Comparison of a neural net-based QSAR algorithm (PCANN) with Hologram- and multiple linear regression-based QSAR approaches: application to 1,4-dihydropyridine-based calcium channel antagonists

A QSAR algorithm (PCANN) has been developed and applied to a set of calcium channel blockers which are of special interest because of their role in cardiac disease and also because many of them interact with P-glycoprotein, a membrane protein associated with multidrug resistance to anticancer agents. A database of 46 1,4-dihydropyridines with known Ca2+ channel binding affinities was employed for the present analysis. The QSAR algorithm can be summarized as follows: (1) a set of 90 graph theoretic and information theoretic descriptors representing various structural and topological characteristics was calculated for each of the 1,4-dihydropyridines and (2) principal component analysis (PCA) was used to compress these 90 into the eight best orthogonal composite descriptors for the database. These eight sufficed to explain 96% of the variance in the original descriptor set. (3) Two important empirical descriptors, the Leo-Hansch lipophilic constant and the Hammet electronic parameter, were added to the list of eight. (4) The 10 resulting descriptors were used as inputs to a back-propagation neural network whose output was the predicted binding affinity. (5) The predictive ability of the network was assessed by cross-validation. A comparison of the present approach with two other QSAR approaches (multiple linear regression using the same variables and a Hologram QSAR model) is made and shows that the PCANN approach can yield better predictions, once the right network configuration is identified. The present approach (PCANN) may prove useful for rapid assessment of the potential for biological activity when dealing with large chemical libraries.

Shared decision-making and the orthopaedic workforce

Studies of physician workforce need a standard of an appropriately sized workforce to compare projections. Although many studies use average rates of healthcare use as a standard, regional benchmarks provide a pragmatic alternative approach to estimating a reasonably sized physician workforce and avoid many of the problems of needs- and demand-based planning. Wide geographic variations in the rates of many procedures, unexplained by differences in population characteristics, suggest that supply-induced demand or physician practice style or both may be the major determinates of the rates for these procedures. In the current study, the authors explore some of these differences in orthopaedic procedure rates and their implications for workforce planning. For example, the rates of hip fracture are fairly uniform across geographic regions, whereas the rates of spine surgery vary sixfold and the rates of spinal fusion vary 10-fold. Shared decision-making is the process of giving patients informed choices about their treatment options based on current best evidence. Careful studies of treatment effectiveness and shared decision-making hold the promise of allowing patients' preferences and values to determine the right rate of healthcare use. These rates could allow workforce projections to be compared with optimal benchmarks for future planning.

Helping patients decide about back surgery: a randomized trial of an interactive video program

STUDY DESIGN

A randomized trial of 100 patients with low back pain who were potential surgical candidates.

OBJECTIVES

To determine whether an interactive videodisc with a booklet is superior to a booklet alone for informing patients about back surgery.

SUMMARY OF BACKGROUND DATA

Substantial geographic variation has been observed in lumbar spine surgery. Informed patient preferences should play an important role in decisions about surgery. However, little is known about optimal strategies for informing patients.

METHODS

Subjects were randomized to receive an interactive videodisc (with a booklet) or a booklet alone. A knowledge test administered at baseline and follow-up was used to measure improvement in knowledge about treatment options for lumbar spine problems. Patients' reactions to the videodisc and booklet and preferences for treatment were also assessed.

RESULTS

The patients' knowledge improved after exposure to either intervention. Multivariate analyses adjusted for baseline score, age, education, gender, and diagnosis showed a significant advantage for the videodisc with booklet over the booklet alone. The videodisc-booklet group showed significantly greater gains in knowledge among subjects with the worst baseline knowledge scores. A larger proportion of subjects in the videodisc-booklet group rated the materials easy to understand (93% vs.- 72%,P = 0.04), containing the right amount of information (93% vs.- 80%,P = 0.3), and adequate to assist in choice of treatment (75% vs.- 51%,P = 0.2). Those who viewed the videodisc expressed a somewhat lower preference for surgery than those who received the booklet alone (23% vs.- 42%,P = 0.4).

CONCLUSIONS

Both the booklet alone and the combination of videodisc and booklet improved knowledge. The combination produced greater knowledge gains than the booklet alone for the subgroup with the least knowledge at baseline. Patients preferred the combination and had a slightly lower preference for surgery if they had viewed the video presentation. For some patients, the video may enhance involvement in clinical decisions.

Searching for pharmacogenomic markers: the synergy between omic and hypothesis-driven research

With 35,000 genes and hundreds of thousands of protein states to identify, correlate, and understand, it no longer suffices to rely on studies of one gene, gene product, or process at a time. We have entered the "omic" era in biology. But large-scale omic studies of cellular molecules in aggregate rarely can answer interesting questions without the assistance of information from traditional hypothesis-driven research. The two types of science are synergistic. A case in point is the set of pharmacogenomic studies that we and our collaborators have done with the 60 human cancer cell lines of the National Cancer Institute's drug discovery program. Those cells (the NCI-60) have been characterized pharmacologically with respect to their sensitivity to >70,000 chemical compounds. We are further characterizing them at the DNA, RNA, protein, and functional levels. Our major aim is to identify pharmacogenomic markers that can aid in drug discovery and design, as well as in individualization of cancer therapy. The bioinformatic and chemoinformatic challenges of this study have demanded novel methods for analysis and visualization of high-dimensional data. Included are the color-coded "clustered image map" and also the MedMiner program package, which captures and organizes the biomedical literature on gene-gene and gene-drug relationships. Microarray transcript expression studies of the 60 cell lines reveal, for example, a gene-drug correlation with potential clinical implications--that between the asparagine synthetase gene and the enzyme-drug L-asparaginase in ovarian cancer cells.

The impact of spinal problems on the health status of patients: have we underestimated the effect?

STUDY DESIGN

A prospective study of 17,774 patients who consulted spine centers in which the impact of spinal disorders and comorbidities on physical functional status were evaluated.

OBJECTIVES

To quantify the effect spinal diagnoses have on patients' physical functional status (SF-36 Physical Component Summary [PCS] score) compared with other common conditions and to quantify the effects of comorbidities on physical functional status in spine patients.

SUMMARY OF BACKGROUND DATA

The burden of spinal conditions on a patient's function and the role that comorbidities play in this affliction are poorly quantified in the literature.

METHODS

Data from the Health Survey Questionnaire were prospectively gathered through the National Spine Network, a nonprofit consortium of spine-focused practices. Each patient's SF-36 score was summarized into a single PCS score. The correlation between diagnosis and comorbidity and PCS score was assessed using multivariate linear regression.

RESULTS

The study patients were a mean of 47.5 years of age, 54.7% were female, 52.3% had lumbosacral diagnoses, and 82.0% had had 3 or more months of pain. The population had a mean PCS score of 30.4 +/- 9.95 (SD) compared with 50.0 +/- 10.00 for the general United States population. The more comorbidities in a patient, the lower the PCS score (Spearman rank correlation = -0.27). The five comorbid conditions that lowered the PCS the most included congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), renal failure, rheumatoid arthritis, and lupus (all P <0.001). In multiple linear regression analysis, age, gender, diagnosis, and comorbidity explained 12.1% of the variance in PCS score.

CONCLUSIONS

The PCS score is greatly affected in patients with spinal problems. The study population's PCS (30.4) was lower or similar to the PCS for patients with other illnesses reported in the literature: CHF (31.0), COPD (33.9), SLE (37.1), cancer (38.4), primary total hip arthroplasty (29.0), primary total knee arthroplasty (32.6), and glenohumeral degenerative joint disease (35.2). Further, the presence of comorbidity in spine patients adds to the burden of spinal conditions on functional status.

Spinal glial activation and cytokine expression after lumbar root injury in the rat

STUDY DESIGN

This study was designed to examine the behaviorial immunohistochemical changes of spinal glial cells and spinal Interleukin (IL)-1beta expression after various nerve root injuries used as models of lumbar radiculopathy.

OBJECTIVES

In order to better understand the role of central inflammation in the pathophysiologic mechanisms that give rise to pain associated with lumbar radiculopathy, this research studied the relationship between pain-related behavior associated with spinal glial activation and IL-1beta expression generated by three types of nerve root injury: loose ligation with chromic gut, loose ligation with silk, and tight ligation with silk.

SUMMARY OF BACKGROUND DATA

An animal model of lumbar radiculopathy originally described by Kawakami and Weinstein involved loose ligation of unilateral L4-L6 nerve roots with chromic gut. Characterization and establishment of such an animal model of low back pain enables further investigation of the nature of the pathophysiologic mechanisms associated with lumbar radiculopathy in humans.

METHODS

Seventy-three rats were divided into four treatment groups. Chromic group (n = 25): The L5 nerve roots (dorsal and ventral) were exposed by hemilaminectomy and loosely ligated with chromic gut. Tight silk group (n = 18): The exposed L5 nerve roots were tightly ligated extradurally with 5-0 silk suture. Loose silk group (n = 15): two loose ligatures of 5-0 silk were placed around the exposed L5 nerve roots. Sham group (n = 15): the rats were subjected to laminectomy alone for exposing nerve roots. Following surgery, thermal hyperalgesia and mechanical allodynia was assessed time-dependently up to 42 days post operatively. At 1, 3, 7, 14, and 42 days postoperatively, the rats in each group were perfused with fixative. The L5 spinal cord segments was harvested and cryosectioned for glial and cytokine immunohistochemistry.

RESULTS

In the chromic and the tight silk group, an immediate and sustained mechanical allodynia was observed in the ipsilateral hind paw up to 35 days postoperatively. The loose silk group also showed an immediate mechanical allodynia that subsided by 14 days postoperatively. Sham-treated animals exhibited mild mechanicalallodynia for the initial 7 days after the surgery. Thermalhyperalgesia was evident in the three primary treatment groups, but not in the sham-treated rats. OX-42 expression was elevated in the gray matter of the L5 spinal section by 3 days in the chromic, the tight silk, and the loose silk groups as compared to the sham group. Astrocytic activation increased over time in all groups except the sham group. There was no direct correlation between degree of microglial response and severity of pain behaviors. In contrast, astrocytic activation demonstrated a direct relationship with the elevation of mechanical allodynia for the first 7 days. In addition, spinal IL-1beta protein expression was increased bilaterally in the superficial layer of the dorsal horn and cell nuclei of the ventral horns in the ligature treated groups as compared with the sham group.

CONCLUSION

Direct mechanical and/or chemical injury to lumbar roots in the rat gives rise to pain behavior suggestive of lumbar radiculopathy. The finding that glial activation and enhanced IL-1beta expression are observed in the spinal cord after root injury supports a central, neuroimmune component in the generation of lumbar radiculopathy. A further understanding of the immunologic consequences of root injury may lead to further development and the novel use of selective cytokine-inflammatory inhibitors for the treatment of low back pain associated with radiculopathy.

Designing an ambulatory clinical practice for outcomes improvement: from vision to reality--the Spine Center at Dartmouth-Hitchcock, year one

Development of a new program for diagnosis and treatment of spine-related problems provided a unique opportunity to design and implement a new model for delivery of health care incorporating outcomes measurement and improvement. Key features include: application of microsystem thinking and interdisciplinary practice; integration of a uniform outcomes measurement tool, the Dartmouth Clinical Value Compass; and touch pad technology for data collection. This, for the first time, provided clinically meaningful point-of-service data and aggregated information for improvement. A further advantage was the ability to integrate a clinical research program within this microsystem. A multisite clinical research trial, the Spine Patient Outcomes Research Trial (SPORT), modeled on the Spine Center microsystem and funded by The National Institute of Arthritis, Musculoskeletal and Skin Diseases and the Office of Research on Woman's Health, the National Institutes of Health, and the National Institute of Occupational Safety and Health, the Centers for Disease Control and Prevention, is currently underway. The significant problems we face today cannot be solved by the same level of thinking that created them.

Cost-effectiveness of fusion with and without instrumentation for patients with degenerative spondylolisthesis and spinal stenosis

STUDY DESIGN

A cost-effectiveness study was performed from the societal perspective.

OBJECTIVE

To evaluate the costs and benefits of laminectomy alone and laminectomy with concomitant lumbar fusion for patients with degenerative lumbar spondylolisthesis and spinal stenosis.

SUMMARY OF BACKGROUND DATA

Costs, probabilities, and utilities were estimated from the literature. Short-term risks considered were perioperative complications, the probability of the fusion healing, and the probability that surgery will relieve symptoms. Long-term risks considered were recurrence of symptoms and reoperation.

METHODS

The 10-year costs, quality-adjusted life years, and incremental cost-effectiveness ratios (reported as dollars per quality-adjusted year of life gained) were calculated using a Markov model. Sensitivity analysis was performed on all variables using clinically plausible ranges.

RESULTS

Laminectomy with noninstrumented fusion costs $56,500 per quality-adjusted year of life versuslaminectomy without fusion. The cost-effectiveness of laminectomy with noninstrumented fusion was most sensitive to the increase in quality-of-life associated with relief of severe stenosis symptoms. The cost-effectiveness ratio of instrumented fusion compared with noninstrumented fusion was $3,112,800 per quality-adjusted year of life. However, if the proportion of patients experiencing symptom relief after instrumented fusion was 90% as compared with 80% for patients with noninstrumented fusion, then the cost-effectiveness ratio of instrumented fusion compared with noninstrumented fusion would be $82,400 per quality-adjusted year of life.

CONCLUSIONS

The cost-effectiveness of laminectomy with noninstrumented fusion compares favorably with other surgical interventions, although it depends greatly on the true effectiveness of these surgeries to alleviatesymptoms and on how patients value the quality-of-life effect of relieving severe stenosis symptoms. Instrumented fusion was very expensive compared with the incremental gain in health outcome. Better data on the effectiveness of these alternative procedures are needed.

Combination gene therapy: synergistic inhibition of human immunodeficiency virus Tat and Rev functions by a single RNA molecule

Current drug combinations can achieve long-term suppression of HIV replication in infected individuals. Unfortunately, complicated dosing schedules and high toxicity make long-term compliance with drug regimens difficult for most patients. Gene therapy may provide a permanent solution for HIV disease by generating cells genetically resistant to virus replication. As with the highly active antiretroviral therapies, genetic drugs must have strong antiviral potency and the ability to prevent the emergence of escape mutants. We have constructed antiviral genes containing unique combinations of Tat- and Rev-binding decoys. The new antiviral molecules are chimeric TAR-RRE RNAs that are expressed only in HIV infected cells in a Tat-regulated manner. One RNA molecule competes for both Tat and Rev binding, and thus blocks the activation and the expression of all viral genes. The two functional Tat- and Rev-binding domains exhibit the highest synergy at the lowest concentration. Conservative quantitative estimates of this synergistic effect were I = 0.24 at 50% inhibition, in terms of the Berenbaum "interaction index," indicating that the combined construct was approximately fourfold more potent than would be predicted on the basis of additive effects. The possibility of HIV escape from this inhibition is unlikely, because it requires simultaneous mutation of TAR and RRE in a manner in which both Tat and Rev preserve their respective functions. TAR-RRE combination decoys represent the first example of mathematically proven synergistic antiviral activity between two domains of the same molecule.

Systematic variation in gene expression patterns in human cancer cell lines

We used cDNA microarrays to explore the variation in expression of approximately 8,000 unique genes among the 60 cell lines used in the National Cancer Institute's screen for anti-cancer drugs. Classification of the cell lines based solely on the observed patterns of gene expression revealed a correspondence to the ostensible origins of the tumours from which the cell lines were derived. The consistent relationship between the gene expression patterns and the tissue of origin allowed us to recognize outliers whose previous classification appeared incorrect. Specific features of the gene expression patterns appeared to be related to physiological properties of the cell lines, such as their doubling time in culture, drug metabolism or the interferon response. Comparison of gene expression patterns in the cell lines to those observed in normal breast tissue or in breast tumour specimens revealed features of the expression patterns in the tumours that had recognizable counterparts in specific cell lines, reflecting the tumour, stromal and inflammatory components of the tumour tissue. These results provided a novel molecular characterization of this important group of human cell lines and their relationships to tumours in vivo.

A gene expression database for the molecular pharmacology of cancer

We used cDNA microarrays to assess gene expression profiles in 60 human cancer cell lines used in a drug discovery screen by the National Cancer Institute. Using these data, we linked bioinformatics and chemoinformatics by correlating gene expression and drug activity patterns in the NCI60 lines. Clustering the cell lines on the basis of gene expression yielded relationships very different from those obtained by clustering the cell lines on the basis of their response to drugs. Gene-drug relationships for the clinical agents 5-fluorouracil and L-asparaginase exemplify how variations in the transcript levels of particular genes relate to mechanisms of drug sensitivity and resistance. This is the first study to integrate large databases on gene expression and molecular pharmacology.

Molecular targets in cancer drug discovery: cell-based profiling

The phrase "molecular target-based drug discovery" usually implies an in vitro biochemical assay or battery of assays. One portion of the U.S. National Cancer Institute's drug discovery program, to the contrary, examines molecular targets for cancer therapy in a cell-based format. That approach has a number of significant limitations, but it has produced databases of significant utility on the activities and structures of tested compounds, as well as on molecular characteristics of the cell types used for testing.

Mining and visualizing large anticancer drug discovery databases

In order to find more effective anticancer drugs, the U.S. National Cancer Institute (NCI) screens a large number of compounds in vitro against 60 human cancer cell lines from different organs of origin. About 70,000 compounds have been tested in the program since 1990, and each tested compound can be characterized by a vector (i.e., "fingerprint") of 60 anticancer activity, or -[log(GI50)], values. GI50 is the concentration required to inhibit cell growth by 50% compared with untreated controls. Although cell growth inhibitory activity for a single cell line is not very informative, activity patterns across the 60 cell lines can provide incisive information on the mechanisms of action of screened compounds and also on molecular targets and modulators of activity within the cancer cells. Various statistical and artificial intelligence methods, including principal component analysis, hierarchical cluster analysis, stepwise linear regression, multidimensional scaling, neural network modeling, and genetic function approximation, among others, can be used to analyze this large activity database. Mining the database can provide useful information: (a) for the development of anticancer drugs; (b) for a better understanding of the molecular pharmacology of cancer; and (c) for improvement of the drug discovery process.

EDGAR: extraction of drugs, genes and relations from the biomedical literature

EDGAR (Extraction of Drugs, Genes and Relations) is a natural language processing system that extracts information about drugs and genes relevant to cancer from the biomedical literature. This automatically extracted information has remarkable potential to facilitate computational analysis in the molecular biology of cancer, and the technology is straightforwardly generalizable to many areas of biomedicine. This paper reports on the mechanisms for automatically generating such assertions and on a simple application, conceptual clustering of documents. The system uses a stochastic part of speech tagger, generates an underspecified syntactic parse and then uses semantic and pragmatic information to construct its assertions. The system builds on two important existing resources: the MEDLINE database of biomedical citations and abstracts and the Unified Medical Language System, which provides syntactic and semantic information about the terms found in biomedical abstracts.

MedMiner: an Internet text-mining tool for biomedical information, with application to gene expression profiling

The trend toward high-throughput techniques in molecular biology and the explosion of online scientific data threaten to overwhelm the ability of researchers to take full advantage of available information. This problem is particularly severe in the rapidly expanding area of gene expression experiments, for example, those carried out with cDNA microarrays or oligonucleotide chips. We present an Internet-based hypertext program, MedMiner, which filters and organizes large amounts of textual and structured information returned from public search engines like GeneCards and PubMed. We demonstrate the value of the approach for the analysis of gene expression data, but MedMiner can also be extended to other areas involving molecular genetic or pharmacological information. More generally still, MedMiner can be used to organize the information returned from any arbitrary PubMed search.