Clinical features predict responsiveness to imatinib in platelet-derived growth factor receptor-alpha-negative hypereosinophilic syndrome.

BACKGROUND: With the exception of the presence of the FIP1L1-PDGFRA fusion gene, little is known about predictors of imatinib response in clinically-defined hypereosinophilic syndrome (HES). METHODS: Subjects with FIP1L1-PDGFRA-myeloid neoplasm (FP; n =12), PDGFRA-negative HES with ≥4 criteria suggestive of a myeloid neoplasm (MHES; n =10), or steroid-refractory PDGFRA-negative HES with <4 myeloid criteria (SR; n = 5) were enrolled in a prospective study of imatinib therapy (NCT00044304: registered at clinicaltrials.gov). The primary outcome was an eosinophil count <1.5 × 109/L at one month and improvement of clinical symptoms. Clinical, molecular, and bone marrow responses to imatinib were assessed. A retrospective cohort of 18 subjects with clinically-defined HES who received imatinib (300-400 mg daily ≥ 1 month) were classified according to the criteria used in the prospective study. RESULTS: Overall, imatinib response rates were 100% in the FP group (n = 16), 54% in the MHES group (n = 13) and 0% in the SR group (n = 16). The presence of ≥ 4 myeloid features was the sole predictor of response. After ≥ 18 months in complete remission, imatinib was tapered and discontinued in 8 FP and 1 MHES subjects. Seven subjects (6 FP, 1 MHES) remain in remission off therapy for a median of 29 months (range 14-36). CONCLUSIONS: Clinical features of MHES predict imatinib response in PDGFRA-negative HES.

Interdigital lesions and frequency of acute dermatolymphangioadenitis in lymphoedema in a filariasis-endemic area.

BACKGROUND: Lymphatic filariasis (LF) is a mosquito-borne nematode infection that causes permanent lymphatic dysfunction in virtually all infected individuals and clinical disease in a subset of these. One major sequel of infection is lymphoedema of the limbs. Lymphoedema of the leg affects an estimated 15 million persons in LF-endemic areas worldwide. Acute dermatolymphangioadenitis (ADLA) in people with filarial lymphoedema causes acute morbidity and increasingly severe lymphoedema. Episodes of ADLA are believed to be caused by bacteria, and it has been shown that entry lesions in the skin play a causative role. Clinical observations suggest that interdigital skin lesions of the feet, often assumed to be fungal, may be of particular importance. OBJECTIVES: To investigate the epidemiology and aetiology of interdigital lesions (IDL) of the feet in filarial lymphoedema. METHODS: The frequency and mycological aetiology of IDL in 73 patients with filarial lymphoedema were compared with 74 individuals without lymphoedema in a region of Guyana highly endemic for Wuchereria bancrofti. RESULTS: More than 50% of patients with lymphoedema had one or more IDL (odds ratio 2.69; 95% confidence interval 1.31-5.66; P<0.005 compared with controls). The number of lesions was the strongest predictor of frequency of ADLA. Only 18% of the lesions had positive microscopy or culture for fungi (dermatophytes and Scytalidium). CONCLUSIONS: These findings highlight the importance of interdigital entry lesions as risk factors for episodes of ADLA and have implications for the control of morbidity from filarial lymphoedema.

Small-sample adjustments for Wald-type tests using sandwich estimators.

The sandwich estimator of variance may be used to create robust Wald-type tests from estimating equations that are sums of K independent or approximately independent terms. For example, for repeated measures data on K individuals, each term relates to a different individual. These tests applied to a parameter may have greater than nominal size if K is small, or more generally if the parameter to be tested is essentially estimated from a small number of terms in the estimating equation. We offer some practical modifications to these robust Wald-type tests, which asymptotically approach the usual robust Wald-type tests. We show that one of these modifications provides exact coverage for a simple case and examine by simulation the modifications applied to the generalized estimating equations of Liang and Zeger (1986), conditional logistic regression, and the Cox proportional hazard model.

Increased risk of acute myocardial infarction associated with beedi and cigarette smoking in Indians: final report on tobacco risks from a case-control study.

BACKGROUND: Tobacco smoking is an important risk factor for ischemic heart disease. In India, tobacco is smoked both as cigarettes and beedies. No studies have evaluated their importance as risk factors for ischemic heart disease among the Indian population. The present study explores the importance of smoking either cigarettes or beedies as risk factors for acute myocardial infarction. METHODS AND RESULTS: The study had a case-control design and was conducted in a tertiary teaching hospital in Bangalore. Three hundred subjects aged 30-60 years with a first acute myocardial infarction and 300 age- and sex-matched controls were recruited prospectively. Smoking, dietary and social history were recorded, body mass index and waist-hip ratio measured, and blood glucose, lipids, fasting plasma and insulin levels estimated. Cases and controls had a mean age of 47.2 years and 46.8 years, respectively. There were 279 (93%) males in each group. Diabetes mellitus (odds ratio 2.69, p<0.0009). hypertension (odds ratio 2.36, p=0.0009), fasting and post-load blood glucose (p<0.0001). and waist-hip ratio (p<0.0001) were found to be important risk factors for acute myocardial infarction. Smoking was an independent risk factor with a clear dose effect. Adjusted odds ratio for smoking > or = 10 cigarettes/day was 3.58 (p<0.0001) and was 4.36 (p<0.0001) for smoking > or = 10 beedies/day. CONCLUSIONS: Smoking > or = 10 cigarettes or beedies/day carries an independent four-fold increased risk of acute myocardial infarction. This reiterates the need for urgent tobacco control measures in India.

Permutation tests for joinpoint regression with applications to cancer rates.

The identification of changes in the recent trend is an important issue in the analysis of cancer mortality and incidence data. We apply a joinpoint regression model to describe such continuous changes and use the grid-search method to fit the regression function with unknown joinpoints assuming constant variance and uncorrelated errors. We find the number of significant joinpoints by performing several permutation tests, each of which has a correct significance level asymptotically. Each p-value is found using Monte Carlo methods, and the overall asymptotic significance level is maintained through a Bonferroni correction. These tests are extended to the situation with non-constant variance to handle rates with Poisson variation and possibly autocorrelated errors. The performance of these tests are studied via simulations and the tests are applied to U.S. prostate cancer incidence and mortality rates.

Comparing several score tests for interval censored data.

I create a general model to perform score tests on interval censored data. Special cases of this model are the score tests of Finkelstein, Sun and Fay. Although Sun's was derived as a test for discrete data and Finkelstein's and Fay's tests were derived under a grouped continuous model, by writing all tests under one general model we see that as long as the regularity conditions hold, any of these three classes of tests may be applied to either grouped continuous or discrete data. I show the equivalence between the weighted logrank form of the general test and the form with a term for each individual, the form often used with permutation tests. From the weighted logrank form of the tests, we see that Sun's and Finkelstein's test are similar, giving constant (or approximately constant) weights to differences in survival distributions over time. In contrast, the proportional odds model (Fay's model with logistic error) gives more weight to early differences.

Use of the logistic organ dysfunction system to study mortality in an Indian intensive care unit.

BACKGROUND: Mortality in Indian intensive care units has not been well studied. Scoring systems are used to predict mortality of patients admitted to such units. Some scoring systems predict hospital mortality while others predict mortality in intensive care units. We used the logistic organ dysfunction system to study the hospital and intensive care unit mortalities in our intensive care unit. METHODS: We prospectively studied 527 consecutively admitted patients in 1997 to the medical intensive care unit in St John's Medical College Hospital, Bangalore. The outcomes studied were death in hospital and death in the intensive care unit. Using standardized mortality ratios, we compared our observed hospital and intensive care unit mortalities with the hospital mortality predicted by the logistic organ dysfunction system. RESULTS: The standardized mortality ratios for hospital deaths was 1.3 with a confidence interval of 1.17-1.49 and for intensive care unit deaths it was 1.0 with a confidence interval of 0.89-1.18. The hospital mortality rates in our setting are significantly higher (p < 0.05) than the predicted hospital mortality rates of the published western model for intensive care unit patients. The intensive care unit mortality rates are not significantly different from the predicted hospital mortality rates of the published western model for intensive care unit patients. CONCLUSION: Our intensive care unit mortality rate is comparable to the western hospital mortality rate. However, after transfer of patients out of the unit, the hospital mortality is higher.

Conditional logistic regression with sandwich estimators: application to a meta-analysis.

Motivated by a meta-analysis of animal experiments on the effect of dietary fat and total caloric intake on mammary tumorigenesis, we explore the use of sandwich estimators of variance with conditional logistic regression. Classical conditional logistic regression assumes that the parameters are fixed effects across all clusters, while the sandwich estimator gives appropriate inferences for either fixed effects or random effects. However, inference using the standard Wald test with the sandwich estimator requires that each parameter is estimated using information from a large number of clusters. Since our example violates this condition, we introduce two modifications to the standard Wald test. First, we reduce the bias of the empirical variance estimator (the middle of the sandwich) by using standardized residuals. Second, we approximately account for the variance of these estimators by using the t-distribution instead of the normal distribution, where the degrees of freedom are estimated using Satterthwaite's approximation. Through simulations, we show that these sandwich estimators perform almost as well as classical estimators when the true effects are fixed and much better than the classical estimators when the true effects are random. We achieve simulated nominal coverage for these sandwich estimators even when some parameters are estimated from a small number of clusters.

A semi-parametric estimate of extra-Poisson variation for vital rates.

We introduce a method for estimating overdispersion in Poisson models for vital rates. We assume smoothness conditions on the counts to obtain pointwise variance estimates that we combine to obtain an estimate of the overdispersion parameter. We create confidence intervals about the observed rates using this estimate and an approximation based on the gamma distribution. The advantage of this method is that the estimates of the superpopulation rates do not depend on the smoothness assumption, yet when this assumption is met we obtain approximately unbiased estimates of the overdispersion parameter. Thus, we may calculate confidence intervals for vital rates under an overdispersed Poisson model without making parametric assumptions on the mean rates.

Effect of different types and amounts of fat on the development of mammary tumors in rodents: a review.

We performed a meta-analysis on data extracted from 97 reports of experiments, involving a total of 12,803 mice or rats, studying the effect on mammary tumor incidence of different types of dietary fatty acids. Fatty acids were categorized into saturated, monounsaturated, n-6 polyunsaturated, and n-3 polyunsaturated. We modeled the relation between tumor incidence and percentage of total calories from these fatty acids using conditional logistic regression and allowing for varying effects between experiments, and for each fatty acid we estimated the effect of substituting the fatty acid calories for nonfat calories. Our results show that n-6 polyunsaturated fatty acids (PUFAs) have a strong tumor-enhancing effect and that saturated fats have a weaker tumor-enhancing effect. The n-3 PUFAs have a small protective effect that is not statistically significant. There is no significant effect of monounsaturated fats. n-6 PUFAs have a stronger tumor-enhancing effect at levels under 4% of total calories, but an effect is still present at intake levels greater than 4% of calories. In addition, when the intake of n-6 PUFAs is at least 4% of calories, the n-6 PUFA effect remains stronger than the saturated fat effect.

Confidence intervals for directly standardized rates: a method based on the gamma distribution.

We offer an approximation to central confidence intervals for directly standardized rates, where we assume that the rates are distributed as a weighted sum of independent Poisson random variables. Like a recent method proposed by Dobson, Kuulasmaa, Eberle and Scherer, our method gives exact intervals whenever the standard population is proportional to the study population. In cases where the two populations differ non-proportionally, we show through simulation that our method is conservative while other methods (the Dobson et al. method and the approximate bootstrap confidence method) can be liberal.

Meta-analyses of dietary fats and mammary neoplasms in rodent experiments.

We review two meta-analyses of experiments on dietary fat and mammary tumor incidence in rodents, emphasizing a recent meta-analysis on the effects of different types of dietary fatty acids. This analysis shows that n-6 polyunsaturated fatty acids most strongly enhance mammary tumors in rodents, and saturated fats also enhance these tumors but less strongly. Further, the analysis shows that energy restriction protects against mammary tumors. We show that these results agree qualitatively with estimates of effects on human breast cancer derived from international correlations.

Rank invariant tests for interval censored data under the grouped continuous model.

This paper creates rank invariant score tests for grouped or interval censored data. This generalizes Finkelstein (1986, Biometrics 42, 845-854), who derived score tests for interval censored data assuming proportional hazards. We frame the problem as a linear rank test of a shift in location with a known error distribution. We discuss adjustments to the test that may be required when the number of observation times is large. We offer a graphical test of the assumption of the location shift model and discuss an alternative interpretation of the test using the logistic error when the location shift assumption does not hold.

Non-parametric two-sample tests for repeated ordinal responses.

Consider data on two groups of clusters, where each cluster consists of many units that respond on an ordinal scale. We develop a Mann-Whitney type test to determine whether a typical response from the first group is larger (or smaller) than a typical response from the second group.