Prognostic risk factors in low stage testicular germ cell tumors: unanswered questions regarding clinically useful prognosticators for extratesticular disease.

Abstract

The counterpoint by Heidenreich et al.1 and the article by Albers et al.2 address the issue of prognostic risk factors in patients with low stage testicular germ cell tumors. Both groups agree that new factors should provide independent prognostic information compared with recognized risk factors and that prospective validation of new markers predicting the risk of occult metastasis is needed. The latter point deserves emphasis because both current staging methods and tumor marker analysis by histopathology and immunohistochemistry inherently are subjective and variable. The literature is replete with new risk factors identified on the basis of retrospective analysis that either have not been validated or fail to fulfill their promise of segregating populations at risk when they are applied in rigorous prospective studies.3 For example, retrospective analysis of our own long term study of 105 Stage I testicular tumor patients on surveillance shows that both vascular space invasion within the primary testicular tumor and embryonal carcinoma histology significantly predict the probability of recurrence compared with patients who do not have these risk factors.4 Conversely, a prospective study in 76 unselected Stage I testicular tumor patients undergoing retroperitoneal lymph node dissection only shows a trend in favor of vascular invasion predicting the risk of occult metastasis (48% of patients with and 33% without vascular invasion had microscopic lymph node metastasis; P = 0.24).5 Furthermore, "newer" prognostic factors must be balanced against improvements in conventional staging methods. For example, two studies5, 6 show that revised computed tomography interpretation of retroperitoneal metastasis based on lymph node size (3-4 mm) and distribution (anterior vs. posterior) can predict > 90% of Stage I testicular tumor patients who harbor occult lymph node disease. A prognostic factor-based staging system has been proposed for patients with metastatic germ cell carcinoma7 and a similar model combining clinical, histologic, and tumor markers needs to be devised for patients with low stage testicular carcinoma. Prospective validation of such a model in multiinstitutional studies will help to define optimal strategy (observation vs. treatment) for patients with low volume testicular tumors. It is time to move forward. Harry W. Herr M.D.*