molecular prognosis. And maybe we can have a predictive molecular evaluation of the endometriosis cases to know really who is at risk for developing these malignancies in a more personalized future management, because I really think ultimately, that's where medicine is going, and it's not so far-fetched.

So, I just want to say think you for your time. Thank you for the opportunity to speak today, and have a good rest of the day.