The History of the Fere Ex Nihilo Theory

In 2010 I wrote a paper about the origin of ovarian cancer.

This is how I arrived to this proposal.

After a small incursion in Internal Medicine, I started in Pathology, and I immediately realized that it is the best specialty. I learned from an exceptional teacher in Argentina, Dr. Oscar Croxatto, that Pathology is not only rendering diagnosis, but that we also have the opportunity to understand how some diseases develop. This view of Pathology and ending up at MD Anderson Cancer Center, have been the best parts of my academic life. Many clinicians do research work with animals or cell lines; they are trying to reproduce what happens in the human body. We Pathologists do not need to reproduce what happens in the body. We can observe this everyday under the microscope. The tissue talks to us; we just need to listen.

I have been always interested in ovarian cancer, trying to understand how it develops. I always thought that in order to understand how high-grade serous carcinoma develops we would need to understand the very early lesions, not the last step, when high-grade serous carcinoma is fully developed. There are numerous issues in common among all serous neoplasms, including high-grade lesions, low-grade tumors, borderline, and benign neoplasms. We have noticed and reported that all serous borderline and benign neoplasms are associated with multiple foci of endosalpingiosis. Therefore, if we understand how endosalpingiosis develops, we might be able to figure out how serous neoplasms develop. In trying to explain how multiple benign glands develop, we cannot invoke a metastatic process, which would be the first explanation for different glands in a malignant process.

In the ovary, many pathologists confuse endosalpingiosis with inclusion cysts. It is true that the epithelium of inclusion cysts might develop tubal metaplasia. But in areas of endosalpingiosis, there are frequently several glands in close proximity, some have dense connective tissue around them, similar to the fibromatous tissue of adenofibromas, there are calcifications in the stroma, and the glands are never connected to the ovarian mesothelial surface. The most common site of endosalpingiosis is the ovary. But it can also be seen in lymph nodes and in different areas of the peritoneum. Therefore any explanation for the development of endosalpingiosis should also be applicable to other areas where invaginations of the surface epithelium are not a recognized feature, like in lymph nodes, or in endosalpingiosis in male patients. In order to understand a lesion we need to see many cases, and endosalpingiosis is a lesion of limited frequency. I believe it is easier to understand endosalpingiosis by reviewing many cases of a lesion related to endosalpingiosis, endometriosis, which is another component of Mullerianosis, and far more common. Some cases of endometriosis could be metastatic from the endometrium, but, again, explanations for the development of endometriois should also be applicable to cases of endometriosis in men, or in women that never menstruated. In these cases, a metaplastic theory has been proposed. Metaplasia usually applies to changes in an epithelium, for example from endometrioid to serous, or in the mesenchyme, for example bone in a fibromatous stroma. However, endometrioid glands in endometriosis are never connected to an epithelial area, and they are in the stroma. This is a classical example of mesenchyme-epithelial transition. We should not be skeptical about these transformations of the mesenchyme into epithelium and vice versa because this is a very active phenomenon in the beginning of our lives. At the beginning of the development of all embryos only epithelial cells are found, first in the trophoectoderm. Later on, the mesenchyme develops from some areas of the epithelial cells, and some cells in the mesenchyme transform again in cells with epithelial features, which, because they come from the mesenchyme, are designated as the mesothelium. I believe that this mesenchymal-epithelial transition explains the development in the stroma of areas of endometriosis and is also a good explanation for the development of endosalpingiosis. Hyperplasia of these areas with epithelial proliferation would be a good explanation for how serous borderline tumors start. And in a similar way we can understand how serous carcinoma develop. This explanation of glands coming from nowhere in the stroma is what I consider “fere ex nihilo”.

A very common example of this mesenchymal-epithelial transition is the presence of glands in adenofibromas. The small glands in adenofibromas are not connected to any epithelial area. They appear and disappear in deeper sections, and the only logical explanation is that they arise in the stroma.

Fere ex nihilo means epithelial cells appear almost out of nothing, anywhere in the stroma. Mesenchyma epithelial transition, and epithelial metaplasia in the stroma, are different terms that refer to the same process explaining how endosalpingiosis and, in some cases endometriosis, start in the stroma. The same process would explain how glands develop in serous borderline tumors, and in serous carcinomas.

I started working on this idea 25 years ago when I was able to induce epithelial tumors in guinea pigs, an animal that rarely develops ovarian neoplasms, using only steroid hormones. The tumor glands in the stroma of the ovary were not connected to the surface epithelium of the ovary. Every time I saw an epithelial ovarian neoplasm I tested the theory, and it grew stronger with every case.

Fifteen years ago I tried to publish the fere ex nihilo theory, but unfortunately this was at the same time when the fallopian tube theory was proposed. I had received numerous rejections from pathology and gynecology journals. Because of my insistence, in 2012 the editorial board of the International Journal of Gynecological Cancer decided to accept just a letter to the editor. In 2016, Advances in Anatomic Pathology published a review article. This year, 10 years after the first publication, Human Pathology will publish a more complete study where we confirm the fere ex nihilo theory and also show that it applies to all serous tumors, benign, borderline, and malignant.

The only way to solve a problem is to understand how it develops. Then we might even be able to prevent it.