EDITORIAL Radiotherapy in Stage I Seminoma of the Testis Alan Horwich Academic Unit of Radiotherapy and Oncology, Royal Marsden National Health Service Trust, and Institute of Cancer Research, Sutton, Surrey, United Kingdom The analysis of excess mortality in a cohort of 453 men cured of early stage testicular seminoma by orchiectomy and radiotherapy at the M.D. Anderson Cancer Center (MDACC) [1], presented in this issue of the Journal of Clinical Oncology, raises questions about current standard practice, especially in stage I disease, in which only 1 in 5 men actually has subclinical metastases. The authors reported a standardized mortality ratio (SMR) of 1.59 emerging 15 years following radiotherapy, and specific analyses revealed a cardiac SMR of 1.61, and a cancer SMR of 1.91. In most countries of the world, adjuvant retroperitoneal node radiotherapy is the postorchiectomy standard of care for patients with stage I testicular seminoma, the most common presentation of testicular cancer. Though radiation is known to have carcinogenic potential, the practice of adjuvant radiotherapy in seminoma is based on the perception that the relatively low dose of radiotherapy delivered as an adjuvant to abdominal lymph nodes should be associated with a low excess risk of cancers and would not cause severe heart damage [2-4]. The report from the MDACC challenges these perceptions. Questions to address include the precision and generalizability of the analyses, their relevance to modern radiotherapy practice and the safety of alternative approaches to management of stage I seminoma. The study was based on patients treated between 1951 and 1999. Although those studies from the earlier years contributed more to the number of person-years at risk and to the number of events, their staging and treatment methods were least like current practice. Whereas it is now accepted that radiation fields should be restricted to the para-aortic region of the abdomen, and the dose to 20 Gy [5,6], the patients in the MDACC series had much more extensive radiotherapy. For example, 31 men received whole-abdominal radiotherapy, and, before 1971, most men had prophylactic mediastinal radiotherapy—a significant risk factor for long-term toxicity in this and other analyses. It is notable that the most common second cancer in the MDACC report was lung cancer. Additionally, even in recent years at the MDACC, patients with bulky abdominal metastases as large as 10 cm in diameter were treated with large radiation fields [7]; whereas in most centers, they would have been treated with chemotherapy [8-10]. Retrospective cohort analyses are prone to the systematic bias that patients suffering an event are more likely to be known about than those who remain well. In this series, the median follow-up of survivors was 13.3 years, even though half the patients had been treated more than 20 years earlier. Data from most of the cohort was censored before the period when excess toxicities occurred, which amplifies the change seen in actuarial risk figures. The SMR is based on comparison with US age-specific and cause-specific mortality. The implicit assumption is that the population of men treated for testicular cancer at the MDACC was typical of the US male population in terms of risk factors. If there were regional variations in risk factors, such as tobacco abuse, ethnicity, diet, deprivation, or sun exposure, then overall US mortality would represent an imprecise control, though there is no evidence to suggest whether this might increase or decrease the relative risks (RRs) that were calculated. Though there is evidence in the literature for increased cardiovascular risk and for increased incidence of second cancer after cure of testicular tumors, most reports either rely on epidemiological data, which may not have accurate and complete treatment records, or arise also from management of nonseminomas, for which a variety of chemotherapeutic agents have been used and for which radiotherapy doses were higher. Travis et al [11] analyzed second malignancies in 28,843 patients surviving at least 1 year after diagnosis of testicular cancer, treated between 1935 and 1993, and who had been registered within 16 population-based cancer registries in the United States, Canada, Sweden, Finland, Denmark, and the Netherlands. Seven thousand four hundred seventy-six men with seminoma had initial treatment with radiotherapy; data on any subsequent treatment were not available. There was a significantly elevated RR of second cancers (RR, 1.45; P = .05), but this was found even during the first 4 years after treatment—an uncharacteristic timescale for radiation carcinogenesis. In the modern era (1975-1993), the predominant elevated cancer risk was leukemia, with six cases among 5,363 men treated. Van Leeuwen et al [12] analyzed 1,909 patients treated in the Netherlands between 1971 and 1985 (a subset of those reported by Travis et al [11]). Seventy-eight developed second cancer (47.6 expected; RR, 1.6 [95% CI, 1.3 to 2.1]), with particular risks for stomach cancer (RR, 3.7), leukemia (RR, 5.1), and contralateral testicular cancer (RR, 35.7). There was no excess risk of lung cancer. Fosså et al [13] analyzed 876 patients treated for testicular cancer at the Norwegian Radium Hospital between 1956 and 1977, representing two-thirds of the total Norwegian cases in the period. Second cancers were identified in the National Cancer Registry. Seminoma patients received 36 to 40 Gy, and nonseminomas were treated with 50 Gy. Patients with metastases all had additional mediastinal radiotherapy. Although in the whole cohort there was a significantly raised risk of second cancer (RR, 1.58; P < .01), for those 579 treated only with infradiaphragmatic radiotherapy, the RR of 1.32 was not statistically significant. Finally, an analysis confined to seminomas treated initially with radiotherapy postorchiectomy in London, England, between 1961 and 1980 [3] included 859 patients, with mean follow-up per case of 10 years. Death certificates did not identify overall, an excess of death from other causes; though the analysis confined to those patients more than 10 years after diagnosis found a relative mortality risk of 1.31 (95% CI, 0.95 to 1.81). There were 20 deaths from second cancers (22.6 expected). In the analysis of second cancer incidence, the RR was 1.27 (95% CI, 0.92 to 1.75); the only notable subtype with excess incidence was acute leukemia with four cases, representing a RR of 6.2. For late cardiac toxic events, there is evidence of increased risk after chemotherapy of germ cell tumors. This has been found both in comparison with population statistics [14] and in comparison with testicular tumor patients treated by surveillance [15]. Both cisplatin and bleomycin [16] can cause vascular side effects, but further data are needed before either of these drugs can be definitively implicated in the etiology of the late cardiac events. Radiotherapy to the heart has been linked to late risk of ischemic heart disease and cardiac events in Hodgkin's disease [17,18] and in left-sided breast cancer [19]. There is some evidence that, in early seminoma, prophylactic mediastinal radiotherapy may be associated with late toxicity [20,21]. For example, the Patterns of Care Study, based on 387 men treated with radiotherapy in 1973 and 1974, identified 10 cardiac deaths (4.4 expected), and eight of the 10 had received mediastinal radiotherapy [21]. Analyses of patients with Hodgkin's disease have suggested that the risk depends on field and dose [18,22], and it is surprising to see any effect with only 25 Gy, typically used in seminoma for prophylactic mediastinal treatment. It is even more puzzling to consider an effect from similarly low doses confined to para-aortic and pelvic radiation fields; however, the relationship is supported by the study we reported recently [15] based on 992 Royal Marsden patients treated for testicular cancer between 1982 and 1992, in which various treatment groups were compared with a group treated by orchiectomy and surveillance for stage I disease. Overall, 1.8% died of cardiovascular disease, and 6.9% had evidence of it. The groups were not matched for disease histology or stage. The radiotherapy group (predominantly stage I seminomas treated with 30 Gy to abdominal and pelvic nodes) had a significantly increased risk of a cardiac event (RR, 2.74; P = .013). In multivariate analysis of the risk in surviving patients, the significant predictive factors were age and serum concentrations of magnesium, free thyroxin, and total protein; treatment group was not significant. There were no differences in blood pressure between treatment groups on follow-up. Even a low risk of long-term toxicity is a cause for concern, especially in the setting of a tumor treated successfully in more than 95% of cases, and with patients in young adult life. Recent research has led to changes in practice to reduce these risks. First, more precise staging techniques and platinum-based chemotherapy have reduced the indications for radiotherapy, now recommended only in the adjuvant therapy for stage I disease or treatment of small-volume stage II disease [23,24]. Second, prophylactic mediastinal radiotherapy is not used [21,23]. Third, a UK Medical Research Council (MRC) prospective randomized trial in 478 patients with stage I seminoma comparing the traditional field (para-aortic and pelvic nodes) with a reduced field (para-aortic nodes) showed the same low recurrence rate, 9 (4%) in each arm; relapse in the pelvis was seen in 4 patients, all treated with the para-aortic field alone [5]. Fourthly, a prospective randomized MRC/European Organization for Research and Treatment of Cancer (EORTC) trial in 625 patients comparing a dose of 30 Gy with one of 20 Gy showed equivalent disease control [6,25]. Thus, at least in the United Kingdom, the standard postorchiectomy treatment for patients with stage I seminoma is adjuvant radiotherapy to para-aortic nodes only to a dose of 20 Gy. The long-term risks of this policy may be very low. Alternatives to adjuvant radiotherapy in stage I seminoma have been investigated, but not yet established, by results from comparative trials. Recently, a pooled analysis was reported of 638 patients treated by surveillance postorchiectomy in Denmark, at the Princess Margaret Hospital (Toronto, Ontario), at the Royal London Hospital, and at the Royal Marsden Hospital [26]. The estimated recurrence risk by 10 years was 21%; although most recurrences were within 3 years, eight of the 121 relapses occurred more than 6 years postorchiectomy. Treatment for recurrence was generally successful, and there were only six (0.9%) cause-related deaths. Multivariate analysis revealed that primary tumor size more than 4 cm diameter and rete testis invasion were independent risk factors for recurrence; the 5-year recurrence risk was 12.2% with neither, 15.9% with one, and 31.5% with both factors present. Though surveillance can be a reasonable option for low-risk patients, and seems acceptable to patients [27], disadvantages of this policy are the long period of careful follow-up required, the lack of a sensitive serum marker, and the risk of moderately advanced disease on recurrence, requiring systemic chemotherapy for some patients. An alternative to surveillance alone is the use of adjuvant chemotherapy, which has been demonstrated to be effective in stage I nonseminoma [28]. There have, as yet, been few reports of this approach in stage I seminoma. Dieckmann et al [29] found an 8.6% relapse rate in 93 patients followed up, for a median of 4 years after a single cycle of adjuvant carboplatin; however, the dose may have been suboptimal. Steiner et al [30] treated 108 patients with two cycles of carboplatin, and with mean follow-up of 5 years, there were only two relapses. An MRC/EORTC intergroup prospective randomized trial recruited more than 1,300 patients and compared adjuvant radiotherapy with a single cycle of adjuvant carboplatin chemotherapy. This was designed to show equivalence, and the result will be reported during 2004. With the caveat that the long-term carcinogenic risk of carboplatin has not been evaluated, the trial may provide level 1 evidence supporting a further option for treatment of patients with stage I seminoma. In conclusion, Zagars et al [1] have highlighted important issues for the management of a highly curable cancer. Patients offered adjuvant radiotherapy following orchiectomy for stage I seminoma should be advised that there is a potential carcinogenic risk. With modern planning techniques and low-dose radiotherapy confined to para-aortic nodes, the risk should be low; however, for those who find the risk unacceptable, there is preliminary evidence that surveillance may be a reasonable alternative. Substantive research results on adjuvant chemotherapy are awaited. The MDACC report also implicates radiotherapy in late cardiac mortality, and with mediastinal fields, there is support in the literature for this concern, albeit associated with higher doses. At present, there is insufficient evidence to regard low-dose para-aortic radiotherapy as a cause of significant cardiac morbidity; mechanistic research would be useful. Nevertheless, it would be prudent to consider whether the upper field border needs to extend as high as the D10 vertebral level, and to use careful planning techniques to minimize renal irradiation. The author indicated no potential conflicts of interest. Originally published as JCO Early Release 10.1200/JCO.2004.12.966 on January 15 2004 Journal of Clinical Oncology, Vol 22, No 4 (February 15), 2004: pp. 585-588 Editorial/RTx testicular J Clin Onc, 2/04 Br J Cancer, 5/04 Remember we are NOT Doctors and have NO medical training. This site is like an Encylopedia - there are many pages, many links on many topics. Support our work with any size DONATION - see left side of any page - for how to donate. You can help raise awareness of CAM.