Original Articles. The Significance of Lymphovascular Invasion of the Spermatic Cord in the Absence of Cord Soft Tissue Invasion

Original Articles The Significance of Lymphovascular Invasion of the Spermatic Cord in the Absence of Cord Soft Tissue Invasion Brandi C. McCleskey, MD; Jonathan I. Epstein, MD; Constantine Albany, MD; Neda Hashemi-Sadraei, MD; Muhammad T. Idrees, MD; Julie M. Jorns, MD; David Y. Lu, MD; Andres Matoso, MD; Soroush Rais-Bahrami, MD; Lauren E. Schwartz, MD; Thomas M. Ulbright, MD; Jennifer Gordetsky, MD Context. Testicular germ cell tumors with lymphovascular invasion (LVI) are staged pt2, and those with spermatic cord involvement are staged pt3. Objective. To study the clinical significance of LVI within the spermatic cord without direct involvement of the cord soft tissues. Design. A retrospective, multi-institutional review was performed on testicular GCTs with spermatic cord LVI in the absence of cord soft tissue invasion. Results. Forty-four germ cell tumors had LVI in the spermatic cord without soft tissue invasion; 37 of 44 patients (84%) had nonseminomatous germ cell tumors (NSGCT), and 7 (16%) had pure seminomas. Patients with NSGCTs and spermatic cord LVI had worse clinical outcomes compared with patients with pure seminoma and spermatic cord LVI (P ¼.008). We then compared patients with NSGCTs and spermatic cord LVI (n ¼ 37) to patients with NSGCTs and LVI limited to the testis (n ¼ 32). A significantly greater percentage of patients with LVI in the spermatic cord presented with advanced clinical stage (76% versus 50%; P ¼.01). There was no statistically significant difference in disease recurrence/progression or death between patients with spermatic cord LVI and patients with LVI limited to the testis (P ¼.40; P ¼.50). There was no significant difference in the presence of embryonal dominant histology (P ¼.30) or rete testis invasion (P ¼.50) between the 2 groups. More hilar soft tissue invasion was seen in patients with LVI present in the spermatic cord (P ¼.004). Conclusions. In patients with NSGCTs, LVI in the spermatic cord, without soft tissue invasion, is associated with worse clinical stage at presentation compared with patients with LVI confined to the testis. (Arch Pathol Lab Med. 2017;141:824 829; doi: 10.5858/ arpa.2016-0226-oa) In the United States, testicular cancer is the most common cancer in men between the ages of 15 and 40 years. 1 Although the incidence of testicular cancer has been increasing during the past century, because of advances in therapy, the prognosis for most patients is excellent. 1 Surgery, chemotherapy, and a close follow-up regimen all Accepted for publication September 23, 2016. Published as an Early Online Release March 31, 2017. From the Departments of Pathology (Drs McCleskey and Gordetsky), Urology (Drs Rais-Bahrami and Gordetsky), and Radiology (Dr Rais-Bahrami), University of Alabama, Birmingham; the Department of Pathology, University of Michigan, Ann Arbor (Dr Jorns); the Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, University of California, Los Angeles (Dr Lu); the Department of Pathology, Brown University, Providence, Rhode Island (Dr Matoso); the Department of Pathology, University of Pennsylvania, Philadelphia (Dr Schwartz); the Departments of Hematology/Oncology (Drs Albany and Hashemi-Sadraei) and Pathology (Drs Idrees and Ulbright), Indiana University School of Medicine, Indianapolis; and the Department of Pathology, The Johns Hopkins Hospital, Baltimore, Maryland (Dr. Epstein). The authors have no relevant financial interest in the products or companies described in this article. Reprints: Jennifer Gordetsky, MD, Department of Pathology and Urology, The University of Alabama, Birmingham, NP 3550, 1802 6th Ave S, Birmingham, AL 35249 (email: jgordetsky@uabmc.edu). have a major role in the management of patients with testicular germ cell tumors. However, overtreatment is both costly and potentially dangerous to patients. Pathologic staging of testicular germ cell tumors is crucial for appropriate patient management. The current American Joint Committee on Cancer staging guidelines define germ cell tumors with lymphovascular invasion (LVI) as pathologic stage pt2 and those with spermatic cord involvement, with or without LVI, as stage pt3. 2 In the absence of metastatic disease, pt2 tumors are considered clinical stage I, and they are either followed with active surveillance or treated with adjuvant chemotherapy. 3 Although overall cure rates approach 99% for testicular germ cell tumors, approximately one-third of nonseminomatous germ cell tumors (NSGCT) will present with advanced disease or relapse. 4 8 To decrease overtreatment, it is important to identify prognostic markers that are associated with increased risk of relapse in those patients. Risk stratification has been accomplished using a variety of clinical variables, such as serum marker elevation and pathologic features. The presence of LVI, a predominant histologic subtype; involvement of the rete testis; and pathologic tumor stage are generally regarded as most significant. 8 13 These prognostic markers help identify patients most at risk for disease relapse and aim to limit overtreatment by chemotherapy or 824 Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al

Lymphovascular invasion within the spermatic cord in the absence of spermatic cord soft tissue invasion (hematoxylin-eosin, original magnification 320). complications of retroperitoneal lymph node dissection. 4,9,14,15 The significance of spermatic cord LVI in the absence of cord soft tissue invasion has not been evaluated, to our knowledge, in testicular germ cell tumors. We reviewed the clinical and pathologic findings in patients with testicular germ cell tumors that presented with LVI of the spermatic cord in the absence of spermatic cord soft tissue invasion to investigate the prognostic significance of this finding. MATERIALS AND METHODS A retrospective, multi-institutional review was performed to identify cases of testicular germ cell tumors that contained LVI within the spermatic cord without soft tissue invasion of the cord (n ¼ 44). Additionally, a contemporary cohort of control patients (n ¼ 32) were identified who had NSGCT diagnosed as pt2 because of the presence of LVI confined to the testis without LVI present in the spermatic cord. Cases from each institution were identified and reviewed by a single pathologist from each respective institution. Specimens had 1 to 3 spermatic cord sections submitted, typically one from the midcord and one from the spermatic cord margin. Lymphovascular invasion was defined as tumor cells adherent to the luminal aspect of a vascular or lymphatic channel (Figure). The presence of fibrin among tumor cells was also used to help discriminate true LVI from artifact. Cases in which the LVI was indeterminate from artifact were excluded from our study. Histologic features evaluated included location of LVI (testis versus cord), LVI identified at spermatic cord margin, tumor size, tumor histology, tumor involvement of the rete testis, hilar soft tissue involvement by tumor, presence of germ cell neoplasia in situ, and presence of intratesticular scar. A chart review was performed to evaluate for the clinical stage at the time of presentation, clinical therapy after orchiectomy (radiation, chemotherapy, and/or retroperitoneal lymph node dissection), and presence of disease progression/recurrence within the duration of clinical follow-up. A 2-tailed Student t test was used to compare continuous variables and the Pearson v 2 test with Fisher exact modification was used, when warranted, to compare categoric values. A predetermined P value of.05 was set as the threshold of statistical significance. RESULTS Forty-four patients were diagnosed with testicular germ cell tumors with LVI in the spermatic cord without soft tissue invasion of the cord during the study period. The clinicopathologic features of the cohort are listed in Table 1. Mean clinical follow-up was 28 months (range, 1 150). Overall survival was 95% (42 of 44). For treatment, 33 of 44 patients (75%) received chemotherapy, 10 of 44 patients (23%) underwent retroperitoneal lymph node dissection, 3 Table 1. Clinical and Pathologic Characteristics of Patients With Lymphovascular Invasion (LVI) Present in the Spermatic Cord Characteristic Results Patients, No. (%) 44 (100) Age, y, mean (SD), range 30.6 (9.8), 14 58 Tumor size, cm, mean (SD), range 5.3 (2.5), 2.4 11.7 Follow-up, mo (range) 27.8 (1 150) Follow-up, mo, median 12 Reported pathologic tumor stage, No. (%) pt2 35 (80) pt3 3 (7) Not specified 6 (14) Clinical stage, No. (%) I 13 (30) II 14 (32) III 15 (34) Not specified 2 (4) LVI (other than in the spermatic cord), No. (%) Yes 36 (82) No 8 (18) Tumor type, No. (%) NSGCT 37 (84) SGCT 7 (16) Dominant tumor histology, NSGCT (n ¼ 37), No. (%) Embryonal 25 (68) Seminoma 5 (13) Yolk sac 4 (11) Teratoma 2 (5) Choriocarcinoma 1 (3) Rete testis involvement, No. (%) Yes 22 (50) No 22 (50) GCNIS present, No. (%) Yes 28 (64) No 16 (36) Hilar soft tissue involvement, No. (%) Yes 14 (32) No 30 (68) Scar present, No. (%) Yes 4 (9) No 40 (91) Abbreviations: GCNIS, germ cell neoplasia in situ; NSGCT, nonseminomatous germ cell tumor; SGCT, seminomatous germ cell tumor. of 44 patients (7%) underwent postorchiectomy radiation therapy, 1 patient (2%) refused postorchiectomy therapy and went on surveillance, and 5 of 44 patients (11%) had an unknown history of postorchiectomy therapy. Of the 44 patients, 37 (84%) were diagnosed with NSGCT, and 7 of 44 (16%) had pure seminomas (Table 2); 8 of 44 patients (18%) had LVI identified only in the spermatic cord, and 36 of 44 patients (82%) had LVI present both in the spermatic cord and testis. Of the 44 patients with LVI in the spermatic cord, 20 (45%) had LVI present at the spermatic cord margin. In addition, 35 of 44 cases (80%) with spermatic cord LVI were given a stage of pt2 in the original pathology report, 3 of 44 (7%) were staged as pt3, and 6 of 44 pathology reports (14%) did not provide a pathologic stage but, instead, provided a comment stating the uncertainty of the pathologic stage given the findings. Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al 825

Table 2. Characteristics of Seminomas and Nonseminomatous Germ Cell Tumors (NSGCTs) With Lymphovascular Invasion (LVI) of the Spermatic Cord Characteristic NSGCT SGCT P Value Patients, No. (%) 37 (84) 7 (16)... Age, y, mean (SD), range 31.1 (9.8), 14 58 27.9 (7.9), 16 39.40 Tumor size, cm, mean (SD), range 5.2 (2.6), 2.4 11.7 6.0 (1.7), 3.9 8.3.30 Follow-up, mo, mean (range) 25.0 (1 117) 41.3 (3 150).30 Follow-up, mo, median 11.5 15... Clinical stage, No. (%).001 I 7 (19) 6 (86) II 13 (35) 1 (14) III 15 (41) 0 (0) Not specified 2 (5) 0 (0) LVI (other than in cord), No. (%).06 Yes 32 (86) 4 (57) No 5 (14) 3 (43) Rete testis involvement, No. (%).20 Yes 17 (46) 5 (71) No 20 (54) 2 (29) Hilar soft tissue involvement, No. (%).50 Yes 11 (30) 3 (43) No 26 (70) 4 (57) GCNIS, No. (%).60 Yes 23 (62) 5 (71) No 14 (38) 2 (29) Scar in testis, No. (%).10 Yes 3 (8) 2 (29) No 34 (92) 5 (71) Any disease recurrence/progression, No. (%).008 Yes 12 (32) 0 (0) No 16 (44) 7 (100) N/A 9 (24) 0 (0) Recurrence/progression after chemotherapy/radiation, No. (%).02 Yes 8 (22) 0 (0) No 13 (35) 6 (86) N/A 16 (43) 1 (14) Death, No. (%).40 Yes 2 (5) 0 (0) No 35 (95) 7 (100) Abbreviations: GCNIS, germ cell neoplasia in situ; N/A, not applicable; SGCT, seminomatous germ cell tumor. Of the 37 patients with NSGCT and spermatic cord LVI, 34 (92%) had tumors of mixed histology, 2 (5%) had pure embryonal carcinoma, and 1 (3%) had pure yolk sac. Embryonal carcinoma was the dominant histology in 25 of 37 tumors (68%). Of the 37 patients with NSGCT, 17 (46%) had involvement of the rete testis, 11 (30%) had involvement of the hilar soft tissues, and 18 (49%) had LVI present at the spermatic cord margin. Some patients had features present in more than one category. Of 37 patients with NSGCT, 28 (76%) presented with an advanced clinical stage and evidence of retroperitoneal lymph node, lung, or liver metastasis. In addition, 12 of 37 patients (32%) had some form of disease progression/recurrence, 8 patients (22%) had disease progression/recurrence after chemotherapy, and 2 patients died (5%). Sixteen patients (44%) had no evidence of disease recurrence/progression, and 9 patients (24%) did not have information available on disease recurrence. Both patients who died presented with clinical stage III metastatic tumors. Both patients had mixed germ cell tumors, with one tumor having choriocarcinoma as the predominant histology and the other having embryonal. Four of the 7 patients (57%) considered to be clinical stage I at the time of orchiectomy had disease progression/ recurrence (Table 2). Of the 7 patients with pure seminoma and spermatic cord LVI, 5 (71%) had rete testis invasion, 3 (43%) had involvement of the hilar soft tissues, and 2 (29%) had LVI present at the spermatic cord margin. Some patients had features present in more than one category. One of the 7 patients (14%) presented with metastatic disease, and none of the 7 patients had disease progression/recurrence or died (Table 2). In patients with spermatic cord LVI, NSGCTs were more likely to have some type of disease recurrence/progression and disease recurrence/progression after chemotherapy/ radiation compared with patients with pure seminoma (P ¼.008; P ¼.02). In addition, patients with NSGCT tended to present at a higher clinical stage (P ¼.001). Compared with pure seminomas, patients with NSGCT had higher cancerspecific mortality rates (5% versus 0%); however that finding did not reach statistical significance (P ¼.40). There was no difference between the 2 groups in tumor size, age at presentation, clinical follow up, rete testis invasion, or hilar soft tissue invasion (Table 2). 826 Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al

Table 3. Comparison of Patients With Nonseminomatous Germ Cell Tumors (NSGCTs) and Spermatic Cord Lymphovascular Invasion (LVI) Versus Patients With NSGCTs and LVI Confined to the Testis Characteristic Spermatic Cord LVI LVI in Testis Only P Value Patients, No. (%) 37 (100) 32 (100) Age, y, mean (SD), range 31.1 (10.1), 14 58 33.1 (11.8), 16 64.46 Tumor size, cm, mean (SD), range 5.2 (2.7), 2.8 11.7 3.5 (2.0), 1.0 8.0.008 Follow-up, mo, mean (range) 25.0 (1 117) 59.4 (3 180),.001 Follow-up, mo, median 11.5 46.5... Clinical stage, No. (%).01 I 7 (19) 16 (50) II 13 (35) 11 (34) III 15 (41) 5 (16) Not specified 2 (5) 0 (0) Dominant tumor histology, No. (%).30 Embryonal 25 (68) 18 (56) Nonembryonal 12 (32) 14 (44) Rete testis involvement, No. (%).50 Yes 17 (46) 12 (38) No 20 (54) 20 (62) Hilar soft tissue involvement, No. (%).004 Yes 11 (30) 1 (3) No 26 (70) 31 (97) Any disease recurrence/progression, No. (%).40 Yes 12 (32) 17 (53) No 16 (44) 14 (44) N/A 9 (24) 1 (3) Recurrence/progression after chemotherapy, No. (%).90 Yes 8 (22) 9 (28) No 13 (35) 16 (50) N/A 16 (43) 7 (22) Death, No. (%).50 Yes 2 (5) 3 (9) No 35 (95) 29 (91) Abbreviation: N/A ¼ not applicable. We compared patients with NSGCT and LVI in the spermatic cord (n ¼ 37) to patients with NSGCTs and LVI limited to the testis (n ¼ 32). There were no significant differences between the 2 groups regarding patient age at presentation, rete testis involvement, or presence of embryonal carcinoma as the dominant histology (Table 3). Longer clinical follow-up was available for patients with LVI confined to the testis. Patients with LVI present in the spermatic cord had larger tumors (P ¼.008). More patients with spermatic cord LVI had hilar soft tissue involvement (P ¼.004). A greater percentage of patients with LVI in the spermatic cord presented with advanced clinical stage disease (28 of 37 [76%] versus 16 of 32 [50%]; P ¼.01). Despite this finding, there was no significant difference in the finding of any disease recurrence/progression or disease recurrence/progression after chemotherapy (P ¼.40; P ¼.90, respectively). In addition, there was no difference in mortality between the 2 groups (P ¼.50). Of the patients who presented as clinical stage 1, 4 of 7 patients (57%) with spermatic cord LVI and 11 of 16 patients (69%) with LVI confined to the testis went on to have disease recurrence/ progression. DISCUSSION Testicular tumors have specific histologic features, which have been shown to be significant indicators of poor prognosis and risk of metastases. 8,9,11,12,16,17 The importance of hilar soft tissue invasion, rete testis invasion, vascular invasion, percentage of histologic components, and tumor size has each been well reported in the literature. 4,8,10,16 18 Valdevenito et al 16 described LVI of the spermatic cord as a significant risk factor for metastasis in seminomas (odds ratio [OR], 4.968; P ¼.02), but it did not maintain significance in multivariate analysis (OR, 1.924, P ¼.50). In addition, the authors did not specify whether the LVI in the spermatic cord was present in the absence of soft tissue invasion of the cord. 16 To our knowledge, the current study is the first to evaluate the significance of spermatic cord lymphovascular invasion without cord soft tissue invasion in testicular germ cell tumors. The current American Joint Committee on Cancer staging guidelines define germ cell tumors with LVI as stage pt2 and those with spermatic cord involvement, with or without LVI, as stage pt3. 2 The presence of LVI within the spermatic cord without direct involvement of the spermatic cord soft tissues is not currently addressed by the American Joint Committee on Cancer staging system. 19 The College of American Pathologists recommends staging testicular tumors with this histologic feature as pt2. 20 However, there is currently no consensus on how to stage this particular histologic finding. In a recent survey conducted by Berney et al, 21 agreement among pathologists varied as to how to stage this finding. 21 Nearly 80% of experts and members of the European Network of Uropathology staged these tumors as pt2 and approximately 20% as pt3. When looking at the original pathology reports, we saw a similar Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al 827

disagreement in our series. Although most of our cases were staged as pt2, 20% (9 of 44) were considered either pt3 or were unclear on which stage best represented the pathologic findings. In our study, we found that patients with NSGCTs and spermatic cord LVI presented with higher clinical stage diseasethandidthosewithlviconfinedtothetestis(p¼.01). Overall, 28 of 37 patients (76%) with NSGCTs and LVI of the spermatic cord presented with advanced clinical stage disease (clinical stage II or III), which is higher than the expected 46% for patients with NSGCT in general. 10 In addition, there was a high rate of any disease recurrence/progression in both groups, with 12 of 37 patients (32%) with spermatic cord LVI and 17 of 32 patients (53%) with LVI confined to the testis having disease recurrence/progression. There was also a high rate of disease recurrence/progression after chemotherapy: 8 of 37 (22%) for the patients with spermatic cord LVI, and 9 of 32 (28%) for the patients with LVI confined to the testis. In the high-risk population, which included patients with LVI, the literature reports a 0% to 5% risk of relapse after chemotherapy. 1 In addition, of the patients presenting with clinical stage I NSGCTs, 4 of 7 (57%) with spermatic cord LVI and 11 of 16 (69%) with LVI limited to the testis experienced any disease progression/recurrence. This is higher than the expected rate of 15% to 30% reported in the literature for clinical stage 1 patients 1,7,10 and is equal to that seen in the highrisk group. 7,14,15 For patients with NSGCTs, there were no statistically significant differences in presence of any disease recurrence/progression, disease recurrence/progression after chemotherapy, or survival when comparing LVI in the spermatic cord to LVI confined to the testis. However, spermatic cord LVI in the absence of soft tissue invasion was associated with advanced clinical stage disease at presentation compared with LVI confined to the testis. One could argue that the pathologic stage should reflect the associated clinical stage at presentation, and as such, spermatic cord LVI should be considered pt3. However, additional research is required before a formal pathologic staging recommendation can be made. In our study, LVI of the spermatic cord in the absence of soft tissue invasion of the cord did not appear to be a significant finding in pure seminomas. Only one patient presented with advanced clinical stage disease, there were no reported incidences of disease progression/recurrence, and patients had a 100% survival rate. This finding is not unexpected, given that other studies have shown LVI to be a less-important prognostic indicator in pure seminomas. 6,16 In addition, pure seminomas overall have a much better prognosis than NSGCT. 6,7 Our findings supported that concept in that patients with spermatic cord LVI had worse clinical outcomes in NSGCT compared with those with pure seminoma. This was demonstrated by differences in advanced clinical stage at presentation (P ¼.001), any disease recurrence/progression (P ¼.008), and disease recurrence/progression after chemotherapy/radiation (P =.02). Seminoma is also a friable germ cell tumor that often produces artifactual LVI. It is possible that the difficulties in identifying true LVI in seminomas have a role in the uncertain meaning of LVI in their prognoses. We followed strict criteria to define LVI in the spermatic cord, to help exclude the possibility that our cases were misplaced tumor. Finally, given that seminomas have an almost absolute response to chemotherapy and radiation, assessing the prognosis of LVI of the spermatic cord without cord soft tissue invasion is difficult in pure seminomas. In our cohort of patients with spermatic cord LVI without cord soft tissue invasion, 8 of 44 (18%) had LVI identified only in the spermatic cord. Adequately sectioning the spermatic cord to evaluate for the presence of LVI is critical, especially when considering this may be the only area of detection. A patient with organ-confined disease and no LVI would be considered at low risk, with a relapse rate of 10% to 20% and could potentially undergo surveillance. 1,7,22 In our study, 4 of 5 patients (80%) with NSGCTs and LVI found only in the spermatic cord had some type of disease relapse or progression. One of these patients did not receive immediate adjuvant therapy and experienced metastases 8 months later while on active surveillance. We were not able to evaluate which component of the mixed germ cell tumor was involved in the vascular invasion within the spermatic cord for all cases. However, of the cases that were able to be evaluated, embryonal carcinoma was the most common morphology found within vascular channels, which has been reported to be the case in other studies. 21 Currently, there is no evidence that reporting which subtype of tumor involved in vascular invasion has any clinical significance. Our study is limited by being a retrospective review of a relatively rare pathologic finding with more clinical followup available in patients with NSGCTs and LVI confined to the testis. Serum tumor markers were also not available for analysis. In addition, because of the limited number of patients, we were unable to compare our patient population in a meaningful manner to a cohort of patients with pt3 tumors. Although the historic data often combined pt2 and pt3 tumors in outcomes studies, one study showed that up to 97% of patients with NSGCTs and spermatic cord soft tissue invasion (pt3) presented with metastatic disease. 23 25 Future studies would include attempts to evaluate pt3 tumors as well as evaluating a larger group of patients presenting in clinical stage I with lymphovascular invasion of the spermatic cord in the absence of cord soft tissue invasion. Again, we were limited by the rarity of patients within each desired cohort. Despite our limited number of patients, our cohort of patients with NSGCTs and spermatic cord LVI was well matched to patients with NSGCT and LVI confined to the testis. To our knowledge, this remains the first study to evaluate LVI of the spermatic cord in the absence of cord soft tissue invasion, which is currently a debated histologic finding in the field of uropathology. CONCLUSIONS In testicular cancer, high-risk pathologic features can help identify patients who are more likely to fail surveillance and may benefit from upfront adjuvant therapy. Our study demonstrated that in patients with NSGCTs, LVI of the spermatic cord, in the absence of cord soft tissue invasion, is associated with advanced clinical stage disease compared with LVI confined to the testis. Although the location of LVI does not appear to affect clinical outcome in disease progression/recurrence or survival, this is likely due to the excellent clinical outcome seen overall in patients with testis tumors. Further studies are required before a formal pathologic staging recommendation can be made. References 1. Beard CJ, Gupta S, Motzer RJ, et al. Follow-up management of patients with testicular cancer: a multidisciplinary consensus-based approach. J Natl Compr Canc Netw. 2015;13(6):811 822. 828 Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al

2. Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti AI, eds. AJCC Cancer Staging Manual. Vol. 649. 7th ed. New York, NY: Springer; 2010. 3. Horwich A, Nicol D, Huddart R. Testicular germ cell tumours. BMJ. 2013; 347:f5526. 4. Divrik RT, Akdoğan B, Ozen H, Zorlu F. Outcomes of surveillance protocol of clinical stage I nonseminomatous germ cell tumors-is shift to risk adapted policy justified? J Urol. 2006;176(4, pt 1):1424 1429; discussion 1429 1430. 5. Hanna NH, Einhorn LH. Testicular cancer discoveries and updates [published correction appears in N Engl J Med. 2014;371(24):2342]. N Engl J Med. 2014;371(21):2005 2016. 6. Kollmannsberger C, Tandstad T, Bedard PL, et al. Patterns of relapse in patients with clinical stage I testicular cancer managed with active surveillance. J Clin Oncol. 2015;33(1):51 57. 7. Motzer RJ, Jonasch E, Agarwal N, et al. Testicular cancer, Version 2.2015. J Natl Compr Canc Netw. 2015;13(6):772 799. 8. Yilmaz A, Cheng T, Zhang J, Trpkov K. Testicular hilum and vascular invasion predict advanced clinical stage in nonseminomatous germ cell tumors. Mod Pathol. 2013;26(4):579 586. 9. Williams SB, Kacker R, Winston D, Bahnson E, Steele GS, Richie JP. Predictors of positive retroperitoneal lymph nodes in patients with high risk testicular cancer. J Urol. 2011;186(6):2245 2248. 10. Lago-Hernandez CA, Feldman H, O Donnell E, et al. A refined risk stratification scheme for clinical stage 1 NSGCT based on evaluation of both embryonal predominance and lymphovascular invasion. Ann Oncol. 2015;26(7): 1396 1401. 11. Heidenreich A, Sesterhenn IA, Mostofi FK, Moul JW. Prognostic risk factors that identify patients with clinical stage I nonseminomatous germ cell tumors at low risk and high risk for metastasis. Cancer. 1998;83(5):1002 1011. 12. Javadpour N, Canning DA, O Connell KJ, Young JD. Predictors of recurrent clinical stage I nonseminomatous testicular cancer: a prospective clinicopathologic study. Urology. 1986;27(6):508 511. 13. Li X, Guo S, Wu Z, et al. Surveillance for patients with clinical stage I nonseminomatous testicular germ cell tumors. World J Urol. 2015;33(9):1351 1357. 14. Nichols CR, Roth B, Albers P, et al. Active surveillance is the preferred approach to clinical stage I testicular cancer [comment in J Clin Oncol. 2013; 31(28): 3477 3479]. J Clin Oncol. 2013;31(28):3490 3493. 15. Kollmannsberger C, Moore C, Chi KN, et al. Non risk-adapted surveillance for patients with stage I nonseminomatous testicular germ-cell tumors: diminishing treatment-related morbidity while maintaining efficacy. Ann Oncol. 2010;21(6):1296 1301. 16. Valdevenito JP, Gallegos I, Fernández C, Acevedo C, Palma R. Correlation between primary tumor pathologic features and presence of clinical metastasis at diagnosis of testicular seminoma. Urology. 2007;70(4):777 780. 17. Moul JW, Heidenreich A. Prognostic factors in low-stage nonseminomatous testicular cancer. Oncology (Williston Park). 1996;10(9):1359 68, 74; discussion 77 78. 18. Berney DM. Staging and classification of testicular tumours: pitfalls from macroscopy to diagnosis. J Clin Pathol. 2008;61(1):20 24. 19. Osunkoya AO, Grignon DJ. Practical issues and pitfalls in staging tumors of the genitourinary tract. Semin Diagn Pathol. 2012;29(3):154 166. 20. Ulbright TM. Protocol for the examination of specimens from patients with malignant germ cell and sex cord-stromal tumors of the testis, exclusive of paratesticular malignancies: a basis for checklists. Cancer Committee, College of American Pathologists. Arch Pathol Lab Med. 1999;123(1):14 19. 21. Berney DM, Algaba F, Amin M, et al. Handling and reporting of orchidectomy specimens with testicular cancer: areas of consensus and variation among 25 experts and 225 European pathologists. Histopathology. 2015;67(3): 313 324. 22. Curreri SA, Fung C, Beard CJ. Secondary malignant neoplasms in testicular cancer survivors. Urol Oncol. 2015;33(9):392 398. 23. Rodriguez PN, Hafez GR, Messing EM. Nonseminomatous germ cell tumor of the testicle: does extensive staging of the primary tumor predict the likelihood of metastatic disease? J Urol. 1986;136(3):604 608. 24. Fung CY, Kalish LA, Brodsky GL, Richie JP, Garnick MB. Stage I nonseminomatous germ cell testicular tumor: prediction of metastatic potential by primary histopathology. J Clin Oncol. 1988;6(9):1467 1473. 25. Raghavan D, Peckham MJ, Heyderman E, Tobias JS, Austin DE. Prognostic factors in clinical stage I non-seminomatous germ-cell tumours of the testis. Br J Cancer. 1982;45(2):167 173. Arch Pathol Lab Med Vol 141, June 2017 Spermatic Cord Lymphovascular Invasion McCleskey et al 829