Abstract
Summary
Total testosterone, which is peripherally converted to its biologically active form dihydrotestosterone (DHT), is the first-line hormone investigation in hyperandrogenic states and infertility in premenopausal women. Polycystic ovary syndrome (PCOS), the most common cause of hyperandrogenism and infertility in young women, is often associated with mild elevations of total testosterone. Whereas very high levels of total testosterone (>2–3 SD of normal reference), are most often associated with hyperandrogenic signs, menstrual irregularity, rapid onset of virilization, and demand a prompt investigation. Herein, we report a case of a 32-year-old woman who was referred to the endocrinology outpatient clinic due to secondary amenorrhea and extremely high testosterone levels without any virilization signs. We initially suspected pitfalls in the testosterone laboratory test. Total serum testosterone decreased after a diethyl-ether extraction procedure was done prior to the immunoassay, but testosterone levels were still elevated. An ovarian steroid-cell tumor (SCT) was then revealed, which was thereby resected. Twenty-four hours post surgery, the total testosterone level returned to normal, and a month later menstruation resumed. This case emphasizes that any discrepancy between laboratory tests and the clinical scenario deserves a rigorous evaluation to minimize misinterpretation and errors in diagnosis and therapeutic approach. Additionally, we describe a possible mechanism of disease: a selective peripheral target-tissue response to high testosterone levels that did not cause virilization but did suppress ovulation and menstruation.
Learning points
-
Total testosterone is the most clinically relevant hormone in investigating hyperandrogenic states and infertility in premenopausal women.
-
Very high total testosterone levels in women (>2–3 SD of normal reference) are most often associated with hyperandrogenic signs, menstrual irregularities, and a rapid onset of virilization.
-
In women with very elevated testosterone levels and the absence of clinical manifestations, laboratory interference should be suspected, and diethyl ether extraction is a useful technique when other methods fail to detect it.
-
Ovarian steroid cell tumors (SCT) encompass a rare subgroup of sex cord-stromal tumors and usually secrete androgen hormones. SCTs are clinically malignant in 25–43% of cases.
-
A selective response of peripheral target tissues to testosterone levels, with clinical manifestations in some tissues and no expression in others, may reflect differences in the conformation of tumor-produced testosterone molecules.
Background
High androgen levels and infertility in reproductive-age women, beyond the most common cause of polycystic ovary syndrome (PCOS), are often challenging diagnoses. The causes include ovarian and adrenal tumors, Cushing’s syndrome, classic and non-classic congenital adrenal hyperplasia, ovarian hyperthecosis, and rare genetic diseases that impair adrenal steroidogenesis (1). Regardless of the cause, very high testosterone levels (reference value (RV): <2.1 nmol/L), higher than 3.5–7 nmol/L, are most often associated with hyperandrogenic signs, menstrual irregularities, and rapid onset of virilization (2).
We report a case of an obese 32-year-old woman that was referred to our department to investigate secondary amenorrhea and extremely high testosterone levels but without any virilization signs. For the last 5 years, the patient was on routine surveillance because of PCOS and a known ovarian mass presumed to be a dermoid cyst. PCOS and obesity can justify amenorrhea but cannot explain the absence of virilization signs when having very high testosterone levels; therefore, laboratory interference was suspected. After a laboratory investigation, testosterone levels were found to be significantly lower but still above 3× the upper normal female range reference. Reevaluation of the known ovarian mass raised malignancy suspicions. Finally, an ovarian SCT was resected, and testosterone levels returned to normal female ranges. This unique case illustrates two hormonal alterations from the same source: (i) pitfalls in testosterone tests and (ii) a testosterone-produced tumor with selective peripheral target-tissue response to high testosterone levels.
Case presentation
Patient history
A 32-year-old female was referred to our endocrinology and metabolic department due to a very high total testosterone level of 22.1 nmol/L. She was married, without children, and was planning a pregnancy. Menarche occurred at the age of 11. At 16 years of age, secondary amenorrhea was diagnosed along with weight gain, without signs of acne or hirsutism. Her total testosterone level was then 2.8 nmol/L, androstenedione 10.5 nmol (RV: 1.2–7.8 nmol/L), without other hormonal abnormalities. An ultrasound revealed a 2 cm ovarian mass, presumed to be a dermoid cyst and PCO features. Combined hormonal treatment with ethinylestradiol 0.03 mg and drospirenone 3 mg was started, followed by ovarian ultrasound surveillance, which did not show significant ovarian mass growth. Three years before the referral, hormonal treatment was stopped with the intention of conceiving, but menstruation did not resume. Her past medical history included obesity and type 2 diabetes mellitus, which was well-controlled with oral antidiabetic drugs. Her family history included diabetes and obesity. No infertility problems were recorded in her parents and siblings. At presentation, except for obesity (body mass index (BMI) 35 kg/m2), her physical examination was normal. There were no signs of hirsutism, masculine musculature build, deep-pitch voice, or male pattern baldness. The gynecological examination was unremarkable. Baseline chemistry and complete blood count laboratory data were normal. The hormone profile showed a total testosterone level of 21.8 nmol/L, free androgen index (FAI) 88.2 (RV: 0.3–4.4), sex hormone binding globulin (SHBG) 24.7 nmol/L (RV: 17.7–138.3 nmol/L), androstenedione 5.9 nmol/L (RV: 1.29–7.8 nmol/L), 17-hydroxyprogesterone 11.2 nmol/L (RV: 0.3–15.2 nmol/L), dehydroepiandrosterone sulfate (DHEA-S) 2.18 µmol/L (RV: 0.9–11.6 µmol/L), follicle-stimulating hormone (FSH) 5.3 IU/L, luteinizing hormone (LH) 2.5 IU/L, estradiol (E-2) 171 pmol/L, and progesterone 3.9 nmol/L. The thyroid function test and prolactin levels were within the normal range (RV: 119–618 nmol/L). Morning serum cortisol of 19 nmol/L (RV: 119–618 nmol/L). A post 1.0 mg overnight dexamethasone suppression test ruled out Cushing's syndrome. A normal response of the 17-hydroxyprogesterone level to an ACTH-stimulation test (basal 18.2 nmol/L, after stimulation 17.0 nmol/L) ruled out a non-classical congenital hyperplasia. The anti-Mullerian hormone (AMH) value was 0.84 ng/mL (RV: 0.07–7.35 ng/mL), not compatible with PCOS.
Investigation
We suspected laboratory interference in the testosterone test, prompting a reevaluation of both laboratory procedures and imaging of the ovarian mass.
Diagnostic assessment of testosterone levels
The results of total testosterone levels are shown in Table 1. Total testosterone concentration was highly elevated: 22.1 nmol/L as measured by a Centaur analyzer (Siemens). The levels of SHBG, FSH, and LH were within normal range. To rule out a possible interference in the testosterone test, serial dilutions were performed at ratios of 1:1, 1:2, 1:4, and 1:8 with the following results: 22.1, 26.1, 26.6, and 22.2 nmol/L, respectively, proving linearity. Linearity was defined as the percent of the observed result over the expected result ratio, in the range of 80–120% for a serial dilution. The sample was processed by two alternative methods, Cobas (Roche) and Architect (Abbott), showing similar testosterone levels, 19.5 and 23.0 nmol/L, respectively (Table 1). The same total testosterone levels were found after blocking heterophile and nonspecific antibodies.
Comparison of testosterone levels with alternative methods.
| Method | Manufacturer | Treatment | Testosterone, nmol/L | RR, nmol/L |
|---|---|---|---|---|
| Centaur (CMIA) | Siemens | Direct | 22.1 | 0.4-2.1 |
| Cobas (ECLIA) | Roche | Direct | 19.5 | 0.29-1.67 |
| Extraction* | 5.66 | |||
| Architect (CMIA) | Abbott | Direct | 23.0 | 0.48-1.85 |
*Extraction procedure for serum testosterone using diethyl ether.
CMIA, chemiluminescence microparticle immunoassay; ECLIA, electrochemiluminescence immunoassay; RR, reference range.
At this point, and according to Roche and our laboratory personal recommendations, an extraction procedure with diethyl ether was performed prior to the immunoassay. Testosterone was then measured on a Cobas/Roche analyzer (Table 1). Steroid extraction from the serum sample was done by adding 3 mL of diethyl ether to 200 µL of serum and vortexing for 4 min. The aqueous and ether layers were separated through freezing (−70ºC), which only freezes the aqueous layer. After ether evaporation, the dried extract was dissolved in 200 µL of PBS, and testosterone was then measured on the Cobas e801 autoanalyzer. Total testosterone levels after extraction were 5.6 nmol/L (Table 1). For quality control, the same extraction procedure was applied to a healthy male sample, with levels of testosterone before and after extraction being 9.8 nmol/L and 9.9 nmol/L, respectively, indicating full recovery. Elimination of the interferences by the extraction procedure demonstrated that the cause of the very high total testosterone was water-soluble fragments of the steroid hormone pathway, which reacted with testosterone antibodies in the direct assay. The total testosterone level after extraction procedure was 5.6 nmol/L, significantly lower than the initial level but still 2.5 times higher than a normal female’s upper limit.
Imaging reevaluation of the ovarian mass
A vaginal ultrasound revealed a left ovarian solid mass measuring 30 × 30 × 32 mm with normal blood flow on color Doppler (Fig. 1A). The mass’s visual appearance was uncharacteristic of a dermoid cyst, which is typically devoid of blood flow. The right ovary had a characteristic polycystic ovary appearance. An abdominal contrast cCT scan showed a 30 × 36 mm round solid mass with clear borders in the left ovary (Fig. 1B). The adrenal glands were normal; no lymphadenopathy or abdominal free fluid was found. The imaging features confirmed the ovarian tumor’s solid nature. We, therefore, referred our patient for a laparoscopic investigation of the ovarian tumor.
(A) Vaginal ultrasound revealed a 30 × 30 × 32 mm left ovarian solid mass with normal blood flow on color Doppler. (B) An abdominal contrast computed tomography scan (CT) showed a 30 mm × 36 mm round solid mass with clear borders in the left ovary. The density of the finding was 90 HU in the venous phase. (C) SCT of left ovary. Diffuse pattern of tumor growth with very scant stroma. (D) SCT of left ovary. The tumor composed of large polygonal cells with distinct cell border, abundant spongy to granular cytoplasm, and clear intracytoplasmic vacuoles. There is central nuclei with nucleoli and minimal nuclear atypia. No mitoses were seen.
Citation: Endocrinology, Diabetes & Metabolism Case Reports 2024, 2; 10.1530/EDM-23-0117
Treatment
Interventions
Upon laparoscopy, a left ovary solid mass of 30 mm in diameter and several left fallopian tube vascular nodules were found. The peritoneum, omentum, diaphragm, intestines, uterus, and right ovary were of normal appearance with no ascites. Peritoneal washings, resection of the left ovarian mass, and a biopsy of the fallopian tube nodules were performed.
Microscopic examinations of the tumor showed diffusely arranged cells with abundant pale lipid-rich focally eosinophilic cytoplasm and small to intermediate round nuclei with frequent nucleoli (Fig. 1C and D). No Reinke crystals, mitosis, necrosis, or atypia were seen. Inhibin was positive in the tumor cells. CD10, PAX7, HMB45, OCT3/4, AFP, EMA, chromogranin, CD15, CEA, CK7, and S100 immunostaining were negative. The fallopian tube lesion had a similar histological appearance suggestive of metastases. Peritoneal washings were negative for tumor cells. The findings were consistent with an ovarian steroid-cell tumor (SCT) not otherwise specified (NOS), with cystic degeneration.
Outcome and follow-up
Twenty-four hours post surgery, the total testosterone level returned to normal, 0.6 nmol/L. Estradiol levels increased to 1000 pmol/L, and the patient resumed menstruation 1 month after surgery. Since ovarian SCT NOS can become malignant, the patient was scheduled for staging laparotomy. Left oophorectomy and left fallopian tube resection were performed. The ovary showed follicular cysts, luteinized follicular cysts, and focus of endosalpingiosis. The fallopian tube was without significant changes.
Four months later, the patient became pregnant through IVF.
Discussion
We describe a case of a childbearing-age woman who was referred for investigation of secondary amenorrhea, infertility, and extremely high serum testosterone levels without virilization signs. She was an obese woman with diagnose of PCOS, which could explain amenorrhea and hyperandrogenism, but did not explain the very high testosterone levels and the absence of virilization. Therefore, a pitfall in the testosterone test was suspected. A decrease in total serum testosterone levels was detected after performing a diethyl ether extraction procedure prior to the immunoassay. However, testosterone levels did not return to female ranges. After a reevaluation of a known left ovarian mass, the patient underwent a surgical resection of an ovarian SCT, thereby returning to total testosterone normal female levels and resuming menstruation.
Levels of total testosterone higher than 3.5–7 nmol/L (>2–3 SD) in women are most often associated with hyperandrogenic signs, menstrual irregularities, and a rapid virilization onset (1, 2). However, except for amenorrhea and consequent infertility, these symptoms and physical signs were not present in our patient.
Pitfalls in endocrine testing occasionally occur, and any discrepancy between laboratory tests and the clinical scenario warrants a complete revision of the clinical case. Mass spectrometry is the gold standard for measuring steroids; however, it is not routinely used in medical laboratories but mainly in research institutes (3). In recent years, immunological methods for steroid hormone measurement underwent improvements and upgrades and are currently used in endocrine practice (4). High specificity and sensitivity techniques are required to avoid false laboratory results. Immunoassays using specific monoclonal antibodies after extractions can meet these demands (4). As Pugeat et al. addressed (2), the vast majority of laboratories performing total testosterone routinely use various immunoassays without prior extraction, and according to the manufacturers’ testosterone inserts, only Roche suggests that laboratories use extraction for exploring implausible elevated testosterone values in women. Other leading manufacturers like Abbott, Siemens, and Beckman Coulter did not mention this option in their ‘limitation of procedure.’ Our case demonstrates the importance of the extraction method for suspected falsely elevated testosterone in women. It is evident that testosterone results of around 20 nmol/L for women without virilization signs need further investigation. The extraction method succeeded in verifying the presence of hydrophilic fragments as the cause for these seemingly contradictory results.
Although the extremely high testosterone levels did not produce virilization, they did suppress ovulation and menstruation. This paradoxical response implies a selective peripheral tissue response to testosterone. Testosterone negatively feeds back on FSH and LH production by negatively modulating GnRH production in the hypothalamus (1, 5) by decreasing the GnRH pulse frequency (6). This suppresses ovulation, resulting in amenorrhea. Tumor resection resulted in a rapid return of normal testosterone levels followed by resumed menstruation. Differences in testosterone’s target-tissue effects in our patient could be explained by different tumor-produced testosterone molecule conformations. The ultrasonographic appearance of the right ovary was typical for PCO. Women with PCOS have elevated AMH levels (>3 ng/mL) due to the high number of ovarian follicles in the menstrual cycle’s early stage. The AMH levels were low in our patient, which can be explained by high testosterone level suppression, as in males, hormonal androgenic treatment using testosterone decreases AMH levels (7).
Sex cord-stromal tumors are ovarian tumors originating from primitive stromal cells (8). They are composed of cells that produce steroid hormones and can be associated with hormone-mediated signs and symptoms (8). Steroid cell tumors (SCT) encompass a rare subgroup of sex cord tumors, which account for <0.1% of all ovarian tumors. The NOS type is the most common of the three subtypes described (stromal luteoma, Leydig cell tumor, and SCT NOS type) (9, 10). SCT NOS type can be found in a wide age range; however, in the overage population, the age of presentation is 43 years old. They are often unilateral and quite large tumors, up to 4.5–8.4 cm. Over 50% of the patients present with androgenic changes up to virilization that could be of many years duration. NOS SCTs are clinically malignant in 25–43% of cases (10), This case is unique because the patient first came for investigation for secondary amenorrhea that appeared immediately after COPC cessation, which is not uncommon, but continued for 3 years until the patient was referred for investigation and despite her very high testosterone level, there were no signs of virilization.
In summary, our case demonstrates that in some instances, ovarian tumor-produced testosterone can have selective influences on peripheral tissues causing only menstrual irregularities without virilization. Since these tumors are potentially malignant, a differential diagnosis of ovarian mass with high levels of testosterone or hydro-soluble fragments of testosterone is essential even without signs of virilization.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
Patient consent
Written informed consent for publication of their clinical details and clinical images was obtained from the patient.
Author contribution statement
VS – conceptualization and design, investigation, methodology, acquisition, analysis and interpretation of data, writing original draft, editing. MU – conceptualization, investigation, methodology, writing. RH – methodology, visualization, investigation. SL – conceptualization, investigation, methodology, writing, review. IH – visualization, investigation. ABA – visualization, investigation. OS – visualization, investigation. TS – visualization, investigation. AK – visualization, investigation. TZ – conceptualization and design, investigation, methodology, analysis and interpretation of data, writing, review and editing.
References
- 1↑
Sharma A, & Welt CK. Practical approach to hyperandrogenism in women. Medical Clinics of North America 2021 105 1099–1116. (https://doi.org/10.1016/j.mcna.2021.06.008)
- 2↑
Pugeat M, Plotton I, de la Perrière AB, Raverot G, Déchaud H, & Raverot V. Hyperandrogenic states in women: pitfalls in laboratory diagnosis. European Journal of Endocrinology 2018 178 R141–R154. (https://doi.org/10.1530/EJE-17-0776)
- 3↑
Handelsman DJ, & Wartofsky L. Requirement for mass spectrometry sex steroid assays in the journal of clinical endocrinology and metabolism. Journal of Clinical Endocrinology and Metabolism 2013 98 3971–3973. (https://doi.org/10.1210/jc.2013-3375)
- 4↑
Haddad RA , Giacherio D, & Barkan AL. Interpretation of common endocrine laboratory tests: technical pitfalls, their mechanisms and practical considerations. Clinical Diabetes and Endocrinology 2019 5 12. (https://doi.org/10.1186/s40842-019-0086-7)
- 5↑
de Kretser DM, Burger HG, & Bremner WJ. Control of FSH and LH secretion. Monographs on Endocrinology 1983 25 12–43. (https://doi.org/10.1007/978-3-642-81912-4_2)
- 6↑
Das N, & Kumar TR. Molecular regulation of follicle-stimulating hormone synthesis, secretion and action. Journal of Molecular Endocrinology 2018 60 R131–R155. (https://doi.org/10.1530/JME-17-0308)
- 7↑
Caanen MR, Soleman RS, Kuijper EA, Kreukels BP, De Roo C, Tilleman K, De Sutter P, van Trotsenburg MA, Broekmans FJ, & Lambalk CB. AntiMüllerian hormone levels decrease in female-to-male transsexuals using testosterone as cross-sex therapy. Fertility and Sterility 2015 103 1340–1345. (https://doi.org/10.1016/j.fertnstert.2015.02.003)
- 8↑
Horta M, & Cunha TM. Sex cord-stromal tumors of the ovary: a comprehensive review and update for radiologists. Diagnostic and Interventional Radiology 2015 21 277–286. (https://doi.org/10.5152/dir.2015.34414)
- 9↑
Kurman RJ, Carcangiu ML, Herrington CS, & Young RH. WHO classification of tumours of female reproductive organs. In World Health Organization Classification of Tumours, vol. 6, pp. 4456. Lyon: IARC, 2014.
- 10↑
Young RH, & Scully RE. Sex cord-stromal, steroid cell and other ovarian tumors with endocrine, paraendocrine and paraneoplastic manifestations. In: Blaustein's Pathology of the Female Genital Tract. Ed: RJ Kurman. Springer, New York, NY, USA. 2019. (https://doi.org/10.1007/978-1-4757-1942-0_19)
