Pathology and development of ovarian cancer

Share :
Published: 9 Oct 2012
Views: 7193
Dr Jamie Prat – Hospital De La Santa Creu I San Pau, Spain

Dr Jamie Prat talks to ecancer at ESMO 2012 about how ovarian cancer develops, and breaks down each of the cancer sub-types.


Dr Prat notes that 40 years ago it was thought that ovarian cancer was a singular disease, but now at least 5 different types have been identified. 


In the clinic, subtypes are identified morphologically and through observation of biomarkers, genetic alterations and clinical collaborations to therapies. 


Filming supported by Amgen

ESMO 2012, Vienna, Austria


Pathology and development of ovarian cancer


Dr Jamie Prat – Hospital De La Santa Creu I San Pau, Spain



Hello Professor Prat, we’re here at ESMO 2012 and you presented some very interesting data on pathology in ovarian cancer. Can you summarise some of your presentation?


I wouldn’t say data, I just presented our views today about how ovarian cancer develops. About forty years ago or thirty years ago it was thought, and still today and still ten years ago, that ovarian cancer was just one disease. Today, because of many developments at the molecular and pathologic and clinical level, we see today a more heterogeneous disease or, better, at least five different diseases. These five different diseases, the best way to identify them is by the microscope, morphologically, by taking a piece of the tumour and looking at it microscopically and then you start with the histologic type, that’s the most important thing. From there you go on and you are able to identify precursors which are different for these five types, genetic alterations at the DNA level and also, of course, clinical correlation with different responses to therapy and different behaviour. So I mean that these not one but five different ovarian cancer types should be treated differently because they really are five different diseases. This is the new thing today.


I’m surprised it’s just five because I look at breast and other cancers that it’s already into the twenties or thirties, different subtypes.


Let me tell you something, ovarian cancer used to be classified according to cell type in five, seven or eight different types – serous, mucinous, endometrioid, clear cell, transitional, squamous, mix and undifferentiated. But today, for instance, undifferentiated and transitional should be lumped with high-grade serous because they really are the same disease and have the very same genetic alterations. After this reduction, you end up with at least five. I’m not saying five exactly, maybe five or six but no more than that. But the striking thing is that one of them, high-grade serous, represents 70% of the total, whereas the other four, endometrioid, clear cell, low-grade serous and mucinous are totally different and represent only 33%. Some investigators, because of this difference in prevalence, lump together these four types and call them type 1 ovarian cancer whereas the other, the high-grade serous, they call type 2. I think this is artificial and doesn’t make any sense because neither the high-grade serous has anything to do with these other four types, nor any of the four types have anything to do with the other three or with high-grade serous so they’re totally different.


So what does all this mean for changes in treatment patterns?


It means a lot because if you consider each of these five types as different diseases, first, the way they are identified when they are diagnosed, they are found at different stages. So staging, figure staging, perhaps should be different; the system that you use for staging these five different diseases because, for instance, high-grade serous is usually found at advanced stage in the omentum, outside the pelvis, whereas clear cell and endometrioid carcinomas, not always but quite frequently, are found confined to the ovary. Why? Because high-grade serous in some cases may not originate in the ovary, may originate in the fallopian tube, in the peritoneum outside of the ovary, in the pelvis and no wonder you have it in the ovary and also in the omentum because it’s metastatic to both, to the ovary and the omentum. Whereas, the clear cell or endometrioid frequently, I wouldn’t say always but frequently, arise in the ovary from endometriosis which results from reflux from the endometrium to the ovary but they are in the ovary from a “benign” quote-unquote lesion; endometriosis by carcinogenesis you have clear cell or endometrial carcinoma presenting as an ovarian primary tumour. So the situation is totally different so there are differences in response to chemotherapy, of course, between the high-grade serous, which is a tumour that because of the BRCA mp53 alterations usually responds, I wouldn’t say always, but most of them do respond to Taxol platinum whereas the clear cell does not.  In those cases in which the clear cell has spread beyond the ovary the results of chemotherapy are very poor.


Ovarian cancer is quite often called the silent killer, as you said, because quite often when it is diagnosed it’s already metastatic. One of the things would be to have biomarkers for early detection, is there anything potentially coming out in this line?


Well, I can say that early detection, the whole thing started with the BRCA and the breast-ovary cancer syndrome with BRCA1 and BRCA2. In 2001 it was found that in the fallopian tube there were lesions; we were usually looking for these precursor lesions in the ovary and couldn’t find any but then some investigators from the Netherlands found, I wouldn’t say precursor, but they found cancer lesions confined to the epithelium in the fallopian tube. Then investigators in the United States have found… I’m always referring to the BRCA patients, talking about BRCA patients which represent about 10% of all ovarian cancers, no more than that. 90% of ovarian cancer is BRCA negative and sporadic. With BRCA positive breast-ovary cancer syndrome, they have in prospective studies, because of their family history of ovarian cancer, they underwent risk-reducing salpingo-oophorectomy. In those fallopian tubes precursor lesions, true precursor lesions, have been found, they are called p53 signature and serous intraepithelial lesion, not carcinoma. These are lesions that can be alert, not the P53 signature, which is found in patients who not necessarily develop ovarian cancer or pelvic peritoneal cancer, but lesions of the fallopian tube epithelium, atypical lesions of the fallopian tube epithelium, in these patients may alert the physician, together with the clinical history or family history, that these patients may develop ovarian cancer and this is the reason why they are followed and, of course, the initial reason why they underwent risk-reducing salpingo-oophorectomy.


Any potential new chemotherapeutic agents for ovarian cancer?


I’m not familiar with that because I’m a pathologist and we don’t prescribe chemotherapy but I think that there are a lot of things cooking, for instance p53 and mTOR alterations are treated with targeted therapy like PARP or erlotinib and other new and others. So many things are being assayed in these patients with, I would say, not very earthquaking results but some results that may be, in the whole context, the entire context of silent killer, may be taken as some good news.


Thank you.


You’re welcome.