Randomized phase III intergroup trial of isotretinoin to prevent second primary tumors in stage I non-small-cell lung cancer.

BACKGROUND: Promising data have suggested that retinoid chemoprevention may help to control second primary tumors (SPTs), recurrence, and mortality of stage I non-small-cell lung cancer (NSCLC) patients. METHODS: We carried out a National Cancer Institute (NCI) Intergroup phase III trial (NCI #I91-0001) with 1166 patients with pathologic stage I NSCLC (6 weeks to 3 years from definitive resection and no prior radiotherapy or chemotherapy). Patients were randomly assigned to receive a placebo or the retinoid isotretinoin (30 mg/day) for 3 years in a double-blind fashion. Patients were stratified at randomization by tumor stage, histology, and smoking status. The primary endpoint (time to SPT) and the secondary endpoints (times to recurrence and death) were analyzed by log-rank test and the Cox proportional hazards model. All statistical tests were two-sided. RESULTS: After a median follow-up of 3.5 years, there were no statistically significant differences between the placebo and isotretinoin arms with respect to the time to SPTs, recurrences, or mortality. The unadjusted hazard ratio (HR) of isotretinoin versus placebo was 1.08 (95% confidence interval [CI] = 0.78 to 1.49) for SPTs, 0.99 (95% CI = 0.76 to 1.29) for recurrence, and 1.07 (95% CI = 0.84 to 1.35) for mortality. Multivariate analyses showed that the rate of SPTs was not affected by any stratification factor. Rate of recurrence was affected by tumor stage (HR for T(2) versus T(1) = 1.77 [95% CI = 1.35 to 2.31]) and a treatment-by-smoking interaction (HR for treatment-by-current-versus-never-smoking status = 3.11 [95% CI = 1.00 to 9.71]). Mortality was affected by tumor stage (HR for T(2) versus T(1) = 1.39 [95% CI = 1.10 to 1.77]), histology (HR for squamous versus nonsquamous = 1.31 [95% CI = 1.03 to 1.68]), and a treatment-by-smoking interaction (HR for treatment-by-current-versus-never-smoking = 4.39 [95% CI = 1.11 to 17.29]). Mucocutaneous toxicity (P<.001) and noncompliance (40% versus 25% at 3 years) were higher in the isotretinoin arm than in the placebo arm. CONCLUSIONS: Isotretinoin treatment did not improve the overall rates of SPTs, recurrences, or mortality in stage I NSCLC. Secondary multivariate and subset analyses suggested that isotretinoin was harmful in current smokers and beneficial in never smokers.

Chromosome alterations in human small cell lung cancer: frequent involvement of 5q.

Deletions of the 3p chromosome region and molecular alterations of the tumor suppressor genes RB1 and TP53, located, respectively, at 13q14 and 17p13, are well-documented in small cell lung cancer (SCLC). Because of technical difficulties, karyotypes of primary SCLC specimens are rarely reported. In this study, detailed cytogenetic analysis was performed on 13 early passage SCLC cell lines and fresh specimens, including 4 lung primaries. Numerous chromosome alterations were found, even in newly diagnosed primary tumors. Consistent with previous molecular studies, chromosomal losses of 3p (13 cases) and 17p13 (12 cases) were frequently observed. Numerical losses of chromosome 13 and structural rearrangements affecting 13q14 were identified in 10 specimens. In addition, losses of chromosome 5 and structural alterations of 5q occurred in 12 tumors; among these, 9 displayed losses of region 5q13-q21. Double minutes were found in 4 cases (3 of 5 specimens from patients who received prior cytotoxic therapy but only 1 of 8 from untreated patients). DNA analysis revealed amplification of either MYC1 or MYCN in cells from each of these 4 tumors. Overall, the cytogenetic findings underscore that progression of SCLC involves multiple genetic changes and suggest further that a tumor suppressor gene(s) on 5q may contribute to SCLC tumorigenesis.

Small cell lung cancer cell line derived from a primary tumor with a characteristic deletion of 3p.

Chemotherapy plus surgery is feasible and potentially effective in selected patients with small cell lung cancer (SCLC) and provides a unique opportunity to study SCLC early in its biological history. The in vitro characteristics of a SCLC cell line derived from a resected lung primary tumor after treatment with 3 courses of chemotherapy is described. The original SCLC cell line UMC-SCLC-1 exhibited features of classic SCLC with typical morphology and growth characteristics, high levels of dopa decarboxylase, bombesin-like peptides, neuron-specific enolase and calcitonin, and the presence of neurosecretory granules and demonstrated the deletion of the short arm of chromosome 3. After multiple passages, UMC-SCLC-1 gradually changed its culture characteristics to a cell line, UMC-SCLC-1A, with morphological features of large cell anaplastic carcinoma, an altered growth pattern, decrease in calcitonin, and increase in radioresistance but retained the other biochemical markers of classic SCLC (bombesin and dopa decarboxylase production). Serial DNA content analyses showed that increased aneuploidy during continuous culture in vitro was associated with the morphological changes. Both UMC-SCLC-1 and UMC-SCLC-1A demonstrated the deletion of chromosome 3p, amplification and abundant expression of N-myc, and increased expression of c-raf. Chemotherapy sensitivities were stable throughout multiple passages and correlated with in vivo response. UMC-SCLC-1A represents a unique SCLC cell line with heterogeneous properties of both classic and morphological variant SCLC cell lines. In addition, the characteristic deletion of 3p, previously described in cultures derived from metastatic lesions and heavily pretreated patients, is seen in a primary lesion early in the natural history of SCLC.

Quantitative assay of human T-cell leukemia/lymphoma virus transformation.

The in vitro transformation of normal T-lymphocytes by human T-cell leukemia/lymphoma virus (HTLV-I) is possible utilizing cocultivation techniques. We now report on a quantitative assay for HTLV-I transformation. Transformed cell lines were produced by cocultivation of either preactivated (phytohemagglutinin and T-cell growth factor) or nonactivated peripheral blood mononuclear cells with an equal number of lethally irradiated HTLV-I-positive donor cells (MT-2). After 14 days in liquid culture, transformed cells were plated in a 2-layer soft agarose system with or without T-cell growth factor (TCGF). Colony formation among 50 normal controls was observed at varying efficiencies with a mean number of 179 colonies (range, 6-599) in the presence of TCGF (up to a 2-log difference). The day 14 T-cell cultures demonstrated relatively low colony-forming efficiencies (less than or equal to 0.1%) and enhanced colony formation in the presence of TCGF. Day 14 after cocultivation was chosen for this assay based on a dose-response relationship between colony formation and the virus-positive donor cell inoculum and the known kinetics of colony growth of normal activated T-cells. An analysis of individual colonies indicated that they were of target cell origin and HTLV-I positive. Recombinant beta-interferon in increasing concentrations caused a decrease in colony formation as measured in this assay. Long-term cell cultures (2-18 months) showed higher colony-forming efficiencies (up to 1.0%) which were not enhanced by TCGF. The ability to quantitatively evaluate transformation via colony counts will provide an opportunity to study differences in transforming efficiencies attributable to varying target cells, donor cells, or blocking factors such as interferons, drugs, or anti-HTLV-I antibodies.

The prognostic significance of neuroendocrine markers and carcinoembryonic antigen in patients with resected stage I and II non-small cell lung cancer.

Non-small cell lung cancer with neuroendocrine differentiation may represent a subset of patients with a more aggressive (like small cell lung cancer) or less aggressive (like carcinoid) biological behavior. To investigate their prognostic significance, immunohistochemical stains for 4 neuroendocrine markers (neuron-specific enolase, chromogranin A, Leu-7, and synaptophysin) and carcinoembryonic antigen (CEA) were studied in 260 patients with surgically resected stage I and II non-small cell lung cancer. The following percentages of cases were positive for each marker: neuron-specific enolase, 70.0%; chromogranin A, 14.2%; Leu-7, 7.7%; synaptophysin, 11.2%; and CEA, 68.5%. Sixty-one (23.5%) were positive for > or = 2 neuroendocrine markers. When compared to adenocarcinoma, squamous cell carcinoma displayed lower positivity for CEA and > or = 2 neuroendocrine markers. There was no significant difference in stage, site of relapse (distant versus local), disease-free, or overall survival for each marker individually or for those with > or = 2 neuroendocrine markers. Multivariate analysis showed that higher nodal stage (N1 versus N0), tumor stage (T2 versus T1), older age, and the presence of mucin predicted for poorer overall survival. Neuroendocrine markers and CEA were not of prognostic significance in this group of patients with resected stage I and II non-small cell lung cancer.

The use of neuroendocrine immunoperoxidase markers to predict chemotherapy response in patients with non-small-cell lung cancer.

Small-cell lung cancer (SCLC), a chemotherapy-responsive disease, is characterized by neuroendocrine properties. In contrast, non-small-cell lung cancer (NSCLC) is at best moderately responsive to chemotherapy, and only 10% to 20% of cases demonstrate neuroendocrine properties. The present study is a retrospective analysis of the use of immunoperoxidase markers for neuron-specific enolase (NSE), Leu-7, and chromogranin A in NSCLC patients treated with chemotherapy. It was designed to determine if the presence of neuroendocrine markers predict for response to chemotherapy. The diagnostic slides and blocks were obtained on 52 NSCLC patients who were treated with chemotherapy (26 responders and 26 nonresponders). Immunoperoxidase studies were performed, and slides were scored without knowledge of the patient's response. Markers were positive in responders and nonresponders, respectively, as follows: NSE, 14 of 26 (54%) versus seven of 26 (27%), P = .04; Leu-7, 11 of 25 (44%) versus five of 26 (19%), P = .08; and chromogranin A, three of 26 (12%) versus 0 of 26 (0%), P = .71. Two markers were positive in 10 of 26 responders (38%) and 0 of 26 nonresponders (0%), P less than .01. Responders with two or more positive markers showed superior survival (median, 79 weeks) compared with responders with fewer than two positive markers (median, 51 weeks) and nonresponders (median, 27 weeks). These data suggest that the presence of neuroendocrine markers in NSCLC is associated with an increased likelihood of response to chemotherapy and may add to the standard parameters (performance status, weight loss) used to select patients for chemotherapy.

Prognostic significance of K-ras codon 12 mutations in patients with resected stage I and II non-small-cell lung cancer.

PURPOSE: The aim of this study was to investigate the prognostic importance of codon 12 K-ras mutations in patients with early-stage non-small-cell lung cancer (NSCLC). PATIENTS AND METHODS: We identified 260 patients with surgically resected stage I (n = 193) and stage II (n = 67) NSCLC with at least a 5-year follow-up. We performed polymerase chain reaction analysis of DNA obtained from paraffin-embedded NSCLC tissue, using mutation-specific probes for codon 12 K-ras. RESULTS: K-ras mutations were detected in 35 of 213 assessable specimens (16.4%). K-ras mutations were detected in 27 of 93 adenocarcinomas (29.0%), one of 61 squamous cell carcinomas (1.6%), five of 39 large-cell carcinomas (12.8%), and two of 20 adenosquamous carcinomas (10%) (P = .001). G to T transversions accounted for 71% of the mutations. There was no statistically significant difference in overall survival for all patients with K-ras mutations (median survival, 39 months) compared with patients without K-ras mutations (median survival, 53 months; P = .33). There was no statistically significant difference in overall or disease-free survival for subgroups with stage I disease, adenocarcinoma, or non-squamous cell carcinoma or for specific amino acid substitutions. The median survival time for stage II patients with K-ras mutations was 13 months, compared with 38 months for patients without K-ras mutations (P = .03). CONCLUSION: Codon 12 K-ras mutations were more common in adenocarcinomas than in squamous cell carcinomas. For the subgroup with stage II NSCLC, there was a statistically significant adverse effect on survival for the presence of K-ras mutations. However, when the entire group was considered, the presence of K-ras mutations was not of prognostic significance in this cohort of patients with resected early-stage NSCLC.

Net costs from three perspectives of using low versus high osmolality contrast medium in diagnostic angiocardiography.

OBJECTIVES: We conducted an economic analysis to assess the extent to which a reduction in adverse drug reactions induced by low osmolality compared with high osmolality contrast media during diagnostic angiocardiography would result in savings to hospitals, society and third-party payers that would offset the substantially higher price of low osmolality contrast medium. BACKGROUND: Substitution of low osmolality for high osmolality contrast media in the approximately 1 million diagnostic angiocardiographic procedures performed each year in the United States could substantially increase health care costs. Cost-effectiveness estimates should include savings that might occur through reduced costs of managing adverse drug reactions. METHODS: In a randomized clinical trial of 505 persons under-going diagnostic angiography with either high osmolality or low osmolality contrast medium, we measured and compared 1) material costs of contrast media, and 2) costs from three perspectives of incremental resources used to manage contrast-related adverse drug reactions. We also performed sensitivity analyses to examine the effect of different assumptions with regard to relative risk, absolute risk and costs of adverse drug reactions on estimates of net cost of use of high osmolality and low osmolality contrast media. RESULTS: One-hundred thirty-seven (54.2%) of 253 patients receiving high osmolality contrast medium and 44 (17.5%) of 252 patients receiving low osmolality contrast medium experienced adverse drug reactions. The average cost (from society's perspective) of resources used to manage adverse drug reactions per patient undergoing angiography was significantly (p = 0.0001) greater for high osmolality (mean $249) versus low osmolality (mean $92) contrast medium. Differential costs (from the hospital's perspective) were $67 greater for high osmolality contrast medium. Charges and professional fees (from the payer's perspective) were $182 greater for high osmolality (mean $312) than for low osmolality (mean $130) contrast medium (p = 0.42, NS). The higher differential and average costs of managing adverse drug reactions with high osmolality contrast medium offset 33% and 75%, respectively, of the $207 difference in mean material costs, but these estimates are sensitive to infrequent high cost cases. CONCLUSIONS: Although low osmolality contrast medium is not cost-saving in diagnostic angiocardiography, its higher price is partially offset by lower management costs of adverse drug reactions. The cost offset for the hospital is lower than that for society and may not be realized by third-party payers. These methods and results may be useful in establishing clinical and payment guidelines for use of alternative contrast media in diagnostic angiocardiography.

Blood group antigen A and flow cytometric analysis in resected early-stage non-small cell lung cancer.

The loss of blood group antigen A on tumor tissue has been reported to be a strong adverse prognostic marker for patients with resected non-small cell lung cancer (NSCLC). Results have varied with respect to the prognostic significance of flow cytometric data. We sought to confirm the prognostic significance of blood group antigen A loss and flow cytometry in a large cohort of patients with early-stage NSCLC. Two hundred and sixty patients with surgically resected stage I (n = 193) and II (n = 67) NSCLC with at least a 5-year follow-up were identified. Using paraffin-embedded primary tumor, immunohistochemical stains for blood group antigen A were performed on 90 patients with blood type A or AB. The DNA index and percentage of cells in S phase were successfully obtained on 188 and 152 patients, respectively. The median survival time of the patients with primary tumors negative for blood group antigen A was 38 months (n = 36), compared with 98 months (n = 54) for those with antigen A-positive tumors (P < 0.01). The median disease-free survival times for antigen A-negative and -positive tumors were 26 months and 98 months, respectively (P < h 0.01). The median survival time of the patients with aneuploid tumors was 51 months (n = 131), compared with 50 months (n = 57) for those with diploid tumors (P = 0.42). The median survival time of the patients with S phase >8% was 44 months (n = 105), compared with 60 months (n = 47) for those with S phase

Phase II trial of weekly paclitaxel and concurrent radiation therapy for locally advanced non-small cell lung cancer.

We conducted a prospective Phase II study to determine the response rate, toxicity, and 2-year survival rate of concurrent weekly paclitaxel and radiation therapy (RT) for locally advanced unresectable non-small cell lung cancer. The weekly paclitaxel regimen was designed to optimize the radiosensitizing properties of paclitaxel. Thirty-three patients with unresectable stage IIIA and IIIB non-small cell lung cancer from six institutions were entered into the study between March 1994 and February 1995. Weekly i.v. paclitaxel (60 mg/m2; 3-h infusion) plus concurrent chest RT (60 Gy over 6 weeks) was delivered for 6 weeks. Twenty-nine patients were evaluable for response. Three patients achieved a complete response (10%), and 22 patients (76%) achieved a partial response, for an overall response rate of 86% (95% confidence interval, 68-96%). One patient progressed during the therapy, and three patients had stable disease. Esophagitis was the principal toxicity. Grade 3 or 4 esophagitis occurred in 11 patients (37%). One patient died of pneumonia after completion of therapy. Additional grade > or =3 toxicities included pneumonitis (12%) and neutropenia (6%). One patient had a grade 3 hypersensitivity reaction. The median overall survival duration for all 33 patients who entered the study was 20 months, and 1-, 2-, and 3-year overall survival rates were 60.6%, 33.3%, and 18.2%, respectively. The median progression-free survival duration for all 33 patients was 10.7 months, and 1-, 2-, and 3-year progression-free survival rates were 39.4%, 12.1%, and 6.1%, respectively. Weekly paclitaxel plus concurrent RT is a well-tolerated outpatient regimen. The survival outcome from this regimen is encouraging and seems to be at least equivalent to that of other chemotherapy/radiation trials. These findings warrant further clinical evaluation of weekly paclitaxel/RT in Phase II trials in the neoadjuvant setting and in combination with other cytotoxic agents.

A study of filgrastim (rG-CSF) priming of etoposide/cisplatin in advanced non-small cell lung cancer.

A previous phase II study (CALGB 9132) of etoposide/cisplatin + rG-CSF in patients with advanced non-small cell lung cancer (NSCLC) showed a marked difference in the absolute neutrophil count (ANC) nadirs between courses 1 and 2. Median ANC nadirs for courses 1 and 2 were 200 and 2500, respectively, suggesting a priming effect for rG-CSF. The present study was designed to determine whether rG-CSF given prior to the first cycle of chemotherapy would decrease the severity and duration of neutropenia. Twelve patients with stage IIIB or IV NSCLC and performance status 0-1 received rG-CSF 5 microg/kg for 5 consecutive days starting 7 days before treatment with etoposide 200 mg/m2 on days 1-3 and cisplatin 35 mg/m2 on days 1-3, repeated every 3 weeks. Patients also received rG-CSF 5 microg/kg s.c. day 4 to postnadir ANC > 10,000. The median WBC nadir, ANC nadir, and platelet nadir after the first cycle of chemotherapy in the historical group (CALGB 9132) were 1300 cells/microl, 200 cells/microl, and 80,000 cells/microl, respectively. In the present study, the median WBC nadir, ANC nadir, and platelet nadir were 1300 cells/microl, 144 cells/microl, and 56,000 cells/microl, respectively. The median time for ANC to reach 10,000 cells/microl was 15 days in both the historical and the present study. For course 2, the WBC, ANC, platelet nadirs, and duration of grade 4 neutropenia were 2600, 1450, 70,000, and 0 days, respectively. This study failed to show a priming effect for rG-CSF when given in this dose and schedule.

Cisplatin-carboplatin therapy in extensive non-small cell lung cancer: a Cancer and Leukemia Group B study.

The response to cisplatin in non-small cell lung cancer (NSCLC) is limited by the renal and neurological toxicities of this agent. Carboplatin has modest activity in NSCLC and when given in conventional doses has a different spectrum of toxicity. Both drugs were administered to 76 eligible patients with advanced NSCLC. No patient had been previously treated with chemotherapy. Cisplatin 50 mg/m2 and carboplatin 350 mg/m2 were administered every 28 days until disease progression occurred. There was 1 complete response and the overall response rate among the 68 evaluable for response patients was 13%. Neither histological subtype nor initial performance status was a significant factor influencing response. Median survival was 5.1 months with significant differences based on initial performance status but not on histological subtype. Severe or life-threatening leukopenia and thrombocytopenia occurred in 23% and 36% of the 76 patients, respectively. There were 2 toxic deaths, 1 each due to infection and haemorrhage. The efficacy of this combination is not different from that of carboplatin alone, and the combination may be of greater benefit in patients with more responsive tumours than NSCLC.

Prognostic significance of Lewis y antigen in resected stage I and II non-small cell lung cancer.

BACKGROUND: The role of Lewis y (Le(y)) antigen expression has been studied extensively in predicting the outcome of various malignancies. We evaluated the expression of Le(y) and its relationship to survival, disease-free survival and other clinicopathologic variables in patients with stage I and II non-small cell lung cancer (NSCLC). OBJECTIVE: To investigate the prognostic significance of Le(y) antigen expression in a large group of well characterized patients with resected stage I and II NSCLC. PATIENTS: Two hundred and sixty patients with surgically resected stage I (n = 193) and II (n = 67) NSCLC with at least 5-year follow-up were identified. RESULTS: The median survival for patients with negative expression of Le(y) (< 50% of cells that were positive) was 46 months, whereas for those with positive expression of Le(y) (> or = 50%), the median survival was 54 months (p = 0.99). The disease-free survival for patients with Le(y)(-) expression was 39 months and 34 months for patients with Le(y)(+) expression (p = 0.3). CONCLUSIONS: We found no relationship between loss of blood group antigen A and expression of Le(y). No statistically significant difference was found in survival between positive and negative expression of Le(y) antigen in patients with resected stage I and II NSCLC.

Phase I study of gemcitabine and carboplatin in advanced non-small cell lung carcinoma.

BACKGROUND: This phase I study was designed to determine the maximum tolerated dose of carboplatin with a fixed dose of gemcitabine without growth factor or hematopoietic precursor support. METHODS: Nineteen patients with previously untreated non-small cell lung cancer (NSCLC) were treated at three dose levels. Initially, the gemcitabine dose was 1000 mg/m(2) given on days 1 and 8. Of the first five patients treated with carboplatin AUC 4, three experienced dose limiting toxicity (DLT). The study was, therefore, amended to decrease the dose of gemcitabine to 800 mg/m(2) given on days 1 and 8 in a 21-day cycle. RESULTS: Dose limiting toxicity (neutropenia and thrombocytopenia) were seen at dose level 2A (carboplatin AUC=5). Thus, no further dose escalation was performed. Grade 3 and 4 toxicities were seen as follows: leukopenia in five of 18 (28%); neutropenia, four of 18 (22%); and thrombocytopenia, four of 18 (22%) evaluable patients. Grade 3 or 4 anemia occurred in one of 18 (6%) patients and no neutropenic fever or treatment related mortality was observed. Partial responses were seen in six patients and one patient with evaluable disease had an objective response. The overall response rate was 37% (seven of 19). Six other patients had stable disease. A total of 89 courses were administered with a median of five courses per patient (range: two to six courses). The median time to progression for all patients was 3.7 months. The median overall survival was 7.4 months with four patients still alive (median follow up 13.5 months). The survival at 6 months and 1 year is 64 and 23%, respectively. CONCLUSION: The maximum tolerated dose (MTD) in this group of patients was defined as carboplatin AUC 4 when administered with gemcitabine 800 mg/m(2) on days 1 and 8 of a 21-day schedule.

Use of neuroendocrine markers, p53, and HER2 to predict response to chemotherapy in patients with stage III non-small cell lung cancer: a Cancer and Leukemia Group B study.

Several studies have suggested that non-small cell lung cancer (NSCLC) patients whose tumors have neuroendocrine (NE) features may be more responsive to chemotherapy. In addition, increased expression of p53 and HER2 may confer relative chemotherapy resistance and shortened survival. The Cancer and Leukemia Group B performed a series of studies involving sequential chemotherapy followed by radiation for patients with unresectable stage III NSCLC. The objectives of this study were to analyze pathological specimens using immunohistochemistry for NE markers, p53 and HER2 to determine if there was a correlation between marker expression and response or survival. Of 160 eligible patients, 28 (18%) were not evaluable because of inadequate material. The percentage of specimens positive for markers was as follows: neuron-specific enolase 38%, Leu-7 2%, chromogranin A 0%, synaptophysin 5%, > or =2+NE markers 3%, p53 61%, and HER2 65%. There was no statistically significant correlation between any individual marker and response to induction chemotherapy or response to combined chemotherapy/radiation except for synaptophysin. Six of 6 (100%) synaptophysin positive tumors responded by the completion of all therapy compared with 69/125 (55%) synaptophysin negative tumors (P=0.04). None of the individual markers had a significant effect on survival in univariate analysis. Neuron-specific enolase was marginally significant in multivariate analysis (P=0.08). In conclusion, this study did not demonstrate that expression of NE markers, p53 and HER2 were predictive of response to chemotherapy, combined chemotherapy/radiation or for survival in this group of patients with stage III NSCLC. Future studies must employ either different markers or be performed on more adequate surgical specimens.

A randomized phase II study of ifosfamide/mesna/cisplatin plus G-CSF or etoposide/cisplatin plus G-CSF in advanced non-small cell lung cancer: a Cancer and Leukemia Group B study.

This Phase II study was designed to determine the efficacy of two chemotherapy regimens with G-CSF support for patients with advanced non-small cell lung cancer (NSCLC). One-hundred and one patients with Stage IIIB or IV NSCLC and performance status 0-1 were randomized to receive ifosfamide 2.0 g/m2 days 1-3, mesna 400 mg/m2 at 0, 4, 6 h days 1-3, cisplatin 33 mg/m2 days 1-3 or etoposide 200 mg/m2 days 1-3, cisplatin 35 mg/m2 days 1-3. Both groups received G-CSF 5 micrograms/kg SQ day 4 to the post day 11 absolute neutrophil count > 10 000. For the 47 eligible patients receiving ifosfamide/mesna/cisplatin, the response rate was 26% (95% confidence interval: 14-40%) and the median survival 7.5 months (95% confidence interval: 5.8-11.0 months). Grade 3 or worse toxicities were: neutropenia 75%, thrombocytopenia 70%, infection 21%. There were two treatment-related deaths due to infection. For course 1, the median absolute neutrophil count nadir was 1.3, platelet nadir 96 000 and incidence of febrile neutropenia 16%. For the 48 eligible patients receiving etoposide/cisplatin, the response rate was 21% (95% confidence interval: 11-35%) and median survival 5.8 months (95% confidence interval: 4.5-9.7 months). Grade 3 or worse toxicities were: neutropenia 90%, thrombocytopenia 58%, infection 29%. There were three treatment-related deaths due to infection. For course 1, the median absolute neutrophil count was 0.2, platelet nadir 80 000 and incidence of febrile neutropenia 33%. For both ifosfamide/mesna/cisplatin and etoposide/cisplatin, median duration of Grade IV neutropenia was short (< or = 4 days), time to subsequent courses 21 days and dose delivered > 95% of planned dose. Although G-CSF allowed full doses of drugs to be delivered on schedule, both ifosfamide/mesna/cisplatin and etoposide/cisplatin produced response rates and survival similar to other cisplatin-based regimens. In view of the significant cost of G-CSF and no obvious improvement in response rate, survival or toxicity profile, G-CSF cannot be recommended with these chemotherapy regimens for patients with advanced NSCLC.

Prognostic significance of Ki-67 immunostaining and symptoms in resected stage I and II non-small cell lung cancer.

The proliferative rate of a tumor has been considered predictive of its clinical course. We evaluated the expression of the proliferative marker Ki-67 and its relationship to survival, disease-free survival and other clinicopathologic variables in both stage I and stage II non-small cell lung cancer (NSCLC). A total of 260 patients with surgically resected stage I (n = 193), and II (n = 67) NSCLC with at least 5 years follow-up were identified. The median survival for patients with low expression of Ki-67 (< or = 25%) was 54 months, while for those with high expression (> 25%), it was 45 months (P = 0.1). The disease-free survival in patients with low expression of Ki-67 was 59 months while it was only 32 months for patients with high Ki-67 (P = 0.1). Out of 136 patients, 84 (62%) had both increased S-phase (> 8%) and high Ki-67 (P = 0.001). A total of 28 of 30 patients who had loss of antigen A had high expression of Ki-67 (93.3%) (P = 0.03). Ki-67 expression was also higher in squamous cell (54/63, 85.7%) compared to nonsquamous cell cancer (70/108, 64%) (P = 0.03). We also analyzed for the presence of symptoms with survival. The presence of symptoms was not found to be statistically significant, for overall survival (P = 0.33) or disease-free survival (P = 0.72). When individual symptoms were analyzed, the presence of cough was statistically significant for both overall and disease-free survival. The median survival was 39 months for patients with cough, and 57 months for patients without cough (P = 0.04). Multivariate analysis showed higher N and T stages, presence of cough and loss of antigen A, predicted for poorer overall survival. Higher N and T stages, loss of antigen A, presence of mucin and cough and increased expression of Ki-67 predicted decreased disease-free survival. Although we did not find a statistically significant difference between low and high Ki-67, there was a trend for a poorer overall and disease-free survival in patients with high Ki-67 expression. Larger studies may be needed to prove a statistically significant effect of Ki-67 on survival. Future studies should assess the potential prognostic significance of the presence of symptoms (particularly cough) in addition to clinical-pathologic variables (such as T and N stage) and biological markers in patients with early stage NSCLC.

Analysis of neuroendocrine markers, HER2 and CEA before and after chemotherapy in patients with stage IIIA non-small cell lung cancer: a Cancer and Leukemia Group B study.

Several studies have suggested that biochemical or molecular markers examined in non-small cell lung cancer carry prognostic or treatment response information. Non-small cell lung cancer patients whose tumors have neuroendocrine (NE) features may be more responsive to chemotherapy. In addition, increased expression of HER2 (c-erbB-2), a membrane-bound receptor with tyrosine kinase activity, has been associated with shortened survival. The Cancer and Leukemia Group B (CALGB) performed a study of patients with stage IIIA (N2 nodes positive) non-small cell lung cancer in which patients received initial chemotherapy followed by surgery, then post-operative therapy consisting of sequential chemotherapy and radiation therapy. Since all patients underwent mediastinoscopy, this provided an opportunity to compare pre- and post-chemotherapy tumor specimens to test the hypothesis that these proteins would predict treatment response. In particular, we hypothesized that the post-chemotherapy specimens would be enriched for NE marker negative cells because of the increased sensitivity of NE positive cells to chemotherapy. We performed immunohistochemical analysis for a panel of NE markers [neuron-specific enolase (NSE), Leu-7, chromogranin A (ChrA), synaptophysin (Syn)], HER2 and CEA to determine if there was an effect of therapy on the percentage of cells expressing these markers. Secondary endpoints were a correlation with chemotherapy response and survival. Slides were scored for intensity (0-4) and percentage of cells positive (0-4). Of 61 eligible patients, there were 38 with both pre- and post-chemotherapy specimens. When both intensity of staining and percentage of positive cells were considered, post-chemotherapy specimens had a higher percentage of positive NE markers compared with pre-chemotherapy. In addition, there was no correlation between NE marker, HER2 or CEA expression (prior to or post treatment) and response to chemotherapy or survival. These data do not support the hypothesis that NE positive tumor cells are preferentially killed by chemotherapy in patients with stage IIIA non-small cell lung cancer.

Efficacy of immune globulin in preventing complications of bone marrow transplantation: a meta-analysis.

To assess the efficacy of immune globulin in preventing CMV infection, interstitial pneumonia, GVHD and death after BMT, we reviewed and synthesized data from 12 published studies (with 1282 patients) in which immune globulin was used prophylactically in BMT patients, controls were included and clinical outcomes were assessed. Data synthesis indicates that immune globulin significantly reduces fatal CMV infection (odds ratio (OR) 0.47, 95% confidence interval (CI) 0.23-0.99), CMV pneumonia (OR 0.61, CI 0.42-0.89), non-CMV interstitial pneumonia (OR 0.57, CI 0.35-0.95) and total mortality (OR 0.74, CI 0.55-0.99). The reduction in acute GVHD was not quite significant (OR CI 0.45-1.02). Complications decrease with both hyperimmune and conventional immune globulin. For CMV-negative transplant recipients, immune globulin decreases symptomatic CMV infection (OR 0.55, CI 0.31-0.94) and interstitial pneumonia (OR 0.34, CI 0.15-0.77). For CMV-positive recipients, immune globulin prevents interstitial pneumonia (OR 0.45, CI 0.26-0.80) but not symptomatic CMV infection (CI 0.41-2.80). We conclude that immune globulin is efficacious in preventing major complications of BMT in both CMV-negative and CMV-positive recipients.

Preoperative evaluation of stage I and stage II non-small cell lung cancer.

The appropriate preoperative evaluation for occult metastasis in patients with potentially resectable lung cancer remains controversial. The records of 265 patients with stage I and II non-small cell lung cancers who underwent resection with curative intent were reviewed to determine if there was a survival benefit of negative preoperative scanning to detect metastases. A minimum of 5 years of follow-up was possible for all long-term survivors. Patients having preoperative bone scans, brain imaging, and abdominal imaging had no increased survival over those without such evaluation (using Kaplan-Meier survival curves). Additionally, no difference was found in the time to first recurrence between these groups, and the site of recurrence was independent of a negative preoperative scan for that location. These data, using patient outcome as the basis of our conclusion, support a policy of reserving expensive preoperative metastatic evaluations only for those patients with clinical evidence of metastatic disease.