Imaging of Pediatric Testicular Tumors: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper

Abstract

Primary intratesticular tumors are uncommon in children but incidence and risk of malignancy both sharply increase during adolescence. Ultrasound is the mainstay for imaging the primary lesion and cross-sectional modalities are often required for evaluation of regional or distant disease. However, variations to this approach are dictated by additional clinical and imaging nuances. This paper offers consensus recommendations for imaging of pediatric patients with a known or suspected primary testicular malignancy at diagnosis and during follow-up.

Introduction

Testicular tumors in prepubertal boys are rare, mostly benign, and are typically teratomas¹. The incidence of testicular tumors sharply increases in the second decade of life when malignant tumors predominate^1,2. Primary testicular tumors in children may be broadly classified into germ cell tumors (GCT) and sex cord-stromal tumors. In 2016, the World Health Organization (WHO) classification of testicular tumors was modified, further dividing GCT based on presence or absence of germ cell neoplasia in situ (GCNIS)³. Table 1 highlights the different types of testicular tumors along with their differing clinical and imaging features.

Table 1:

Intratesticular tumor imaging and clinical features classified by WHO

Tumor Type	Arise from GCNIS*?	Imaging features on US	Clinical features
Seminoma	Yes	Homogenous solid mass	Rare in children
Embryonal carcinoma	Yes	Poorly circumscribed; heterogenous	May be aggressive
Choriocarcinoma	Yes	Hypoechoic; poorly defined	Hemorrhagic metastatic potential
Post-pubertal teratoma	Yes	Heterogenous cyst/solid mass with calcification	Majority are malignant
Pre-pubertal teratoma	No	Heterogenous cyst/solid mass with calcification	More likely component of mixed GCT. Often benign or indolent.
Post-pubertal Yolk Sac Tumor	Yes	Heterogenous cyst/solid mass with calcification	Very rare in pure form; elevated α-fetoprotein
Pre-pubertal Yolk Sac Tumor	No	Solid hypervascular, homogenous mass; diffuse testicular enlargement	More often pure form; elevated α-fetoprotein
Epidermoid Cyst	No	“Onion ring” appearance; avascular	Non-neoplastic
Leydig cell tumor	No	Hypoechoic, often small, peripherally situated mass.	Precocious puberty
Sertoli cell tumor	No	Variable. But often calcified and multiple in children	Associated with Carney and Peutz-Jeghers syndromes
Gonadoblastoma	No	Findings not well described; rare	Undescended testis or gonadal dysgenesis

^*GCNIS – germ cell neoplasia in situ

Testicular cancer is the most common solid tumor diagnosed in males between 15 and 34 years of age⁴. GCT are the most common tumor type and account for more than 15% of all testicular malignancies diagnosed during adolescence^5,6. Patients typically present with a firm, painless mass. Signs of precocious puberty suggest that a sex cord stromal tumor, typically a Leydig cell tumor, is present. Patients with cryptorchidism have a greater risk (3.6 relative risk) of developing a testicular tumor⁷. Additional risk factors for testicular malignancy include cancer in the contralateral testis and a family history of the disease⁸.

Staging:

Both the American Joint Committee on Cancer (AJCC) and the Children’s Oncology Group (COG) have published staging systems. In prepubertal children, the COG staging system is most often followed. This staging system includes a mix of surgical and imaging components. Imaging plays an important role in the assessment of local disease, regional nodes (retroperitoneal nodes) and distant metastasis. For post-pubescent boys, there is some controversy regarding the use of the adult-oriented National Comprehensive Cancer Network (NCCN) versus the COG approach as data suggests better outcomes using the former which is itself predicated on use of the AJCC Tumor-Nodes-Metastases (TNM) system^1,9. In 1997, the International Germ Cell Cancer Collaborative Group (IGCCCG) published a risk classification scheme which included both adults and children but also included extra-testicular germ cell tumors. The IGCCCG system has been revised and validated based on effective prognostication^10-12. Though limited, there is evidence for improved outcomes in adolescents with metastatic GCT when adhering to IGCCCG guidelines as the biology of tumors in this group is more similar to adults¹³. The IGCCCG Risk Classification scheme is based on the location of disease, the presence and location of metastatic disease, and laboratory values.

Imaging at Diagnosis:

In a boy suspected of having a testicular mass, US is the initial imaging test of choice [GRADE A; SOR 1]. Its reported sensitivity between 92 and 98% allows reliable exclusion of an intratesticular mass when none is present^14,15. Additionally, US may reveal another cause for a palpable mass such as hydrocele, epididymal cyst, or an extra-testicular mass. While US can distinguish the above entities from a testicular tumor, it cannot reliably differentiate the histologic types of intratesticular masses^16,17. Benign tumors, though, are more common in prepubertal boys and testicular-sparing surgery can be considered in this population. To this end, the radiologist’s impression of spared, viable testicular tissue may be helpful to the surgeon^17,18. In post-pubertal boys the pre-test probability of malignancy is considerably higher and testicular-sparing surgery is not commonly performed.

Currently US is the only imaging modality required for evaluation of a primary intratesticular mass [GRADE B, SOR 1.30]. MRI can be employed in cases with inconclusive scrotal US findings or when scrotal US findings are inconsistent with clinical findings^19-21. For example, MRI can support the suspicion of an epidermoid cyst after equivocal US findings²². However, to date, data supporting the routine use of MRI for intratesticular tumors remains limited^23-25.

In all patients with a malignant intratesticular mass, CT or MRI of the remainder of the abdomen is mandatory to assess retroperitoneal lymphadenopathy and metastatic disease [GRADE A, SOR 1.00]. In a recent systematic analysis of young adult men with testicular tumors, the reported sensitivity and specificity of CT to detect metastatic lymph nodes has ranged between 66.7 – 84% and 66.7 – 95.2% respectively ^26-28. There is less data on the performance of MRI. However, studies have shown it to have a sensitivity of 78 – 80% and a specificity of 91 – 96% for the evaluation of retroperitoneal lymph nodes^27-29. When describing lymph nodes, it is suggested that the radiologist report two dimensions of any suspicious nodes as the AJCC staging system relies on the long axis, but the COG guidelines are based on short axis [GRADE B, SOR 1.40]^6,15,28.

Chest CT should be performed at diagnosis in all adolescent patients with a malignant testicular mass to identify pulmonary metastatic disease [GRADE C, SOR 1.40]. In adults with non-seminomatous GCT, the American Urological Association (AUA) recommends either chest radiograph or CT at diagnosis³⁰. Adolescent event-free survival is less favorable than both prepubescent boys and adult men with non-seminomatous GCT³¹. Additionally, there is a trend toward higher grade tumors in adolescents. CT offers much higher spatial resolution and is more sensitive for the diagnosis of pulmonary metastases.

Intravenous contrast is required for abdominal CT scans while for MRI, intravenous contrast is preferred [GRADE D, SOR 1.10].

Follow-up Imaging:

In post-pubescent adolescent males with testicular tumors, limited data is available to recommend a detailed imaging surveillance schedule. As such, there is current variability in practice patterns between institutions. Fortunately, varying the interval between scans by several months has not been shown to yield significantly different outcomes³². When considering follow-up imaging, several principles should be kept in mind: 1) Most childhood testicular neoplasms recur within the first two years. In fact, it has been reported that 80% of all relapses of non-seminomatous germ cell tumors occur within one year, 90% by two years and nearly all by three years post-orchiectomy^33,34. 2) Recurrence of seminomas can be more delayed compared with non-seminomatous GCT. Therefore, imaging surveillance data derived from adult patients - which is heavily populated by seminomatous tumors - cannot be readily extrapolated to the pediatric population.

Periodic US surveillance of the contralateral testicle is recommended in patients previously treated for a testicular malignancy – especially in the setting of an additional risk factor such as history of cryptorchid testis [GRADE D, SOR 2.0]^35,36. In adult series, overall risk of metachronous malignancy in the contralateral testis has been shown to be 12 times higher than the general population^37,38. Moreover, the higher risk was skewed toward younger patients³⁷. The recommendation is based on the concept that a small tumor may not be palpable. Incidentally detected testicular lesions have been shown to only rarely harbor malignancy³⁹. However, data on nonpalpable, radiologically detected tumors in the population of patients with a history of testicular malignancy is lacking.

Either CT or MRI of the abdomen should be obtained to surveil for recurrent nodal disease in all adolescent patients with testicular tumors [GRADE B, SOR 1.80] ^29,40. Most authors recommend following COG guidelines for follow-up surveillance in pre-pubertal patients and following adult guidelines in post-pubertal patients^41,42. There are differing recommendations on the frequency for imaging surveillance. The AUA recommends abdominal CT every 3-6 months in year 1, every 4-12 months in year 2, once in year 3, and once in year 4 or 5 in adult men with Stage I non-seminomatous GCT³⁰. The NCCN recommends surveillance for 3 to 5 years with the frequency of imaging depending on the histology of the tumor, the type of therapy, the stage of the tumor at diagnosis^43,44. In both guidelines, imaging is used along with physical exam and laboratory evaluation.

For children with stage I disease, chest radiographs suffice at follow-up⁴⁵ [GRADE C, SOR 2.40]. This is consistent with adult guidelines which recommend chest radiography for all patients. In the NCCN guidelines, chest CT with IV contrast is recommended for patients with thoracic symptoms or in those with supradiaphragmatic disease at baseline⁴⁴. Thus, surveillance chest CT is recommended for patients with COG Stage II or higher disease and should be performed in any patient with metastatic disease (stage IV) at initial presentation⁴⁶ [GRADE C, SOR 1.50].

Routine use of ¹⁸F fluorodeoxyglucose (FDG) PET/CT for staging or surveillance of non-seminomatous GCT is not recommended. Instead, its use is reserved to identify occult disease in patients with rising tumor markers and a negative CT or MR, or to evaluate equivocal findings that could affect surgical planning or therapy [GRADE C, SOR 1.78]⁴⁷. There is an emerging role for fluorodeoxyglucose F 18 (FDG)-PET/CT in problem-solving equivocal findings and also in patients with rising tumor markers in the context of negative CT or MR.⁴⁷ Although PET offers excellent negative predictive value in the setting of seminomas, The International Global Germ Cell Cancer Group Registry cautions against clinical decisions based on PET positivity alone due to risk of false positives^7,48. Currently, there is little evidence to support its routine use in staging or follow-up surveillance in non-seminomatous GCT in the pediatric population however it should be considered in equivocal cases.

Tumor Response Assessment:

All regional lymph nodes ≥ 8 mm in their short axis should be reported [GRADE B, SOR 1.80]. Based on work in adult populations with testicular cancer, it has been suggested that lymph nodes ≥ 8 mm in the short axis be deemed suspicious at diagnosis^40,49,50. Because most patients with testicular tumors undergo up-front orchiectomy, tumor response assessment is mainly focused on regional nodes. However, even retroperitoneal nodes less than 1 cm should be reported – especially if ipsilateral to the primary tumor.

During routine post-treatment surveillance imaging, the radiologist should identify complications such as incisional hernia, bowel adhesions, and radiation enteritis. Early evidence of pulmonary fibrosis should be sought in patients receiving bleomycin therapy. Secondary malignancies are the most common cause of death in survivors of testicular cancer, occurring in 1% to 2%. However, to date, there is no data to support long-term screening for secondary malignancy. Other potential late effects from testicular tumor therapy include hypogonadism, infertility, cisplatin-related nephrotoxicity, and atherosclerotic disease. Imaging is currently not used to screen for any of these effects^4,51.

Testicular Microlithiasis:

Post-pubertal boys with testicular microlithiasis (TM) should be educated on the importance of regular self-exams in addition to being referred to a urologist for consideration of US screening [GRADE D; SOR 2.20]. The topic of screening for testicular tumors in patients with TM is controversial. Currently, most groups only recommend screening for patients with TM and additional risk factor(s) for testicular malignancy which include cryptorchidism, testicular atrophy and previous testicular cancer^52-57. Nevertheless, recent studies in children with TM, including one large multi-institutional retrospective study confirm a statistically significantly strong association between TM and testicular malignancy^58,59. It is still possible that the two are not causally related and that TM may be merely a marker of risk in patients with already-known risk factors. However, given the known association and this open question, there is some hesitancy in universally extending the recommendation of no screening to the child and adolescent population⁶⁰.

If TM is seen in the setting of an additional risk factor, both self-examinations and annual screening US is recommended in post-pubertal boys [GRADE A; SOR 1.40].

The Future of Imaging in testicular tumors:

Although MRI cannot reliably differentiate malignant from benign tumors, any imaging assistance to this end may be helpful in the prepubescent boy in whom testicular-sparing surgery may be feasible. Much work in advancing oncologic MRI is focused on diffusion weighted imaging (DWI). However, in addition to malignant tumors, the testicular parenchyma normally demonstrates restricted diffusion⁶¹, an observation which has historically tempered enthusiasm for DWI. Despite this, there is emerging evidence to suggest use of MRI, including DWI with attention to technical details in discriminating benign from malignant lesions^62,63. For example, marked restricted diffusion (more than the normal testicular parenchyma) in intratesticular epidermoid cysts has already shown clinical value²².

Additional MRI sequences such as diffusion tensor imaging, magnetization transfer imaging and proton spectroscopy, which can provide further information about the microstructure and microenvironment of the testicular parenchyma, are being explored⁶⁴. Lastly, radiomics is a promising emerging field that may provide imaging biomarkers for the characterization of testicular masses as well as for the assessment of retroperitoneal metastatic disease^65-68,69-73.

Abbreviation

GCT

Germ Cell Tumor

WHO

World Health Organization

GCNIS

Germ Cell Neoplasia In Situ

AJC

American Joint Committee on Cancer

COG

Children’s Oncology Group

NCCN

National Comprehensive Cancer Network

IGCCCG

International Germ Cell Cancer Collaborative Group

SOR

Strength Of Recommendation

AUA

American Urological Association

TM

Testicular Microlithiasis

DWI

Diffusion Weighted Imaging