Ovarian teratomas have been divided into two groups, regarding their mature or immature characteristics.1 The first kind is the mature teratoma, also called mature cystic teratoma or dermoid cyst, which is the most common form and represents 99% of ovarian teratomas. In its pure form, mature cystic teratoma is always benign, and the treatment is surgical excision. Secondly, there is the immature teratoma, which is an uncommon tumor, making up less than 1% of ovarian teratomas. The tumor is composed of tissues derived from the three germ layers and contains immature or embryonal structures or both. Its prognosis is correlated to the histologic grading (from 0 to 3), which corresponds to the degree of tissue differentiation. Immature teratoma may be combined with mature elements (most frequents forms) or with other neoplasic germ cell elements (choriocarcinoma, endodermal sinus tumor, dysgerminoma), called mixed forms. The chemosensitivity of immature teratoma is high, and the treatment combines surgical excision and, most often, chemotherapy.
Growing teratoma syndrome was first described in 1982 by Logothetis et al.2 It is observed in subjects surgically treated for nonseminomatous germ cell tumor. Three criteria are required for the definition of growing teratoma syndrome:
- Clinical or radiological enlargement of tumors during or after chemotherapy administered for nonseminomatous germ cell tumor;
- Normalization of previously elevated tumor markers (alpha fetoprotein [AFP] and human chorionic gonadotrophin [hCG] or both); and
- Absence of any nonseminomatous germ cell tumor component other than mature teratoma at histological examination of the entirely resected tumor.
Growing teratoma syndrome was first described in male patients treated for testicular nonseminomatous germ cell tumor. Its incidence is about 3% of patients treated with chemotherapy (2.8% in the study by Tonkin et al3 and 2.6% in that by Sonneveld et al4). In female patients, the first description was made in 1977, with a series of three patients affected by immature teratoma, who presented with all growing teratoma syndrome criteria, even though the syndrome had not been defined at the time.5 The second description was made in 1983 with two case reports.6 In a recent review, Amsalem et al7 have emphasized that growing teratoma syndrome is associated with chemotherapeutic retroconversion. We report here a series of 12 patients with growing teratoma syndrome who were treated for immature teratoma.
PARTICIPANTS AND METHODS
This is a single-institution retrospective study. We collected all cases registered prospectively in our database as ovarian immature teratoma (pure or mixed forms) and treated in the Institut Gustave-Roussy from 1980 to 2002. Among those patients, we have selected those who presented with a growing teratoma syndrome according to the definition by Logothetis et al.2
During that period, patients with ovarian immature teratoma were treated with initial surgery for the complete or partial resection of the primary tumor, followed by platinum-based chemotherapy regimen, except in case of stage I tumor (according to the International Federation of Gynecology and Obstetrics [FIGO] classification8) and grade 1 or 2 pure immature teratoma. After chemotherapy, and in case of residual masses and normalization of serum tumoral markers (AFP/hCG), our policy was the resection—as completely as possible—of those masses. Quality of surgical resection was classified into three types:
- R0: macroscopic and microscopic complete resection;
- R1: persistence of microscopic tumoral residues with no macroscopic residues (positive systematic biopsies);
- R2: incomplete resection with macroscopic residues.
Follow-up of patients with ovarian immature teratoma and growing teratoma syndrome was based on physical and gynecologic examinations, serum tumoral marker (AFP/hCG) measurements, and computed tomography (CT) scan. All slides (ovarian immature teratoma and growing teratoma syndrome) were analyzed by the same anatomopathologist from our Institute.
During the 22-year study period, 103 patients were treated at the Institut Gustave-Roussy for ovarian immature teratomas. Among those 103 patients, 12 presented a growing teratoma syndrome, allowing in this series an estimation of the incidence of at least 12%. Patients’ characteristics are summarized in Table 1.
The median age of participants was 15.5 years (range 9–29). Serum AFP was initially elevated in all patients but two and was not available for one patient.
In nine cases (75%), mature and immature components were associated. Only three cases of pure immature teratoma have been observed. There were six mixed germ cell tumors containing teratoma and another associated component (embryonal carcinoma and/or endodermal sinus tumor). Nine tumors (75%) contained predominantly immature neuroectodermic components.
Ten patients (83%) had a FIGO8 stage III tumor with peritoneal involvement in nine cases and a para-aortic lymph node involvement in only one. One patient had a disease limited to the ovary (stage Ic), and stage was not assessable in one patient.
The treatment of the primary tumor was surgery. Adnexectomy was performed in all cases: unilateral and isolated for four patients, unilateral and associated with peritoneal resection for five patients, bilateral with hysterectomy for two patients. Details of the initial surgery were not available for one patient. Finally, six patients had an initial R0 resection, four patients an incomplete resection with macroscopic residues (R2), and information was not available for two patients.
All patients but one were treated with platinum-based chemotherapy. At the end of the initial treatment, four patients out of 12 did not present any sign of disease. Regarding the eight patients left, residual disease was noted after the end of adjuvant chemotherapy, and a second surgery was then performed. Two of those eight patients presented persistent lesions of immature teratoma (peritoneal lesions in both cases) and received more chemotherapy. The other six patients had only growing teratoma syndrome lesions (peritoneal deposits for two patients, para-aortic adenopathy for one patient, and peritoneal nodules for the other three patients).
The median interval between the diagnosis of the primary ovarian germ cell tumor and the first growing teratoma syndrome occurrence was 9 months (range 4–55 months). Growing teratoma syndrome occurred during chemotherapy in one case with initial R2 resection; chemotherapy was stopped and surgery done 2 months later. For the 11 patients left, the diagnosis of growing teratoma syndrome was made no earlier than one month after the end of chemotherapy, with a mean time of 9 months after the end of chemotherapy (range 1–50 months). Growing teratoma syndrome occurred regardless of the macroscopic character, complete or not, of the initial resection of ovarian immature teratoma.
The growing teratoma syndrome was generally revealed (9 cases, 75%) by systematic radiological examinations during the follow-up (ultrasonography, CT scan, or magnetic resonance imaging). For two patients, growing teratoma syndrome was discovered during a second-look procedure, which was done after chemotherapy for an initial incomplete surgical resection (R2), whereas the patient was apparently in complete remission (evaluation of the residual disease after chemotherapy). In both cases, the surgeon discovered diffuse small-sized peritoneal deposits. In one case only, growing teratoma syndrome was discovered through clinical examination: that was a mass in the Douglas pouch after adjuvant chemotherapy.
In all cases but one, growing teratoma syndrome occurred in the site involved by the ovarian immature teratoma: the nine patients who had an initial stage III tumor with peritoneal involvement had a growing teratoma syndrome localized in the peritoneum. In the same way, the patient with only a retroperitoneal lymph node involvement at initial presentation had a growing teratoma syndrome only localized in those nodes. The only exception was a patient who presented with an initial stage Ic FIGO ovarian immature teratoma. In this patient, the growing teratoma syndrome was localized in para-aortic nodes. We can wonder whether the ovarian immature teratoma stage was or was not underestimated during initial surgery; the latter had consisted in a right adnexectomy isolated with normal macroscopic exploration of the abdominal cavity. Initial FIGO stage was not available for one patient. One patient was lost to follow-up after diagnosis of growing teratoma syndrome, which was a residual mass after adjuvant chemotherapy and mature teratoma on the biopsy. So this patient is evaluable only for the incidence of growing teratoma syndrome. For the other 11 patients, treatment of the first episode of growing teratoma syndrome was surgical. A complete resection could be performed in eight patients; there were two incomplete resections with macroscopic residues (R2) and one resection with microscopic residues (positive systematic peritoneal) (R1). No further treatment, such as chemotherapy or radiotherapy, was done after the resection of growing teratoma syndrome. One patient with R2 resection remains free of disease 9 years after the growing teratoma syndrome incomplete resection.
Four patients among 12 presented with a growing teratoma syndrome recurrence (second event) after a relatively long interval because it was radiologically observed respectively at 53, 67, 82, and 120 months after the surgery for initial growing teratoma syndrome. In all cases, the growing teratoma syndrome-recurrence occurred in the same place as the first event: peritoneal deposits in two cases, peritoneal nodules in one case, and inter hepatodiaphragmatic lesion in another one.
In one patient, growing teratoma syndrome recurrence was diagnosed because of mechanical complications (pain in the right hypochondrium). The three other recurrences were diagnosed by systematic radiological examinations (CT scan) during follow-up, including one observed more than 14 years after the initial diagnosis of ovarian immature teratoma. One of those patients experienced two episodes of intestinal obstruction that was treated medically before the diagnosis of growing teratoma syndrome recurrence (diagnosis performed by CT scan).
In all cases, the treatment of the recurrence consisted only in surgical resection. Among the four patients who experienced a growing teratoma syndrome recurrence, two had a complete resection (R0) of their first growing teratoma syndrome, one had a resection with microscopic residues (R1), and the last one had an incomplete resection (R2). So, patients with a first growing teratoma syndrome with incomplete resection did not have a recurrence because the second event occurred in the same place as the first one. For the patient with R1 resection, first growing teratoma syndrome occurred as disseminated peritoneal lesions, which would not allow a complete microscopic resection. For the patient with R2 resection, the lesion was not completely removed the first time because of its localization under the diaphragm but was totally removed 7 years later (slow growing). The patient remains free of disease 7 years after surgery.
After the treatment of growing teratoma syndrome recurrence, complications were observed in one patient. A bowel obstruction on adhesion occurred 5 years afterward. The obstruction was treated surgically, and the surgeon did not observe a new lesion of teratoma.
The study was ended in July 2003. The median follow-up from the initial diagnosis of ovarian immature teratoma was 144 months (range 9–198). One patient was lost to follow-up, immediately after the diagnosis of growing teratoma syndrome. Of the 11 patients alive, 10 are in complete remission. The patient who experienced a second event more than 14 years after the initial diagnosis of ovarian immature teratoma had an incomplete resection (R2) of this second event in February 2003. The growing teratoma syndrome lesions were localized in the peritoneum with numerous nodules. The medical follow-up is ongoing, and at this time, the patient does not demonstrate any symptoms related to the disease.
Growing teratoma syndrome is a complication observed in subjects treated for ovarian immature teratoma. Its incidence was established after testicular nonseminomatous germ cell tumor (from 3–10% in the literature3,4,9), but not for ovarian immature teratoma. In our series, the incidence is evaluated to be 12% of ovarian immature teratoma.
The pathogenesis of growing teratoma syndrome remains unclear. Two different mechanisms have been mentioned.6,10 The first hypothesis is that chemotherapy may induce malignant cell differentiation of immature teratoma into mature teratoma. Cells could acquire a benign phenotype insensitive to chemotherapy and grow on their own. Studies have demonstrated that malignant germ cells can differentiate alone in mature elements under chemotherapy (Merrin C, Baumgartner G, Wajsman Z. Benign transformation of testicular carcinoma by chemotherapy [letter]. Lancet 1975;2:43–4.).11 The second hypothesis is that chemotherapy can induce a selective destruction of immature components, whereas mature elements, resistant to chemotherapy, persist and grow alone as growing teratoma syndrome. One fact that seems to favor this hypothesis is that the mature teratoma component is very often associated with immature teratoma in the primary ovarian immature teratoma and developing into growing teratoma syndrome (nine cases, 75% in our series).10 The comparison between the chromosomal abnormalities found in mature residual teratomas following chemotherapy and those in primary testicular nonseminomas suggests that residual teratomas result from selection of clones from the primary tumor with a less abnormal karyotype.10,12
In our series, growing teratoma syndrome occurred in 10 cases in the peritoneal area. We can stress that nine of those patients had a peritoneal FIGO stage III ovarian immature teratoma. In two patients, growing teratoma syndrome was diagnosed in lymph nodes: one patient had an initial FIGO stage III ovarian immature teratoma without peritoneal disease, and the second one had a stage Ic. So, localizations of ovarian growing teratoma syndrome are different from those observed after testicular nonseminomatous germ cell tumor, most often localized in the retroperitoneal area9 or less frequently in viscera. This is in accordance with natural dissemination of ovarian and testicular tumors.
Despite their relative benign phenotype, two types of clinical complications can be observed in growing teratoma syndrome. Firstly, mechanical complications due to the growth of the tumoral masses, which can compress the surrounding organs. These complications are different, according to the site of growing teratoma syndrome: pain, intestinal obstruction, renal failure by ureteral compression, thrombophlebitis, and tissular necroses.10 In our study, these complications were observed in two patients.
Malignant transformation is the second complication. It occurred in 3% of the cases in the series by André et al.10 Mature teratomas have been reported to hold a malignant potential,10 including transformation into ovarian immature teratoma,13 sarcoma,14 squamous cell carcinoma,15 adenocarcinoma,16 carcinoid tumor,17 or primitive neuroectodermal tumor.18 So, histologically mature teratoma may have a benign appearance but with genomic malignancy.12,19 In our series, no case of malignant transformation was observed.
Given the possibilities of these complications, growing teratoma syndrome lesions have to be removed as completely as possible. In our study, a complete surgical resection could be done in eight cases. The problems that have been most often observed were due to the localization of growing teratoma syndrome, like peritoneal deposits in ovarian growing teratoma syndrome, which often does not allow a microscopic complete resection of the lesions.
The spontaneous evolution of growing teratoma syndrome is not possible to predict. Besides progressions or long stabilizations, few spontaneous regressions have been reported. In this study, all patients have been operated on for growing teratoma syndrome. One patient had a resection R2 of the growing teratoma syndrome, and recurrence was not observed during a 20 year follow-up, suggesting a spontaneous regression or stabilization of the residual peritoneal lesions or both. For the other two patients who had an incomplete resection of the growing teratoma syndrome lesions, recurrence occurred, respectively, 68 and 82 months after growing teratoma syndrome surgery.
Clinical responses and stabilizations have been reported with α-interferon in nonoperable diseases.3,9 Follow-up of patients treated for growing teratoma syndrome must be prolonged for many years because recurrences being observed up to 14 years later in our series.
Because of the rarity of nonseminomatous germ cell tumor and, even more, of growing teratoma syndrome, it is difficult to characterize factors predicting the subsequent development of growing teratoma syndrome. In male patients, the presence of mature teratoma in the primary nonseminomatous germ cell tumor is a predicting factor.10 In our series, we frequently found mature teratoma in the primary ovarian immature teratoma (nine cases, 75%), FIGO stage III disease with peritoneal involvement (eight cases, 66%), and the presence of predominantly immature neuroectodermic components in the primary germ cell tumor (nine cases, 75%). Because of the small number of patients, it is not possible to treat these factors as predictive. In fact, only incomplete resection of the primary ovarian immature teratoma seems to be a logical risk factor, but in our series, six patients experienced growing teratoma syndrome after an apparently complete initial resection (R0) of the ovarian immature teratoma. However, the disease was disseminated to the peritoneum in five of these cases, suggesting the persistence of microscopic peritoneal lesions of mature teratoma, even if there were macroscopically complete resection.
In conclusion, growing teratoma syndrome is a complication observed during the natural history of ovarian immature teratoma. With an incidence of at least 12% in our series, we show that it is not an exceptional evolution. Treatment of growing teratoma syndrome consisted in a surgical resection as complete as possible to avoid complications (mechanical and malignant transformations). Most of the patients are then in complete remission, and no further treatment is indicated. The medical follow-up of the patients has to be prolonged because of the possibility of very late recurrences. The prognosis remains good.
1. Talerman A. Germ cell tumours of the ovary: Blaustein’s pathology of the female genital tract. 3rd ed. New York (NY): Springer-Verlag; 1987. p. 659–721.
2. Logothetis C, Samuel M, Trindade A, Johnson DE. The growing teratoma syndrome. Cancer 1982;50:1629–35.
3. Tonkin KS, Rustin GJS, Wignall B, Paradinas F, Bennett M. Successful treatment of patients in whom germ cell tumour masses enlarged on chemotherapy while serum markers decreased. Eur J Cancer Clin Oncol 1989;25:1739–43.
4. Sonneveld DJ, Sleijer DT, Koops HS, Keemers-Gels ME, Molenaar WM, Hoekstra HJ. Mature teratoma identified after post-chemotherapy surgery in patients with disseminated non seminomatous testicular germ cell tumour. Cancer 1998;82:1343–51.
5. Di Saia P, Saltz A, Kagan AR, Morrow CP. Chemotherapeutic retroconversion of immature teratoma of the ovary. Obstet Gynecol 1977;49:346–50.
6. Aronowitz J, Estrada R, Lynch R, Kaplan AL. Retroconversion of malignant immature teratomas of the ovary after chemotherapy. Gynecol Oncol 1983;16:414–21.
7. Amsalem H, Nadjari M, Prus D, Hiller N, Benshushan A. Growing teratoma syndrome vs chemotherapeutic retroconversion: case report and review of the literature. Gynecol Oncol 2004;92:357–60.
8. Staging announcement. FIGO Cancer Committee. Gynecol Oncol 1986;25:383–5.
9. Jeffery GM, Theaker JM, Blaquiere RM, Smart CJ, Mead GM. The growing teratoma syndrome. Br J Urol 1991;67:195–202.
10. André F, Fizazi K, Culine S, Droz J, Taupin P, Lhomme C, et al. The growing teratoma syndrome: results of therapy and long-term follow-up of 33 patients. Eur J Cancer 2000;36: 1389–94.
11. Hong WK, Wittes RE, Hajdu ST, Cvitkovic E, Whitmore WF, Golbey RB. The evolution of mature teratoma from malignant testicular tumour. Cancer 1977;40:2987–92.
12. Castedo SM, De Jong B, Oosterhuis JW, Idenburg VJ, Seruca R, Buist J, et al. Chromosomal changes in mature residual teratomas following polychemotherapy. Cancer Res 1989;49:672–6.
13. Ahmed T, Bosl GJ, Hajdu SI. Teratoma with malignant transformation in germ cell tumours in men. Cancer 1985;56:860–3.
14. Hughes DF, Allen DC, O’Nell JJ. Angiosarcoma arising in a testicular carcinoma. Histopathology 1991;18:81–3.
15. Bontis J, Vakiani M, Vavilis D, Agorastos T, Dragoumis K. Squamous cell carcinoma arising from mature cystic teratoma: a report of three cases. Eur J Gynecol Oncol 1996;17:49–52.
16. Ahmed FY, Gore ME, Shepherd D, Fisher C. Transformation of residual mature teratoma to adenocarcinoma 10 years after initial treatment. Clin Oncol 1996;8:125–6.
17. Lancaster KJ, Liang CY, Myers JC, McCabe KM. Goblet cell carcinoid arising in a mature teratoma of the mediastinum. Am J Surg Pathol 1997;21:109–13.
18. Motzer RJ, Amsterdam A, Prieto V, Sheinfeld J, Murty VV, Mazumdar M, et al. Teratoma with malignant transformation: diverse malignant histologies arising in men with germ cell tumour. J Urol 1998;159:133–8.
19. Sella A, el Naggar A, Ro JY, Dexeus FH, Amato RJ, Lee JS, et al. Evidence of malignant features in histologically mature teratoma. J Urol 1991;146:1025–8.