Significant risk factors for malignant transformation of ovarian endometrioma during dienogest treatment: a case report and retrospective study

Endometriosis is among the most common gynecological diseases and is known to affect 2–10% of women of reproductive age [1, 2]. This disease most commonly affects the ovaries, forming a cystic mass called an ovarian endometrioma (OMA) [3], and is diagnosed by an outpatient ultrasound examination. In many cases, magnetic resonance imaging (MRI) is needed for a definitive diagnosis [4]. Because endometriosis is frequently associated with infertility and pelvic pain [5], laparoscopic surgery or drug treatment is often performed for symptom relief [6]. The use of dienogest therapy (DNGT), a drug-based treatment, has recently expanded because DNGT has a curative effect in terms of pelvic pain [7] and is expected to reduce the size of an OMA [8]. This therapy can also reduce postoperative lesion recurrence and extend the pain-free period [9]. Thus, dienogest is a new and effective drug. However, several adverse effects, including uterine bleeding, have been reported [10]. Among these side effects, some cases of malignant transformation (MT) during DNGT have been reported [11], although MT during DNGT is very rare. However, a detailed analysis of this risk factor has not been performed. For this reason, in addition to the present report of several cases of MT during DNGT in our hospital, we have performed a multivariate analysis of the risk factors for MT during DNGT.

Recent studies have indicated that endometrioid carcinoma and clear cell carcinoma are associated with OMA [12]. Ovarian endometrioma is a subtype of endometriosis, affecting 17–44% of women with endometriosis [13]. In most cases, this disease is diagnosed as endometrioma by an outpatient ultrasound examination [4], and laparoscopic surgeries are often performed to treat pelvic pain and subfertility [6]. In a previous report, these cancers reportedly developed in 0.72% of women with a previous diagnosis of OMA who had been managed expectantly [14]. Conversely, some reports showed that a certain ratio of patients with ovarian cancers have a history of endometriosis [15, 16]. However, no significant relationship between previous treatments with hormone drugs (or surgery) and the risk of MT has been validated [17, 18]. Therefore, our analysis may provide new insight into preventing future cases of MT of OMA.

A 43-year-old Japanese woman, gravida 0, para 0, was referred to our hospital because of aggravated dysmenorrhea and hypermenorrhea. An ovarian cyst had been noted in this patient 3 years before her visit to our hospital, and she had dysmenorrhea beginning at approximately 30 years of age. MRI that was performed before her visit to our hospital suggested a left endometrioma that was 7 cm in diameter and a right endometrioma that was 3 cm in diameter, with no solid component in either endometriotic cyst. Bilateral laparoscopic cystectomy (LC) was performed with the patient under general anesthesia, and the diameters of the right and left endometriomas were 3 cm and 7 cm, respectively (Fig. 1a and b). After both ovaries were subjected to a surgical procedure, including complete Douglas pouch obliteration, to release them from their severe pelvic adhesion, the left endometrial cyst was completely excised, and the right endometrial cyst was ablated. The rASRM score was 81 points, and this case was diagnosed as stage IV endometriosis. The ovarian cyst did not contain a solid component (Fig. 1c), and pathological examination showed a left endometriotic cyst and no malignancy (Fig. 1d). After surgery, DNGT was chosen for the prevention of recurrence, and the patient was monitored by transvaginal ultrasound every 6 months.

At 1 year and 1 month after surgery (that is, 1 year after the start of the administration of DNGT), when the patient was 44 years old, a recurrent left-sided OMA that was 6 cm in diameter with a solid component inside the left ovarian tumor was detected by outpatient transvaginal ultrasound examination. Contrast-enhanced MRI showed that in addition to a 3-cm right endometrioma, there was a 7-cm left endometrioma with a solid component inside the cavity in which MT was strongly suspected (Fig. 1e). In addition, contrast-enhanced computed tomography exhibited numerous para-aortic lymph node enlargements, including a maximum-sized node that was 10 mm in diameter. The serum cancer antigen 125 level, which was 51.4 U/ml at the time of the first visit to our hospital, was increased to 515 U/ml. Soon after the imaging diagnosis was performed, complete abdominal surgery, including removal of the uterus, bilateral ovaries, omentum, pelvic lymph nodes, and para-aortic lymph nodes, was performed. The final diagnosis was clear cell carcinoma of the left ovary (Fig. 1f, g, h), International Federation of Gynecology and Obstetrics (FIGO) stage 1C1, pT1C1N0M0. After the necessity of chemotherapy was explained to her, she received chemotherapy of 175 mg/m² paclitaxel and carboplatin (TC; area under the curve, 6 mg/ml/min) for six cycles and was without recurrence during follow-up.

Currently, DNGT is being used for patients with symptomatic endometriosis with or without surgery [21]. This progestin agent is reported to decrease the risk of recurrence by inhibiting the growth of endometriotic tissue [21, 22]. This mechanism can also theoretically prevent MT [23], and reported cases of MT are very rare [24]. However, in our hospital, we had four cases in which ovarian carcinoma was diagnosed during DNGT for OMA. MT in two of the four cases was detected entirely in our hospital, and the probability of MT after starting DNGT was 1.2% (n = 2 of 172). This probability was similar to the ability of endometriosis to transform into malignancy, which has been reported to be approximately 1% of OMAs in previous studies [14]. In this study, we extracted the details of the patients’ clinical histories. Because all four patients were over 40 years old and nulliparous, we hypothesized that there were several risk factors, including the patient’s age and parity, for using DNGT. To verify the significant influence of risk factors on MT, a multivariate analysis of six factors extracted before starting DNGT was performed (Table 2). Of the six factors, advanced age, nullipara and increased size and recurrence showed significant increases in the probability of MT. In contrast to a previous study in which the association between spontaneous pregnancy and the disease progression of endometriosis was unclear [25], these results were roughly consistent with those of past reports [26, 27]. In particular, 19 patients who had all three factors showed a probability of over 20% for the occurrence of ovarian carcinoma (21.1%, n = 4 of 19). Because the sample number was very small, the result for this extracted high-risk group might be coincidental. Therefore, a larger sample size will be needed to determine the risk factors for MT. However, in contrast to previous studies, large cyst [26‐28] and surgical history [29] did not show a significantly higher probability of MT. “Bilateral cysts” were also not significant. These multivariate analysis results presented the possibility that the characteristics of the patient have a stronger influence on MT than the characteristics and treatment approach of the OMA itself. Because two cases among the four cases of MT exhibited recurrences of OMA after LC and because the other two cases exhibited cyst size enlargement after starting DNGT, careful observation of the cyst size is important, regardless of the use of surgical treatment. However, similar to previous reports [11], in our hospital, we could not prevent the two cases of MT (Table 1, case 1 and case 2) by follow-up ultrasound examination at an interval of 6 months.

Although DNGT seems to be effective in controlling OMA, careful observation of the cyst size is needed. Because the probability for ovarian carcinoma occurrence could exceed 20% in some cases, particularly for nulliparous women of advanced age, a more individualized choice of therapy might be needed. For a more accurate determination of the treatment approach, a larger sample size will be needed to determine the risk factors for MT.