Prediction of response to antiretroviral therapy by human experts and by the EuResist data-driven expert system (the EVE study).

OBJECTIVES: The EuResist expert system is a novel data-driven online system for computing the probability of 8-week success for any given pair of HIV-1 genotype and combination antiretroviral therapy regimen plus optional patient information. The objective of this study was to compare the EuResist system vs. human experts (EVE) for the ability to predict response to treatment. METHODS: The EuResist system was compared with 10 HIV-1 drug resistance experts for the ability to predict 8-week response to 25 treatment cases derived from the EuResist database validation data set. All current and past patient data were made available to simulate clinical practice. The experts were asked to provide a qualitative and quantitative estimate of the probability of treatment success. RESULTS: There were 15 treatment successes and 10 treatment failures. In the classification task, the number of mislabelled cases was six for EuResist and 6-13 for the human experts [mean±standard deviation (SD) 9.1±1.9]. The accuracy of EuResist was higher than the average for the experts (0.76 vs. 0.64, respectively). The quantitative estimates computed by EuResist were significantly correlated (Pearson r=0.695, P<0.0001) with the mean quantitative estimates provided by the experts. However, the agreement among experts was only moderate (for the classification task, inter-rater κ=0.355; for the quantitative estimation, mean±SD coefficient of variation=55.9±22.4%). CONCLUSIONS: With this limited data set, the EuResist engine performed comparably to or better than human experts. The system warrants further investigation as a treatment-decision support tool in clinical practice.

A highly divergent picornavirus in a marine mammal.

Nucleic acids from an unidentified virus from ringed seals (Phoca hispida) were amplified using sequence-independent PCR, subcloned, and then sequenced. The full genome of a novel RNA virus was derived, identifying the first sequence-confirmed picornavirus in a marine mammal. The phylogenetic position of the tentatively named seal picornavirus 1 (SePV-1) as an outlier to the grouping of parechoviruses was found consistently in alignable regions of the genome. A mean protein sequence identity of only 19.3 to 30.0% was found between the 3D polymerase gene sequence of SePV-1 and those of other picornaviruses. The predicted secondary structure of the short 506-base 5'-untranslated region showed some attributes of a type IVB internal ribosome entry site, and the polyprotein lacked an apparent L peptide, both properties associated with the Parechovirus genus. The presence of two SePV-1 2A genes and of the canonical sequence required for cotranslational cleavage resembled the genetic organization of Ljungan virus. Minor genetic variants were detected in culture supernatants derived from 8 of 108 (7.4%) seals collected in 2000 to 2002, indicating a high prevalence of SePV-1 in this hunted seal population. The high level of genetic divergence of SePV-1 compared to other picornaviruses and its mix of characteristics relative to its closest relatives support the provisional classification of SePV-1 as the prototype for a new genus in the family Picornaviridae.

A 6-basepair insert in the reverse transcriptase gene of human immunodeficiency virus type 1 confers resistance to multiple nucleoside inhibitors.

While many point mutations in the HIV-1 reverse transcriptase (RT) confer resistance to antiretroviral drugs, inserts or deletions in this gene have not been previously characterized. In this report, 14 RT inhibitor-treated patients were found to have HIV-1 strains possessing a 6-basepair insert between codons 69 and 70 of the RT gene. Known drug resistance mutations were also observed in these strains, with T215Y appearing in all strains. Genotypic analysis indicated that the inserts had substantial nucleotide variability that resulted in relatively restricted sets of amino acid sequences. Linkage of patients' treatment histories with longitudinal sequencing data showed that insert strains appeared during drug regimens containing ddI or ddC, with prior or concurrent AZT treatment. Drug susceptibility tests of recombinant patient isolates showed reduced susceptibility to nearly all nucleoside RT inhibitors. Site- directed mutagenesis studies confirmed the role of the inserts alone in conferring reduced susceptibility to most RT inhibitors. The addition of AZT-associated drug resistance mutations further increased the range and magnitude of resistance. These results establish that inserts, like point mutations, are selected in vivo during antiretroviral therapy and provide resistance to multiple nucleoside analogs.

Bayesian network analysis of resistance pathways against HIV-1 protease inhibitors.

Interpretation of Human Immunodeficiency Virus 1 (HIV-1) genotypic drug resistance is still a major challenge in the follow-up of antiviral therapy in infected patients. Because of the high degree of HIV-1 natural variation, complex interactions and stochastic behaviour of evolution, the role of resistance mutations is in many cases not well understood. Using Bayesian network learning of HIV-1 sequence data from diverse subtypes (A, B, C, F and G), we could determine the specific role of many resistance mutations against the protease inhibitors (PIs) nelfinavir (NFV), indinavir (IDV), and saquinavir (SQV). Such networks visualize relationships between treatment, selection of resistance mutations and presence of polymorphisms in a graphical way. The analysis identified 30N, 88S, and 90M for nelfinavir, 90M for saquinavir, and 82A/T and 46I/L for indinavir as most probable major resistance mutations. Moreover we found striking similarities for the role of many mutations against all of these drugs. For example, for all three inhibitors, we found that the novel mutation 89I was minor and associated with mutations at positions 90 and 71. Bayesian network learning provides an autonomous method to gain insight in the role of resistance mutations and the influence of HIV-1 natural variation. We successfully applied the method to three protease inhibitors. The analysis shows differences with current knowledge especially concerning resistance development in several non-B subtypes.

Human Immunodeficiency Virus Reverse Transcriptase and Protease Sequence Database: an expanded data model integrating natural language text and sequence analysis programs.

The HIV Reverse Transcriptase and Protease Sequence Database is an on-line relational database that catalogs evolutionary and drug-related sequence variation in the human immunodeficiency virus (HIV) reverse transcriptase (RT) and protease enzymes, the molecular targets of anti-HIV therapy (http://hivdb.stanford.edu). The database contains a compilation of nearly all published HIV RT and protease sequences, including submissions from International Collaboration databases and sequences published in journal articles. Sequences are linked to data about the source of the sequence sample and the antiretroviral drug treatment history of the individual from whom the isolate was obtained. During the past year 3500 sequences have been added and the data model has been expanded to include drug susceptibility data on sequenced isolates. Database content has also been integrated with didactic text and the output of two sequence analysis programs.

Highly drug-resistant HIV-1 clinical isolates are cross-resistant to many antiretroviral compounds in current clinical development.

OBJECTIVES: To assess the in-vitro drug susceptibility of a panel of five well-characterized drug-resistant HIV variants to recently developed anti-HIV compounds including seven reverse transcriptase (RT) inhibitors and seven protease inhibitors. METHODS: Drug-resistant viral strains were selected on the basis of the prevalence of these mutants in patient samples from local area HIV clinics. The isolates included one multinucleoside-resistant virus containing the Q151M mutation, and four clinical isolates containing multiple RT and protease resistance mutations. The activity of the experimental compounds against these isolates was determined using drug susceptibility assays and measuring the viral antigen p24 end-point. RESULTS: These clinically relevant highly drug-resistant viruses were resistant to many of the new compounds in clinical development. In most cases the resistance mutations of the clinical isolate were different from those selected in vitro for the particular experimental compound. CONCLUSIONS: It is critical to expand the preclinical development of new drugs to include the assessment of their activity against currently circulating highly drug-resistant clinical strains, in order to develop appropriate salvage therapies for patients harboring resistant strains.

Phenotypic hypersusceptibility to non-nucleoside reverse transcriptase inhibitors in treatment-experienced HIV-infected patients: impact on virological response to efavirenz-based therapy.

BACKGROUND: Enhanced susceptibility to non-nucleoside reverse transcriptase inhibitors (NNRTI) was recently described in association with increased resistance to nucleoside analogs (nucleoside reverse transcriptase inhibitors; NRTI). OBJECTIVES: To determine the prevalence of NNRTI hypersusceptibility, the genotypic correlates, and its impact on virologic response to efavirenz-based salvage therapy. METHODS: Genotype and phenotype testing was performed retrospectively on baseline isolates from 30 patients who received salvage therapy containing efavirenz. NNRTI hypersusceptibility was defined as a 50% inhibitory concentration (IC(50)) of < 0.5 that of the wild-type control. RESULTS: Eight isolates had major NNRTI mutations. Among the 22 isolates with no major NNRTI mutations, 11 (50%) were hypersusceptible to efavirenz, 10 (45%) to delavirdine, and eight (36%) to nevirapine. Among eight isolates with NNRTI mutations, NNRTI resistance was present, but at lower than expected levels. The number of NRTI mutations was correlated inversely with the fold decrease in susceptibility to efavirenz (Spearman's rho, -0.57; P = 0.005), delavirdine (rho, -0.43; P = 0.04), and nevirapine (rho, -0.69; P < 0.001). Excluding subjects with NNRTI mutations, subjects with efavirenz hypersusceptibility at baseline had significantly better virologic suppression over 24 weeks than those without efavirenz hypersusceptibility (P < 0.001). CONCLUSION: NNRTI hypersusceptibility is common in heavily treated but NNRTI naive patients and is related directly to NRTI resistance mutations. Among patients receiving efavirenz-containing regimens, NNRTI hypersusceptibility was associated with an improved virologic outcome after 24 weeks of therapy. A reversal of phenotypic resistance was seen in patients with NNRTI mutations in the presence of multiple NRTI mutations, but no obvious virologic benefit of this phenomenon was seen in this study.

HIV prevalence, immunosuppression, and drug resistance in patients with tuberculosis in an area endemic for AIDS.

From October 1987 to June 1988, we attempted to determine the prevalence of HIV infection among patients hospitalized with tuberculosis and the extent of immunosuppression among those tuberculosis patients infected with HIV. Of 178 consecutive patients, 18-65 years of age, who were hospitalized with newly diagnosed, previously untreated tuberculosis, 46% (82 out of 178) had clinical or serological evidence of HIV infection, 30% (54 out of 178) were HIV-seronegative, and 24% (42 out of 178) could not be assessed for the presence of HIV infection. Among the HIV-seropositive patients without an AIDS-defining diagnosis by non-tuberculous criteria, the median CD4 lymphocyte (CD4) count was 133 x 10(6) cells/l (range: 11-677 x 10(6]; among the HIV-seronegative patients, the median CD4 count was 613 x 10(6) cells/l (range: 238-1614 x 10(6); P less than 0.001). Among the HIV-seropositive patients, those with disseminated tuberculosis (median CD4 = 79 x 10(6) cells/l) and those with pulmonary tuberculosis who had radiographic evidence of mediastinal or hilar adenopathy (median CD4 = 45 x 10(6) cells/l) had the most severe CD4 depletion, whereas those with localized extrapulmonary tuberculosis (median CD4 = 242 x 10(6) cells/l) and those with pulmonary tuberculosis without adenopathy (median CD4 = 299 x 10(6) cells/l) were less severely immunosuppressed. Of the 178 patients, 6% (11 out of 178) were infected with strains of Mycobacterium tuberculosis resistant to both isoniazid and rifampin.

Predictors of survival in HIV-infected tuberculosis patients.

OBJECTIVE: To ascertain predictors of survival in HIV-infected tuberculosis (TB) patients. DESIGN: Retrospective cohort study. SETTING: New York City public hospital. PATIENTS: Fifty-four consecutive HIV-seropositive patients with newly diagnosed TB and no other AIDS-defining illnesses. MAIN OUTCOME MEASURES: CD4+ T-lymphocyte counts, completion of anti-TB therapy, repeat hospitalizations with TB, and survival. RESULTS: Forty-five (84%) of the 54 patients died a median of 15 months after TB diagnosis (range, 1-80 months), five (9%) were alive after a median of 81 months (range, 75-84 months), and four (7%) were lost to follow-up after a median of 42 months (range, 30-66 months). In univariate analyses, disseminated TB, intrathoracic adenopathy, oral candidiasis and CD4 count depletion were each associated with decreased survival. In a multivariate analysis, CD4 count depletion was the only independent predictor of decreased survival. Repeat hospitalization with TB occurred in 10 out of 15 patients who did not complete anti-TB therapy compared with one out of 21 patients who completed anti-TB therapy (P < 0.001). CONCLUSION: The clinical presentation of TB and CD4 count at TB diagnosis are each predictive of survival in HIV-seropositive TB patients. The CD4 count is the only independent predictor of survival.

Extrapulmonary tuberculosis in patients with human immunodeficiency virus infection.

The annual number of cases of culture-proven extrapulmonary tuberculosis (TB) at our hospital increased from 47 cases in 1983 to 113 cases in 1988. At least 43% (199) of 464 consecutive patients with extrapulmonary TB during this 6-year period were infected with the human immunodeficiency virus (HIV); since HIV serologic testing was not performed routinely the true HIV prevalence is likely to be higher. Of the HIV-infected patients, 59% were intravenous drug users, 31% were Haitian, 3% were homosexual males, 1% were perinatally-infected infants, and 6% did not have a known risk factor for HIV infection. Ninety-eight percent of the HIV-infected patients were black (84%) or hispanic (14%). The HIV-infected patients were more likely than the control patients to have either disseminated, genitourinary, intra-abdominal, mediastinal, or concurrent pulmonary TB. Fever was nearly universal among the HIV-infected patients, but was absent in about one-third of the control patients. Among untreated HIV-infected patients, disease progression was rapid and nearly always fatal. Among HIV-infected patients who received treatment, the response to therapy, as judged by hospital survival and time to defervescence, was similar to that of the control patients. Despite the extensive tuberculous dissemination among the HIV-infected patients, the diagnosis of TB was difficult and often delayed. In addition to the decrease in tuberculin reactivity and the atypical chest radiograph patterns, there was a need to consider other HIV-related infections in the differential diagnosis. Although sputum specimens grew M. tuberculosis in greater than 90% of the HIV-infected patients in whom they were obtained, sputum AFB stains were positive in less than 50%. Blood and urine specimen cultures were positive in 56% and 77% of the HIV-infected patients in whom these specimens were obtained, but did not provide a means of early diagnosis. Cerebrospinal fluid and pleural fluid were abnormal in nearly all patients with involvement of these sites but were rarely AFB-positive and were, therefore, only suggestive of TB. Procedures such as biopsies and aspirates of peripheral lymph nodes, visceral lymph nodes, liver, and bone marrow provided the highest immediate diagnostic yields with rates between 50% and 90%. These procedures must be considered early in the course of illness in HIV-infected patients with suspected extrapulmonary TB due to the rapidly progressive nature of this often fatal but usually treatable infection.

Successful prophylaxis of Pneumocystis carinii pneumonia with trimethoprim-sulfamethoxazole in AIDS patients with previous allergic reactions.

Thirty-four homosexual patients with AIDS were treated for Pneumocystis carinii pneumonia between April 1984 and November 1985. All 31 survivors were treated with oral trimethoprim-sulfamethoxazole (TMP-SMX) prophylaxis immediately upon completion of intravenous therapy, despite the prior occurrence of hypersensitivity reactions to intravenous TMP-SMX in 21 of these patients. Only four patients had subsequent reactions to oral TMP-SMX requiring the drug's discontinuation. None of the patients remaining on prophylaxis developed recurrent Pneumocystis pneumonia. Oral TMP-SMX appears effective at preventing recurrent Pneumocystis pneumonia in patients with AIDS. Hypersensitivity reactions during therapy with TMP-SMX may not be a contraindication to continuation of therapy and subsequent oral prophylaxis.

HIV-1 syncytium-inducing phenotype, virus burden, codon 215 reverse transcriptase mutation and CD4 cell decline in zidovudine-treated patients.

The variable rate of disease progression in HIV-1-infected patients treated with zidovudine may be related to certain viral characteristics, such as, antiviral drug resistance, virus burden, and viral syncytium-inducing (SI) capacity. Thirty-two HIV-1-infected patients treated with zidovudine (mean of 34 months) were studied to determine the relationship of SI phenotype and the codon 215 pol gene mutation (a marker of zidovudine resistance) to virus burden and CD4 cell decline. Patients with SI strains and the codon 215 mutation in their proviral DNA had a 54% decline in CD4 cells and a virus burden of 21,480 proviral DNA copies/10(6) CD4 cells. In contrast, patients with non-SI (NSI) strains and wild-type at codon 215 had a 10% increase in CD4 cells and had a viral burden 1/46 that of patients with SI and the 215 mutation. Among patients with NSI strains, changes in CD4 cells depended on the presence of the codon 215 mutation (-160 CD4 cells/microliters), compared with those wild-type at codon 215 (+28 CD4 cells/microliters) (p < 0.01). There was a concordant rise in virus burden between proviral DNA and plasma HIV RNA depending on HIV phenotype and genotype. Using multiple linear regression, SI phenotype and the codon 215 mutation were found to independently predict CD4 cell decline and increased virus burden in zidovudine-treated patients.

HIV type 1 envelope subtype C sequences from recent seroconverters in Zimbabwe.

HIV-1 envelope sequence patterns have implications for virus cell tropism and for the development of an effective vaccine. To identify the sequence characteristics of recently transmitted HIV-1 isolates in southern Africa, we sequenced the V3-V5 envelope regions of 24 male seroconverters in Harare, Zimbabwe. Each of the sequences clustered with previously reported subtype C isolates and there was a mean 17% intersequence pairwise genetic distance between the Zimbabwean isolates. Three isolates were syncytium inducing (SI). One of the SI isolates had an unusual GIGK crown and a deletion at codon 23; one had the codon 23 deletion alone; and one had a high net positive charge in the V3 loop. The extensive genetic diversity within the envelope of subtype C HIV-1 isolates must be considered in vaccine development. Further analysis of subtype C SI isolates and site-directed mutagenesis experiments are required to determine the molecular basis of SI activity in global HIV-1 isolates.

Reproducibility of human immunodeficiency virus type 1 (HIV-1) protease and reverse transcriptase sequencing of plasma samples from heavily treated HIV-1-infected individuals.

The reproducibility of population-based human immunodeficiency virus type 1 (HIV-1) protease and reverse transcriptase (RT) sequencing was assessed using replicate aliquots of cryopreserved plasma samples obtained from seven heavily treated HIV-1-infected individuals. The sequence of each sample replicate was compared with the consensus sequence for that sample and 99.4% of 35128 amino acids were found to be concordant with the sample consensus. Partial discordances were present at 0.5% of positions and complete discordances were present at <0.1% of positions. To assess the reproducibility at detecting mutations (defined here as differences from the subtype B consensus sequence), the proportion of sequences having a mutation when at least two sequences from that sample had the same mutation were examined. There was a median of 13 protease and 18 RT mutations per sample for a total of 3126 mutations; 95% of these mutations were detected. However, sequencing of multiple clones from two samples demonstrated that those mutations present in a minority of clones were often not detected by population-based sequencing. These results suggest that HIV-1 protease and RT sequencing of circulating plasma virus is highly reproducible but that the sensitivity at detecting mutations may be low if those mutations are present as minor variants.

Zidovudine susceptibility testing of human immunodeficiency virus type 1 (HIV) clinical isolates.

Traditional antiviral susceptibility testing methods using cell lines can be applied to no more than about 30% of clinical HIV isolates (Larder et al., 1989a; Fenyo et al., 1989). We tested the cell-free supernatant from low passage clinical HIV isolates using donor peripheral blood mononuclear cells (PBMC). Drug susceptibility was assessed by measuring the effect of increasing zidovudine (ZDV) concentrations on HIV P24 antigen production. Susceptibility results were obtained on 24/27 consecutive clinical isolates and 6/6 laboratory isolates. The mean IC90 of isolates from untreated patients was 0.008 microM ZDV (range: 0.002-0.038). The IC90s of isolates from ZDV-treated patients ranged from 0.007 to greater than 10 microM ZDV. All isolates with an IC90 < 0.1 microM ZDV had a wild type sequence at codon 215 of the HIV pol gene; 11/12 isolates with an IC90 > 0.1 microM ZDV had a mutation at codon 215 (P < 0.001). Among 16 ZDV-treated patients, there was a modest correlation between the change in CD4 count from the start of ZDV treatment and the IC90 of the patient's isolate following treatment (r = 0.51). Susceptibility testing using donor PBMC can be a sensitive means of testing a broad range of clinical HIV isolates.

Highly active antiretroviral therapy (HAART) for the treatment of infection with human immunodeficiency virus type 1.

Highly active antiretroviral therapy (HAART) refers to a broad category of treatment regimens usually comprised of three or more antiretroviral drugs that, in previously untreated HIV-1-infected patients, are expected to reduce plasma virus levels below the limits of detection. Most HAART regimens include drugs from at least two of the three classes of antiretroviral therapy (nucleoside analog reverse transcriptase (RT) inhibitors, non-nucleoside analog RT inhibitors, and protease inhibitors). In deciding when to initiate antiretroviral therapy, physicians and their patients must balance the virological and immunological benefits of early treatment with the costs of drug therapy, the risk of drug side effects, and the risk of drug resistance if adherence is suboptimal. In previously untreated patients, HIV-1 replication can be suppressed indefinitely with certain HAART regimens. In previously treated patients, the benefits of HAART are often significantly diminished.