Soluble ST2 as a marker for risk of mortality in HIV-infected subjects

Abstract

The present invention relates to materials and methods concerning the IL-1 receptor family protein ST2. Use of soluble ST2 as a marker for cardiovascular disease or disease outcome is provided, in particular as a marker of the risk of mortality.

Claims

1. A method of predicting a risk of mortality within one year following a cardiovascular event, of an HIV-infected subject, and, based on the risk of mortality, treating the HIV-infected subject using a suitable method, comprising (i) measuring, in an HIV-infected subject who has no pre-existing cardiovascular disease and who has not yet commenced anti-HIV therapy, a level of sST2 in a biological sample from said subject, (ii) comparing said level of sST2 with a control level derived from HIV-infected subjects with no pre-existing cardiovascular disease, wherein a level of sST2 in the subject sample which is higher than that of the control is indicative of an increased risk of mortality; and (iii) treating the HIV-infected subject with an aggressive cardiovascular disease treatment when an elevated level of sST2 is detected in the subject sample compared to the control and treating the HIV-infected subject with a conservative cardiovascular disease treatment when the level of sST2 detected in the subject sample is not elevated compared to the control.

2. The method of claim 1, wherein the biological sample is whole blood, plasma or serum.

3. The method of claim 1, wherein the aggressive treatment is one or both of surgical intervention and administering a multiple drug regime, and wherein the conservative treatment comprises one or both of lifestyle changes and administering a selected drug.

4. The method of claim 3, wherein the surgical intervention is selected from the group consisting of a coronary bypass procedure, valve repair or replacement, insertion of one or more stents and a heart transplant operation.

5. The method of claim 3, wherein the multiple drug regime includes two or more drugs selected from the group consisting of angiotensin receptor blockers, ACE inhibitors, anti-arrhythmics, antiplatelet therapeutics, aspirin, beta-blockers, calcium channel blockers, anti-thrombotic agents, cholesterol lowering agents, digitalis medicines, diuretics, inotropic agents and nitrates and the selected drug is a single drug selected from the group consisting of angiotensin receptor blockers, ACE inhibitors, anti-arrhythmics, antiplatelet therapeutics, aspirin, beta-blockers, calcium channel blockers, anti-thrombotic agents, cholesterol lowering agents, digitalis medicines, diuretics, inotropic agents and nitrates.

6. The method of claim 3, wherein the lifestyle changes are selected from the group consisting of increased exercise, stress reduction techniques, weight loss, dietary improvements increased exercise, stress reduction techniques, weight loss and dietary improvements.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows sST2 levels (pg/ml) in healthy donors (open bar) versus HIV-infected subjects (filled bar).

(2) FIG. 2 shows CD4 cell count versus ST2 level (pg/ml).

(3) FIG. 3 shows viral load versus ST2 level (pg/ml).

(4) FIG. 4 shows ST2 values in subjects with no history of CVE (cardiovascular event; open bar) and those with history of CVE (filled bar). The dark grey bars show ST2 level in CVE subjects in which a CVE occurred before ST2 measurement. The light grey bars show ST2 levels in subjects in which a CVE occurred after ST2 measurement.

EXPERIMENTAL SECTION

(5) Methods

(6) Cross-Sectional Study of Samples from HIV-Infected Patients and Healthy Donors

(7) In the cross-sectional study, 116 men and 38 female were enrolled from one center in Henri Mondor Hospital (Créteil, France). The study was approved by the local ethics committee. Blood was collected in tubes containing EDTA (potassium ethylenediamine tetra-acetic acid) for viral load and sST2. To establish reference values for sST2 biomarker, 35 HIV-uninfected healthy donors 21-73 years of age with BMI between 19.9 and 36.7 were recruited.

(8) SMART Study Samples

(9) The methods and results of the SMART trial have been published (El-Sadr et al, 2006, N Engl J Med 355:2283). Between January 2002 and January 2006, 5472 HIV-infected patients with a CD4.sup.+ cell count above 350 cells/μl were randomized to episodic ART (drug conservation) or continuous ART (viral suppression). Viral suppression patients taking ART at entry continued taking it, and those not taking it initiated ART after randomization. For the viral suppression group, available ART was to be used in an uninterrupted manner with the goal of maximal and continuous suppression of HIV replication. The experimental drug conservation strategy entailed intermittent use of ART for periods defined by CD4.sup.+ cell count until January 2006, at which time the study participants were advised to re-introduce ART due to an increased risk of opportunistic disease and risk. Patients were asked to consent to storing blood for future research, and only samples for consenting patients were used. The SMART study was approved by the ethics committee of each clinical site and of the University of Minnesota.

(10) Criteria for CVD events, defined as coronary heart disease (CHD), atherosclerotic non-CHD (stroke and peripheral arterial disease), congestive heart failure (CHF) and CVD or unwitnessed death, have been previously described (Phillips et al, 2008, Antivir Ther 13:177). For CVD cases with a blood specimen available for analysis (216 patients) and for controls (n=399) matched on the country, age, sex, sST2 was measured at baseline (study entry) on stored serum.

(11) Biomarker Methods

(12) sST2 assays were performed by ELISA (ST2/IL-1R4 R&D systems, Oxford, UK) according to the manufacturer's instructions with a coefficient of variation <10% and a lower limit of detection of 2 pg/ml and a upper limit of detection >2000 pg/ml. All samples were analyzed blinded to case and control status and to treatment group.

(13) Statistical Methods

(14) Odd ratios (OR) associated with baseline levels of sST2 for CVD were estimated using conditional logistic regression. Models which adjusted for race, use of HAART, viral load, smoking, prior CVD, metabolic factors, co-infection, major ECG abnormalities, hsCRP, IL-6 and D-dimer in addition to matching factors were considered as well as unadjusted models (matching factors only).
Results
sST2 is Increased in HIV Infected Patients

(15) Serum sST2 levels were analyzed in 131 HIV infected patients and 35 healthy volunteers. As shown in FIG. 1, serum sST2 concentration, expressed as median (interquartile range [IQR]) is significantly higher in HIV infected patients (174.8 pg/ml [IQR 129.9 to 235.3] than in the healthy volunteers (24.9 pg/ml [IQR 20.3 to 48.05]; p<0.001). sST2 does not correlate with age (data not shown). The relationships between sST2, viral load and CD4.sup.+ T cell count are shown in FIGS. 2 and 3. No association is found with CD4.sup.+ T count (r=−0.05, p=0.51) and viral load (r=0.037, p=0.64).

(16) Among these 131 infected patients, 21 patients had a cardiovascular event (CVE) either before or after sST2 measurement (12 myocardial infarctions, 4 ischemic strokes, 2 peripheral vascular diseases, 2 left ventricular insufficiency, 1 pulmonary hypertension). As previously reported for HIV negative patients, a clear association between history of CVE and elevated levels of sST2 in HIV infected patients is observed (FIG. 4). However, when a CVE event occurred before sST2 measurement, the highest sST2 levels were observed, suggesting that sST2 may be a predictive factor of CVD.

(17) sST2 is Correlated with Cardiovascular Disease Mortality in HIV-Infected Subjects

(18) Table 1 gives baseline characteristics for CVD cases and matched controls in HIV-infected subjects. In univariate analyses, prior AIDS, current smoking status, diabetes, prior CVD, major resting ECG abnormality and use of blood pressure lowering drugs were associated with an increased risk of CVD. In addition, HDL cholesterol was lower in CVD cases compared to controls. Baseline hsCRP, IL-6 and D-dimer were significantly higher in CVD cases than controls. At baseline, median sST2 level was 85.0 pg/ml (IQR 24 to 247) in patients who developed a CVD event and 78.0 pg/ml (IQR 22 to 208) in the matched control patients. There was no difference in median sST2 levels with age, gender and co-infection with hepatitis B/C. No difference was observed in median ST2 levels by time of CVD event.

(19) As in the first study, there was no association between sST2 and CD4.sup.+ T count (r=−0.02, p=0.6) and HIV RNA levels (r=−0.03, p=0.39). At baseline, sST2 were correlated with IL-6 (r=0.11, p=0.008). However, there was no association between sST2 and hsCRP (r=0.05, p=0.26) and D-dimer (r=0.07, p=0.09). A significant correlation were found between sST2 and HDL-cholesterol (r=−0.09, p=0.03) and total/HDL ratio (r=0.10, p=0.02). There was a significant correlation between sST2 and triglycerides (r=0.08, p=0.04) and Apo/Apo A1 Ration (r=0.09, p=0.03).

(20) TABLE-US-00001 TABLE 1 Baseline Characteristics, cardiovascular disease risk factors and lipid profile of cardiovascular disease cases and matched controls with ST2 results CVD Cases Controls (N = 216) (N = 399) P-value.sup.* Treatment group (% DC) 56.5 45.6 0.01 Demographics Age (median, IQR) 50 (44, 56)  49 (43, 56) NA Gender (% female) 18.1 17.8 NA Black (%) 38.0 37.1 0.90 CD4+ (cells/mm.sup.3) 576 (456, 839) 635 (483, 0.73 (median, IQR) 803) CD4+ nadir (cells/mm.sup.3) 233 (117, 351) 244 (148, 0.81 (median, IQR) 346) HIV-RNA ≦ 400 67.0 66.8 0.93 copies/mL (%) Prior AIDS-related 37.5 25.1 0.001 illnesses (%) Hepatitis B (%) 1.4 1.3 0.80 Hepatitis C (%) 20.8 20.3 0.83 CVD Risk Factors Current smoker (%) 52.3 40.9 0.009 Diabetes (%) 16.7 9.8 0.01 Prior CVD (%) 13.9 5.5 <0.001 Major ECG abnormality (%) 21.2 10.5 <0.001 Blood pressure lowering 44.0 31.6 0.001 drugs (%) Lipid lowering drugs (%) 28.7 24.3 0.29 Lipids Total cholesterol (mg/dl)  195 (169, 233) 193 (168, 0.77 (median, IQR) 232) HDL cholesterol (mg/dl) 38 (31, 48) 42 (33, 52) 0.03 (median, IQR) LDL cholesterol (mg/dl) 111 (88, 141) 110 (89,  0.80 (median, IQR) 134) Triglycerides (mg/dl)  190 (133, 312) 181 (124, 0.59 (median, IQR) 300) Total/HDL cholesterol  5.5 (3.9, 6.9) 4.7 (3.5, 6.1) 0.17 (median, IQR) Biomarkers hsCRP (μg/mL) 3.66 (1.54, 2.18 (0.92, <0.001 (median, IQR) 7.75) 4.93) IL-6 (pg/mL) 3.20 (1.98, 2.22 (1.44, <0.001 (median, IQR) 4.91) 3.53) D-dimer (μg/mL) 0.31 (0.20, 0.24 (0.15, 0.001 (median, IQR) 0.60) 0.47) ST2 (pg/mL) (median, IQR)  85 (24, 247) 78 (22, 208) 0.35 .sup.*P-value obtained from univariate conditional logistic model. Biomarkers are natural log transformed

(21) Table 2 summarizes the ORs of the upper quartiles versus the lower quartile of sST2. The unadjusted OR for CVD for the highest quartile (4.2% of cases and 4% of controls) versus lower quartile (12.5% of cases and 15.3% of controls) was 1.2 (95% Cl: 0.8 to 2.5; p=0.22). After adjustment for baseline covariates this OR was 1.6 (95% Cl: 0.8 to 3.2; p=0.19). Moreover, after adjustment for covariates that also included baseline levels of log.sub.e IL-6, hsCRP and D-dimer, this OR was 1.7 (95% Cl: 0.8-3.6, p=0.2).

(22) TABLE-US-00002 TABLE 2 Odds ratio for cardiovascular disease for upper versus lower quartile of ST2 levels ST2 level (pg/mL) OR (95% Cl) p-value Unadjusted <2 (ref.) — —  2-22 1.4 (0.7-3.0) 0.33 23-79 1.2 (0.6-2.1) 0.62 80-221 1.2 (0.7-2.1) 0.59 222+ 1.4 (0.8-2.5) 0.22 Adjusted (1) <2 (ref.) — —  2-22 1.8 (0.8-4.1) 0.16 23-79 1.1 (0.6-2.3) 0.71 80-221 1.2 (0.6-2.3) 0.61 222+ 1.6 (0.8-3.2) 0.19 Adjusted (2) <2 (ref.) — —  2-22 1.7 (0.7-4.2) 0.27 23-79 1.0 (0.5-2.1) 0.97 80-221 1.0 (0.5-2.2) 0.97 222+ 1.7 (0.8-3.6) 0.20 (1) adjusted for age, race, ART and HIV-RNA, smoking, prior CVD, diabetes, BP lowering drugs, total/HDL cholesterol, CD4+, BMI and major ECG abnormalities (2) adjusted for (1) plus log.sub.e hsCRP, IL-6 and D-dimer

(23) Median sST2 for cases and controls by type of event are given in Table 3. Of the 216 CVD events, 53 patients died. Thirty three of the deaths were not preceded by a non-fatal event. Twenty deaths were preceded by a non-fatal event. Median sST2 levels are particularly high among patients who died. Median sST2 was 114 pg/ml (IQR 37 to 359) in patients who died without history of CVD as compared to 78 pg/ml (IQR 32 to 149) in matched controls, and 158 pg/ml (IQR 48 to 289) in patients who died with history of CVD as compared to 94 pg/ml (IQR 27 to 228) in matched controls. For the patients suffering non-fatal events that did not subsequently die (163 cases of 216 CVD cases), there were no differences between sST2 levels for cases (72 pg/ml; [IQR 23 to 222]) and matched controls (77 pg/ml; [IQR 20 to 220]). Moreover, the unadjusted OR was greater when only the 33 CVD/unwitnessed deaths were considered: 2.8 (95% Cl: 0.8 to 10.3; p=0.12).

(24) TABLE-US-00003 TABLE 3 ST2 levels (pg/ml) by CVD event/death and case history of CVD or CHF Cases Matched Controls N Median (IQR) N Median (IQR) CVD or unwitnessed deaths Overall 33 114 (37, 359) 61 78 (32, 149) Prior CVD 8 114 (43, 158) 15 106 (52, 157)  Prior CHF 3 102 (85, 179) 6 128 (62, 208)  No prior CVD or CHF 24 136 (35, 498) 44 68 (18, 146) Non-fatal CVD and subsequently died Overall 20 158 (48, 289) 37 94 (27, 228) Prior CVD 3 16 (2, 275) 6 84 (48, 119) Prior CHF 2 140 (4, 275)  4 54 (28, 144) No prior CVD or CHF 16 170 (91, 336) 29 94 (27, 366) Non-fatal CVD and did not die Overall 163 72 (23, 222) 301 77 (20, 220) Prior CVD 19 91 (50, 107) 35 70 (8, 286)  Prior CHF 2 1054 (107, 2000) 3 68 (2, 2000) No prior CVD or CHF 144 69 (22, 225) 266 77 (22, 220)