Increase in tumor markers (human chorionic gonadotropin [HCG] and α-fetoprotein [AFP]) are correlated to relapse in non-seminomatous germ cell tumors (NSGCT). Thirty percent of NSGCT patients do not produce one of these markers. Much effort has been devoted to the search for new tumor markers. The TRA-1-60 antigen has been related to the presence of embryonal germ cell carcinoma (EC) and carcinoma in situ and can be detected by way of a 2-step immunoenzymometric assay in the blood of patients with EC.1– The antigen, characterized by Badcock et al.1 is a sialylated keratan sulfate proteoglycan. Our study further investigated the clinical efficacy of TRA-1-60 as a serum tumor marker for germ cell cancer in the testis.
TRA-1-60 antigen has been related to the presence of embryonal germ cell carcinoma (EC) and carcinoma in situ. Our study further investigated the clinical efficacy of TRA-1-60 as a serum tumor marker for germ cell cancer in the testis. Three groups of patients with germ cell tumors were included: Group 1, 34 patients with disseminated disease (24 nonseminomatous germ cell tumors [NSGCT] and 10 seminomatous germ cell tumors [SGCT]); this group of patients were followed during the course of chemotherapy with measurements of TRA-1-60, HCG and AFP; Group 2, 28 patients with Stage I NSGCT (22 with embryonal carcinoma [EC]-component and 6 without EC-component, median follow-up 15 months); and Group 3, 40 patients with Stage I pure SGCT (median follow-up 15 months). Seventy-eight percent of patients with disseminated EC-positive NSGCT had increased levels of TRA-1-60 before chemotherapy. After chemotherapy, levels of TRA-1-60 had dropped significantly (p < 0.01). Levels of TRA-1-60 did not normalize in 15% of NSGCT and 30% of SGCT patients after chemotherapy. This was not associated with recurrent disease. Approximately one-third of patients with Stage I NSGCT had increased values of TRA-1-60 during follow-up without having a relapse. Contrary to earlier reports TRA-1-60 is not at present useful as a tumor marker in patients with germ cell tumors. Although detecting a few early relapses the rate of false positive elevations in the tumor marker makes it unreliable in the clinical setting. Our study did confirm that elevated levels of TRA-1-60 were present in ∼80% of patients with disseminated EC-positive NSGCT before start of chemotherapy and chemotherapy induced a significant decrease in levels of TRA-1-60. Thus, the TRA-1-60 antigen might still prove clinically useful provided that the reliability of the assay can be increased. © 2002 Wiley-Liss, Inc.
MATERIAL AND METHODS
Three groups of patients with germ cell tumors were included: Group 1, 34 patients with disseminated disease (24 NSGCT and 10 seminomatous-GCT); this group of patients were followed during the course of chemotherapy and until evaluation 4 weeks after chemotherapy with measurements of TRA-1-60, HCG and AFP; Group 2, 28 patients with stage I NSGCT (22 with EC-component and 6 without EC-component, median follow-up 15 months); and Group 3, 40 patients with Stage I pure SGCT (median follow-up 15 months). The local ethics committee approved the study and informed consent was obtained from all patients according to Helsinki II. Patients with disseminated disease received either 3 or 4 courses of BEP. Difference in mean was established using non-parametric test (Wilcoxon) in SPSS software.
TRA-1-60 was measured by a 2-step enzyme immunometric assay as described previously in detail by Marrink et al.3 In the first step of this assay antigen (sample or standard) is incubated over night with excess monoclonal anti-TRA-1-60 antibody, after which the antigen/antibody complex is precipitated by 2.5% polyethylene-glycol (PEG). In the second step the non-bound antibodies in the supernatant are transferred to antigen-coated wells of a micro-titer plate. After 2 hr incubation and subsequent 4 rounds of washing the amount of unbound antibody is quantified by its ability to bind to the coated antigen as detected by a peroxidase-coupled second antibody (code P260, DAKO, Denmark).
The NT2/D1 cell clone was used as source of antigen. The NT2/D1 cells were harvested and washed in phospate-buffered saline (PBS), centrifuged 6 min at 1,500 rpm and subsequently resuspended in PBS. The cell suspension was incubated with slow rotation for 24 hr at room temperature. After incubation the cells were centrifuged 10 min at 3,000 rpm and the supernatant collected. This supernatant contained the TRA-1-60 antigen exfoliated from the surface of the cells during incubation and was used as antigen preparation for coating of microtiter plates and for preparation of standard dilutions for calibration of the assay. The standard dilutions were calibrated in the laboratory of Jan Marrink (Groningen, The Netherlands) against the TRA-1-60 assay used there. TRA-1-60 concentrations were given as arbitrary U/ml. Both the NT2/D1 cell clone and the monoclonal TRA-1-60 antibody (ClAp11) were provided by professor Peter W. Andrews (Sheffield, UK). The value of 230 U/ml defined by Marrink et al.2 as the upper limit of the normal was used accordingly in our present study.
At the initiation of chemotherapy, levels of TRA-1-60 were elevated (>230 U/ml) in 79% (19/24) of NSGCT patients (range 369–2,649 U/ml) and in 80% (8/10) of SGCT patients (239–987 U/ml) (Fig. 1). In 96% (23/24) of patients with NSGCT, embryonal carcinoma was present in the primary tumor and 78% (18/23) had increased levels of TRA-1-60. After chemotherapy, levels in this group had dropped significantly (p < 0.01), but were still marginally increased in 17% (4/23) (269–352 U/ml) (Fig. 1). The persisting elevations of TRA-1-60 were not associated with remaining disease and HCG and AFP normalized in all patients. Increased levels of TRA-1-60 at the end of chemotherapy were also found in 30% (3/10) patients with SCGT (range 246–299 U/ml). The 1 patient with EC-negative NSCGT had increased levels of TRA-1-60 both before and after chemotherapy.
Immediately before chemotherapy elevations in HCG, AFP or TRA-1-60 tumor markers were found in 96% (22/23) EC-positive NSGCT patients (Table I). Twenty-two percent (5/23) patients had normal values of TRA-1-60. Four of these had increased levels of AFP or HCG. In 2 patients TRA-1-60 was found to be the only elevated tumor marker.
|AFP/HCG +/+||AFP/HCG +/−||AFP/HCG −/+||AFP/HCG −/−||n|
Stage I disease
Fifty-seven percent (16/28) of patients with Stage I NSGCT had elevated levels of TRA-1-60 at the first follow-up 2–4 weeks after orchidectomy and after 3 months of follow-up, TRA-1-60 was persistently increased in 10 patients (Fig. 2). Seventy-nine percent (22/28) of patients had EC-component in their primary tumor. Sixty-two percent (13/21) of EC-positive patients had elevated levels of TRA-1-60 at first follow-up as compared to 50% (3/6) of EC-negative patients. After 3 months follow-up 33% (7/21) of EC-positive patients had persisting elevations in TRA-1-60 as compared to 50% (3/6) of EC-negative patients (Fig. 2). Fourteen percent (4/28) of patients had recurrent disease during follow-up. Three of 4 patients had increasing levels of TRA-1-60 at time of recurrence but also increasing levels of AFP or HCG.
Thirteen percent (5/40) of patients with Stage I SGCT presented with elevated levels of TRA-1-60 at the first follow-up 2–4 weeks after orchidectomy. All of these patients had persistently high values despite being free of relapse (Fig. 2). Fifteen percent (6/40) of patients had recurrence during follow-up. In 1 of these patients recurrence was associated with an elevation in TRA-1-60.
The observations of our study were not in accordance with earlier reports by Marrink et al.2, 3 who have presented the only former clinical applications of serum TRA-1-60. They did find that TRA-1-60 was released into the serum of 76% of EC-positive disseminated NSGCT patients in concordance with the 78% found in our study. In contrast they found sera from all seminoma patients to contain normal levels of TRA-1-60, whereas our present study found elevations in 80% of these patients with disseminated disease. The reasons for this discrepancy are unclear and although earlier in vitro studies have not been consistent as to the expression of TRA-1-60 in seminomatous cell lines differences in antigen expression does not seem a reasonable explanation in this case.4
In our present study, 1 of the major problems identified was the lack of normalization of TRA-1-60 after chemotherapy. Again, this is in contrast to the earlier findings by Marrink et al.2 who found that all elevated TRA-1-60 values normalized after chemotherapy. Our main concern, however, relates to the rate of false positive values found in the Stage I NSGCT group. This problem was only addressed briefly in the study by Marrink et al.2 Approximately one-third of Stage I NSCGT patients in our study had elevated levels of TRA-1-60 at 2 months follow-up without developing recurrence and with normal AFP and HCG (Fig. 2). This figure was the same when only patients with EC-positive NSGCT were considered. At 4 months 8/19 NSGCT patients without subsequent recurrence had TRA-1-60 levels that would necessitate further examinations. Again, AFP and HCG were normal in all patients. The problem is illustrated in Figure 3.
In patients with disseminated testicular cancer, normalization in TRA-1-60 is not seen in all patients after treatment with chemotherapy despite no other signs of residual disease. Approximately one-third of patients with Stage I NSGCT have increased values of TRA-1-60 without having a relapse. TRA-1-60 is not at present useful as a tumor marker in patients with germ cell tumors. Although detecting a few early relapses the rate of false positive elevations in the tumor marker makes it unreliable in the clinical setting. Nevertheless, our study did confirm that elevated levels of TRA-1-60 were present in ∼80% of patients with disseminated EC-positive NSGCT at the start of chemotherapy and chemotherapy induced a significant decrease in levels of TRA-1-60. Thus, the TRA-1-60 antigen might still prove clinically useful provided the reliability of the assay can be improved.