Outcomes of Surveillance Protocol of Clinical Stage I Nonseminomatous Germ Cell Tumors—Is Shift to Risk Adapted Policy Justified?

Outcomes of Surveillance Protocol of Clinical Stage I Nonseminomatous Germ Cell Tumors—Is Shift to Risk Adapted Policy Justified?

Outcomes of Surveillance Protocol of Clinical Stage I Nonseminomatous Germ Cell Tumors—Is Shift to Risk Adapted Policy Justified? Rauf Taner Divrik,* ...

105KB Sizes 0 Downloads 1 Views

Outcomes of Surveillance Protocol of Clinical Stage I Nonseminomatous Germ Cell Tumors—Is Shift to Risk Adapted Policy Justified? Rauf Taner Divrik,* Bülent Akdog˘an, Haluk Özen and Ferruh Zorlu From the Department of Urology, SB Tepecik Research and Training Hospital, I˙zmir (RTD, FZ), and Department of Urology, Hacettepe University, Ankara, Turkey

Purpose: We evaluated the potential risk factors for disease relapse in patients with clinical stage I nonseminomatous germ cell tumors treated with surveillance and reevaluated our treatment of these patients. Materials and Methods: A total of 211 consecutive patients with clinical stage I nonseminomatous germ cell tumors treated with surveillance after orchiectomy between 1993 and 2005 were included in this retrospective study. Risk factors evaluated were presence of vascular invasion, proportion of embryonal carcinoma, age, tumor size, preoperatively increased serum ␣-fetoprotein and the absence of yolk sac component. Results: Of the 211 patients 66 (31.3%) had disease relapse. Recurrence ranged from 2 to 32 months after orchiectomy (median 6). A total of 52 (78.8%) cases of relapse were diagnosed in year 1 of followup, 11 (16.7%) during year 2 and only 3 cases were diagnosed thereafter. The first evidence of relapse was most commonly the increase in serum tumor markers alone (28.8%) or in combination with other modalities (66.7%, overall 95.5%). While 40.9% of patients with more than 50% embryonal carcinoma had disease relapse, the relapse rate was 20.8% in patients with less than 50% embryonal carcinoma (p ⫽ 0.002). Relapse rates in patients with and without vascular invasion were 75.5% and 17.9%, respectively (p ⫽ 0.000). The relapse rates were 6.1% and 75.7% in patients with no risk factors (no vascular invasion and less than 50% embryonal carcinoma) and 2 risk factors (vascular invasion and more than 50% embryonal carcinoma), respectively (p ⬍0.001). Multivariate analysis revealed that vascular invasion was the most powerful predictor of relapse (OR 16.350, 95% CI 5.582– 47.893). Disease-free and disease specific survival rates were 97.6% at a median followup of 75 months. Conclusions: In light of our results we suggest that all patients with vascular invasion should receive chemotherapy. However, patients with no risk factors and those with more than 50% embryonal carcinoma but without vascular invasion should be on surveillance after orchiectomy since the relapse rate is less than 30%. Although strict followup in the first year is justified, followup schemas may be reassessed for the frequency of radiological investigations. Key Words: testicular neoplasms, recurrence, risk factors

he standard treatment of patients with clinical stage I NSGCT of the testis following orchiectomy remains controversial since patients with CSI NSGCT have excellent survival either with RPLND,1 surveillance2 or primary chemotherapy.3 Even with modern staging techniques with CTA and monitoring of serum tumor markers, approximately 30% of patients with clinical stage I nonseminomatous testicular germ cell tumors have occult metastatic disease and will have relapse if they are only observed and followed up after orchiectomy. If RPLND is performed disease staging will be more accurate but many patients will have undergone unnecessary surgery and, despite histologically verified negative lymph nodes, 10% to 17% will have metastases4 outside the surgical boundaries. Moreover, 50% to 55% of patients who are given primary chemotherapy after orchiectomy will be over treated. Because of the inaccuracy of clinical staging methods and the risk of metastasis,

clinical research focusing on the development of prognostic risk factors remains one of the most important clinical challenges. In previous studies vascular invasion and the percentage of embryonal carcinoma in an orchiectomy specimen have been identified as significant histopathological risk factors for disease relapse.5–10 We report on the results of a surveillance program for CSI NSGCT and an analysis of prognostic factors for relapse to revisit our treatment of these patients.

Submitted for publication November 2, 2005. Nothing to disclose. * Correspondence: 1394 SK. NO:11/13, Alsancak-I˙zmir, Türkiye (telephone: ⫹90 232 4650888; FAX: ⫹90 232 4646050; e-mail: [email protected] or [email protected]).

Editor’s Note: This article is the fourth of 5 published in this issue for which category 1 CME credits can be earned. Instructions for obtaining credits are given with the questions on pages 1690 and 1691.

T

0022-5347/06/1764-1424/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION

PATIENTS AND METHODS Of 251 total patients who were offered surveillance treatment at 2 institutes 64 patients missed at least 1 visit. These 64 patients were invited by telephone, and of them 24 responded and were evaluated for relapse. Thus, 211 available patients with CSI NSGCT from SB Tepecik Research and

1424

Vol. 176, 1424-1430, October 2006 Printed in U.S.A. DOI:10.1016/j.juro.2006.06.012

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS

every 6 months in year 3 and annually thereafter. Computerized tomography of the abdomen and thorax were performed every 4 months in the first year, 6 months in the second year and annually thereafter. Disease recurrence was defined as serum tumor marker increase and/or tumor growth seen on radiographic study. The patients with disease relapse received mostly 3 courses of BEP chemotherapy and supplementary surgical resection of any residual disease after chemotherapy. The chi-square test and Fisher’s exact test were used for univariate analysis. Multivariate logistic regression analysis with stepwise selection was performed for variables.

TABLE 1. Patient characteristics No. Pts (%) Age: 30 or Younger Older than 30 Relapse AFP (preop, abnormal)* Death Presence of yolk sac component Vascular invasion present Presence of EC (more than 50%) Cm3 primary tumor vol:* Greater than 3 3 or Less Condition: Histopathology Teratoma EC Mixed germ cell Yolk sac Teratocarcinoma

1425

130 (61.6) 81 (38.4) 66 (31.3) 127 (71.3) 5 (2.4) 47 (22.3) 49 (23.2) 110 (52.1) 98 (67.6) 47 (32.4) 13 (6.2) 41 (19.4) 85 (40.3) 8 (3.8) 64 (30.3)

RESULTS

* We could find partial results in some cases.

Training Hospital and Hacettepe University Hospital treated with surveillance between January 1993 and May 2005 were included in this retrospective study. After undergoing radical orchiectomy cases were clinically staged with CXR, CT of the chest and abdomen, and repeated serum tumor marker analysis. All patients with normalized markers according to the half-life of the markers and normal tomography were offered surveillance. For the prediction of risk factors we used the previously published cutoff values of these parameters, that is the absence or presence of VI, 50% or more vs less than 50% EC, age, primary tumor size (cutoff 3 cm), presence of abnormal serum AFP levels before orchiectomy and presence of YS component. Vascular invasion was documented when tumor adhered to a vessel wall or almost completely filled a space lined with flat endothelial cells. We used the term vascular invasion to indicate the presence of venous or LI with no distinction between venous and lymphatic structures. Followup consisting of clinical history, physical examination, tumor markers and CXR was performed every 2 months in year 1, every 3 months in year 2,

Median patient age was 28 years (range 2 to 56). These patients were followed for a median of 75 months after orchiectomy (mean 87.7, range 12 to 148). Disease progression occurred in 66 patients (31.3%) and was observed 2 to 32 months following orchiectomy (median 6 months). Of 62 patients who had relapse 52 (78.8%) were diagnosed within 12 months, 11 (16.7%) had relapse during the second year and the rest after 24 months of followup. Tumor dimensions were available for evaluation in 145 patients, with 47 (32.4%) having a tumor less than 3 cm and 98 (67.6%) greater than 3 cm. Serum AFP levels before orchiectomy were available in 178 patients. AFP was increased in 127 (71.3%) patients. Primary histopathological findings were teratoma in 13 patients (6.2%), EC in 41 patients (19.4%), mixed germ cell tumors in 85 patients (40.3%), YS in 8 patients (3.8%) and teratocarcinoma in 64 patients (30.3%). More than 50% of EC was found in 110 (52.1%) patients. Vascular invasion was detected in 49 (23.2%) patients. Patient characteristics are summarized in table 1. Neither age at orchiectomy nor preoperative AFP levels were significantly associated with the risk of relapse. Furthermore, tumor dimensions and the presence of YS component were not found to be significant on univariate analysis. Two factors were determined as significant risk factors for relapse, that is more than 50% EC and the presence of

TABLE 2. Univariate and multivariate analyses of variables in terms of relapse Relapse

AFP: Normal High % EC: Greater than 50 Less than 50 VI: Present Absent VI absent & less than 50% EC (no risk factor) VI present or greater than 50% EC (1 risk factors) VI present & greater than 50% EC (2 risk factors) Yolk sac elements: Present Absent Age: Younger than 31 Older than 30 Primary tumor vol (cm3): 3 or Less Greater than 3

Univariate Analysis (p value)

% No

% Yes

Multivariate Analysis (p value)

70.6 68.5

29.4 31.5

0.859

Not significant

59.1 79.2

40.9 20.8

0.002

0.045 (OR 2.583, 95% CI 1.020–6.540)

24.5 82.1 93.9 64.1 24.3

75.5 17.9 6.1 35.9 75.7

0.000

⬍0.001 (OR 16.350, 95% CI 5.582–47.893)

0.000

⬍0.001

59.6 71.3

40.4 28.7

0.153

0.701 (OR 1.258, 95% CI 0.390–4.053)

71.5 64.2

28.5 35.8

0.287

0.859 (OR 1.091, 95% CI 0.418–2.845)

72.4 68.1

27.6 31.9

0.696

0.828 (OR 1.118, 95% CI 0.407–3.070)

1426

References

Risk Factors

No. Pts

Median Mos Observation

% Overall Survival

% Overall Relapse

0 Risk Factors

1 Risk Factor

Peckham et al3 Thompson et al13

EC present Lymphatic invasion present EC present, VI, pT greater than 1 EC present, VI, lymphatic invasion, absence of YS pT greater than 1 80% EC, VI, (AFP greater than 80) EC present, VI EC present, VI VI VI, absence of teratoma VI, 50% EC present, tumor greater than 5 cm EC present VI, greater than 50% EC

53 36

15 36

100 97.3

17 33.3

3.4 6.2

42.8 55.5

59

30

98.3

30.5

259

30

46

40⫹

Pizzocaro et al7 Freedman et al6

Raghavan et al12 7

Wishnow et al * Sturgeon et al8 Sogani et al9 Colls et al14 Alexandre et al15 Roeleveld et al

Atsü et al16 Present study

10

20

50

29

14.6

45

5.6 12 17 0

Not mentioned

105 105 248 88

60 135 53 51.6

99 97.2 98 96

35.2 25.7 28 27

90

97.2

98.9

26

38 75

99 97.6

24 31.3

132 211

* Total of 80% EC and VI were accepted predominant risk factors.

Not mentioned

9–22

28

82

3 Risk Factors

40–50 29

24–25

32 35.9

EC* Present

57

VI Absent

Less Than 50% EC

13.3

50–58

58

52

57.1 71

60.6 60 46

50 40 46 Not significant

23.6 17 17

19

61 55

7.5 6.1

VI Present

42.8

60–67

27

95.7

2 Risk Factors

75.7

51.5

Not significant 75.5

33

10.9

32 40.9

17.9

7.5 20.8

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS

TABLE 3. Risk factors and relapse rates of patients with CSI NSGCT treated with surveillance from published series

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS

1427

TABLE 4. Number of chemotherapy courses according to hypothetical management protocols No. Pts Needing Chemotherapy/Total No.

Total surveillance At least 1 risk factor present (VI or greater than 50% EC or both) VI present Greater than 50% EC present 2 Risk factors present

As Adjuvant Therapy

As Therapy For Relapse

0 129/129

66/211 5/82

66 ⫻ 3 ⫽ 198 129 ⫻ 2 ⫹ 5 ⫻ 3 ⫽ 273

49/49 110/110 37/37

29/162 21/101 38/174

49 ⫻ 2 ⫹ 29 ⫻ 3 ⫽ 185 110 ⫻ 2 ⫹ 21 ⫻ 3 ⫽ 283 37 ⫻ 2 ⫹ 38 ⫻ 3 ⫽ 188

vascular invasion. There were 45 (40.9%) and 21 (20.8%) patients with more than 50% EC and less than 50% EC with relapse, respectively (p ⫽ 0.002). Of 49 patients with vascular invasion 37 (75.5%) had relapse, while 29 (17.9%) of 162 patients without vascular invasion had relapse (p ⫽ 0.000). There were 82 (38.9%), 92 (43.6%) and 37 (17.5%) patients with no risk factors, with 1 risk factor and both risk factors, respectively. The relapse rates were 6.1%, 35.9% and 75.7% in patients with no risk factors (no VI and less than 50% EC), with 1 risk factor (VI or more than 50% EC) and with 2 risk factors (VI and more than 50% EC), respectively (p ⬍0.001). On multivariate analysis only 2 factors correlated with relapse, vascular invasion being the most powerful predictor of relapse (OR 16.350, 95% CI 5.582– 47.893) (table 2). In 66 patients with relapse, relapse sites were retroperitoneal nodes only in 43 (65.2%) patients, retroperitoneal nodes and lung in 4 (6%) patients, and increased markers only in 19 (28.8%) patients. The first evidence of relapse was most commonly the increase in serum tumor markers, alone (28.8%) or in combination with the other modalities (66.7%, overall 95.5%). Five patients (2.4%) died of disease during followup. At the end of followup disease specific survival was 97.6%. DISCUSSION There is a 97% to 99% cure rate for patients with CSI NSGCT no matter which therapeutic approach is selected.11 Thus, it is important to decide on the most costeffective and the least morbid approach. In the presented cohort the relapse rate was 31% with surveillance, which is in accordance with relevant studies (table 3).3,5–9,12–16 In this study all patients with CSI NSGCT regardless of poor prognostic parameters (vascular invasion and presence of greater than 50% EC) were subjected to this approach in the hope of avoiding unnecessary chemotherapy and/or surgery. When patients were analyzed according to their defined risk factors, the relapse rates were 6.1% and 75.7% in patients with no risk factors (no VI and less than 50% EC) and both risk factors (VI and more than 50% EC), respectively. On the other hand VI remained the most significant prognostic parameter on multivariate analysis. The relapse rates were 17.9% and 75.5% in patients without VI vs with VI. These results suggest that patients without both of these risk factors should not receive therapy. The importance of VI is highlighted in previous published series.3,5–10,14,15,17 Patients with no risk factors may safely be monitored

No. Chemotherapy Courses

during surveillance due to a low relapse rate. However, nerve sparing RPLND or surveillance may be offered to patients with EC only (relapse rate 28%) according to our results and reports from the literature. Compliance may be the drawback of surveillance while morbidity of a major surgery and metastasis outside retroperitoneum may be regarded as the pitfalls of RPLND. In a study of cost-effectiveness Link et al demonstrated that primary chemotherapy is significantly superior to surveillance and RPNLD if the relapse rate is greater than 46%.11 In patients with VI the disease relapse rate in our patients was greater than 50%. Thus, according to the aforementioned study the recommendation for our patients with VI with or without more than 50% EC should be chemotherapy instead of surveillance alone. However, it must be emphasized that the study by Link et al depends on a set of critical assumptions, some of which are open to discussion.11 Furthermore, Cullen et al have shown that patients preferred chemotherapy if the relapse rate was estimated to be higher than 50%.18 In fact a retrospective hypothetical analysis shows that initial chemotherapy for our patients with VI would have been cost-effective and less morbid compared to surveillance alone (table 4). In our present approach 66 of 211 patients had disease relapse and received 3 courses of chemotherapy resulting in 198 courses of chemotherapy, excluding patients needing extra courses following postchemotherapy surgery. However, if we had initially administered chemotherapy to the 49 patients with VI (2 courses) plus 3 courses to the 29 patients without VI who would have experienced relapse, this would have resulted in 185 courses (table 4). Although the difference is not great it must be emphasized that morbidity due to chemotherapy is known to increase significantly with the third course. The low total dose of etoposide in 2 courses should have a minute risk of secondary leukemia and, moreover, it has been shown that 2 courses of BEP chemotherapy did not adversely affect fertility status.19 Thus, not only is the number of courses decreased with this approach but also, and more importantly, the major morbidity due to chemotherapy is reduced because more patients receive less than 3 courses of chemotherapy. For patients with risk factors the incidence of retroperitoneal metastasis ranges from 41% to 92%.17 Although most of the patients with low volume metastasis may be cured with RPLND only, patients with extensive disease present more than a 50% relapse rate following RPLND and necessitate adjuvant chemotherapy. Thus, the role of RPLND in this high risk cohort is controversial. The substantial advan-

1428

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS TABLE 5. Time of relapse in different studies

References

No. Relapse (%)

No. First 12 Mos (%)

No. First 24 Mos (%)

46 36 105 105 248 88 90 132 211

13 (28.2) 12 (33.3) 37 (35.2) 27 (25.7) 70 (28.2) 24 (27.3) 23 (25.6) 32 (24.2) 66 (31.3)

10 (76.9) 10 (83.3) 34 (91.9) 23 (85.1) 60 (85.7) 21 (87.5) 16 (69.6) 28 (87.5) 52 (78.8)

11 (84.6) 10 (83.3) 37 (100.0) 27 (100.0)

1,061

304 (28.8)

254 (82.9)

224 (93.8)

No. Pts

Raghavan et al12 Thompson et al13 Sturgeon et al8 Sogani et al9 Colls et al14 Alexandre et al15 Roeleveld et al10 Atsü et al16 Present study Overall

tage of RPLND compared to chemotherapy is for late relapse. Late relapse after properly performed RPLND is reported to be as low as 0.6% exclusively in the lung.1 However, late relapse after primary chemotherapy usually occurs in the retroperitoneum and since it is chemorefractory the cure rate in this subset is low.1 The important point is to decide on the most sufficient and the least complicated treatment with the best patient compliance. With this in mind we have to choose the risk adapted treatment modality which has a minimal relapse rate. The risk of overall long-term relapse in these patients is 27% to 35% if orchiectomy is followed by surveillance alone.5– 8,12–15 We can further reduce this relapse rate and improve patient compliance if we choose to treat the aforementioned group that has a high chance of disease relapse, namely patients with VI. A second unresolved issue is the frequency of CT investigations during surveillance. The majority of disease relapse occurred within the first year. In our study the relapse rate diagnosed in year 1 of followup was 78.8% and within 2 years 95.5% of all relapse was diagnosed. In published series overall this rate was 82.9% (range 70% to 92%) for the first year and 93.8% (range 83% to 100%) for 24 months (table 5). 8,9,12–16 On the other hand, the first evidence of relapse was most commonly the increase in serum tumor markers, alone (28.8%) or in combination with other modalities (66.7%), overall reaching 95.5% in the present study. The increase in serum tumor markers with or without other modalities was found in the range of 58.3% to 84.3% and the mean rate of the increased markers with or without the other modalities was 74.2% in total of the 8 published and our series (table 6).8 –10,13–16,20 The high predictivity of the markers and the fact that

No. First 28 Mos (%) 12 (100.0) 69 (99.0)

24 (100.0) 20 (87.0) 32 (100.0) 63 (95.5)

78.8% of patients had disease relapse in the first year suggests that prospective studies are needed to define the schema for surveillance CT. Decreasing the number of CT investigations would decrease cost and increase patient compliance. We are not able to comment further on this subject with the present data. Another important point in this study is the noncompliance of 64 of the 251 patients who missed at least 1 visit. More importantly, of 64 patients 40 were noncompliant even after telephone contact. This high figure again emphasizes the need to offer chemotherapy to high risk patients. Motivating the patients who would be offered surveillance in a risk adapted policy is still important, albeit the relapse risk is relatively low in this group.

CONCLUSIONS The most significant predictor of relapse was vascular invasion followed by the presence of more than 50% EC. VI is associated with a higher probability of relapse. The patients with no risk (no VI and less than 50% EC) should be treated with surveillance after orchiectomy because the relapse rate is less than 10%. Our results suggest that patients with vascular invasion with or without more than 50% EC should be treated with chemotherapy because they have a significant risk of disease relapse. By removing the patients with a high risk of recurrence the relapse-free survival of the remainder on surveillance will be improved. The strict followup schedule may induce stress and invite noncompliance. Our findings suggest the need for well designed studies to define the surveillance schema especially in regard to the frequency of CT.

TABLE 6. The first evidence for detecting relapse in different studies %

References

Markers Only

Markers & Others

Overall Markers

Thompson et al13 Sturgeon et al8 Klepp et al20 Sogani et al9 Colls et al14 Alexandre et al15 Roeleveld et al11 Atsü et al16 Present study

33.3 16.2 15.0 7.0 24.3 29.2 4.3 31.3 28.8

25.0 67.6 55.0 59.7 60.0 45.8 60.9 37.5 66.7

58.3 83.8 70.0 66.7 84.3 75.0 65.2 68.8 95.5

CTA Only

CTA & Others

16.2

25.0 67.6

Chest CT

63.0 27.1

1.4 58.3 78.3

CXR

Clinical

16.7 8.1 20.0 19.0 5.7 12.5 13.0

25.0 8.1

Lymphangiography of Retroperitoneal Lymphatic

Radiology (not specified)

4.3 21.7 31.3

24.0

16.4

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS

1429

14.

Abbreviations and Acronyms AFP ⫽ ␣-fetoprotein BEP ⫽ bleomycin, etoposide, cisplatin CSI NSGCT ⫽ clinical stage I nonseminomatous germ cell tumors CT ⫽ computerized tomography CTA ⫽ computerized tomography of abdomen and pelvis CXR ⫽ chest radiography EC ⫽ embryonal carcinoma LVI ⫽ lymphovascular invasion RPLND ⫽ retroperitoneal lymph node dissection VI ⫽ vascular invasion YS ⫽ yolk sac

REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

Stephenson, A. J. and Sheinfeld, J.: The role of retroperitoneal lymph node dissection in the management of testicular cancer. Urol Oncol, 22: 225, 2004 Culine, S., Theodore, C., Terrier-Lacombe, M. J. and Droz, J. P.: Primary chemotherapy in patients with nonseminomatous germ cell tumors of the testis and biological disease only after orchiectomy. J Urol, 155: 1296, 1996 Peckham, M. J., Barrett, A., Husband, J. E. and Hendry, W. F.: Orchiectomy alone in testicular stage I nonseminomatous germ cell tumors. Lancet, 2: 678, 1982 Donohue, J. P., Thornhill, J. A., Foster, R. S., Rowland, R. G. and Bihrle, R.: Retroperitoneal lymphadenectomy for clinical stage A testis cancer (1965 to 1989): modifications of technique and impact on ejaculation. J Urol, 149: 237, 1993 Pizzocaro, G., Zanoni, F., Milani, A., Salvioni, R., Piva, L., Pilotti, S. et al: Orchiectomy alone in clinical stage I nonseminomatous testis cancer: a critical appraisal. J Clin Oncol, 4: 35, 1986 Freedman, L. S., Parkinson, M. C., Jones, W. G., Oliver, R. T., Peckham, M. J., Read, G. et al: Histopathology in the prediction of relapse of patients with stage I testicular teratoma treated by orchidectomy alone. Lancet, 2: 294, 1987 Wishnow, K. I., Johnson, D. E., Swanson, D. A., Tenney, D. M., Babaian, R. J., Dunphy, C. H. et al: Identifying patients with low-risk clinical stage I nonseminomatous testicular tumors who should be treated by surveillance. Urology, 34: 339, 1989 Sturgeon, J. F. G., Jewett, M. A. S., Alison, R. E., Gospodarowicz, M. K., Blend, R., Herman, S. et al: Surveillance after orchidectomy for patients with clinical stage I nonseminomatous testis tumors. J Clin Oncol, 10: 564, 1992 Sogani, P. C., Perrotti, M., Herr, H. W., Fair, W. R., Thaler, H. T. and Bosl, G.: Clinical stage I testis cancer: long-term outcomes of patients on surveillance. J Urol, 159: 855, 1998 Roeleveld, T. A., Horenblas, S., Meinhardt, W., Van de Vijver, M., Kooi, M. and Huinink, W. W. T. B.: Surveillance can be the standard of care for stage I nonseminomatous testicular tumors and even high risk patients. J Urol, 166: 2166, 2001 Link, R. E., Allaf, M. E., Pili, R. and Kavoussi, L. R.: Modeling the cost of management options for stage I nonseminomatous germ cell tumors: a decision tree analysis. J Clin Oncol, 23: 5762, 2005 Raghavan, D., Colls, B., Levi, J., Fitzharris, B., Tattersall, M. H., Atkinson, C. et al: Surveillance for stage I nonseminomatous germ cell tumours of the testis: the optimal protocol has not yet been defined. Br J Urol, 61: 522, 1988 Thompson, P. I., Nixon, J. and Harvey, V. J.: Disease relapse in patients with stage I nonseminomatous germ cell tumor of the testis on active surveillance. J Clin Oncol, 6: 1597, 1988

Colls, B. M., Harvey, V. J., Skelton, C. M. A., Frampton, P. I., Thompson, P. I., Bennett, M. et al: Late results of surveillance of clinical stage I nonseminoma germ cell testicular tumours: 17 years’ experience in a national study in New Zealand. BJU Int, 83: 76, 1999 15. Alexandre, J., Fizazi, K., Mahe, C., Culine, S., Droz, J.-P., Theodore, C. et al: Stage I non-seminomatous germ cell tumours of the testis: identification of a subgroup of patients with a very low risk of relapse. Eur J Cancer, 37: 576, 2001 16. Atsu, N., Eskicorapçi, S., Uner, A., Ekici, S., Gungen, Y., Erkan, I. et al: A novel surveillance protocol for stage I nonseminomatous germ cell testicular tumours. BJU Int, 92: 32, 2003 17. Heidenreich, A., Sesterhenn, I. A., Mostofi, F. K. and Moul, J. W.: Prognostic risk factors that identify patients with clinical stage I nonseminomatous germ cell tumors at low risk and high risk for metastasis. Cancer, 83: 1002, 1998 18. Cullen, M. H., Billingham, L. J., Cook, J. and Woodroffe, C. M.: Management preferences in stage I non-seminomatous germ cell tumours of the testis: an investigation among patients, controls and oncologists. Br J Cancer, 74: 1487, 1996 19. Böhlen, D., Burkhard, F. C., Mills, R., Sonntag, R. W. and Studer, U. E.: Fertility and sexual function following orchiectomy and 2 cycles of chemotherapy for stage I high risk nonseminomatous germ cell cancer. J Urol, 165: 441, 2001 20. Klepp, O., Dahl, O., Flodgren, P., Stierner, U., Olsson, A. M., Oldbring, J. et al: Risk-adapted treatment of clinical stage 1 non-seminoma testis cancer. Eur J Cancer, 33: 1038, 1997

EDITORIAL COMMENTS This article verifies that surveillance is a reasonable method of treatment of patients with clinical stage I nonseminoma. Other methods of management include nerve sparing RPLND or chemotherapy with 2 courses of BEP. The chance for cure is essentially the same regardless of the chosen option so patient involvement in the choice of therapy is mandatory. Patients choose the treatment method based upon psychological issues, differences in side effects of each type of therapy and the local capabilities of the treating physicians. The article by Link et al has excellent methodology but is based upon assumptions gleaned from old literature which, in some cases, is not relevant to current clinical practice (reference 11 in article). It will be interesting to see the results of the current German cooperative trial comparing 1 course of BEP to RPLND in clinical stage I nonseminoma. This is a large randomized trial and the results should be available soon. If 1 course of BEP is equivalent to standard therapy, the way we approach these patients may change dramatically. Richard S. Foster Department of Urology Indiana University Indianapolis, Indiana The authors recommend surveillance for patients with CSI NSGCT with a low risk of relapse (based on the absence of LVI in the primary tumor), but that chemotherapy with BEP ⫻ 2 be administered to those with LVI based on a 76% (95% CI, 62– 88) risk of relapse. We agree that surveillance is an appropriate option for low risk patients, however the evidence

1430

SURVEILLANCE OF CLINICAL STAGE I NONSEMINOMATOUS GERM CELL TUMORS

presented in this study and published reports supports RPLND compared to BEP ⫻ 2 in those at higher risk for relapse. The main observations of interest in this study were that disease specific survival with surveillance was 98% and that 65% of relapse occurred in the retroperitoneum compared to only 6% in the lung. Coupled with a recent study reporting only a 1.3% risk of systemic progression and a 33% incidence of retroperitoneal metastases in patients with low stage NSGCT undergoing RPLND,1 these findings suggest that the risk of occult distant metastases in appropriately staged CS I is low. Thus, the therapeutic focus should be control of the retroperitoneum, and we believe the available evidence supports RPLND as preferable to chemotherapy to achieve this goal. BEP ⫻ 2 represents inadequate therapy for approximately a third of patients with CS I with metastasis based on the 20% to 30% rate of teratoma among patients with pathological stage II disease and the 8% incidence of viable malignancy after induction chemotherapy for CS IIA-B NSGCT.1,2 If this disease is left unresected these patients are at risk for late relapse of chemoresistant disease in the retroperitoneum, which is associated with a median survival of 2 years.3 In contrast, patients who have disease relapse on surveillance or after RPLND are cured with chemotherapy with or without resection of residual masses in virtually all cases. RPLND alone is curative in 90% of patients with low volume (pN1) retroperitoneal metastases,1 and only an estimated 28% of patients with LVI and/or more than 50% embryonal carcinoma after RPLND will require chemotherapy as adjuvant therapy or for treatment of relapse.4 While the authors refer to the perioperative morbidity of RPLND as a major pitfall, the operative mortality is negligible and major complications are reported in only 4% of patients (reference 1 in article). The long-term morbidity is a 1% risk of small bowel obstruction, a 5% rate of ejaculatory dysfunction after nerve sparing RPLND and a midline scar. In

contrast, BEP ⫻ 2 is associated with potentially greater longterm toxicity including cardiovascular disease, secondary leukemia, Raynaud’s phenomenon, peripheral neuropathy and ototoxicity. Although the incidence of these complications is dose dependent, there is no safe lower limit (reference 1 in article). For these reasons we favor RPLND compared to chemotherapy for patients with CS I NSGCT at high risk for relapse after orchiectomy. We believe BEP ⫻ 2 is suitable only for high risk patients unwilling to undergo surveillance who do not have access to surgeons with sufficient experience in performing RPLND. Andrew J. Stephenson and Eric A. Klein Glickman Urological Institute Cleveland Clinic Foundation Cleveland, Ohio 1.

2.

3.

4.

Stephenson, A. J., Bosl, G. J., Motzer, R. J., Kattan, M. W., Stasi, J., Bajorin, D. F. et al: Retroperitoneal lymph node dissection for nonseminomatous germ cell testicular cancer: impact of patient selection factors on outcome. J Clin Oncol, 23: 2781, 2005 Stephenson, A. J., Bosl, G. J., Bajorin, D. F., Motzer, R. J., Stasi, J. P., Carver, B. S. et al: Primary retroperitoneal lymph node dissection (P-RPLND) versus induction chemotherapy (IC) for the management of clinical stage (CS) IIA and IIB nonseminomatous germ cell testicular cancer (NSGCT). J Urol, 173: 196, abstract 720, 2005 Ronnen, E. A., Kondagunta, G. V., Bacik, J., Marion, S., Bajorin, D. F., Sheinfeld, J. et al: Incidence of late-relapse germ cell tumor and outcome to salvage chemotherapy. J Clin Oncol, 23: 6999, 2005 Stephenson, A. J., Bosl, G. J., Bajorin, D. F., Stasi, J., Motzer, R. J. and Sheinfeld, J.: Retroperitoneal lymph node dissection in patients with low stage testicular cancer with embryonal carcinoma predominance and/or lymphovascular invasion. J Urol, 174: 557, 2005