Classification and transitions in chorionic tumors

Media

Part of Acta Medica Philippina

Title: Classification and transitions in chorionic tumors
Creator: Canlas, Benjamin D.
Language: English
Source: Volume XV (4) April-June 1959
Year: 1958
Subject: Choriocarcinoma; Medicine -- Periodicals
Rights
Fulltext: CLASSIFICATION AND TRANSITIONS IN CHORIONIC TUMORS BENJAMIN ». CANLAS, JR, MJ). Department of Pathology, College of Medicine University of the Philippines Attempts at classification of tumors arising from the pla centa started with Marchand who first recognized the chorionic rather than the maternal origin of the tumor we now know as choriocarcinoma. He divided chorionic tumors into: (a) typ ical — which consisted of the actively growing and metastasiz ing tumors and (2) atypical — which consisted of the less active tumors composed mainly of giant cells (3,11)- Ewing introduced the term chorioma to designate a tumor arising from the cells of the chorion (4). The classification by Ewing in 1910 into choriocarcinoma, chorioadenoma destruens, and syn cytioma has gained the greatest popularity and has persisted lo the present with slight modification. Other terms have been used to signify the same entities: chorion-epithelioma which Marchand originally gave for choriocarcinoma; invasive mole, destructive mole, penetrating mole, malignant mole and metas tasizing mole for chorioadenoma destruens; and syncytia! en dometritis for syncytioma. Hydatidiform mole, because of its obvious close relationship to the other choriomas, has since been added to this group of tumors. In 1947 Hertig and Sheldon subdivided the ordinary hydatidiform mole into six groups for purposes of prognostication which later was reduced to 3 grades (6, 7). This grading depended on the degree of hyperplasia, anaplasia, and neoplasia of the tumor. More re cently, Sutomo in 1956 proposed the term villous choriocarci noma to designate the invasive variants of mole and non-villous choriocarcinoma to those showing no villous patterns, including the syncytioma and the typical choriocarcinoma (18). These are aside from the ordinary hydatidiform mole. Obviously, these classifications were intended to correlate the clinical ma nifestations with the pathologic findings in order to reduce the confusion that these tumors have created in the clinician as 223 224 ACTA HEDICA PHILIPPINa well as the pathologist The confusion is the result of many variable and inexplicable outcomes encountered very frequently in these tumors. Our purpose is mainly to give emphasis to certain observa tions on the progression or regression of this group of tumors that may account for their apparent clinical unpredictability. Only by knowing these possible pathways of transition can one fully understand why these tumors behave in such an enigmatic manner. The key to the enigma lies in the different pathways that trophoblasts may follow, depending on factors not yet fully understood and probably residing not only in the cells them selves but also in the environment of the host tissues. It is definitely established that choriomas, as implied in the name given by Ewing, arise from the chorion of the product of conception. In this sense choriomas must all be derived from the developing fertilized ovum. The tumor cells in these neoplasms arise from one of the earliest cells that differentiate in the developing embryo: the trophoblasts. It is theoretically possible that the fertiltized ovum will, from the very start, give rise only to primitive malignant trophoblasts without producing the embryonic cells of the inner cell mass from which the future embryo finally develops. Because of the lapse of time that precedes choriomas and the relatively short period within which trophoblasts and the inner cel) mass differentiate in the early 4-day-old fertilized ovum (9), it is more likely that almost always they arise from already well-formed trophoblasts that have become malignant. Comparatively speaking, the histological appearance of choriocarcinoma mimics very closely the pattern of the tropho blasts in the early previllous developing ovum (7, 9). It is also of interest that the destructiveness of the trophoblasts at this early age of pregnancy simulates very closely the destruc tiveness of choriocarcinoma (19). In short, choriocarcinoma can be classed as a tumor whose pattern assumes the most immature form of differentiation considering that it redupli cates cells the maturity of which is only a few days old (12-15 days). It can be said that the morphologic aspects of this tumor repeats the earliest stages of embryogenesis. From this point of view, it is also theoretically possible that occasionally CHORIONIC TUMORS 225 rare villous formation may occur depending upon the degree of differentiation or undifferentiation the malignant tropho blasts may follow. These villi are, of course, aside from those that may remain from the original pregnancy, abortion, or hy datidiform mole. While in most instances, therefore, villi may be regarded as tending to support a diagnosis other than chorio carcinoma, in rare cases where there is overwhelming tropho blastic activity and anaplasia, one or two villi need not neces sarily rule out the possibility of choriocarcinoma. It is generally agreed that choriocarcinoma follows a mole, an abortion or a term pregnancy all of which represent an end product of conception. An exception to this general rule is, however, pertinent The germ cell of either male or female or any pluripotentia) cell that has the potentiality of giving rise to teratogenous growths may also be capable of trophoblastic differentiation. In this way, therefore, teratogenous growths may give rise to choriocarcinoma not necessarily resulting di rectly from a product of conception. This genesis explains the rare cases of choriocarcinoma found in male or pre-pubertal females usually of gonadal origin, as well as still rarer chorio carcinomas of extrauterine, extragonadal, and usually of mid line origin. Other than these exceptions, therefore, all chorio carcinomas must arise from trophoblastic cells which ultimately must be regarded as sequential to conception. It is to be assumed that choriocarcinoma following a term pregnancy or an abortion arises from trophoblasts that have somehow gained a nidus in the uterus and have failed to be ex pelled with the rest of the placental elements. Initial manifes tations in such instances are primarily in the uterus although occasionally the primary focus in the uterus may remain small and undetected clinically while the metastases manifest them selves more prominently in whatever organ they may be located. In such instances, the genital manifestations are absent or nil so that the clinician is often confused or even misled in his diagnosis. Not infrequently one encounters benign trophoblastic de portation to the lungs in pregnant woman (3, 10, 13, 20). This should be regarded as a consequence of the unusual invasive ca pacity of trophoblasts including invasion of the uterine sinuses, 226 ACTA MED1CA PHILIPPINA even in normal pregnancies. From these sinuses the tropho blasts, no doubt, can easily be transported to the lungs where they are filtered. These must, however, be regarded as mere mechanical benign deportations and not aa true metastases. The chances of jnolar trophoblasts to be transported to the lungs are similar, if not greater, because of a neoplastic aspect present in the hydatidiform moles, If perchance these trans ported trophoblasts cells remain viable and are somehow ac tivated into malignant growth, then they can give rise to a choriocarcinoma manifesting itself initially in the lungs. Thia possibility may also explain certain choriocarcinomas without any apparent primary lesions in the uterus (13, 19). The fact that the majority of choriocarcinomas foltow hy datidiform moles suggests the presence of a factor or factors in moles, not sufficiently present in pregnancies terminating either as term deliveries or as abortions that predispose to the development of this highly malignant and fatal tumor. Hertig and Edmonds in their classic study of the genesis of hydatidi form mole concluded that moles usually resulted from either a failure or a defective development of the embryo resulting in uondevelopment of the fetal circulation in the presence of an intact maternal circulation and functioning trophoblasts (8). This apparently explains the degenerative aspects of moles but does not satisfactorily explain the proliferative aspects of moles bordering neoplasm or actually neoplastic in character. Opi nions as to the possible explanation of this phenomenon may be postulated, but the final explanation will probably await the perennial problem of what is the ultimate cause of neoplasm. It is this neoplastic tendency, prominently noted in moles and not found in the other types of pregnancy, which undoubtedly predisposes to the development of choriocarcinoma. The malig nant potentiality of the trophoblasts in moles obviously is great er than that seen in either abortions or in term deliveries which ordinarily manifests proliferative activity within bounds of what is regarded as normal in contrast to the varying degrees of proliferation, anaplasia, and neoplasia observed in moles. The neoplastic nature of moles contributing to their malignant potentiality is further supported by the general correlation be tween morphologic variation and clinical malignancy aa attested by the studies of various authors (6, 7,10). CHORIONIC TUMORS 227 The term “invasive” mole should be used to designate a mole that has shown invasion of the structures beyond the or dinary confines of trophoblastic invasion in a normal preg nancy. Invasion of any structure by direct contiguity beyond these limits in the uterus serves to differentiate an “invasive” mole from a “metastasizing” mole which should show molar elements transported to other organs or structures not show ing any direct contiguity with the original uterine molar mass. These two types of mole can be rightfully classed under the more encompassing term “malignant” hydatidiform mole. It is admitted that the term "malignant” in this instance is a “mo dified” form of malignancy because the true malignant nature of the “metastasizing" and invasive moles as implied in the terms used to qualify them, is not in the true sense of malig nancy as their clinical outcomes are not always predictable in spite of their "metastasizing” or invasive tendencies. The term "malignant” mole, therefore, is used more to distinguish these variants of mole from the ordinary noninvasive and nonmetas tasizing mole rather than imply a dark prognostic significance as is usual for truly malignant tumors. To the moles that show neither invasion nor "metastasis” the term “benign” mole is assigned in contradistinction to the “malignant” ones as already described. Because of their different behavior simulating ma lignancy, “malignant" moles have to be designated under a different category that will make them distinguishable from their more benign counterparts. In view of these distinguish ing characteristics, therefore, we have divided hydatidiform moles into 2 big categories: the ordinary “benign" mole which is limited to the endometrial cavity and the "malignant” mole which, in turn, may be subdivided into the "invasive" and “me tastasizing” varieties. The term "benign” as used in this sense, however, does not necessarily eliminate the potentiality of these types of hydatidiform moles' to precede choriocarcinoma as is generally agreed. These moles can vary morphologically from the very innocuous to the very hyperplastic and anaplastic pat terns. Distinguishing these types of mole from the evacuated molar material and curettings would naturally also follow the general correlation between morphologic variation and clinical malignancy as stated earlier. 228 ACTA MED1CA PHILIPPINA The invasive property of trophoblastic tissue is apparently accentuated in the invasive mole as defined above. Invasion is a property ordinarily ascribed to malignant tumors, but be cause the prognosis in these instances is not necessarily dark due to the usually localized nature of the invasion, it is doubt ful if classification under the truly malignant tumors is war ranted. On the other hand, there is no doubt that if the inva sion is extensive it can lead to perforation of the uterus, wide spread infection and hemoperitoneum which may be fatal in outcome. The invasive potentiality of a mole, like its potential ity to precede choriocarcinoma, can be correlated with its mor phologic variation. On the premise that any trophoblasts when activated can give rise to choriocarcinoma, it is not presump tuous to assume that invasive moles can also precede chorio carcinoma if and when the factor or factors responsible for malignant transformation are present. The metastasizing mole apparently exhibits a feature of greater malignancy than the invasive mole. In this variant of mote the metastatic lesions may be found in any part of the body but more prominently in the lungs and vagina (1, 15, 17, 18). There are 2 possible ways by which villi may get into the lungs or to any other distant organ not in contiguity with the original mole in the uterus. The villi may be carried by the blood stream to these organs or they may develop within these organs by a process of differentiation of trophoblasts that have been carried to these sites. In the latter instance the villi are formed "in situ” in the organa where the differentiating tro phoblasts have lodged. In invasive moles, the possibility that trophoblasts with their accompanying villi may invade sinuses and from there be transported to the tungs is great. The trans ported structures, if capable of autonomous growth, may per sist and thus can now be called a metastasizing mole, A me tastasizing mole may, therefore, coexist with an invasive mole. If on the other hand, these structures are not capable of auto nomous growth, then they undergo degeneration (see later dis cussion of syncytioma) and finally disappear. The extent, size, aa well as the location of the dissemination will obviously de termine the prognosis in these instances. In a case reported by Delfs (2), the metastases are in relatively unimportant struc tures and this apparently led to an extended course. If on the CHORIONIC TUMORS 229 other hand, these are located in strategic positions as the brain, they can no doubt be rapidly fata). One would expect also that multiple lesions in the lungs though benign morphologically may give rise to massive bleeding. This type of mole, there fore, though morphologically benign, is clinically malignant. It may be likened to the well-differentiated carcinoma of the thyroid which morphologically is indistinguishable from the normal such that it waa previously designated as a “lateral aberrant thyroid." The good morphologic differentiation may be indicative of its slow growth as observed clinically. The over-all evaluation of the metastasizing mole appears to indicate predominantly malignant features rather than benign ones but a categorical statement to this effect has to await further stud ies on this unusual tumor, A malignant transformation into frank choriocarcinoma is not beyond reasonable bounds in this instance where cells composing the tumor are even more cap able of autonomous growth. It should be emphasized, however, that the transported trophoblasts do not always remain viable in their secondary sites of deposition and therefore the criteria for metastasis are not always satisfied <1). True metastasis calls for transportation of tumor cells to a distant site, followed by growth of these cells in this secondary site. Syncytioma or syncytial endometritis has been of even more controversial nature. The true neoplastic nature of this tumor has been questioned (3, 7, 13) so that the designation syncytial endometritis has been applied especially when the inflammatory response is very prominent. Ewing believed, to which the author concurs, that the trophoblasts found in this lesion are actually regressing rather than progressing (3, 4). The lesion appears to be an accentuation of the invasive prop erty of the placental site giant cells which normally may be found as deep as the myometrium especially at the site of im plantation of the original placenta (7), Because of its close relationship to trophoblastic .tumors, however, in spite of its doubtful neoplastic nature, this lesion should rightfully belong to this class of tumors. The placenta) giant cells have been regarded as individual trophoblasts that have broken off from the original tropho blastic shell and have wandered into the decidual or myometrial layers (3, 4, 7, 13, 14). Ordinarily this is of no consequence 230 ACTA MEDICA PHILIPPINA but when several of these cells are left in the placental site, they apparently prevent the normal healing of the denuded en dometrial cavity and induce considerable hemorrhage. These effects may be attributed to the retention of vasodilating and anti-coagulating substances, enzymes, and other factors usually found in trophoblasts (9, 21). Recently histochemical studies seem to indicate that these cells are of cytotrophoblastic origin (12), The inflammatory cells incident to the presence of these cells, which normally should have undergone spontaneous re gression, appear to be a reaction to the lesion and vary con siderably in quantity. The symptoms and even the pathologic changes that may be produced by this lesion may closely si mulate choriocarcinoma. Since these cells are derived from trophoblasts, it is not improper to deduce that they may create the same lesion wher ever trophoblasts may be transported. While, heretofore, syn cytioma has been described only in the uterus, there are indi cations that its location is not limited to this organ (1.5). Wherever trophoblasts are transported, a syncytioma may also develop (1). This type of lesion is, however, more frequently seen when the transported trophoblasts have a greater malig nant potentiality as in the higher grades of mole described by Hertig and Mansell than when they come from an ordinary pregnancy. The cells in this category produce much less in jury to the host tissue than metastasizing mole and certainly are not malignant like choriocarcinoma. When found in the uterus, this lesion may regress spontaneously or may subside after a curettage, not necessitating a more radical procedure as a hysterectomy. In the same token, lesions produced else where in the body are similarly more innocuous and may dis appear spontaneously. The symptoms manifested and the out come produced wil) no doubt depend on the location, size, and number of lesions present. Since these are viable trophoblasts, although possibly undergoing regressive changes, the theoretical possibility that they may still possess malignant potentialities cannot be completely ruled out. This possibility has, up to the present, not been substantiated. For want of a more appro priate name, "extra-uterine" syncytioma is proposed for this type of extraruterine lesion whose genesis ib similar to that of CHORIONIC TUMORS 231 its uterine counterpart. Obviously, the name syncytia) endo metritis cannot be applied to a lesion outside the endometrial confines. Frequently, densities in Use lungs arc seen radiological!)' after an ordinary benign or an invasive mole (16). These le sions that regress spontaneously are probably extra-uterine syncytiomas. It is, however, conceded that even well developed choriocarcinoma, for unknown reasons, may undergo sponta neous regression. It has been noted that such regressing choriocarcinomatous foci may produce changes morphologically similar to those of extra-uterine syncytioma (1). In view of the foregoing discussion, the following diagram to illustrate the development and transitions in choriomas is presented: Invasive mole and metastasizing mole are blocked to indicate that both may give rise to either syncytioma or choriocarcinoma. FERTILIZED OVUM PLURIPOTENTIAL CELL The following modified classification of chorionic tumors is also proposed: CHORIONIC TUMORS A. Hydatidiform mole 1. Benign hydatidiform mole 2. Malignant hydatidiform mole 232 ACTA MEDIC A PHILIPPINA a. Invasive mole (chorioadenoma destruens) b. Metastasizing mole B. Choriocarcinoma C. Syncytioma J. Uterine syncytioma (syncytial endometritis) 2. Extra-uterine syncytioma Since tumors need not necessarily include only neoplastic conditions, there can be no objection to this classification even when syncytioma and hydatidiform mole may be claimed to be not truly neoplastic in nature. In both conditions tumoral le sions are produced. SUMMARY The enigmatic manner with which chorionic tumors behave has been conceded by most clinicians and pathologists. This uncertainty in the behavior of these tumors has created some confusion in the original classification by Ewing into chorio carcinoma, chorioadenoma destruens, and syncytioma. Without deviating considerably' from this original and classic classifica tion, a better understanding of these tumors can be obtained by a study of the possible pathways of transition that tropho blasts may follow depending on factors not yet fully understood. The genesis of choriocarcinoma from trophoblasts and plu ripotential cells of the body explains choriocarcinoma after a mole, abortion, or a term pregnancy and teratogenous chorio carcinoma. The very great similarity between this tumor and early pregnancy is stressed. The possibility of potentiality malignant trophoblasts to be deported via the blood stream to other organs and there developing extra-uterine choriocarci noma is explained. The fact that majority of choriocarcinomas follow hydatidiform mole suggests that there is a neoplastic aspect of mole not found in term pregnancy or abortions. Three varieties of mole are postulated. These are the ordinary "benign" mole which is limited to the uterine cavity, the "invasive” mole which invades by contiguity structures ad jacent to the uterine cavity (chorioadenoma destruens), and the CHORIONIC TUMORS 233 “metastasizing” mole which, shows distant “metastasis" of vil lous structures. The latter two may be considered as “malig nant” moles. Again it is conceded that great difficulty is en countered in differentiating these three from each other on the basis of histological examination of the evacuated molar ma terial. Most often diagnosis can be made only after the entire uterus is removed or a biopsy of the lesion outside the uterus is made. In none of these instances, however, is the prognosis usually dark as in choriocarcinoma, although it appears to be a little worse in the order that they are mentioned above. Any one of these three may also precede choriocarcinoma. The mo dified meanings of "metastasis” and "malignant” as used in this terminology is explained. Syncytioma is definitely tumorous in that it is a space occupying lesion but not definitely neoplastic. In fact, it is probably regressive in nature. The lesion appears to be an accentuation of the invasive property of placental site giant cells which are derived from the trophoblasts. More and more cases have been encountered by the author proving that these cells may be seen not only in the myometrium but also in any site where trophoblasts may be transported. Therefore, aside from the uterine syncytioma (syncytial endometritis}, the ca tegory of extra-uterine syncytioma is proposed. The usual spontaneous regression of these latter lesions may explain den sities in the lungs, heretofore considered as choriocarcinomatous, that spontaneously regress. While seemingly benign histolo gically, they may be fatal depending on their number, size, and location. A diagram illustrating the origin and transition of chorio nic tumors is presented. A new, modified classification of these tumors is proposed. REFERENCES 1- CANLAS, B. D., Jr,: Unpublished data. 2. DELFS, E,: Quantitative Chorionic Gonadotropin— Prognostic Value in Hydatidiform Mole and Chorionepithelioma, Obat. & Gynec., 9:1-24, 1957. 3. EWING, J.: Neoplastic Diseases— A Treatise on Turnon, 3rd Ed.r 1928, W- B. Saunders Co., Philo. r pp, 608’623. 234 ACTA MEDICA PHIL1PPINA 4, EWING, J.: Chorioma — A Clinical and Pathological Study, SwObst. & Gynec,, 10:336-892, 1910. 5. FLEISCHMAN: Quoted by Ewing (3), p. 617, 6. HERTIG, A. T„ and SHELDON, W. H.: Hydatidiform Mole—A Clinico-Pathological Correlation of 200 Cases. Am. 3. Obst & Gynec., 58:1-36. 19477. HERTIG, A. T., and MANSELL, H.: Atlas of Tiimor Pathology, Section IX, Fascicle 33, Tamora of the Female Sex Organa, Part I: Hydatidiform Mole and Choriocarcinoma, Armed Forces Insti tute of Pathology, Washington 25, 1956, 8, HERTIG, A. T„ and EDMONDS, H. W.: Genesis of Hydatidiform Mole, Arch. Path., 30:260-291, 1940. 9. HERTIG. A. T., ROCK, J., and ADAMS. E. C.: A Description of 34 Human Ova within the First 17 Daye nf Development, Am. J. Anat. 98:435-494, 1956. 10. HUNT, W.. DOCKERTY, M. B.> and RANDALL, L. M.: Hydatidi form Mole — A Clinico-Pathologic Study Involving Grading as u Measure of Possible Malignant Change. Obst. & Gynec., 1:593-609, 1953. 11. MARCHAND, F.: Quoted by Ewing (3), p. 608. 12 McKAY. D. G„ HERTIG, A. T., ADAMS. E. C., and RICHARDSON, M. V.: Histo-chemieal Observations on Human Placenta, Obst. & Gynec.. 12:1-36, 1958. 13. NOVAK. E,: Pathological Aspects of Hydatidiform Mole and Chorio carcinoma, Am, J. Obst, & Gynec., 59:1355-1370, 1950. 14. NOVAK, E. and SEAH, C. S.: Benign Trophoblastic Lesions in the Mathieu Chorionepithelioma Registry, Am. J. Obst. & Gynec., 68:976-390, 1954. 15. NOVO. A.: Transition from Mole to the Chorionepithelioma; Various Aspects (Abstract), S.G.O. w/Intem. Abstr. Surg., 102:63-64, 1966. 16. SAVAGE, M. B«: Trophoblastic Lesions of the Lungs Following Benign Hydatid Mole, J, Obst. & Gynec., 62:346-352, 1961. 17. STA. CRUZ. J. Z,: Personal Communication. 18. SUTOMO, T.: Choriocarcinoma in Indonesia, Scientific Exhibit, Firet Asiatic Congress of Obstetrics and Gynecology, Tokyo, Japan, April 4-6, 1957. 19. WILLIAMS. T. J.: Hydatidiform Mole and Associated Tumors of the Chorion, Am. J, Obst. & Gynec., 46:432-444, 1943. 20. WILLIS, R. A.: Pathology of Tumors, 2nd Ed., 1953, C. V. Mosby Co., St Louis, Mo., p. 986. 21, WISLOCK1, G. B., DEMPSEY, E. W., and FAWCETT, D. W.: Some Functional Activities of the Placental Trophoblast, Obst. Gynec. Burg., 8:604-614, 1948,
pages: 223-234