There are no randomized studies of surveillance alone compared to adjuvant therapy. This creates a challenge in articulating the options that optimize cure, expeditiously allow all patients to return to their lives, and avoid patient exposure to interventions that may lead to permanent long-term adverse events. The other challenge to recommending a management option for stage I seminoma is that the available long-term toxicity/survival data are retrospective, with all the inherent problems associated with retrospective data, and yet these data show a clear pattern of treatment-related deaths that cannot be ignored.
The data that exist suggest that virtually all patients with stage I testicular seminoma are cured regardless of the post-orchidectomy management. The five-year survival reported in all the studies identified in this systematic review was over 95%, regardless of management strategy, including surveillance alone with no adjuvant therapy. In non-randomized studies of surveillance alone, the five-year relapse-free rate was consistently reported as over 80%, with no reduction in cause-specific or overall survival. Therefore, it appears that the majority of patients are cured by orchidectomy alone, and those that are not, rarely die from their disease. The available data therefore support the conclusion that surveillance as a management option does not compromise survival. Given this fact, and the acute and long-term toxicity of adjuvant treatment, especially in terms of second malignancies, the use of any form of adjuvant therapy must be given careful consideration. Any adjuvant treatment regime would expose the 80% of patients who would never have a relapse and would be cured by orchidectomy alone to the risk of treatment-related toxicity, a serious consideration given the retrospective data concerning second malignancies and cardiac effects.
The studies that have evaluated radiation therapy (RT) in testicular cancer all report clinically important increases in second malignancy, and treatment is associated with other significant toxicities such as cardiac toxicity. Although changes in the field size and RT dose did occur during the time period examined in these studies, such changes are unlikely to have a large effect on the estimation of risk, as any RT given (regardless of the dose/field delivered) is associated with increase in second malignancy, and it is the absolute size of the risk that may be affected by dose and field size issues. Although the RT treatment given today is not exactly the same as that given to the patients in these long-term toxicity studies, it is sufficiently similar that these issues cannot be ignored or dismissed as being irrelevant to current treatment practices. While further prospective study of these issues would in many ways be ideal, the large numbers of patients needed, and also the long periods of time over which such data needs to be collected, limits the ways in which this information can be obtained. Further clarification of the issue will always be hampered by the inherent difficulties associated with retrospective and non-randomized studies. Further, only a small minority, if any, of patients in the long-term toxicity studies, are likely to have received single-agent carboplatin; thus it is not currently possible to comment definitively on any associated long-term toxicity associated with that treatment.
Surveillance may have an advantage over adjuvant therapy in that both acute and long-term toxicity may potentially be avoided; however, surveillance requires a commitment to more intense and prolonged follow-up from both patients and clinicians. Patient compliance is essential, as the failure to detect relapse at an early stage may compromise survival. In addition, it must be noted that repeated exposure to serial computed tomography (CT) scans poses some potential risk of second malignancy, albeit less significant than that posed by adjuvant RT. Therefore, the disadvantage of surveillance as a management strategy is that follow-up for surveillance requires more frequent visits and imaging to detect relapse when compared to patients who have received adjuvant therapy. Despite these drawbacks, all the guidelines found and evaluated included surveillance as a treatment option for stage I seminoma, and where the treatments were ranked in order of preference, surveillance was the primary option.
Surveillance has become a well-established management option worldwide. It seems that all men with stage I seminoma should be suitable candidates for surveillance as long as they are able to undergo the follow-up and CT scan procedures. More importantly, these men should have full commitment to be compliant with the designated surveillance schedule. Noncompliance may lead to more advanced disease when relapse is detected clinically, potentially requiring more aggressive treatment for a cure.
There will still be many patients who may choose to receive adjuvant therapy. When adjuvant therapy is chosen, RT remains an option for patients. In the randomized trial reported by Jones et al, 20 Gy (2 Gy/day) was shown to be equivalent to 30 Gy in terms of disease control. One of the rationales for using 20 Gy was to reduce toxicity. While acute toxicity was improved, the follow-up in this trial is insufficient, and it may be underpowered to identify if there is a benefit with respect to long-term toxicity or second malignancy; however, 20 Gy has the advantage of an overall shorter treatment time with good disease control. There is some variation as to what is considered to be the standard radiation dose for stage I seminoma. Consideration should be given not only to the total dose but also to the dose per fraction. In some non-randomized studies, a total dose of 25 Gy given in 1.25 Gy per fraction has provided good in-field local control with low rates of acute toxicity.
In the randomized trial reported by Fossa et al a reduced para-aortic field size was compared to standard extended-field ("dog-leg") RT, with the hypothesis that a reduced field size would lead to reduced toxicity and second malignancy. While the trial demonstrated equivalence between the field sizes in terms of overall prevention of relapse, and also showed reduced acute toxicity, the follow-up is not sufficient to judge any reduction in long-term toxicity or second malignancy. One issue that does arise from this trial and the one reported by Jones et al was that the pattern of relapse was altered. While there was no difference in the overall number of recurrences, in both randomized controlled trials the pelvis was the most common site of relapse in patients treated with para-aortic RT, while pelvic relapse was rare for patients treated with extended-field RT. This is supported by an examination of patterns of relapse in patients in the non-randomized studies. Although only a small proportion of patients ultimately relapse in the pelvis, a pelvic recurrence is a serious event that is not easily detected at an early stage unless a CT scan is used. Therefore, all patients treated with para-aortic RT still require follow-up CT scans of the pelvis, an investigation that is not needed in patients treated with extended-field RT.
Neither of these modified treatments is likely to completely eliminate the risk of second malignancy, and any associated risk reduction remains unknown at this time. Thus, while para-aortic RT to a minimum dose of 20 Gy in 2 Gy fractions is the RT option that may best reduce acute toxicity, owing to concerns about the additional follow-up needed and pelvic relapses, extended-field RT may still be appropriate.
Data regarding the effects of adjuvant carboplatin therapy are limited, and the duration of follow-up is relatively short; thus, in contrast to RT, more questions remain regarding its use. The conclusion of the randomized trial reported by Oliver et al was that carboplatin was equivalent to RT for prevention of short-term relapse, with improved acute toxicity. However, similar to the reduced-field RT trial discussed above, the pattern of relapse in patients treated with carboplatin was altered such that the majority of the relapses occurred in the retroperitoneal/para-aortic lymph nodes. Given these findings, continued CT monitoring for relapse cannot be eliminated from the follow-up schedule: indeed it should mirror that recommended for surveillance. This trial also has insufficient follow-up to evaluate the durability of disease control and the long-term toxicity of carboplatin in this patient population, as compared to RT. In a meta-analysis of sarcoma patients performed by Tierney et al, adjuvant chemotherapy showed a short-term benefit in the recurrence rate; however, overall survival did not appear to be affected, implying that recurrences may have been delayed as opposed to prevented. Without long-term survival data for chemotherapy in the treatment of seminoma, there is the possibility that recurrences have just been delayed and that late recurrences may still occur. In light of these issues, the use of carboplatin might be best restricted to situations in which there is a contraindication to RT or within a clinical trial.
Given that there are several management options, none of which have proven to have absolute superiority for patients with stage I testicular seminoma, men should be counselled concerning their treatment and the trade-offs associated with the different options after orchidectomy. While physicians may view one management approach as preferable, individual patient preferences must be considered. An individual treatment plan that takes into account the patient's wishes and is developed in consultation with an expert in the treatment of stage I seminoma should be developed for each patient. A summary of the benefits and risks of the different management strategies that physicians may wish to share with their patients appears in Table 5 of the original guideline document.
