- Email: [email protected]
Surveillance for stage I testicular seminoma
Urologic Oncology 6 (2001) 139–143
Review article
Surveillance for stage I testicular seminoma: a review A. Bayley, M.D., P. Warde, M.B.*, M. Milosevic, M.D., M. Gospodarowicz, M.B. Department of Radiation Oncology, University of Toronto and Princess Margaret Hospital, 610 University Avenue, Toronto, ON M5G 2MN, Canada Received 14 August 2000; received in revised form 7 November 2000; accepted 4 December 2000
Abstract There is good evidence that patients with stage I seminoma can be managed equally well after orchidectomy with surveillance and adjuvant retroperitoneal radiation therapy. There is considerable reluctance amongst many physicians to accept surveillance as a management option in stage I seminoma and this is largely based on the excellent results achieved with adjuvant retroperiteonal radiation for many years. However, patients with stage I seminoma have a long life span and it is possible that the long-term sequelae of radiation treatment could have a negative impact on quality of life and longevity. It is of utmost importance to continue the study of the long-term effects of all current treatment approaches, in particular the risk of induction of second malignancies. However, the psychosocial impact on patients of surveillance and other management strategies must also be assessed. Stage I testicular seminoma is highly curable with currently available management approaches and the current challenge for clinicians is to maintain these excellent results while minimizing toxicity and individualising treatment to the specific social, economic and emotional circumstances of each patient. Surveillance should be one of the management options offered to patients with stage I seminoma. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Management; Seminoma; Surveillance
1. Introduction Testicular germ cell tumors are uncommon neoplasms, accounting for only 1% to 2% of malignancies in North American males [1]. They are, however, the commonest malignancies in men aged 20–34 and epidemiological studies have shown a doubling of the incidence rate in the past 30 years [2]. Approximately 45% of germ cell tumors are seminomas and the majority of patients will present with clinical stage I disease. Only 15% to 20% of patients have infradiaphragmatic lymph node involvement (stage II disease) and less than 5% present with distant metastatic disease. Post orchiectomy treatment options in patients with stage I disease include adjuvant retroperitoneal radiotherapy, surveillance with treatment reserved for those who relapse, retroperitoneal lymph node dissection, and adjuvant chemotherapy. The cause-specific survival with all the above strategies approaches 100%. Post-orchidectomy adjuvant radiation therapy (RT) to the para-aortic and ipsilateral pelvic lymph nodes remains the treatment of choice with expected relapse rates of 4–8% [3–10]. However, the routine use of postoperative radiotherapy has been re-examined over the past 15 years, primarily because of the success of * Corresponding author. Tel.: 1-416-946-2122; fax: 1-416-9464586. E-mail address: [email protected] (P.Warde).
surveillance studies in stage I non-seminomatous germ cell tumors [11,12]. In addition, the availability of improved imaging of the retroperitoneum and increasing concern regarding the risk of induction of second malignancies have led to a re-appraisal of the routine use of adjuvant RT and has stimulated an increased interest in observation after orchidectomy in patients with stage I disease. This article presents an overview of the available information about the use of surveillance in stage I testicular seminoma and indicates where surveillance fits into the armamentarium of clinicians dealing with this disease. 2. Relapse in surveillance studies Several large prospective nonrandomized studies of surveillance have been conducted over the past 15 years (Table 1). The relapse rate in most studies is approximately 20% and the cause-specific survival on studies with 36 months of follow-up is 99.3%. The median time to relapse is 14 to 18 months, however, late relapses have been observed with 25% of relapses occurring greater than four years from the original diagnosis. The largest reported experiences are from the Danish Testicular Carcinoma Study Group (DATECA), the Princess Margaret Hospital (PMH) and the Royal Marsden Hospital (RMH) [13]. In the DATECA study, conducted from 1985 to 1988, 261 patients were enrolled in a nationwide
1078-1439/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S1078-1439(00)00 1 2 9 - 0
140
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
study of surveillance. With a median follow-up of 48 months, the relapse rate was 19% (49 patients). Three patients died of seminoma, one due to progressive disease and two from complications of treatment following relapse. In the PMH series, 256 patients have been followed for a median of 6.1 years (range 1.3–12.3 years) and 37 have relapsed with an actuarial risk of relapse at 5 years of 15.6% [14]. Only one patient has died of seminoma and the causespecific survival is 99.5%. In the Royal Marsden Hospital series 103 patients were followed for a median of 62 months and 17 patients relapsed. The actuarial risk of relapse at 5 years was 18%. No patient died of disease. 2.1. Sites of relapse The predominant site of relapse is in the para-aortic lymph nodes and most patients are asymptomatic at the time of detection. In the DATECA study 41 of 49 relapses (82%) occurred in the para-aortic region and similarly, in the PMH experience 34 of 37 relapses (92%) occurred in the paraaortic lymph nodes. Other sites of relapse included the pelvic lymph nodes (approximately 3% overall), and very rarely the inguinal nodes and the lungs. 2.2. Treatment at relapse At relapse most patients are treated with radiotherapy to the ipsilateral pelvic and para-aortic lymph nodes. It must be noted that the likelihood of detecting progression in the retroperitoneal lymph nodes when their size is less than 5 cm, and therefore can safely be treated with RT, depends on the frequency of follow-up CT scans of the abdomen and pelvis. If these are done infrequently then it is more likely that relapse will be detected at an advanced stage and require treatment with chemotherapy rather than RT. In the DATECA, PMH and RMH studies, the proportions of patients treated with radiotherapy at relapse were 76%, 73% and 74% respectively. In the Danish study 37 patients were treated with RT for first relapse, and four experienced a second relapse. The sites of second relapse included the mediastinal lymph nodes, supraclavicular and neck nodes, lung and pleura. All patients with second relapse received chemotherapy; three achieved complete response and were alive and disease-free at last follow-up (18–38 months), while the fourth patient died of progressive disease. Two of the twelve patients who received chemotherapy at first relapse died of treatment induced neutropenia and infection. In the PMH series, of the 37 patients with disease progression on surveillance, 27 patients were treated with RT to the para-aortic and pelvic lymph nodes, eight with chemotherapy and two with retroperitoneal lymphadenectomy (RPLND). All patients treated with RT achieved local control. Five of the patients treated with RT for relapse developed a second relapse and were treated with cisplatin based chemotherapy regimens. They are all alive and free of disease since completing treatment. Of the 8 patients treated at
initial relapse with cisplatin-based chemotherapy, seven had bulky retroperitoneal disease (6–9 cm), and one had mediastinal disease. Relapse with bulky retroperitoneal disease while on surveillance has not occurred over the past 10 years in patients adhering to the surveillance protocol and this likely represents improvements in diagnostic imaging over the past decade. One patient with mediastinal relapse failed first and second line chemotherapy, and died of disease while undergoing high dose chemotherapy and an autologous bone marrow transplant. Two patients, who previously received RT to the para-aortic area for a contralateral testis tumor and progressed while on surveillance for a second primary, were treated with retroperitoneal lymphadenectomy (RPLND) and neither has developed a second relapse. 3. Prognostic factors for relapse Prognostic factors for relapse have been studied in a number of the surveillance studies. In the DATECA study, tumor size, histological subtype, necrosis and invasion of the rete testis predicted for relapse on univariate analysis (Table 1). However, on multivariate analysis, tumor size was the only significant prognostic factor. The cumulative risk of relapse after 4 years was 6% in those with tumors 3 cm in diameter, 18% with tumors 3 to 6 cm, and 36% with tumors 6 cm or larger. In the Royal Marsden Hospital series of 103 patients managed with surveillance, the only significant factor predicting for relapse was the presence of lymphatic and/or vascular invasion (9% vs. 17% relapse rate). In the PMH series, tumor size, age and presence of small vessel invasion were predictive of relapse on univariate analysis (Fig. 1 and Table 2). On multivariate analysis, age and tumor size were predictive of relapse while small vessel invasion (SVI) had borderline significance. Only 3 of 57 patients (6%) with no adverse prognostic factor (age 34, tumor size 6 cm and absent SVI) relapsed. Age was not
Table 1 Results of surveillance in stage I seminoma patients
Series Von der Maase et al. [13] Warde et al. [15] Horwich et al. [24] Ramakrishnan et al. [25] Oliver et al. [20] Germa Luch & Climent [26] Alhoff et al. [27] Miki et al. [28] Yonese et al. [29] Charig et al. [30]
Relapse (n)
Crude relapse rate %
Causespecific survival (%)
48 50 62
49 27 17
19 16 17
98.9 99 100
72 67
44 61
13 18
18 32
100 97
45 33 23 22 15
34 – 43 – 31
5 3 2 5 5
11 9 9 23 33
100 100 – 100 100
Number of patients
Median follow-up (months)
261 172 103
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
found to be a prognostic factor in the DATECA study, and was not analyzed in the RMH study. A number of other histopathological factors have been assessed for prognostic importance including DNA ploidy status, mitotic rate, DNA S phase fraction, the presence of syncytiotrophoblasts, the degree of lymphocytic infiltration of the primary tumor, and expression of BHCG and low molecular weight keratin on immunohistochemistry [15]. While none of these factors has been shown to be associated with a significantly higher risk of relapse, given the small number of events in the studies, conclusions regarding these factors must be regarded as preliminary. Warde et al. have recently pooled data from the DATECA study, PMH, RMH and the Anglian Germ Cell Cancer Group to examine the prognostic factors and the outcomes in a group of 638 patients with a median followup of 7 years on surveillance protocols. Preliminary analysis from this study has indicated that tumor size and invasion of the rete testis are independent prognostic factors for relapse on surveillance.
4. Economic issues The follow-up schedule for surveillance programs has developed over time. The optimal schedule has not yet been determined. Currently, our patients are seen at four-monthly intervals for the first three years, 6-monthly in years 4–7, and at yearly intervals in years 8–10 (Table 1). At each visit a CT abdomen and pelvis is performed, chest X-rays are done at alternate visits, and serum tumor marker estimation (-HCG, AFP, LDH) is done at each visit for the first 3 years of surveillance. The surveillance programs are not without cost and concerns have been raised about the price of surveillance. Economic analyses of the various treatment approaches are very important. If the outcomes of the two management strategies are the same, then the least expensive option should be adopted as standard treatment. However, in any economic
141
analysis it is very important to capture all potential costs (and outcomes) and in this setting the costs of the increased risk of second malignancy must be considered when comparing the cost of prophylactic radiation treatment vs. surveillance. Two analyses have looked at the economic impact of surveillance. Sharda et al. [16], using a cost minimization model, found that surveillance was more expensive than immediate adjuvant RT with the assumption of equivalent outcomes. Costs were analyzed from the perspective of the payer. Follow-up costs represented 91% of all costs while salvage chemotherapy represented the remaining 9%. The cost of surveillance surpassed the cost of upfront treatment after 2.5 years. Surveillance cost $20,487 compared to $14,722 in 1994 US dollars. In a recent analysis of the PMH data, surveillance was more expensive with an average additional cost per patient of only Can $2,620.00 spread over 10 years [14]. However, in these analyses, the outcome using the two strategies was assumed to be identical and the costs associated with the acute and long term toxicity of radiotherapy (in particular the induction of second malignancies) were not included. There are considerable social and economic costs to the patient, society as a whole and the health care system associated with the use of routine postoperative RT. These patients are often off work for four to six weeks or longer. The work-related costs associated with the follow-up
Table 2 Prognostic factors for relapse on surveillance in stage I testicular seminoma1 Univariate prognostic factors Series DATECA
n 261
Factor 2
Size
Histology
Necrosis Rete Testis PMH
201
Size2 Age Small vessel
RMH
103
Small vessel
Pooled analysis
638
Size2 Rete Testis2 Small vessel
Fig. 1. Stage I testicular seminoma managed by surveillance 1981–93. Effect of capillary lymphatic space invasion.
Stratification
Relapse
3.0 cm 3.0–5.9 cm 6.0 cm Spermatocytic Classical Anaplastic No Yes No Yes 6cm 6cm 34 years 34 years No Yes No Yes 4cm 4cm No Yes No Yes
6% 18% 36% 0% 16% 33% 14% 23% 14% 23% 12% 33% 9% 21% 14% 31% 10% 20% 13% 24% 14% 24% 14% 23%
1 Adapted from Milosevic, et al. Management of testicular seminoma. Sem Surg Oncol 1999;17:243. 2 Significant on multivariate analysis.
142
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
necessary in surveillance programs must also be taken into account, but patients are seen at most 3 times per year and in many instances no loss of time from work is necessary. The cost of any surveillance protocol is determined primarily by the frequency of follow-up and the optimal frequency and duration of patient assessment on surveillance protocols is currently being determined. With further analysis of prognostic factors for occult disease, it may well be possible to tailor follow-up strategy based on individual patients’ risk of relapse. At the present time, CT scans of the entire abdomen and pelvis are routinely performed as part of the surveillance protocols and with further data on the sites of relapse it may be possible to refine this strategy and be more selective (and therefore more cost efficient) in our approach. Further data is necessary on the optimal followup strategy for both approaches before a meaningful economic analysis can be done.
5. Discussion The available data from surveillance and adjuvant RT series suggests that almost 100% of patients with stage I testicular seminoma are cured, whichever management strategy is adopted after orchidectomy. One of the most attractive features of surveillance is the ability to avoid treatment in the majority of patients. However, surveillance must not compromise survival and it is reassuring to find that there have been only four disease-related deaths reported in the 826 patients in the surveillance studies. While isolated occurrences of second non-testicular tumor following treatment of seminoma with RT have been reported for decades, there is now convincing data from long-term studies that patients with seminoma treated with RT have an increased risk of developing a second malignancy. An increased risk of second cancers has been documented in the Netherlands and Denmark, and since this increased risk is expressed more than 10 to 15 years following RT, it may not be apparent in series with shorter follow-up [17,18]. The largest study of second cancers in long-term survivors of testicular cancer was conducted by Travis et al. at the National Cancer Institute Cancer Epidemiology Division [19]. Over 28,000 patients with testis cancer, including over 15,000 with seminoma from 16 population-based registries worldwide, were evaluated. Overall, 1,406 second cancers, excluding contralateral testis, occurred against 981 expected (observed/expected [O/E]1.43). An excess of rectal and small intestine cancer was seen in seminoma patients. The actuarial risk of developing a second malignancy, excluding contralateral cancers, increased over time from diagnosis of testicular cancer up and was 18.2 percent at 25 years. Secondary leukemia was linked with RT and chemotherapy, while an excess of the stomach, bladder, and possibly pancreas tumours was associated with prior RT. This is a major reason to consider surveillance and recommend it to patients with stage I seminoma.
One of the concerns about surveillance is the potential for more patients to require chemotherapy, which involves a prolonged and more toxic course of treatment. In the 1981– 1994 PMH experience, similar numbers of patients managed with surveillance (12/225, 5.3%) and adjuvant RT (10/245, 4.1%) received chemotherapy [9]. This concern is difficult to address, because, in some centres, chemotherapy was used in all patients at relapse. This approach increases the proportion of patients receiving chemotherapy but likely reduces the number of patients developing a second relapse. Also, some patients were placed on surveillance because they had received prior RT or had contraindications to RT, such as inflammatory bowel disease. Chemotherapy with cisplatin and etoposide is very effective in advanced seminoma and is an alternative to RT or surveillance in stage I seminoma. The use of 1 to 2 courses of carboplatin as adjuvant therapy after orchidectomy has been reported by Oliver et al. [20], who treated 78 patients (53 patients had 2 courses; 25 patients had 1 course). With a median follow-up of 44 months, only 1 patient has relapsed. Dieckman et al. (with a median follow-up of 48 months) reported 8 relapses in a cohort of 125 patients treated in a similar fashion (93 patients had 1 course of carboplatin; 32 had 2 courses) [21]. Treatment was well tolerated and acute and intermediate term toxicity was mild. The carboplatin dosage scheme used in these 2 studies was different; in the Oliver et al. study the carboplatin dose was based on area under the curve formula of Calvert et al (AUC) [22], while in the Dieckman report the dosage of Carboplatin was 400 mg/m2. Basing the dose on body surface area rather than AUC leads to a significantly lower dose being given (e.g., 34-year-old man with normal renal function — GFR of 110 ml/minute based on EDTA clearance and BSA of 1.787 would receive 715 mg under the BSA approach but 945 mg using the Calvert formula). A single adjuvant carboplatin injection may well be an attractive treatment option for patients at moderate to high risk of relapse on surveillance protocols. However, before this approach can be generally adopted, further study is necessary. The MRC UK are presently conducting a phase III study comparing adjuvant carboplatin to standard retroperitoneal RT in stage I seminoma and the National Cancer Institute of Canada Clinical Trials Group is currently planning a study comparing adjuvant carboplatin to surveillance. RPLND is another treatment option available in seminoma. It is a reasonable approach for patients unwilling to comply with surveillance or those unable to be treated with RT, such as those with inflammatory bowel disease or those previously treated with retroperitoneal RT [23]. It may also be the best approach in patients with concurrent or previous malignancy in whom histologic examination of lymph nodes is essential to plan treatment. While physicians may view one management approach as preferable, individual patients’ preferences must also be considered. These preferences may be based on many socioeconomic factors (e.g., a desire to travel and consequent inabil-
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
ity to be available for follow up) as well as the side effects associated with both approaches. Also, some patients may view the risk of second malignancy 20–25 years in the future as being of little importance and opt for adjuvant RT to minimize their more immediate risk of relapse. The challenge for physicians involved in treating patients with stage I seminoma is to recommend the most appropriate management for individual patients. Clearly, with surveillance programs, many patients can avoid unnecessary therapy but a proportion of patients will develop disease progression and require treatment. In the future, the identification of prognostic factors for occult disease will allow patients and clinicians to choose management based on a more accurate assessment of an individual patient’s risk of relapse. Patients at a high risk of relapse may opt for adjuvant treatment after surgery. In summary, surveillance is a good option for patients with stage I seminoma. With further research it may be possible to refine our treatment recommendations for individual patients based on prognostic factors for relapse. We currently offer patients surveillance to all patients with stage I seminoma and we believe that the routine use of post-operative RT in stage I seminoma should be reconsidered and a surveillance program offered to all patients as an alternative management option.
References [1] Greenlee R, Murray T, Bolden S, Wingo P. Cancer Statistics 2000. CA: A cancer journal for clinicians 2000;50(1):7–33. [2] Buetow SA. Epidemiology of testicular cancer. Epidemiol Rev 1995; 17(2):433–49. [3] Dosmann MA, Zagars GK. Post-orchiectomy radiotherapy for stages I and II testicular seminoma. Int J Radiat Oncol Biol Phys 1993; 26(3):381–90. [4] Duncan W, Munro AJ. The management of testicular seminoma: Edinburgh 1970–1981. Br J Cancer 1987;55(4):443–8. [5] Fossa SD, Aass N, Kaalhus O. Radiotherapy for testicular seminoma stage I: treatment results and long-term post-irradiation morbidity in 365 patients. Int J Radiat Oncol Biol Phys 1989;16(2):383–8. [6] Giacchetti S, Raoul Y, Wibault P, Droz JP, Court B, Eschwege F. Treatment of stage I testis seminoma by radiotherapy: long-term results — a 30-year experience. Int J Radiat Oncol Biol Phys 1993; 27(1):3–9. [7] Kellokumpu-Lehtinen P, Halme A. Results of treatment in irradiated testicular seminoma patients. Radiother Oncol 1990;18(1):1–7. [8] Sommer K, Brockmann WP, Hubener KH. Treatment results and acute and late toxicity of radiation therapy for testicular seminoma. Cancer 1990;66(2):259–63. [9] Warde P, Gospodarowicz MK, Panzarella T, et al. Stage I testicular seminoma: results of adjuvant irradiation and surveillance. J Clin Oncol 1995;13(9):2255–62. [10] Willan BD, McGowan DG. Seminoma of the testis: a 22–year experience with radiation therapy. Int J Radiat Oncol Biol Phys 1985; 11(10):1769–75.
143
[11] Sturgeon JF, Jewett MA, Alison RE, et al. Surveillance after orchidectomy for patients with clinical stage I nonseminomatous testis tumors. J Clin Oncol 1992;10(4):564–68. [12] Nicolai N, Pizzocaro G. A surveillance study of clinical stage I nonseminomatous germ cell tumors of the testis: 10-year followup. J Urol 1995;154(3):1045–9. [13] von der Maase H, Specht L, Jacobsen GK, et al. Surveillance following orchidectomy for stage I seminoma of the testis. Eur J Cancer 1993;14(4):1931–34. [14] Warde P, Gospodarowicz M, Panzarella T, et al. Long term outcome and cost in the management of stage 1 testicular seminoma. Can J Urol 2000;7:967–72. [15] Warde P, Gospodarowicz M, Banerjee D, et al. Prognostic factors for relapse in stage I testicular seminoma treated with surveillance. 1997; J Urol 157:1705–09. [16] Sharda N, Kinsella T, Ritter M. Adjuvant radiation versus observation: a cost analysis of alternate management schemes in early-stage testicular seminoma. J Clin Oncol 1996;14:2933–9. [17] van Leeuwen F, Stiggelbout AM, van den Belt-Dusebout AW, et al. Second cancer risk following testicular cancer: A follow-up study of 1,909 patients. J Clin Oncol 1993;11(3):415–24. [18] Moller H, Mellemgaard A, Jacobsen GK, Pedersen D, Storm HH. Incidence of second primary cancer following testicular cancer. Eur J Cancer 1993;5:672–6. [19] Travis L, Curtis R, Storm H, et al. Risk of second malignant neoplasms among long-term survivors of testicular cancer. J Natl Cancer Inst 1997;89(19):1429–39. [20] Oliver RT, Edmonds PM, Ong JY, et al. Pilot studies of 2 and 1 course carboplatin as adjuvant for stage I seminoma: should it be tested in a randomized trial against radiotherapy? Int J Radiat Oncol Biol Phys 1994;29(1):3–8. [21] Dieckmann KP, Krain J, Kuster J, Bruggeboes B. Adjuvant carboplatin treatment for seminoma clinical stage I. J Cancer Res Clin Oncol 1996;122(1):63–6. [22] Calvert H, Judson I, van der Vijgh W. Platinum complexes in cancer medicine: pharmacokinetics and pharmacodynamics in toxicity and therapeutic activity. Cancer Surv 1993;17:189–217. [23] Willett C, Ooi C, Zietman A, et al. Acute and late toxicity of patients with inflammatory bowel disease undergoing irradiation for abdominal and pelvic neoplasms. Int J Radiat Oncol Biol Phys 2000;46:995–8. [24] Horwich A. Surveillance for stage I seminoma of the testis. In: Horwich A, editor. Testicular cancer. Investigation and management. London: Chapman and Hall Medical, 1991. pp 109–16. [25] Ramakrishnan S, Champion AE, Dorreen MS, Fox M. Stage I seminoma of the testis: is post-orchidectomy surveillance a safe alternative to routine postoperative radiotherapy? Clin Oncol 1992;4(5): 284–6. [26] Germa Lluch JR, Climent MA, Villavicencio H, et al. Treatment of stage I testicular tumours. Br J Urol 1993;71(4):473–77. [27] Allhoff EP, Liedke S, de Riese W, Stief C, Schneider B. Stage I seminoma of the testis. Adjuvant radiotherapy or surveillance? Br J Urol 1991;68(2):190–4. [28] Miki T, Maeda O, Saiki S, et al. Surveillance after orchiectomy for stage I testicular seminoma. Nippon Hinyokika Gakkai Zasshi 1992; 83:1789–94. [29] Yonese J, Kawai T, Yamauchi T, Ueda T, Kawakami S, Ishibashi K. Surveillance following orchiectomy for stage I testicular tumor. Nippon Hinyokika Gakkai Zasshi 1993;84:1804–10. [30] Charig MJ, Hindley AC, Lloyd K, Golding SJ. ‘Watch policy’ in patients with suspected stage I testicular seminoma: CT as a sole staging and surveillance technique. Clin Radiol 1990;42(1):40–1.
Review article
Surveillance for stage I testicular seminoma: a review A. Bayley, M.D., P. Warde, M.B.*, M. Milosevic, M.D., M. Gospodarowicz, M.B. Department of Radiation Oncology, University of Toronto and Princess Margaret Hospital, 610 University Avenue, Toronto, ON M5G 2MN, Canada Received 14 August 2000; received in revised form 7 November 2000; accepted 4 December 2000
Abstract There is good evidence that patients with stage I seminoma can be managed equally well after orchidectomy with surveillance and adjuvant retroperitoneal radiation therapy. There is considerable reluctance amongst many physicians to accept surveillance as a management option in stage I seminoma and this is largely based on the excellent results achieved with adjuvant retroperiteonal radiation for many years. However, patients with stage I seminoma have a long life span and it is possible that the long-term sequelae of radiation treatment could have a negative impact on quality of life and longevity. It is of utmost importance to continue the study of the long-term effects of all current treatment approaches, in particular the risk of induction of second malignancies. However, the psychosocial impact on patients of surveillance and other management strategies must also be assessed. Stage I testicular seminoma is highly curable with currently available management approaches and the current challenge for clinicians is to maintain these excellent results while minimizing toxicity and individualising treatment to the specific social, economic and emotional circumstances of each patient. Surveillance should be one of the management options offered to patients with stage I seminoma. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Management; Seminoma; Surveillance
1. Introduction Testicular germ cell tumors are uncommon neoplasms, accounting for only 1% to 2% of malignancies in North American males [1]. They are, however, the commonest malignancies in men aged 20–34 and epidemiological studies have shown a doubling of the incidence rate in the past 30 years [2]. Approximately 45% of germ cell tumors are seminomas and the majority of patients will present with clinical stage I disease. Only 15% to 20% of patients have infradiaphragmatic lymph node involvement (stage II disease) and less than 5% present with distant metastatic disease. Post orchiectomy treatment options in patients with stage I disease include adjuvant retroperitoneal radiotherapy, surveillance with treatment reserved for those who relapse, retroperitoneal lymph node dissection, and adjuvant chemotherapy. The cause-specific survival with all the above strategies approaches 100%. Post-orchidectomy adjuvant radiation therapy (RT) to the para-aortic and ipsilateral pelvic lymph nodes remains the treatment of choice with expected relapse rates of 4–8% [3–10]. However, the routine use of postoperative radiotherapy has been re-examined over the past 15 years, primarily because of the success of * Corresponding author. Tel.: 1-416-946-2122; fax: 1-416-9464586. E-mail address: [email protected] (P.Warde).
surveillance studies in stage I non-seminomatous germ cell tumors [11,12]. In addition, the availability of improved imaging of the retroperitoneum and increasing concern regarding the risk of induction of second malignancies have led to a re-appraisal of the routine use of adjuvant RT and has stimulated an increased interest in observation after orchidectomy in patients with stage I disease. This article presents an overview of the available information about the use of surveillance in stage I testicular seminoma and indicates where surveillance fits into the armamentarium of clinicians dealing with this disease. 2. Relapse in surveillance studies Several large prospective nonrandomized studies of surveillance have been conducted over the past 15 years (Table 1). The relapse rate in most studies is approximately 20% and the cause-specific survival on studies with 36 months of follow-up is 99.3%. The median time to relapse is 14 to 18 months, however, late relapses have been observed with 25% of relapses occurring greater than four years from the original diagnosis. The largest reported experiences are from the Danish Testicular Carcinoma Study Group (DATECA), the Princess Margaret Hospital (PMH) and the Royal Marsden Hospital (RMH) [13]. In the DATECA study, conducted from 1985 to 1988, 261 patients were enrolled in a nationwide
1078-1439/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S1078-1439(00)00 1 2 9 - 0
140
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
study of surveillance. With a median follow-up of 48 months, the relapse rate was 19% (49 patients). Three patients died of seminoma, one due to progressive disease and two from complications of treatment following relapse. In the PMH series, 256 patients have been followed for a median of 6.1 years (range 1.3–12.3 years) and 37 have relapsed with an actuarial risk of relapse at 5 years of 15.6% [14]. Only one patient has died of seminoma and the causespecific survival is 99.5%. In the Royal Marsden Hospital series 103 patients were followed for a median of 62 months and 17 patients relapsed. The actuarial risk of relapse at 5 years was 18%. No patient died of disease. 2.1. Sites of relapse The predominant site of relapse is in the para-aortic lymph nodes and most patients are asymptomatic at the time of detection. In the DATECA study 41 of 49 relapses (82%) occurred in the para-aortic region and similarly, in the PMH experience 34 of 37 relapses (92%) occurred in the paraaortic lymph nodes. Other sites of relapse included the pelvic lymph nodes (approximately 3% overall), and very rarely the inguinal nodes and the lungs. 2.2. Treatment at relapse At relapse most patients are treated with radiotherapy to the ipsilateral pelvic and para-aortic lymph nodes. It must be noted that the likelihood of detecting progression in the retroperitoneal lymph nodes when their size is less than 5 cm, and therefore can safely be treated with RT, depends on the frequency of follow-up CT scans of the abdomen and pelvis. If these are done infrequently then it is more likely that relapse will be detected at an advanced stage and require treatment with chemotherapy rather than RT. In the DATECA, PMH and RMH studies, the proportions of patients treated with radiotherapy at relapse were 76%, 73% and 74% respectively. In the Danish study 37 patients were treated with RT for first relapse, and four experienced a second relapse. The sites of second relapse included the mediastinal lymph nodes, supraclavicular and neck nodes, lung and pleura. All patients with second relapse received chemotherapy; three achieved complete response and were alive and disease-free at last follow-up (18–38 months), while the fourth patient died of progressive disease. Two of the twelve patients who received chemotherapy at first relapse died of treatment induced neutropenia and infection. In the PMH series, of the 37 patients with disease progression on surveillance, 27 patients were treated with RT to the para-aortic and pelvic lymph nodes, eight with chemotherapy and two with retroperitoneal lymphadenectomy (RPLND). All patients treated with RT achieved local control. Five of the patients treated with RT for relapse developed a second relapse and were treated with cisplatin based chemotherapy regimens. They are all alive and free of disease since completing treatment. Of the 8 patients treated at
initial relapse with cisplatin-based chemotherapy, seven had bulky retroperitoneal disease (6–9 cm), and one had mediastinal disease. Relapse with bulky retroperitoneal disease while on surveillance has not occurred over the past 10 years in patients adhering to the surveillance protocol and this likely represents improvements in diagnostic imaging over the past decade. One patient with mediastinal relapse failed first and second line chemotherapy, and died of disease while undergoing high dose chemotherapy and an autologous bone marrow transplant. Two patients, who previously received RT to the para-aortic area for a contralateral testis tumor and progressed while on surveillance for a second primary, were treated with retroperitoneal lymphadenectomy (RPLND) and neither has developed a second relapse. 3. Prognostic factors for relapse Prognostic factors for relapse have been studied in a number of the surveillance studies. In the DATECA study, tumor size, histological subtype, necrosis and invasion of the rete testis predicted for relapse on univariate analysis (Table 1). However, on multivariate analysis, tumor size was the only significant prognostic factor. The cumulative risk of relapse after 4 years was 6% in those with tumors 3 cm in diameter, 18% with tumors 3 to 6 cm, and 36% with tumors 6 cm or larger. In the Royal Marsden Hospital series of 103 patients managed with surveillance, the only significant factor predicting for relapse was the presence of lymphatic and/or vascular invasion (9% vs. 17% relapse rate). In the PMH series, tumor size, age and presence of small vessel invasion were predictive of relapse on univariate analysis (Fig. 1 and Table 2). On multivariate analysis, age and tumor size were predictive of relapse while small vessel invasion (SVI) had borderline significance. Only 3 of 57 patients (6%) with no adverse prognostic factor (age 34, tumor size 6 cm and absent SVI) relapsed. Age was not
Table 1 Results of surveillance in stage I seminoma patients
Series Von der Maase et al. [13] Warde et al. [15] Horwich et al. [24] Ramakrishnan et al. [25] Oliver et al. [20] Germa Luch & Climent [26] Alhoff et al. [27] Miki et al. [28] Yonese et al. [29] Charig et al. [30]
Relapse (n)
Crude relapse rate %
Causespecific survival (%)
48 50 62
49 27 17
19 16 17
98.9 99 100
72 67
44 61
13 18
18 32
100 97
45 33 23 22 15
34 – 43 – 31
5 3 2 5 5
11 9 9 23 33
100 100 – 100 100
Number of patients
Median follow-up (months)
261 172 103
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
found to be a prognostic factor in the DATECA study, and was not analyzed in the RMH study. A number of other histopathological factors have been assessed for prognostic importance including DNA ploidy status, mitotic rate, DNA S phase fraction, the presence of syncytiotrophoblasts, the degree of lymphocytic infiltration of the primary tumor, and expression of BHCG and low molecular weight keratin on immunohistochemistry [15]. While none of these factors has been shown to be associated with a significantly higher risk of relapse, given the small number of events in the studies, conclusions regarding these factors must be regarded as preliminary. Warde et al. have recently pooled data from the DATECA study, PMH, RMH and the Anglian Germ Cell Cancer Group to examine the prognostic factors and the outcomes in a group of 638 patients with a median followup of 7 years on surveillance protocols. Preliminary analysis from this study has indicated that tumor size and invasion of the rete testis are independent prognostic factors for relapse on surveillance.
4. Economic issues The follow-up schedule for surveillance programs has developed over time. The optimal schedule has not yet been determined. Currently, our patients are seen at four-monthly intervals for the first three years, 6-monthly in years 4–7, and at yearly intervals in years 8–10 (Table 1). At each visit a CT abdomen and pelvis is performed, chest X-rays are done at alternate visits, and serum tumor marker estimation (-HCG, AFP, LDH) is done at each visit for the first 3 years of surveillance. The surveillance programs are not without cost and concerns have been raised about the price of surveillance. Economic analyses of the various treatment approaches are very important. If the outcomes of the two management strategies are the same, then the least expensive option should be adopted as standard treatment. However, in any economic
141
analysis it is very important to capture all potential costs (and outcomes) and in this setting the costs of the increased risk of second malignancy must be considered when comparing the cost of prophylactic radiation treatment vs. surveillance. Two analyses have looked at the economic impact of surveillance. Sharda et al. [16], using a cost minimization model, found that surveillance was more expensive than immediate adjuvant RT with the assumption of equivalent outcomes. Costs were analyzed from the perspective of the payer. Follow-up costs represented 91% of all costs while salvage chemotherapy represented the remaining 9%. The cost of surveillance surpassed the cost of upfront treatment after 2.5 years. Surveillance cost $20,487 compared to $14,722 in 1994 US dollars. In a recent analysis of the PMH data, surveillance was more expensive with an average additional cost per patient of only Can $2,620.00 spread over 10 years [14]. However, in these analyses, the outcome using the two strategies was assumed to be identical and the costs associated with the acute and long term toxicity of radiotherapy (in particular the induction of second malignancies) were not included. There are considerable social and economic costs to the patient, society as a whole and the health care system associated with the use of routine postoperative RT. These patients are often off work for four to six weeks or longer. The work-related costs associated with the follow-up
Table 2 Prognostic factors for relapse on surveillance in stage I testicular seminoma1 Univariate prognostic factors Series DATECA
n 261
Factor 2
Size
Histology
Necrosis Rete Testis PMH
201
Size2 Age Small vessel
RMH
103
Small vessel
Pooled analysis
638
Size2 Rete Testis2 Small vessel
Fig. 1. Stage I testicular seminoma managed by surveillance 1981–93. Effect of capillary lymphatic space invasion.
Stratification
Relapse
3.0 cm 3.0–5.9 cm 6.0 cm Spermatocytic Classical Anaplastic No Yes No Yes 6cm 6cm 34 years 34 years No Yes No Yes 4cm 4cm No Yes No Yes
6% 18% 36% 0% 16% 33% 14% 23% 14% 23% 12% 33% 9% 21% 14% 31% 10% 20% 13% 24% 14% 24% 14% 23%
1 Adapted from Milosevic, et al. Management of testicular seminoma. Sem Surg Oncol 1999;17:243. 2 Significant on multivariate analysis.
142
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
necessary in surveillance programs must also be taken into account, but patients are seen at most 3 times per year and in many instances no loss of time from work is necessary. The cost of any surveillance protocol is determined primarily by the frequency of follow-up and the optimal frequency and duration of patient assessment on surveillance protocols is currently being determined. With further analysis of prognostic factors for occult disease, it may well be possible to tailor follow-up strategy based on individual patients’ risk of relapse. At the present time, CT scans of the entire abdomen and pelvis are routinely performed as part of the surveillance protocols and with further data on the sites of relapse it may be possible to refine this strategy and be more selective (and therefore more cost efficient) in our approach. Further data is necessary on the optimal followup strategy for both approaches before a meaningful economic analysis can be done.
5. Discussion The available data from surveillance and adjuvant RT series suggests that almost 100% of patients with stage I testicular seminoma are cured, whichever management strategy is adopted after orchidectomy. One of the most attractive features of surveillance is the ability to avoid treatment in the majority of patients. However, surveillance must not compromise survival and it is reassuring to find that there have been only four disease-related deaths reported in the 826 patients in the surveillance studies. While isolated occurrences of second non-testicular tumor following treatment of seminoma with RT have been reported for decades, there is now convincing data from long-term studies that patients with seminoma treated with RT have an increased risk of developing a second malignancy. An increased risk of second cancers has been documented in the Netherlands and Denmark, and since this increased risk is expressed more than 10 to 15 years following RT, it may not be apparent in series with shorter follow-up [17,18]. The largest study of second cancers in long-term survivors of testicular cancer was conducted by Travis et al. at the National Cancer Institute Cancer Epidemiology Division [19]. Over 28,000 patients with testis cancer, including over 15,000 with seminoma from 16 population-based registries worldwide, were evaluated. Overall, 1,406 second cancers, excluding contralateral testis, occurred against 981 expected (observed/expected [O/E]1.43). An excess of rectal and small intestine cancer was seen in seminoma patients. The actuarial risk of developing a second malignancy, excluding contralateral cancers, increased over time from diagnosis of testicular cancer up and was 18.2 percent at 25 years. Secondary leukemia was linked with RT and chemotherapy, while an excess of the stomach, bladder, and possibly pancreas tumours was associated with prior RT. This is a major reason to consider surveillance and recommend it to patients with stage I seminoma.
One of the concerns about surveillance is the potential for more patients to require chemotherapy, which involves a prolonged and more toxic course of treatment. In the 1981– 1994 PMH experience, similar numbers of patients managed with surveillance (12/225, 5.3%) and adjuvant RT (10/245, 4.1%) received chemotherapy [9]. This concern is difficult to address, because, in some centres, chemotherapy was used in all patients at relapse. This approach increases the proportion of patients receiving chemotherapy but likely reduces the number of patients developing a second relapse. Also, some patients were placed on surveillance because they had received prior RT or had contraindications to RT, such as inflammatory bowel disease. Chemotherapy with cisplatin and etoposide is very effective in advanced seminoma and is an alternative to RT or surveillance in stage I seminoma. The use of 1 to 2 courses of carboplatin as adjuvant therapy after orchidectomy has been reported by Oliver et al. [20], who treated 78 patients (53 patients had 2 courses; 25 patients had 1 course). With a median follow-up of 44 months, only 1 patient has relapsed. Dieckman et al. (with a median follow-up of 48 months) reported 8 relapses in a cohort of 125 patients treated in a similar fashion (93 patients had 1 course of carboplatin; 32 had 2 courses) [21]. Treatment was well tolerated and acute and intermediate term toxicity was mild. The carboplatin dosage scheme used in these 2 studies was different; in the Oliver et al. study the carboplatin dose was based on area under the curve formula of Calvert et al (AUC) [22], while in the Dieckman report the dosage of Carboplatin was 400 mg/m2. Basing the dose on body surface area rather than AUC leads to a significantly lower dose being given (e.g., 34-year-old man with normal renal function — GFR of 110 ml/minute based on EDTA clearance and BSA of 1.787 would receive 715 mg under the BSA approach but 945 mg using the Calvert formula). A single adjuvant carboplatin injection may well be an attractive treatment option for patients at moderate to high risk of relapse on surveillance protocols. However, before this approach can be generally adopted, further study is necessary. The MRC UK are presently conducting a phase III study comparing adjuvant carboplatin to standard retroperitoneal RT in stage I seminoma and the National Cancer Institute of Canada Clinical Trials Group is currently planning a study comparing adjuvant carboplatin to surveillance. RPLND is another treatment option available in seminoma. It is a reasonable approach for patients unwilling to comply with surveillance or those unable to be treated with RT, such as those with inflammatory bowel disease or those previously treated with retroperitoneal RT [23]. It may also be the best approach in patients with concurrent or previous malignancy in whom histologic examination of lymph nodes is essential to plan treatment. While physicians may view one management approach as preferable, individual patients’ preferences must also be considered. These preferences may be based on many socioeconomic factors (e.g., a desire to travel and consequent inabil-
A. Bayley et al. / Urologic Oncology 6 (2001) 139–143
ity to be available for follow up) as well as the side effects associated with both approaches. Also, some patients may view the risk of second malignancy 20–25 years in the future as being of little importance and opt for adjuvant RT to minimize their more immediate risk of relapse. The challenge for physicians involved in treating patients with stage I seminoma is to recommend the most appropriate management for individual patients. Clearly, with surveillance programs, many patients can avoid unnecessary therapy but a proportion of patients will develop disease progression and require treatment. In the future, the identification of prognostic factors for occult disease will allow patients and clinicians to choose management based on a more accurate assessment of an individual patient’s risk of relapse. Patients at a high risk of relapse may opt for adjuvant treatment after surgery. In summary, surveillance is a good option for patients with stage I seminoma. With further research it may be possible to refine our treatment recommendations for individual patients based on prognostic factors for relapse. We currently offer patients surveillance to all patients with stage I seminoma and we believe that the routine use of post-operative RT in stage I seminoma should be reconsidered and a surveillance program offered to all patients as an alternative management option.
References [1] Greenlee R, Murray T, Bolden S, Wingo P. Cancer Statistics 2000. CA: A cancer journal for clinicians 2000;50(1):7–33. [2] Buetow SA. Epidemiology of testicular cancer. Epidemiol Rev 1995; 17(2):433–49. [3] Dosmann MA, Zagars GK. Post-orchiectomy radiotherapy for stages I and II testicular seminoma. Int J Radiat Oncol Biol Phys 1993; 26(3):381–90. [4] Duncan W, Munro AJ. The management of testicular seminoma: Edinburgh 1970–1981. Br J Cancer 1987;55(4):443–8. [5] Fossa SD, Aass N, Kaalhus O. Radiotherapy for testicular seminoma stage I: treatment results and long-term post-irradiation morbidity in 365 patients. Int J Radiat Oncol Biol Phys 1989;16(2):383–8. [6] Giacchetti S, Raoul Y, Wibault P, Droz JP, Court B, Eschwege F. Treatment of stage I testis seminoma by radiotherapy: long-term results — a 30-year experience. Int J Radiat Oncol Biol Phys 1993; 27(1):3–9. [7] Kellokumpu-Lehtinen P, Halme A. Results of treatment in irradiated testicular seminoma patients. Radiother Oncol 1990;18(1):1–7. [8] Sommer K, Brockmann WP, Hubener KH. Treatment results and acute and late toxicity of radiation therapy for testicular seminoma. Cancer 1990;66(2):259–63. [9] Warde P, Gospodarowicz MK, Panzarella T, et al. Stage I testicular seminoma: results of adjuvant irradiation and surveillance. J Clin Oncol 1995;13(9):2255–62. [10] Willan BD, McGowan DG. Seminoma of the testis: a 22–year experience with radiation therapy. Int J Radiat Oncol Biol Phys 1985; 11(10):1769–75.
143
[11] Sturgeon JF, Jewett MA, Alison RE, et al. Surveillance after orchidectomy for patients with clinical stage I nonseminomatous testis tumors. J Clin Oncol 1992;10(4):564–68. [12] Nicolai N, Pizzocaro G. A surveillance study of clinical stage I nonseminomatous germ cell tumors of the testis: 10-year followup. J Urol 1995;154(3):1045–9. [13] von der Maase H, Specht L, Jacobsen GK, et al. Surveillance following orchidectomy for stage I seminoma of the testis. Eur J Cancer 1993;14(4):1931–34. [14] Warde P, Gospodarowicz M, Panzarella T, et al. Long term outcome and cost in the management of stage 1 testicular seminoma. Can J Urol 2000;7:967–72. [15] Warde P, Gospodarowicz M, Banerjee D, et al. Prognostic factors for relapse in stage I testicular seminoma treated with surveillance. 1997; J Urol 157:1705–09. [16] Sharda N, Kinsella T, Ritter M. Adjuvant radiation versus observation: a cost analysis of alternate management schemes in early-stage testicular seminoma. J Clin Oncol 1996;14:2933–9. [17] van Leeuwen F, Stiggelbout AM, van den Belt-Dusebout AW, et al. Second cancer risk following testicular cancer: A follow-up study of 1,909 patients. J Clin Oncol 1993;11(3):415–24. [18] Moller H, Mellemgaard A, Jacobsen GK, Pedersen D, Storm HH. Incidence of second primary cancer following testicular cancer. Eur J Cancer 1993;5:672–6. [19] Travis L, Curtis R, Storm H, et al. Risk of second malignant neoplasms among long-term survivors of testicular cancer. J Natl Cancer Inst 1997;89(19):1429–39. [20] Oliver RT, Edmonds PM, Ong JY, et al. Pilot studies of 2 and 1 course carboplatin as adjuvant for stage I seminoma: should it be tested in a randomized trial against radiotherapy? Int J Radiat Oncol Biol Phys 1994;29(1):3–8. [21] Dieckmann KP, Krain J, Kuster J, Bruggeboes B. Adjuvant carboplatin treatment for seminoma clinical stage I. J Cancer Res Clin Oncol 1996;122(1):63–6. [22] Calvert H, Judson I, van der Vijgh W. Platinum complexes in cancer medicine: pharmacokinetics and pharmacodynamics in toxicity and therapeutic activity. Cancer Surv 1993;17:189–217. [23] Willett C, Ooi C, Zietman A, et al. Acute and late toxicity of patients with inflammatory bowel disease undergoing irradiation for abdominal and pelvic neoplasms. Int J Radiat Oncol Biol Phys 2000;46:995–8. [24] Horwich A. Surveillance for stage I seminoma of the testis. In: Horwich A, editor. Testicular cancer. Investigation and management. London: Chapman and Hall Medical, 1991. pp 109–16. [25] Ramakrishnan S, Champion AE, Dorreen MS, Fox M. Stage I seminoma of the testis: is post-orchidectomy surveillance a safe alternative to routine postoperative radiotherapy? Clin Oncol 1992;4(5): 284–6. [26] Germa Lluch JR, Climent MA, Villavicencio H, et al. Treatment of stage I testicular tumours. Br J Urol 1993;71(4):473–77. [27] Allhoff EP, Liedke S, de Riese W, Stief C, Schneider B. Stage I seminoma of the testis. Adjuvant radiotherapy or surveillance? Br J Urol 1991;68(2):190–4. [28] Miki T, Maeda O, Saiki S, et al. Surveillance after orchiectomy for stage I testicular seminoma. Nippon Hinyokika Gakkai Zasshi 1992; 83:1789–94. [29] Yonese J, Kawai T, Yamauchi T, Ueda T, Kawakami S, Ishibashi K. Surveillance following orchiectomy for stage I testicular tumor. Nippon Hinyokika Gakkai Zasshi 1993;84:1804–10. [30] Charig MJ, Hindley AC, Lloyd K, Golding SJ. ‘Watch policy’ in patients with suspected stage I testicular seminoma: CT as a sole staging and surveillance technique. Clin Radiol 1990;42(1):40–1.