Ultrasound features of immature ovarian teratomas: Case series and review of literature

Abstract

Introduction

Immature ovarian teratomas are rare but account for 10–20% of ovarian cancers in women under the age of 20 years. This study aimed to characterise immature ovarian teratomas using grey-scale and Doppler ultrasonography and review the literature to refine the diagnosis of immature ovarian teratomas.

Methods

Patients with a confirmed histological diagnosis of immature ovarian teratoma from years 2006–2018, who had undergone a transvaginal ultrasound at two large teaching hospitals, were identified. The imaging was retrieved from the centres clinical databases. Ultrasound scans were performed by experienced ultrasound examiners and described according to International Ovarian Tumour Analysis criteria.

Results

Eight patients were identified in total with a mean age of 26 years (range 13–35). Half of the patients had a past history of a mature ovarian teratoma (3 ipsilateral, 1 contralateral). The cysts were generally large (median 115 mm), fast growing unilateral lesions with a single, peripheral predominantly solid component arising from the cyst wall. The solid component was hyperechoic with multiple foci of fibrosis and numerous small cysts. The cystic component typically formed less than 75% of the lesion and the cyst fluid was of low-level echogenicity. Subjective assessment of vascularity of the solid part of the tumours varied between scores of 1 and 2. Tumour markers showed a raised serum a-fetoprotein level in 42% of these patients.

Conclusion

Although there were no ultrasound features that were pathognomonic of immature teratoma, the diagnosis should be suspected in a young woman with a large ovarian cyst with a fibrotic, microcystic solid component, particularly if she has a past history of a dermoid cyst.

Keywords

Ultrasound immature ovarian teratoma histopathology dermoid cyst

Introduction

Immature ovarian teratomas (IOTs) are a subtype of female malignant ovarian germ cell tumours (MOGCTs). In the 1960s, Thurlbeck and Scully described IOTs as pure or mixed ovarian tumours¹ that represent less than 1% of all ovarian cancers.^1–3 They account for 10–20% of ovarian malignancies in women under the age of 20 years.³ The mean age at presentation is 27 years (range 18–36 years) and patients most frequently present with rapid abdominal distension (81%), abdominal pain (11%) and irregular periods.⁴

In the pathology literature, they have been described as large cystic-solid tumours. The tumours are mainly unilateral (approximately 80%)⁴ and more than 90% are diagnosed at International Federation of Gynecology and Obstetrics (FIGO)stage I.⁵ Microscopically, they can be found in pure form or as a mixed germ cell tumour. When mixed, immature neuroepithelium is the most frequently detected component in an IOT.^2,6,7 IOTs are graded 1–3 depending on the amount of atypia of the immature tissue.^2,8 There is a paucity of literature to date regarding the ultrasound features of immature ovarian teratomas, so the aim of this study was to characterise IOTs using grey-scale and Doppler ultrasonography and review the literature to help to refine the diagnosis and increase awareness of these tumours.

Methods

Patients with a confirmed histological diagnosis of an IOT between years 2006 and 2018 were included in this study. Clinical and ultrasound data were initially collected prospectively and stored electronically (ViewPoint, GE Healthcare). The ultrasound scans were performed pre-operatively by experienced gynaecological sonographers and the features described prospectively according to IOTA criteria.⁹ The images were retrieved and also retrospectively analysed by two senior clinicians independently for additional descriptive morphological features. Seven out of eight patients had a transvaginal ultrasound scan performed; one patient had a transabdominal ultrasound only. The following parameters were assessed; size of the cyst, laterality, the percentage of solid component present, the histopathological presence of hair, fat and sebum and the presence of fine echogenic spicules on ultrasound thought to represent fibrosis or fine calcification. Subjective assessment of vascularity at power Doppler examination was described in accordance with Timmerman et al.⁹ (1 = no vascularisation; 2 = minimal vascularisation; 3 = moderate vascularisation and 4 = strong vascularisation). Serum tumour markers and the demographics of the patients collected at the time of the scans were summarised. Ethical approval was sought by the local Research & Development teams and granted.

Results

Eight patients were identified in total with a mean age of 26 years (range 13–35 years) (Table 1). The patients presented with an abdominal mass, pelvic pain, amenorrhoea with suspicion of pregnancy or a mass that was found incidentally during early pregnancy scans (2/8 patients). Half of the patients had a previous history of an ovarian dermoid. The original presumptive diagnoses were dermoid cyst, Krukenberg tumour, ‘complex ovarian cyst’ and immature teratoma. The rapid growth of these tumours was evident in 3/8 (38%) patients who had a normal scan in the 6–18 months preceding the diagnosis of IOT. Serum CA125 was mildly raised in 6/8 (75%), alpha-fetoprotein in 5/7 (72%) of patients, and CA19-9 in 2/5 (40%). All tumours were diagnosed at FIGO Stage I. Histopathological grading of the tumours was G3 for 4/8 (50%) tumours, G2 for 2/8 (25%) and G1 for 2/8 (25%). There was no correlation seen between the grading of an IOT and the degree of elevation of the serum tumour markers.

Table 1.

Patients’ demographics and clinical details.

Age	Presenting complaint	Previous history of dermoid?	Interval from negative scan (months)?	CA125 kU/L (0–35)	CA19-9KU/L (0–37)	AFP kU/L (<7)	LDH U/L (0–175)	Grade and stage
26	Pain and pelvic mass	No	No	125	2724	57		G3 1C
25	Incidental in pregnancy	Yes – ipsilateral	No	7	–	3	–	G1 1A
35	Incidental in pregnancy	Yes – ipsilateral	No	125	2724	57	161	G3 1C
24	Pain	No	9	204	<3	13	193	G2 1A
13	Pain	Yes – contralateral	No	54	–	141	–	G3 1C
31	Incidental in pregnancy	No	18	163	<3	–	–	G3 1A
25	Pelvic mass	No	No	16	–	5	–	G2 1A
18	Pelvic mass	Yes – ipsilateral	6	177	<3	2484	288	G1 1A

On ultrasonography the predominant features were that the lesions were large (median 115 mm, range 48–158 mm) with a solitary solid component arising from the cyst wall. They were all unilateral. They contained low-level echogenic fluid contents and the solid compartment made up ≥15% of the cyst. On correlation with histopathology it was clear that the multiple small hyperechoic areas represented sebum. Adjacent to these were several small echogenic foci which signified calcification. There were also numerous small cysts seen within the solid compartment of the tumour which contained serous, mucinous or haemorrhagic fluid.

The case in Figure 1(a) shows a single, solid polypoid mass with a coarsely papillary surface protruding from the wall of the cyst. The fine echogenic spicules seen in the cyst fluid represent hair. Areas of calcification are seen within the solid components (Figures 1 and 2). One patient had a focal histopathology finding of an embryonal carcinoma within the immature teratoma (Figure 1(c)). The cystic-solid masses in Figure 2(a) to (d) of immature teratomas are large, predominantly solid masses interspersed by multiple cystic areas containing serous/mucinous fluid. Pockets of haemorrhage are adjacent to these cysts. No acoustic shadowing was seen behind the solid components, a typical finding in benign teratomas. Subjective assessment of vascularity of the tumours varied between scores of 1–2 as seen in Figure 1(b) and (d). Extensive implants of mature teratoma, gliomatosis peritonei, were discovered during one patient’s fertility sparing surgery (Figure 1(d)). Only 1/8 (12.5%) patients had ascites present. The ultrasonography findings have been summarised in Table 2.

Figure 1.

Immature ovarian teratomas showing low-level echogenic fluid contents and solid compartment made up ≥15% of the cyst.

Figure 2.

Numerous focal cystic areas and haemorrhage in predominantly solid immature ovarian teratomas.

Table 2.

Ultrasound features of immature ovarian teratomas.

Large unilateral cystic-solid tumour (median 115 mm, range 48–158 mm)

Solitary solid component arising from cyst wall comprising ≥15% of the cyst

No acoustic shadowing behind the solid component

Within solid compartment – multiple small hyperechoic areas represent sebum; echogenic foci signify calcification

Low-level echogenic fluid in cystic area with fine echogenic spicules seen representing hair

Subjective assessment of vascularity of the tumours varied between scores of 1 and 2

Discussion

To date, there is limited literature on the ultrasound appearances of IOTs, most likely due to their rarity. Despite it being a retrospective study, we have presented the largest case series to characterise IOTs to date using grey-scale/power/colour Doppler transvaginal ultrasonography. Our study was limited by the small number of cases presented. Immature ovarian teratoma is a rare tumour and the analysis could have been improved with a larger cohort of patients across a wider network of tertiary hospitals with gynae-oncology services. The retrospective analysis of data meant that we relied on images and information available and thus were unable to adjust parameter such as the pulse repetition frequency (PRF) for an improved Doppler score analysis.

Transvaginal ultrasonography is used as the first-line technique to characterise ovarian lesions. There have been many approaches as to what are the best scoring systems and imaging techniques for characterising adnexal masses preoperatively.³ The International Ovarian Tumour Analysis (IOTA) group defined the terms used to describe ovarian lesions on ultrasound in order to standardise the sonographic assessment. The group has also devised several models to attempt to discriminate benign from malignant ovarian tumours including the Simple Rules, Logistic regression 1&2 and ADNEX models.¹⁰ The subjective assessment of benign and malignant adnexal masses by Doppler ultrasound, based on pattern recognition by experts, has a high sensitivity (77–98%), and specificity (89–100%),^11–13 and is superior to or at the very least equal to ultrasound models.¹⁰ Subjective examination of fibromas, para-ovarian cysts and rare benign tumours still remains a diagnostic challenge with lower sensitivity rates (8–17%).¹¹ A careful clinical history needs to be correlated with the imaging findings to increase the sensitivity and specificity of diagnosis of ovarian masses.¹⁵

The use of computed tomography (CT) and magnetic resonance imaging (MRI) for the diagnosis of ovarian malignancy increases neither the sensitivity nor specificity as compared to the assessment of ovarian masses by TVUS.^16,17

It has been historically difficult to ascertain the imaging characteristics of IOTs as they are rare, fast growing and can vary in their appearance over time.⁶ There have been only a few ultrasound images available in the literature and no case series describing the features of IOTs on Doppler ultrasound examination to date.⁴ Sectional imaging by CT and MRI have been the main techniques used to describe their imaging features. A review by Outwater et al.¹⁸ concluded that there were no specific features seen on ultrasound examination of IOTs and they are described as heterogenous solid masses with scattered calcifications on MRI and CT.¹⁹ A retrospective MRI study by Yamaoka et al.²⁰ of 10 patients (mean age 19.0 years) with pathologically proven IOTs described the lesions as solid/cystic masses with punctuate foci of fat in contrast to the predominantly sebaceous cystic fluid seen in mature cystic teratomas. Up to 60% of patients demonstrated ascites in this study. A further review by Peterson et al.²¹ suggests that it is difficult to distinguish between mature and immature teratomas using imaging alone and histological examination is required. A summary of the previous imaging studies is shown in Table 3.

Table 3.

Histopathology and imaging findings of immature ovarian teratomas.

Author	Macroscopic features	MRI/CT	US
Norris et al,¹⁴ n = 58	Encapsulated, smooth, visibly cystic lesions(serous), haemorrhage and necrosis present	–	–
Mlikotic et al.,²² n = 8	Large encapsulated masses, median diameter 18 cm, large cystic components, areas of necrosis and heamorrhage		Complex mass with an irregular branching solid component (isoechoic), thin septations.
Outwater et al.,²¹ n = 1	Large (14–25 cm) solid/cysts	Irregular solid component containing coarse calcifications and small foci of fat.	Non-specific, heterogeneous, partially solid lesions, scattered calcifications with small foci of fat within the solid component.
Yamaoka et al.,¹⁹ n = 10	–	Fat-containing tumours with solid components consisting of numerous cysts.Wide variety of signal intensities on T2-weighted images.	–
Saba et al.⁷	Tumours are large, encapsulated masses with a rich solid component, composed of immature neuroectodermal tissue	Small foci of fat, and calcifications tend to be coarse or ill-defined. Cystic areas in IOTs are usually filled with serous, mucinous fluid.
Peterson et al.,²⁰n = 10	–	Prominent soft tissue components. Calcifications scattered throughout lesions	Predominantly cystic retroperitoneal mass
Nishida et al.,¹⁸n = 3	–	No solid components.High signal intensity on T1-weighted images and low to iso-intensity signal on T2-weighted images.Fatty elements were identified on fat-saturated T1-weighted images.

US: ultrasound; CT: computed tomography; MRI: magnetic resonance imaging; IOTs: immature ovarian teratomas.

Women of reproductive age are the largest group of patients who present with ovarian cysts.³ The characterisation of an ovarian tumour allows patients to be appropriately triaged to the care of the gynaeoncologist where they can receive optimum care and the correct surgery yielding improved survival rates.^15,16 It is therefore important to achieve an early diagnosis and referral to a specialist centre as this should reduce patient morbidity and risk to fertility. Differentiating between common benign cysts and ovarian malignancies (overall lifetime risk of 1.4%)¹⁰ using grey scale and Doppler ultrasonography has become a necessary skill in gynaecological ultrasound.

IOTs are large unilocular-solid tumours that contain fluid of low-level echogenicity. The solid mass typically contains multiple small cystic spaces, arising from the cyst wall, with no papillary projections. The solid mass is heterogeneous, with fine hyperechoic spicules and occasionally small foci of fat. This distinguishes them from typical benign dermoids that are usually smaller (median size 60–70 mm) and contain hyperechoic sebum. Dermoids commonly have a Rokitansky nodule (Figure 3(a)), distinct hyperechoic mural tubercles that are associated with a strong acoustic shadow, or a fluid level due to an interface with fat as seen in Figure 3(b).^15,22

Figure 3.

Dermoid cysts with (a) Rokitansky nodule and (b) cystic dermoid.

The cystic component of an IOT contains sparse bright echoes in the low-level echogenic fluid. Dermoid cysts on the other hand contain a greater proportion of bright echoes due to strands of hair. In a cystic teratoma, an isoechoic branching solid component has been suggested to be significantly associated with a malignancy. This suggests that the most discriminating feature of an IOT may be its solid component.^7,23

These morphological characteristics are also different from other ovarian lesions found in young women. Small benign unilocular cysts such as corpus luteal cysts (median 20–50 mm) or endometriomas (median 58 mm)²⁴ are easily distinguishable from the large unilocular-solid cyst of IOTs. Corpus luteal cysts are usually thick walled, cystic/solid/haemorrhagic cysts that demonstrate peripheral circumferential blood flow, termed the ‘ring of fire’¹⁵ (Figure 4). Endometriomas with a glass-ground appearance, shown in Figure 4(b), and functional haemorrhagic cysts, can appear to have solid tissue due to bleeding within the cyst (Figure 4(c)). This is easily demonstrated as an avascular jelly-like mass when on Doppler examination and nudging with the transducer.²⁵ Decidualisation of an endometrioma sometimes seen in pregnancy shows multiple vascularised papillary projections unlike an IOT.¹⁵ However, malignant transformation of an endometrioma (median 107 mm)²⁵ has similarities on transvaginal ultrasound scan to IOTs. Endometrioid ovarian carcinoma is characterised by the presence of a solid mass with heterogeneous cystic contents and is more likely to show a Doppler score of 3–4. Patients with endometrioid ovarian carcinoma are also likely to be older (median 58 years old, range 28–59) and postmenopausal.²⁶

Figure 4.

(a) Solid corpus luteum with ‘ring of fire’; (b) typical ground-glass appearance of endometrioma; (c) haemorrhage within endometrioma.

An IOT with a solid mass of less 20% may look similar to a mucinous cystadenoma. Mucinous tumours may be benign, borderline or malignant.²⁶ A mucinous cystadenoma (median 180 mm) is the most common type of benign mucinous tumour of the ovary and is typically a multilocular cyst with mixed echogenicity of the cyst contents.^26,27 It shows typically thin branching septations within the cyst rather than a solid component (Figure 5a). Cystadenofibromas are either a unilocular-solid or multilocular-solid masses with the presence of hyperechoic solid components and spicules representing fibrosis (Figure 5(b)). However, these cysts are common in postmenopausal women rather than the younger age group with IOT. Acoustic shadows are present and a vascularity of 1–2 on Doppler examination.²⁸

Figure 5.

(a) Mucinous cystadenoma; (b) cystadenofibroma; (c) fibroma.

Borderline mucinous cystadenomas (Figure 6) on the other hand usually have a cluster of ‘honeycomb’ septations within the cyst, due to the arrangements of columnar cells lining the cyst wall.^15,27 Multiple papillary projections are typical of serous or sero-mucinous borderline tumours. The single solid mass in an IOT has small multicystic spaces, fills at least ≥15% of its cyst content and is clearly seen separate from any septations.

Figure 6.

Borderline mucinous cystadenomas.

Conclusion

In conclusion, IOTs are rare, rapidly growing ovarian cysts that typically have a large predominantly solid, poorly or moderately vascularised component. This component is typically less cystic than the multilocular ‘honeycomb’ nodules of mucinous borderline tumours and much larger than the small papillary projections seen in serous and sero-mucinous borderline tumours. Although some of the tumours contained hair and fat, the hyperechoic sebaceous material that is pathognomonic of dermoid cysts was not a typical feature. The images and features described should help to improve the presumptive preoperative diagnosis of IOT in young patients who present with large ovarian cysts, particularly if they have a history of a previous dermoid cyst.

Footnotes

Contributors

SAI and JAR researched literature and conceived the study. SAI, KH and JAR did the data analysis. SAI wrote the first draft of the manuscript. SAI, KH and JAR wrote the final version of the manuscript. All authors reviewed and approved the final version of the manuscript. JAR was the project supervisor and the department clinical lead.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics approval

Local approval was obtained from the Research and Development Committee at King’s College Hospital in September 2018.

Guarantor

JA Ross.

ORCID iDs

S Abdullahi Idle

JA Ross

References

Thurlbeck

Scully

RE.

Solid teratoma of the ovary. A clinicopathological analysis of 9 cases.

Cancer 1960; 13: 804–811.

Alwazzan

Popowich

Dean

, et al.

Pure immature teratoma of the ovary in adults: thirty-year experience of a single tertiary care center.

Int J Gynecol Cancer 2015; 25: 1616–1622.

RCOG. Management of female malignant ovarian germ cell tumours. Scientific Impact Paper Nos. 52, 2016.

Vicus

Beiner

Clarke

, et al.

Ovarian immature teratoma: treatment and outcome in a single institutional cohort.

Gynaecol Oncol 2011; 123: 50–53.

Ting

Yan

Xiao

, et al.

Ovarian cystectomy in the treatment of apparent early-stage immature teratoma.

J Int Med Res 2017; 45: 771–780.

Trabelsi

Conan-Charlet

Lhomme

, et al.

Peritoneal glioblastoma: recurrence of ovarian immature teratoma.

Ann Diagn Pathol 2002; 22: 130–133.

Saba

Guerriero

Sulcis

, et al.

Mature and immature ovarian teratomas: CT, US and MR imaging characteristics.

Eur J Radiol 2009; 72: 454–463.

Jorge

Jones

Chen

, et al.

Characteristics, treatment and outcomes of women with immature ovarian teratoma, 1998-2012.

Gynaecol Oncol 2016; 142: 261–266.

Timmerman

Valentin

Bourne

, et al.

Terms, definitions and measurements to describe the sonographic features of adnexal tumors: a consensus opinion from the International Ovarian Tumor Analysis (IOTA) Group.

Ultrasound Obstet Gynaecol 2000; 16: 500–505.

10.

Kaijser

Vandecaveye

Deroose

, et al. Imaging techniques for the pre-surgical diagnosis of adnexal tumours. Best Pract Res Clin Obstet Gynaecol 2014; 28: 683–695.

11.

Sokalska

Timmerman

Testa

, et al.

Diagnostic accuracy of transvaginal ultrasound examination for assigning a specific diagnosis to adnexal masses.

Ultrasound Obstet Gynecol 2009; 34: 462–470.

12.

Meys

EMJ

Jeelof

Achten

NMJ

, et al.

Estimating risk of malignancy in adnexal masses: external validation of the ADNEX model and comparison with other frequently used ultrasound methods,

Ultrasound Obstet Gynaecol 2017; 49: 784–792.

13.

Valentin

Hagen

Tingulstad

, et al.

Comparison of ‘pattern recognition’ and logistic regression models for discrimination between benign and malignant pelvic masses: a prospective cross validation.

Ultrasound Obstet Gynaecol 2001; 18: 357–365.

14.

Norris HJ, Zirkin HJ, Benson WL. Immature (malignant) teratoma of the ovary: a clinical and pathologic study of 58 cases. Cancer 1976; 37: 2359--2372.

15.

Sayasneh

Ekechic

Ferrara

, et al.

The characteristic ultrasound features of specific types of ovarian pathology (Review).

Int J Oncol 2015; 46: 445–458.

16.

Timmerman

Testa

Bourne

, et al.

Logistic regression model to distinguish between the benign and malignant adnexal mass before surgery: a multicenter study by the International Ovarian Tumor Analysis Group.

J Clin Oncol 2002; 23: 8794–8801.

17.

Royal College of Obstetricians & Gynaecologists. Management of suspected ovarian masses in premenopausal women. Green-top Guideline No. 62, 2011.

18.

Outwater

Siegelman

Hunt

Ovarian teratomas: tumour types and imaging characteristics.

RadioGraphics 2001; 21: 475–490.

19.

Nishida

Kawano

Yuge

, et al. Three cases of immature teratoma diagnosed after laparoscopic operation. Clin Med Insights 2014; 7: 91–94.

20.

Yamaoka

Togashi

Koyama

, et al.

Immature teratoma of the ovary: correlation of MR imaging and pathologic findings.

Eur Radiol 2003; 13: 313–319.

21.

Peterson

Buckley

Holley

, et al.

Teratomas: a multimodality review.

Curr Prob Diagn Radiol 2012; 41: 210–219.

22.

Cohen

Sabbagha

Echo patterns of benign cystic teratomas by transvaginal ultrasound.

Ultrasound Obstet Gynaecol 1993; 3: 120–123.

23.

Mlikotic

McPhaul

Hansen

, et al.

Significance of the solid component in predicting malignancy in ovarian cystic teratomas.

J Ultrasound Med 2001; 20: 859–866.

24.

Testa

Timmerman

Van Holsbeke

, et al.

Ovarian cancer arising in endometrioid cysts: ultrasound findings.

Ultrasound Obstet Gynaecol 2011; 38: 99–106.

25.

Valentin

Use of morphology to characterize and manage common adnexal masses.

Best Pract Res Clin Obstet Gynaecol 2004; 18: 71–89.

26.

Caspi

Hagay

Appelman

Variable echogenicity as a sonographic sign in the preoperative diagnosis of ovarian mucinous tumors.

J Ultrasound Med 2006; 25: 1583–1585.

27.

Brown

Frumovitz

Mucinous tumours of the ovary: current thoughts on diagnosis and management.

Curr Oncol Rep 2014; 16: 389.

28.

Yen

Khong

Lamba

, et al.

Ovarian fibromas and fibrothecomas: sonographic correlation with computed tomography and magnetic resonance imaging: a 5-year single-institution experience.

J Ultrasound Med 2013; 32: 13–18.