Summary Forty-six patients with recurrent ovarian cancer were reoperated, and cancer samples for the subrenal capsule assay (SRCA) were collected from 23 of them, whereas this test was not done in the remaining 23 control patients. The SRCA was evaluable in 22 cases (96%). Taken together, no significant difference appeared in the 3 years' survival figures between the groups: seven of 22 patients (32%) with the evaluable SRCA and six of 23 control patients (26%) were alive. However, a further analysis of the data revealed that the SRCA guided the selection of chemotherapy only in 15 patients, whereas tumour samples were resistant to all cytostatics tested in six cases and toxic side-effects limited the clinical application of the test results in the remaining one case. Four of the 11 patients (36%) whose further chemotherapy was strictly chosen based on the SRCA and seven of the 24 patients (29%) whose treatment was based on physician's choice survived at least 3 years. Our conclusion is that the SRCA is of limited value in the selection of second-line chemotherapy in recurrent ovarian cancer.The subrenal capsule assay (SRCA) (Bogden et al., 1981) correctly predicts the response to cytostatic chemotherapy in 63-85% of patients with previously untreated ovarian cancer (Bogden, 1985; Griffin et al., 1983; Maenpaa et al., 1985a,b;Stratton et al., 1984Stratton et al., , 1986 Stratton et al., , 1988 (Maenpaa, 1985c). There are some data to suggest that the SRCA could be useful in the selection of the second-line chemotherapy (Griffin et al., 1983;Maenpaa, 1985c), but so far no data exist on improved survival rates in patients whose second-line chemotherapy has been guided by the use of the SRCA. We therefore assessed the value of the SRCA in patients with recurrent ovarian cancer that were followed up to death or for a minimum of 3 years.
Materials and methodsForty-six patients between 25 and 76 years of age with recurrent ovarian cancer were studied (Table I). The primary operation was radical in seven women, whereas tumours smaller than 2 cm of size remained in nine women and tumours larger than 2cm remained in 30 women. Serous cystadenocarcinoma was the most common type of cancer (22 women) followed by anaplastic cancer (11 women), miscellaneous cancer (nine women with mesonephric cancers and one with endometrioid cancer) and mucinous cystadenocarcinoma (three women) ( Table I). The cancer had spread to clinical stages III-IV in 37 women. All patients following surgery had been treated with 4-28 courses of various cytostatics, mostly with the combination of doxorubicin, cisplatin and cyclophosphamide (n = 34 patients). Thirty-five had responded favourably to the therapy, as assessed by the standard criteria (Miller et al., 1981), whereas in 11 women no response to therapy and/or progression of the disease was seen.These 46 patients entered our study with the approval of the local committee of ethics between 5-48 months after primary operation, when they came for a routine second-look (n = 32) and third-look (n ...