Abstract
Objective:
Sertoli-Leydig cell tumors (SLCTs) are rare sex cord-stromal tumors, representing <0.5% of all ovarian tumors. We analyze the role of germline DICER1 status in outcomes of ovarian SLCT.
Methods:
Patients with SLCT were enrolled in the International Pleuropulmonary Blastoma/DICER1 Registry and/or the International Ovarian and Testicular Stromal Tumor Registry. Medical records were systematically abstracted, and those with known germline DICER1 status were selected for analysis.
Results:
Of 162 patients with SLCT, 60% had a germline DICER1 pathogenic or likely pathogenic (P/LP) variant. Adjusted 3-year recurrence-free survival (RFS) was 87.2% (95% CI: 79.4-95.8%) for patients with a germline DICER1 P/LP variant compared to 78.1% (95% CI: 66.4-91.9%) for those without a germline DICER1 P/LP variant (p = .043). Adjusted 3-year and 5-year overall (OS) was 93.9% (95% CI: 87.3-100.0%) for those with a germline DICER1 P/LP variant compared to 3-year OS of 91.3% (95% CI: 83.4-100.0%) and 5-year OS of 78.2% (95% CI: 63.8-95.9%) for those without a germline DICER1 P/LP variant (p = .021). Among patients with a germline DICER1 P/LP variant, the risk of a subsequent, non-recurrent event was 36.2% (95% CI: 21.4-48.1%) within 10 years. Prior/concurrent and subsequent neoplasms were rare among those without a germline DICER1 P/LP variant.
Conclusion:
This cohort study of patients with SLCT demonstrated that those with germline DICER1 P/LP variants had superior RFS and OS even when adjusting for other prognostic factors. Beyond prognostic implications of a germline DICER1 P/LP variant, germline testing helps identify patients at risk of subsequent neoplasms, including metachronous SLCT.
Keywords: Sertoli-Leydig Cell Tumor, Sex Cord-Stromal Tumor, DICER1, DICER1-Related Tumor Predisposition, Gynandroblastoma
INTRODUCTION
Ovarian Sertoli-Leydig cell tumors (SLCTs) are rare sex cord-stromal tumors comprising less than 0.5% of all ovarian tumors and primarily occur in adolescents and young adults.1
SLCTs are associated with variants in DICER1 and are a manifestation of DICER1-related tumor predisposition (MONDO: 0100216, OMIM: 606241). The International Pleuropulmonary Blastoma (PPB), established in 1987, noted several probands and relatives with SLCT and established the linkage between DICER1 and familial PPB in 2009.2 A parallel registry, the International Ovarian and Testicular Stromal Tumor (OTST) Registry, was established in 2011 to investigate further the clinicopathologic and genetic features of sex cord-stromal tumors, including SLCT. The Registries enroll patients worldwide and include direct review of records, standardized abstraction, and longitudinal follow-up.
In most moderately and poorly differentiated SLCTs, biallelic DICER1 loss-of-function and RNase IIIb hotspot variants are identified.3,4 Around 60% of these patients have a predisposing germline DICER1 loss-of-function variant.3 In 2018, the Registry first published surveillance guidelines that included screening for gynecologic neoplasms.5 Surveillance guidelines were updated in 2024 and recommend pelvic ultrasound every 6 months from detection of a germline DICER1 variant until at least age 40 for individuals with female reproductive organs.6,7
Most ovarian SLCTs present with low-stage disease with a favorable prognosis. However, specific risk factors, such as mesenchymal heterologous elements, are associated with a higher risk for recurrence.3,8,9 In a 2018 report and a subsequent follow-up with an expanded cohort in 2024, a germline DICER1 pathogenic or likely pathogenic (P/LP) variant had a protective effect on SLCT recurrence-free survival.3,4 One explanation could be that SLCT is detected at earlier stages in these patients due to surveillance. However, since these findings were present even before DICER1 surveillance guidelines were available, we hypothesized that the protective effect of a germline DICER1 P/LP variant is only partially explained by early detection. Herein, using the cohort from our 2024 analysis, we analyze the role of germline DICER1 status, focusing on associations with survival outcomes while controlling for other prognostic factors. Additionally, we systematically assess the risk of metachronous SLCT and other neoplasms stratified by germline DICER1 status.
METHODS
Enrollment
Individuals were enrolled in the International PPB/DICER1 Registry (ClinicalTrials.gov identifier: NCT03382158) and/or International OTST Registry (ClinicalTrials.gov identifier: NCT01970696). Written informed consent (including assent when applicable) was obtained. All study procedures were approved by the relevant human subjects’ committees.
Data Collection
Medical records were requested from treating institution(s) and systematically abstracted. Genetic and molecular data were obtained using previously described techniques or were abstracted from clinical records.10 Patients enrolled through the Registries with germline DICER1 testing available were selected for analysis and grouped based on the presence or absence of a germline DICER1 P/LP variant. Patients with known mosaicism for a loss-of-function variant or an RNase IIIb missense variant were excluded from the analysis. All variants were validated with VariantValidator (transcript NM_177438.3).11 Molecular, demographic, clinical, surgical, and histologic data variables were reported based on the number of evaluable patients (i.e., when patients had missing information, they were excluded from reporting within that variable). Follow-up was attempted annually.
Surgical Staging and Pathology
Surgical staging was abstracted using the International Federation of Gynecology and Obstetrics (FIGO) criteria; missing staging criteria were handled per prior analyses.3 Pathology was centrally reviewed when available by a Registry pathologist (JNS/RHY/LPD/DAH). Well-differentiated tumors were excluded as a recent study has suggested they are genetically distinct tumors and lack variants in DICER1.12
Statistical Analysis and Data Presentation
Associations with patient characteristics and germline DICER1 status were assessed using the χ2 test (or Fisher exact test) and the Mann-Whitney U test for categorical and numeric data. R (version 4.1.0) package trackViewer was used to create a lollipop plot to map DICER1 variants along the DICER1 gene.
Time-to-event (from diagnosis) outcomes included overall survival (OS) and SLCT recurrence-free survival (RFS); the former events included all-cause mortality, and the latter events were defined as recurrent SLCT or death. Metachronous SLCT and subsequent neoplasms were not classified as recurrent events. Metachronous SLCT was defined as intrinsic to the contralateral ovary and/or differing DICER1 RNase IIIb hotspots when hotspot testing was available. SLCT RFS and OS, stratified by DICER1 germline status, were estimated using Kaplan-Meier (KM) curves. Survival analyses were only performed among symptomatic patients to control for the impact of surveillance and incidental findings in earlier detection. The risk of subsequent malignancies and neoplasms was estimated using KM curves and stratified by DICER1 germline status.
Propensity score overlap weights were used to adjust for differences in prognostic factors between positive and negative DICER1 P/LP groups to ensure the groups were comparable.13 Overlap weights were used to estimate adjusted survival curves and test associations via log-rank test; both used robust standard errors to account for the weighting strategy.14 The propensity score modeled positive and negative germline DICER1 groups using logistic regression based on previously reported risk factors.3 The balance between comparison groups was assessed via standardized differences.15 Statistical analyses were performed using R Statistical Software (v4.3.0).16
RESULTS
In total, 162 participants with SLCT and known germline DICER1 status were enrolled in the Registries (Table 1).
Table 1:
Demographics, molecular, clinical, surgical, histology, treatment, and outcome data by germline DICER1 status.
| Variable | Positive Germline DICER1 P/LP Variant (n = 97) n/N (%) |
Negative Germline DICER1 P/LP Variant (n = 65) n/N (%) |
p-value | Total (n = 162) n/N (%) |
|---|---|---|---|---|
| Molecular Data and Demographics | ||||
| Somatic (tumor) DICER1 RNase IIIb hotspot variant | 45/45 (100) | 39/42 (93) | .108 | 84/87 (97) |
| Age at Diagnosis: Median [Q1, Q3], years | 14 [12, 16] | 15 [14, 19] | .001 | 15 [13, 17] |
| Age at Diagnosis: Mean [range], years | 14.8 [0 – 65] | 18.9 [2 – 60] | 16.5 [0 – 65] | |
| Clinical Data | ||||
| How Diagnosed | ||||
| Symptomatic | 80/96 (83) | 60/64 (94) | 0.001 | 140/160 (88) |
| DICER1 Surveillance | 13/96 (14) | 0/64 (0) | 13/160 (8) | |
| During Pregnancy | 1/96 (1) | 2/64 (3) | 3/160 (2) | |
| Incidental | 2/96 (2) | 2/64 (3) | 4/160 (3) | |
| Maximum tumor dimension: Median [range], cm | 11.0 [1.8 – 33.0] n=92 | 14.7 [4.1 – 35.0] n=63 | .001 | 12.4 [1.8 – 35.0] n=155 |
| Surgical Data | ||||
| FIGO Stage | ||||
| Stage IA | 59 (61) | 38 (58) | 0.926 | 97 (60)a |
| Stage IC | 30 (31) | 22 (34) | 52 (32)b | |
| Stage II-IV | 8 (8) | 5 (8) | 13 (8)c | |
| Histology | ||||
| Level of differentiation | ||||
| Intermediately | 75 (77) | 44 (68) | 0.173 | 119 (73) |
| Poorly | 22 (23) | 21 (32) | 43 (27) | |
| Retiform pattern | 27 (28) | 17 (26) | .814 | 44 (27) |
| Mesenchymal heterologous elements | 13 (13) | 18 (28) | .023 | 31 (19) |
| Epithelial heterologous elements | 21 (22) | 17 (26) | .507 | 38 (23) |
| JGCT-like foci | 11 (11) | 11 (17) | .309 | 22 (14) |
| Treatment and Outcome | ||||
| Follow up: Median [range] | 54 [0 – 355] | 29 [2 – 139] | .009 | 42 [0 – 355] |
| Adjuvant chemotherapy | 39 (40) | 31 (48) | .346 | 70 (43) |
| Metachronous SLCT | 11 (11) | 0 (0) | .003 | 11 (7) |
| Recurrence | 10 (10) | 16 (25) | .015 | 26 (16) |
| Death | 4 (4) | 12 (18) | .003 | 16 (10) |
| Cause of Death | ||||
| SLCT-related | 3/4 (75) | 10/12 (83) | 0.607 | 13/16 (81) |
| Other malignancy | 1/4 (25) | 1/12 (8) | 2/16 (13) | |
| Non-malignancy related | 0/4 (0) | 1/12 (8) | 1/16 (6) | |
FIGO Stage IA = 85 and IX = 12
FIGO Stage IC1 = 28, IC2 = 8, IC3 = 6, and ICX = 10
FIGO Stage IIA = 2, IIB = 5, IIIA2 = 1, IIIC = 4, and IVB = 1
FIGO, International Federation of Gynecology and Obstetrics; JGCT, juvenile granulosa cell tumor, N = number of evaluable patients; P/LP, pathogenic or likely pathogenic; SLCT, Sertoli-Leydig cell tumor
Molecular Analysis/Testing
Overall, 60% (97/162) of patients were found to have a germline DICER1 P/LP variant (Fig. 1). Nonsense variants were most prevalent (67%, 62/93), followed by splice site/intronic variants in 22% (20/93). (Fig. 1-2).
Figure 1:
Flow diagram with the distribution of patients by germline DICER1 status and subsequent somatic (tumor) DICER1 testing information.
*Of the remaining 10 with an RNase IIIb hotspot variant in which the somatic (tumor) loss-of-function variant was unknown, four tumors were only tested for an RNase IIIb “hotspot” variant, four did not have a DICER1 loss-of-function variant detected, and two did not have tumor testing reports available.
Figure 2:
Lollipop plot of loss-of-function variant distribution within the DICER1 gene among participants with (top) the loss-of-function variant in the germline and participants with (bottom) the loss-of-function variant confined to the tumor (i.e., no germline DICER1 variant).
RNase IIIb hotspot variants were present in 97% (84/87) of tumors with testing available. The most common hotspot location was E1813 (44%, 35/80), followed by D1709 (38%, 30/80) (fig 2). The most common sequence variant was c.5125G>A (p.D1709N), identified in 35% (28/80) of tumors.
Table 2:
Distribution of somatic (tumor) RNase IIIb “hotspot” variants among those with tumor testing with a detected RNase IIIb missense variant and known variant information (n=80).
| Sequence Variant |
Amino Acid Sequence |
Number | Percent | “Hotspot” Location |
Number | Percent |
|---|---|---|---|---|---|---|
| c.5113G>A | p.E1705K | 9 | 11% | E1705 | 10 | 13% |
| c.5114A>T | p.E1705V | 1 | 1% | |||
| c.5125G>A | p.D1709N | 28 | 35% | D1709 | 30 | 38% |
| c.5126A>G | p.D1709G | 1 | 1% | |||
| c.5126A>T | p.D1709V | 1 | 1% | |||
| c.5127T>A | p.D1709E | 0 | 0% | |||
| c.5425G>A | p.G1809R | 1 | 1% | G1809 | 1 | 1% |
| c.5428G>C | p.D1810H | 2 | 3% | D1810 | 4 | 5% |
| c.5428G>T | p.D1810Y | 2 | 3% | |||
| c.5429A>T | p.D1810V | 0 | 0% | |||
| c.5437G>A | p.E1813K | 7 | 9% | E1813 | 35 | 44% |
| c.5437G>C | p.E1813Q | 8 | 10% | |||
| c.5438A>C | p.E1813A | 1 | 1% | |||
| c.5438A>G | p.E1813G | 6 | 8% | |||
| c.5438A>T | p.E1813V | 1 | 1% | |||
| c.5439G>A | p.E1813D | 1 | 1% | |||
| c.5439G>C | p.E1813D | 1 | 1% | |||
| c.5439G>T | p.E1813D | 10 | 13% |
In patients without a germline DICER1 P/LP variant, tumor testing for DICER1 was available in 42 cases, with 93% having an RNase IIIb hotspot variant (Fig. 1). Of the 39 tumors with an RNase IIIb hotspot variant, the somatic (tumor) loss-of-function variant was known in 29 cases. Nonsense variants were identified in 79% (23/29), splice sites in 10% (3/29), exon-level deletions in 7% (2/29), and missense variants in 3% (1/29) of tumors.
Demographics, Clinical and Surgical Data
Median age at diagnosis was 14 years (range: 0-65 years) among those with a germline DICER1 P/LP variant and 15 years (range: 2-60 years) for those without a germline variant (Table 1). The age distribution for those without a germline DICER1 P/LP variant was skewed toward older age (mean: 14.8 vs 18.9 years, p = .001) (Supplemental Fig. 1). Most (88%, 140/160) tumors were detected symptomatically. FIGO staging was similar among both groups (p=.926).
Histology and Treatment
Tumors in patients without a germline DICER1 P/LP variant were poorly differentiated in 32% (21/65) compared to 23% (22/97) for patients with a germline variant (p=.173). Mesenchymal heterologous elements were present in 28% (18/65) of tumors in patients without a germline DICER1 P/LP variant compared to 13% (13/97) of tumors in patients with a germline variant (p=.023).
When considering all patients with known germline DICER1 status, there was no difference in age when considering those with mesenchymal heterologous elements versus those without (median: 15 vs. 15 years, p=.976) or considering those with intermediately differentiated versus poorly differentiated tumors (median: 15 vs. 15 years, p=.841).
Treatment
Overall, 43% (70/162) of patients received chemotherapy, including 40% (39/97) of patients with a germline DICER1 P/LP variant and 48% (31/65) of those without (p = .346).
Impact of DICER1 Surveillance
Thirteen patients had a primary SLCT detected via DICER1 surveillance either due to a germline DICER1 P/LP variant or for a known familial or personal history of DICER1-related condition. The median age at detection for these patients was 13 years (range: 4-24 years). The median maximum tumor dimension was 8.1 cm (range: 2.4-22.0 cm). Nearly all (92%, 12/13) were resected as stage IA. One patient had stage IIA disease and was the only patient in this subset to receive upfront chemotherapy and the only patient with a recurrence. This individual was undergoing pelvic ultrasounds every 6 months with interval development of an 8.7 cm complex cystic mass. The patient had a subsequent pelvic ultrasound 3 months later with interval growth to 13.7 cm, then underwent resection and was found to have fallopian tube involvement.
Outcome
Median follow-up was 54 months (range: 0-355 months) for those with a germline DICER1 P/LP variant and 29 months (range: 2-139 months) for those without a germline variant (p=.009). All (n=11) metachronous SLCTs were identified in patients with a germline DICER1 P/LP variant. There were 26 recurrences, 10 with a germline DICER1 P/LP variant and 16 without a germline variant. There were 16 deaths, with most (81%, 13/16) being SLCT-related. Of the non-SLCT-related deaths, one patient with a germline DICER1 P/LP variant died from anaplastic sarcoma of the kidney.17 The two non-SLCT-related deaths among those without a germline variant included metastatic Wilms’ tumor and complications from a COVID-19 infection.
Survival estimates were assessed among patients who presented with symptomatic SLCT (Fig. 3A-B). Among patients with a germline DICER1 P/LP variant, 3-year RFS was 86.7% (95% CI: 78.9-95.3%) compared to 81.0% (95% CI: 70.8-92.5%) for patients without a germline variant (p=.014). Three-year and 5-year OS was 95.1% (95% CI: 89.7-100.0%) among patients with a germline DICER1 P/LP variant compared to 3-year OS of 90.7% (95% CI: 82.3-100.0%) and 5-year OS of 78.5% (95% CI: 64.9-95.0%) for those without a germline variant (p<.001).
Figure 3:
A. Sertoli-Leydig cell tumor (SLCT) recurrence-free survival (RFS) and B. overall survival (OS) stratified by germline DICER1 status among patients that presented symptomatically. C. SLCT RFS and D. OS adjusted for other prognostic factors among patients that presented symptomatically and stratified by germline DICER1 status.
Propensity score overlap weights for germline DICER1 status improved balance among prognostic factors (Supplemental Fig. 2). Adjusted 3-year RFS was 87.2% (95% CI: 79.4-95.8%) for patients with a germline DICER1 P/LP variant compared to 78.1% (95% CI: 66.4-91.9%) for those without a germline variant (p=.043) (Fig. 3C). Adjusted 3-year and 5-year OS was 93.9% (95% CI: 87.3-100.0%) for those with a germline DICER1 P/LP variant compared to 3-year OS of 91.3% (95% CI: 83.4-100.0%) and 5-year OS of 78.2% (95% CI: 63.8-95.9%) for those without a germline variant (p=.021) (Fig. 3D).
Prior and Subsequent Conditions
There were 36 prior and/or concurrent neoplasms (typically DICER1-related conditions) detected among 29 patients (Supplemental Table 1). Nearly all occurred in patients with a germline DICER1 P/LP variant, with 33 total neoplasms in 26 patients. Three patients without a germline variant had a prior and/or concurrent neoplasm, including a prior medulloblastoma, a prior Wilms’ tumor, and a concurrent renal cell carcinoma.
In addition to the metachronous SLCT diagnoses listed, subsequent neoplasms are reported (Fig. 4, Supplemental Table 2). Overall, 28 neoplasms were identified among 23 of the 97 patients with a germline DICER1 P/LP variant. There was one subsequent neoplasm (thyroid carcinoma) in a patient without a germline variant.
Figure 4:
A. Proportion of patients with a non-recurrent Sertoli-Leydig cell tumor (SLCT) event stratified by germline DICER1 status. B. Proportion of patients with a germline DICER1 variant with a subsequent event subcategorized by neoplasm type. A complete list of subsequent neoplasms is included in Supplemental Table 2.
Among patients with a germline DICER1 P/LP variant, the proportion with a subsequent, non-recurrent event was 25.8% (95% CI: 14.3-35.8%) at 5 years and 36.2% (95% CI: 21.4-48.1%) at 10 years (Fig. 4A). In contrast, the proportion with a subsequent non-recurrence event for those without a germline variant was 1.8% (95%: 0.0-5.3%) at both 5 and 10 years (p=.001).
Metachronous SLCT was the most common subsequent neoplasm among patients with a germline DICER1 P/LP variant, with a rate of 18.9% (95% CI: 7.0-29.3%) at 10 years (Fig. 4B). Additionally, the rate of subsequent other malignancy (metachronous SLCT or thyroid carcinoma captured separately) was 9.8% (95% CI: 0.5-18.3%) at 10 years.
DISCUSSION
We found that having a germline DICER1 P/LP variant was associated with improved RFS and OS in patients diagnosed with SLCT. The protective effect was sustained even when balancing for other prognostic factors and restricting to patients detected symptomatically to control for the known protective impact of early detection with DICER1 surveillance.3,6
We highlight a higher prevalence of mesenchymal heterologous elements among patients without a germline DICER1 P/LP variant. We initially hypothesized that early detection among those with a germline variant may have resulted in fewer tumors developing higher-risk histology. However, among all patients with known germline DICER1 status, those with mesenchymal heterologous elements presented at the same age, on average, as those without mesenchymal heterologous elements. The same held for the level of differentiation, as poorly and intermediately differentiated tumors presented, on average, at the same age. This finding suggests that the higher-risk histology is not necessarily driven by age. Further study is needed to understand the driving factors for developing higher-risk histology in ovarian SLCT.
When considering the impact of germline predisposition on prognosis, a similar protective effect was demonstrated among those with a germline DICER1 P/LP variant, with type II and III PPB treated with upfront chemotherapy, although further study is warranted.18 In contrast, a study of rhabdomyosarcoma (RMS) found that germline cancer-predisposition variants (CPVs) in previously reported RMS-predisposition genes were associated with significantly worse event-free survival and OS.19 The results were driven by germline variants in TP53 and HRAS, whereas NF1 and BRCA2 CPVs did not demonstrate the same association.19 For TP53 variants, the outcomes were independent of secondary malignancies.19 Study of the effect of DICER1 heterozygosity in the tumor microenvironment of animal models may provide further insight into differences in tumorigenesis that impact outcomes in germline predisposition and should be a future research direction. We hypothesize that the genetic/epigenetic events required for tumor development vary between those with a germline DICER1 variant and those with tumor-confined biallelic DICER1 variants, driving the difference in outcomes.
In a study of individuals with germline DICER1 P/LP variants, the risk of developing an SLCT by age 50 was just over 5%.20 Meanwhile, among patients with SLCT and a germline DICER1 P/LP variant in this cohort, we found a subsequent risk of a metachronous SLCT at 10 years from diagnosis to be almost 20%. This raises the question of whether the underlying germline variant impacts the phenotype in DICER1-related tumor predisposition. In a report of patients with PPB, of those with germline predisposition, 8% (7/90) were found to have a splice site/intronic variant, whereas just over 20% of patients with an SLCT and germline predisposition had a splice site/intronic DICER1 variant in this cohort.10 Study of genotype-phenotype correlation may provide important information for stratifying tumor risk in individuals with DICER1-related tumor predisposition.
Notably, a patient with a prior Wilms’ tumor (a known DICER1-related neoplasm) had negative germline DICER1 testing that included deletion/duplication testing. Tumor testing was not available on either the Wilms’ tumor or SLCT. This case raises the question of whether there was an undetected germline DICER1 variant, undetected mosaicism for loss-of-function variant, or a variant in a related gene.
Among the participants without a germline DICER1 P/LP variant and with tumor sequencing, there were four tumors in which only an RNase IIIb missense variant was identified. Biallelic variants (a loss-of-function and an RNase IIIb missense variant) are expected, and identification of only an RNase IIIb variant on tumor testing should prompt consideration of additional germline testing, especially if this group is enriched for splice site/intronic variants. These four patients were included in the germline DICER1 negative group, although it is possible they had an undetected germline variant. Germline DICER1 testing should consist of deletion/duplication analysis, as exon-level deletions are described in DICER1-related tumor predisposition.6,10,21 Additionally, a recent study has detected deep intronic pathogenic variants that may not be detected on standard sequencing, and intronic sequencing and splicing analysis should be considered.22 These findings also raise the question of whether individuals without a detectable germline DICER1 variant may have a distinct non-DICER1-related genetic/epigenetic event in a related or unrelated pathway contributing to tumorigenesis; additional sequencing of such cases may provide important insights.
Patients with SLCT detected via surveillance are an important cohort to help evaluate the efficacy of surveillance guidelines for DICER1-related tumor predisposition. As with prior findings, surveillance facilitated asymptomatic detection of SLCT with resection at a lower stage, smaller maximum dimension, and lower rate of chemotherapy.
Genetic counseling and germline DICER1 testing are critical in patients who present with an ovarian SLCT, as 60% of patients were found to have a germline DICER1 P/LP variant. DICER1 alterations in patients with testicular Sertoli-stromal/sex cord-stromal cell tumors have only rarely been reported.23,24 Germline DICER1 P/LP variants are associated with a spectrum of benign and malignant neoplasms as well as lung cysts and thyroid nodules.25-27 Surveillance recommendations have demonstrated efficacy in the early detection of malignancies in their earliest, most treatable forms.6,28 While primary SLCT-specific outcomes were more favorable, we highlight the risk of subsequent malignancies and other DICER1-related neoplasms among patients with a germline DICER1 P/LP variant. Prior/concurrent and subsequent non-recurrence events were rare among patients without germline DICER1 predisposition. Among patients with a germline DICER1 P/LP variant, the risk of subsequent DICER1-related conditions may be underestimated as both benign thyroid nodules and unresected lung cysts were excluded due to incomplete ascertainment.
Nearly all intermediately and poorly differentiated SLCTs are associated with a somatic (tumor) RNase IIIb missense variant.3 Given the risk of a subsequent metachronous SLCT of the contralateral ovary among patients with germline predisposition, tumor testing can provide further insight and guide therapeutic decision-making, given worse outcomes noted in recurrent disease. SLCT recurrence can be isolated to the contralateral ovary, although in our experience, metachronous disease is a far more common etiology of disease isolated to this location. Different tumor-specific RNase IIIb hotspot variants are indicative of metachronous disease. Identical tumor-specific RNase IIIb hotspot variants are more suggestive of recurrent disease. However, it is possible to have a metachronous tumor with the same hotspot, and careful clinical judgment is needed in situations where the same hotspot is detected in a subsequent SLCT confined to the contralateral ovary. Histologic differences and tumor sequencing beyond DICER1 may also provide insight into the determination of metachronous versus recurrent disease in instances where the same hotspot is detected.
Expanded somatic testing beyond DICER1 sequencing was infrequent among our cohort, and thus, associations with additional somatic variants and outcomes could not be assessed. It is unknown if additional tumor sequencing would identify coexisting variations in other genes and/or chromosome instability that drive tumor histology and/or prognosis; this is a direction of future research. Similarly, expanded germline testing was available in only a small subset of patients. Expanded germline testing remains an area of interest and may provide additional insights into the risk for tumor development.
This analysis is strengthened using propensity score methods to balance covariates among the study groups based on published prognostic factors.3 Although developed for causal inference of interventions, propensity score methods are helpful for association studies because they can adjust for several potential confounding factors even when the effective sample size (i.e., number of events) is small.
Limitations to this analysis include that treatment decisions were at the treating physician’s discretion; thus, specific chemotherapy regimens and indications for chemotherapy were not uniform. Individuals with a germline DICER1 variant may be more likely to enroll in the Registries; thus, our cohort may overestimate the percentage of patients with SLCT that harbor a germline DICER1 variant. The data are at risk of overestimating the neoplasms accounted for in this analysis, as those with subsequent conditions are expected to be more likely to seek care. We mitigate this risk within the Registries by systematically pursuing follow-up from treating institutions and enrolled patients. Additionally, we exclude patients with mosaicism for a loss-of-function DICER1 variant, who are expected to have similar-to-less severe phenotypes, and those with an RNase IIIb missense variant, who are expected to have more severe phenotypes and warrant expert consultation and individualized surveillance.10 Nonetheless, this report provides the most comprehensive assessment of the impact of germline DICER1 predisposition in patients diagnosed with ovarian SLCT.
CONCLUSION
This cohort study of patients with SLCT with known germline DICER1 status demonstrated that patients with germline predisposition had superior RFS and OS. Although further research is needed, germline DICER1 status may provide important prognostic information among patients with ovarian SLCT and provide a basis for preclinical study on the effect of DICER1 heterozygosity on tumorigenesis. Germline testing helps identify patients at risk of subsequent neoplasms, including metachronous SLCT.
Supplementary Material
CONTEXT.
Key Objective:
We analyze the role of germline DICER1 status in ovarian Sertoli-Leydig cell tumor (SLCT) outcomes, focusing on associations with survival outcomes while controlling for other prognostic factors. Additionally, we systematically assess the subsequent risk of metachronous SLCT and other malignancies and neoplasms stratified by germline DICER1 status.
Knowledge Generated:
This cohort study of patients with SLCT demonstrated that those with germline DICER1 pathogenic/likely pathogenic (P/LP) variants had superior recurrence-free and overall survival even when adjusting for other known prognostic factors. Prior/concurrent and subsequent neoplasms were rare among those without a germline DICER1 P/LP variant. Metachronous SLCT was the most common subsequent neoplasm among patients with a germline DICER1 P/LP variant, with a rate of nearly 20% at 10 years.
Relevance:
Germline DICER1 status may provide important prognostic information for individuals with ovarian SLCT and provide a basis for preclinical study on the effect of DICER1 heterozygosity on tumorigenesis.
Acknowledgments:
The authors wish to thank the many treating physicians, genetic counselors, patients, and families who collaboratively support the International PPB/DICER1 Registry and International Ovarian and Testicular Stromal Tumor (OTST) Registry as well as the Pine Tree Apple Classic Fund, whose volunteers, tennis players, and donors have provided more than 39 years of continuous support for PPB and DICER1 Research. The authors gratefully acknowledge the contributions of #livelikemegan to ongoing Sertoli-Leydig cell tumor research initiatives.
Funding Statement:
The International PPB/DICER1 Registry is supported by the Pine Tree Apple Classic Fund, Children’s Minnesota Foundation, Mendon F. Schutt Foundation, and Rein in Sarcoma. This analysis was supported by a grant from the Children’s Minnesota Internal Research Grant Program, Cancer Prevention and Research Institute of Texas (RR180071), and the Intramural Research Program of the Division of Cancer Epidemiology and Genetics of the National Cancer Institute, Rockville, MD. Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Numbers R37CA244940 and R01CA143167. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Conflicts of Interest:
Dr. Hill is the owner of ResourcePath LLC, a company that does research and development of laboratory tests, including for DICER1 cancers. That work is unrelated to the information presented in this article. Dr. Stewart provides telegenetics services for Genome Medical, Inc., in accordance with relevant National Cancer Institute policies. The remaining authors have no conflicts to disclose.
REFERENCES
- 1.WHO Classification of Tumours Editorial Board: World Health Organization classification of tumours. 5th ed. Female genital tumours, IARC Press, 2020 [Google Scholar]
- 2.Hill DA, Ivanovich J, Priest JR, et al. : DICER1 mutations in familial pleuropulmonary blastoma. Science 325:965, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nelson AT, Harris AK, Watson D, et al. : Outcomes in ovarian Sertoli-Leydig cell tumor: A report from the International Pleuropulmonary Blastoma/DICER1 and Ovarian and Testicular Stromal Tumor Registries. Gynecologic Oncology 186:117–125, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schultz KAP, Harris AK, Finch M, et al. : DICER1-related Sertoli-Leydig cell tumor and gynandroblastoma: Clinical and genetic findings from the International Ovarian and Testicular Stromal Tumor Registry. Gynecol Oncol 147:521–527, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Schultz KAP, Williams GM, Kamihara J, et al. : DICER1 and Associated Conditions: Identification of At-risk Individuals and Recommended Surveillance Strategies. Clin Cancer Res 24:2251–2261, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schultz KAP, Nelson AT, Mallinger PHR, et al. : DICER1-Related Tumor Predisposition: Identification of At-risk Individuals and Recommended Surveillance Strategies. Clinical Cancer Research, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Schultz KAP, MacFarland SP, Perrino MR, et al. : Update on Pediatric Surveillance Recommendations for PTEN Hamartoma Tumor Syndrome, DICER1-Related Tumor Predisposition, and Tuberous Sclerosis Complex. Clinical Cancer Research:OF1-OF 11, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Prat J, Young RH, Scully RE: Ovarian Sertoli-Leydig cell tumors with heterologous elements. II. Cartilage and skeletal muscle: a clinicopathologic analysis of twelve cases. Cancer 50:2465–75, 1982 [DOI] [PubMed] [Google Scholar]
- 9.Plastini T, Staddon A: Sertoli-Leydig Cell Tumor with Concurrent Rhabdomyosarcoma: Three Case Reports and a Review of the Literature. Case Rep Med 2017:4587296, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brenneman M, Field A, Yang J, et al. : Temporal order of RNase IIIb and loss-of-function mutations during development determines phenotype in pleuropulmonary blastoma / DICER1 syndrome: a unique variant of the two-hit tumor suppression model. F1000Res 4:214, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Freeman PJ, Hart RK, Gretton LJ, et al. : VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat 39:61–68, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.McCluggage WG, Rivera B, Chong AS, et al. : Well-differentiated Sertoli-Leydig Cell Tumors (SLCTs) Are Not Associated With DICER1 Pathogenic Variants and Represent a Different Tumor Type to Moderately and Poorly Differentiated SLCTs. Am J Surg Pathol 47:490–496, 2023 [DOI] [PubMed] [Google Scholar]
- 13.Thomas LE, Li F, Pencina MJ: Overlap Weighting: A Propensity Score Method That Mimics Attributes of a Randomized Clinical Trial. JAMA 323:2417–2418, 2020 [DOI] [PubMed] [Google Scholar]
- 14.Li F: Overlap Weighting, Handbook of Matching and Weighting Adjustments for Causal Inference, Chapman and Hall/CRC, 2023, pp 263–282 [Google Scholar]
- 15.Austin PC, Stuart EA: Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 34:3661–79, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.R Core Team. A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria, 2021 [Google Scholar]
- 17.Schoettler PJ, Smith CC, Nishitani M, et al. : Anaplastic sarcoma of the kidney (DICER1-sarcoma of the kidney): A report from the International Pleuropulmonary Blastoma/DICER1 Registry. Pediatr Blood Cancer:e31090, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schultz KAP, Harris AK, Nelson AT, et al. : Outcomes for Children With Type II and Type III Pleuropulmonary Blastoma Following Chemotherapy: A Report From the International PPB/DICER1 Registry. J Clin Oncol 41:778–789, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Martin-Giacalone BA, Li H, Scheurer ME, et al. : Germline Genetic Testing and Survival Outcomes Among Children With Rhabdomyosarcoma: A Report From the Children’s Oncology Group. JAMA Netw Open 7:e244170, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Stewart DR, Best AF, Williams GM, et al. : Neoplasm Risk Among Individuals With a Pathogenic Germline Variant in DICER1. J Clin Oncol 37:668–676, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.de Kock L, Wu MK, Foulkes WD: Ten years of DICER1 mutations: Provenance, distribution, and associated phenotypes. Hum Mutat 40:1939–1953, 2019 [DOI] [PubMed] [Google Scholar]
- 22.Fraire CR, Mallinger PR, Hatton JN, et al. : Intronic Germline DICER1 Variants in Patients With Sertoli-Leydig Cell Tumor. JCO Precis Oncol 7:e2300189, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Golmard L, Vasta LM, Duflos V, et al. : Testicular Sertoli cell tumour and potentially testicular Leydig cell tumour are features of DICER1 syndrome. J Med Genet 59:346–350, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Acosta AM, Sholl LM, Maclean F, et al. : Testicular Neoplasms With Sex Cord and Stromal Components Harbor a Recurrent Pattern of Chromosomal Gains. Mod Pathol 37:100368, 2024 [DOI] [PubMed] [Google Scholar]
- 25.González IA, Stewart DR, Schultz KAP, et al. : DICER1 tumor predisposition syndrome: an evolving story initiated with the pleuropulmonary blastoma. Mod Pathol 35:4–22, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Nelson AT, Vasta LM, Watson D, et al. : Prevalence of lung cysts in adolescents and adults with a germline DICER1 pathogenic/likely pathogenic variant: a report from the National Institutes of Health and International Pleuropulmonary Blastoma/DICER1 Registry. Thorax 79:644–651, 2024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Khan NE, Bauer AJ, Schultz KAP, et al. : Quantification of Thyroid Cancer and Multinodular Goiter Risk in the DICER1 Syndrome: A Family-Based Cohort Study. J Clin Endocrinol Metab 102:1614–1622, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.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 61:1695–7, 2014 [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.