Efficient nonparametric confidence bands for receiver operating-characteristic curves.

Receiver operating-characteristic curve is a popular graphical method frequently used in order to study the diagnostic capacity of continuous (bio)markers. In spite of the existence of a huge number of papers devoted to both theoretical and practical aspects of this topic, the construction of confidence bands has had little impact in the specialized literature. As far as the authors know, in the CRAN there are only three R packages providing receiver operating-characteristic curve confidence regions: plotROC, pROC and fbroc. This work tries to fill this gap studying and proposing a new nonparametric method to build confidence bands for both the standard receiver operating-characteristic curve and its generalization for nonmonotone relationships. The behavior of the proposed procedure is studied via Monte Carlo simulations and the methodology is applied on two real-world biomedical problems. In addition, an R function to compute the proposed and some of the previously existing methodologies is provided as online supplementary material.

Receiver operating characteristic curve generalization for non-monotone relationships.

The receiver operating characteristic curve is a popular graphical method frequently used in order to study the diagnostic capacity of continuous markers. It represents in a plot true-positive rates against the false-positive ones. Both the practical and theoretical aspects of the receiver operating characteristic curve have been extensively studied. Conventionally, it is assumed that the considered marker has a monotone relationship with the studied characteristic; i.e., the upper (lower) values of the (bio)marker are associated with a higher probability of a positive result. However, there exist real situations where both the lower and the upper values of the marker are associated with higher probability of a positive result. We propose a receiver operating characteristic curve generalization, [Formula: see text], useful in this context. All pairs of possible cut-off points, one for the lower and another one for the upper marker values, are taken into account and the best of them are selected. The natural empirical estimator for the [Formula: see text] curve is considered and its uniform consistency and asymptotic distribution are derived. Finally, two real-world applications are studied.

Impact of the incorporation of tyrosine kinase inhibitor agents on the treatment of patients with a diagnosis of advanced renal cell carcinoma: study based on experience at the Hospital Universitario Central de Asturias.

INTRODUCTION: For nearly the past two decades, cytokines (CKs) have been the only systemic treatment option available for advanced renal cell carcinoma (RCC). In recent years, tyrosine kinase inhibitors (TKIs) have demonstrated clinical activity on this tumour. Our purpose is to describe one centre's experience with the use of CKs and TKIs in the treatment of patients with advanced RCC. MATERIALS AND METHODS: This study was designed as a retrospective chart review of RCC patients who were treated with CKs and/or TKIs in our department between July 1996 and June 2008. Efficacy and toxicity were assessed using World Health Organization (WHO) criteria. The Kaplan-Meier method was used to estimate progression-free (PFS) and overall (OS) survival. RESULTS: Ninety-four patients were classified into three groups depending on the modality of treatment administered: 46 were treated with CKs alone and/or chemotherapy (27 with immunotherapy, one with chemotherapy and 18 with both), 28 with TKIs alone (25 with sunitinib and 13 with sorafenib) and 20 with TKIs in second-line treatment following failure with CKs (17 with sunitinib, eight with sorafenib, four with bevacizumab and one with lapatinib). The median age was 60 years in the CK group and 65 and 62, respectively, in TKI in first and second-line treatment groups. Eighty-five percent of patients treated with CKs and 75% in the TKI group in first-line treatment and 80% in second-line treatment were men. Overall, 89% of patients had favourable risk, and 11% had intermediate risk. All patients were considered evaluable for toxicity. The main grade 3-4 (%) toxicity was asthenia for both groups, (ten in TKIs and 15 in CKs). Other grade 1-2 toxicities were mucositis (39), bleeding (8), hypertension (19), skin toxicity (33) and hypothyroidism (12.5) associated with TKIs; and anaemia (33), cough (29), asthenia (39) and emesis (14) associated with CKs. The objective response rate among 80 patients evaluable for activity was 10.6% with CKs and 46.5% and 35%, respectively, with TKIs in first- and second-line treatments. Disease stabilisation with CKs was recorded at 59% of patients and with TKIs 25% and 50% in first- and second-line treatment groups, respectively. The median progression-free survival (PFS) with CKs was 122 days [95% confidence interval (CI) 82-162] and with TKIs 201 days (65-337) in the first and 346 days (256-436) in second-line treatment groups. The median overall survival (OS) was 229 days (142-316) and 2,074 days (1,152-2,996) for patients treated with CKs and TKIs. CONCLUSIONS: Our results are in line with the activity and survival rates previously reported in the literature regarding the use of TKIs for patients with advanced RCC in first- and second-line treatment, which has demonstrated an acceptable toxicity level.

Cisplatin plus gemcitabine with or without vinorelbine as induction chemotherapy prior to radical locoregional treatment for patients with stage III non-small-cell lung cancer (NSCLC): results of a prospective randomized study.

To evaluate possible improvement in objective response of adding vinorelbine (V) to the combination of cisplatin/gemcitabine (CG) in induction chemotherapy for stage III NSCLC, patients (n=154) aged < or =75 years, Karnofsky index > or =70%, were stratified by stage (IIIA versus IIIB) and randomly assigned to receive: C (50mg/m(2) i.v.) plus G (1250mg/m(2) i.v.) or CG plus V (25mg/m(2) i.v.). All drugs were administered on days 1 and 8 of an every 3-week cycle. At conclusion, local treatment (LT) with surgery and/or radiotherapy was scheduled. The results indicated that, following a median of 3 cycles, the overall efficacy was 65% in the CG and 61% in the CGV group. Most patients in both groups received radiotherapy as part of their LT. Pathological complete response was confirmed by surgery in 18% in the CG and 25% in the CGV group. Median progression-free survival was 368 days in the CG and 322 days in the CGV group. There were no statistically significant differences in toxicities between groups. We conclude that the CG and CGV combinations had similar efficacy and moderate toxicity, without accruing to the triplet combination.

Gemcitabine and vinorelbine (GV) versus cisplatin, gemcitabine and vinorelbine (CGV) as first-line treatment in advanced non small cell lung cancer: results of a prospective randomized phase II study.

The objective of this study was to assess whether adding cisplatin to gemcitabine/vinorelbine combination improves the clinical outcome in patients with non-small-cell lung cancer (NSCLC). Chemotherapy-naïve patients with advanced NSCLC; age < or = 75 years: Karnofsky performance status > or = 60%, and with adequate hematological, renal and hepatic function, were randomized into 2 treatment groups to receive Gemcitabine 1250 mg/m2 + vinorelbine 30 mg/m2 (GV group), or cisplatin 50 mg/m2 + gemcitabine 1000 mg/m2 + vinorelbine 25 mg/m2 (CGV group). All drugs were administered on days 1 and 8 every three weeks: From September 1999 to March 2003, 114 patients were enrolled. No statistically significant difference was observed in GV vs CGV group in objective response (37 versus 47%, respectively; P = 0.5), median time to progression (5 versus 5.8 months; P = 0.6), overall survival (9 versus 10 months; P = 0.9) and 1-year survival (26 versus 28%; P = 0.9). Conversely, toxicities were significantly higher for CGV, including grade 3-4 neutropenia (24 versus 45%); neutropenic fever (4 versus 14%, including one toxic death); grade 3-4 thrombocytopenia (2 versus 14%); and grade 3-4 emesis (2 versus 14%). Our results suggest that the combination of gemcitabine and vinorelbine is less toxic than three-drug combination with cisplatin while showing similar efficacy.

Combination of docetaxel, epirubicin and vinorelbine administered every 2 weeks as first-line therapy in patients with metastatic breast cancer: a dose-finding study.

AIMS: To assess efficacy and optimum combination dosage of intravenous docetaxel (T), epirubicin (E) and vinorelbine (N) administered every 2 weeks and without colony stimulating factor (CSF) support in patients with metastatic breast cancer (MBC). PATIENTS AND METHOD: Patients (n = 5 1) with MBC were consecutively assigned to four different dose levels (DL) to receive (in mg/m2): Level I = T35 + E30 + N25; Level II = T30 + E30 + N25; Level III = T30 + E25 + N25; and Level IV = T25 + E25 + N25. Consecutive cycles were delayed if absolute neutrophil and/or platelet counts fell below 1.5 x 10(9) and 100 x 10(9) l(-1), respectively. Treatment at a given dose level was suspended if 33% or more of patients included in a given cohort had unacceptable toxicity. RESULTS: The patients evaluable for toxicity (n = 48) received 448 cycles (median 9; range 1-23). There was neutropenia G 3-4 in 30 patients (63%) with fever in 3 (6%). The G 2-3 non-hematological toxicities were alopecia in 39 patients (81%), mucositis in 11 (23%), and nausea/vomiting in 8 (17%). There were no toxic deaths. Treatment delay or dose reduction after first cycle occurred in > or = 30% of patients treated in all DLs, except the fourth. Objective response was achieved in 29 of the 47 evaluable patients (58%; 95% CI: 50-66). The median duration of response, time-to-progression and overall survival were 13, 11 (range 8-14) and 20 (range 16-24) months, respectively. CONCLUSION: The combination of docetaxel, epirubicin and vinorelbine without CSF support ought not to exceed 25 mg/m2 every 2 weeks. The efficacy is no greater than other existing regimens for first-line treatment of MBC.

Phase I/II study of gemcitabine and vinorelbine plus cisplatin in non-small cell lung cancer.

PURPOSE: We determine the maximum tolerated dose (MTD) and efficacy of gemcitabine plus vinorelbine combined with cisplatin in patients with non-small cell lung cancer (NSCLC). PATIENTS AND METHODS: Chemo naive patients with stage IIIA to IV non-small cell lung cancer received outpatient administration of gemcitabine 1,000 mg/m2 and vinorelbine 25 mg/m2 intravenously on days 1 and 8 every 21 days. Doses of gemcitabine and vinorelbine were escalated by 250 mg/m2 and 5 mg/m2, respectively, at each dose level. Cisplatin was administered at a fixed dose of 50 mg/m2 on days 2 and 9. After the MTD was reached, the study was continued as a phase II trial. RESULTS: From January 1998 to March 1999, sixty-five patients were enrolled. The first 38 patients participated in the phase I evaluation. After 130 cycles, the dose-limiting toxicities were neutropenia, stomatitis, asthenia, and hepatotoxicity occurring at the third and fourth dose levels (doses of gemcitabine/vinorelbine of 1,500/25 and 1,000/30 mg/m2). For the subsequent phase II evaluation, 27 additional patients, out of a total of 53, receiving the MTD of gemcitabine and vinorelbine (1000-1250/25 mg/m2) followed (24 hours later) by cisplatin 50 mg/m2. Thirty one (58%) of 53 assessable patients responded. Objective response for patients with stages III and IV disease, respectively, were 65% and 47%. The median time to progression and the overall survival time were 9 months (95% CI: 5-12) and 11 months (95% CI: 9-13), respectively. World Health Organization toxicity > or = grade 3 neutropenia was registered in 28 (54%) of 52 assessable patients (2% with febrile neutropenia), and > or = grade 3 thrombocytopenia in 15 (29%) patients (4% with bleeding). Nausea/vomiting (> or = grade 2) and asthenia (moderate to severe) occurred in 24 (46%) and 14 (27%) patients, respectively. CONCLUSION: Gemcitabine 1,000-1,250 mg/m2 plus vinorelbine 25 mg/m2 on days 1 and 8, followed by cisplatin 50 mg/m2 24 hours later, is safe for outpatient administration and active in patients with NSCLC.

Phase I/II study of gemcitabine plus vinorelbine in non-small cell lung cancer.

BACKGROUND: Because gemcitabine and vinorelbine have demonstrated single-agent activity in non-small cell lung cancer (NSCLC), we conducted this phase I/II study to determine the maximum tolerated dose (MTD) and activity of these drugs combined. PATIENTS AND METHODS: Patients with inoperable or advanced NSCLC and no prior chemotherapy were treated with gemcitabine plus vinorelbine on days 1 and 8 every 21 days. The initial doses of gemcitabine 1,000 mg/m2 and vinorelbine 25 mg/m2 were escalated by 250 mg/m2 and 5 mg/m2, respectively, in separate patient cohorts until the MTD was established. RESULTS: In phase I, 32 patients received a total of 115 cycles. Dose-limiting toxicities were neutropenia and hepatotoxicity, occurring at the dose level of 1,500 mg/m2 and 30 mg/m2. Thus, the MTD used for phase II was 1,250 mg/m2 and 30 mg/m2. Of 41 patients in phase II, 16 (39%) achieved objective responses (95% confidence interval [CI] 24% to 54%), with a median time to progression of 4.2 months. Overall survival was 9 months (95% CI 5.7 to 12.7 months) and the 1-year survival rate was 31%. World Health Organization (WHO) > or = grade 3 neutropenia and reversible thrombocytosis occurred in 15% and 65% of patients, respectively. Non-hematologic toxicity was mild at all dose levels. Grades 3 and 4 hepatotoxicity were reported in one patient each. CONCLUSION: The combination of 1,250 mg/m2 gemcitabine and 30 mg/m2 vinorelbine on days 1 and 8 every 21 days is well tolerated and active in patients with NSCLC. These results should be confirmed in comparative studies.