Andrzej P. Kudelka

Phase II study of intravenous topotecan as a 5-day infusion for refractory epithelial ovarian carcinoma.

PURPOSE To determine the efficacy and toxicity of topotecan administered as a 5-day intravenous infusion in patients with advanced ovarian cancer refractory to cisplatin-based chemotherapy. PATIENTS AND METHODS Thirty patients with advanced epithelial ovarian cancer refractory to cisplatin-based chemotherapy received intravenous infusions of topotecan 1.5 mg/m2 delivered over 30 minutes each day for 5 days. A course was repeated every 21 days. The patient eligibility requirements included age > or = 18 years, Zubrod score < or = 2, measurable disease, adequate hepatic and renal function, neutrophil count > or = 1,500/microL, platelet count > or = 100,000/microL, and anticipated survival > or = 3 months. RESULTS Twenty eight patients were assessable for response and toxicity. All patients were assessable for survival. The major toxicity from administration of topotecan at this dose schedule was myelosuppression; 21 patients required dose reductions. Four patients had neutropenic fever that required hospitalization, and seven patients required platelet transfusions. Maculopapular pruritic exanthema occurred in 20% of patients; gastrointestinal side effects were mild. No deaths were reported on the study. At dose levels of 1.5, 1.25, and 1.0 mg/m2, 61%, 31%, and 25% of patients, respectively, required dose reductions. Of 28 assessable patients, four (14%; 95% confidence interval [CI], 4% to 34%) achieved a partial response (PR) at a median of 1.4 months and lasting 8.9 months, and 17 had stable disease (SD). The overall median survival time was 10.0 months (95% CI, 8.1 to 13.5). CONCLUSION Topotecan shows modest clinical activity against cisplatin-refractory ovarian cancer, although the dose-intensity is compromised by the depth of the granulocyte nadir and the duration of granulocytopenia. Further studies of topotecan may necessitate a reevaluation of optimal dose schedule, with the possible incorporation of multilineage cytokines, and its activity in taxane-resistant tumors.

The taxoids: paclitaxel (Taxol®) and docetaxel (Taxotere®)☆

The taxoids, paclitaxel (Taxol) and docetaxel (Taxotere), represent a novel class of antineoplastic drugs. Paclitaxel and docetaxel share a similar mechanism of action: the promotion of microtubule assembly and inhibition of microtubule disassembly. The clinical development of paclitaxel was initially hampered by hypersensitivity reactions (HSRs). The use of premedications and prolongation of the infusion time to 24h has reduced these reactions and allowed this drugs clinical development. Although paclitaxels clinical activity has not been fully investigated, clinical trials have demonstrated its activity against ovarian, breast, and bronchial carcinomas. Because phase I studies of docetaxel noted occasional HSRs and these observations increased with further clinical experiences, those premedications employed with paclitaxel have now been instituted in many phase II studies of docetaxel. Docetaxel is currently being investigated in ovarian, breast, and bronchial carcinomas and has shown impressive clinical activity. The dose-limiting toxicity of both these agents is neutropenia; myalgias, mucositis, neuropathies, and alopecia have also been observed with both drugs. Additionally, a fluid retention syndrome and cutaneous toxicities have been noted in patients treated with docetaxel. Future studies of the taxoids will allow further comparisons of the toxicity and efficacy of these agents.

Recombinant human thrombopoietin attenuates carboplatin-induced severe Thrombocytopenia and the need for platelet transfusions in patients with gynecologic cancer

Myelosuppression is a serious complication in patients who are receiving chemotherapy for cancer. The use of myeloid growth factors has reduced the incidence of febrile neutropenia (1, 2). However, thrombocytopenia has been managed predominantly by platelet transfusion and modification of the chemotherapy dose. Although platelet transfusion can reduce hemorrhagic complications, repeated transfusions increase the risk for transfusion reactions, alloimmunization, and transmission of infectious agents and increase health care costs (3). Recently, thrombopoietin was identified and cloned and became available for clinical investigations (4-7). In early clinical trials, therapy with thrombopoietin increased platelet counts before chemotherapy and enhanced platelet recovery after moderately myelosuppressive regimens (8-11). However, the value of this agent in preventing severe thrombocytopenia and averting the need for platelet transfusions has not been established. Carboplatin has a broad spectrum of antitumor activity and is highly effective in the treatment of gynecologic cancer; however, cumulative thrombocytopenia is frequently dose-limiting (12, 13). We therefore initiated a phase I/II trial of recombinant human thrombopoietin (rhTPO, a full-length glycosylated molecule) to evaluate the clinical safety and activity of this agent in patients with gynecologic cancer who are at high risk for chemotherapy-induced severe thrombocytopenia. Methods Patients Patients with platinum-sensitive, recurrent, or advanced gynecologic cancer were referred from multiple sources to our comprehensive cancer center. Patients were eligible if they had adequate Karnofsky performance status ( 80%); adequate bone marrow, renal, cardiac, and hepatic function; and life expectancy of at least 3 months. Patients with a history of rapidly progressive disease (marked increase in tumor size [>50%], ascites, or symptoms related to underlying cancer in the preceding 4-week period), pelvic irradiation, surgery within the previous 2 weeks, chemotherapy or radiotherapy within the previous 4 weeks, or thromboembolic or bleeding disorders were excluded. Design The study had two phases: a cohort dose-escalation (safety) phase and a dose-expansion (activity) phase. During the dose-escalation phase, a single dose of rhTPO was administered to assess clinical tolerance and hematopoietic effects. Three weeks later, patients entered cycle 1, in which they received carboplatin at a dose calculated to provide an area under the curve (AUC) of 11 by using the Calvert formula (carboplatin dose=target AUC [glomerular filtration rate + 25]). Three weeks later, upon recovery (absolute neutrophil count 1.5 109 cells/L and platelet count 100 109 cells/L), patients entered cycle 2, in which they received carboplatin (AUC, 11) followed by rhTPO every other day for four doses (on days 2, 4, 6, and 8). Cycle 1 served as an internal control for cycle 2. At least three patients were enrolled for each dose level of rhTPO (0.6, 1.2, 2.4, and 3.6 g/kg of body weight per day). The optimal biological dose was defined as the lowest active dose at which platelet count response reached a plateau. Once the optimal biological dose was established, prechemotherapy treatment with rhTPO was eliminated and 12 patients received carboplatin alone until thrombocytopenia (platelet count nadir<30 109 cells/L) was observed. In subsequent cycles, the optimal biological dose of rhTPO was given as secondary prophylaxis on the same schedule (days 2, 4, 6, and 8 [n=6]) or a modified schedule (days 1, 1, 3, and 5 [n=6]). Patients with stable or responsive disease who had not experienced prolonged thrombocytopenia (platelet count<20 109 cells/L for>7 days) in cycle 2 were eligible to receive four additional cycles. Patients received platelet transfusion (single donors when available or 4 units from random donors) for severe thrombocytopenia (platelet count<20 109 cells/L). Clinical and Laboratory Monitoring Patients were monitored by history taking; physical examinations; and laboratory tests, including complete blood counts (three times weekly and daily during the expected platelet count nadir), serum chemistry, urinalysis, chest radiography, and electrocardiography. Serum samples were screened for antibodies three times before the study and weekly during the study by using enzyme-linked immunosorbent assays based on full-length TPO, truncated TPO, and c-mpl receptor (14). Reactive sera were tested by using a bioassay based on inhibition of the TPO-dependent cell line. Neutralizing antibodies were defined as those that were inhibitory on bioassay and associated with clinically significant thrombocytopenia. Statistical Analysis Hematologic toxicity in cycle 1 (no rhTPO) was compared with that in cycle 2 (during which rhTPO was given) as the degree and duration of thrombocytopenia and time to platelet recovery (Wilcoxon signed-rank test) and the proportion of patients requiring platelet transfusions (McNemar test). The difference in the platelet count nadirs between two cycles (cycle 2 cycle 1) was analyzed for dose response by using one-factor (dose) analysis of variance with a linearity test and multiple comparisons test (SPSS, Inc., Chicago, Illinois). Role of the Funding Source Recombinant human thrombopoietin and partial funding for the study were provided by Genentech, Inc. (South San Francisco, California). The collection, analysis, and interpretation of data and the decision to submit the manuscript for publication were under the control of the principal investigator. Results Patients Twenty-nine patients were enrolled in the trial. All were evaluable except for one who declined treatment after one chemotherapy cycle. Twenty-five patients had previously received chemotherapy. Recombinant Human Thrombopoietin Treatment before Chemotherapy Treatment with a single dose of rhTPO before chemotherapy (n=16) resulted in a dose-dependent increase in platelet counts (mean count at baseline, 277 109 cells/L; maximum mean count, 462 109 cells/L [P<0.001]). After a peak response around day 15, platelet counts gradually decreased. No major changes were seen in leukocyte counts (mean count at baseline, 6.74 109 cells/L; mean count after treatment, 7.26 109 cells/L) or hemoglobin values (baseline value, 119.3 2.7 g/L; post-treatment value, 121.4 2.7 g/L). Recombinant Human Thrombopoietin Treatment after Chemotherapy Twenty-eight patients received rhTPO after carboplatin therapy (16 in the dose-escalation phase and 12 in the dose-expansion phase). Dose-Escalation Phase Therapy with rhTPO significantly reduced both the degree of thrombocytopenia (platelet count nadir, 53 109 cells/L and 35 109 cells/L [P=0.005]) and its duration (days on which platelet count was<50 109 cells/L, 3 and 6 [P=0.002]) in cycle 2 compared with cycle 1 (Table). At an rhTPO dose of 0.6 g/kg, the mean platelet count nadir did not differ between cycle 1 and cycle 2, but in cycle 2 it was twofold higher at 1.2 g/kg, 1.7-fold higher at 2.4 g/kg, and 1.2-fold higher at 3.6 g/kg. No linear dose response was seen (P=0.181). However, the difference in platelet count nadir was greater (P=0.027) for patients who received rhTPO at 1.2 and 2.4 g/kg than in those that received it at 0.6 and 3.6 g/kg. Because no greater benefit was seen at 2.4 g/kg, 1.2 g/kg was the lowest active dose and was considered the optimal biological dose for this regimen. Table. Effect of Recombinant Human Thrombopoietin Treatment on Carboplatin-Induced Thrombocytopenia and Platelet Recovery Dose-Expansion Phase To better assess the safety and activity of rhTPO at the optimal biological dose, six patients received cycles of carboplatin alone until they experienced thrombocytopenia (platelet count<30 109 cells/L). Recombinant human thrombopoietin was used in subsequent cycles as a secondary prophylaxis (1.2 g/kg), with the same schedule (days 2, 4, 6, and 8). As shown in the Figure, five of the six patients required platelet transfusion in cycle 1. Therapy with rhTPO decreased the severity of thrombocytopenia and eliminated the need for platelet transfusion in three patients and reduced the need for transfusion in one patient. Figure. Platelet counts after therapy with recombinant human thrombopoietin (1.2 g/kg administered on days 2, 4, 6, and 8) used as secondary prophylaxis after carboplatin treatment (cycle 2) ( solid line ) compared with those obtained by using carboplatin treatment alone (cycle 1) ( dotted line ). arrowheads stars Six additional patients received rhTPO (1.2 g/kg) 1 day before chemotherapy (day 1) and on days 1, 3, and 5 to determine whether the degree of protection could be further augmented. Five of the 6 patients experienced thrombocytopenia (platelet count<30 109 cells/L) in cycle 1 and received rhTPO in cycle 2. As with the original schedule, the need for platelet transfusion was eliminated in 3 of these 5 patients. The sixth patient experienced severe thrombocytopenia in cycle 2 and received rhTPO in cycle 3. After rhTPO therapy, this patients platelet count nadir increased and the need for platelet transfusion was eliminated. Thus, of 12 patients who received rhTPO as secondary prophylaxis, 11 initiated rhTPO therapy in cycle 2 and 1 initiated it in cycle 3. In this group, rhTPO increased the platelet count nadir by twofold and reduced the duration of severe thrombocytopenia by 4 days (Table). Platelet Transfusion and Recovery The need for platelet transfusion was markedly reduced with rhTPO (Table). Specifically, in the group that received rhTPO at the optimal biological dose (1.2 g/kg), 75% of patients required platelet transfusion in cycle 1 compared with 25% in cycle 2 (P=0.013). The number of transfusions required was reduced by 69% (16 compared with 5 transfusions). Therapy with rhTPO significantly enhanced platelet recovery (P<0.001) (Table). In cycle 2, 67% of patients recovered platelet counts of at least 100 109 cells/L by day 21,

Docetaxel for Patients With Paclitaxel-Resistant Müllerian Carcinoma

PURPOSE To determine the efficacy and toxicity of docetaxel in patients with müllerian carcinoma resistant to paclitaxel. PATIENTS AND METHODS Thirty-two patients with epithelial ovarian cancer, fallopian tube cancer, or primary peritoneal cancer who failed paclitaxel-based chemotherapy received either 100 or 75 mg/m(2) of docetaxel every 3 weeks. Resistance to paclitaxel was defined as either progression of disease during treatment, failure to achieve regression of disease after at least four courses, or rapid recurrence (within 6 months) after completion of therapy. RESULTS Eighteen patients were treated on a formal protocol and fourteen with the commercially available docetaxel. Thirty were assessable for response. Toxicities were thoroughly evaluated in the 18 patients on protocol. Twenty-seven patients (85%) had epithelial ovarian cancer. The overall response rate was 23% (one complete and six partial responses), with a median survival time of 44 weeks (9.5 months). Nine patients had stable disease and 14 progressive disease. Among 19 patients who progressed during prior paclitaxel treatment, two (11%) responded to docetaxel, compared with five (45%) of 11 patients in other paclitaxel-resistance categories. The responders had a median taxane-free interval (ie, the time between the last paclitaxel and first docetaxel treatment) of 73 weeks, compared with 19 weeks for the nonresponder group. Toxic effects were as expected. CONCLUSION Docetaxel is an active chemotherapeutic agent in patients with müllerian carcinoma previously treated with paclitaxel-based chemotherapy, especially in the patients who had a long taxane-free interval after a previous short response to paclitaxel.

Phase II study of irinotecan in prior chemotherapy-treated squamous cell carcinoma of the cervix.

PURPOSE A phase II study was performed to evaluate the antitumor activity and toxicity of irinotecan (CPT-11), a water-soluble derivative of camptothecin, in patients with prior chemotherapy-treated squamous cell cancer of the cervix. PATIENTS AND METHODS Forty-two patients were included in the study. The median age was 44 years (range, 24 to 59 years). The median Zubrod performance status was 1. All patients were refractory to first-line chemotherapy and 88% had received prior radiotherapy. The initial dose of CPT-11 was 125 mg/m2 given as a weekly 90-minute intravenous infusion for 4 weeks, every 6 weeks. Subsequent doses were unchanged, reduced, or omitted according to toxicity grade. RESULTS Forty-two patients were assessable for response. The overall response rate was 21%. The median time to response was 6 weeks and the median duration of response was 12 weeks. The overall median duration of survival was 6.4 months. A statistically significant survival advantage (median of 12.6 v 5.1 months) was found in patients whose disease responded to the treatment (P < .015). The major dose-limiting toxic effects (grade > or = 3) were nausea and vomiting (45%), diarrhea (24%), and granulocytopenia (36%). Grade > or = 3 anemia was encountered in 62% of patients and the incidence of thrombocytopenia was negligible. Less severe side effects were alopecia (48%), drug fever (43%), anorexia (33%), fatigue (33%), skin rash (21%), stomatitis (14%), and allergic reaction (9%). The gastrointestinal intolerance was dose-related. The incidence of bone marrow depression did not decrease with dose reduction, possibly because of a cumulative effect or hematologic intolerance by a subset of patients. CONCLUSIONS CPT-11 has significant activity in refractory cervical carcinoma. Gastrointestinal intolerance and hematologic toxicity must be monitored carefully. Further studies of alternative schedules may improve the tolerance and response rate.

Carboplatin reinduction after taxane in patients with platinum-refractory epithelial ovarian cancer.

PURPOSE To determine the activity of carboplatin in patients with ovarian cancer who progressed on taxane (paclitaxel or docetaxel) therapy. PATIENTS AND METHODS Thirty-three patients with ovarian cancers refractory to platinum and taxane therapy were treated with single-agent carboplatin reinduction once the disease progressed on a taxane. The starting dose of carboplatin was 300 mg/m2 at 28-day intervals. RESULTS Patients were a median age of 56 years (range, 31 to 80), had a median Zubrod performance status of 1 (range, 0 to 2) and had received a median of three prior chemotherapy regimens (range, two to eight) and one pretaxane platinum regimen (range, one to three). Twenty-six patients had a platinum-free interval of at least 12 months at the time of posttaxane re-treatment with carboplatin. There were seven of 33 (21%) partial responses, with a median duration of 7+ months (range, 2+ to 12+). Responses were noted only in patients with at least a 12-month platinum-free interval and an initial sensitivity to a taxane. The therapy was well tolerated and neurotoxicity was absent. CONCLUSION A subset of patients with platinum-refractory disease that initially responded to a taxane and who eventually have a platinum-free interval of at least 1 year may respond to carboplatin reinduction. This finding may be secondary to paclitaxel or docetaxel therapy that leads to the reversal of platinum resistance, or the prolonged platinum-free interval permits the loss of resistance to platinum by the tumor. Carboplatin reinduction should be considered in the treatment of patients whose ovarian cancer progresses after an initial sensitivity to a taxane and who had a prolonged platinum-free interval.

Increased expression of fascin, motility associated protein, in cell cultures derived from ovarian cancer and in borderline and carcinomatous ovarian tumors

Fascin bundles actin microfilaments within dynamic cellular structures such as microspikes, stress fibers and membrane ruffles. Fascin overexpression induces membrane protrusions and increased cell motility, and is highly expressed in various transformed cells, and in specialized normal cells including neuronal, endothelial and dendritic cells. In breast cancer, fascin expression correlates with high-grade tumors. To investigate whether fascin might be a predictor factor for ovarian cancer progression, eighteen cell cultures derived from ovarian cancer, and thirty four archival paraffin-embedded material of normal versus borderline and carcinomatous ovaries were stained by immunocytochemistry and immunohistochemistry with fascin Mab 55K-2. Overall expression of the fascin protein was found in 50% (9/18) of cell cultures derived from original samples of ovarian tumors. Expression of fascin protein was found in 67% (6/9) of cell cultures derived from patients diagnosed with stage IV disease, and 33% (3/9) of cell cultures from patients diagnosed with stage II/III. There was no clear relationship between fascin expression and histologic types, tumor grade, or DNA ploidy. However, 75% of cell cultures, which developed into a xenograft after intraperitoneal inoculation, showed fascin expression, while 86% of non-tumorigenic cell cultures did not show fascin expression. Expression of fascin in these established ovarian tumor cell cultures was significantly associated with the ability for these cells to grow intraperitoneally (P<0.05). Furthermore, fascin was never expressed in normal epithelial ovarian tissues, but was present in all pathologic ovaries. Both diffuse and focal patterns were observed in borderline ovarian tumors (67% and 33%), advanced primary ovarian cancer (67% and 33%) and metastatic ovarian cancer (89% and 11%). Therefore, our data suggest that fascin could serve as a prognostic factor for abnormal ovarian epithelial pathology and could be a novel target for the treatment of ovarian cancer.

Intraperitoneal adoptive immunotherapy of ovarian carcinoma with tumor-infiltrating lymphocytes and low-dose recombinant interleukin-2: A pilot trial

A pilot study was conducted in patients who had advanced epithelial ovarian carcinoma, and who were refractory to platinum-based chemotherapy, to determine the feasibility and clinical effects of a schedule of intraperitoneal (IP) tumor-infiltrating lymphocytes (TIL) expanded in recombinant interleukin-2 (rIL-2), and low-dose rIL-2 IP. TIL were expanded from solid metastases or malignant effusions in serum-free AIM V medium supplemented with low concentrations (600 IU/ml) or rIL-2 using a four-step method of expansion that included a hollow fiber bioreactor (artificial capillary culture system). Patients received IP TIL suspended in dextrose 5% in sodium chloride 0.2% containing 0.1% human albumin and 6 x 10(5) IU rIL-2 on day 1, followed by 6 x 10(5) IU rIL-2/m2 body surface area, administered daily by bolus IP injection, on days 2-4, 8-11, and 15-18. In the absence of disease progression, two additional 4-day cycles of IP rIL-2 were administered. Patients (n = 3) whose TIL failed to grow in vitro received IP IL-2 alone. Eight patients received rIL-2 expanded TIL (10(10)-10(11) range) plus rIL-2 followed by several cycles of rIL-2 alone. One of these patients was treated twice with TIL plus rIL-2. Expanded TIL were primarily CD3+CD4+TCR alpha beta+ (eight TIL-derived T-cell lines). One TIL-derived T-cell line was comprised mostly of CD3+CD8+TCR alpha beta+ cells. Eleven patients (eight treated with TIL plus rIL-2 and three patients treated with rIL-2 alone) received a total of 38 cycles of rIL-2 without TIL. Grade 3 clinical toxicity (peritonitis) occurred in 1 of 9 cycles of TIL plus rIL-2 and 1 of 38 cycles of rIL-2 alone. Each cycle was 4 days long. Grade 3 anemia occurred in 1 of 9 TIL plus rIL-2 cycles and 3 of 38 cycles of rIL-2 alone. There were no measurable responses; however, four of eight patients treated with IP TIL plus rIL-2 had some indication of clinical activity: ascites regression (two patients), tumor and CA-125 reduction (one patient), and surgically confirmed stable tumor and CA-125 values (one patient). The schedule of IP TIL plus low-dose rIL-2 shows manageable toxicity and is worthy of further evaluation in patients with epithelial ovarian cancer who have less tumor burden.

A phase I clinical and pharmacological study of oral 9-nitrocamptothecin, a novel water-insoluble topoisomerase I inhibitor.

9-nitrocamptothecin (9NC) is a water-insoluble topoisomerase I inhibitor with a broad antitumor activity in animal models. To determine the maximum tolerated oral dose (MTD), a phase I study was performed in patients with advanced cancer refractory to conventional chemotherapy. 9NC was administered orally with escalating doses to cohorts of five patients beginning at 1 mg/m2/day for five consecutive days every week for 4 weeks. Increments were 0.5 mg/m2/day for each cohort. Toxicity was evaluated in 28 patients diagnosed with various malignancies. Seven patients received 1 mg/m2/day for 28 weeks; 10 patients, 1.5 mg/m2/day for 68 weeks; and 26 patients, 2 mg/m2/day for 159 weeks. At 1.5 mg/m2/day or higher, the dose-limiting toxicity was hematologic, with grade 4 anemia in eight (29%); neutropenia in seven (25%) and thrombocytopenia in five (18%). Grade 2 or higher toxic effects occurred at each dose level: nausea and vomiting in 15 (54%), diarrhea in nine (32%), chemical cystitis in seven (25%), neutropenic sepsis in six (21%) and weight loss in five (18%) (N=28). Responses were observed after 2-8 weeks of therapy in five patients with pancreatic, breast, ovarian and hematologic tumors. Fourteen patients had a disease stabilization and one patient received treatment up to 18 months. The MTD of 9NC given orally has been estimated at 1.5 mg/m2/day for five consecutive days weekly. 9NC may be tolerated for sustained periods of time, but has the potential for significant hematologic, gastrointestinal and urinary bladder toxicity. Significant antitumor activity was observed, warranting further clinical investigations.

Large-scale expansion in interleukin-2 of tumor-infiltrating lymphocytes from patients with ovarian carcinoma for adoptive immunotherapy

Tumor infiltrating lymphocytes (TIL) from malignant ascites or solid tumor specimens obtained from patients with ovarian carcinoma were expanded to large numbers in vitro (10(10)-10(11)) by a four-step method using AIM V medium and low concentrations of recombinant interleukin-2 (rIL-2). The expansion procedure employed 24-well culture plates, T-flasks, polyolefin gas-permeable bags (PGPB), and an artificial capillary culture system (ACCS). The mean number of mononuclear leukocytes introduced into the 24-well plates was 16.5 +/- 4.2 x 10(6) cells. TIL from a total of 16 patients were expanded only through the first three steps of the process (24-well-plates, T-flasks, and PGPB) with an overall expansion of 255 +/- 99 fold and mean duration of 27.4 +/- 2.2 days. TIL from 9 of 16 patients were expanded further through the fourth step (ACCS) of the expansion method. The cumulative fold-expansion in nine patients was 8044 +/- 4807 (mean +/- SEM), the median was 2876 and the mean expansion time was 47.1 +/- 4.7 days. TIL from seven additional patients did not grow in rIL-2. Six of these 7 patients received chemotherapy at least four weeks before the specimens were collected. Two ACCS were used in parallel to facilitate expansion of TIL. Viable rIL-2-expanded TIL in the range of 1 x 10(10)-1 x 10(11) were recovered from the two ACCS, a number sufficient for adoptive immunotherapy of patients with ovarian carcinoma. The rIL-2-expanded TIL were predominantly CD3+ CD4+ CD8- alpha beta TCR+, although CD3+ CD4- CD8+ alpha beta TCR+ T cell lines were obtained from certain patients. An increase (43 +/- 8 vs 75 +/- 13; P = 0.05) in the proportion of CD4+ cells was observed over the duration of the four expansion steps. However, CD8+ TIL-derived T cells lines were also expanded in the ACCS. The four-step expansion method described here has several significant advantages over existing techniques. It requires substantially less personnel, equipment and space and the risk of contamination during expansion of the cultures is decreased. These results demonstrate that the four-step method described here can be effectively used for the large-scale expansion of ovarian TIL for the treatment of patients with ovarian carcinoma by adoptive immunotherapy.