Long-term Clinical Outcome in Patients with Stage-I Nonseminomatous Germ Cell Cancer . A Critical Review of Own Treatment Modalities in a Retrospective Study

Purpose: The optimal management of patients with clinical stage I non-seminomatous germ cell testicular cancer (NSGCT I) was considered controversial until the European Germ Cell Cancer Consensus Group determined unambiguous treatment strategies. In order to assess the long-term outcome we evaluated the data of patients with NSGCT I. Materials and Methods: In a retrospective evaluation, we included 52 patients with a mean age of 26 years (range 15-58) who were treated with different modalities at our department between 1989 and 2003. Mean follow-up was 5.9 years (range 2-14 years). After orchiectomy, 39 patients were treated with chemotherapy, 7 patients underwent retroperitoneal lymph node dissection and 6 men were managed using a surveillance strategy. Survival, recurrence rate and time of recurrence were evaluated. The histological staging and treatment modality was related to the relapse. Results: Tumor specific overall mortality was 3.8%. The mortality and relapse rate of the surveillance strategy, retroperitoneal lymph node dissection and chemotherapy was 16.7% / 50%, 14.3% / 14.3% and 0% / 2.5% respectively. All relapsed patients in the surveillance group as well as in the RPLND group had at least one risk factor for developing metastatic disease. Conclusions: Following the European consensus on diagnosis and treatment of germ cell cancer in patients with NSGCT Stage I any treatment decision must be individually related to the patient according to prognostic factors and care capacity of the treating centre. In case of doubt, adjuvant chemotherapy should be the treatment of choice, as it provides the lowest risk of relapse or tumor related death.


INTRODUCTION
The incidence of testicular cancer has increased over the last 50 years and is the most common malignancy in men in the 15-35 year age group.Nonseminomatous germ cell cancer occurs in slightly younger patients than in those with seminomas.Stage 1 disease is treated initially by orchiectomy, Clinical Urolog� Clinical Urolog� which assures accurate histological diagnosis.The importance of this pathological staging is reflected in the decision for the adjuvant treatment modalities.Although standardized recommendations for follow-up are not defined, patients without increased relapse risk such as vascular or lymphatic invasion, predominant component of embryonal carcinoma and undifferentiated elements (1), are recommended for active surveillance.The relapse rate in this treatment strategy is approximately 30 %. Metastases will occur in the retroperitoneum in 54-78% and in the lung in 13-31% of the relapsed patients (2) and can be salvaged with cisplatin-based chemotherapy protocols (3).Following the recommendations of the European Germ Cell Cancer Consensus Group, in patients with reservations against the surveillance strategy, needing a high rate of compliance, adjuvant chemotherapy is the treatment of choice (relapse rate 3%).In case of reservations against the two afore mentioned options, nerve sparing retroperitoneal lymph node dissection (NS-RPLND) is suggested (3).
Vascular invasion is the most important prognostic indicator for developing metastatic disease in up to 48%.Patients with risk factors should be given two cycles of BEP (standard dose of cisplatin, etoposide, bleomycin) (3).However, cure rates of about 99% can be reached in patients with clinical stage I non-seminomatous germ cell testicular cancer (NS-GCT I) with or without risk factors, independently of the treatment strategy.
Any decision for the optimal adjuvant treatment modality for non-seminomatous Stage 1 needs close co-operation between physician and patient to make sure, that the patient's compliance is in line with the chosen therapy.If a patient cannot deal with the psychological distress of a recurrence rate of approximately 30% (low risk) to 58% (high risk) (4) or if the compliance for regular follow-up intervals must be questioned, an adjuvant treatment should be preferred instead of the surveillance strategy.
In this study we retrospectively reviewed the patients with NSGCT I treated at our urological department from 1989 to 2003 to evaluate the long-term outcome.The results were compared with those obtained from other studies in the literature.Preferentially, we offered adjuvant chemotherapy with excellent cure rates, accepting an overtreatment in selected patients.The aim of this study was to critically review the applied treatment strategies with special emphasize on the relapsed patients.Data acquisition -The records were reviewed for histological classification, clinical and pathological staging, serum tumor markers, adjuvant therapy and last follow-up.

Patients
Inclusion criteria -All patients had a histologically proven non-seminomatous germ cell cancer.Forty-three of the evaluated patients had a pT1 tumor, in 8 patients the histological work-up showed a pT2 stage and 1 patient had a pT3 tumor.Staging evaluations (chest X-ray, pre-and postoperative serum tumor markers, abdominal CT scan) excluded metastatic disease.
Therapy and follow-up -All patients had undergone radical orchiectomy.Following orchiectomy 39 patients were treated with chemotherapy of two (n = 29) or three (n = 5) courses of bleomycine, etoposide and cisplatin (BEP), or, in case of prior lung problems ifosfamide instead of bleomycin (n = 5).Seven patients underwent retroperitoneal lymph node dissection and 6 men were managed in a surveillance strategy.Follow-up evaluations included physical examination, chest radiographs, serum tumor markers, abdominal and testicular ultrasound and abdominal CT scan periodically as seen in Table -1.Patients were encouraged to be followed for at least 10 years.
Data evaluation -All patients with a documented follow-up of at least two years were included.Three men were lost to follow-up.The data of 52 patients could be evaluated based on a follow-up until June 2008.Evaluation included survival, recurrencerate and time of recurrence.The previous histological staging and treatment modality was related to the relapse.

RESULTS
Up to the evaluation date three patients had died.One of these developed gastric cancer and died 9 years after treatment for non-seminomatous germ cell cancer.The initial histological workup showed a pT1 tumor and the patient was treated by retroperitoneal lymph node dissection (RPLND) after radical orchiectomy.
As regards RPLND -Six patients from the RPLND group (n = 7) were disease free and well at the time of evaluation.Five of them had a pT1 and the other a pT2 tumor.The relapsed patient who had a pT1 tumor with teratoma and embryonal carcinoma was initially treated by orchiectomy.Subsequently an adjuvant modified RPLND was performed.Histological work-up did not show any pathologic lymph nodes.Twelve months after the diagnosis a recurrence occurred revealed by serum tumor markers.Initially, the patient refused further imaging and therapy.One year later, he had a CT scan showing a compression of the vena cava with a large retroperitoneal mass.Chemotherapeutic treatment (cisplatin, etoposide, ifosfamid) was started and a secondary RPLND was performed after completing this therapy.The histological workup revealed mature teratoma.Six months later elevated serum tumor markers again indicated tumor recurrence again.Because the CT scan did not show any pathology, biopsies of the remaining testis were performed that histologically showed only atrophic parenchyma without malignancy.However, another 2 months later the abdominal CT scan showed multiple liver metastases.A high dose chemotherapy (POMP-ACE: prednisone, vincristine, methotrexate, mercaptopurine, adriamycin, cyclophosphamide, etoposide) followed but tumor mass could not be downsized.The patient refused any further therapy and died of his disease a few months later.
Surveillance -Six of our patients entered a surveillance protocol.Three of them relapsed.Histological evaluation after orchiectomy had shown a pT 2 tumor with seminomatous and embryonal cell components in one of the relapsed patients.Despite regular follow-up, two years after primary diagnosis, retroperitoneal recurrence of the tumor was detected when the patient complained of flank pain and weight loss.Tumor markers were increased.At first, the patient refused any further imaging and therapy.After one year, the CT scan showed multiple metastases in the retroperitoneum and upper abdomen.Due to renal insuffiency a carboplatin (instead of cisplatin) based chemotherapy was initiated.Nevertheless, a few days later the patient died of complications caused by the chemotherapy with renal failure and tumor lysis syndrome.
Relapse of the tumor was recorded in another two patients (Table -2).The first patient had been managed in a surveillance strategy after orchiectomy for a pT1 tumor with components of embryonal carcinoma (predominant) and seminoma that relapsed at 11 years after initial treatment.Serum tumor markers were prominent in the follow-up.CT scan showed enlarged para-aortic lymph nodes.After chemotherapy treatment with cisplatin, etoposide and bleomycin he is disease-free up to now at a follow-up of 6 years since diagnosis.
The second patient presented with a large bulk of interaortocaval lymph node metastases and elevated serum tumor markers (AFP and ß-HCG) one year after primary diagnosis of a nonseminomatous germ cell tumor with components of yolk sac tumor and predominant embryonal carcinoma.The relapse was treated with three courses of bleomycin, etoposide, and cisplatin (BEP) and residual masses were removed by secondary RPLND.Histological examination revealed mature teratoma tissue.Seven years postoperatively the patient remained disease free.Chemotherapy -Thirty-nine patients underwent adjuvant chemotherapy after primary orchiectomy.One of them relapsed.
The contralateral biopsy of the patient performed at the time of orchiectomy of a pT1 tumor showed intratubular germ cell neoplasia (TIN).Radiation treatment of the remaining testis with a total dose of 20 Gy followed.One year later another biopsy of the testis did not show any malignancy.Another year later, serum tumor marker increased.The CT scan did not detect any pathologically enlarged lymph nodes.A further biopsy of the testis followed.The histological workup again showed a TIN.Assuming that it would be a generalized problem rather than a local tumor growth, the patient was treated with two courses of chemotherapy (BEP) and we included him in the recurrence group.He remained disease free at a follow-up of 5 years (Table-2, Pt. 5).Taken together, the incidence of TIN of the contralateral testis in our group of patients was 3% (1 of 33 biopsies, others refused biopsy or orchiectomy was performed in an external hospital without obtaining a biopsy).Our patient's cohort included 20 patients with one or more risk factors (embryonal carcinoma and/or vascular invasion) and 19 without.

COMMENTS
The incidence of testicular cancer has been increasing in recent years (5).The optimal management of these patients was considered controversial until the European Germ Cell Cancer Consensus Group primarily established clearly defined diagnostic and therapeutic strategies in 2004 and then updated in 2008 (3).Recommendations for active surveillance in patients with low recurrence risk (without evidence of vascular invasion, a predominant component of embryonal cell carcinoma or undifferentiated element) are uniformly accepted as long as the patients compliance is in line with the repeated diagnostic testing to detect relapses at an early stage (3,6).However, for patients who cannot manage the psychological distress of recurrence rates between 14% and 22 % (1) or those not candidates for surveillance for other reasons the adjuvant management remains controversial.Chemotherapy or retroperitoneal lymph node dissection (RPLND) are possible options.There is no consensus about, which strategy should be preferred.Krege et al. suggest chemotherapy with two cycles of BEP (3), whereas Stephenson and Sheinfeld prefer RPLND in these patients (6).In cases with a high risk of recurrence the same recommendation dilemma exists.Because of relapse rates up to 50%, most authors suggest an adjuvant treatment (7).However, some authors propose active surveillance even in this patients group (8).
The major advantage of the surveillance strategy is that up to 86% of the patients do not need any further treatment.Furthermore, relapses can be cured in nearly 100% of cases.However, the problem might be the patient's compliance and the psychological distress of the recurrence rates with a more intensive chemotherapy in case of relapse.A strict follow-up scheme and a compliant patient are mandatory otherwise an adjuvant treatment has to be recommended.
Advantages of RPLND over chemotherapy are the surgical removal of chemoresistant teratoma, as its biological potential is unpredictable, and, furthermore, the lower long-term toxicity (6).Relapses can be cured with chemotherapy in nearly all cases.Otherwise, patients will be exposed to surgery-associated side effects.The retrograde ejaculation with consecutive potential infertility based on surgical damage of the postganglionic sympathetic fibers (Th 12-L 3) forming the hypogastric plexus near the aortic bifurcation is an essential problem for young patients.However, even selective RPLND has significantly reduced but not eliminated ejaculatory problems (1).
The major disadvantage of adjuvant chemotherapy is potential overtreatment in up to 70% of unselected Stage I patients.Short-term side effects (Nausea, vomiting) can be managed with potent clinical agents and leucocytopeny and thrombozytopeny are usually mild.Long-term side effects are a possibly decreased fertility and the development of secondary malignancies, as seen in high dose chemotherapy with etoposide (9).Concerning fertility, it has to be considered that in patients with malignant germ cell tumors, semen quality of the unaffected contralateral testes is significantly worse than in the healthy male (10) even before any chemotherapeutic treatment was applied.Furthermore, whether spermatogenesis is affected irreversibly by chemotherapy is determined by the cumulative dose of cisplatin.Pont and co-workers point out that the dose of even four courses of BEP is unlikely to cause any irreversible damage as the cisplatin dose generally remains below the critical dose of 400 mg/m 2 (11).Secondary malignancies are potentially caused by etoposide.The risk should be low because the critical dose is 1500-2000 mg/m 2 and the applied dose in adjuvant BEP will generally remain below (12).Furthermore, platin based therapy increases the risk of cardiovascular events (13).The above-mentioned studies only included patients with three or more courses and late effects seem to be dose dependant.
Considering these various factors, independently of the therapeutic regimen cure rates up to 99% (3) can be reached in patients with NSGCC.The decision about optimal adjuvant treatment after orchiectomy has to include risk factors as well as the patient's wishes and psychological situation and includes surveillance, adjuvant chemotherapy and RPLND.
To evaluate the outcome depending on the treatment strategies and risk factors of our own patients with NSGCT Stage I we retrospectively evaluated patients treated in our department between 1989 and 2003.Our preferred adjuvant modality was a chemotherapeutical treatment with two courses of BEP or ifosfamide instead of bleomycin in case of prior lung problems, accepting a potentially overtreatment in up to 50% of the cases but with very low recurrence rates (3).The minimal follow-up of the patients was defined as at least 2 years because the majority of relapses will occur within this period (14,15).
With respect to the low patient's number and the inhomogeneous group the cancer related mortality of the included 52 patients was 4% and comparable with the data obtained from the literature (16).All the patients relapsed in the retroperitoneum, independently from the chosen adjuvant modality.

Surveillance Group
Three out of six patients managed with a surveillance strategy after orchiectomy relapsed.Data published in the literature ranged from 30 -75% (3,14,17) depending on risk factors.Whether adjuvant therapy was not recommended, or the patients primarily decided to active surveillance, or the concerned patients refused any further therapy, remains unclear from the retrospective evaluation of the records.However, all of our relapsed patients had one or two risk factors for developing metastatic disease, questioning the chosen strategy in these patients retrospectively.In two cases embryonal carcinoma was the dominating histological feature in the ablated testes in addition to vascular invasion of the tumor, including the patients retrospectively in the "high risk" group.The other patient did not have any vascular invasion but primarily rather embryonal carcinoma in the resected tumor.Time to recurrence was 1, 2 and 11 years, respectively.There was one tumor related death in these 6 patients on surveillance.At the point of recurrence, the respective patient foremost refused any further diagnosing and therapy (Table-2, Pt. 2), presuming that he did not show appropriate compliance to the primarily chosen surveillance strategy.In previous studies the rate was between 1, 2 and 2.8% (8,17).The high value in our group might be explained by the low number of patients included in this evaluation.After initiating an intensified chemotherapy one year after the metastatic spread the patient died because of chemotherapeutic induced side effects.
From the patients under surveillance which did not relapse one was "high risk" (embryonal carcinoma) and the other two were "low risk" cases (seminoma and embryonal carcinoma and seminoma, yolk sac tumor and embryonal carcinoma respectively, both with seminoma being the predominant component).

RPLND
Even though it is uncommon, one of our patients treated with RPLND after primary orchiectomy relapsed.Unfortunately, the patient refused any further imaging and therapy after suspicion of recurrence and the large retroperitoneal bulk made it difficult to evaluate the site of relapse.Therefore, it was impossible to assess whether it was an infield recurrence with lymphatic tissue left behind during the primary RPLND or an outside the border of the primary RPLND recurrence, as has also been described by other groups (18).The relapse rate after RPLND reported in the literature was between 5.8 -21% (18,19).As seen from Table-2 (Pt. 1) the affected patient, although showing a stage pT1, belonged to the high-risk group as the tumor showed embryonal carcinoma as predominating histological component.Although intensified chemotherapeutic treatment was initiated, the patient died of tumor progression.

Chemotherapy
One of the 39 patients (2.5%) treated with adjuvant chemotherapy relapsed.He suffered from a pT1 teratocarcinoma.Therefore, he initially did not have any risk factors for developing metastatic disease.Additionally, the histological workup showed a TIN of the contralateral testis.He was treated with two courses BEP and a radiation therapy was applied to the remaining testis.Two years later tumor marker increased but imaging did not show enlarged lymph nodes.Another therapy with two courses BEP was given.The patient has remained disease free to date.Comparably, formerly published studies showed a mortality of patients with NSGCT I treated with CTX between 0 and 4% (9,20,21).Relapse rates between 0 and 7% were reported (9,20,21).All these studies administered chemotherapeutic treatment only in high risk patients.Our patient cohort included 20 patients with one or more risk factors (embryonal carcinoma and/or vascular invasion) and 19 without.Retrospectively, most of the patients of the low-risk group were overtreated.
Limitations of the study -The indication for the particularly chosen treatment strategy could not be determined from the retrospectively assessed data.The reason for the inhomogeneous groups remains unclear and results in small numbers of patients especially in the surveillance and RPLND group.Small numbers can affect the reliability and confidentiality of results.
Rate is likely to be imprecise and the comparability of percentages is limited.Nevertheless, we attempted to include the data in this context and tied to discuss the results with respect to the low patient's number.

CONCLUSION AND CRITICAL REVIEW
The results of the evaluation of 52 own patients with NSGCT I treated with different adjuvant modalities were comparable with those obtained in the literature.All relapsed patients from the surveillance group had at least one risk factor for developing metastatic disease, presuming them to better candidates for adjuvant chemotherapy.One of these patients seemed to be fairly incompliant and was not a candidate for this follow up scheme retrospectively.Nevertheless, although the treatment strategies for our patients with NSGCT Stage I were highly inconsistent during the chosen observation period of 24 years, neither relapse rates nor mortality were mainly affected.
Considering all the factors involved, the decision for the correct adjuvant approach in patients with stage I nonseminomatous germ cell tumors should include risk factors for developing metastatic disease as well as patient related factors.Furthermore, individual clinical expertise should be considered in the decision.Summarizing the results in line with recent data from the literature, patients with NSGCT I should be treated following the recommendations of the European Germ Cell Cancer Consensus Group (3) avoiding an inhomogeneous therapeutic regimen and providing the optimal treatment for every single patient.

EDITORIAL COMMENT
Overall, the paper lacks strength due to the small number of patients in the RPLND and surveillance arms.In the introduction it is stated that "in patients with reservations to surveillance, adjuvant chemotherapy is the treatment of choice".In my opinion this is not true.Furthermore, in the discussion it states that "chemotherapy or RPLND are possible options".Overall there is not survival difference between any of the 3 modalities and RPLND is curable in high risk patients.

Department of Urology and Oncology
Indiana University Medical Center Indianapolis, Indiana, USA E-mail: sdwbeck@iupui.edu

EDITORIAL COMMENT
The introduction of cisplatin-based combination chemotherapy has revolutionized the treatment of metastatic testicular cancer.Owing to the high success rate in the salvage of disseminated cancer, it has become reasonable to argue for managing clinical Stage I nonseminomatous germ cell testicular tumors (CS I NSGCTT) patients with orchiectomy alone followed by surveillance.Patients, who relapse are treated with systemic chemotherapy, whereas those, who do not relapse, are spared unnecessary treatment.
The surveillance after orchiectomy alone has gained a lot of popularity in the management of CS I NSGCTT.Preliminary results were enthusiastic, but critical voices have been raised against general use of this option as a routine management.With longer observation, the relapse rate has been found to increase to 25 % or more after orchiectomy.Recent investigations have focused on determining the factors that identify a group of patients at high risk of the relapse, who might therefore benefit from a program other than surveillance.
The optimal management of CS I NSGCTT patients after an orchiectomy has been controversial for several decades, because of the difficulty of distinguishing true Stage I patients from those with occult retroperitoneal and distant metastases.Over the last 20 years, a surveillance strategy has been in practice at various centers to save patients in Stage I from unnecessary treatment-related morbidity.A number of primary tumor prognostic factors have been discovered that may be useful in stratifying CS I patients as to their likelihood of harboring occult disease.Up to 30 % of CS I NSGCTT patients have subclinical metastases and will relapse if surveillance alone is applied after orchiectomy.
The utility of vascular invasion (venous and lymphatic invasion) as a prognostic marker in CS I NSGCTT was first recognized in the 1980`s and during the years it became the main predictor of relapse in CS I NSGCTT managed by surveillance.Importance of embryonal carcinoma as a prognostic factor in low stage NSGCTT was discovered when surveillance studies were analyzed for relapse factors.Therefore, embryonal carcinoma is extremely important as a prognostic marker for occult disease in CS I NSGCTT.The presence of teratoma elements in testicular germ cell tumors has been known to have a favorable impact on prognosis.In contemporary era of prognostic factors in CS I NSGCTT, the presence of teratoma lessens the likelihood of occult disease.Teratomatous elements in the orchiectomy specimen predict for retroperitoneal teratoma, therefore primary RPLND in CS I NSGCTT patients was recommended for cases with the finding of teratoma in the primary tumor.Patients can be stratified according to risk factors into different prognostic groups with different recurrence rates.According to EAU guidelines on testicular cancer and to reports of the European Germ Cell Cancer Consensus Group risk-adapted treatment is recommended as treatment of first choice in CS I NSGCTT patients, however, there is no worldwide consensus on the management of high-risk CS I NSGCTT.High risk patients, with vascular invasion are recommended to undergo adjuvant chemotherapy with two cycles of BEP regimen, intermediate risk patients are recommended to undergo primary RPLND and low risk patients, without vascular invasion are recommended to undergo surveillance.
It is generally accepted that surveillance is appropriate for patients with a low risk of relapse (without vascular invasion), however, there is no universally accepted standard protocol for surveillance of patients with CS I NSGCTT.The main advantage of surveillance being that 70-86 % of patients do not need any further treatment after orchiectomy.The disadvantages are the psychological and practical difficulties of intense follow-up for some patients.
The interesting article by Seseke et al. describes long-term experiences with CS I NSGCTT.Their information that vascular invasion is the most important prognostic indicator for a risk of develop-ing metastatic disease is correct, but percentage up to 48% is too high.Also the results of the authors that patients managed with a surveillance strategy after orchiectomy showed a relapse rate 50 % is too high.Therefore, the results of authors are not comparable with those obtained in literature because of small number error causes (inhomogeneous number of patients treated by particular therapeutic modalities).

Table 2 -
Recurrences in the follow-up of 52 patients with NSGCT stage I.