Complementary Log-log Transformation Research Articles

A phase 2 basket study of tucatinib and trastuzumab in previously treated solid tumors with HER2 alterations: Urothelial cancer cohort (SGNTUC-019).

TPS587 Background: Tucatinib, a highly selective human epidermal growth factor receptor 2 (HER2)-directed tyrosine kinase inhibitor approved in multiple regions for HER2-positive (HER2+) metastatic breast cancer, is being investigated as a novel therapy for metastatic colorectal cancer, gastric cancer, and other solid tumors. In xenograft models of HER2+ and HER2-mutated tumors, dual targeting of HER2 with tucatinib plus trastuzumab showed superior activity to either alone (Kulukian 2020; Peterson 2020). Despite development of several new therapies for metastatic urothelial cancer (UC), most patients are refractory to subsequent therapies and die from the disease, highlighting the need for additional therapeutic approaches. Given that 20%–30% of metastatic UCs have molecular alterations of the ErbB family, tucatinib plus trastuzumab warrants further evaluation. SGNTUC-019 (NCT04579380) is a multi-cohort, open-label, phase 2 study evaluating tucatinib plus trastuzumab in patients with previously treated solid tumors displaying HER2+ or HER2-mutated solid tumors, including a cohort with locally advanced or metastatic UC. Methods: Eligible patients in the UC cohort must have HER2+ locally advanced or metastatic disease, with progression during or after, or intolerance of, the most recent line of systemic therapy. Patients must have an ECOG PS ≤1; adequate hepatic, hematologic, renal, and cardiac function; and no prior exposure to HER2-directed therapy. HER2 alterations can be demonstrated by HER2 overexpression/amplification in tumor tissue by prior IHC/ISH or by HER2 amplification in prior or prescreening NGS assay of ctDNA or prior tissue NGS assay. The HER2 overexpression/amplification UC cohort will enroll 12 response-evaluable patients per RECIST 1.1. If ≥2 responses are observed, the cohort will be expanded to 30. The primary objective is antitumor activity, with confirmed ORR as the primary endpoint and DCR, duration of response, PFS, and OS as secondary efficacy endpoints. Confidence intervals (CIs) will be calculated for confirmed ORR and DCR. For time-to-event endpoints, median survival time will be estimated by the Kaplan-Meier method, and the associated 95% CI will be calculated using complementary log-log transformation. Safety endpoints will be assessed using descriptive statistics. Patients will receive tucatinib 300 mg orally twice daily and trastuzumab 8 mg/kg intravenously on Cycle 1 Day 1 and 6 mg/kg every 21 days from Cycle 2 Day 1. Disease assessments per RECIST 1.1 will occur every 6 weeks for 24 weeks, then every 12 weeks. Trough concentrations of tucatinib will be evaluated in Cycles 2–6, with peak concentration sampled in Cycle 3. Quality of life will be evaluated every second cycle using the EQ-5D-5L. Sites are currently enrolling within the US, EU, and Asia Pacific. Clinical trial information: NCT04579380 .

Analysis of Patient-Level Data from 3 Cooperative Group Trials Confirms a Survival Advantage for NPM1m Patients Achieving MRD-Negative CR after Intensive Induction

Background:NPM1 mutations (NPM1m), predominantly tetra-nucleotide insertions in exon 12, are clonal driver events in acute myeloid leukemia (AML) and are found exclusively in leukemic cells, making them ideal for sensitive detection of measurable residual disease (MRD) using quantitative polymerase chain reaction (qPCR)- or next generation sequencing (NGS)-based methods. Over 50 peer-reviewed publications across multiple cooperative group trials have shown that NPM1m AML patients in complete remission (CR), CR with incomplete count recovery (CRi), or CR with partial hematologic recovery (CRh) have significantly better survival if they are MRD negative compared to MRD positive. Based on this body of evidence, the European LeukemiaNet recommends monitoring molecular MRD in NPM1m AML patients during treatment to help guide treatment decisions. Purpose: To provide further evidence for the value of NPM1 MRD as a surrogate endpoint for prospective, randomized trials in patients receiving novel agents in combination with intensive chemotherapy by evaluating the relationship between MRD negativity in patients with CR after 2 cycles of chemotherapy and event-free survival (EFS) and overall survival (OS) in a pooled analysis of patient-level data from 3 large cooperative group trials. MRD in all 3 studies was assessed using reverse transcriptase-mediated qPCR (RT-qPCR) of NPM1m transcripts normalized to ABL transcripts. Methods: Deidentified patient data for a total of 1,128 patients who achieved CR, CRi, or CRh, and who had MRD data after 2 cycles of chemotherapy, were provided by the UK National Cancer Research Institute AML Study Group (SG) for the AML17 (N = 539), the AMLSG for the AMLSG 09-09 (N = 497), and the Study Alliance Leukemia (SAL) for the AML2003 (N = 105) trials and standardized using Standard Data Tabulation Models. Data included demographic information, disease history, induction and consolidation treatment received, morphologic response, MRD data after 2 cycles of treatment, relapse, and survival. Cutoffs for MRD negativity were defined according to the definitions used in the individual studies (Ivey A, et al. N Engl J Med. 2016; Kapp-Schwoerer S, et al. Blood. 2020; Shayegi N, et al. Blood. 2013). In addition, sensitivity analyses were conducted using a range of cutoff values. CR was defined as <5% blasts in the bone marrow (BM) with count recovery, occurring within 42 days of the start of the last induction cycle; only these patients were included in the MRD analysis. EFS and OS were estimated using the Kaplan-Meier method and defined according to FDA criteria. Briefly, EFS was measured from day 1 of randomization to the date of treatment failure, hematologic relapse from CR or death from any cause; treatment failure was defined as not achieving CR after two cycles of chemotherapy. OS was measured from day 1 to the date of death from any cause. The absolute risk difference among MRD response status groups in EFS/OS at the 2-month, 1-year, 2-year, 3-year, and 5-year landmarks and their corresponding 95% confidence intervals (CI) were calculated using complementary log-log transformation. Hazard ratios (HR) were fitted using a Cox regression model with EFS/OS as the dependent variable and with MRD-response status and study as independent variables. Results: Among the 515 patients who achieved a CR within 42 days at the start of chemotherapy Cycle 2, MRD negativity confers a substantial survival advantage with EFS HR (95% CI) = 1.6 (1.1-2.3) when measured in BM and 1.8 (1.4-2.4) in peripheral blood (PB) using the study-defined cutoffs for positive and negative. This advantage is also observed using cutoff definitions of MRD negativity ranging from <10 to <1,000 NPM1m/104ABL transcripts. A survival advantage for OS was also observed with HR (95% CI) = 1.5 (1.0-2.4) and 1.7 (1.2-2.3) in BM and PB, respectively. Analysis of the predictive value of MRD negativity in the 613 patients who achieved CRi/CRh within 42 days of the start of Cycle 2 is underway. Conclusions: Analysis of pooled patient-level data confirms the survival advantage of achieving MRD-negative CR as previously reported in 3 independent, cooperative group trials using different induction regimens. The prognostic value of MRD negativity in NPM1m AML is consistent across a broad range of cutoff definitions in both BM and PB providing further evidence for the value of NPM1 MRD as a surrogate endpoint for clinical trials.

Incidence-severity relationships in non-treated and fungicide-treated wheat head blast epidemics in Brazil

The relationship between wheat head blast incidence (I; proportion of diseased heads in a sample) and severity (S; the average diseased area in a sample of heads) was studied with the aim of determining whether severity could be reliably predicted from incidence. Data were collected from two main sources: a) the EPAMIG (non-treated) field experiments spanning seven years (2013–2019 with three cultivars) designed to evaluate the effect of planting dates on wheat blast (no fungicide use); and b) the UFTs (fungicide-treated) field experiments conducted by researchers of the network of uniform fungicide trials conducted during nine years (2012–2020) across 10 locations (totaling 14 cultivars). An ordinary linear regression model fitted to the complementary log-log transformation of the data provided a good fit to the data based on the square correlation of the predicted Ŝ, and the observed S (R2): The R2 was 0.95 for the pooled (all years) non-treated dataset and 0.88 for the pooled fungicide-treated dataset. A covariance analysis indicated that year, but not cultivar, significantly affected the parameters of the I-S relationship in the non-treated dataset. We conclude that wheat head blast S can be predicted from I but more reliably in non-treated epidemics given the larger influence of chemical treatments on the relationship than cultivar or year, a surrogate for varying weather.

A note on confidence intervals for the restricted mean survival time based on transformations in small sample size.

Restricted mean survival time (RMST) is one measure now used to summarize time-to-event type data, but it has been pointed out that the distribution of differences in RMST deviates markedly from a normal distribution for controlled clinical trials with small sample sizes. Therefore, we conducted a numerical simulation of the RMST in which the one-sample survival time follows a Weibull distribution, comparing eight different confidence intervals combining two types of variance with four types of variable transformations, including no transformation. The evaluation items were the coverage probability and the above and below error probabilities for the true value. The variance types were based on Greenwood's formula and its Kaplan-Meier correction. The arcsine square root transformation, logit transformation, and complementary log-log transformation were used as the variable transformations. When the sample size was small and the event rate was low, the confidence interval of the untransformed RMST tended to have a small coverage probability and to be overestimated. Variance by Kaplan-Meier correction improved the coverage. The problems of coverage and overestimation were also improved by variable transformations, and in particular, applying the logit transformation and the complementary log-log transformation both resulted in substantial improvements. Our study suggested that it is preferable to construct the confidence intervals of RMST using the logit transformation for variances based on Greenwood's formula in small sample size trials. The SAS code to replicate the analyses is available at https://github.com/HiroyaHashimoto/SAS-Programs.

A change of heart: Preliminary results of the US 2018 adult heart allocation revision.

In 2018, the Organ Procurement and Transplantation Network (OPTN) modified adult heart allocation to better stratify candidates and provide broader access to the most medically urgent candidates. We analyzed OPTN data that included waiting list and transplant characteristics, geographical distribution, and early outcomes 1year before (pre: October 18, 2017-October 17, 2018) and following (post: October 18, 2018-October 17, 2019) implementation. The number of adult heart transplants increased from 2954 pre- to 3032 postimplementation. Seventy-eight percent of transplants in the post era were for the most medically urgent (statuses 1-3) compared to 68% for status 1A in the pre era. The median distance between the donor hospital and transplant center increased from 83 to 216 nautical miles, with an increase in total ischemic time from 3 to 3.4hours (all P<.001). Waiting list mortality was not different across eras (14.8 vs 14.9 deaths per 100 patient-years pre vs post respectively). Posttransplant patient survival was not different, 93.6% pre and 92.8% post. There is early evidence that the heart allocation policy has enhanced stratification of candidates by their medical urgency and broader distribution for the most medically urgent candidates with minimal impact on overall waiting list mortality and posttransplant outcomes.

Prostate cancer: net survival and cause-specific survival rates after multiple imputation.

BackgroundEstimations of survival rates are diverse and the choice of the appropriate method depends on the context. Given the increasing interest in multiple imputation methods, we explored the interest of a multiple imputation approach in the estimation of cause-specific survival, when a subset of causes of death was observed.MethodsBy using European Randomized Study of Screening for Prostate Cancer (ERSPC), 20 multiply imputed datasets were created and analyzed with a Multivariate Imputation by Chained Equation (MICE) algorithm. Then, cause-specific survival was estimated on each dataset with two methods: Kaplan-Meier and competing risks. The two pooled cause-specific survival and confidence intervals were obtained using Rubin’s rules after complementary log-log transformation. Net survival was estimated using Pohar-Perme’s estimator and was compared to pooled cause-specific survival. Finally, a sensitivity analysis was performed to test the robustness of our constructed multiple imputation model.ResultsCause-specific survival performed better than net survival, since this latter exceeded 100 % for almost the first 2 years of follow-up and after 9 years whereas the cause-specific survival decreased slowly and than stabilized at around 94 % at 9 years. Sensibility study results were satisfactory.ConclusionsOn our basis of prostate cancer data, the results obtained by cause-specific survival after multiple imputation appeared to be better and more realistic than those obtained using net survival.Electronic supplementary materialThe online version of this article (doi:10.1186/s12874-015-0048-4) contains supplementary material, which is available to authorized users.

Analysis of Incidence-Severity Relationships for Strawberry Powdery Mildew as Influenced by Cultivar, Cultivar Type, and Production Systems.

The relationships between strawberry powdery mildew incidence (I) and severity (S) were investigated for various cultivars, for June-bearing and day-neutral cultivars, and for production systems (open-field and plastic-tunnel) with the objective of deriving a simple relationship for predicting severity (proportion of leaf area diseased [PLAD]) from incidence (proportion of diseased leaves). Data were collected from 2006 to 2011 at 11 commercial and experimental sites, for a total of 2,326 observations (n). For the cultivars grown in open fields, higher severity was observed on 'Seascape', with mean PLAD of 0.299 (n = 427); followed by 'Chambly', with 0.133 (n = 334); 'Cavendish', with 0.115 (n = 250); 'Darselect', with 0.111 (n = 321); and 'Jewel', with 0.105 (n = 276). In general, mean severity was higher when the strawberry plants were grown in plastic tunnels, with PLAD of 0.204, 0.199, and 0.181 for Chambly (n = 204), Darselect (n = 261), and Jewel (n = 253), respectively. A linear model based on complementary log-log transformation of I and S provided a good fit of the data (coefficient of determination [R2] adjusted for degrees of freedom from 0.82 to 0.96). A covariance analysis indicated that the sampling year and site of sampling did not significantly influence the estimated slope of the I-S relationship, nor did the specific cultivar among the June-bearing ones, whereas the production system (open-field versus plastic-tunnel) and the cultivar type (June-bearing versus day-neutral) significantly influenced the estimated slope. From this analysis, we were able to develop three specific models for open-field-grown June-bearing cultivars (R2 = 0.90), for the open-field-grown day-neutral cultivar (Seascape, R2= 0.91), and for June-bearing cultivars grown in plastic tunnels (R2 = 0.92). From these results, it was concluded that strawberry powdery mildew leaf severity can be accurately estimated from incidence of diseased leaves. The I-S relationships developed in the present study may be used in making practical disease management decisions, especially for management programs that use information on disease level in the field to initiate fungicide spraying programs or to time the interval between sprays.

Nascer em Hospital Amigo da Criança no Rio de Janeiro, Brasil: um fator de proteção ao aleitamento materno?

This article aims to investigate the impact of the Baby-Friendly Hospital Initiative (BFHI) on exclusive breastfeeding (EBF). This was a cross-sectional study with 811 mothers of infants under five months of age, randomly selected at five health centers in Rio de Janeiro, Brazil. The exposure variable was hospital of birth, categorized in accredited hospitals (BFH), certified hospitals (CBFH), and hospitals without accreditation. The data were analyzed by complementary log-log transformation models, which capture cohort longitudinal experience (current status data). Even after adjusting the analysis for maternal socio-demographic, lifestyle, and psychological factors, health services use, and infants' age and health status, duration of EBF was longer in infants born in BFH and CBFH. The findings suggest the effectiveness of BFHI in maintaining EBF throughout the early months of life. To extend EBF through the first six months of life would require not only strengthening the BFHI but also developing and encouraging more actions in favor of breastfeeding, focusing on primary healthcare facilities.

Development and comparison of two models to predict survival rates of young larvae of Stegobium paniceum (L.) (Coleoptera: Anobiidae) under heat treated temperatures

Predicting mortality or survival rate of insects at heat treatment temperatures is critical for commercial heat treatment of food and storage facilities. Two models were developed to predict the survival rate of young larvae of drugstore beetle, Stegobium paniceum (L.) (Coleoptera: Anobiidae) under steady state or transient temperature condition: modified fundamental kinetic model and modified complementary log-log transformation model. Published young larvae mortalities and raw data of the temperature history, determined at different heat treatment temperatures, were used to develop these models. The models were verified and compared by using the field data collected in a medium sized mill. Both of the developed models overestimated the insect mortality in the mill when temperature was > 53°C and underestimated insect mortality when temperature was < 50°C. The lowest mean of the absolute difference between the predicted and measured insect mortality in the mill was 16.7±1.1% which was generated by using the modified complementary log-log transformation model. The possible reason of this divergence from the model is discussed. Keywords : Stegobium paniceum (L.), Survival, Heat treatments, Modeling.

Relationships Between Incidence and Severity of Fusarium Head Blight on Winter Wheat in Ohio

ABSTRACT To determine the relationship between incidence (I; proportion of diseased spikes) and severity (S; mean proportion of diseased spikelets per spike) for Fusarium head blight of wheat and to determine if severity could be predicted reliably from incidence data, disease assessments were made visually at multiple sample sites in artificially and naturally inoculated research and production fields between 1999 and 2002. Ten distinct data sets were collected. Mean disease intensity ranged from 0.023 to 0.975 for incidence and from 0.0003 to 0.808 for severity. A model based on complementary log-log transformation of incidence and severity performed well for all data sets, based on calculated coefficients of determination and random residual plots. The I-S relationship was consistent among years and locations, with similar slopes for all data sets. For 7 of the 10 data sets and for the pooled data from all locations and years, the estimated slope from the fit of the model ranged from 1.03 to 1.26. Time of disease assessment affected the relationship between incidence and severity; however, the estimated slopes from each assessment time were also close to 1. Based on the width of the 95% prediction interval, severity was estimated more precisely at lower incidence values than at higher values. The number of sampling units and the index of dispersion of disease incidence had only minor effects on the precision with which S was predicted from I. The estimation of mean S from I would substantially reduce the time required to assess Fusarium head blight in field surveys and treatment comparisons, and the observed relationship between I and S could be used to identify genotypes with some types of disease resistance.

The theoretical basis and practical application of relationships between different disease intensity measurements in plants

SummaryThe literature concerning relationships among different measures of plant disease intensity is reviewed. Some previous confusion over the definition of the terms “incidence” and “severity” is noted and clarified. The review highlights the common features of relationships between incidence and severity, incidence and disease density, and incidence at pairs of scales in a spatial hierarchy. These relationships often show a similar saturation curve form that can frequently be described empirically using the complementary log‐log transformation. A catalogue of alternative functional forms is provided. Practical applications in varietal evaluation are discussed.

Power-Transformation for Ordered Categorical Data

We consider in this paper a model of asymmetric power-transformation of response probability explained by linear function of some covariates. This model includes logistic and complementary log-log transformation models as its specific case, so we can use the model to evaluate the appropriateness or the goodness of fit of these models. Then the performances of the asymmetric power-transformation model are evaluated and examined, based on data used in published literatures. And we discuss various issues of diagnoses which may occur in the process of applying the model. Further we consider alternative transformations, and then in comparison with them, we point out the advantages of the asymmetric power-transformation.

A Survival Model for a Wood Preservative Trial

Summary A trial is considered where the progress of individuals is recorded at relatively long intervals. By assuming the proportional hazards model it is shown that the conditional probability offailure or death in an interval, given survival up to that interval, can be described using a generalised linear model. That is, the complementary log-log transformation of this conditional probability is assumed to be a linear function of parameters for each treatment and parameters for each time interval. The model is fitted to a set of data from a wood preservative trial and a survival distribution is obtained that leads to estimates of mean lifetime and lifetime quantiles.

Models and designs for agricultural vaccine trials

AbstractAgricultural vaccine trials are costly, and it is important that they should provide the maximum amount of information. The traditional probit model often fails to provide a reliable estimate of median protective dilution. The complementary loglog transformation is to be preferred because it provides a probabilistic basis for fixing one of the parameters, thereby increasing the precision of estimation.