An optimal three-stage design for phase II clinical trials

A phase II clinical trial in cancer therapeutics is usually a single-arm study to determine whether an experimental treatment (E) holds sufficient promise to warrant further testing. When the criterion of treatment efficacy is a binary endpoint (response/no response) with probability of response p, we propose a three-stage optimal design for testing H₀: p ≤ p₀ versus H₁: p ≥ p₁, where p₁ and p₀ are response rates such that E does or does not merit further testing at given levels of statistical significance (α) and power (1 ? β). The proposed design is essentially a combination of earlier proposals by Gehan and Simon. The design stops with rejection of H₁ at stage 1 when there is an initial moderately long run of consecutive treatment failures; otherwise there is continuation to stage 2 and (possibly) stage 3 which have decision rules analogous to those in stages 1 and 2 of Simon's design. Thus, rejection of H₁ is possible at any stage, but acceptance only at the final stage. The design is optimal in the sense that expected sample size is minimized when p = p₀, subject to the practical constraint that the minimum stage 1 sample size is at least 5. The proposed design has greatest utility when the true response rate of E is small, it is desirable to stop early if there is a moderately long run of early treatment failures, and it is practical to implement a three-stage design. Compared to Simon's optimal two-stage design, the optimal three-stage design has the following features: stage 1 is the same size or smaller and has the possibility of stopping earlier when 0 successes are observed; the expected sample size under the null hypothesis is smaller; stages 1 and 2 generally have more patients than stage 1 of the two-stage design, but a higher probability of early termination under H₀; and the total sample size and criteria for rejection of H₁ at stage 3 are similar to the corresponding values at the end of stage 2 in the two-stage optimal design.     

Gestational age and intrauterine growth retardation among white and black very low birthweight infants: a population-based cohort study

Summary. Very low birthweight (VLBW) is a commonly used endpoint in perinatal epidemiology, but the population of VLBW infants comprises a wide range of gestational ages and rates of fetal growth. We used data from a population-based study of all 1072 black and white VLBW liveborn infants born in 29 counties in Georgia between April 1986 and March 1988. Less than 1% of the VLBW infants were ≥ 37 weeks gestation; most were 29–32 weeks (26%) or 25 to 28 weeks (40%); 12% were 22 weeks or less. All infants 33 weeks gestation or greater were growth retarded. The population of VLBW infants seems to comprise three groups: approximately 11% very immature infants of 22 weeks or less; the majority of infants, born between 23 and 30 weeks, 90% of which are of normal weight for their gestational age; and a group of less premature, growth-retarded infants from 31 to 36 weeks. We found little or no difference in the distribution of gestational age or the percentage of intrauterine growth rates (IUGR) between black and white infants. In the USA the VLBW rate among black infants is over three times greater than that among white infants and consequently the rates of the three types of VLBW among black infants are likely to be triple those among white infants.     

Low peripheral plasma renin activity as a critical marker in pediatric hypertension

In evaluating hypertensive children and adolescents, the etiological considerations should include a set of inherited disorders that share very low plasma renin activity (PRA) as a common feature. In particular among these disorders, glucocorticoid remediable aldosteronism (GRA) appears to be emerging as an important etiology of hypertension in the pediatric population. We report the evaluation of a 9-year-old Caucasian girl who presented with severe hypertension and a strong family history of early-onset hypertension. Her suppressed PRA, her family history, and her failure to respond to conventional antihypertensive therapy raised GRA as a potential etiology. The diagnosis was confirmed by an elevated ratio of urinary 18-oxotetrahydrocortisol to urinary tetrahydroaldosterone and genetic testing, which demonstrated the chimeric gene duplication. The molecular pathogenesis of GRA and the clinical implications are reviewed. Received May 15, 1996; received in revised form and accepted September 16, 1996     

Choline: Dietary Requirements and Role in Brain Development

Choline is needed for the maintenance of the structural integrity and signaling functions of cell membranes, for neurotransmission, and for transport of lipids and as a source of methyl groups. Choline can be made de novo in the body, but some individuals must also obtain choline in the diet to prevent deficiency symptoms. A number of environmental and genetic factors influence dietary requirements for choline, and average intakes in the population vary widely. Therefore, certain individuals may be at greater risk of choline deficiency. Choline is critical during fetal development, particularly during the development of the brain, where it can influence neural tube closure and lifelong memory and learning functions.     

Delineating the sites and progression of in vivo atrophy in multiple system atrophy using fluid-registered MRI.

We describe the pattern and progression of atrophy delineated using fluid registration of serial magnetic resonance imaging scans in a case of multiple system atrophy (MSA). The in vivo findings were consistent with those found at postmortem, including significant supratentorial atrophy concurrent with an unusual degree of cognitive impairment for MSA.     

Multiplex PCR analysis of in vivo-arising deletion mutations in the hprt gene of human T-lymphocytes

A multiplex polymerase chain reaction (PCR) procedure was adapted for the rapid and efficient evaluation of deletions of the hypoxanthine guanine phosphoribosyltransferase (hprt) gene in human T-lymphocytes. The hprt clonal assay was used to isolate in vivo-arising hprt-deficient T-cells from six healthy males. Mutant frequencies ranged from 9-27 × 10^?6. Simple crude cellular extracts from 223 mutants were analyzed for hprt gene deletion. Sixteen (7.2%) were found to be due to total gene deletion and 22 (9.9%) were due to partial gene deletion. The relatively high frequency of total gene deletions was caused by replicate isolates of a single mutational event as shown by single-strand conformation polymorphism (SSCP) analysis of rearranged T-cell receptor (TCR)-γ genes. Eighteen of the 22 partial hprt gene deletion mutants were determined to be of independent origin based on a unique hprt mutation or SSCP-TCR-γ pattern. One-half (9/18) of the partial deletion mutants involved all or part of exon 4 alone, suggesting that this region of the hprt gene is prone to deletion. The small deletions effecting exon 1 (1 mutant), exon 2 (2 mutants), and exon 4 (6 mutants) would not have been detected by conventional Southern blot analysis and may represent a new, previously unrecognized class of mutations. The ready isolation of such intragenic deletions will allow the characterization of breakpoint junctions and may provide insights into the important processes of DNA breakage and rejoining. © 1994 Wiley-Liss, Inc.     

Cloning and expression analysis of the chick DAN gene, an antagonist of the BMP family of growth factors.

Differential screening-selected gene aberrative in neuroblastoma (DAN) is a member of a cystine knot protein family that includes Cerberus and Gremlin. First isolated in a screen to identify genes down-regulated in transformed rat fibroblasts, DAN has subsequently been cloned in Xenopus, mouse, and human. Overexpression of DAN suppresses the transformed phenotype and retards the cell's entry into S phase. Biochemical analyses have demonstrated DAN's ability to bind bone morphogenetic proteins and antagonize their signaling activity. In this study, chick DAN was cloned and sequenced, revealing a conserved cystine knot region as well as an N-glycosylation site. A riboprobe was designed from the 3' chick DAN coding sequence and used for analysis of DAN in the developing chick embryo by in situ hybridization. Chick DAN was expressed beginning at stage 10 in the developing somites and the medial otic epithelium. Expression in the neural layer of the eye became apparent at stage 14. By stage 17, expression had expanded to the base of the hindbrain. Limb bud labeling began at stage 20, whereas expression in the branchial arches appeared at stage 25. Chick DAN expression generally corresponded to that of mouse DAN expression as shown by comparative in situ hybridization. However, chick DAN was found in the otic epithelium and notochord, whereas mouse DAN was restricted to the overlying otic ectomesenchyme and was absent from the notochord. This observation suggests that DAN may play different roles in chick and mouse otic and notochord development.     

In vivo vs in vitro anticariogenic behavior of glass-ionomer and resin composite restorative materials.

OBJECTIVE: To evaluate the in vivo vs the in vitro anticariogenic potential of glass-ionomer and resin composite restoratives, utilizing a standardized interfacial gap model. METHODS: (a) In vitro study. Box shaped cavities were prepared at the buccal surfaces of extracted premolars limited to enamel. The incisal cavity walls received no treatment and were covered with 40 microm-thick metal spacers. The cavities were restored with a glass-ionomer (Ketac-Fil, n=8) and a fluoride-free resin composite (Scotchbond MP Plus/Z100, n=8). After 4 weeks immersion in an acidic gel (pH 4), thin sections were produced and examined under polarized-light microscopy.(b) In vivo study. Four low caries activity volunteers, with first four premolars, each planned to be extracted for orthodontic reasons, participated in the study. Cavities were prepared as before and filled contralaterally per patient with glass-ionomer (n=8) and resin composite (n=8). After 6 months in vivo, the teeth were extracted, sectioned and investigated by polarized-light microscopy, Raman microspectroscopy and SEM-EDS X-ray microanalysis. Unpaired t-test (lesion dimensions) and one-way ANOVA and Newman-Keuls tests (Ca, P wt%, Ca/P ratios) were used to identify statistically significant differences in lesion analysis (alpha=0.05). RESULTS: (a) In vitro study. All restorations developed lesions at incisal and cervical margins. At gap-free regions glass-ionomers showed reduced lesion dimensions compared to those of composites (p<0.05). At regions with gaps, no significant differences were found in lesion depth between the restorative groups tested. Lesion length was increased in composite, and decreased in glass-ionomer, whereas lesion depth in both restorative groups was increased in comparison to gap-free regions (p<0.05).(b) In vivo study. No lesions were observed at gap-free regions. At gap regions, 75.5% of glass-ionomer and 62.5% of composite restorations developed lesions. The lesion dimensions were significantly greater in glass-ionomer (p<0.05). A reduction in PO4(3-), CO3(2-), Ca and P was found in lesions compared to intact tissues. No F was detected and no CaF2 lattice vibrations were found at the enamel margins facing the gap adjacent to glass-ionomers. SIGNIFICANCE: In the presence of a standardized interfacial gap, no preventive effect was exerted in vivo from the glass-ionomer to protect the adjacent enamel wall from secondary caries attack. The lack of any correlation between the in vivo and in vitro models tested implies that artificial caries experiments have a negligible clinical relevance in predicting the in vivo effect.