Performance of robust regression methods in real-time polymerase chain reaction calibration.

The ordinary least squares (OLS) method is routinely used to estimate the unknown concentration of nucleic acids in a given solution by means of calibration. However, when outliers are present it could appear sensible to resort to robust regression methods. We analyzed data from an External Quality Control program concerning quantitative real-time PCR and we found that 24 laboratories out of 40 presented outliers, which occurred most frequently at the lowest concentrations. In this article we investigated and compared the performance of the OLS method, the least absolute deviation (LAD) method, and the biweight MM-estimator in real-time PCR calibration via a Monte Carlo simulation. Outliers were introduced by replacement contamination. When contamination was absent the coverages of OLS and MM-estimator intervals were acceptable and their widths small, whereas LAD intervals had acceptable coverages at the expense of higher widths. In the presence of contamination we observed a trade-off between width and coverage: the OLS performance got worse, the MM-estimator intervals widths remained short (but this was associated with a reduction in coverages), while LAD intervals widths were constantly larger with acceptable coverages at the nominal level.

An Italian program of External Quality Control for chromogranin A (CgA) assay: performance evaluation of CgA determination.

BACKGROUND: Chromogranin A (CgA) is an acidic glycoprotein produced by many neuroendocrine cells and neurons. Currently, two different methods for assaying CgA, immunoradiometric assay (IRMA) and enzyme-linked immunosorbent assay (ELISA), are widely used in routine practice. Within the framework of a Ministry of Health project, an External Quality Control program was developed to investigate the state of the art of CgA determination in Italy and to monitor the performance of laboratories carrying out this assay. This paper reports the results regarding laboratory performance. METHODS: A total of 43 laboratories participated in this program, in which 21 used the ELISA method and 22 the IRMA method. Each laboratory received six samples, three aliquots of serum and three of plasma, at high, intermediate and low concentrations. The results provided by the two assay methods were analyzed separately using two statistical approaches, the principal component analysis and the control chart method. RESULTS: For the IRMA method, questionable results for all samples were obtained by two laboratories, while in two other laboratories performance was questionable for only one sample. For the ELISA method, questionable performances were obtained in only one laboratory for the low and intermediate concentration samples, whereas in three laboratories performance was questionable for only one sample. Interestingly, the coefficients of variation increased approximately five-fold when shifting from the IRMA to the ELISA method. CONCLUSIONS: This program demonstrated both the requirement and demand for external quality assessment of CgA assay.

[Statistical notes. Statistical significance and clinical relevance: are they the two sides of the same medal?]. / Note statistiche. Significatività statistica e rilevanza clinica: due facce della stessa medaglia?

The aim of this statistical note is to draw the attention of the cardiologists to the aspects pertinent to the clinical relevance of the result of a clinical controlled randomized trial. The difference between clinical relevance and statistical significance is shown by using the results of GISSI and GUSTO III clinical controlled randomized trials.

EQUAL-qual: a European program for external quality assessment of genomic DNA extraction and PCR amplification.

BACKGROUND: Despite the rapid transition into routine clinical practice of molecular techniques based on PCR, external quality assessment (EQA) is still not widely available. The European Union and European Communities Confederation of Clinical Chemistry have supported the EQUAL project as a series of 3 different EQA programs for the assessment of molecular methods independently from analytes. We present the results from the EQUAL-qual program designed to evaluate the analytical aspects of DNA analysis by means of a conventional qualitative PCR experiment. METHODS: The EQUAL-qual program provided DNA, blood samples, and primer sets to participant laboratories to assess DNA extraction and PCR amplification. We have developed statistical procedures to identify laboratories performing poorly in DNA extraction (quality and quantity), PCR efficiency, and data interpretation after electrophoresis. RESULTS: An application to participate was obtained from 213 laboratories (from 25 countries), and 175 (82%) of laboratories submitted results for assessment. Questionable results in terms of quality and/or quantity of DNA derived from blood extractions were returned by 27% of laboratories (46 of 166). PCR efficiency showed high variability, with 3% of laboratories (5 of 163) showing a consistently low rate of amplification and 10% (18 of 175) not reporting the expected number of bands of the amplified targets. CONCLUSIONS: The results showed considerable variability in all phases of the experiment. The approach confirms the validity of EQA as a method for evaluating analytical aspects of PCR-based tests.

[What does "p" mean at conclusion of a test of hypothesis in a randomized controlled clinical trial of superiority?]. / Note statistiche. Cosa significa "p"a conclusione di un test d'ipotesi in una sperimentazione clinica controllata di superiorità?

The aim of this statistical note, the sixth in the series, is to introduce the rationale of the test of hypothesis suitable for comparing the effect of two treatments in a randomized controlled clinical trial of superiority. The presentation takes advantage of the analogy with a criminal trial debate based upon circumstantial evidence in an Italian Court. The results of three randomized controlled clinical trials: ISIS-1, AIMS and RESTORE are introduced and proper ways for their interpretation are suggested.

Axillary lymph node nanometastases are prognostic factors for disease-free survival and metastatic relapse in breast cancer patients.

PURPOSE: Early breast cancer presents with a remarkable heterogeneity of outcomes. Undetected, microscopic lymph node tumor deposits may account for a significant fraction of this prognostic diversity. Thus, we systematically evaluated the presence of lymph node tumor cell deposits

EQUAL-quant: an international external quality assessment scheme for real-time PCR.

BACKGROUND: Quantitative gene expression analysis by real-time PCR is important in several diagnostic areas, such as the detection of minimum residual disease in leukemia and the prognostic assessment of cancer patients. To address quality assurance in this technically challenging area, the European Union (EU) has funded the EQUAL project to develop methodologic external quality assessment (EQA) relevant to diagnostic and research laboratories among the EU member states. We report here the results of the EQUAL-quant program, which assesses standards in the use of TaqMan probes, one of the most widely used assays in the implementation of real-time PCR. METHODS: The EQUAL-quant reagent set was developed to assess the technical execution of a standard TaqMan assay, including RNA extraction, reverse transcription, and real-time PCR quantification of target DNA copy number. RESULTS: The multidisciplinary EQA scheme included 137 participating laboratories from 29 countries. We demonstrated significant differences in performance among laboratories, with 20% of laboratories reporting at least one result lacking in precision and/or accuracy according to the statistical procedures described. No differences in performance were observed for the >10 different testing platforms used by the study participants. CONCLUSIONS: This EQA scheme demonstrated both the requirement and demand for external assessment of technical standards in real-time PCR. The reagent design and the statistical tools developed within this project will provide a benchmark for defining acceptable working standards in this emerging technology.

[Statistical notes. Which risks do the patient incur if his own cardiologist accepts to be involved in a clinical trial without deep consideration of the a priori available evidence?]. / Note statistiche. A quali rischi espone i propri pazienti il cardiologo che accetta di partecipare ad una sperimentazione clinica senza un'approfondita riflessione sulle evidenze disponibili a priori?

The aim of this statistical note is to describe the results of the randomized controlled clinical trial TARGET, which compared the effect of tirofiban (new treatment) and abciximab (standard treatment) in patients who were expected to undergo coronary stenting. Primary aim of TARGET was to evaluate the non-inferiority of tirofiban with respect to abciximab, but it concluded in favor of superiority of the standard treatment. The authors of this study point out that a deep consideration regarding a priori available evidence could have avoided to expose 2398 patients randomized to tirofiban to the risk of death or non-fatal myocardial infarction.

Artificial neural network for the joint modelling of discrete cause-specific hazards.

OBJECTIVE: Artificial neural network (ANN) based regression methods have been introduced for modelling censored survival data to account for complex prognostic patterns. In the framework of ANN extensions of generalized linear models for survival data, PLANN is a partial logistic ANN, suitable for smoothed discrete hazard estimation as a function of time and covariates. An extension of PLANN for competing risks analysis (PLANNCR) is now proposed for discrete or grouped survival times, resorting to the multinomial likelihood. METHODS AND MATERIALS: PLANNCR is built by assigning input nodes to the explanatory variables with the time interval treated as an ordinal variable. The logistic function is used as activation for the hidden nodes of the network, whereas the softmax, which corresponds to the canonical link of generalized linear models for polytomous regression, is adopted for multiple output nodes, to provide a smoothed estimation of discrete conditional event probabilities for each event. The Kullback-Leibler distance is used as error function for the target vectors, amounting to half of the deviance of a multinomial logistic regression model. PLANNCR can jointly model non-linear, non-proportional and non-additive effects on cause-specific hazards (CSHs). The degree of smoothing is modulated by the number of hidden nodes and penalization of the error function (weight decay). Model optimisation is achieved by quasi-Newton algorithms, while non-linear cross-validation (NCV) and the Network Information Criterion (NIC) were adopted for model selection. PLANNCR was applied to data on 1793 women with primary invasive breast cancer, histologically N-, who underwent surgery at the Milan Cancer Institute between 1981 and 1986. RESULTS: Differential effects of covariates and time on the shape of the CSH for the three main failure causes, namely intra-breast tumor recurrences, distant metastases and contralateral breast cancer, have been enlightened. CONCLUSIONS: PLANNCR can be suitably adopted in an exploratory framework for a thorough evaluation of the disease dynamics in the presence of competing risks.

Molecular subtyping of breast cancer from traditional tumor marker profiles using parallel clustering methods.

PURPOSE: Recent small-sized genomic studies on the identification of breast cancer bioprofiles have led to profoundly dishomogenous results. Thus, we sought to identify distinct tumor profiles with possible clinical relevance based on clusters of immunohistochemical molecular markers measured on a large, single institution, case series. EXPERIMENTAL DESIGN: Tumor biological profiles were explored on 633 archival tissue samples analyzed by immunohistochemistry. Five validated markers were considered, i.e., estrogen receptors (ER), progesterone receptors (PR), Ki-67/MIB1 as a proliferation marker, HER2/NEU, and p53 in their original scale of measurement. The results obtained were analyzed by three different clustering algorithms. Four different indices were then used to select the different profiles (number of clusters). RESULTS: The best classification was obtained creating four clusters. Notably, three clusters were identified according to low, intermediate, and high ER/PR levels. A further subdivision in two biologically distinct subtypes was determined by the presence/absence of HER2/NEU and of p53. As expected, the cluster with high ER/PR levels was characterized by a much better prognosis and response to hormone therapy compared to that with the lowest ER/PR values. Notably, the cluster characterized by high HER2/NEU levels showed intermediate prognosis, but a rather poor response to hormone therapy. CONCLUSIONS: Our results show the possibility of profiling breast cancers by means of traditional markers, and have novel clinical implications on the definition of the prognosis of cancer patients. These findings support the existence of a tumor subtype that responds poorly to hormone therapy, characterized by HER2/NEU overexpression.

Bivariate statistical approach to evaluate laboratory performance by analysis of standard curves in an External Quality Assurance program for quantitative assays based on real-time PCR with Taq-Man probes.

Recently a revolutionary technique for quantitative PCR determination was introduced in diagnostic laboratories. To determine the influence of technical variability on the reliability of the quantitative assay, it is crucial to use External Quality Assurance (EQA) programs. An EQA program was developed in Italy to check the analytical performance of real-time PCR procedures based on Taq-Mantrade mark probes. This article suggests a new statistical approach to discriminate, using a bivariate technique, laboratory performance that appears to be questionable, by separately considering the two main features of the standard curve: analytical sensitivity and efficiency. Furthermore, specific indexes to evaluate the impact of these two features on the determination of the initial number of molecules are given to help to improve the assay procedure.

[Statistical notes. Randomized controlled clinical trials of both superiority and non-inferiority: critical considerations]. / Note statistiche. Sperimentazioni cliniche controllate randomizzate sia di superiorità che di non inferiorità: considerazioni critiche.

The aim of this statistical note, the fourth in the series, is to describe and critically appraise the randomized controlled clinical trial (RCCT) GUSTO V, which combined in a single RCCT both the superiority and non-inferiority hypotheses. In this note we present the logical path that the authors have presumably followed in planning a RCCT of such a kind. The results are reported and possible critical aspects are highlighted and debated upon. Finally the cardiologist reader is stimulated to give his own opinion on this kind of approach.

An Italian program of external quality control for quantitative assays based on real-time PCR with Taq-Man probes.

Quantitative real-time PCR techniques are increasingly being used for the measurement of nucleic acids in research applications as well as in the clinical laboratory. It is therefore important that external quality control programs (EQA) are implemented for the evaluation of the analytical aspects common to molecular tests based on quantitative PCR. The aim of this study was the development of an Italian program of external quality control for quantitative assays based on real-time PCR with Taq-Mantrade mark probes to compare the analytical performance of 42 laboratories. Participants were provided with a set of reagents (cDNA for reference curve preparation, primers-probe mix and three unknown samples) and requested to perform a conventional assay using the master mix employed in their laboratories. The quantitative results in unknown samples were analyzed. The results of our study showed clear heterogeneity in performance. Two of the 42 laboratories provided results indicating contamination during the experiment, whereas six did not provide values for at least one of the six standard points. Only 12 laboratories gave results that were both precise and accurate for all the samples tested. Regarding imprecision, 17 laboratories appeared to deviate in at least one result, whereas inaccuracy showed an inverse dose-dependent trend. Finally, 12 laboratories were not able to measure the sample with the lowest concentration. Ten of these laboratories were equipped with the same instruments. The results of this first round of analytical EQA of real-time PCR-based methods seem to indicate high variability among laboratories carrying out the same experimental protocol. These findings could have implications for any assay based on this type of technique. This survey demonstrates the importance of experimental EQAs of methodological proficiency testing. Our approach has proved useful for comparing the analytical aspects shared by all diagnostic laboratories applying quantitative assays for the measurement of nucleic acids based on the use of Taq-Mantrade mark probes and real-time platforms.

[Statistical notes. Hazard curves of sudden death: their role in the cardiologist's clinical decision]. / Note statistiche. Curve di rischio istantaneo di morte: il loro ruolo nella decisione clinica del cardiologo.

In long-term studies researchers are mainly concerned with occurrence of death during the follow-up period. This statistical note is focused on survival analysis which is the main tool to process this kind of data. Survival curve, cumulative mortality curve and hazard curve are here introduced together with an appropriate effect indicator: hazard ratio. In particular, the use of the latter is shown by resorting to the randomized controlled clinical trial TARGET.

[Statistical notes. From superiority to non-inferiority clinical trials: a leap in the dark?]. / Note statistiche. Dalle sperimentazioni cliniche di superiorità alle sperimentazioni cliniche di non inferiorità: un salto nel vuoto?

In these "statistical notes", equivalence and non-inferiority randomized controlled clinical trials (RCCT) are considered. Equivalence trials are designed to confirm the absence of a meaningful difference between the effect of two treatments. Non-inferiority trials are designed to prove that the new treatment is no less effective than an existing one: it may be more effective or it may have a similar effect. In this note the attention is addressed to suitable criteria for the choice of the tolerance margin epsilon, i.e. the largest difference which is clinically acceptable, so that a difference bigger than that would matter in practice. In particular, the procedures for the determination of the margin epsilon, used by the authors of the non-inferiority RCCT COBALT and INJECT, are presented and discussed in detail. The ethical implications of equivalence and non-inferiority RCCT are here considered and the reader is repeatedly invited to consider the appropriateness of the basic arguments asserted by the supporters of this kind of studies.

[Statistical notes. Absolute and relative effect measures]. / Note statistiche. Misure di effetto assolute e relative.

The aim of these statistical notes was to give the cardiologist the suitable tools for the understanding of the statistical aspects in the reading of papers presenting the results of randomized controlled clinical trials, in the most simple and intuitive way, without any previous knowledge of basic statistics. The fundamentals of the most common tools for the comparison of two experimental treatments, are developed by getting cue from classical cardiological examples and by guiding the reader, step by step, toward the understanding and critical discussion of the proposed examples. In particular, this note focuses the attention on the main "indicators" of absolute and relative effect measures by evidencing their most important differences.

Picotamide, a combined inhibitor of thromboxane A2 synthase and receptor, reduces 2-year mortality in diabetics with peripheral arterial disease: the DAVID study.

AIMS: Patients with diabetes are at excessive risk of mortality and cardiovascular morbidity. Previous studies suggest that aspirin may be less effective in diabetic patients. In this multi-centre, randomized, double blind trial picotamide, a dual inhibitor of thromboxane A2 synthase and receptor, was compared with aspirin for the prevention of mortality and major cardiovascular events in diabetics with peripheral arterial disease (PAD). METHODS AND RESULTS: A total of 1209 adults aged 40-75 years with type 2 diabetes and PAD were randomized to receive picotamide (600 mg bid) or aspirin (320 mg od) for 24 months. The cumulative incidence of the 2 years overall mortality was significantly lower amongst patients who received picotamide (3.0%) than in those who received aspirin (5.5%) with a relative risk ratio for picotamide versus aspirin of 0.55 (95% CI: 0.31-0.98%). Events were reported in 43 patients (7.1%) on picotamide and 53 (8.7%) on aspirin. The combined endpoint of mortality and morbidity had a slightly lower incidence in the picotamide group but this difference did not reach statistical significance. CONCLUSION: Picotamide is significantly more effective than aspirin in reducing overall mortality in type 2 diabetic patients with associated PAD.

[Equivalence studies in cardiovascular diseases]. / Gli studi di equivalenza nelle malattie cardiovascolari.

An increasing number of equivalence and non-inferiority trials appeared in recent years among randomized clinical trials, particularly for the evaluation of efficacy of treatments in cardiovascular medicine. These studies are characterized by remarkable methodological issues and important ethical implications. The aim of this review was to underscore some topic methodological problems: setting the equivalence boundary, the special logic used to establish equivalence, and the determinants of sample size. A Medline search was performed in order to identify the main problems in study design and in the interpretation of results of the equivalence and non-inferiority trials in cardiovascular diseases.