If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The World Health Organization classification of urologic cancer 2016 describes 3 noninvasive precursor lesions for testicular germ cell tumor type II (TGCT) of young adults. Germ cell neoplasia in situ is the initial precursor lesion. Intratubular seminoma (ITSE), and intratubular embryonal carcinoma (ITEC) are 2 intermediate precursor lesions. Microinvasive testicular germ cell tumor (MGCT) is an invasive precursor lesion.
Materials and Methods
We undertook a retrospective study of testes obtained using orchiectomy for TGCT and examined precursor lesions. The examinations included immunohistochemical staining of the precursor lesions for octamer-binding transcription factor 4 (OCT4), CD117, and CD30. We examined 63 consecutive and evaluable patients.
Results
Of the patients, 44 had seminoma and 19 had a nonseminomatous TGCT. MGCT was more frequent than ITSE (P = .002; χ2 test). None of the testes had ITEC. Immunohistochemistry showed that 72 of 77 precursor lesions (93%) stained positive for OCT4 and CD117 and negative for CD30. The pattern represents a seminomatous immunophenotype. A meta-analysis of the published studies regarding precursor lesions included 1007 patients. Overall, the pooled rate of ITSE was 29% (95% confidence interval [CI], 18%-40%) and the pooled rate of MGCT was 21% (95% CI, 15%-27%).
The World Health Organization (WHO) classification of urologic cancer 2016 describes 3 precursor lesions for TGCT: germ cell neoplasia in situ (GCNIS), intratubular SEM (ITSE) and intratubular EC (ITEC).
Intermediate precursor lesions, such as ITSE, ITEC, and MGCT, are mainly found in testes concomitant with GCNIS and TGCT. However, the literature about intermediate precursor lesions is limited.
Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group Best practices recommendations in the application of immunohistochemistry in testicular tumors: report from the International Society of Urological Pathology consensus conference.
Octamer-binding transcription factor 4 (OCT4) is a transcription factor, CD117 is a proto-oncogene that is receptor for the V-Hardy-Zuckerman feline sarcoma viral oncogene homolog Kit (KIT) ligand (stem cell factor), and CD30 is a member of the tumor necrosis factor superfamily of receptors. Precursor lesions, SEM, and EC stain positive for OCT4.
Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group Best practices recommendations in the application of immunohistochemistry in testicular tumors: report from the International Society of Urological Pathology consensus conference.
CD30 is specific for EC. Thus, SEM has an OCT4-positive (OCT4+), CD117-poitive (CD117+), and CD30-negative (CD30−) immunophenotype. Previous studies have not examined whether MGCT has a seminomatous or nonseminomatous immunophenotype.
The first aim of the present study was to evaluate the frequency of precursor lesions. The second aim was to elucidate the immunophenotype of precursor lesions. The third aim was to compare the frequency of intermediate precursor lesions in previous studies and in our study. The fourth aim was to indicate how OCT4 could contribute to the pathogenesis of TGCT.
Materials and Methods
Patients and Materials
In this study we selected consecutive patients with TGCT who were examined at the Department of Pathology at Aarhus university hospital in Denmark from December 1, 2011 to December 1, 2013. The selection included 69 patients. The pathologists were unable to obtain paraffin wax blocks of testicular specimens from 6 patients. Therefore, the study included blocks of the testis from 63 patients. Table 1 shows clinical characteristics of the patients. The pathologist had access to blocks of TGCT and the nonmalignant parts of the testis. Tissue specimens of the nonmalignant parts included testicular parenchyma, rete testes, epididymis, spermatic cord, and upper resection border of the orchiectomy specimen. The tissue specimens had been fixed in 10% formalin and embedded in paraffin wax.
Table 1Clinical Characteristics of the Patients in Our Study
Characteristic
Value
Median Age (Range), y
SEM
38.5 (19-64)
Non-SEM
36 (19-53)
Histology
SEM
44 (70)
Non-SEM
19 (30)
Non-SEM Elements
EC
2 (3)
TER
2 (3)
EC + TER
3 (5)
EC + SEM
1 (2)
TER + YST
1 (2)
TER + SEM
1 (2)
YST + SEM
1 (2)
EC + TER + YST
3 (5)
EC + TER + SEM
1 (2)
EC + CC + SEM
2 (3)
EC + TER + CC
1 (2)
EC + TER + YST + SEM
1 (2)
Data are presented as n (%) except where otherwise stated.
Abbreviations: CC = choriocarcinoma; EC = embryonal carcinoma; SEM = seminoma; TER = teratoma; YST = yolk sac tumor.
For the present study, 2 pathologists, M.B.J. and S.H., stained slides with hematoxylin and eosin. The review confirmed the diagnosis of TGCT. The pathologists excluded slides with poor preservation. The pathologist selected paraffin wax blocks for IHC of OCT4, CD117, and CD30. The IHC used an indirect avidin-biotin peroxidase method that detected antigen-antibody complexes with the biomarkers. The used antibodies are indicated in Table 2. As positive controls, the IHC was undertaken for specimens known to be positive. As negative controls, the primary antibody was replaced with the dilution buffer. The pathologists evaluated independently and semiquantitatively the intensity of the IHC staining. They used ++ for a strong staining in all cells of the precursor lesions, +/− for a weak staining in single cells of the precursor lesions, and − for no staining. The pathologists discussed their evaluations at a consensus conference.
Table 2Immunohistochemistry for the Precursor Lesions
Primary Antibody
Company
Clone Number
Dilution
OCT4
Ventana Medical Systems, Inc, Tucson, AZ F. Hoffmann-La Roche, Basel, Switzerland
The pathologists reported GCNIS, ITSE, ITEC, and the histologic types of TGCT as described in the WHO classification. The pathologists also diagnosed MGCT as described previously.
For most MGCT, the malignant germ cells were present as single cells or small clusters of cells in the testicular interstitium. But for a small proportion of MGCT, the malignant germ cells were spread massively. GCNIS, ITSE, and MGCT have common cellular characteristics and do not destruct the tubular architecture. Also ITEC respect the tubular architecture of the testicular interstitium but differ in cellular pattern from ITSE. TGCT destructs the tubular architecture of the testis. In our study we aimed to limit the risk of false-positively classifying intratubular expansions of SEM and EC as ITSE and ITEC. Therefore, the pathologists only included precursor lesions that were >3 high power fields (×40, equivalent to >3 mm) away from the TGCT.
Systematic Review
To compare findings of previous studies with those of the present study, one author (F.E.v.E.) updated a systematic review regarding intermediate precursor lesions for TGCT. A search for literature in PubMed used the search words: ((“testicular neoplasms” or “testis cancer” or “testicular germ cell tumors”) and (“precursor lesions” or “carcinoma in situ” or “intratubular seminoma” or “intratubular embryonal carcinoma” or “microinvasive germ cell tumor”)). He also undertook a manual search in reference lists of original papers and reviews. The systematic review included articles if they reported human studies with at least 10 patients with testes with TGCT. The articles should have a frequency of at least 1 intermediate precursor lesion and should have been published in a European language before July 2017.
Statistical Methods
In our study and meta-analysis we calculated the frequency of a precursor lesion as the proportion of testes with a positive finding of all evaluated testes. We estimated that our study would need ≥ 50 patients to detect whether the frequency of ITSE and ITEC differed significantly from the frequency of MGCT. We used χ2 tests as we compared the frequency of lesions in 2 groups of patients. We used the metaprop program and STATA version 14.2 (StataCorp, College Station, TX) as we summarized the frequency of lesions in the studies. However, the recent version of metaprop excludes studies with a minimal or maximal rate when it summarizes the studies. We considered a P value of ≤ .05 as statistically significant. One of the authors (F.E.v.E.) undertook descriptive and analytic evaluations using STATA version 14.2 (StataCorp).
Ethical Issues
As of July 10, 2013, the regional Committee for Medical Research Ethics in Southern Denmark approved the study and its use of archived human tissues (ID 20130077). The committee also approved that the participants in the study did not have to permit the present reuse of the testicular specimens.
Results
Present Study
Table 1 shows clinical characteristics of the included patients. The age distribution of the patients in our study was characteristic for TGCT type II.
Review of the pathology supported that all patients had TGCT type II. Thus, none of the patients had spermatocytic SEM. As for GCNIS, ITSE, ITEC, and MGCT, the urologists had not reported a macroscopically overt tumor at orchiectomy. Table 3 details the frequency of precursor lesions in the testicular interstitium. GCNIS was more frequent than ITSE, ITEC, and MGCT. MGCT was more frequent than ITSE (P = .002; χ2 test). Surprisingly, testis with nonseminomatous TGCT had a higher frequency of ITSE than testes with SEM (P = .04; χ2 test). None of the testes had ITEC.
Table 3Frequency of Precursor Lesions for Patients According to the Main TGCT Histology in the Present Study
Main TGCT Histology
Total Patient n
Patients With GCNIS, n (%)
Patients With ITSE, n (%)
Patients With ITEC, n (%)
Patients With MGCT, n (%)
Seminoma
44
33 (75)
2 (5)
0 (0)
15 (34)
Nonseminomatous TGCT
19
17 (89)
4 (21)
0 (0)
6 (32)
Total
63
50 (79)
6 (10)
0 (0)
21 (33)
Abbreviations: GCNIS = germ cell neoplasia in situ; ITEC = intratubular embryonal carcinoma; ITSE = intratubular seminoma; MGCT = microinvasive germ cell tumor; TGCT = testicular germ cell tumor type II.
Histologically, in hematoxylin and eosin staining, GCNIS, ITSE, and MGCT had a common cellular structure. The malignant germ cells had clear cytoplasm, large nuclei with hyperchromatic chromatin, and 1 to 2 nucleoli, as shown in Figures 1 and 2. GCNIS, ITSE, and MGCT had not destructed the tubular architecture of the testis.
Figure 1Intratubular Seminoma (ITSE): (A) Immunohistochemistry (IHC) Showing Octamer-Binding Transcription Factor 4 (OCT4); (B) IHC Showing CD117; (C) IHC Not Showing CD30; and (D) Hematoxylin and Eosin Staining Showing the H&E Pattern
The immunophenotype was OCT4+, CD117+, and CD30− for 72 of 77 precursor lesions (93%). Figure 1 shows the typical pattern for ITSE. Figure 2 shows the typical pattern for MGCT. Three precursor lesions (4%) had an OCT4+, CD117−, and CD30− pattern. Two precursor lesions (3%) had an OCT4−, CD117+, and CD30− pattern.
In our study we found a predominance of 2 sequences for the pathogenesis from GCNIS to overt TGCT. For half of the patents, the sequence was directly from GCNIS to TGCT. For a quarter of the patients, the sequence was from GCNIS through MGCT to TGCT. Only 2 testes (3%) had a sequence of GCNIS through ITSE and MGCT to TGCT. Two other testes (3%) with only TER had no precursor lesions.
All Studies
A systematic review and meta-analysis summarized the frequency of GCNIS, ITSE, and MGCT in all selected studies (Tables 4 and 5, and Figure 3, Figure 4, Figure 5, Figure 6).
Combined, the previous studies and our study reported 1077 patients. Most previous studies had used IHC staining for placental alkaline phosphatase. Overall, GCNIS was a histology with a very high frequency, as shown in Figure 3. ITSE had a pooled frequency of 29% (95% confidence interval [CI], 18%-40%) as shown in Figure 4. Overall, in the studies, testes with nonseminomatous TGCT had a higher frequency of ITEC than testes with SEM (Table 5). Two studies did not use a border zone around the nonseminomatous TGCT in the search for precursor lesions.
used a border zone around the TGCT to limit the risk of intratubular extensions of the nonseminomatous TGCT being diagnosed as ITEC. These 2 studies had a lower frequency of ITEC than the 2 studies that did not use a border zone, 0% to 13%. One previous study was a potential outlier.
That study had an extraordinarily high frequency of ITEC. Overall, MGCT had a pooled frequency of 21% (95% CI, 15%-27%), as shown in Figure 6. Previous studies and our study were consistent regarding the frequency of GCNIS, ITSE, and MGCT.
Table 4Frequency of GCNIS, ITSE, and MGCT According to the Main TGCT Histology in Published Studies
Abbreviations: GCNIS = germ cell neoplasia in situ; ITSE = intratubular seminoma; MGCT = microinvasive germ cell tumor; NR = not recorded; TGCT = testicular germ cell tumor type II.
Figure 3Frequency of Germ Cell Neoplasia in Situ (GCNIS) in Studies of Testes With Testicular Germ Cell Tumor Type II (TGCT). The Figure Summarizes Frequency as a Subtotal for Each of the Main TGCT Histologies and as a Total for Both Types
Figure 4Frequency of Intratubular Seminoma (ITSE) in Studies of Testes With Testicular Germ Cell Tumor Type II (TGCT). The Figure Summarizes Frequency as a Subtotal for Each of the Main TGCT Histologies and as a Total for Both Types
Figure 5Frequency of Intratubular Embryonal Carcinoma (ITEC) in Studies of Testes With Testicular Germ Cell Tumor Type II (TGCT). The Figure Summarizes Frequency as a Subtotal for Each of the Main TGCT Histologies and as a Total for Both Types
Figure 6Frequency of Microinvasive Germ Cell Tumor (MGCT) in Studies of Testes With Testicular Germ Cell Tumor Type II (TGCT). The Figure Summarizes Frequency as a Subtotal for Each of the Main TGCT Histologies and as a Total for Both Types
As histologic findings concomitant with TGCT, intermediate precursor lesions had a substantial frequency. As expected, GCNIS had the highest frequency of among the precursor lesions. ITSE and MGCT had a higher frequency than ITSE. GCNIS, ITSE, and MGCT had a common OCT4+, CD117+, and CD30− immunophenotype. Further, previous studies and our study had grossly similar frequency of ITSE and MGCT. The transcription factor OCT4 can contribute to the pathogenesis of TGCT.
Thomas Ulbright, a leading expert of TGCT pathology, argued that the lesion we defined as MGCT was histologically identical to the lesion that he had given the designation intertubular SEM.
Further, the WHO classification describes that intertubular SEM also and predominantly is found in the periphery of seminomas. Thus, despite the cellular similarity, MGCT, as defined in our study, refers to a histology that differs from most of the lesions that the WHO classification describes as intertubular SEM. Henley et al warned against the use of the term microinvasion for MGCT because they believe that a stage of local microinvasion never gives rise to metastases.
In their study, 2 of 12 patients with intertubular SEM presented with metastases. However, Lau et al contradicted this opinion because they found that intertubular SEM increased the metastatic potential of SEM.
Further, we believe that it is a defining histologic characteristic whether malignant germ cells do not destruct the tubular architecture of the testicular interstitium, as in GCNIS, ITSE, and MGCT, or destruct the tubular architecture, as in SEM. Finally, as a stage for other cancers, microinvasion has a well documented metastatic potential.
Previous studies stressed that gene expressions in embryonic stem cells and gonocytes were like those of malignant germ cells.
In addition, induced stem cells and malignant germ cells of precursor lesions have a similar high expression of OCT4. A panel of OCT4, homeobox transcription factor NANOG (NANOG), Krüppel-like factor 4 (KLF4) and the oncogene avian myelocytomatosis viral oncogene homolog, MYC (C-MYC) might induce reprogramming of testicular germ cells to a stem cell pattern.
Correspondingly, malignant germ cells of GCNIS have an especially high expression of OCT4, NANOG, C. elegans homolog of LIN28 (LIN28), SRY-box 17 (SOX17), and KLF4.
These genetic similarities might be significant for the pathogenesis of TGCT. Transfection with OCT4 and SRY-box 2 (SOX2) might make primordial germ cells undergo reprogramming to an induced pluripotent stem cell pattern, like that of EC.
Also, a combination of the transcription factors OCT4, NANOG, and KFL4 and the oncogene MYC might induce the stem cell pattern. However, in a study of adult mice, a panel of transcription factors did not induce a pluripotent pattern in germ cells.
The panel of transcription factors might also be the background for the seminomatous immunophenotype in the malignant germ cells of precursor lesions. GCNIS and SEM express SOX17 but not SOX2, whereas EC expresses SOX2 but not SOX17.
OCT4 and SOX17 might associate with an OCT4/SOX17 motif in enhancers of genes that limit differentiation compared with the effect obtained with the combination of OCT4 and SOX2.
Observational studies support that MGCT might be classified as an invasive, extratubular, precursor lesion. Physicians have followed patients with GCNIS over time with only observation and repeat testicular biopsies.
showed that malignant germ cells of GCNIS can migrate from the seminiferous tubules into the testicular interstitium. Further, Henley et al reported a series of 12 men with intertubular SEM/MGCT without a macroscopic mass of TGCT.
For half of adult patients with GCNIS, the precursor progressed to MGCT within 5 years without having formed a macroscopically overt TGCT. MGCT was mainly found concomitant with GCNIS. Thus, malignant germ cells from GCNIS might invade the extratubular interstitium as an active process. The invasion might not have been the result of overflow when malignant cells completely fill the seminiferous tubules, as in ITSE.
The receptor CD117 might stimulate growth of SEM. The receptor CD30 might stimulate growth of EC. However, few previous studies of precursor lesions have undertaken immunotyping as in our study. One study of GCNIS reported that the precursor lesions in 44 of 44 patients (100%) had staining for OCT4.
The authors also proposed that GCNIS develops into a nonseminomatous TGCT through a SEM-like stage. In our study, ITSE as well as MGCT had a seminomatous immunophenotype. Hence, ITSE and MGCT could be the proposed intermediate SEM-like stage in the process leading to a nonseminomatous TGCT.
The common immunophenotyped of GCNIS, ITSE, and MGCT indicates that malignant germ cells might develop into ITSE and MGCT before the transition to EC. Engrafting cells of the TCam-2 SEM cell line to the flank of mice exemplifies the transition from SEM to EC.
Thus, although the transition from SEM to EC was driven mainly by the shift from expression of SOX17 to expression of SOX2, OCT4 might contribute to the transition.
In our study, EC was often combined with TER among the nonseminomatous TGCT. A previous study reported that 46 of 149 patients with nonseminomatous TGCT had EC combined with TER.
Future studies might further elucidate how an OCT4 panel of transcription factors contributes to the pathogenesis of TGCT.
Limitations
The study has limitations. In the investigation of patients with TGCT we used only 2 methods, a retrospective cohort study and a systematic review, as we estimated the frequency of concomitant precursor lesions. The recent metaprop program (Stata Corp) excluded studies with minimal and maximal rates when it summarized the rates in several studies. Thus, our meta-analysis excluded studies with a maximal frequency of GCNIS and MGCT. Thereby, our analyses of the literature underestimated the frequency of GCNIS and MGCT. The study did not carry out IHC in the TGCT. The protocol for the study did not plan for analyzing the relation between the distance from the precursor lesion to the TGCT and the frequency of the precursor lesions.
Conclusion
MGCT is a frequent intermediate precursor lesion.
Clinical Practice Points
•
The recent WHO classification of urologic cancer 2016 included 2 intermediate precursor lesions, intratubular SEM and ITEC, but not an invasive precursor named MGCT.
•
Precursor lesions are mainly found concomitant with TGCT.
•
The present consensus favors a fetal-origin hypothesis for TGCT.
•
However, studies show that most precursor lesions express a transcription factor OCT4.
•
OCT4 can interact with a panel of other transcription factors and an oncogene to induce a stem cell pattern in postfetal germ cells.
•
Induced in adult differentiated cells, the panel can cause reprogramming to a pluripotent stem cell pattern, like that in EC.
•
Testicular biopsies of men at risk, such as infertile men, might detect a precursor lesion in a seemingly nonmalignant testis.
•
If andrologists leave precursor lesions untreated, they most often develop into a macroscopically overt TGCT during the following 5 to 10 years.
•
Germ cell neoplasia in situ is cured with local radiotherapy. MGCT is treated with radical orchiectomy.
•
Precursor lesions do not recur after local treatment.
•
Thus, if possible, a precursor lesion should be treated when it is diagnosed.
•
A precursor lesion may be treated less aggressively than a macroscopically overt TGCT.
Disclosure
The authors have stated that they have no conflicts of interest.
Acknowledgments
The authors thank the Minister Erna Hamilton's Legacy for Science and Art, the family Hede Nielsen's foundation, and director Jacob Madsen and wife Olga Madsen's foundation for grants to support this work.
References
Oosterhuis J.W.
Looijenga L.H.
Testicular germ-cell tumours in a broader perspective.
Members of the ISUP Immunohistochemistry in Diagnostic Urologic Pathology Group
Best practices recommendations in the application of immunohistochemistry in testicular tumors: report from the International Society of Urological Pathology consensus conference.
30265-3&id=gr1.jpg){kind=link}
30265-3&id=gr2.jpg){kind=link}
30265-3&id=gr3.jpg){kind=link}
30265-3&id=gr4.jpg){kind=link}
30265-3&id=gr5.jpg){kind=link}
30265-3&id=gr6.jpg){kind=link}