Carcinoma of unknown primary with gastrointestinal profile: immunohistochemistry and survival data for this favorable subset.

BACKGROUND: Carcinoma of unknown primary with a "gastrointestinal profile" is an emerging, favorable entity. Distinguishing this entity is of increasing significance given the progress in the treatment of colorectal cancer. PATIENTS AND METHODS: 74 carcinoma of unknown primary (CUP) patients with CDX2+ tumors were chosen from the databases at M.D. Anderson and Sarah Cannon Cancer Centers between 2004 and 2010. Data on clinical and pathological characteristics including therapy and survival were recorded. RESULTS: 20 patients had ascites on presentation; the predominant sites of metastases included liver (30 %), carcinomatosis (50 %), and nodes (51 %). Based on immunohistochemistry, 2 cohorts were created: Cohort 1-"consistent with lower GI profile" included 34 patients [CDX-2+, CK20+, CK7-] and Cohort 2-"probable lower GI profile" included 40 patients [CDX2+, irrespective of CK7/CK20 status]. Excluding 6 outliers, Cohorts 1 and 2 had 32 and 36 patients, respectively; their median survivals were 37 and 21 months, respectively. On multivariate Cox regression analysis, only liver metastases were found to negatively influence survival. CONCLUSIONS: Our retrospective study provides encouraging indications that CUP patients with gastrointestinal profiles benefit from site-specific therapy. We recommend all CUP patients, especially those with abdominal nodes, isolated carcinomatosis or liver metastases, to undergo optimal immunohistochemistry (IHC) to check for a gastrointestinal profile of CUP.

Cancer of unknown primary: progress in the search for improved and rapid diagnosis leading toward superior patient outcomes.

This paper explores the enigma of cancer of unknown primary (CUP) in relation to rapidly improving molecular diagnostic approaches. It is based on the first global collaboration meeting on improving research and clinical outcomes in CUP organized by the CUP Foundation. We review the difficulties of classifying this widely heterogeneous disease and the available diagnostic and pathological evaluative techniques, focusing on molecular profiling. Retrospective studies in CUP patients are shown to provide indirect validation of the accuracy of several platforms of gene expression profiling assays that may identify CUP subsets that respond favorably to active chemotherapy regimens. This review concludes that the recent major improvements in pathologic and molecular diagnostics, coupled with new improved therapies for several specific advanced solid tumors, need to be harmonized with more evidence from clinical-translational trials. All patients with CUP could thus be appropriately managed without the constant uncertainty that has previously severely hampered patient care and optimal outcomes. The longer-term objective is to understand the biology of highly metastatic disease, leading to the development of future global therapeutic programs. Current clinical studies, such as CUP-ONE, will address some of these issues.

Phase I clinical study of N-[(4-chlorophenyl)amino]carbonyl-2,3-dihydro-1H-indene-5-sulfonamide (LY186641).

Between February 1987 and July 1988, 45 patients with advanced refractory cancer were treated with LY186641, a diarylsulfonylurea that has shown a broad spectrum of activity in preclinical testing. Patients received a weekly p.o. dose of LY186641 for 6 consecutive weeks; responding and stable patients continued weekly therapy until progression occurred. Using a standard phase I study design, the first three patients received LY186641 at 30 mg/m2 week; the dose was escalated in subsequent patients until dose-limiting toxicity occurred. Methemoglobinemia was the major toxicity observed and was dose related. Methemoglobin levels peaked approximately 24 h after LY186641 was administered and fell to low levels after 48 h. Six patients developed fatigue, cyanosis, and dyspnea associated with serum methemoglobinemia levels of greater than 20%; four of these patients were subsequently removed from the study. Hemolytic anemia was also observed but was clinically significant in only 10 patients. Other side effects were mild and infrequent. The maximum tolerated dose of LY186641, when given at this schedule, was 2550 mg/m2/week. No objective tumor responses were observed.

Potential utility of serum neuron-specific enolase levels in small cell carcinoma of the lung.

To assess the value of neuron-specific enolase (NSE) as a possible biomarker of small cell lung cancer, serum levels were determined by radioimmunoassay in 93 newly diagnosed untreated patients and were compared to the NSE levels of 20 healthy adult controls [9.6 +/- 0.7 (S.E.) ng/ml]. Serum NSE was elevated (greater than 20 ng/ml) in 73% of all patients including 23 of 39 (59%) with limited-stage disease and 45 of 54 (83%) with extensive-stage disease. The mean serum NSE was significantly higher in extensive-stage disease (94.5 +/- 13.8 ng/ml) compared to the mean value for limited-stage disease (33.7 +/- 4.7 ng/ml) (p less than 0.001). NSE was elevated in all patients with three or more sites of metastatic disease. Serial NSE determinations were obtained on 57 small cell lung cancer patients. NSE levels fell in 40 of 50 (80%) of patients responding to treatment, increased in 5 of 7 (71%) of patients with progressive disease, and increased in 30 of 35 (86%) of patients who relapsed. A persistent rise in serum NSE occurred as many as 12 weeks before the clinical recognition of relapse in 15 of 23 (65%) of patients for whom adequate serial NSE data were available. These findings indicate that serum NSE may be a useful marker for staging, monitoring treatment, and predicting relapse in patients with small cell lung cancer.

Pharmacokinetics of high-dose etoposide (VP-16-213) administered to cancer patients.

Plasma urine, and cerebrospinal fluid etoposide concentrations have been measured in 12 adult patients after administration of high-dose (400 to 800 mg/sq m) etoposide in order to determine the pharmacokinetics of this drug at these elevated dosages. Increasing the drug dosage produced proportionally higher peak plasma etoposide concentrations (27 to 114 micrograms/ml) and total areas under the concentration-time curve (9,200 to 48,000 micrograms/ml X min). The etoposide mean (+/- S.D.) terminal half-life of 8.05 +/- 4.3 hr and plasma clearance of 28.0 +/- 9.7 ml/min/sq m, however, were independent of the dosage given. The mean etoposide renal clearance in 5 patients was 10.0 +/- 4.3 ml/min/sq m, representing from 35 to 40% of the total clearance of this drug from plasma. Cerebrospinal fluid etoposide concentrations ranged from 0.1 to 1.4 micrograms/ml, as measured in 6 patients at 1 to 8 hr after high-dose etoposide therapy, and were 1.8 +/- 1.7% of the simultaneously measured plasma levels. Pleural fluid removed from one patient at 18 hr posttherapy contained etoposide at 1.8 micrograms/ml. Our data, combined with data published previously, indicate that the pharmacokinetics of high-dose etoposide is linear within the dosage range tested and similar to that seen with lower drug doses. They also suggest that etoposide penetrates poorly into the cerebrospinal fluid.

Combination intraventricular chemotherapy for meningeal neoplasia.

Twenty two patients with meningeal neoplasia were treated with biweekly combination intraventricular chemotherapy using methotrexate, cytosine arabinoside, and thiotepa. Patients with the following malignancies were included: breast cancer, ten patients; lung cancer, seven; non-Hodgkin's lymphoma, two; malignant melanoma, one; transitional cell carcinoma of the bladder, one; and malignant glioma, one. Eight of 22 patients (36%) had a Karnofsky performance status of less than 50%. Eleven of 22 patients received radiotherapy to symptomatic areas, and seven received systemic chemotherapy in addition to combination intraventricular therapy. Patients were evaluated for both toxicity and response to therapy. Myelosuppression was the major toxic condition and occurred in 17 of 22 patients (77%). Ten patients (45%) had a nadir WBC count of less than 1000/microL or a platelet count of less than 25,000/microL. No patient achieved a complete response (CR), although nine patients (41%) had partial responses (PRs) lasting 4 to 24 + weeks. Median survival for the entire group was 10 weeks (range, 6 to 24+ weeks). In this small group of patients, simultaneous triple-drug intraventricular chemotherapy caused unacceptable myelosuppression without increasing the response rate, response duration, or survival when compared with single-agent methotrexate and radiotherapy.

Mitoxantrone, fluorouracil, and high-dose leucovorin: an effective, well-tolerated regimen for metastatic breast cancer.

Between September 1988 and August 1990, we treated 35 women with metastatic breast cancer with a novel regimen containing mitoxantrone, fluorouracil (5-FU), and high-dose leucovorin. This regimen was designed to take full advantage of the favorable toxicity profiles of these agents while maintaining a high level of activity. All patients had received previous chemotherapy (adjuvant only, 15 patients; at least one metastatic regimen, 20 patients). Seven patients had received previous doxorubicin, but none within 6 months of study entry. Of 31 assessable patients, 20 (65%) had objective responses (two complete, 18 partial), with a median response duration of 6 months (range, 3 to 16+ months). Four patients with bone metastases (abnormal bone scan only) and pain were not considered assessable by strict response criteria; two of these patients had sustained symptomatic relief for 6 and 8 months, respectively. Myelosuppression was the most frequent toxicity but was mild in most patients; only four hospitalizations for fever and neutropenia were required (2% of courses). No severe thrombocytopenia occurred and no RBC transfusions were required. Alopecia, mucositis, and nausea/vomiting were uncommon and were not severe in any patient. The combination of mitoxantrone, 5-FU, and high-dose leucovorin is well tolerated and active as a first- or second-line treatment for metastatic breast cancer. Comparison with other standard regimens for breast cancer is indicated.

Lymphoproliferative diseases in immunocompromised hosts: the role of Epstein-Barr virus.

Epstein-Barr virus (EBV) is a ubiquitous transforming virus of the herpes group showing tropism for B lymphocytes. Primary infection in normal hosts results in a transient lymphoproliferative disorder, acute infectious mononucleosis (IM), that is restricted by cytotoxic and suppressive lymphocytes. However, in the immunodeficient host, EBV-induced lymphoproliferation may behave in a biologically malignant fashion. Patients with primary immunodeficiencies and those with immune incompetence resulting from suppressive therapy in allograft transplantation or infection with human immunodeficiency virus (HIV) have EBV-related illness ranging from fulminant mononucleosis and invasive polyclonal B cell hyperplasia to monoclonal B cell malignancies. While the direct link between EBV and malignant B cell proliferation in these patients has not been elucidated, the association has been increasingly recognized with improved techniques of viral detection. Clinical management can be guided by the location and extent of tumor, histologic features, and clonality. Regional and node-based polyclonal proliferations may respond to prompt reduction of immunosuppressive therapy and efforts to interrupt the replicative cycle with antiviral agents. Systemic cytotoxic therapy often leads to further immunosuppression and should be reserved for patients with progressive disease, advanced visceral involvement, and monoclonal lymphoid malignancies.

Poorly differentiated carcinoma of unknown primary site: correlation of light microscopic findings with response to cisplatin-based combination chemotherapy.

We have previously reported complete responses and long-term survival in patients with metastatic poorly differentiated carcinoma (PDC) of unknown primary site who received intensive cisplatin-containing chemotherapy regimens. We reviewed the light microscopic specimens from 113 patients with PDC in an attempt to identify common histopathologic features in the chemotherapy-responsive subgroup, and to rule out the presence of previously unrecognized germ cell tumors. Relatively few diagnoses more specific than PDC could be made. We could identify no histopathologic features by light microscopy that distinguished responsive from unresponsive neoplasms. Only one patient was found to have a previously unrecognized yolk sac carcinoma, and in five other patients the possibility of a germ cell neoplasm was considered in the differential diagnosis by at least one reviewer. The remaining tumors had no histologic features suggestive of germ cell neoplasms. Ninety-six patients had received combination chemotherapy (89 with cisplatin-containing regimens); 27 patients (28%) achieved complete remission, and 16 remain free of disease at a median of 65 months after completion of therapy. Patients with PDC of unknown primary site who are responsive to cisplatin-containing chemotherapy regimens cannot be reliably identified by light microscopy. At present, all such patients should be considered for an empiric trial of chemotherapy with cisplatin-based regimens, since cure is achievable in a minority.

Chronic daily administration of oral etoposide--a phase I trial.

In this phase I study, we administered etoposide (VP-16) orally for 21 consecutive days to patients with advanced refractory cancers. All patients had received previous chemotherapy, and 50% of patients had received more than one combination regimen. When given for 21 consecutive days, the maximum-tolerated dose of oral VP-16 was 50 mg/m2/d. Myelosuppression was the dose-limiting toxicity, and occurred between days 21 and 28. In most patients, blood counts had recovered sufficiently by day 35 to begin another 3-week course. WBC count nadirs of less than 1,000/microL occurred in four of 20 courses at this dose, and three patients required hospitalization for treatment of neutropenia and fever. Alopecia occurred in most patients; gastrointestinal (GI) and other toxicities were uncommon. Five of 16 patients with measurable tumor had partial responses of 3 to 4 months duration. Four of these five patients had malignancies that are usually unresponsive to VP-16 when administered by previously investigated schedules. This method of VP-16 administration is well tolerated, convenient, and may optimize antitumor efficacy by exploiting the schedule dependency of this drug. Phase II studies are necessary to define the level of activity of this schedule of VP-16.

Cisplatin-based combination chemotherapy in the treatment of poorly differentiated carcinoma and poorly differentiated adenocarcinoma of unknown primary site: results of a 12-year experience.

PURPOSE: We previously reported excellent responses to cisplatin-based chemotherapy in a minority of patients with poorly differentiated carcinoma (PDC) or poorly differentiated adenocarcinoma (PDA) of unknown primary site. We have continued to study and to treat these patients, and now report clinical characteristics, treatment results, and prognostic factors in a large group of patients identified prospectively. PATIENTS AND METHODS: Between February 1978 and December 1989, we treated 220 patients with PDC or PDA of unknown primary site. The median age was 39 years; 48% of patients had predominant tumor location in the mediastinum, retroperitoneum, or peripheral lymph nodes. Specialized pathologic studies resulted in the identification of specific tumor types in only a few cases. All patients received cisplatin-based chemotherapy; between 1978 and 1984, 116 patients received cisplatin, vinblastine, and bleomycin (PVeB) +/- doxorubicin, and 104 patients treated since January 1985 received cisplatin and etoposide +/- bleomycin. RESULTS: One hundred thirty-eight patients (63%) had objective responses to therapy, and 58 (26%) had complete response. Thirty-six patients (16%) are currently disease-free at a median of 61 months following therapy (range, 11 to 142 months). Actuarial 10-year survival is 16%. Favorable prognostic factors identified by Cox regression analysis include: (1) predominant tumor location in the retroperitoneum or peripheral lymph nodes, (2) tumor limited to one or two metastatic sites, (3) no history of cigarette use, and (4) younger age. CONCLUSION: Patients with PDC or PDA of unknown primary site represent another group of patients for whom potentially curative therapy is available. Patients with this syndrome should be distinguished from patients with well-differentiated adenocarcinoma of unknown primary site, and should receive a trial of cisplatin-based chemotherapy.

Overexpression of Her-2 in patients with poorly differentiated carcinoma or poorly differentiated adenocarcinoma of unknown primary site.

PURPOSE: To determine the frequency of Her-2 overexpression in patients with poorly differentiated carcinoma or poorly differentiated adenocarcinoma of unknown primary site. PATIENTS AND METHODS: Tumor specimens from 100 patients with poorly differentiated carcinoma or poorly differentiated adenocarcinoma were stained for the Her-2 protein using the Dako immunohistochemical method. Clinical and pathologic characteristics of patients with and without Her-2 overexpression were compared. RESULTS: Staining for Her-2 overexpression was successful in 94 of 100 patients. Ten (11%) of 94 tumor specimens overexpressed Her-2. Eight of 10 overexpressors had poorly differentiated adenocarcinoma, and all overexpressors had predominant tumor location above the diaphragm, usually in the mediastinum or lungs. CONCLUSION: Her-2 overexpression occurs in a minority of patients with poorly differentiated carcinoma/adenocarcinoma of unknown primary site. Because most overexpressors had poorly differentiated adenocarcinoma, further evaluation of patients with adenocarcinoma of unknown primary site is necessary to determine the frequency of Her-2 overexpression in this common subgroup. Evaluation of the efficacy of trastuzumab in Her-2 overexpressors with carcinoma of unknown primary site is indicated.

Paclitaxel by 1-hour infusion: an active drug in metastatic non-small-cell lung cancer.

PURPOSE: Paclitaxel is an active single agent when administered as a 24-hour continuous infusion in the treatment of stage IV non-small-cell lung cancer. We evaluated the efficacy and toxicity of paclitaxel administered by 1-hour infusion in the outpatient setting to patients with stage IV or relapsed non-small-cell lung cancer. PATIENTS AND METHODS: Fifty-nine patients with stage IV or relapsed non-small-cell lung cancer were treated with 1-hour infusions of paclitaxel. The first 17 patients received a dose of 135 mg/m2 and the remaining 42 patients received 200 mg/m2. By random assignment, 31 patients received a single-day, 1-hour infusion of paclitaxel, and 28 patients received a 3-day, divided-dose schedule, with each dose administered by 1-hour infusion. Both regimens were repeated every 21 days. All patients received premedication with dexamethasone, diphenhydramine, and cimetidine. Cytokines were not routinely used. RESULTS: Thirteen of 53 assessable patients (25%) had partial responses to treatment. An additional five patients had minor responses. The median survival duration of the entire group was 8 months and the actuarial 1-year survival rate was 33%. Patients who received 200 mg/m2 of paclitaxel had a higher response rate than those who received 135 mg/m2 (31% v 12%, respectively). Six of 16 patients (38%) previously treated with cisplatin-based regimens responded to 200 mg/m2 of paclitaxel. No significant differences in activity were seen when the 1-day and 3-day paclitaxel schedules were compared. Paclitaxel was well tolerated at both doses and schedules, and no severe hypersensitivity reactions occurred. Only 18 of 154 courses (12%) given at 200 mg/m2 resulted in grade 3 or 4 leukopenia. CONCLUSION: Paclitaxel administered by 1-hour infusion is an active and well-tolerated new agent in the treatment of metastatic non-small-cell lung cancer. These results suggest that a paclitaxel dose of 200 mg/m2 is more effective than 135 mg/m2 and can produce responses in patients previously treated with cisplatin-based regimens. Incorporation into combination regimens is indicated.

Paclitaxel in combination chemotherapy with radiotherapy in patients with unresectable stage III non-small-cell lung cancer.

PURPOSE: The addition of combination chemotherapy to standard radiation therapy has improved treatment for locally unresectable non-small-cell lung cancer. In this phase II study, we evaluated the toxicity and efficacy of a novel chemotherapy regimen that included paclitaxel, cisplatin, and etoposide plus concurrent radiation therapy in this group of patients. PATIENTS AND METHODS: Thirty-three patients with previously untreated, unresectable stage III non-small-cell lung cancer (stage IIIA, 11 patients; stage IIIB, 22 patients) initially received two courses of chemotherapy, which included paclitaxel 135 mg/m2 by 1-hour infusion on day 1, cisplatin 60 mg/m/ intravenously (i.v.) on day 2, and etoposide 100 mg/m2 i.v. on days 1, 2 and 3. On week 6, radiation therapy (60 Gy in 30 fractions) was initiated in conjunction with two additional courses of chemotherapy: paclitaxel 135 mg/m2 i.v. by 1-hour infusion on day 1, cisplatin 5 mg/m2 i.v. on days 2- to 10, and etoposide 25 mg/m2 on days 1 to 10. RESULTS: This combined modality program was feasible and well tolerated by most patients. During the two courses of induction chemotherapy, grade 3 or 4 myelosuppression occurred in only six patients (18%). Esophagitis was common during combined modality therapy (grade 3, 10 patients; grade 4 five patients). Forty-two percent of patients had partial response after two courses of induction therapy, and 82% of patients had an objective response at completion of therapy. Twelve patients (36%) had a complete response. Nineteen patients remain progression-free at a median of 8 months; the median survival time has not been reached. CONCLUSION: This paclitaxel-containing combined modality therapy is feasible and highly active in patients with inoperable stage III lung cancer. Esophagitis is the most common severe toxicity with this program. Further studies with paclitaxel-containing combination regimens in patients with stage III non-small-cell lung cancer are indicated.

Lung cancer in Hodgkin's disease: association with previous radiotherapy.

Seven cases of lung cancer were observed in patients with Hodgkin's disease (HD) since 1970. The risk ratio for the development of lung cancer among HD patients was 5.6 times that expected in the general population. The pertinent clinical data from these patients are described and compared to 28 additional patients reported from other institutions. Small-cell lung cancer represented the predominant histologic type of lung cancer encountered in both smoking and nonsmoking patients with HD, accounting for 42% of cases overall and greater than 55% of cases reported in reviews of second malignancies. Tobacco use was noted in only 53% of patients. Twenty-eight (94%) of 30 patients developing metachronous lung cancer received supradiaphragmatic irradiation as primary therapy for HD. Nineteen (68%) of these patients received subsequent chemotherapy salvage. The median age at diagnosis of HD and lung cancer was 39 and 45 years, respectively. The interval between diagnosis of HD and metachronous lung cancer averaged seven years but appeared to vary inversely with age. HD patients treated with supradiaphragmatic irradiation or combined modality therapy may be at increased risk for developing lung cancer. The high frequency of in-field malignancies that we observed and the prevalence of small-cell lung cancer in both smoking and nonsmoking patients suggests that chest irradiation may influence the development of metachronous lung cancer in these patients. The finding of a mean latent interval in excess of seven years emphasizes the need for close long-term observation.

High-dose VP-16-213 monotherapy for refractory germinal malignancies: a phase II study.

VP-16-213 at standard dose is one of the more active agents for the treatment of germ cell tumors. In previous phase I studies, VP-16-213 has been investigated in suprastandard dose when hematopoietic reconstitution was assured by autologous bone marrow transplantation (ABMT). This phase II study was performed to explore the possibility that an augmented dose of VP-16-213 may be more active than standard dose against germ cell tumors. Eleven patients with progressive refractory germ cell tumors were treated with high-dose VP-16-213: 2,400 mg/m2 with ABMT every three to four weeks followed by 1,200 mg/m2 without ABMT. Seven patients had received VP-16-213 at standard dose prior to high-dose VP-16-213. Toxicity to high-dose VP-16-213 included severe myelosuppression, nausea, vomiting, alopecia, mucositis, and hepatitis. Of 10 evaluable patients, two complete responses and four partial responses, all of short duration, were obtained. However, some patients unresponsive to standard-dose VP-16-213 exhibited responses to the augmented-dose VP-16-213. Therefore, although more myelosuppressive, VP-16-213 may have increased activity against germ cell tumors when administered at augmented dose. High-dose VP-16-213 may be considered in designing new approaches for initial management of patients with germ cell tumors not expected to be cured with standard chemotherapy.

Prolonged administration of oral etoposide in patients with relapsed or refractory small-cell lung cancer: a phase II trial.

Twenty-two patients with recurrent small-cell lung cancer (SCLC) were treated with single-agent etoposide 50 mg/m2/d by mouth for 21 consecutive days. Eleven patients had received previous chemotherapy with cyclophosphamide, doxorubicin, and vincristine (CAV) or etoposide (CAE) or both (CAVE). Four of the latter patients also received salvage treatment with cisplatin and etoposide (EP). Nine patients had been treated with EP as induction therapy, while two patients had received high-dose cyclophosphamide, etoposide and cisplatin (HDCEP). Altogether, 18 patients had received previous intravenous etoposide. The median time off chemotherapy was 4.5 months (range, 1 to 28.9 months). Ten patients (45.5%; 95% confidence interval [CI], 27% to 65%) achieved a complete or partial response. Responses were most common in patients who had responded to previous chemotherapy and who had not received any treatment in the 90 days before initiation of oral etoposide. Median response duration was 4 months (range, 1.5 to 9.5 months) and median survival was 3.5+ months (range, 1.0 to 15+ months). Leukocyte and platelet nadirs were 1,800/microL and 160,000/microL, respectively, during cycle 1 of treatment and occurred between days 21 and 28. Overall, total leukocyte count decreased to less than 1,000/microL during 10 of 56 cycles (18%). Five patients required six hospitalizations for neutropenia and fever. There were two toxic deaths due to sepsis. Platelet counts less than 50,000/microL occurred in 14 cycles (25%). Alopecia developed in all patients; gastrointestinal toxicity was uncommon. This schedule of etoposide administration warrants further study in combination with other active agents in previously untreated patients with SCLC.

Prolonged administration of oral etoposide in non-small-cell lung cancer: a phase II trial.

PURPOSE: The trial was undertaken to investigate the activity and toxicity of a prolonged schedule of oral etoposide in the treatment of advanced non-small-cell lung cancer (NSCLC). PATIENTS AND METHODS: Between March 1989 and August 1990, 25 patients with advanced NSCLC were treated with oral etoposide 50 mg/m2/d for 21 consecutive days, repeated every 28 to 35 days. The median patient age was 60 years (range, 38 to 84 years); male:female ratio was 12:13. Eight patients had stage IIIB disease; 17 had stage IV. Seventy-six percent of patients had Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. No patient had received previous chemotherapy with standard agents; nine patients had received previous or concurrent radiation therapy. Plasma etoposide concentrations were measured to estimate etoposide bioavailability and kinetics. RESULTS: Five of 22 patients (23%; 95% confidence interval [CI], 10% to 43%) had partial responses. Median response duration was 5 months (range, 2 to 6 months). Four of five responders were female. Besides alopecia, which occurred in all patients, myelosuppression was the most common toxicity, but was mild or moderate in most patients. Median leukocyte nadir during course 1 was 3,200/microL; only four of 69 courses produced a leukocyte nadir less than 1,000/microL. Severe thrombocythemia (less than 75,000/microL) did not occur. Gastrointestinal toxicity was uncommon. Median peak etoposide concentration was 3.4 micrograms/mL. A mean serum etoposide concentration greater than 1 microgram/L was maintained for more than 13 hours; the plasma concentration-time curve (AUC) was estimated to be 90% of that predicted after an identical dose of etoposide given intravenously. CONCLUSIONS: Etoposide given by this dose and schedule has moderate activity as first-line systemic therapy for advanced NSCLC. In previously untreated patients, chronic oral etoposide is well tolerated, and incorporation into combination regimens should be feasible. Etoposide bioavailability may be increased at lower oral doses.