Abstract
Background
Ovarian sex cord-stromal tumors (OSCST) include juvenile granulosa cell tumors (JGCT), Sertoli-Leydig cell tumor (SLCT) and gynandroblastoma (GAB) among others. These ovarian sex cord-stromal tumors as well as other tumors including pleuropulmonary blastoma (PPB) may be associated with DICER1 mutations. We sought to describe the clinical and genetic findings from the first 107 individuals enrolled in the International Ovarian and Testicular Stromal Tumor Registry.
Methods
Medical and family history were obtained for individuals consecutively enrolled in the International Ovarian and Testicular Stromal Tumor Registry. Pathology was centrally reviewed. DICER1 sequencing was performed on blood and tumor tissue.
Results
Of the 107 participants, 49 had SLCT, 25 had JGCT and 5 had GAB. Nearly all (36/37) SLCTs and 4/4 GAB tested had a DICER1 mutation in an RNase IIIb domain hotspot; approximately half of these individuals had a predisposing germline DICER1 mutation. Metachronous SLCTs were seen in 3 individuals with germline DICER1 mutations. Other DICER1-associated conditions were seen in 19% of patients with SLCT or GAB. Three children of women with SLCT were diagnosed with PPB based on genetic testing and clinical screening during the course of this study. All were diagnosed with PPB in its earliest and most curable form (Type I), were treated with surgery alone, and are alive without evidence of disease.
Conclusions
Recognition of the distinct genetic basis for a group of these tumors improves precise classification in difficult cases and promotes mutation-based screening and early detection.
Keywords: DICER1, Sertoli-Leydig cell tumor, gynandroblastoma, sex cord-stromal tumor, ovary
BACKGROUND
Ovarian sex cord-stromal tumors (OSCSTs) account for approximately 10% of all primary ovarian neoplasms during childhood and adolescence.1 Certain morphologic types of OSCSTs, especially juvenile granulosa cell tumors (JGCTs) and Sertoli-Leydig cell tumors (SLCTs), present primarily in the first two decades of life.2 In the original large series on JGCTs, over 40% of tumors were diagnosed in girls 10 years of age or less; some were seen in infants.3 By contrast, SLCTs tend to occur in adolescents and young adult women.4 Gynandroblastomas (GAB), now classified as sex cord stromal tumors of mixed forms, are composed of both JGCT and SLCT morphologic components and may be diagnosed at any age.5 Adult granulosa cell tumors are most common in older women but may rarely occur in adolescents and are characterized by somatic FOXL2 mutations.6
The International Ovarian and Testicular Stromal Tumor (OTST) Registry was established in December 2011 to develop a more complete understanding of the clinicopathologic features and genetic basis of this heterogeneous and understudied group of neoplasms. An association between SLCT and other conditions such as thyroid nodules or embryonal rhabdomyosarcoma has been recognized in the literature since the early 1970s.4,7 The International Pleuropulmonary Blastoma (PPB) Registry has enrolled children with PPB since its inception in 1988 and noted several probands and relatives with SLCT.8,9 The linkage between DICER1 and familial PPB was first reported in 2009, and since that time, many additional studies have documented the association between SLCT and DICER1 mutations.8,10–16
DICER1 encodes an RNaseIII endonuclease which cleaves precursor microRNAs into active miRNA. Mutations in DICER1 cause aberrant cleavage of mature 5p miRNAs resulting in altered expression of mRNAs with an accompanying risk for various types of neoplasms.13,17,18,19 Individuals with germline mutations in DICER1 are also at increased risk for several benign and malignant tumors including PPB, cystic nephroma and renal sarcoma, Wilms’ tumor, nodular thyroid hyperplasia and thyroid cancer, pineoblastoma and pituitary blastoma.20–24,25–30
PPB is the most lethal manifestation of the DICER1 tumor predisposition syndrome and is primarily seen in infants and young children.9 The latter tumor progresses from a multiloculated cyst (Type I) to a mixed cystic and solid (Type II) and solid (Type III) high grade multi-patterned primitive sarcoma which fills the hemithorax.31 With pathologic progression, the survival rate diminishes from 91% in Type I to 74% in Type II to only 53% in Type III.20 Testing and imaging surveillance of family members with DICER1-related disorders may allow detection of PPB in its earliest and most curable form. Likewise, most OSCST may be treated with surgery alone when found as International Federation of Gynecological Oncology (FIGO) stage Ia [T1aN0M0], thus also highlighting the importance of early detection.2
The aims of this study are to characterize the clinical and genetic characteristics of sequentially enrolled individuals with SLCT, JGCT and GAB. We determined the frequency of the DICER1 mutations, and evaluated the impact of predisposing mutations on clinical presentation, outcome, familial surveillance and directed intervention.
METHODS
Study subjects
Individuals in this report were enrolled in the International OTST Registry from December 2011 to March 2016. This study was approved by the Institutional Review Board at Children’s Minnesota and Children’s National Medical Center. Written informed consent was provided by the patient if 18 years of age or older or by the parent or guardian of each child under 18 years of age. Eligible diagnoses include any OSCST diagnosed at any age with the exception of adult granulosa cell tumors, which were eligible only if diagnosed before age 31, or if co-occurring with a personal or family history of DICER1-related conditions. Participants interested in receiving results of germline DICER1 testing underwent genetic counseling. Individuals also completed family history questionnaires. Pedigrees were reviewed when available. Age at diagnosis/recurrence was classified as the age of the patient at diagnostic surgery/surgery for recurrence. Follow-up data was requested annually. Medical records including operative and pathology reports and treatment data were collected. All available pathology material was centrally reviewed (LPD, DAH, RHY) and classified according to the World Health Organization (WHO) classification.4 Central pathology review was separated from genetic testing results. If tissue was not available for central pathology review, the original diagnosis was accepted. Tumors were staged according to the International Federation of Gynecological Oncology (FIGO)/TNM system for ovarian cancer.1 Patients with neoplasms originally classified as OSCST but found to be other entities (e.g. germ cell tumors) on central review were excluded from this analysis.
Molecular analyses
DICER1 gene sequencing was performed on blood and/or saliva and tumor tissue using either Sanger sequencing or a next generation sequencing assay designed to detect base substitutions and small insertions/deletions in both coding and intron-exon junction flanking regions.9,18,25 DICER1 deletion testing was performed on germline DNA on a subset of individuals. DICER1 mutations identified in blood or saliva at 38–62% variant allele frequency were considered germline. Mutations that were present in multiple tissue types/sites but at a lower variant allele frequency than typical heterozygous were classified as mosaic. Both mosaic and germline mutations were grouped as “predisposing” for statistical analysis with the assumption they were present during embryogenesis. A more detailed description of the methods is provided in Supplemental Methods.
Statistical Analyses
The Kaplan-Meier test was used to examine both time to death and time to recurrence of primary OSCST.32 Time to additional events such as additional DICER1-related conditions (including metachronous ovarian tumors) was calculated separately. Tests of equality of survival distributions were calculated using Breslow and Tarone-Ware. Mood’s Median Test was used to measure differences in median age, stage and differentiation.33 A Chi square was used to test for differences in the age distribution for individuals with and without predisposing DICER1 mutations. Analyses were completed using SPSS V23.
RESULTS
Between December 2011 and March 2016, 107 individuals with OSCST had enrolled (Table 1). Pathologic materials from 92% (98/107) of tumors were centrally reviewed. For nine cases (5 JGCT and 4 SLCT) in which no pathologic material was available, the local pathology diagnosis was used. Central review pathologists concurred with the local diagnosis in 80 of 98 cases (82%) (Supplemental Table 1). Thirty nine of 49 (82%) individuals with stage Ia disease were treated with surgery alone; however, three of four individuals (75%) with poorly differentiated stage Ia SLCT also received adjuvant chemotherapy. Chemotherapy was generally given to individuals with JGCT or SLCT if greater than stage Ia with the exception of intraoperatively ruptured JGCT which was generally treated with surgery alone (58.3%, n=14/24). The most common chemotherapy regimen was cisplatin, etoposide and bleomycin (PEB or BEP; SLCT=12, JGCT=4, sex cord stromal tumor with annual tubules (SCTAT)=1 and sex cord stromal tumor (SCST) NOS= 2). Two patients with SLCT received only cisplatin and etoposide. One patient with JGCT and one patient with SLCT received carboplatin, etoposide and bleomycin. Additional individuals received carboplatin and paclitaxel (SLCT=3 and Sertoli cell=1). Four additional individuals (SLCT=2, GAB=1 and AGCT=1) received chemotherapy but the regimen details were unavailable. SLCT, GAB, and JGCT accounted for 79 of 107 cases (74%) Since the focus of this report is DICER1, individuals with SLCT, GAB, and JGCT are the focus of further detailed analyses.
Table 1.
Demographics and outcome for all enrolled individuals (n=107) with ovarian sex cord- stromal tumors.
| N=107 | SLCT | JGCT | GAB | Sex Cord tumor with annular tubules | Sertoli cell tumor | Steroid cell tumor | AGCT | Sclerosing stromal tumor | Sex Cord tumor, NOS |
|---|---|---|---|---|---|---|---|---|---|
| Patients (n) | 49 (46% ) | 25 (23% ) | 5 (5%) | 5 (5%) | 5 (5%) | 2 (2%) | 10 (9%) | 4 (4%) | 2 (2%) |
| Caucasian (%) | 37 (76% ) | 10 (40% ) | 4 (80%) | 4 (80%) | 5 (100% ) | 0 | 6 (60% ) | 2 (50%) | 1 (50% ) |
| Age @ Diagnosis, Median (years), range | 17 (2– 61) | 9 (<1– 28) | 16 (14– 32) | 12 (3– 50) | 44 (33– 52) | 38 (17– 58) | 37 (13– 49) | 14 (10– 22) | 12 (10– 15) |
| Median Time from Diagnosis (months), range | 32 (0– 624) | 15 (0– 415) | 11 (3– 117) | 23 (0– 47) | 33 (18– 68) | 11 (1–– 20) | 46 (3– 203) | 35 (1–76) | 7.5 (1– 14) |
| Status | |||||||||
| NED | 40 (82%) | 21 (84%) | 5 (100%) | 5 (100%) | 5 (100%) | 2 (100%) | 8 (80%) | 4 (100%) | 1 (50%) |
| In TRT | 2 (4%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 (50%) |
| In Recur TRT | 2 (4%) | 1 (4%) | 0 | 0 | 0 | 0 | 2 (20%) | 0 | 0 |
| In Additional TRT | 1 (2%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Death | 4 (8%) | 3 (12%) | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Current Age, Median (years), range | 26 (4–68) | 14 (2–54) | 23 (16–35) | 20 (7– 53) | 48 (36–57) | 40 (18–62) | 40 (17–61) | 18 (13–30) | 15 (11–19) |
| Germline DICER1 mutation (n=89) | 25/41 | 0/19 | 3/5 | 0/5 | 0/5 | 0/1 | 0/9 | 0/3 | 0/1 |
| Tumor DICER1 mutation (n=48) | 36/37 | 0/2 | 4/4 | 0/0 | 0/1 | 0/1 | 0/1 | 0/0 | 0/0 |
SLCT= Sertoli-Leydig cell tumor, JGCT= juvenile granulosa cell tumor, GAB= gynandroblastoma, AGCT= adult granulosa cell tumor, NOS= not otherwise specified. NED= no evidence of disease, TRT = treatment, Recur= recurrence
Sertoli-Leydig cell tumors
Almost 50% of enrolled individuals had SLCT (n=49; 46%) with a median age at diagnosis of 17 years (range, 2–61 years) (Table 1). Hormonal symptoms were common (Supplemental Table 2). The majority of SLCTs were intermediately or poorly differentiated (Table 2). A retiform pattern was seen in 8/45 centrally reviewed tumors. Nearly half had heterologous histologic elements defined as non-Sertoli-Leydig components, most often mucinous glandular epithelium. Ten tumors had sarcomatous features. A representative example is shown in Supplemental Fig. 1E. About 50% of individuals had stage Ia. All individuals with stage Ia SLCT are presently free of tumor after a median follow-up of 19 months (range, 0–359 months). Overall, 8/49 (16.3%) individuals with SLCT had a recurrence and 4 of these with recurrent disease (50%) died of tumor progression at a median follow-up interval of 35.5 months (range, 8–74 months).
Table 2.
Stage, level of differentiation, treatment and outcome for 49 individuals with SLCT.
| Stage | Differentiation | Adjuvant Chemo | Surgery Only | NED | Metachronous | Relapse | DOD |
|---|---|---|---|---|---|---|---|
|
| |||||||
| IA | Well (n=2) | 0 | 2 | 2 | 0 | 0 | 0 |
| [T1aN0M0] (n=27) | Intermediate (n=20) | 3 | 17 | 20 | 2 | 0 | 0 |
| Poor (n=4) | 3 | 1 | 4 | 1 | 0 | 0 | |
| Unknown (n=1) | 1 | 0 | 1 | 0 | 0 | 0 | |
|
| |||||||
| IC [T1cN0M0] (n=6) | Intermediate (n=1) | 0 | 1 | 0 | 0 | 1 | 1 |
| Poor (n=5) | 2 | 3 | 2 | 0 | 3 | 1 | |
|
| |||||||
| IC1 [T1C1N0M0] (n=6) | Intermediate (n=2) | 2 | 0 | 2 | 0 | 0 | 0 |
| Poor (n=3) | 1 | 2 | 2 | 0 | 1 | 1 | |
| Unknown (n=1) | 1 | 0 | 1 | 0 | 0 | 0 | |
|
| |||||||
| >IC [T1cN0M0] (n=4) | Intermediate (n=2) | 2 | 0 | 2 | 0 | 0 | 0 |
| Poor (n=2) | 2 | 0 | 0 | 0 | 1 | 0 | |
|
| |||||||
| Unknown (n=6) | Well (n=1) | 0 | 1 | 1 | 0 | 0 | 0 |
| Intermediate (n=1) | 0 | 1 | 1 | 0 | 0 | 0 | |
| Unknown (n=4) | 1 | 3 | 2 | 0 | 2 | 1 | |
|
| |||||||
| All stages (n=49) | Well (n=3) | 0 | 3 | 3 | 0 | 0 | 0 |
| Intermediate (n=26) | 7 | 19 | 25 | 2 | 1 | 1 | |
| Poor (n=14) | 8 | 6 | 8 | 1 | 5 | 2 | |
| Unknown (n=6) | 3 | 3 | 4 | 0 | 2 | 1 | |
FIGO 1a: limited to one ovary, capsule intact, negative washings; 1c1: surgical spill. SLCT= Sertoli-Leydig cell tumor, NED= no evidence of disease, DOD= died of disease, Chemo= chemotherapy, NOS= not otherwise specified
DICER1 RNase IIIb hotspot mutations were identified in 36/37 (97%) Sertoli-Leydig cell tumors sequenced (Supplemental Table 3). In these 36 patients with DICER1-mutated SLCTs, 22 individuals had heterozygous germline loss of function (LOF) mutations; and 3 individuals were mosaic for a DICER1 LOF or RNase IIIb mutation. Eleven individuals had DICER1 mutations limited to the tumor (Supplemental Table 3). Only one individual with intermediately differentiated SLCT lacked DICER1 mutations after complete tumor and germline testing; pathologic review did not reveal anything unique about that tumor (Supplemental Figure 1). Of the three patients with well differentiated SLCT histology, two of three were not tested for DICER1 and the third was negative for a germline DICER1 mutation.
Gynandroblastoma
Five patients with GAB were enrolled with a median age at diagnosis of 16 years (range, 14–32 years). The SLCT component of each evaluable GAB showed intermediate differentiation. Most were stage Ia (n=4). All five individuals are alive without evidence of disease (Table 3).
Table 3.
Stage, treatment and outcome characteristics of five individuals with gynandroblastoma.
| Stage | Differentiation | Adjuvant Chemo | Surgery Only | NED | Relapse | DOD |
|---|---|---|---|---|---|---|
| IA [T1aN0M0] (n=4) | Intermediate | 1 | 3 | 4 | 0 | 0 |
| IIC [T2cN0M0] (n=1) | Intermediate | 1 | 0 | 1 | 0 | 0 |
| All stages (n=5) | Intermediate | 2 | 3 | 5 | 0 | 0 |
FIGO 1a: limited to one ovary, capsule intact, negative washings; IIb Chemo= chemotherapy, NED= no evidence of disease, DOD =died of disease
Four GABs had tumor tissue available for testing and each contained a DICER1 RNase IIIb hotspot mutation. Three of these individuals had germline LOF mutations. The fourth patient, whose tumor had an RNase IIIb mutation with an allele frequency of 62%, did not have any detectable germline DICER1 mutations by sequencing or duplication/deletion analysis.
Juvenile granulosa cell tumors
Twenty-five individuals with JGCT were enrolled. The median age at presentation for JGCT was 9 years (range, 1 month-28 years). Most had hormonal symptoms. Seven individuals had stage Ia; another tumor was stage Ib [T1bN0M0]. All the stage Ia/Ib patients were treated with surgery alone, and are free of disease after a median follow-up of 9 months (range, 1–77 months). Of 11 patients with tumor stage Ic [T1cN0M0] or greater, 3 (27%) died of disease at a median follow-up time of 17 months (range, 10–44 months) (Table 4). Germline DNA was analyzed in 19 individuals with JGCT; none was positive for a germline DICER1 mutation. For two individuals with JGCT, tumor tissue was sequenced and no DICER1 mutation was found.
Table 4.
Stage, treatment and outcome characteristics of 25 individuals with JGCT.
| Stage | Adjuvant Chemo | Surgery Only | NED | Relapse | DOD |
|---|---|---|---|---|---|
| IA [T1aN0M0] (n=7) | 0 | 7 | 7 | 0 | 0 |
| IB [T1bN0M0] (n=1) | 0 | 1 | 1 | 0 | 0 |
| IC [T1CN0M0] (n=4) | 1 | 3 | 2 | 2 | 1 |
| IC2[T1C2N0M0] (n=4) | 2 | 2 | 4 | 0 | 0 |
| >Ic [T1CN0M0] (n=3) | 2 | 1 | 1 | 2 | 2 |
| Unknown (n=6) | 0 | 6 | 6 | 1 | 0 |
| All Stages (n=25) | 5 | 20 | 21 | 5 | 3 |
Abbreviations: JGCT= juvenile granulosa cell tumor, NED= no evidence of disease, DOD= died of disease, Chemo= chemotherapy
DICER1 mutation analysis in other tumor types
Germline DICER1 sequencing was performed for 24 individuals with other types of OSCST (Table 1). No DICER1 mutations were identified in groups other than SLCT and GAB. One individual with a DICER1 mutation had a history of undifferentiated ovarian sarcoma and later developed SLCT.
Prognostic Analyses of SLCT and GAB groups
Individuals with predisposing (germline or mosaic) DICER1 mutations presented at younger median age than those with tumor-limited mutations (16 years, range 4–61 years) vs. (21 years, range 2–61 years) (p =0.151) (Tables 5 & 6). Eighty-two percent vs. 55% of patients with predisposing mutations were younger than 21 years of age at diagnosis (p=0.022). Overall and recurrence free survival in patients with SLCT and GAB was significantly better when disease was limited to the ovary (stage Ia). Poor differentiation was a negative prognostic factor for recurrence free survival (Table 6; Supplemental Fig. 3). Overall and recurrence-free survival was also significantly better for patients with predisposing DICER1 mutations compared to patients with tumor-limited DICER1 mutations (Supplemental Figures 2 and 3).
Table 5.
Clinical and pathologic characteristics by predisposing DICER1 mutation category
| Germline Loss of Function | Mosaic | Tumor-limited mutations | Germline & Tumor Negative | Incompletely Tested* | ||
|---|---|---|---|---|---|---|
| Loss of function | RNase IIIb hotspot | |||||
| # of Patients (n=54) | 25 | 2 | 1 | 12 | 1 | 13 |
| Age @Diagnosis (year), Median (range) | 16 (4–61) | 21 (4–27) | 5 | 21 (2–61) | 37 | 23 (4–54) |
| Diagnosis | ||||||
| SLCT (n=49) | 22# | 2 | 1 | 11 | 1 | 12 |
| GAB (n=5) | 3 | 0 | 0 | 1 | 0 | 1 |
| Additional disease foci | ||||||
| PPB | 2 | 0 | 1 | 0 | 0 | 0 |
| Cystic Nephroma | 2 | 0 | 1 | 0 | 0 | 0 |
| Thyroid Nodules | 14 | 2 | 1 | 0 | 0 | 1 |
| Thyroid Cancer | 4 | 2 | 0 | 0 | 0 | 0 |
| ERMS | 1 | 0 | 0 | 0 | 0 | 0 |
| NCMH | 1 | 0 | 0 | 0 | 0 | 0 |
| Intestinal polyps | 1 | 0 | 0 | 0 | 0 | 0 |
| Metachronous SLCT | 3 | 0 | 0 | 0 | 0 | 0 |
| Outcome (months) Diagnosis to Follow up | ||||||
| # Patients NED | 24 | 2 | 0 | 9 | 1 | 9 |
| # Patients AWD | 0 | 0 | 1 | 3 | 0 | 2 |
| # Patients Recur | 2 | 0 | 0 | 5 | 0 | 1 |
| # Patients alive (Median, months) | 24 (61) | 2 (22) | 1 (82) | 10 (12) | 1 (2) | 12 (12) |
| # Patients deceased (Median, months) | 1& (57) | 0 | 0 | 2 (14) | 0 | 1 (8) |
= 1 patient also developed an undifferentiated ovarian sarcoma, NOS
= DICER1 mutation by inference, patient not directly tested but patient’s sister tested positive for DICER1 following patient’s death
= 6 germline negative, tumor not tested
Abbreviations: SLCT= Sertoli-Leydig cell tumor, GAB= gynandroblastoma, PPB= pleuropulmonary blastoma, NED= no evidence of disease, AWD= alive with disease, Recur= recurrence, ERMS= embryonal rhabdomyosarcoma, NCMH= nasal chondromesenchymal hamartoma
Table 6.
Sertoli-Leydig cell tumor and Gynandroblastoma: Prognostic factors
| DICER1 | Stage | Differentiation | ||||
|---|---|---|---|---|---|---|
|
| ||||||
| Predisposing mutation | Tumor- specific mutation | Ia/T1aN0M0 | >Ia /T1a- N0-M0 | Poorly | Well & Intermediate | |
|
| ||||||
| Overall Survival | ||||||
| % | 96.4% | 81.8% | 100% | 82.4% | 85.7% | 97.1% |
|
|
||||||
| p-value | 0.020* | 0.038* | 0.262 | |||
|
| ||||||
|
Recurrence Free Survival % |
92.9% | 72.7% | 100% | 70.6% | 71.4% | 97.1% |
|
|
||||||
| p-value | 0.050* | 0.007* | 0.039* | |||
|
| ||||||
| DICER1 (predisposing vs. tumor-specific) | ---------------- | 55% | 76% | 75% | 55% | |
|
|
||||||
| p- value | 0.252 | 0.263 | ||||
|
| ||||||
|
Age
Median, years (Range) |
16 (4–61) | 21 (2–61) | 16 (2–40) | 16 (4–61) | 19 (4–61) | 16 (2–40) |
|
|
||||||
| p-value | 0.151 | 0.860 | 0.490 | |||
Other DICER1 syndrome conditions in women with SLCT and GAB
Additional DICER1 related conditions were frequently seen in our study cohort of SLCT and GAB (Table 5). Three individuals with predisposing DICER1 mutations developed metachronous SLCT within a range of 5 to 14 years from original SLCT diagnosis. All three metachronous tumors were stage Ia and treated with surgical resection only. Eighteen women (33%) with SLCT or GAB self-reported thyroid nodules and 6 individuals (11%) reported well-differentiated thyroid carcinoma. Thyroid nodules and thyroid carcinoma were not reported in individuals without predisposing DICER1 mutations in this cohort.
Impact on family surveillance and early diagnosis of PPB
Three infants of women with SLCT were diagnosed with PPB based on genetic testing and clinical screening during the course of this study. These infants were diagnosed with Type I PPB at ages 3, 4 and 7 months of age, were treated with surgery alone and are alive without evidence of disease at a follow-up of 33, 21 and 7 months.
DISCUSSION
This study reports the clinical outcomes, management, and DICER1 sequencing results in a group of centrally reviewed OSCST. We found that SLCT and GAB are nearly always DICER1-related tumors. This study confirms and extends previous reports showing DICER1 mutations in association with SLCT. Slade et al. showed germline DICER1 mutations in 4/6 SLCTs. Witkowski et al. found somatic mutations in the RNase IIIb domain in 8/15 SLCTs and Heravi-Moussavi et al reported somatic DICER1 mutations in the RNase IIIb domain of 29% of nonepithelial ovarian tumors including 26 of 43 SLCTs. This study includes both germline and tumor testing along with central review, the latter of which removed some non-SLCT tumors from the study population (see Supplemental Table 1). The combination of these factors may be responsible for a stronger association between SLCT and DICER1 than was noted in previous reports.11,15,34–36 None of the individuals with a centrally reviewed diagnosis of JGCT, steroid cell tumor, sex cord-stromal tumor with annual tubules or Sertoli cell tumor had germline DICER1 mutations, although one individual with a germline DICER1 mutation developed undifferentiated ovarian sarcoma and later SLCT.37
Based on the importance of identifying DICER1 pathogenic variants in an individual, as well as the moderate rate of discordance (18%) between local and central review (which is typical of rare tumors), we recommend careful and, when feasible, central histopathologic review for all individuals with OSCST tumors with DICER1 testing for all individuals with SLCT and GAB and consideration of DICER1 testing in individuals with other OSCST if the medical or family history suggests DICER1-tumor predisposition (e.g. thyroid nodules).
In our study, 97% of SLCT are DICER1-related which is significantly higher than other series. One potential reason for this difference is that we sequenced both germline and tumor DNA. The vast majority of DICER1 syndrome cancers including pleuropulmonary blastoma, SLCT and rhabdomyosarcoma have biallelic DICER1 mutations consisting of a loss of function mutation in one allele and a missense RNase IIIb domain mutation in the other allele. Our previous study of SLCT only had germline testing data available; in that data approximately half of individuals with SLCT had germline DICER1 mutations, however, tumor tissue was not sequenced. The current study combines germline and tumor testing for DICER1 and when that testing is combined, nearly all the SLCTs are linked to DICER1 mutations. Several SLCTs had biallelic DICER1 mutations limited to tumor tissue, which may represent mutations confined to the tumor or very low level mosaicism. A negative germline result does not indicate that the tumor is DICER1-unrelated.
Also, in this study, histologic classification included central review which removed some non-SLCT/GAB tumors from the SLCT/GAB cohorts and thus may have contributed to a higher concordance of pathologic to genetic findings. SLCT, GAB and JGCT are uncommon tumors with a broad range of morphologic appearances. Two individuals with DICER1 mutations were diagnosed with JGCT at the treating institution but after central review reclassified as GAB. For difficult to classify cases, DICER1 mutation testing of tumor tissue may be useful.
As in previous reports, the outcome for individuals with SLCT was strongly correlated with stage and level of differentiation.4,38 Individuals with predisposing DICER1 mutations had significantly better overall and recurrence free survival although additional DICER1-related conditions were frequently noted. Metachronous SLCTs were seen in 3 individuals with predisposing DICER1 mutations up to 14 years after initial diagnosis and associated with a favorable prognosis. Thus, the finding of a contralateral ovarian mass in an individual with DICER1 mutation and a previously diagnosed ovarian tumor cannot be assumed to be a recurrence. This consideration has importance for therapeutic planning as a metachronous tumor may be more sensitive to first-line chemotherapy than a true tumor recurrence. Individuals with germline mutations or mosaicism should be followed for the emergence of metachronous disease even after the highest risk period for recurrence has passed.
Identifying a germline DICER1 mutation or mosaicism in a young woman has considerable implications for the individual and her family.39 Post treatment surveillance regimens must consider the possibility of underlying tumor predisposition. In addition to the risk for contralateral ovarian tumors, individuals with germline DICER1 mutations are also at risk for other conditions including nodular thyroid hyperplasia or carcinoma.24,40 Identification of germline DICER1 mutations may be especially critical for young children in the family who may be at risk for PPB and other conditions. Early diagnosis improves morbidity and mortality in PPB and may also be relevant for minimizing complications of other DICER1-related tumors including cystic nephroma, Wilms tumor, ovarian tumors and certain childhood brain tumors.
In individuals with no detectable germline mutation, there may be a role for tumor specific testing for biallelic mutations in DICER1. Conventional treatment algorithms for SLCT empirically use platinum-based chemotherapy similar to the management of epithelial and germ cell tumors; however, a more directed treatment design may be beneficial, particularly in the case of tumors with sarcomatous elements. In contrast to most individuals with germ cell tumors, those with SLCT and GAB lack a reliable blood-based tumor marker. Imaging studies are limited by lack of sensitivity. The nearly universal presence of DICER1 RNase IIIb mutations could potentially be useful as circulating tumor DNA (ctDNA ) biomarkers for monitoring of response to therapy and post treatment surveillance.
Limitations
Tissue for central pathology review was not available for 9/107 individuals. In some cases, although slides were available, sufficient tissue for molecular analysis was not and thus only germline DNA results are available for some individuals. Although this represents the largest series to date, the rarity and heterogeneous clinical presentation of this tumor limits the power of multiple variable analyses.
CONCLUSIONS
Nearly all SLCTs and GAB are DICER1-related. No other centrally reviewed OSCST types were found to harbor germline or somatic DICER1 mutations. These results clearly demonstrate the distinct nature of the pathophysiology of SLCT and GAB when compared to germ cell tumors, epithelial ovarian tumors, and even other OSCST tumors. Recognition of SLCT and GAB as a unique and well-defined molecular group may pave the way for precise diagnosis, rational development of therapies and opportunities to implement mutation-based screening in patients and their young families.
Supplementary Material
Highlights.
DICER1 RNase IIIb mutations were identified in 36/37 SLCT’s and 4/4 GABs sequenced.
Germline or mosaic mutations were found in more than half of those with SLCT.
Predisposing DICER1 mutations were associated with higher recurrence free survival.
DICER1 testing in women with SLCT facilitated screening of their children for PPB.
Three children were diagnosed with PPB in its earliest and most curable form.
Acknowledgments
The authors wish to thank Mercedes Wilhelm, Ann Mason, Gretchen Williams, Ann Carr, Laura Harney, Mark Brenneman and Kristen Griffin as well as the families and treating providers who participate in the International OTST Registry. We gratefully acknowledge the Pine Tree Apple Tennis Classic players, donors and volunteers who have provided 31 years of continuous funding for PPB and DICER1-related research.
Source of Funding: This work is supported by National Institutes of Health grant NCI R01CA143167 and The Parson’s Foundation (DAH,YM). The International Ovarian and Testicular Stromal Tumor Registry is supported by St. Baldrick’s Foundation, Pine Tree Apple Tennis Classic Foundation, Hyundai Hope on Wheels and the Randy Shaver Cancer Research and Community Fund. DTS is supported by the German Childhood Cancer Foundation. Supported in part by the Intramural Research Program of the Divisions of Cancer Epidemiology and Genetics, National Cancer Institute.
Footnotes
Conflicts of InterestThe authors each state that he/she has no conflicts of interest to disclose.
References
- 1.Schneider DT, Terenziani M, Cecchetto G, Olson TA. Gonadal and extragonadal germ cell tumors, sex cord stromal tumors and rare gonadal tumors. In: Schneider DT, Brecht IB, Oslon TA, AF, editors. Rare tumors in children and adolescents. Heidelberg: Springer; 2012. pp. 327–402. [Google Scholar]
- 2.Schneider DT, Calaminus G, Wessalowski R, et al. Ovarian sex cord-stromal tumors in children and adolescents. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2003;21(12):2357–2363. doi: 10.1200/JCO.2003.05.038. [DOI] [PubMed] [Google Scholar]
- 3.Young RH, Dickersin GR, Scully RE. Juvenile granulosa cell tumor of the ovary. A clinicopathological analysis of 125 cases. Am J Surg Pathol. 1984;8:575–596. doi: 10.1097/00000478-198408000-00002. [DOI] [PubMed] [Google Scholar]
- 4.Young RH, Scully RE. Ovarian Sertoli-Leydig cell tumors. A clinicopathological analysis of 207 cases. Am J Surg Pathol. 1985;9(8):543–569. doi: 10.1097/00000478-198508000-00001. [DOI] [PubMed] [Google Scholar]
- 5.Young RH. Sex cord-stromal tumors of the ovary and testis: their similarities and differences with consideration of selected problems. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc. 2005;18(Suppl 2):S81–98. doi: 10.1038/modpathol.3800311. [DOI] [PubMed] [Google Scholar]
- 6.Shah SP, Kobel M, Senz J, et al. Mutation of FOXL2 in granulosa-cell tumors of the ovary. The New England journal of medicine. 2009;360(26):2719–2729. doi: 10.1056/NEJMoa0902542. [DOI] [PubMed] [Google Scholar]
- 7.McClean GE, Kurian S, Walter N, Kekre A, McCluggage WG. Cervical embryonal rhabdomyosarcoma and ovarian Sertoli-Leydig cell tumour: a more than coincidental association of two rare neoplasms? Journal of clinical pathology. 2007;60(3):326–328. doi: 10.1136/jcp.2006.039206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Schultz KA, Pacheco MC, Yang J, et al. Ovarian sex cord-stromal tumors, pleuropulmonary blastoma and DICER1 mutations: a report from the International Pleuropulmonary Blastoma Registry. Gynecol Oncol. 2011;122(2):246–250. doi: 10.1016/j.ygyno.2011.03.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hill DA, Ivanovich J, Priest JR, et al. DICER1 mutations in familial pleuropulmonary blastoma. Science. 2009;325(5943):965. doi: 10.1126/science.1174334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Oost EE, Charles A, Choong CS, et al. Ovarian sex cord-stromal tumors in patients with probable or confirmed germline DICER1 mutations. International journal of gynecological pathology : official journal of the International Society of Gynecological Pathologists. 2015;34(3):266–274. doi: 10.1097/PGP.0000000000000150. [DOI] [PubMed] [Google Scholar]
- 11.Heravi-Moussavi A, Anglesio MS, Cheng SW, et al. Recurrent somatic DICER1 mutations in nonepithelial ovarian cancers. The New England journal of medicine. 2012;366(3):234–242. doi: 10.1056/NEJMoa1102903. [DOI] [PubMed] [Google Scholar]
- 12.Witkowski L, Mattina J, Schonberger S, et al. DICER1 hotspot mutations in non-epithelial gonadal tumours. Br J Cancer. 2013;109(10):2744–2750. doi: 10.1038/bjc.2013.637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wang Y, Chen J, Yang W, et al. The oncogenic roles of DICER1 RNase IIIb domain mutations in ovarian Sertoli-Leydig cell tumors. Neoplasia. 2015;17(8):650–660. doi: 10.1016/j.neo.2015.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Rio Frio T, Bahubeshi A, Kanellopoulou C, et al. DICER1 mutations in familial multinodular goiter with and without ovarian Sertoli-Leydig cell tumors. JAMA. 2011;305(1):68–77. doi: 10.1001/jama.2010.1910. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang Y, Chen J, Yang W, et al. The Oncogenic Roles of DICER1 RNase IIIb Domain Mutations in Ovarian Sertoli-Leydig Cell Tumors. 2015;17(8):650–600. doi: 10.1016/j.neo.2015.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Witkowski L, McCluggage WG, Foulkes WD. Recently characterized molecular events in uncommon gynaecological neoplasms and their clinical importance. Histopathology. 2016;69(6):903–913. doi: 10.1111/his.13058. [DOI] [PubMed] [Google Scholar]
- 17.Pugh TJ, Yu W, Yang J, et al. Exome sequencing of pleuropulmonary blastoma reveals frequent biallelic loss of TP53 and two hits in DICER1 resulting in retention of 5p-derived miRNA hairpin loop sequences. Oncogene. 2014;33(45):5295–5302. doi: 10.1038/onc.2014.150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Brenneman M, Field A, Yang J, et al. Temporal order of RNase IIIb and loss-of-function mutations during development determines phenotype in DICER1 syndrome: a unique variant of the two-hit tumor suppression model. F1000Research. 2015;4:214. doi: 10.12688/f1000research.6746.1. [v1; ref status: approved with reservations 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Anglesio MS, Wang Y, Yang W, et al. Cancer-associated somatic DICER1 hotspot mutations cause defective miRNA processing and reverse-strand expression bias to predominantly mature 3p strands through loss of 5p strand cleavage. J Pathol. 2013;229(3):400–409. doi: 10.1002/path.4135. [DOI] [PubMed] [Google Scholar]
- 20.Messinger YH, Stewart DR, Priest JR, et al. Pleuropulmonary Blastoma (PPB): A Report on 350 Central Pathology Reviewed Confirmed Cases by the International PPB Registry. Cancer. 2015;121(2):276–285. doi: 10.1002/cncr.29032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Foulkes WD, Bahubeshi A, Hamel N, et al. Extending the Phenotypes Associated with DICER1 Mutations. Hum Mutat. 2011 doi: 10.1002/humu.21600. [DOI] [PubMed] [Google Scholar]
- 22.de Kock L, Foulkes WD. Sarcoma and germ-line DICER1 mutations. Lancet Oncol. 2016;17(11):e470. doi: 10.1016/S1470-2045(16)30522-8. [DOI] [PubMed] [Google Scholar]
- 23.de Kock L, Wang YC, Revil T, et al. High-sensitivity sequencing reveals multi-organ somatic mosaicism causing DICER1 syndrome. Journal of medical genetics. 2016;53(1):43–52. doi: 10.1136/jmedgenet-2015-103428. [DOI] [PubMed] [Google Scholar]
- 24.Hill DA, Doros L, Schultz KA, et al. DICER1-related disorders. In: Pagon RA, editor. GeneReviews [Internet] Seattle: University of Washington, Seattle; 2014. http://www.ncbi.nlm.nih.gov. [Google Scholar]
- 25.Doros LA, Rossi CT, Yang J, et al. DICER1 mutations in childhood cystic nephroma and its relationship to DICER1-renal sarcoma. Mod Pathol. 2014;27:1267–1280. doi: 10.1038/modpathol.2013.242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.de Kock L, Sabbaghian N, Druker H, et al. Germ-line and somatic DICER1 mutations in pineoblastoma. Acta Neuropathol. 2014;128:583–595. doi: 10.1007/s00401-014-1318-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.De Kock L, Sabbaghian N, Plourde F, et al. Pituitary blastoma: a pathognomonic feature of germ-line DICER1 mutations. Acta Neuropathol. 2014;128:111–122. doi: 10.1007/s00401-014-1285-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Foulkes WD, Priest JR, Duchaine TF. DICER1: mutations, microRNAs and mechanisms. Nat Rev Cancer. 2014;14(10):662–672. doi: 10.1038/nrc3802. [DOI] [PubMed] [Google Scholar]
- 29.Rakheja D, Chen KS, Liu Y, et al. Somatic mutations in DROSHA and DICER1 impair microRNA biogenesis through distinct mechanisms in Wilms tumours. Nat Commun. 2014;2:4802. doi: 10.1038/ncomms5802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wu MK, Sabbaghian N, Xu B, et al. Biallelic DICER1 mutations occur in Wilms tumours. J Pathol. 2013;230(2):154–164. doi: 10.1002/path.4196. [DOI] [PubMed] [Google Scholar]
- 31.Sabapathy DG, Guillerman RP, Orth RC, et al. Radiographic screening of infants and young children with genetic predisposition for rare malignancies: DICER1 mutations and pleuropulmonary blastoma. AJR Am J Roentgenol. 2015;204(4):W475–482. doi: 10.2214/AJR.14.12802. [DOI] [PubMed] [Google Scholar]
- 32.Kaplan EL, Meier P. Nonparametric Estimation from Incomplete Observations. Journal of the American Statistical Association. 1958;53(282):457–481. [Google Scholar]
- 33.Mood AM. On the Asymptotic Efficiency of Certain Nonparametric Two-Sample Tests. 1954:514–522. [Google Scholar]
- 34.Chen J, Wang Y, McMonechy MK, et al. Recurrent DICER1 hotspot mutations in endometrial tumours and their impact on microRNA biogenesis. J Pathol. 2015;237(2):215–225. doi: 10.1002/path.4569. [DOI] [PubMed] [Google Scholar]
- 35.Witkowski L, Mattina J, Schonberger S, et al. DICER1 hotspot mutations in non-epithelial gonadal tumours. Br J Cancer. 2013;109:2744–2750. doi: 10.1038/bjc.2013.637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Stewart CJ, Charles A, Foulkes WD. Gynecologic Manifestations of the DICER1 Syndrome. Surg Pathol Clin. 2016;9(2):227–241. doi: 10.1016/j.path.2016.01.002. [DOI] [PubMed] [Google Scholar]
- 37.Schultz KA, Harris A, Messinger Y, et al. Ovarian tumors related to intronic mutations in DICER1: a report from the international ovarian and testicular stromal tumor registry. Fam Cancer. 2016;15(1):105–110. doi: 10.1007/s10689-015-9831-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Schneider DT, Orbach D, Cecchetto G, et al. Ovarian Sertoli Leydig cell tumours in children and adolescents: an analysis of the European Cooperative Study Group on Pediatric Rare Tumors (EXPeRT) Eur J Cancer. 2015;51(4):543–550. doi: 10.1016/j.ejca.2014.11.013. [DOI] [PubMed] [Google Scholar]
- 39.Schultz KA, Harris A, Williams GM, et al. Judicious DICER1 testing and surveillance imaging facilitates early diagnosis and cure of pleuropulmonary blastoma. Pediatr Blood Cancer. 2014;61(9):1695–1697. doi: 10.1002/pbc.25092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Khan NE, Bauer AJ, Schultz KA, et al. Quantification of thyroid cancer and multinodular goiter risk in the DICER1 syndrome: a family-based cohort study. J Clin Endocrinol Metab. 2017 doi: 10.1210/jc.2016-2954. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.