Tumor necrosis factor alpha (TNF-α), which is produced by monocytes and macrophages in response to bacterial and viral infections, is an integral part of the host immune response. Excessive TNF-α, however, may lead to some of the pathophysiology associated with chronic infections. Recently, increased TNF-α and soluble TNF-α receptor levels have been documented in serum samples of patients infected with the human immunodeficiency virus type 1 (HIV-1) (1-4) and may be involved in the pathogenesis of HIV-1-associated wasting (5,6). The mechanism of this interaction is not understood and has been the subject of intensive investigation (5,7). TNF-α has been shown to play an important role in enhancing replication of HIV-1 in cell lines and in cells from HIV-1-infected patients (8-10). TNF-α is a strong inducer of NF-κB, a transcriptional factor used by the virus (11). Tuberculosis, a chronic mycobacterial infection, is characterized by TNF-α-associated toxicities, including fevers, weakness, fatigue, and progressive weight loss (13). Components of the mycobacterial cell wall have been found to stimulate TNF-α production in vitro (12,14-17). It has been suggested that tuberculosis may accelerate the progression of HIV-1 disease by enhancing TNF-α production, thereby activating HIV-1 replication (5,18). In addition, it has been recently shown that mycobacteria directly activate HIV-1 expression in vitro (19,20).
The synthesis of TNF-α is down-regulated by some cytokine inhibitors. For example, thalidomide has been shown to reduce serum TNF-α levels in patients with tuberculosis and with lepromatous leprosy (erythema nodosum leprosum), with a concomitant abrogation of clinical symptoms (13,21). In addition, thalidomide has been shown to suppress the activation in vitro of latent HIV-1 in a monocytoid line, U1 (22). Furthermore, thalidomide inhibited the activation in vitro of virus replication in the peripheral blood mononuclear cells of patients with HIV-1 infection (22). This inhibition was associated with a reduction in TNF-α production. These results suggested that thalidomide might be useful in a clinical context.
We therefore undertook a pilot double-blind, placebo-controlled randomized study of thalidomide treatment in HIV-1-infected patients with or without co-infection with tuberculosis. The effect of the drug on CD4+ T-cell counts, plasma TNF-α levels, HIV-1 levels, clinical responsiveness, and body weight were all evaluated, to determine the clinical, immunological, and virologic effects of thalidomide treatment in patients with HIV-1 and tuberculosis-associated wasting.
Thirty-nine HIV-1-seropositive male patients who were treated at the chest and venereal diseases clinics in Chiang Mai, Thailand, in November 1993, who were referred either by self or physicians and who self-reported ≥10% loss of body weight in the past 6 months, were invited to participate in a randomized placebo-controlled outpatient study of thalidomide. Patients were excluded if they were suspected of having multidrug-resistant tuberculosis (as determined by clinical response to therapy and confirmed by Bactec cultures) or had a history of peripheral neuropathy because the latter is a recognized side effect of thalidomide. Likewise, patients were excluded if they had been treated with antiretrovirals within the 2 weeks before the study, or had a known opportunistic infection other than tuberculosis, or had a life expectancy of <3 months. No antiretroviral drugs were administered to the patients during the study.
Patients were grouped into two blocks: HIV-1 infected only and HIV-1 infected with tuberculosis. Individual patients were then randomized within each block based on a table of random numbers and allocated to either treatment or placebo. Eighteen patients who entered the study were being treated for active pulmonary tuberculosis, including nine who were diagnosed with tuberculosis during the 2 weeks before entering the study. Chest radiographs of 15 patients showed findings suggestive of active pulmonary tuberculosis; two patients had hilar adenopathy; one had changes resembling miliary disease. All patients received dietary consultation, but no dietary supplementation was provided. The study was approved by the institutional review boards of Rockefeller University and Chiang Mai University, and all patients gave written informed consent.
Thalidomide at 300 mg/day or placebo was given p.o. nightly for 21 days followed by a 7-day washout period. Thalidomide and placebo were provided by Grunenthal GMBH (Aachen, Germany). Patients in whom tuberculosis was diagnosed were given ≥1 week of antituberculosis multidrug therapy before beginning the study (Table 1). Multidrug therapy consisted of 300 mg isoniazid, 600 mg rifampin, 1.2 g pyrazinamide, and 800 mg ethambutol, given daily. The study drug was discontinued if severe adverse reactions became evident.
Initial complete medical histories were taken, and each patient underwent physical examination by one of the study physicians. All patients had baseline complete blood counts, T-cell subset determinations, and confirmatory HIV-1 testing (Cobas Core Anti-HIV-1/HIV-2 enzyme immunoassay; Roche, Bangkok, Thailand; Serodia-HIV-1 Particle Agglutination Test, Fujirebio, Tokyo, Japan). The patients underwent chest radiography and measurement of body weight. The two study groups were similar with respect to age and starting weight (Table 1). If a patient had pulmonary symptoms, sputum examination and culture for mycobacteria were obtained. The patients were interviewed and examined weekly by a study physician and study nurse who were unaware of the treatment regimen of each patient. The evaluations included standardized assessments of symptoms, signs, and compliance.
Blood for CD4+ T-cell counts and plasma determinations was collected in heparin-containing vacutainers on days 0, 7, 14, 21, and 28 and in ethylenediaminetetraacetic acid (EDTA)-containing vacutainers on days 0 and 21. Specimens were transported to the local laboratory and refrigerated at 4°C within 3 h of collection for CD4+ T-cell counts. Heparin- and EDTA-containing plasma was separated, stored at - 70°C for ≤6 weeks, and transported to New York on dry ice for TNF-α and HIV-1 analyses.
TNF-α Determination in Plasma Samples
Heparinized and EDTA-treated plasma samples used for cytokine determination were frozen and thawed only once, and serial samples from a given patient were assayed at the same time. Total TNF-α (unbound and bound to the soluble TNF-α receptor) was determined in undiluted plasma samples by a commercial enzyme immunoassay (Medgenix, Fleurus, Belgium), as described (7). Assays were done according to the manufacturer's guidelines. No significant difference in TNF-α levels between the heparin-treated and EDTA-treated plasma samples was observed in these assays. Cytokine levels are expressed as picograms per milliliter. This assay has a minimum detection level of ≈3 pg/ml.
HIV-1 Determination in Plasma Samples
Plasma HIV-1 RNA was quantified in EDTA-containing samples collected on days 0 and 21 and in heparin-containing samples collected on days 0, 7, 14, 21, and 28 by branched-chain DNA hybridization (Chiron Corporation, Emeryville, CA, U.S.A.) (23). Results are expressed as RNA equivalents per milliliter. We compared HIV-1 levels in different samples. Baseline (day 0) HIV-1 levels were greater in EDTA-containing samples than in heparin-containing samples, as has recently been reported (24); the EDTA-containing samples had a 92% greater viral particle concentration. HIV-1 levels in the heparin- and EDTA-containing specimens had similar relative values and were highly concordant for changes over time. However, because of the increased sensitivity, EDTA-containing samples were used for all comparative analyses on day 0 and day 21 (Figs. 1-3; Table 2).
T-Cell Subset Determination
Whole blood was collected in heparinized vacutainers on days 0 and 21. Peripheral blood mononuclear cells (PBMCs) were separated by Ficoll-Hypaque gradient centrifugation. Anti-CD4 monoclonal antibodies (Zymed, South San Francisco, U.S.A.) were used to determine CD4+ T-cell numbers in PBMCs using a FACScan flow cytometer (Becton Dickinson, San Jose, CA, U.S.A.).
Data were analyzed with the Statistical Program for Social Sciences (SPSS PC3+). Log transformation was performed to normalize nonparametric data. The Student's t test for group means was used to analyze parametric data between groups (Figs. 1, 3). Standard correlation analysis was carried out for the TNF-α versus HIV-1 correlation using a Pearson products correlation coefficient. Significance was evaluated using a t test for the correlation (SAS software) (Fig. 2). Repeated-measures analyses of variance were performed to identify treatment effects in serial data (Fig. 4). The Wilcoxon signed-rank test was used to compare paired samples in Fig. 2.
Thirty-nine HIV-1-infected male patients (18 dually infected with HIV-1 and tuberculosis), with ≥10% loss of normal body weight and who were not on antiretroviral therapy, were enrolled in the study. Of the 32 patients who successfully completed the study, 16 were co-infected with tuberculosis and HIV-1, and 16 were infected with HIV-1 only. Eight patients with HIV-1 infection and eight dually infected with HIV-1 and tuberculosis received thalidomide; eight HIV-1-infected patients and eight dually infected with HIV-1 and tuberculosis received placebo. Data from the 32 patients who completed the study form the basis of this report (Table 1). Seven patients did not complete the study (Table 3).
TNF-α Levels and HIV-1 Levels Before Drug Treatment
Several differences were observed between the baseline characteristics of patients with HIV-1 infection alone and those who were dually infected with tuberculosis. Despite similar mean CD4+ T-cell counts, patients with tuberculosis had a higher mean plasma TNF-α level (120 ± 29 pg/ml) than was observed in HIV-1 patients without tuberculosis (70 ± 13 pg/ml, p < 0.04 by t test for logtransformed data) (Fig. 1). Furthermore, patients with both HIV-1 and tuberculosis had a higher mean HIV-1 level (487 ± 135 × 103 RNA equivalents versus 211 ± 43 × 103 RNA equivalents, p < 0.05). A striking positive correlation was seen between TNF-α levels and HIV-1 levels (r = 0.74, p = 0.0001) (Fig. 2). Thus, most patients with high TNF-α levels had high HIV-1 levels. The analysis also showed very high correlation coefficients for the two populations (r = 0.69, p = 0.0029 for HIV-1; r = 0.77, p = 0.0007 for HIV-1 + TB). Statistical analyses to determine whether the slopes for the two populations were different showed no significant difference at p = 0.3420 (a p value <0.05 would indicate significant differences). Thus, the correlation between TNF-α levels and HIV-1 levels was observed whether or not HIV-1-infected patients were co-infected with tuberculosis.
Response to Thalidomide
During the course of treatment patients were asked to comment on their subjective well-being. A somewhat higher percentage of patients receiving thalidomide reported an improvement in appetite and strength and a reduction in fatigue (Table 4).
Six of the 20 patients randomized to receive thalidomide developed drug rashes between days 7 and 14 (median day 12), which were characterized by pruritic erythematous macular skin lesions over the trunk and back, but not in the axillae and groin. In five patients the rash resolved upon discontinuation of medication. One patient with the rash, in whom the drug was not discontinued owing to a missed follow-up appointment, went on to develop Stevens-Johnson syndrome with desquamation, which resolved upon discontinuation of the medication. Six of the patients who developed the drug-related rash had significantly lower mean CD4+ T-cell counts than patients who did not develop the rash (17 cells/mm3/ versus 216 cells/mm3, respectively, p < 0.005) (Tables 1 and 3). Two of the patients who experienced rashes left the study because of the rash. Two additional patients randomized to the thalidomide group developed opportunistic infections: Pneumocystis carinii pneumonia (day 11) and cryptococcal meningitis (day 21). They also left the study (Table 3).
No rash was seen in any of the 19 patients randomized to receive placebo. Three patients in the placebo group did not complete the study. Cryptococcal meningitis appeared in one (day 21) and sepsis in another (day 28); the third was lost to follow-up (day 14) (Table 3). Other adverse effects of thalidomide therapy included sedation, dryness of mouth, and constipation, all known side effects of thalidomide (Table 4). There were no reported adverse neurological effects. No significant changes in CD4+ T-cell counts were observed during the study in any of the groups (data not shown).
Body weight was measured in all patients throughout the study (Tables 1 and 3). Figure 4 shows the mean percentage weight gain in the 32 patients completing the study. Patients with HIV-1 infection alone had starting weights similar to patients with both HIV-1 and tuberculosis infection. Weight gain was highly significant in the co-infected group (8.2% ± 2.6%, p = 0.003 by repeated-measures analysis of variance) and in the group infected with HIV-1 only (4.4% ± 1.1, p = 0.012 by repeated-measures analysis of variance). Duration of antituberculosis therapy was not a confounder in this study. In patients co-infected with tuberculosis, the percentage weight gain by day 21 of thalidomide treatment was greater, but not statistically significant (Fig. 4). All patients who received thalidomide gained a significant amount of weight steadily ≤21 days of treatment (6.5% ± 1.2%, p < 0.02 by repeated-measures analysis of variance) compared with those treated with placebo (Table 1). Within a few days after thalidomide was discontinued, weight gain stopped or was even reversed in some patients (Table 1 and Fig. 4).
Plasma TNF-α levels were evaluated to determine the effect of thalidomide or placebo treatment on TNF-α production (Table 2). The range of TNF-α levels measured in all patients was 23-493 pg/ml at baseline. This wide range of values in HIV-1-infected individuals has been observed by us and other investigators using the same assay and different assays (7, 25). By comparison, TNF-α levels previously reported for normal controls ranged from 0 to 35 pg/ml, with a mean of 5-8 pg/ml (7,25). For patients with tuberculosis only, the serum TNF-α levels reported previously ranged from 0 to 125 pg/ml (13, 25).
At the start of this study the eight patients with HIV-1 infection only who were randomized to receive placebo had relatively low plasma TNF-α levels (mean ± SEM = 77 ± 19 pg/ml). The TNF-α levels did not change significantly in this group during the course of the study (Table 2 and Fig. 5). Similarly, although plasma TNF-α levels were higher in HIV-1 patients co-infected with tuberculosis who were receiving placebo (134 ± 60 pg/ml), no significant changes in plasma TNF-α levels were observed during the study (p = 0.6 and 0.8, respectively, by the Wilcoxon signed-rank test). Therefore, placebo treatment had no clear effect on TNF-α levels whether patients were HIV-1 infected or were dually infected with HIV-1 and tuberculosis.
On the other hand, thalidomide therapy was associated with ≥10% reduction in plasma TNF-α levels in seven of the eight patients co-infected with HIV-1 and tuberculosis (Table 2 and Fig. 5) (p = 0.06 for the change during treatment, by the Wilcoxon signed-rank test). This group of dually infected patients had starting plasma TNF-α levels higher than those of patients with HIV-1 infection alone. The thalidomide effect on plasma TNF-α levels was not observed consistently in those patients with HIV-1 only (Table 2 and Fig. 5) (p = 0.8). Only four of the eight patients in the HIV-1-infected group showed a reduction in TNF-α in response to thalidomide, while three showed an increase, and one remained the same.
Plasma HIV-1 RNA was studied by branched-chain DNA hybridization in all patients. HIV-1 levels assayed in samples of plasma containing EDTA obtained at days 0 and 21 are presented in Table 2.
Patients with HIV-1 infection only who were randomized to receive placebo (eight patients) had relatively low plasma HIV-1 levels at the start of the study (303 ± 60 RNA equivalents × 103/ml). HIV-1 levels did not change significantly in this group during the course of the study (p = 0.08, by the paired t test) (Table 2 and Fig. 3). At the start of the study, mean plasma HIV-1 levels were higher in patients receiving placebo who were co-infected with HIV-1 and tuberculosis (418 ± 198 RNA equivalents × 103/ml). A reduction in viral levels was observed in four of the eight placebo-treated patients with HIV-1 and tuberculosis, but this reduction was not statistically significant for the group as a whole (p = 0.3, paired t test).
In patients with HIV-1 infection only who received thalidomide, no significant change in HIV-1 levels was observed (p = 0.1 paired t test) (Table 2 and Fig. 3). However, in patients with both HIV-1 and tuberculosis who had higher HIV-1 levels initially, thalidomide treatment was associated with a reduction in HIV-1 viral levels (Table 2 and Fig. 3). For this group, the reduction in HIV-1 is significant (p = 0.01, paired t test). When the drug treatment and placebo groups were combined and stratified according to infection with tuberculosis, there was a significant decrease in viral levels in tuberculosis patients compared with the patients without tuberculosis (p < 0.01, two-sample test), suggesting that antituberculosis therapy in itself may contribute to a reduction in viral levels.
In recent years, the pleiotropic effects of TNF-α have been well documented (26,27). TNF-α, which has an important role in immune regulation, was first associated with wasting when Rouzer and Cerami demonstrated that a factor (cachectin) suppresses lipoprotein lipase activity in rodent macrophages, resulting in elevated serum levels of triglycerides and weight loss (28). Recent investigations have indicated that TNF-α may also play a pivotal role in HIV-1 infection. TNF-α has been shown to augment the replication of HIV-1 both in cell lines and in cells from HIV-1-infected patients (8-11,22,29). Moreover, TNF-α levels are elevated in the serum samples of patients with AIDS (1-3).
In this study we show a positive correlation between patients' plasma TNF-α levels and HIV-1 viremia (Fig. 2) assayed in the same sample of EDTA-treated plasma. HIV-1-infected patients with concomitant tuberculosis had significantly higher mean starting levels of TNF-α compared with mean levels of TNF-α in HIV-1-infected patients without tuberculosis (Fig. 1), suggesting that mycobacterial infection may stimulate TNF-α production in vivo. This possibility is supported by the observation that infections with microorganisms such as M. tuberculosis up-regulate TNF-α production by monocytes, presumably through direct stimulation of the phagocytes by mycobacterial cell-wall components (12-17). In our study, patients with both HIV-1 and tuberculosis also showed increased HIV-1 levels compared with patients with HIV-1 infection only (Figs. 1, 3; Table 2). This finding is in agreement with results reported from recent studies that have documented that mycobacteria can directly activate HIV-1 expression in vitro (19,20,25). Our observations support the hypothesis that an opportunistic infection such as tuberculosis may cause increased production of TNF-α, thereby leading to enhanced HIV-1 replication and increased HIV-1 viremia (5,18).
Thalidomide, which has been shown to inhibit monocyte production of TNF-α in vitro (22,30), has important potential as a therapeutic agent to ameliorate the actions of TNF-α in vivo. Thalidomide treatment has been shown to lower serum TNF-α and improve symptoms, including wasting in patients with tuberculosis (13) and in erythema nodosum leprosum in leprosy patients (31). In our present study, we show that thalidomide treatment caused a reduction in TNF-α levels in those patients with both HIV-1 and tuberculosis infections receiving antituberculosis treatment (Fig. 5). The reduction is most pronounced in patients who had increased TNF-α levels at the start of the study. Patients on antituberculosis therapy and placebo did not show a significant reduction in TNF-α levels.
There was no statistically significant reduction in plasma HIV-1 levels in thalidomide-treated patients who had HIV-1 infection only (Fig. 3). Since this is a limited investigation with a small sample size and short duration of therapy, the study may not have had adequate power to detect a thalidomide effect in patients with low starting HIV-1 levels. In addition, although the mean CD4+ T-cell counts in patients without tuberculosis are not significantly different, they are dissimilar and may confound our ability to detect a drug effect. Furthermore, it is likely that HIV-1 activation in vivo is a multifactorial process regulated by factors other than just TNF-α. This theory is supported by the observation that some tuberculosis-infected patients in the placebo group who were receiving antituberculosis therapy showed a decrease in HIV-1 viremia.
Our results show that thalidomide treatment was accompanied by a persistent and significant weight gain in patients with concurrent HIV-1 and M. tuberculosis infections and wasting. Antituberculosis therapy alone did not induce significant weight gain (Fig. 4) during the short study period. Our hypothesis has been that TNF-α production in response to chronic infections such as tuberculosis and HIV is associated with wasting. This hypothesis has supported by a previous study in which we observed an association between thalidomide-induced inhibition of TNF-α production and weight gain in patients with tuberculosis (13). It is important to note that TNF-α levels in either serum or plasma of patients may fluctuate, because of its short half-life (26), and may therefore not always accurately reflect the extent of actual TNF-α production and/or the degree of TNF-α-related toxicities.
In order to address this problem, alternative methods for measuring TNF-α-producing capacity in the leukocytes of patients have been used (13). We have compared serum TNF-α levels, TNF-α mRNA levels in PBMCs and lipopolysaccharide (LPS)-induced TNF-α release in PBMCs studied ex vivo. Serum TNF-α levels showed the least amplitude of change in response to treatment with a TNF-α inhibitor (thalidomide). Although this is the easiest assay to perform under field conditions, especially when the study is carried out with relatively limited laboratory facilities, as in the present pilot study, its limitations must be recognized and the results interpreted cautiously. By comparison, we have shown that thalidomide treatment in vivo has a significant inhibitory effect on TNF-α mRNA levels in freshly isolated PBMCs and on LPS-induced TNF-α release in vitro (13). These observations may explain why, in the present study, significant weight gain was noted in thalidomide-treated patients even in the absence of substantial TNF-α inhibition.
Whether the weight gain observed in patients treated with thalidomide is due to an increase in body water, fat, or muscle mass was not ascertained in this preliminary study. In ongoing studies, changes in caloric intake, basal metabolic rates, and body fat composition are being assessed. If, indeed, thalidomide induces an increase in lean body mass, this drug may prove to be a clinically important way of reversing cachexia in patients with HIV-1 and concurrent mycobacterial infections.
A significant adverse effect and potential drawback to the use of thalidomide in AIDS patients may be the high incidence of drug rash. An increased incidence of rashes in response to thalidomide or other drug therapies has been reported in patients with advanced AIDS (32,33). At the relatively high dose of thalidomide used in our study, the drug-induced rashes necessitated stopping therapy. We have recently obtained preliminary data suggesting that lower doses of thalidomide (e.g., 100-200 mg/day) may cause fewer hypersensitivity reactions while still achieving the desired effects, and ongoing studies are using lower doses.
In summary, our studies establish that high plasma levels of TNF-α are associated with higher plasma HIV-1 levels. Furthermore, opportunistic infections such as tuberculosis, which stimulates TNF-α production, appear to be associated with increased HIV-1 viremia. Thalidomide, a selective TNF-α inhibitor, lowers elevated plasma levels of TNF-α and HIV-1 levels in patients with concomitant HIV-1 and tuberculosis infections. Thalidomide treatment is associated with weight gain in patients infected with HIV-1, with or without M. tuberculosis infection. We are currently studying whether thalidomide can be administered safely for longer periods of time and whether this treatment will maintain lower TNF-α levels and ultimately lower the level of HIV-1 viremia in infected patients, thereby slowing the progression of AIDS.
Acknowledgments: We thank Judy Adams for repeatedly preparing the figures and Marguerite Nulty for repeatedly typing the manuscript. This work was supported by grants AI 24775 and AI 33124 from the National Institutes of Health and by Celgene Corp.
1. Lahdevirta J, Maury CP, Teppo AM, Repo H. Elevated levels of circulating cachectin/tumor necrosis factor in patients with acquired immunodeficiency syndrome. Am J Med
2. Hober D, Haque A, Wattre P, Beaucaire G, Mouton Y, Capron A. Production of tumour necrosis factor-alpha (TNF-α) and interleukin-1 (IL-1) in patients with AIDS: Enhanced level of TNF-α is related to a higher cytotoxic activity. Clin Exp Immunol
3. Kobayashi S, Hamamoto Y, Kobayashi N, Yamamoto N. Serum level of TNF-α in HIV-infected individuals. AIDS
4. Godfried MH, van der Poll T, Weverling GJ, et al. Soluble receptors for tumor necrosis factor as predictors of progression to AIDS in asymptomatic human immunodeficiency virus type 1 infection. J Infect Dis
5. Matsuyama T, Kobayashi N, Yamamoto N. Cytokines and HIV infection: Is AIDS a tumor necrosis factor disease? AIDS
6. Grunfeld C, Kotler DP. Pathophysiology of the AIDS wasting
syndrome. AIDS Clin Rev
7. Aukrust P, Liabakk N-B, Muller F, Lien E, Espevik T, Freland SS. Serum levels of tumor necrosis factor-α (TNF-α) and soluble TNF receptors in human immunodeficiency virus type 1 infection: correlations to clinical, immunologic and virologic parameters. J Infect Dis
8. Clouse KA, Powell D, Washington I, et al. Monokine regulation of human immunodeficiency virus-1 expression in a chronically infected human T cell clone. J Immunol
9. Folks TM, et al. Tumor necrosis factor alpha induces expression of human immunodeficiency virus in a chronically infected T-cell clone. Proc Natl Acad Sci USA
10. Michihiko S, Yamamoto N, Shinozaki F, et al. Augmentation of in-vitro HIV replication in peripheral blood mononuclear cells of AIDS and ARC patients by tumour necrosis factor. Lancet
11. Duh EJ, Maury WJ, Folks TM, Fauci AS, Rabson AB. Tumor necrosis factor alpha activates human immunodeficiency virus type 1 through induction of nuclear factor binding to the NF-kappa B sites in the long terminal repeat. Proc Natl Acad Sci USA
12. Barnes PF, Fong SJ, Brennan PJ, et al. Local production of tumor necrosis factor and IFN-γ in tuberculous pleuritis. J Immunol
13. Tramontana JM, Utaipat U, Molloy A, et al. Thalidomide
treatment reduces TNF-α production and enhances weight gain in patients with pulmonary tuberculosis
. Mol Med
14. Valone SE, Rich E, Wallis RS, Ellner JJ. Expression of tumor necrosis factor in vitro by human mononuclear phagocytes stimulated with whole Mycobacterium bovis
BCG and mycobacterial antigens. Infect Immunol
15. Moreno C, Javerne J, Mehlert A, et al. Lipoarabinomannan from Mycobacterium tuberculosis
induces the production of tumour necrosis factor from human and murine macrophages. Clin Exp Immunol
16. Wallis RS, Amir-Tahmasseb M, Ellner JJ. Induction of interleukin 1 and tumor necrosis factor by mycobacterial proteins: the monocyte Western blot. Proc Natl Acad Sci USA
17. Zhang Y, Doerfler M, Lee TC, Guillemin B, Rom WN. Mechanisms of stimulation of interleukin-1 beta and tumor necrosis factor-alpha by Mycobacterium tuberculosis
components. J Clin Invest
18. Wallis RS, Vjecha M, Amir-Tahmasseb M, et al. Influence of tuberculosis
on human immunodeficiency virus (HIV-1): enhanced cytokine expression and elevated beta 2-microglobulin in HIV-1-associated tuberculosis
. J Infect Dis
19. Lederman MM, Georges DL, Kusner DJ, Mudido P, Giam CZ, Toossi Z. Mycobacterium tuberculosis
and its purified protein derivative activate expression of the human immunodeficiency virus. J AIDS
20. Zhang Y, Nakata K, Weiden M, Rom WN. Mycobacterium tuberculosis
enhances human immunodeficiency virus-1 replication by transcriptional activation at the long terminal repeat. J Clin Invest
21. Sampaio EP, Moreira AL, Sarno EN, Malta AM, Kaplan G. Prolonged treatment with recombinant interferon gamma induces erythema nodosum leprosum in lepromatous leprosy patients. J Exp Med
22. Makonkawkeyoon S, Limson-Pobre RN, Moreira AL, Schauf V, Kaplan G. Thalidomide
inhibits the replication of human immunodeficiency virus type 1. Proc Natl Acad Sci USA
23. Dewar RL, Highbarger HC, Sarmiento MD, et al. Application of branched DNA signal amplification to monitor human immunodeficiency virus type 1 burden in human plasma. J Infect Dis
24. Holodniy M, Mole L, Yen-Lieberman B, et al. Comparative stabilities of quantitative human immunodeficiency virus RNA in plasma from samples collected in Vacutainer CPT, Vacutainer PPT, and standard Vacutainer tubes. J Clin Microbiol
25. Kaplan G, Moreira AL. TNF-α regulation of HIV-1: biology and therapy. Res Immunol
26. Beutler B, Cerami A. The common mediator of shock, cachexia and tumor necrosis. Adv Immunol
27. Tracey KJ, Cerami A. Tumor necrosis factor in the malnutrition (cachexia) of infection and cancer. Am J Trop Med Hyg
1992;47:suppl. 1, Pt 2:2-7.
28. Rouzer CA, Cerami A. Hypertriglyceridemia associated with Trypanosoma brucei
infection in rabbits: role of defective triglyceride removal. Mol Biochem Parasitol
29. Poli G, Kinter A, Justement JS, et al. Tumor necrosis factor alpha functions in an autocrine manner in the induction of human immunodeficiency virus expression. Proc Natl Acad Sci USA
30. Sampaio EP, Sarno EN, Galilly R, Cohn ZA, Kaplan G. Thalidomide
selectively inhibits tumor necrosis factor alpha production by stimulated human monocytes. J Exp Med
31. Sampaio EP, Kaplan G, Miranda A, et al. The influence of thalidomide
on the clinical and immunologic manifestation of erythema nodosum leprosum. J Infect Dis
32. Williams I, Weller IVD, Malni A, Anderson J, Waters MF. Thalidomide
hypersensitivity in AIDS. Lancet
33. Gunzler V. Thalidomide
in human immunodeficiency virus (HIV) patients: a review of safety considerations. Drug Safety