Estimating gestational age at birth from fundal height and additional anthropometrics: a prospective cohort study

OBJECTIVE

Accurate assessment of gestational age (GA) is critical to paediatric care, but is limited in developing countries without access to ultrasound. Our objectives were to assess the accuracy of prediction of GA at birth and preterm birth classification using routinely collected anthropometry measures.

DESIGN

Prospective cohort study.

SETTING

United States.

POPULATION OR SAMPLE

A total of 2334 non-obese and 468 obese pregnant women.

METHODS

Enrolment GA was determined based on last menstrual period, confirmed by first-trimester ultrasound. Maternal anthropometry and fundal height (FH) were measured by a standardised protocol at study visits; FH alone was additionally abstracted from medical charts. Neonatal anthropometry measurements were obtained at birth. To estimate GA at delivery, we developed three predictor models using longitudinal FH alone and with maternal and neonatal anthropometry. For all predictors, we repeatedly sampled observations to construct training (60%) and test (40%) sets. Linear mixed models incorporated longitudinal maternal anthropometry and a shared parameter model incorporated neonatal anthropometry. We assessed models' accuracy under varied scenarios.

MAIN OUTCOME MEASURES

Estimated GA at delivery.

RESULTS

Prediction error for various combinations of anthropometric measures ranged between 13.9 and 14.9 days. Longitudinal FH alone predicted GA within 14.9 days with relatively stable prediction errors across individual race/ethnicities [whites (13.9 days), blacks (15.1 days), Hispanics (15.5 days) and Asians (13.1 days)], and correctly identified 75% of preterm births. The model was robust to additional scenarios.

CONCLUSIONS

In low-risk, non-obese women, longitudinal FH measures alone can provide a reasonably accurate assessment of GA when ultrasound measures are not available.

TWEETABLE ABSTRACT

Longitudinal fundal height alone predicts gestational age at birth when ultrasound measures are unavailable.

Comparison of methods for identifying small-for-gestational-age infants at risk of perinatal mortality among obese mothers: a hospital-based cohort study

OBJECTIVE

To assess differences in small-for-gestational age (SGA) classifications for the detection of neonates with increased perinatal mortality risk among obese women and subsequently assess the association between prepregnancy body mass index (BMI) status and SGA.

DESIGN

Hospital-based cohort.

SETTING

Twelve US clinical centres (2002-08).

POPULATION

A total of 114 626 singleton, nonanomalous pregnancies.

METHODS

Data were collected using electronic medical record abstraction. Relative risks (RR) with 95% CI were estimated.

MAIN OUTCOME MEASURES

SGA trends (birthweight < 10th centile) classified using population-based (SGA_POP ), intrauterine (SGA_IU ) and customised (SGA_CUST ) references were assessed. The SGA-associated perinatal mortality risk was estimated among obese women. Using the SGA method most associated with perinatal mortality, the association between prepregnancy BMI and SGA was estimated.

RESULTS

The overall perinatal mortality prevalence was 0.55% and this increased significantly with increasing BMI (P < 0.01). Among obese women, SGA_IU detected the highest proportion of perinatal mortality cases (2.49%). Perinatal mortality was 5.32 times (95% CI 3.72-7.60) more likely among SGA_IU neonates than non-SGA_IU neonates. This is in comparison with the 3.71-fold (2.49-5.53) and 4.81-fold (3.41-6.80) increased risk observed when SGA_POP and SGA_CUST were used, respectively. Compared with women of normal weight, overweight women (RR = 0.82, 95% CI 0.78-0.86) and obese women (RR = 0.80; 95% CI 0.75-0.83) had a lower risk for delivering an SGA_IU neonate.

CONCLUSION

Among obese women, the intrauterine reference best identified neonates at risk of perinatal mortality. Based on SGA_IU , SGA is less common among obese women but these SGA babies are at a high risk of death and remain an important group for surveillance.

TWEETABLE ABSTRACT

SGA is less common among obese women but these SGA babies are at a high risk of death.

Differences in risk factors for incident and recurrent small-for-gestational-age birthweight: a hospital-based cohort study

OBJECTIVE

Examine whether small-for-gestational-age (SGA) risk factors differed by prior SGA birth.

DESIGN

Hospital-based cohort study.

SETTING

Utah, USA.

POPULATION

Electronic medical record data from 25,241 women who were nulliparous at study entry with ≥2 subsequent consecutive singleton deliveries (2002-2010).

METHODS

Estimated adjusted relative risks (RR) and 95% confidence intervals (95% CI) for the association between second pregnancy characteristics and SGA risk. Tested for risk factor differences between recurrence and incidence (Pdifference).

MAIN OUTCOME MEASURES

Second pregnancy incident (n = 1067) and recurrent SGA (n = 484) determined using a population-based reference.

RESULTS

SGA complicated 20.3 and 4.5% of deliveries to women with and without a prior SGA birth, respectively. Young maternal age (Pdifference = 0.01) and pregnancy hypertensive diseases (Pdifference = 0.03) were associated with incident but not recurrent SGA. Significant risk factors for incidence and recurrence were smoking (incident RR = 1.64 [95% CI 1.22-2.19]; recurrent RR = 1.59 [95% CI 1.17-2.17]), short stature (incident RR = 1.34 [95% CI 1.16-1.54]; recurrent RR = 1.54 [95% CI 1.31-1.82]), prepregnancy underweight (incident RR = 1.32 [95% CI 1.07-1.64]; recurrent RR = 1.30 [95% CI 1.03-1.64]), and inadequate weight gain (incident RR = 1.41 [95% CI 1.22-1.64]; recurrent RR = 1.33 [95% CI 1.10-1.60]). Race-ethnicity, marital or insurance status, alcohol, diabetes, asthma, thyroid disease, depression, or interpregnancy interval were not associated with incidence or recurrence.

CONCLUSION

There was considerable overlap in the risk factors for SGA recurrence and incidence. Recurrence and incidence risk factors included smoking, short stature, underweight, and inadequate weight gain. Maternal age and hypertensive diseases increased the risk for incidence only. Regardless of the SGA definition, some potentially modifiable risk factors for recurrence were identified.

Estimation of gene-environment interaction by pooling biospecimens

Case-control studies are prone to low power for testing gene-environment interactions (GXE) given the need for a sufficient number of individuals on each strata of disease, gene, and environment. We propose a new study design to increase power by strategically pooling biospecimens. Pooling biospecimens allows us to increase the number of subjects significantly, thereby providing substantial increase in power. We focus on a special, although realistic case, where disease and environmental statuses are binary, and gene status is ordinal with each individual having 0, 1, or 2 minor alleles. Through pooling, we obtain an allele frequency for each level of disease and environmental status. Using the allele frequencies, we develop a new methodology for estimating and testing GXE that is comparable to the situation when we have complete data on gene status for each individual. We also explore the measurement process and its effect on the GXE estimator. Using an illustration, we show the effectiveness of pooling with an epidemiologic study, which tests an interaction for fiber and paraoxonase on anovulation. Through simulation, we show that taking 12 pooled measurements from 1000 individuals achieves more power than individually genotyping 500 individuals. Our findings suggest that strategic pooling should be considered when an investigator designs a pilot study to test for a GXE.

Moderate alcohol consumption and 24-hour urinary levels of melatonin in postmenopausal women

CONTEXT

Low overnight urinary melatonin metabolite concentrations have been associated with increased risk for breast cancer among postmenopausal women. The Postmenopausal Women's Alcohol Study was a controlled feeding study to test the effects of low to moderate alcohol intake on potential risk factors for breast cancer including serum and urinary levels of hormones and other biomarkers. Previously, we observed significant increases in concentrations of serum estrone sulfate and dehydroepiandrosterone sulfate in participants after consumption of 15 or 30 g (one or two drinks) of alcohol per day.

OBJECTIVE

In the present analysis, we evaluated the relationship of alcohol consumption with 24-h urinary 6-sulfatoxymelatonin (6-SMT) concentration (micrograms per 24 h).

DESIGN AND PARTICIPANTS

Healthy postmenopausal women (n = 51) consumed a controlled diet plus each of three treatments (a nonalcoholic placebo beverage or 15 or 30 g alcohol/d) during three 8-wk periods in random order under conditions of weight maintenance.

MEASURES

6-SMT was measured in 24-h urine samples that were collected at entry into the study (baseline) and at the midpoint (4 wk) and end (8 wk) of each of the three diet periods.

RESULTS

Concentration of 6-SMT was not significantly modified by the alcohol treatment after adjustment for body mass index, hours of sleep, daylight hours, and baseline level of 6-SMT.

CONCLUSIONS

These results suggest that low to moderate daily alcohol consumption does not significantly affect 24-h urinary levels of melatonin among healthy postmenopausal women.

A two-part model for reference curve estimation subject to a limit of detection

Reference curves are commonly used to identify individuals with extreme values of clinically relevant variables or stages of progression which depend naturally on age or maturation. Estimation of reference curves can be complicated by a technical limit of detection (LOD) that censors the measurement from the left, as is the case in our study of reproductive hormone levels in boys around the time of the onset of puberty. We discuss issues with common approaches to the LOD problem in the context of our pubertal hormone study, and propose a two-part model that addresses these issues. One part of the proposed model specifies the probability of a measurement exceeding the LOD as a function of age. The other part of the model specifies the conditional distribution of a measurement given that it exceeds the LOD, again as a function of age. Information from the two parts can be combined to estimate the identifiable portion (i.e. above the LOD) of a reference curve and to calculate the relative standing of a given measurement above the LOD. Unlike some common approaches to LOD problems, the two-part model is free of untestable assumptions involving unobservable quantities, flexible for modeling the observable data, and easy to implement with existing software. The method is illustrated with hormone data from the Third National Health and Nutrition Examination Survey.

A high-legume low-glycemic index diet reduces fasting plasma leptin in middle-aged insulin-resistant and -sensitive men

Fasting leptin and ghrelin levels were measured in 36 insulin-sensitive (IS) and 28 insulin-resistant (IR) men who consumed a legume-enriched low-glycemic index (LG) diet or healthy American (HA) diet in a randomly ordered cross-over feeding study consisting of two 4-week periods. Weight remained stable over the entire study. Fasting plasma leptin was significantly reduced from pre-study levels by both the LG (18.8%, P < 0.001) and HA (16.1%, P < 0.001) diets, whereas fasting ghrelin did not change. By subgroup analysis according to prestudy insulin status, leptin was reduced in IR subjects after both the LG (17.1%, P < 0.01) and the HA (33.3%, P < 0.001) diets, whereas IS subjects responded only after the LG diet (23.1%, P < 0.01). Thus, a legume-rich LG index diet may be a beneficial strategy for reducing circulating leptin concentrations, even under conditions of weight maintenance.

Serum cytokine concentrations, flavonol intake and colorectal adenoma recurrence in the Polyp Prevention Trial

Background:

Serum cytokine concentrations may reflect inflammatory processes occurring during the development of colorectal neoplasms. Flavonols, bioactive compounds found in plant-based foods and beverages, may inhibit colorectal neoplasms partly by attenuating inflammation.

Methods:

Using logistic regression, we estimated odds ratios (ORs) and 95% confidence intervals (CIs) to investigate the association between serum concentrations of interleukin (IL)1β, 2, 8, 10, 12p70, granulocyte macrophage colony stimulating factor, interferon-γ, and tumour necrosis factor-α, measured over time, flavonol intake, estimated from a flavonol database used in conjunction with a food frequency questionnaire, and adenoma recurrence in 872 participants from the intervention arm of the Polyp Prevention Trial.

Results:

Decreased IL-2 concentration during the trial increased the risk of any adenoma recurrence (4th vs 1st quartile, OR=1.68, 95% CI=1.13–2.49), whereas decreased IL-1β or IL-10 reduced the risk of advanced adenoma recurrence (OR=0.37, 95% CI=0.15–0.94; OR=0.39, 95% CI=0.15–0.98, respectively). Individuals with flavonol intake above the median (29.7 mg per day) and decreased cytokine concentrations had the lowest risk of advanced adenoma recurrence.

Conclusion:

Overall, no consistent associations were observed between serum cytokine profile and colorectal adenoma recurrence; however, decreased cytokine concentrations during high flavonol consumption may indicate prevention of colorectal neoplasms.

Diversity of chromosomal karyotypes in maize and its relatives

Maize is a highly diverse species on the gene sequence level. With the recent development of methods to distinguish each of the 10 pairs of homologues in somatic root tip spreads, a wide collection of maize lines was subjected to karyotype analysis to serve as a reference for the community and to examine the spectrum of chromosomal features in the species. The core nested association mapping progenitor collection and additional selections of diversity lines were examined. Commonly used inbred lines were included in the analysis. The centromere 4 specific repeat and ribosomal RNA loci were invariant. The CentC centromere repeat exhibited extensive differences in quantity on any particular chromosome across lines. Knob heterochromatin was highly variable with locations at many sites in the genome. Lastly, representative examples from other species in the genus Zea (teosintes) were examined, which provide information on the evolution of chromosomal features.

Components of variation in serum carotenoid concentrations: the Polyp Prevention Trial

OBJECTIVES

The intra- and interindividual variations and season and center effects were estimated from a series of serum carotenoid concentrations in the Polyp Prevention Trial (PPT) participants.

SUBJECTS/METHODS

Fasting blood was collected annually for 4 years in all 1905 participants, and a subcohort of 901 participants were selected within each (of eight) center(s), by gender and dietary arm of the study, for measurement of five major carotenoid peaks. Using variance of component methods, the variation in serum carotenoid concentrations about the underlying mean was partitioned into explanatory components attributed to various sources.

RESULTS

The contributions of the inter- and intraindividual variances to the overall variation in carotenoid concentrations were in the range of 61-70 and 20-35%, respectively, whereas center and center-by-season effects provided 2.6-9.5 and 0.2-1.4%, respectively. The highest percent (35%) of intraindividual variation was exhibited by lycopene, and the highest percent (70% apiece) of interindividual variation was exhibited by lutein/zeaxanthin and beta-carotene. Serum lycopene had the highest ratio of intra- to interindividual variation of 0.57, whereas lutein had the lowest ratio of 0.29. We estimate that the ratio of intra- to interindividual variance around the mean carotenoid concentration can be reduced greatly by collecting 3-4 compared to 1 blood measurement in large-scale trials like the PPT.

CONCLUSION

In the largest study of components of variation in individuals at high risk for colorectal cancer, the largest contributors to variation in serum carotenoid concentrations were intra- and interindividual effects followed by center and center-by-season effects.

Pulmonary function following total body irradiation (with or without lung shielding) and allogeneic peripheral blood stem cell transplant

Our purpose was to determine if total body irradiation (TBI) with lung dose reduction protects against subsequent radiation-induced deterioration in pulmonary function. Between July 1997 and August 2004, 181 consecutive patients with hematologic malignancies received fractionated TBI before allogeneic peripheral blood stem cell transplant. The first 89 patients were treated to a total dose of 13.6 Gy. Thereafter, total body dose was decreased to 12 Gy with lung dose reduction to 9 or 6 Gy. All patients underwent pulmonary function test evaluation before treatment, 90 days post-treatment, then annually. Median follow-up was 24.0 months. Eighty-nine patients were treated with lung shielding, and 92 without. At 1-year post transplant, there was a small but significant difference in lung volume measurements between patients with lung shielding and those without. This was not observed at the 2-year time point. When stratified by good (>100% predicted) or poor (</=100% predicted) baseline lung function, patients with poor function demonstrated protection at 1 year with lung shielding, while those with good initial lung function did not. TBI with or without lung dose reduction has a small but statistically significant effect on pulmonary function measured at 1 year but not 2 years post irradiation.

Identification of a unique epigenetic sub-microenvironment in prostate cancer

The glutathione S-transferase P1 (GSTP1) gene promoter is methylated in tumour cells in more than 90% of prostate carcinomas. Recently, GSTP1 promoter methylation was identified in tumour-associated stromal cells in addition to the tumour epithelium. To define the extent and location of stromal methylation, epigenetic mapping using pyrosequencing quantification of GSTP1 promoter methylation and an anatomical three-dimensional reconstruction of an entire human prostate specimen with cancer were performed. Normal epithelium and stroma, tumour epithelium, and tumour-associated stromal cells were laser capture-microdissected from multiple locations throughout the gland. As expected, the GSTP1 promoter in both normal epithelium and normal stromal cells distant from the tumour was not methylated and the tumour epithelium showed consistently high levels of promoter methylation throughout. However, tumour-associated stromal cells were found to be methylated only in a localized and distinct anatomical sub-field of the tumour, revealing the presence of an epigenetically unique microenvironment within the cancer. Morphologically, the sub-field consisted of typical, non-reactive stroma, representing a genomic alteration in cells that appeared otherwise histologically normal. Similar epigenetic anatomical mapping of a control prostate gland without cancer showed low background methylation levels in all cell types throughout the specimen. These data suggest that stromal cell methylation can occur in a distinct sub-region of prostate cancer and may have implications for understanding tumour biology and clinical intervention.

An analysis of erectile function after intensity modulated radiation therapy for localized prostate carcinoma

Radiation therapy for prostate cancer can cause erectile dysfunction (ED). Intensity Modulated Radiation Therapy (IMRT) can reduce the amount of radiation to surrounding tissues associated with ED. We characterize the incidence of and factors associated with ED in prostate cancer patients after IMRT at the National Naval Medical Center (NNMC). Patients potent by definition of the Sexual Health Inventory for Men (SHIM) before treatment completed the specific erectile questions of the SHIM after IMRT. Statistical analyses were performed to examine the relationships between several factors and ED. Thirty-two of 45 patients with mean age of 68.2 years (50-86 years) completed the SHIM. The median follow-up was 36.8 months (16-63.6 months) as defined by the time from completion of therapy to reassessment with the SHIM. Eight of 32 patients (25%) had no post-treatment ED (SHIM score 22-25), three of 32 (9%) had mild post-treatment ED (SHIM score 17-21), five of 32 (16%) had mild to moderate ED (SHIM score 12-16), five of 32 (16%) had moderate ED (SHIM score 8-11) and 11 of 32 (34%) had severe post-treatment ED (SHIM score<8). Post-treatment potency was significantly associated with the pre-treatment SHIM score (P=0.001) and history of hypertension (P=0.03). The mean radiation dose to the penile bulb and volume of penile bulb treated were not associated with post-treatment potency (P=0.38, 0.76, respectively). IMRT maintains potency in the majority of patients. This analysis compares favorably in preserving erectile function to previously reported series using conventional external beam radiation therapy techniques. The dose of radiation received by the penile bulb and volume of penile bulb were not associated with post-treatment ED in this analysis.

Sensitive fluorescence in situ hybridization signal detection in maize using directly labeled probes produced by high concentration DNA polymerase nick translation

The signal produced by fluorescence in situ hybridization (FISH) often is inconsistent among cells and sensitivity is low. Small DNA targets on the chromatin are difficult to detect. We report here an improved nick translation procedure for Texas red and Alexa Fluor 488 direct labeling of FISH probes. Brighter probes can be obtained by adding excess DNA polymerase I. Using such probes, a 30 kb yeast transgene, and the rp1, rp3 and zein multigene clusters were clearly detected.

Efficacy of the Stone Cone for Treatment of Proximal Ureteral Stones

BACKGROUND AND PURPOSE

Whereas ureteroscopic lithotripsy is more efficacious than SWL in treating lower-ureteral stones, the same has not been universally said of its ability to treat proximal ureteral stones. Because failed proximal lithotripsy is often attributable to the complications associated with stone migration into the renal pelvis and calices, an instrument that can prevent this migration is a potentially important tool in the ureteroscopic armamentarium. This study sought to assess the role of just such an instrument, the Stone Cone, in proximal-ureteral lithotripsy.

PATIENTS AND METHODS

We treated 19 consecutive patients having proximal-ureteral stones using semirigid ureteroscopy, a Stone Cone Nitinol urologic retrieval coil, and holmium:YAG laser lithotripsy with a 200- or 365-microm fiber. In all patients, both the Stone Cone and the laser fiber were utilized under direct visual guidance through the working channel(s) of the ureteroscope.

RESULTS

All 19 patients were rendered stone free after Ho:YAG laser lithotripsy in conjunction with a Stone Cone. No stone fragments were noted to migrate into the renal pelvis, and the Stone Cone did not break or become entrapped in any of the 19 cases.

CONCLUSION

The Stone Cone is a powerful new tool for proximal-ureteral lithotripsy and will likely revolutionize the treatment of proximal-ureteral stones. The savings in morbidity, time, and money associated with not having to chase stone fragments using flexible ureteroscopy are considerable.

Moderate alcohol consumption and levels of antioxidant vitamins and isoprostanes in postmenopausal women

BACKGROUND

Although alcohol intake has been positively associated with breast cancer risk in epidemiologic studies, the mechanisms mediating this association are speculative.

OBJECTIVE

The Postmenopausal Women's Alcohol Study was designed to explore the effects of moderate alcohol consumption on potential risk factors for breast cancer. In the present analysis, we evaluated the relationship of alcohol consumption with antioxidant nutrients and a biomarker of oxidative stress.

DESIGN

Participants (n=53) consumed a controlled diet plus each of three treatments (15 or 30 g alcohol/day or a no-alcohol placebo beverage), during three 8-week periods in random order. We measured the antioxidants, vitamin E (alpha (alpha)- and gamma (gamma)-tocopherols), selenium, and vitamin C in fasting blood samples which were collected at the end of diet periods, treated and frozen for assay at the end of the study. We also measured 15-F(2t)-IsoP isoprostane, produced by lipid peroxidation, which serves as an indicator of oxidative stress and may serve as a biomarker for conditions favorable to carcinogenesis.

RESULTS

After adjusting for BMI (all models) and total serum cholesterol (tocopherol and isoprostane models) we observed a significant 4.6% decrease (P=0.02) in alpha-tocopherol and a marginally significant 4.9% increase (P=0.07) in isoprostane levels when women consumed 30 g alcohol/day (P=0.06 and 0.05 for overall effect of alcohol on alpha-tocopherol and isoprostanes, respectively). The other antioxidants were not significantly modified by the alcohol treatment.

CONCLUSIONS

These results suggest that moderate alcohol consumption increases some biomarkers of oxidative stress in postmenopausal women.

Common genetic variants of TP53 and BRCA2 in esophageal cancer patients and healthy individuals from low and high risk areas of northern China

TP53 and BRCA2 are frequently mutated in cancer and polymorphisms of these genes may modify cancer risk. We used SSCP and DNA sequencing to assess and compare frequencies of R72P (TP53) and 5'UTR203G>A, N372H, and K1132K (BRCA2) polymorphisms in healthy Chinese subjects at varying risk for esophageal squamous cell carcinoma (ESCC) and in ESCC patients. Suggestive overall differences in the distributions of genotypes by risk groups were seen for all genotypes except K1132K. Differences in R72P and N372H were most likely a reflection of lack of Hardy-Weinberg equilibrium (HWE), however, the difference in 203G>A was due to low prevalence of GG in ESCC patients (0.22 versus 0.36 in high risk group (P=0.047), and 0.22 versus 0.40 in low risk group (P=0.010)), consistent with a disease association. These data suggest that the 203G>A polymorphism in BRCA2 may be associated with risk of ESCC.

Hormone replacement therapy and colorectal adenoma recurrence among women in the Polyp Prevention Trial

BACKGROUND

Epidemiologic studies have suggested that estrogen may protect against the development of colorectal cancers and adenomatous polyps. We conducted a prospective study to evaluate the association between hormone replacement therapy (HRT) and adenoma recurrence among perimenopausal and postmenopausal women participating in the Polyp Prevention Trial, a randomized dietary intervention study of individuals with colorectal adenomas.

METHODS

We used a questionnaire and interviews to collect detailed information, at baseline and at each of four annual study visits, from 620 women regarding hormone use, menopausal status, diet, alcohol consumption, and other risk factors. Adenoma recurrence was ascertained by complete colonoscopy at baseline and after 1 and 4 years. Logistic regression models were used to evaluate the association between hormone use and adenoma recurrence after adjusting for intervention group and for age and body mass index at baseline. All statistical tests were two-sided.

RESULTS

Adenomas recurred in 200 women. There was no overall association between adenoma recurrence and either overall hormone use (odds ratio [OR] = 1.01; 95% confidence interval [CI] = 0.70 to 1.45), combined estrogen and progestin use (OR = 0.94; 95% CI = 0.57 to 1.56), or unopposed estrogen use (OR = 1.04; 95% CI = 0.68 to 1.59). HRT use was associated with a reduction in risk for recurrence of distal adenomas (OR = 0.56; 95% CI = 0.32 to 1.00) and a statistically nonsignificant increase in risk for recurrence of proximal adenomas (OR = 1.39; 95% CI = 0.85 to 2.26). We observed a statistically significant interaction between the HRT-adenoma recurrence association and age (P =.02). HRT was associated with a 40% reduced risk of adenoma recurrence among women older than 62 years (OR = 0.58; 95% CI = 0.35 to 0.97) but with an increased risk among women younger than 62 years (OR = 1.99; 95% CI = 1.11 to 3.55).

CONCLUSIONS

HRT was not associated with a reduced risk for overall adenoma recurrence in this trial cohort, although there was a suggestion of an age interaction. The effect of age on the association needs to be confirmed in other adenoma recurrence trials.

Sphingolipids as biomarkers of fumonisin exposure and risk of esophageal squamous cell carcinoma in china

OBJECTIVE

Ecologic studies of esophageal squamous cell carcinoma (ESCC) have reported an association with consumption of maize contaminated with Fusarium verticillioides, which produce fungal toxins referred to as fumonisins. Fumonisins disrupt sphingolipid metabolism and serum sphingolipids have been proposed as biomarkers of fumonisin exposure. We conducted a prospective nested case-control study to examine the relationship between serum sphingolipids and ESCC incidence.

METHODS

Cases and controls were selected from a large prospective trial conducted in Linxian, People's Republic of China. Ninety-eight ESCC cases were randomly selected from the 639 incident ESCC ascertained during the initial 5.25 years of follow-up: 185 controls were also randomly selected based on the distribution of cases among six age and sex strata. Concentrations of sphinganine and sphingosine were determined by high-performance liquid chromatography in serum collected at the study baseline.

RESULTS

No significant associations were found between serum sphingosine, sphinganine, or the sphinganine/ sphingosine ratio and ESCC incidence in conditional and unconditional logistic regression models with adjustment for age, sex, tobacco use. and alcohol use.

CONCLUSION

Our study is the first prospective study to assess the relationship between sphingolipid levels, as biomarkers of fumonisin exposure, and cancer incidence. We found no significant association between sphingolipid levels and risk of ESCC.

Limitations of the case-only design for identifying gene-environment interactions

The case-only design, which requires only diseased subjects, allows for estimation of multiplicative interactions between factors known to be independent in the study population. The design is being used as an alternative to the case-control design to study gene-environment interactions. Estimates of gene-environment interactions have been shown to be very efficient relative to estimates obtained with a case-control study under the assumption of independence between the genetic and environmental factors. In this paper, the authors explore the robustness of this procedure to uncertainty about the independence assumption. By using simulations, they demonstrate that inferences about the multiplicative interaction with the case-only design can be highly distorted when there is departure from the independence assumption. They illustrate their results with a recent study of gene-environment interactions and risk of lung cancer incidence in a cohort of miners from the Yunnan Tin Corporation in southern China. Investigators should be aware that the increased efficiency of the case-only design is a consequence of a strong assumption and that this design can perform poorly if the assumption is violated.

A cross-sectional study of human serum sphingolipids, diet and physiologic parameters

Sphinganine and sphingosine, the two sphingoid base backbones of sphingolipids, are highly bioactive compounds that are of increasing interest to nutritionists because they occur in food and their metabolism can be altered by fungal toxins that contaminate some foods. Nonetheless, no studies of diet and sphinganine or sphingosine concentrations in serum have yet been reported. Here we describe a cross-sectional study of 265 residents of Linxian, People's Republic of China, which examines potential demographic, physiologic and dietary correlates of serum sphinganine and sphingosine in this population. Median concentrations of serum sphinganine and sphingosine were compared among strata for 29 different variables. For sphinganine, no significant differences were found. For sphingosine, significant differences were seen among strata of age, menstruation status, serum cholesterol, carotenoids, retinol, tocopherols, fresh and dried vegetable and fresh fruit consumption. Using multivariate linear regression with stepwise selection, we found that the significant predictors for serum sphingosine included total tocopherols, age, serum selenium and retinol, with a final R(2) = 0.22; for sphinganine, tooth loss was the sole correlate, with R(2) = 0.015. Analyses using ranked sphingolipid data or principal components analysis, to simplify the food variables, did not materially alter these results. This study represents the largest report of human serum sphingolipid concentrations to date and provides insight into potential explanatory variables that can be incorporated into future studies.

Therapeutic effects of viral vector-mediated antiangiogenic gene transfer in malignant ascites

Malignant ascites is a common complication of advanced intraabdominal neoplasms for which standard treatments are suboptimal. Evidence suggests that tumor-mediated angiogenesis and enhanced vascular permeability in the peritoneal wall due to high levels of vascular endothelial growth factor play a fundamental role in the pathogenesis of malignant ascites. To explore the advantage of viral vector-mediated "targeted antiangiogenic therapy" in ascites formation, we constructed and administered adenoviral vectors encoding several different antiangiogenic proteins (angiostatin, endostatin, platelet factor 4, and a fusion protein between angiostatin and endostatin) alone or in combination intraperitoneally in mice with peritoneal carcinomatosis from breast cancer (TA3 cells) and ovarian cancer (SKOV-3 i.p. and ES-2 cell lines) to explore the potential of additive or synergistic activity. Our data demonstrated statistically significant downregulation of ascites formation, tumor growth, vascularity, and prolongation of animal survival after intraperitoneal treatment with antiangiogenic adenoviral vectors in three different ascites tumor models. Combined treatment proved to be more effective than treatment with one vector alone. Reduced ascites formation was accompanied by decreased microvascular density in the peritoneal wall and increased apoptosis of tumor cells after administration of antiangiogenic vectors in vivo. Of interest was the observation that AdPF4 caused a significant decrease in the level of VEGF secreted by tumor cells both in vitro and in TA3 ascites tumor-bearing animals in vivo. These data suggest that adenoviral vector-mediated delivery of genes encoding antiangiogenic proteins may represent a potentially new treatment modality for malignant ascites.

Comparison of one-, two-, and three-dimensional measurements of childhood brain tumors

BACKGROUND

End points for assessing drug activity in brain tumors are determined by measuring the change in tumor size by magnetic resonance imaging (MRI) relative to a pretreatment or best-response scan. Traditionally, two-dimensional (2D) tumor measurements have been used, but one-dimensional (1D) measurements have recently been proposed as an alternative. Because software to estimate three-dimensional (3D) tumor volume from digitized MRI images is available, we compared all three methods of tumor measurement for childhood brain tumors and clinical outcome.

METHODS

Tumor size from 130 MRI scans from 32 patients (32 baseline and 98 follow-up scans, for a total of 130 scans; median, three scans per patient; range, two to 18 scans) was measured by each method. Tumor-response category (partial response, minor response, stable disease, or progressive disease) was determined from the percentage change in tumor size between the baseline or best-response scan and follow-up scans. Time to clinical progression was independently determined by chart review. All statistical tests were two-sided.

RESULTS

Concordances between 1D and 2D, 1D and 3D, and 2D and 3D were 83% (95% confidence interval [CI] = 67% to 99%), 61% (95% CI = 47% to 75%), and 66% (95% CI = 52% to 80%), respectively, on follow-up scans. Concordances for 1D and 3D and for 2D and 3D were statistically significantly lower than the concordance for 1D and 2D (P< .001 and P = .003, respectively). Concordance among 1D, 2D, and 3D methods in detecting partial response was high; there was less concordance in classifying tumors in the minor response and progressive-disease categories. Median times to progression measured by the 1D, 2D, and 3D methods were 154, 105, and 112 days, respectively, compared with 114 days based on neurologic symptoms and signs (P = .09 for overall comparison).

CONCLUSIONS

Detection of partial responses was not influenced by the measurement method, but estimating time to disease progression may be method dependent for childhood brain tumors.

Latent class modeling approaches for assessing diagnostic error without a gold standard: with applications to p53 immunohistochemical assays in bladder tumors

Improved characterization of tumors for purposes of guiding treatment decisions for cancer patients will require that accurate and reproducible assays be developed for a variety of tumor markers. No gold standards exist for most tumor marker assays. Therefore, estimates of assay sensitivity and specificity cannot be obtained unless a latent class model-based approach is used. Our goal in this article is to estimate sensitivity and specificity for p53 immunohistochemical assays of bladder tumors using data from a reproducibility study conducted by the National Cancer Institute Bladder Tumor Marker Network. We review latent class modeling approaches proposed by previous authors, and we find that many of these approaches impose assumptions about specimen heterogeneity that are not consistent with the biology of bladder tumors. We present flexible mixture model alternatives that are biologically plausible for our example, and we use them to estimate sensitivity and specificity for our p53 assay example. These mixture models are shown to offer an improvement over other methods in a variety of settings, but we caution that, in general, care must be taken in applying latent class models.

Serum hormones and the alcohol-breast cancer association in postmenopausal women

BACKGROUND

Alcohol ingestion is associated with an increased risk of breast cancer in most epidemiologic studies. Results, however, are heterogeneous at lower levels of alcohol intake, and a biologic mechanism for the association has not been clearly identified. To determine whether alcohol consumption by postmenopausal women elevates serum levels of hormones associated with an increased risk of breast cancer, we performed a controlled feeding study.

METHODS

Participants were 51 healthy postmenopausal women not using hormone replacement therapy. Each participant rotated through three 8-week dietary periods in which she consumed 15 or 30 g of alcohol per day or an alcohol-free placebo beverage. The order of assignment to the three alcohol levels was random. During the dietary periods, all food and beverages were supplied by the study, and energy intake was adjusted to keep body weight constant. Levels of estradiol, estrone, estrone sulfate, testosterone, androstenedione, progesterone, dehydroepiandrosterone (DHEA), DHEA sulfate (DHEAS), and androstenediol were measured by radioimmunoassays in serum collected at the end of each dietary period. All statistical tests are two-sided.

RESULTS

When women consumed 15 or 30 g of alcohol per day, respectively, estrone sulfate concentrations increased by 7.5% (95% confidence interval [CI] = -0.3% to 15.9%; P =.06) and 10.7% (95% CI = 2.7% to 19.3%; P =.009) and DHEAS concentrations increased by 5.1% (95% CI = 1.4% to 9.0%; P =.008) and 7.5% (95% CI = 3.7% to 11.5%; P<.001) relative to levels when women consumed placebo. None of the other hormones measured changed statistically significantly when women consumed alcohol.

CONCLUSIONS

Results suggest a possible mechanism by which consumption of one or two alcoholic drinks per day by postmenopausal women could increase their risk of breast cancer.

Retrospective stratification of a consecutive cohort of prostate cancer patients treated with a combined regimen of external-beam radiotherapy and brachytherapy

PURPOSE

The evaluation of clinical variables that influence biochemical relapse-free survival in a cohort of patients treated by combined radiotherapy over a fixed interval.

METHODS AND MATERIALS

Three hundred forty-eight patients diagnosed with clinical Stage T1--T3a prostate cancer were treated with a course of (103)Pd or (125)I brachytherapy followed by a limited course of external beam radiation formed the basis for study. All censored patients had a minimum 2-year follow-up. Biochemical relapse-free survival (BRFS) was estimated using a modified American Society for Therapeutic Radiology and Oncology consensus definition. Discrete "risk groups" were developed based on BRFS as influenced by pretreatment parameters.

RESULTS

Significant risk factors contributing to biochemical failure were serum prostate-specific antigen (PSA) greater than 20 ng/mL, Gleason sum of 7 or greater, or clinical stage T2c or greater. Five-year biochemical control for those exhibiting no risk factor was 88%; one risk factor, 75%; two or more risk factors, 51%. The differences in BRFS among all three risk groups were statistically significant. Outcomes for patients presenting with PSA 10 to 20 ng/mL, but otherwise low-risk disease, fared no differently from those low risk patients presenting with PSA less than 10 ng/mL.

CONCLUSIONS

Combined radiotherapy with (103)Pd or (125)I followed by external beam radiotherapy achieves a high rate of biochemical and clinical control in patients with low- to intermediate-risk clinically organ confined disease.

Concurrent paclitaxel and radiation in the treatment of locally advanced head and neck cancer

PURPOSE

To determine the feasibility of an organ preservation regimen consisting of infusional paclitaxel administered concurrently with radiotherapy to patients with locally advanced head and neck squamous cell carcinoma (HNSCC).

PATIENTS AND METHODS

Thirty-three previously untreated patients with stage III or IV tumors were enrolled onto the study. Paclitaxel was administered as a 120-hour continuous infusion every 3 weeks during the course of radiation therapy. Sixteen patients received a paclitaxel dose of 105 mg/m(2), and 17 patients received 120 mg/m(2). Radiation was delivered in a standard format at 1.8 Gy/d to a total dose of 70.2 to 72 Gy.

RESULTS

Three months after therapy, a 76% complete response (CR) at the primary site and a 70% overall CR was achieved. At 36 months, locoregional control was 55.7%, overall survival was 57.8%, and disease-free survival was 51.1%. The median survival duration for all 33 patients was greater than 50 months at the time of this report. Local toxicities including mucositis, dysphagia, and skin reactions were severe but tolerable. All patients retained functional speech, and all but four patients were swallowing food 3 months after treatment. Steady-state plasma concentrations for paclitaxel were not achieved during a 120-hour infusion, suggesting a nonlinear process. Tumor volume quantified by pretreatment computerized tomography imaging was associated with likelihood of response and survival.

CONCLUSION

Paclitaxel administered as a 120-hour continuous infusion in combination with radiotherapy is a feasible and promising treatment for patients with advanced HNSCC.

Adjusting for drop-out in clinical trials with repeated measures: design and analysis issues

Recently, Wu and Follmann developed summary measures to adjust for informative drop-out in longitudinal studies where drop-out depends on the underlying true value of the response. In this paper we evaluate these procedures in the common situation where drop-out depends on the observed responses. We also discuss various design and analysis strategies which minimize the bias obtained with this type of drop-out. Of particular interest is the use of multiple measurements of the response at each visit to reduce bias. These strategies are evaluated with a simulation study. The results are highlighted with applications to both a hypertensive and a respiratory disease clinical trial, where multiple measurements of the primary response were made for all participants at each visit.

Electroacupuncture for control of myeloablative chemotherapy-induced emesis: A randomized controlled trial

CONTEXT

High-dose chemotherapy poses considerable challenges to emesis management. Although prior studies suggest that acupuncture may reduce nausea and emesis, it is unclear whether such benefit comes from the nonspecific effects of attention and clinician-patient interaction.

OBJECTIVE

To compare the effectiveness of electroacupuncture vs minimal needling and mock electrical stimulation or antiemetic medications alone in controlling emesis among patients undergoing a highly emetogenic chemotherapy regimen.

DESIGN

Three-arm, parallel-group, randomized controlled trial conducted from March 1996 to December 1997, with a 5-day study period and a 9-day follow-up.

SETTING

Oncology center at a university medical center.

PATIENTS

One hundred four women (mean age, 46 years) with high-risk breast cancer.

INTERVENTIONS

Patients were randomly assigned to receive low-frequency electroacupuncture at classic antiemetic acupuncture points once daily for 5 days (n = 37); minimal needling at control points with mock electrostimulation on the same schedule (n = 33); or no adjunct needling (n = 34). All patients received concurrent triple antiemetic pharmacotherapy and high-dose chemotherapy (cyclophosphamide, cisplatin, and carmustine).

MAIN OUTCOME MEASURES

Total number of emesis episodes occurring during the 5-day study period and the proportion of emesis-free days, compared among the 3 groups.

RESULTS

The number of emesis episodes occurring during the 5 days was lower for patients receiving electroacupuncture compared with those receiving minimal needling or pharmacotherapy alone (median number of episodes, 5, 10, and 15, respectively; P<.001). The electroacupuncture group had fewer episodes of emesis than the minimal needling group (P<.001), whereas the minimal needling group had fewer episodes of emesis than the antiemetic pharmacotherapy alone group (P =.01). The differences among groups were not significant during the 9-day follow-up period (P =.18).

CONCLUSIONS

In this study of patients with breast cancer receiving high-dose chemotherapy, adjunct electroacupuncture was more effective in controlling emesis than minimal needling or antiemetic pharmacotherapy alone, although the observed effect had limited duration. JAMA. 2000;284:2755-2761.

Modeling repeated count data subject to informative dropout

In certain diseases, outcome is the number of morbid events over the course of follow-up. In epilepsy, e.g., daily seizure counts are often used to reflect disease severity. Follow-up of patients in clinical trials of such diseases is often subject to censoring due to patients dying or dropping out. If the sicker patients tend to be censored in such trials, estimates of the treatment effect that do not incorporate the censoring process may be misleading. We extend the shared random effects approach of Wu and Carroll (1988, Biometrics 44, 175-188) to the setting of repeated counts of events. Three strategies are developed. The first is a likelihood-based approach for jointly modeling the count and censoring processes. A shared random effect is incorporated to introduce dependence between the two processes. The second is a likelihood-based approach that conditions on the dropout times in adjusting for informative dropout. The third is a generalized estimating equations (GEE) approach, which also conditions on the dropout times but makes fewer assumptions about the distribution of the count process. Estimation procedures for each of the approaches are discussed, and the approaches are applied to data from an epilepsy clinical trial. A simulation study is also conducted to compare the various approaches. Through analyses and simulations, we demonstrate the flexibility of the likelihood-based conditional model for analyzing data from the epilepsy trial.

A transitional model for longitudinal binary data subject to nonignorable missing data

Binary longitudinal data are often collected in clinical trials when interest is on assessing the effect of a treatment over time. Our application is a recent study of opiate addiction that examined the effect of a new treatment on repeated urine tests to assess opiate use over an extended follow-up. Drug addiction is episodic, and a new treatment may affect various features of the opiate-use process such as the proportion of positive urine tests over follow-up and the time to the first occurrence of a positive test. Complications in this trial were the large amounts of dropout and intermittent missing data and the large number of observations on each subject. We develop a transitional model for longitudinal binary data subject to nonignorable missing data and propose an EM algorithm for parameter estimation. We use the transitional model to derive summary measures of the opiate-use process that can be compared across treatment groups to assess treatment effect. Through analyses and simulations, we show the importance of properly accounting for the missing data mechanism when assessing the treatment effect in our example.

Therapeutic Strategies for Localized Prostate Cancer II: Perineal Prostatectomy, X-Rays, Protons, Neutrons, and Combination Brachytherapy

Application of improved imaging, diagnostic, and computer techniques is beginning to have an impact on the management of localized prostate cancer. It is possible to perform a range of surgical and radiation procedures with less morbidity than in the past. The changes in therapy for patients with localized disease derive from better knowledge of anatomy for invasive procedures and optimization of virtual planning for noninvasive methods. Perineal prostatectomy and combinations of beam and seed radiation offer both patient and physician reasonable therapeutic options.

Latent model for correlated binary data with diagnostic error

We propose a methodology for modeling correlated binary data measured with diagnostic error. A shared random effect is used to induce correlations in repeated true latent binary outcomes and in observed responses and to link the probability of a true positive outcome with the probability of having a diagnosis error. We evaluate the performance of our proposed approach through simulations and compare it with an ad hoc approach. The methodology is illustrated with data from a study that assessed the probability of corneal arcus in patients with familial hypercholesterolemia.

A mover-stayer model for longitudinal marker data

Studies of chronic disease often focus on estimating prevalence and incidence in which the presence of active disease is based on dichotomizing a continuous marker variable measured with error. Examples include hypertension, asthma, and depression, where active disease is defined by setting a threshold on a continuous measure of blood pressure, respiratory function, and mood, respectively. This paper proposes a model for inference about prevalence and incidence when active disease is determined by dichotomizing a continuous marker variable in a population-based study. In this formulation, it is postulated that there are three groups of people, those that are not susceptible to the disease, those who are always in the disease state, and those who have the potential to transition between the disease and the disease-free states over time. The model is used to estimate the prevalence and incidence of the disease in the population while accounting for measurement error in the marker. An EM algorithm is used for parameter estimation and the methodology is illustrated on Framingham heart study hypertension data. A simulation study is conducted in order to demonstrate the importance of accounting for measurement error in estimating prevalence and incidence for this example.

Longitudinal data analysis (repeated measures) in clinical trials

Longitudinal data is often collected in clinical trials to examine the effect of treatment on the disease process over time. This paper reviews and summarizes much of the methodological research on longitudinal data analysis from the perspective of clinical trials. We discuss methodology for analysing Gaussian and discrete longitudinal data and show how these methods can be applied to clinical trials data. We illustrate these methods with five examples of clinical trials with longitudinal outcomes. We also discuss issues of particular concern in clinical trials including sequential monitoring and adjustments for missing data. A review of current software for analysing longitudinal data is also provided. Published in 1999 by John Wiley & Sons, Ltd. This article is a US Government work and is the public domain in the United States.

Bayesian monitoring of event rates with censored data

A Bayesian approach to monitoring event rates with censored data is proposed. A Dirichlet prior for discrete time event probabilities is blended with discrete survival times to provide a posterior distribution that is a mixture of Dirichlets. Approximation of the posterior distribution via data augmentation is discussed. Practical issues involved in implementing the procedure are discussed and illustrated with a simulation of the single arm Cord Blood Transplantation Study where 6-month survival is monitored.

Repeated probit regression when covariates are measured with error

This paper develops a model for repeated binary regression when a covariate is measured with error. The model allows for estimating the effect of the true value of the covariate on a repeated binary response. The choice of a probit link for the effect of the error-free covariate, coupled with normal measurement error for the error-free covariate, results in a probit model after integrating over the measurement error distribution. We propose a two-stage estimation procedure where, in the first stage, a linear mixed model is used to fit the repeated covariate. In the second stage, a model for the correlated binary responses conditional on the linear mixed model estimates is fit to the repeated binary data using generalized estimating equations. The approach is demonstrated using nutrient safety data from the Diet Intervention of School Age Children (DISC) study.

Two pleiotropic classes of daf-2 mutation affect larval arrest, adult behavior, reproduction and longevity in Caenorhabditis elegans

The nematode Caenorhabditis elegans responds to overcrowding and scarcity of food by arresting development as a dauer larva, a nonfeeding, long-lived, stress-resistant, alternative third-larval stage. Previous work has shown that mutations in the genes daf-2 (encoding a member of the insulin receptor family) and age-1 (encoding a PI 3-kinase) result in constitutive formation of dauer larvae (Daf-c), increased adult longevity (Age), and increased intrinsic thermotolerance (Itt). Some daf-2 mutants have additional developmental, behavioral, and reproductive defects. We have characterized in detail 15 temperature-sensitive and 1 nonconditional daf-2 allele to investigate the extent of daf-2 mutant defects and to examine whether specific mutant traits correlate with each other. The greatest longevity seen in daf-2 mutant adults was approximately three times that of wild type. The temperature-sensitive daf-2 mutants fell into two overlapping classes, including eight class 1 mutants, which are Daf-c, Age, and Itt, and exhibit low levels of L1 arrest at 25.5 degrees. Seven class 2 mutants also exhibit the class 1 defects as well as some or all of the following: reduced adult motility, abnormal adult body and gonad morphology, high levels of embryonic and L1 arrest, production of progeny late in life, and reduced brood size. The strengths of the Daf-c, Age, and Itt phenotypes largely correlated with each other but not with the strength of class 2-specific defects. This suggests that the DAF-2 receptor is bifunctional. Examination of the null phenotype revealed a maternally rescued egg, L1 lethal component, and a nonconditional Daf-c component. With respect to the Daf-c phenotype, the dauer-defective (Daf-d) mutation daf-12(m20) was epistatic to daf-2 class 1 alleles but not the severe class 2 alleles tested. All daf-2 mutant defects were suppressed by the daf-d mutation daf-16(m26). Our findings suggest a new model for daf-2, age-1, daf-12, and daf-16 interactions.

A two-state Markov chain for heterogeneous transitional data: a quasi-likelihood approach

Many chronic diseases are measured by repeated binary data where the scientific interest is on the transition process between two states of disease activity. Examples include: depression; schizophrenia; multiple sclerosis, and respiratory illness. The course for many of these diseases is inherently heterogeneous, making it difficult to make inference on the transition process. This paper presents a model that incorporates heterogeneity by allowing the transition probabilities to vary randomly across subjects. In the proposed quasi-likelihood formulation for a two-state Markov chain, only the first two moments of the bivariate distribution on the transition probabilities are specified, and we develop a generalized estimating equations (GEE) approach for estimating the mean and variance of this distribution. In addition to estimating the model parameters, we discuss the estimation of derived quantities of the transition matrix such as estimating the expected first passage times and we discuss how we can introduce covariate dependence into the model. We use this methodology to summarize the transitioning pattern of respiratory illness in a group of children with intra-uteral growth retardation, and we conduct a simulation to investigate the finite sample properties of our procedure and to demonstrate marked bias if heterogeneity is ignored.

Characterization of MRI response to treatment with interferon beta-1b: contrast-enhancing MRI lesion frequency as a primary outcome measure

MRI is a valuable tool to examine the pathophysiology and natural history of multiple sclerosis (MS), and several large multicenter trials have utilized MRI as a secondary outcome measures. We previously examined the effect of interferon beta-1b on contrast-enhancing lesions on MRI using a baseline versus treatment design, and found that on treatment there is a reduction in mean frequency of enhancing lesions over the group. Using an expanded number of patients and the same trial design, we examined the individual response to treatment more extensively. We find that the effect seen previously is still present, and that there is heterogeneity in the amount of decrease in contrast-enhancing lesions. This expanded number of patients and trial design allows for the discussion of new criteria for individual response to treatment, which are applied in the current trial. These approaches may be useful in the examination, early testing, and comparison of experimental therapeutic agents in MS as well as in the characterization of patients who do or do not have a response seen on MRI.

A generalized estimating equation approach for modeling random length binary vector data

A common measure in clinical trials and epidemiologic studies is the number of events such as seizures, hospitalizations, or bouts of disease. Frequently, a binary measure of severity for each event is available but is not incorporated in the analysis. This paper proposes methodology for jointly modeling the number of events and the vector of correlated binary severity measures. Our formulation exploits the notion that a given covariate may affect both outcomes in a similar way. We functionally link the regression parameters for the counts and binary means and discuss a generalized estimating equation (GEE) approach for parameter estimation. We discuss conditions under which the proposed joint modeling approach provides marked gains in efficiency relative to the common procedure of simply modeling the counts, and we illustrate the methodology with epilepsy clinical trial data.

A latent process regression model for spatially correlated count data

This paper proposes a regression model for spatially correlated count data that generalizes the work of Zeger (1988, Biometrika 75, 621-629) developed in a time-series setting. In this approach, spatial correlation is introduced through a latent process, and the marginal mean function may contain spatial trends and covariates. Generalized estimating equations are used to estimate and perform marginal inference on the spatial trend and covariate effects. The feasibility of this approach is demonstrated using an example of the distribution of neuronal cell counts in a laboratory culture dish.

Control of C. elegans larval development by neuronal expression of a TGF-beta homolog

The Caenorhabditis elegans dauer larva is specialized for dispersal without growth and is formed under conditions of overcrowding and limited food. The daf-7 gene, required for transducing environmental cues that support continuous development with plentiful food, encodes a transforming growth factor-beta (TGF-beta) superfamily member. A daf-7 reporter construct is expressed in the ASI chemosensory neurons. Dauer-inducing pheromone inhibits daf-7 expression and promotes dauer formation, whereas food reactivates daf-7 expression and promotes recovery from the dauer state. When the food/pheromone ratio is high, the level of daf-7 mRNA peaks during the L1 larval stage, when commitment to non-dauer development is made.

Environmental induction and genetic control of surface antigen switching in the nematode Caenorhabditis elegans

Nematodes can alter their surface coat protein compositions at the molts between developmental stages or in response to environmental changes; such surface alterations may enable parasitic nematodes to evade host immune defenses during the course of infection. Surface antigen switching mechanisms are presently unknown. In a genetic study of surface antigen switching, we have used a monoclonal antibody, M37, that recognizes a surface antigen on the first larval stage of the free-living nematode Caenorhabditis elegans. We demonstrate that wild-type C. elegans can be induced to display the M37 antigen on a later larval stage by altering the growth conditions. Mutations that result in nonconditional display of this antigen on all four larval stages fall into two classes. One class defines the new gene srf-6 II. The other mutations are in previously identified dauer-constitutive genes involved in transducing environmental signals that modulate formation of the dauer larva, a developmentally arrested dispersal stage. Although surface antigen switching is affected by some of the genes that control dauer formation, these two process can be blocked separately by specific mutations or induced separately by environmental factors. Based on these results, the mechanisms of nematode surface antigen switching can now be investigated directly.

Relationship between sleep and mood in patients with rapid-cycling bipolar disorder

The relationship between sleep and mood was examined in a longitudinal, naturalistic data set derived from out-patients with rapid-cycling bipolar disorder. Eleven patients completed daily self-ratings of mood and sleep logs for 18 months. Using logistic regression with autoregressive terms, we examined the effect of prior sleep (sleep duration, time of sleep onset, and time of wake onset) on the probability of being in a depressed, manic, or hypomanic episode on one or more subsequent days. Of the three sleep parameters, decreased sleep duration was the best predictor of mania or hypomania the next day, followed by wake onset time. The association between sleep duration and subsequent mood was less consistent for depression than for mania or hypomania. Four of the patients showed no relationship between mood and any of the sleep variables measured. These results reinforce the importance of monitoring, and perhaps controlling, sleep duration and wake onset time in at least some patients with rapid-cycling bipolar disorder.

Guidelines for the use of magnetic resonance techniques in monitoring the treatment of multiple sclerosis. US National MS Society Task Force

Because of the major difficulties in measuring clinical end points in multiple sclerosis (MS) treatment trials, there has been much enthusiasm for using magnetic resonance imaging (MRI) findings as an alternative outcome. To provide international consensus guidelines for the use of MRI in MS clinical trials, a task force of the US National MS Society was convened. The recommendations of the task force are presented in this review. Given the high sensitivity for detecting pathological activity in relapsing-remitting and secondary progressive MS, monthly T2-weighted and gadolinium-enhanced brain MRI is an excellent tool for short-term exploratory trials of new agents where it serves as the primary end point; in particular, failure to demonstrate a reduction in lesion activity avoids the time, cost, and risks of a larger clinical end point study. However, conventional MRI findings have a limited correlation with disability in established MS. The primary end point of a definitive trial should therefore be clinical, although serial MRI at 6- to 12-month intervals is a useful secondary end point in providing an index of pathological progression. In trials of patients presenting with clinically isolated syndromes suggestive of MS, MRI findings can be used in the entry criteria, and as a secondary outcome measure, but conversion to clinically definite MS should be the primary outcome. The pathological substrates of irreversible disability are demyelination and axonal loss. Putative magnetic resonance markers for these processes include decreased N-acetylaspartate on proton magnetic resonance spectroscopy, decreased magnetization transfer ratios, hypointensity on T1-weighted images, and loss of short T2 water fractions, some of which relate more closely to disability than conventional MRI findings. Further technical developments should lead to more accurate quantitation, greater pathological specificity, and stronger clinical correlations.

Changes in the amount of diseased white matter over time in patients with relapsing-remitting multiple sclerosis

MRI is a sensitive technique for assessing disease activity in MS. Diseased white matter (WM) can be identified on T2-weighted images, and active disease is reflected by abnormalities in the blood-brain barrier (BBB) shown on T1-weighted images after administration of paramagnetic contrast agents. Active disease may be demonstrated by contrast-enhanced MRI in patients with early, mild relapsing-remitting (RR) MS even during periods of clinical stability, which indicates that MS is an active process even during the early phase of the illness. To examine the amount of abnormal WM at frequent intervals over time, we studied seven mildly affected RRMS patients, all of whom had frequent contrast-enhancing lesions. These RRMS patients were imaged monthly for 26 to 36 months at 1.5 tesla; the area of abnormal increased WM signal was calculated by image-processing software that utilizes both the T2- and T1-weighted images. All patients showed fluctuations over time in amount of abnormal WM signal, which reflected factors such as the amount of BBB breakdown (measured by number or area of enhancing lesions) and measurement error. All seven RRMS patients, however, showed an overall increase in abnormal WM. Because of the fluctuations between individual measurements, the increase was most accurately reflected when the mean of the first 6 months' measurements was compared with the mean of the final 6 months' measurements, or when a linear regression model was applied.(ABSTRACT TRUNCATED AT 250 WORDS)

A generalized estimating equations approach for spatially correlated binary data: applications to the analysis of neuroimaging data

This paper proposes a generalized estimating equations approach for the analysis of spatially correlated binary data when there are large numbers of spatially correlated observations on a moderate number of subjects. This approach is useful when the scientific focus is on modeling the marginal mean structure. Proper modeling of the spatial correlation structure is shown to provide large efficiency gains along with precise standard error estimates for inference on mean structure parameters. Generalized estimating equations for estimating the parameters of both the mean and spatial correlation structure are proposed. The use of semivariogram models for parameterizing the correlation structure is discussed, and estimation of the sample semivariogram is proposed as a technique for choosing parametric models and starting values for generalized estimating equations estimation. The methodology is illustrated with neuroimaging data collected as part of the National Institute of Neurological Disorders and Stroke (NINDS) Stroke Data Bank. A simulation study demonstrates the importance of accurate modeling of the spatial correlation structure in data with large numbers of spatially correlated observations such as those found in neuroimaging studies.

Blood-brain barrier disruption on contrast-enhanced MRI in patients with mild relapsing-remitting multiple sclerosis: relationship to course, gender, and age

MRI has provided insight into the pathophysiology and course of MS, particularly through the use of a paramagnetic contrast agent that allows visualization of blood-brain barrier (BBB) breakdown. Neither the overall frequency of BBB breakdown in MS patients nor the characteristics associated with BBB breakdown in MS are known. We studied 68 relapsing-remitting MS (RRMS) patients with three monthly MRIs to examine these questions. Seventy-eight percent of the RRMS patients studied had evidence of BBB breakdown on at least one MRI. While there was a great deal of variability among patients in terms of mean enhancing lesion frequency, BBB breakdown was associated with younger age at onset of disease, measured by age at first symptom or age at diagnosis, and more severe disease as measured by Expanded Disability Status Scale scores equal to or greater than 4.0. We found no relationship between BBB breakdown and duration of disease or gender. We conclude that BBB breakdown is a relatively common phenomenon in RRMS patients and may be most commonly found in patients with more aggressive disease and younger onset. These findings have implications for clinical trials that use MRI as an outcome measure.