Drs. Foster and Nichols, both recognized experts on testicular cancer, are to be congratulated for their concise overview of continuing controversies and challenges in the care of young men with this disease. Both the research group at Indiana University led by Dr. Foster and our group at Walter Reed Army Medical Center and the Armed Forces Institute of Pathology have been very interested in using primary tumor factors to predict the risk of occult retroperitoneal disease in men with clinical stage I disease. Although the two groups disagree about the clinical utility of Ki-67 (MIB-1) immunohistochemical staining of the primary tumor, we agree that a prospective, multicenter, consecutive clinical trial is necessary to determine which combination of factors has the best prognostic accuracy. Under the guidance of Dr. Foster and the Indiana group, a trial was recently opened by the Eastern Cooperative Oncology Group (ECOG) to address this issue. I applaud and support this initiative.
Updated Results From Walter Reed
Until this clinical trial is completed, our group continues to find that the percentage of embryonal carcinoma component combined with the presence or absence of vascular invasion in the primary tumor are powerful prognostic factors. We recently updated our experience examining vascular invasion, percentage of embryonal carcinoma, p53 expression, bcl-2 expression, MIB-1(Ki-67) expression, cathepsin D expression, and E-cadherin expression in 149 patients with clinical stage I nonseminomatous germ-cell tumor.
Table 1 provides results in the cases stratified by findings at retroperitoneal lymph node dissection (pathologic stage I or II), along with statistics from the univariate and multivariate analyses. Like our prior work, vascular invasion and percentage of embryonal carcinoma were the only primary tumor factors that remained significant in our more recent multivariate analysis.
We agree with Foster and Nichols that it is of paramount importance to be able to identify clinical stage I nonseminoma patients who can be managed by surveillance. To this end, our updated study of 149 patients can identify this low-risk group using a threshold of £ 45% embryonal carcinoma and the absence of vascular invasion. Table 2 provides the various cut-off points for percentage of embryonal carcinoma with or without vascular invasion, and Table 3 provides the sensitivity, specificity, and predictive values for these parameters.
Until the ECOG prospective trial is completed, we feel that it is important to carefully assess the primary tumor for vascular invasion and percentage of embryonal carcinoma. We further believe that an embryonal cut-off percentage of £ 45% can be used to select patients at very low risk of occult metastases/recurrence.
In our experience, none of the molecular or proliferation staining bio-markers studied to date adds to the prognostic ability of percentage of embryonal carcinoma and vascular invasion. We do not feel that these biomarkers should be used outside of a research setting.