One-sided Alternatives Research Articles

On Tests for Special Two-Sample Location Problem Based on Subsample Minimum and Median

Shetty and Umarani(2005) considered a special type of two-sample location problem which has some potential applications, particularly in comparing the performances of two packing machines. In this paper, a new class of test statistics is proposed which is based on sub-sample minima of first sample and sub-sample median of second sample. The performance of members of the new class is evaluated in terms of Pitman asymptotic relative efficiency in comparision with Shetty and Umarani(2005). It can be concluded that the performance of the new proposed class of tests is better for medium and heavy tailed distributions such as Normal, Logistic, Laplace and Cauchy. For light tailed distributions the performance of new testis better in specific cases. It is proved that the test is consistent for one-sided alternatives.

A phase II pilot study of the efficacy and safety of nivolumab and ipilimumab in T1aN0M0 renal cell carcinoma patients ineligible for surgical treatment.

TPS769 Background: Surgery remains the standard curative-intent therapy for localized renal cell carcinoma (RCC). Thus, systemic therapy for RCC should only be considered in patients who have contraindications to surgery. The phase 3 CheckMate 214 trial showed that a combination of nivolumab and ipilimumab has a significant impact on tumor burden in intermediate- and poor-risk metastatic RCC patients with a complete response rate of 11%. Median time to objective response was 2.8 months. Among all complete responders to nivolumab plus ipilimumab in the intention-to-treat population, 5% achieved a complete response at the first scan. We hypothesize that this combination could eliminate the primary tumor in T1aN0M0 patients ineligible for surgical treatment. Methods: This is a prospective, multicenter, non-randomized phase 2 pilot study. Patients with biopsy-proven clear-cell RCC of ≤4 cm (cT1a), no evidence of any metastases, and unable to have surgery or other nephron-sparing interventions (ablation) for any reason (functional single kidney with a central, high-complexity RCC, high risk of nephrectomy and dialysis, patients with complex coagulation disorders, etc.) or with preference to have no surgery or intervention for any reason will receive combination of ipilimumab (1 mg/kg intravenously) every 3 weeks for four doses, and nivolumab (240 mg intravenously) every 2 weeks during 16 weeks. The primary endpoint is complete response rate. Simon's two-stage design (Simon, 1989) is used. The null hypothesis that the true complete response rate is 11% will be tested against a one-sided alternative. This design yields a type I error rate of 0.05 and power of 0.9 when the true complete response rate is 60% (alternative hypothesis). In the first stage, 3 patients will be accrued. If there are no complete responses in these 3 patients, the study will be stopped. Otherwise, 5 additional patients will be accrued for a total of 8. The null hypothesis will be rejected if 3 or more responses are observed in 8 patients. Clinical trial information: NCT04134182.

Statistical Procedures for Assessing the Need for an Affirmative Action Plan: A Reanalysis of Shea v. Kerry

In the 1980s, reports from Congress and the Government Accountability Office (GAO) presented statistical evidence showing that employees in the Foreign Service were overwhelmingly White male, especially in the higher positions. To remedy this historical discrimination, the State Department instituted an affirmative action plan during 1990–1992 that allowed females and race-ethnic minorities to apply directly for mid-level positions. A White male employee claimed that he had been disadvantaged by the plan. The appellate court unanimously held that the manifest statistical imbalance supported the Department’s instituting the plan. One judge identified two statistical issues in the analysis of the data that neither party brought up. This article provides an empirical guideline for sample size and a one-sided Hotelling’s T2 test to answer these problems. First, an approximate rule is developed for the minimum number of expected minority appointments needed for a meaningful statistical analysis of under-representation. To avoid the multiple comparison issue when several protected groups are involved, a modification of Hotelling’s T2 test is developed for testing the null hypothesis of fair representation against a one-sided alternative of under-representation in at least one minority group. The test yields p-values less than 1 in 10,000 indicating that minorities were substantially under-represented. Excluding secretarial and clerical jobs led to even larger disparities.Supplemental materials for this article are available online.

A Phase 1b/2 Study of TAK-981, a First-in-Class Sumoylation Inhibitor, in Combination with Rituximab in Patients with Relapsed/Refractory (r/r) CD20-Positive Non-Hodgkin Lymphoma (NHL)

A Phase 1b/2 Study of TAK-981, a First-in-Class Sumoylation Inhibitor, in Combination with Rituximab in Patients with Relapsed/Refractory (r/r) CD20-Positive Non-Hodgkin Lymphoma (NHL)

Abnl Marro: An International Cooperative Trial for Patients with MDS/MPN Overlap Syndromes

Abnl Marro: An International Cooperative Trial for Patients with MDS/MPN Overlap Syndromes

BrUOG 345: Fractionated Gemtuzumab Ozogamicin Followed By Non-Engraftment Donor Leukocyte Infusions for Relapsed/Refractory Acute Myeloid Leukemia

Background: Allogeneic stem cell transplant (allo-SCT) is a cornerstone in the treatment of Acute Myeloid Leukemia (AML) exerting much of its therapeutic efficacy through the graft-versus-leukemia effect. The administration of immunoregulatory cells via engrafted donor lymphocytes is essential to the eradication of residual malignancy and long-term survival. Many patients, however, either do not have an allo-SCT donor or cannot withstand the toxicity of allo-SCT. Previous studies have shown durable responses in AML patients following donor lymphocyte infusions in the absence of engraftment in both the frontline and relapsed setting (Dey et al, BJH 2005, Colvin et al, BBMT 2009, Ai et al, Blood 2010). In this clinical trial we propose a role for donor leukocyte infusions (DLI) in the absence of engraftment. Without the need for engraftment patients will not need to receive high dose chemotherapy or radiation and the toxicities that accompany these therapies. Instead, allogeneic donor cells are infused into patient with relapsed/refractory (R/R) AML to serve as a potent immune stimulator. Prior to DLI, patients will receive fractionated dosing of gemtuzumab ozogamicin (GO). GO is an anti-CD-33 antibody drug conjugate approved in combination with induction therapy for de novo AML and in R/R disease. Patients who demonstrate a CR or CRi to therapy will go on to have up to 2 additional GO + DLI treatments. Bone marrow and blood samples will be obtained from patients before, during, and after treatment to determine immune effector cells (both donor and patient), cytokine release profiles, and extracellular vesicle components. Study Design and Methods: Our study, BrUOG 345 [NCT03374332], evaluates the combination of fractionated GO with non-engraftment DLI in the treatment of patients with R/R AML. Adults patients 18 years of age and older with histologically confirmed R/R AML who have had recurrence or progression after at least 1 prior standard treatment are eligible. Enrollees must have no active systemic infections and have adequate lung, liver, cardiac, and renal function with an ECOG PS 0-1. Fractionated GO 6-9mg/m2 will be administered on days 1, 4, and 7 followed by infusion of 1-2x108 CD3 cells/kg from a 0-3/6 HLA mismatched related donor cell. Patients that are in CR or CRi can receive up to 2 additional treatments with GO+DLI (Figure 1). The primary objective of the phase 1 portion is to determine the maximum tolerated dose (MTD) of DLI in combination with GO while the primary objective of Phase 2 portion is to assess the response rate after one cycle of fractionated GO followed by non-engraftment DLI in patients with relapsed/refractory AML. The study will initially utilize a 3+3 design in Phase 1 to determine if 1-2x108 CD3 cells/kg can be safely administered with GO. This study will target a response rate of 57% considered to be interesting enough to warrant further study in a randomized setting. With this hypothesis in mind, the phase 2 portion of the study will use Simon's two-stage design. The null hypothesis that the true response rate (CR and CRi) is only 29% will be tested against a one-sided alternative. In the first stage, 9 subjects will be accrued. Patients treated at MTD in the Phase 1 portion of the study will be included in this cohort. If there are 3 or fewer responses in these 9 patients, the study will be stopped for futility. Otherwise, 6 additional subjects will be accrued for a total of 15. The null hypothesis will be rejected if 7 or more responses are observed in 15 patients. This design yields a type I error rate of 0.1 and power of 80% if the true response rate is 57%. A continuous assessment of toxicity will be utilized for the Phase 2 portion of this study. Sequential boundaries will be used to monitor dose-limiting toxicity rate for patients after the initial MTD is determined. Accrual will be halted if excessive numbers of dose-limiting toxicities are seen. The primary outcome of the Phase 2 portion is the CR/CRi rate following GO and non-engraftment DLI. Secondary outcomes will include survival, both progression-free (PFS) and overall (OS) until two years post treatment, and dose limiting toxicities until 16 weeks post-infusion. Additional lab correlative studies will be performed including CD33 expression before, during, and after GO infusion and T-cell activation markers, antigen presenting cell/macrophage amounts, cytokine release profiles, and extracellular vesicle measurements (Figure 1). Disclosures Olszewski: TG Therapeutics: Research Funding; Spectrum Pharmaceuticals: Research Funding; Adaptive Biotechnologies: Research Funding; Genentech: Research Funding. Reagan:Pfizer: Research Funding; Incyte: Membership on an entity's Board of Directors or advisory committees.

Testing for superiority between two variance functions

This paper focuses on the problem of testing the null hypothesis that the variance functions of two populations are equal versus one-sided alternatives under a general nonparametric heteroscedastic regression model. The asymptotic distribution of the test statistic is studied.

A New Test for Multivariate Ordered Alternatives: Medical Application and Geometric Interpretation

In the present paper, the geometric interpretation and medical application of ordered one-sided hypothesis testing are presented. We consider tests for the hypothesis that the mean vectors are zero against one-sided alternatives when the observation vectors are independently and identically distributed as normal with known and unknown covariance matrices. This problem is an extension of Kudo (1963) and Glimm et al. (2002) to several multivariate normal distributions. When the covariance matrices are known, the test statistic is derived and we give suitable conditions which under the one-sided restricted alternative hypothesis the obtained estimator is the maximum likelihood estimate. We shall discuss a geometric interpretation of the test statistic based on the closed convex polyhedral cone and also obtain its null distribution in the special case. Also, when the covariance matrices are unknown and common a proper test statistic is proposed. To evaluate the results, the effect of a drug is investigated on the body of one-day babies for suspected Sepsis disease.

Multi-Aspect Tests with Sample Quantiles

In this research, we consider multi-aspect tests based on the quantile statistics. The multi-aspect tests would be efficient in terms of power when the underlying distributions may not be known or quite different types. First of all, we use the quadratic form for the test statistic to combine several quantile statistics, which would be suitable for the general type of alternative. Also we consider sum type of statistic for the quantile functions which may be applied for the one-sided alternative. We derive the asymptotic normalities for both cases applying the large sample approximation theorem. Further, we apply the permutation principle with the Monte-Carlo method to obtain the exact null distributions and state briefly the order of application of the permutation principle. Then we illustrate our procedure with a numerical example by obtaining p-values using both methods. Finally we discuss the combination functions and resampling methods as concluding remarks.

A phase II randomised (calibrated design) study on the activity of the single-agent trabectedin in metastatic or locally relapsed uterine leiomyosarcoma

BackgroundPatients with recurrent/metastatic uterine leiomyosarcoma (U-LMS) have a dismal prognosis. This phase II study aims to evaluate trabectedin efficacy and safety in advanced U-LMS.MethodsEligible patients had received ≥ one line of chemotherapy. Gemcitabine ± docetaxel naive patients were randomised to Arm A: trabectedin 1.3 mg/m2 or calibration Arm B: gemcitabine 900 mg/m2 and docetaxel 75 mg/m2. Patients who had already received gemcitabine ± docetaxel directly entered Arm A. Primary end-point: 6-month progression-free rate (PFS-6). The null hypothesis that the true PFS-6 = 14% was tested against a one-sided alternative. This design yielded a 5% type I error rate and 90% power when the true PFS-6 is 25%.ResultsOverall, 126 patients entered Arm A (45 from randomisation and 81 directly) and 42 Arm B. Arm A patients characteristics: median age = 57; ≥2 previous chemotherapy lines = 37.4%; metastatic disease = 93%. The study met the condition for trabectedin activity: PFS-6 = 35.2% (95% CI: 26.2–45). No difference in PFS by the number of previous chemotherapy lines emerged. Median OS = 20.6 months (IQR: 8–36.4). In Arm B, the PFS-6 = 51.5% (95% CI: 33.5–69.2). No toxic deaths occurred. In Arm A, only 4 patients interrupted treatment for toxicity.ConclusionsTrabectedin is active and well tolerated, retaining similar efficacy across one to three previous lines of chemotherapy.

A New Class of Robust Two-Sample Wald-Type Tests

Parametric hypothesis testing associated with two independent samples arises frequently in several applications in biology, medical sciences, epidemiology, reliability and many more. In this paper, we propose robust Wald-type tests for testing such two sample problems using the minimum density power divergence estimators of the underlying parameters. In particular, we consider the simple two-sample hypothesis concerning the full parametric homogeneity as well as the general two-sample (composite) hypotheses involving some nuisance parameters. The asymptotic and theoretical robustness properties of the proposed Wald-type tests have been developed for both the simple and general composite hypotheses. Some particular cases of testing against one-sided alternatives are discussed with specific attention to testing the effectiveness of a treatment in clinical trials. Performances of the proposed tests have also been illustrated numerically through appropriate real data examples.

A phase II, noncomparative, open label, multicentre, study of AZD9291 in patients with locally advanced or metastatic EGFR mutated “T790M undetectable or unknown” non-small cell lung cancer (stage IIIB-IV) after no immediate prior EGFR TKI (OSIRIS study).

TPS9107 Background: Osimertinib (OSI), is an oral, potent, irreversible inhibitor of both epidermal growth factor receptor (EGFR) sensitizing and resistance mutations (T790M) indicated for the treatment of pts with advanced EGFR T790M mutation-positive NSCLC. In the AURA study, OSI was associated with an ORR of 21% (13/61) among all patients with T790M negative mutation. Response rate broken down by immediate versus no immediate prior EGFR TKI was 11% (4/36 pts) versus 36% (9/25) respectively. This better activity with deferred OSI, drug able to inhibit also the EGFR sensitizing mutations, could be explained by a selection of sensitive tumor cells during chemotherapy (re-challenge strategy). Aim of the current study is prospective evaluate the efficacy of OSI in EGFR mutated, T790M “undetectable or unknown” patients as third-line therapy after a first-line EGFR TKI and a subsequent chemotherapy. Methods: OSIRIS study is a prospective single-arm, phase 2, open label, italian multicenter study. T790M “undetectable or unknown" is defined by the following conditions: inconclusive/negative tumor test result for T790M at the time of disease progression or medical inaccessible/contraindications/declined tumor biopsy or insufficient tumor tissue for testing. Pts are treated with OSI 80 mg once daily until disease progression or unacceptable toxicity. The single-arm design is appropriate, as there is no accepted standard therapy for these pts after chemotherapy. The primary endpoint is ORR according to RECIST version 1.1. The null hypothesis that the true response rate is 9% will be tested against a one-sided alternative. In the first stage, 32 pts will be accrued. If there are 3 or fewer responses in these 32 pts, the study will be stopped. Otherwise, 49 additional pts will be accrued for a total of 81. This design yields a type I error rate of 0.05 and power of 80% when the true response rate is 19%. Secondary endpoints are PFS, OS and safety. Exploratory: mutational analysis of a panel of genes involved in resistance to EGFR-TKIs is planned. Clinical trial information: 2016-002555-17.

Phase II trial of ribociclib and everolimus in p16 low anaplastic thyroid cancer (ATC).

TPS6098 Background: ATC is a rare thyroid cancer with a highly aggressive clinical course. Median OS is 3-4 months and 90% of patients die within 1 year of the diagnosis. There are no effective treatment options in metastatic disease. Targeted therapies to ALK and BRAF are occasionally associated with dramatic responses. Retinoblastoma (Rb) inhibits cell cycle progression and is inactivated by CDK 4/6, which is the target for ribociclib. Nearly all differentiated thyroid cancers are Rb negative, conversely nearly 100% of ATC expresses intact Rb which may be crucial for rapid cell cycle progression. Ribociclib is a CDK 4/6 inhibitors that slows cell cycle progression and DNA replication in tumors with functional Rb. The p16 protein similarly inhibits CDK4/6; if p16 levels are high Rb is phosphorylated and thus inactive. p16 is low in ATC, and from our comprehensive genomic analysis 30% of ATC lacks the p16 gene CDKN2a. Most ATC demonstrates PI3K/Akt/mTOR abnormalities, this pathway represents an attractive target in ATC. The mTOR inhibitor everolimus has shown promising efficacy in ATC cell lines, especially in those with TSC2 mutation whose wild type negatively regulates mTOR. Methods: The combination of everolimus and ribociclib targets mutations/abnormalities that are frequently seen in ATC, and is tolerable based on Phase I/II trial results in the latest arm of the biomarker/oncogene driven ATC Master Protocol . This open-label trial treats metastatic Rb+ ATC patients with p16-/ CDKN2a-. Treatment is ribociclib 400 mg + everolimus 5 mg QD. The primary endpoint is the overall response rate; secondary endpoints are PFS, OS, safety and toxicity. Exploratory objectives include if tissue biomarkers or mutations noted on Next Generation Sequencing correlate for enhanced/impaired response to combination therapy. Simon's two-stage design will be used with the null hypothesis that the true response rate is 5%, this will be tested against a one-sided alternative. In the first stage, 9 patients will be accrued. If no response in these 9 patients, the study will be stopped. Otherwise, 21 additional patients will be accrued for a total of 30. Clinical trial information: NCT02289144.

Asymptotic [formula omitted] and [formula omitted] tests in an efficient GMM setting

This paper considers two-step efficient GMM estimation and inference where the weighting matrix and asymptotic variance matrix are based on the series long run variance estimator. We propose a simple and easy-to-implement modification to the trinity of test statistics in the two-step efficient GMM setting and show that the modified test statistics are all asymptotically F distributed under the so-called fixed-smoothing asymptotics. The modification is multiplicative and involves the J statistic for testing over-identifying restrictions. This leads to convenient asymptotic F tests whose critical values, i.e., the standard F critical values, are readily available from standard statistical tables and programming environments. For testing a single restriction with a one-sided alternative, an asymptotic t test theory using the standard t distribution as the reference distribution is also developed.

Abstract OT2-01-09: PALINA: A phase II safety study of palbociclib in combination with letrozole in African American women with hormone receptor positive HER2 negative advanced breast cancer

Abstract Background:Insufficient data exist to describe the hematological safety of palbociclib in African American women (AAW) who are known to have a high incidence of benign ethnic neutropenia (BEN). The studies that led to the FDA approval of palbociclib (PALOMA 1 and 3) only included participants with baseline absolute neutrophil count (ANC) of ≥1500/mm3. The standard lower limit of ANC of 1500/mm3 for initiation of treatment in those with BEN has been previously challenged. In this current study, we propose to lower the ANC cutoff for enrollment to 1000/mm3. Trial design: PALINA is a phase II study evaluating the hematological safety of palbociclib with letrozole in 35 AAW with hormone receptor (HR) positive HER2 negative advanced breast cancer and ANC ≥1000/mm3. Patients enrolled will receive palbociclib 125mg daily for 21 days followed by 7 days off and letrozole 2.5mg daily. For patients enrolled with baseline ANC between 1000-1499/mm3, initial dose of palbociclib will be 100mg daily for 21 days followed by 7 days off. Presence of Duffy Null Polymorphism (SNP rs2814778) as a predictive marker for neutrophil count will be assessed at baseline. Metabolite and exosomal signature (proteins and RNA) of drug resistance will be evaluated at different time points. Main eligibility criteria: Self-identified Black, African or AAW of ≥ 18 years of age with proven diagnosis of advanced HR-positive, HER2-negative breast cancer; ECOG performance status 0-2; ANC ≥ 1,000/mm3 and no prior receipt of CDK4/6 inhibitors. Specific aims: The primary endpoint is the proportion of patients who complete planned oncologic therapy without the development of a hematological event defined as episodes of febrile neutropenia and treatment discontinuation due to neutropenia. Additional endpoints include: number of patients who required dose delays or dose reductions in palbociclib attributed to neutropenia; rate of grade 3/4 neutropenia; clinical benefit rate at 24 weeks; correlations between metabolite and exosomal signature with disease response; correlations between baseline ANC prior to cancer diagnosis and the Duffy Null polymorphism with hematological safety. Statistical methods: The study is designed to assess the rate of completion of planned therapy in the absence of a hematological event defined as episodes of febrile neutropenia and treatment discontinuation due to neutropenia. Simon's two-stage design with a maximum of 35 patients is used. The null hypothesis that the true completion rate is 60% will be tested against a one-sided alternative. This design yields a type I error rate of 0.05 and power of 80% when the true completion rate is 80%. Present accrual and target accrual: The Institutional Review Board at Georgetown University Medical Center (GUMC) has approved the study. Enrollment of the first patient is expected in July 2016 with a total of 35 patients planned to be recruited. The recruitment sites are MedStar Georgetown University Hospital and other hospitals of the Georgetown MedStar Cancer Network, Hackensack University Medical Center and University of Alabama at Birmingham. This trial is funded by an ASPIRE Breast Cancer Research Award from Pfizer. Citation Format: Lynce F, Shajahan-Haq A, Cai L, Graham D, Gallagher C, Mohebtash M, Kamugisha L, Novielli N, Castle J, Forero A, Isaacs C. PALINA: A phase II safety study of palbociclib in combination with letrozole in African American women with hormone receptor positive HER2 negative advanced breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr OT2-01-09.

A Bayesian bird's eye view of 'Replications of important results in social psychology'.

We applied three Bayesian methods to reanalyse the preregistered contributions to the Social Psychology special issue ‘Replications of Important Results in Social Psychology’ (Nosek & Lakens. 2014 Registered reports: a method to increase the credibility of published results. Soc. Psychol. 45, 137–141. (doi:10.1027/1864-9335/a000192)). First, individual-experiment Bayesian parameter estimation revealed that for directed effect size measures, only three out of 44 central 95% credible intervals did not overlap with zero and fell in the expected direction. For undirected effect size measures, only four out of 59 credible intervals contained values greater than (10% of variance explained) and only 19 intervals contained values larger than . Second, a Bayesian random-effects meta-analysis for all 38 t-tests showed that only one out of the 38 hierarchically estimated credible intervals did not overlap with zero and fell in the expected direction. Third, a Bayes factor hypothesis test was used to quantify the evidence for the null hypothesis against a default one-sided alternative. Only seven out of 60 Bayes factors indicated non-anecdotal support in favour of the alternative hypothesis (), whereas 51 Bayes factors indicated at least some support for the null hypothesis. We hope that future analyses of replication success will embrace a more inclusive statistical approach by adopting a wider range of complementary techniques.

450TiP - Safety of lanreotide 120 mg ATG in combination with metformin in patients with advanced well-differentiated gastro-intestinal (GI) or lung carcinoids. A pilot, one-arm, open-label, prospective study: The MetNET-2 trial

In the CLARINET trial, lanreotide (LAN) 120 mg resulted in a 53% reduction in the risk of death or disease progression, compared with placebo, in GI-NET patients. Several studies have identified diabetic patients (pts) as having increased risk for the development of cancer and have associated metformin (MET) treatment with a decrease of cancer risk. MET may have anti-proliferative activity due to its ability to decrease insulin and IGF1 levels; in addition, it promotes AMPK activation and TSC1-2/mTOR inhibition. Our retrospective experience in a large group of pancreatic NET, has suggested that MET may provide additional clinical benefit in diabetic pts receiving everolimus and/or somatostatin analogues. This study evaluates the safety of LAN 120 mg in combination with MET in pts with advanced well-differentiated GI or lung carcinoids. Trial design: Methods: Pts with advanced GI or lung carcinoids will receive a combination of LAN 120 mg/month and MET 2550 mg/day, until progression or inacceptable toxicity. Radiological progression will be assessed every 4 months. The primary objective is to evaluate the incidence of adverse events (AEs) and severe AEs (SAEs). A 1-stage Hern design will be used. The null hypothesis that the SAEs rate related to the treatment is [25%] will be tested against a one-sided alternative. 20 pts with available tissue specimens will be enrolled. The null hypothesis will be rejected if ≤2 pts will experience a severe toxicity. This design yields a type I error rate of 10% and power of 85% when the true toxicity rate is ≥5%. The expression of 111 genes potentially involved in the pathways related to MET (e.g., LKB1, TP53, KRAS, IGF1R, VEGFR, PDGFR, AKT, PIK3CA, PTEN, mTOR) will be assessed by a target NGS approach. Other endpoints include TTP and RR. Results: In March 2016, the trial received institutional ethic board approval and in April 2016 the enrollment was started (EudraCT 2015-004626-34). Recruitment will be completed in April 2017. Conclusions: This study will investigate the safety of the combination of LAN 120 mg and MET and the correlation between tumor mutations and response to this therapy. Clinical trial identification: Protocol EUDRACT Number: 2015-004626-34 approved by Ethical committee of Fondazione IRCCS Istituto Tumori Milan on 8 March 2016 Legal entity responsible for the study: Fondazione IRCCS Istituto Tumori Milano

Robust testing for superiority between two regression curves

The problem of testing the null hypothesis that the regression functions of two populations are equal versus one-sided alternatives under a general nonparametric homoscedastic regression model is considered. To protect against atypical observations, the test statistic is based on the residuals obtained by using a robust estimate for the regression function under the null hypothesis. The asymptotic distribution of the test statistic is studied under the null hypothesis and under root−n local alternatives. A Monte Carlo study is performed to compare the finite sample behaviour of the proposed tests with the classical one obtained using local averages. A sensitivity analysis is carried on a real data set.

Lenalidomide Maintenance Significantly Improves Progression-Free Survival (PFS) in Patients with Chemosensitive Relapse of Diffuse Large B-Cell Lymphoma (DLBCL) Who Are Not Eligible for Autologous Stem Cell Transplantation (ASCT) or Experienced Relapse after Transplantation: Results of a Multicentre Phase II Trial

BACKGROUND: Patients (pts) with relapsed DLBCL who cannot be treated with consolidative ASCT or allogeneic transplantation exhibit a poor prognosis, with a 1-year PFS of 30-40%. Despite a high response rate to salvage therapy, these pts invariably experience early relapse and die of lymphoma. Single-drug maintenance may be a good alternative to intensified consolidation to prolong response duration in these high-risk pts. Lenalidomide is an oral immunomodulatory agent, active against DLBCL, which can be taken for years with an excellent safety profile. Accordingly, we designed a multicenter phase II trial addressing safety and efficacy of lenalidomide maintenance in pts with chemosensitive relapse of DLBCL not eligible for consolidative ASCT or experiencing relapse after ASCT (NCT00799513). Herein, we report the primary endpoint analysis.METHODS: Selection criteria were: 1) adult HIV-negative pts; 2) histologically-proven de novo or transformed DLBCL; 3) relapsed disease responsive (partial or complete response) to conventional-dose rituximab-containing salvage therapy; 4) ECOG PS ≤3; 5) time to progression (TTP) from the previous line ≥3 months. After confirmation of objective response to salvage therapy, pts were registered and treated with lenalidomide 25 mg/day once daily for 21 days out of 28, for two years or until lymphoma failure or unacceptable toxicity. Primary endpoint was the 1-yr PFS. Simon's two-stage optimal design was used. The null hypothesis that the true 1-yr PFS is 30% was tested against a one-sided alternative. The trial design yields a type I error rate of 5% and power of 80% when the true 1-year PFS is 50%. To demonstrate this PFS improvement, 47 pts were needed. The null hypothesis would be rejected if 19 or more pts progression-free at one year were observed.RESULTS: 41 pts were enrolled (median age 72, range 34-86; M:F ratio: 1.5). Thirty pts had a de novo DLBCL, 11 had a transformed DLBCL; 29 pts were enrolled after the first relapse, 12 after the 2nd - 4th relapse. All pts were previously treated with anthracycline- and rituximab-based combination, plus ASCT in 6 pts. The median TTP after the previous line was 17 months (range 3-121). Most pts had unfavourable features: IPI ≥2 in 34 (83%) pts, advanced disease in 33 (80%), extranodal disease in 29 (71%), high LDH in 18 (44%). Twelve pts had HCV and/or HBV infection. Salvage combination included high-dose-cytarabine in 23 pts, high-dose-ifosfamide in 6, anthracycline in 6 and bendamustine in 6. Response to salvage therapy was complete in 25 pts and partial in 16.Twenty-three pts received the planned maintenance; lenalidomide was interrupted due to lymphoma relapse in 8 pts, toxicity in 5 (diarrhoea in 2, rash, prolonged neutropenia, intestinal infarction), and patient's refusal in 5 (diarrhoea in 4, rash). Dose reduction to 10 or 15 mg/d, mostly due to rash or neutropenia, was indicated in 17 pts. Toxicity was mild; there were 6 SAEs (febrile neutropenia in 4, diarrhoea, intestinal infarction) in 5 pts. Grade 4 haematological toxicity consisted of neutropenia in 17 pts and thrombocytopenia in 2. Grade 3-4 non-haematological toxicity consisted of diarrhoea in 3 pts and rash in 5. HBV/HCV seropositivity was not associated with higher toxicity.Six of the 16 pts in partial response after salvage therapy achieved a complete remission during lenalidomide maintenance. At a median follow-up of 16 months, 29 pts remained relapse free, with a 1-year PFS of 75±8%. Importantly, 21 pts were progression-free at one year, with the early achievement of the primary endpoint. Thirty-two pts are alive (NED in 25), 8 pts died of lymphoma and one of intestinal infarction, with a 1-yr OS of 84±7%. Age ≤70 years, normal LDH level, and complete response at registration were independently associated with better PFS and OS, whereas gender, DLBCL category, HBV/HCV infection, and TTP after previous line (< vs. ≥12 months) were not. Assessment of prognostic effect of ontogenic stratification by NanoString is ongoing.CONCLUSIONS: With the early achievement of the primary endpoint, this is the first prospective trial showing a positive effect of maintenance in pts with relapsed DLBCL. Lenalidomide maintenance is feasible and well tolerated in this elderly population, but diarrhoea and rash remain frequent dose-limiting side effects. The evident improvement in survival figures warrants further investigation of immunomodulators maintenance in these high-risk pts. DisclosuresFerreri:celgene: Consultancy, Research Funding. Rusconi:Roche: Honoraria.

Is the locally best invariant test uniformly most powerful for a wider class of invariant tests?

In the context of a general regression model in which some regression coefficients are of interest and others are purely nuisance parameters, Bhowmik and King (5) constructed the locally best invariant (LBI) test against one-sided alternatives. This paper investigates whether this LBI test is uniformly most powerful invariant (UMPI) or not through simulation results. A test that is locally best invariant against one-sided alternative hypotheses is found to be uniformly most powerful invariant (UMPI) in a wider class of tests than the invariant tests for the standard F test. The results of a simulation study conducted to prove that the LBI test is UMPI are presented. Through simulation results this study proves that the LBI test is UMPI.