- Email: [email protected]
Circulating Long Noncoding RNAs in Personalized Medicine
2914
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
NEP activity was significantly reduced after therapy
*Department of Cardiology
conversion to ARNI (68.7 [interquartile range (IQR):
Mediical University of Vienna
45.9 to 169.6] ng[Ang1-7]/ml/h vs. 35.9 [IQR: 17.1 to
Währinger Gürtel 18-20
48.2] ng[Ang1-7]/ml/h; p ¼ 0.004), confirming an effi-
1090 Vienna
cient uptake of the drug. Likewise NT-proBNP levels
Austria
decreased significantly (1,933 [IQR: 862 to 3,134] pg/ml
E-mail: [email protected]
vs. 1,309 [IQR: 469 to 2,184] pg/ml; p ¼ 0.041) in line
http://dx.doi.org/10.1016/j.jacc.2016.10.017
with
observations
from
the
clinical
study
(2).
Interestingly there was a trend for higher renin levels after induction of ARNI (425 [IQR: 33 to 944] mIE/ml vs. 1,016 [IQR: 132 to 2,016] m IE/ml; p ¼ 0.071). RAS fingerprints are displayed in Figure 1. For the converted ACE-I group a significant increase in AngII, Ang1-5, and AngIV was apparent. When converting from ARB to ARNI (introduction of NEP inhibition [NEPi]), the proportions of angiotensins remained similar.
However,
patients
displayed
higher
concentrations of all systemic angiotensins on ARNI with significantly higher levels of the downstream metabolites Ang1-5 and AngIV. Although NEP is principally a membrane-bound endopeptidase, the detection of therapy response in plasma suggests that circulating NEP might serve as surrogate for systemic NEP actions and that soluble NEP may be a useful biomarker in HFrEF, as already discussed (3). The introduction of NEPi was not associated
Please note: The authors thank Susanne Weinreder from the Medical University of Vienna for her valuable help with patient recruitment and data collection. This study was supported by an unrestricted grant of the Austrian Heart Foundation (Österreichischer Herzfonds). The study has been partially funded by an unrestricted grant of the Austrian Heart Foundation (Österreichischer Herzfonds). Dr. Wurm has received speaker fees from Novartis. Dr. Poglitsch is an employee of Attoquant Diagnostics (Vienna, Austria). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
REFERENCES 1. Yancy CW, Jessup M, Bozkurt B, et al. 2016 ACC/AHA/HFSA focused update on New Pharmacological Therapy for Heart Failure: an update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. J Am Coll Cardiol 2016;68:1476–88. 2. McMurray JJ, Packer M, Desai AS, et al. Angiotensin-neprilysin inhibition versus enalapril in heart failure. N Engl J Med 2014;371:993–1004. 3. Bayes-Genis A, Barallat J, Richards AM. A test in context: neprilysin: function, inhibition, and biomarker. J Am Coll Cardiol 2016;68:639–53. 4. Gu J, Noe A, Chandra P, et al. Pharmacokinetics and pharmacodynamics of LCZ696, a novel dual-acting angiotensin receptor-neprilysin inhibitor (ARNi). J Clin Pharmacol 2010;50:401–14.
with qualitative alterations in systemic RAS patterns but rather with elevation of all angiotensins. The profound clinical benefits of NEPi may lie in its net effects on the interplay of vasoactive peptides, yet it seems that NEPi also triggers RAS activation. Potential controversial effects of NEPi have long been discussed before the conduction of the PARADIGM-HF (AngiotensinNeprilysin Inhibition versus Enelapril in Heart Failure) trial, and pharmacokinetic data showed that also
Circulating Long Noncoding RNAs in Personalized Medicine Response to Pioglitazone Therapy in Type 2 Diabetes
LCZ696 (ARNI) leads to a significant dose-dependent elevation of AngII but also renin in healthy volunteers (4). RAS activation affects all angiotensin metabolites equally, including AngII and Ang1-7, which in their higher concentration could exert beneficial RAS effects alongside an efficient AT1R blockade. Excess of the downstream angiotensins as AngIV, which is discussed to
exert
vasoprotective
properties,
enhance beneficial ARNI effects. Noemi Pavo, MD Raphael Wurm, MD Georg Goliasch, MD Johannes Franz Novak, Cand Med Guido Strunk, PhD Mariann Gyöngyösi, MD Marko Poglitsch, PhD Marcus D. Säemann, MD *Martin Hülsmann, MD
may
further
Pioglitazone is a thiazolidinedione insulin sensitizer that improves left ventricle (LV) diastolic function in patients with type 2 diabetes mellitus (T2DM) (1). Unfortunately, the clinical use of pioglitazone is limited by the risk of adverse effects (2). Predictive tools
are
essential
to
monitor
therapeutic
effectiveness. Long noncoding RNAs (lncRNAs) are novel
biomarkers
of
cardiac
dysfunction
(3).
Nonetheless, no study has investigated the use of circulating lncRNAs as biomarkers of therapeutic efficiency. Here, we hypothesize that pre-treatment levels of circulating lncRNAs predict the response to pioglitazone therapy. The PIRAMID (Pioglitazone Influence on tRiglyceride Accumulation in the Myocardium In Diabetes) study was a prospective intervention designed to evaluate the effect of pioglitazone on myocardial
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
T A B L E 1 Associations Between Percentage of Change in Parameters of Diastolic Function After 24 Weeks of Pioglitazone Therapy and
Potential Predictors at Baseline
DE-dec peak b –0.014
Age
p Value
0.955
DE-decmean b 0.145
p Value
0.554
DE/A Peak Flow b –0.073
p Value
0.767
DE/E0 b –0.389
p Value
0.110
Body mass index, kg/m2
–0.009
0.971
–0.126
0.608
0.077
0.755
–0.117
0.644
HbA1c, %
–0.153
0.531
–0.132
0.591
0.002
0.994
–0.050
0.844
LDL cholesterol, mmol/l
–0.440
0.078
–0.242
0.350
–0.489
0.047*
–0.049
0.859
HDL cholesterol, mmol/l
0.061
0.810
0.257
0.304
0.093
0.714
0.111
0.670
–0.246
0.325
–0.308
0.213
–0.350
0.155
0.198
0.446
Triglycerides, mmol/l Systolic blood pressure, mm Hg
0.045
0.854
0.133
0.587
0.091
0.710
–0.241
0.336
Diastolic blood pressure, mm Hg
0.392
0.097
0.336
0.160
0.415
0.078
–0.219
0.382
Heart rate, beats/min
0.109
0.657
–0.105
0.670
0.269
0.266
–0.055
0.827
LV end-systolic volume, ml
0.273
0.258
0.099
0.688
0.344
0.150
–0.181
0.472
LV end-diastolic volume, ml
–0.190
0.436
–0.170
0.485
–0.035
0.887
0.178
0.479
LV ejection fraction, %
–0.585
0.008*
–0.291
0.228
–0.537
0.018*
0.441
0.067
E-decpeak, ml/s2 *10–3
–0.273
0.259
–0.336
0.160
–0.480
0.037*
0.245
0.327
E-decmean, ml/s2 *10–3
0.269
0.266
0.476
0.039*
0.328
0.171
–0.075
0.766
–0.336
0.160
–0.480
0.037*
0.245
0.327
0.012
0.956
E/A peak flow E/E0
0.436
0.071
0.556
0.017*
–0.596
0.009*
0.718
–0.214
0.380
–0.045
0.854
–0.144
0.654
–0.440
0.068
–0.467
0.051
–0.380
0.120
0.582
0.009*
0.584
0.492
NT-proBNP, ng/l
–0.089
us-CRP, mg/l SENCR, a.u
0.038*
0.009*
0.604
0.006*
0.351 –0.621
0.167 0.006*
*Statistically significant. a.u ¼ arbitrary units; E-decmean ¼ mean deceleration gradient of the early filling phase; E-decpeak ¼ peak deceleration gradient of the early filling phase; E/A ¼ early filling phase and atrial contraction; E/E0 ¼ left ventricle filling pressures; HbA1c ¼ glycosylated hemoglobin; HDL ¼ high-density lipoprotein; LDL ¼ low-density lipoprotein; LV ¼ left ventricular; NT-proBNP ¼ N-terminal pro–B-type natriuretic peptide; SENCR ¼ smooth muscle and endothelial cell-enriched migration/differentiation-associated long noncoding RNA; us-CRP ¼ ultrasensitive C-reactive protein.
function in patients with well-controlled T2DM (1).
associated with improved LV diastolic function (1).
Seventy-eight T2DM men were assigned to pioglita-
Neither fluid retention nor congestive heart failure
zone (30 mg/day) or metformin (2000 mg/day)
was detected during the study (1).
for 24 weeks (1). A detailed description of the
Baseline levels of circulating SENCR (smooth
characteristics of patients and study procedures
muscle and endothelial cell-enriched migration/
has been previously published (1). The study was
differentiation-associated long noncoding RNA) were
conducted in full compliance with the Declaration
associated with the percentage of change in 4 LV
of Helsinki and with the approval of the ethics
diastolic function parameters, peak deceleration
committees. Written informed consent was obtained
gradient of the early filling phase, mean deceleration
from all participants. A panel of lncRNAs was
gradient of the early filling phase, ratio of the peak
evaluated: LIPCAR (Long intergenic non-coding RNA
filling rates of the early filling phase and atrial
predicting cardiac remodeling), MALAT1 (Metastasis-
contraction, and LV filling pressures (E/E 0 ) after
associated
24 weeks of pioglitazone therapy (Table 1). The
lung
adenocarcinoma
transcript
1),
MIAT (Myocardial infarction-associated transcript),
observed associations remained statistically signifi-
H19, HOTAIR (HOX transcript antisense RNA) and
cant after correction for multiple comparisons using
SENCR (Smooth muscle and endothelial cell-enriched
Benjamini–Hochberg false discovery rate test. The
migration/differentiation-associated long noncoding
association observed was higher than that found for
RNA). LncRNA quantification was performed as
other variables. No association was found in the
previously described by our group (4).
metformin group (p > 0.050 for all associations).
In an initial study cohort (48 randomly selected
Baseline circulating SENCR levels were higher in the
samples, mean 57.5 5.4 years of age, glycosylated
responder group (those patients showing a decrease
hemoglobin 7.2 1.0%), no differences were observed
in E/E 0 >10%; 45%) versus the nonresponder group
between study groups in baseline clinical character-
(responder ¼ 2.13 0.50 arbitrary units; non-
istics or expression levels of selected lncRNAs
responder ¼ 1.59 0.25 arbitrary units; p ¼ 0.020).
(p > 0.050 for all comparison). Pioglitazone was
No differences were observed for other variables
2915
2916
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
(p > 0.050 for all comparisons). An association between the response to therapy (decrease in E/E
0
>10%) and inclusion in the highest tertile of baseline SENCR was observed (odds ratio [95% confidence interval]: 0.067 [0.005 to 0.823]; p ¼ 0.035; area under the curve [95% CI]: 0.763 [0.558 to 0.967]).
study was supported by Eli Lilly, the Netherlands, which has a partnership with Takeda, the manufacturer of pioglitazone. Metformin tablets and matching placebos were kindly provided by Merck, the Netherlands. Dr. Thum filed patents about the diagnostic use of lncRNAs in cardiovascular diseases. Dr. Thum is founder of Cardior Pharmaceutics GmbH. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Llorente-Cortes and Thum contributed equally to this work.
In a validation cohort of 30 men with wellcontrolled T2DM (mean 55.0 5.7 years of age, glycosylated hemoglobin 7.0 1.0%), SENCR was associated with the percentage change in E/E 0 ( b ¼ –0.708, p ¼ 0.007). Furthermore, pre-treatment SENCR levels were higher in the responder group (responder ¼ 2.10 0.44 arbitrary units; nonresponder ¼ 1.53 0.27 arbitrary units; p ¼ 0.014). We suggest for the first time that circulating lncRNAs are blood-based biomarkers for therapeutic guidance of cardiovascular conditions. We demonstrate that: 1) pre-treatment levels of circulating SENCR are associated with the rate of change in dia-
REFERENCES 1. van der Meer RW, Rijzewijk LJ, de Jong HW, et al. Pioglitazone improves cardiac function and alters myocardial substrate metabolism without affecting cardiac triglyceride accumulation and high-energy phosphate metabolism in patients with well-controlled type 2 diabetes mellitus. Circulation 2009;119:2069–77. 2. Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. Pioglitazone and risk of cardiovascular events in patients with type 2 diabetes mellitus: a metaanalysis of randomized trials. JAMA 2007;298:1180–8. 3. Boon RA, Jae N, Holdt L, Dimmeler S. Long noncoding RNAs: from clinical genetics to therapeutic targets? J Am Coll Cardiol 2016;67:1214–26. 4. Kumarswamy R, Bauters C, Volkmann I, et al. Circulating long noncoding RNA, LIPCAR, predicts survival in patients with heart failure. Circ Res 2014; 114:1569–75.
stolic function; 2) levels of circulating SENCR before therapy allow distinction of responders from nonresponders; and 3) SENCR is a better predictor of the response to therapy than are other traditional clinical
Trimethylamine N-Oxide in Seafood
variables. Thus, assessment of lncRNA levels could constitute a major breakthrough in the development
Rotten or Forgotten?
of personalized medicine to guide medical therapy, with important implications in terms of adverse effects and treatment costs.
We read the paper by Senthong et al. (1) with great interest. The research article was accompanied by
*David de Gonzalo-Calvo, PhD Franziska Kenneweg, MSc Claudia Bang, PhD Rocío Toro, MD, PhD Rutger W. van der Meer, MD, PhD Luuk J. Rijzewijk, MD Johannes W.A. Smit, MD, PhD Hildo J. Lamb, MD, PhD *Vicenta Llorente-Cortes, PhD Thomas Thum, MD, PhD
an editorial comment by Schmid-Schönbein (2).
*Cardiovascular Research Center (CSIC-ICCC)
directly involved in development of atherosclerosis
Biomedical Research Institute Sant Pau (IIB Sant Pau)
therefore seems like a paradox, as intake of seafood
Av. Sant Antoni Maria Claret 167
is generally accepted as cardioprotective.
Pavelló del Convent
During
recent
years,
trimethylamine
N-oxide
(TMAO) has emerged as a putative risk factor linking development of atherosclerosis to intake of socalled unhealthy foods (beef, egg) containing the TMAO precursors carnitine and choline. However, seafood is an important dietary source of free TMAO, in addition to the TMAO precursors carnitine and choline, a fact that appears to be commonly overlooked. To us, the hypothesis that TMAO is
In fish and marine invertebrates, TMAO exerts
08025 Barcelona
important osmoregulatory functions, and the mole-
Spain
cule can be found in high concentrations in these
E-mail: [email protected] OR [email protected]
organisms (3). TMAO from fish may be absorbed
http://dx.doi.org/10.1016/j.jacc.2016.10.014
directly from the gastrointestinal tract and give rise
Please note: The authors are part of the Cardiolinc network. The authors thank Ignasi Gich, MD, PhD (IIB Sant Pau, Barcelona, Spain) and Pablo Martínez Camblor, PhD (Geisel School of Medicine, Hanover, New Hampshire) for their excellent statistical support. This work was supported by the Merck Health Foundation (Merck Serono Research Award in Cardiometabolism 2014), Fundación Pública Andaluza Progreso y Salud (PI-0011/2014), Instituto de Salud Carlos III (FIS PI14/01729), ERC Consolidator grant LongHeart, DFG Th903/10-1, the IFB-Tx (BMBF), the 7th FP-funded EU project HOMAGE and the Foundation Leducq; and cofinanced by the European Fund for Regional Development (E.F.R.D.) and Fundació Marató TV3 (201521 10). The PIRAMID
to plasma levels that are orders of magnitude higher than after beef or egg intake (4). Indeed, TMAO has previously been suggested as a marker of seafood consumption (5). In contrast, TMAO derived from carnitine and choline is produced in the liver, by oxidation of trimethylamine (TMA) produced by gut bacteria. Hepatic oxidation of TMA to TMAO may
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
NEP activity was significantly reduced after therapy
*Department of Cardiology
conversion to ARNI (68.7 [interquartile range (IQR):
Mediical University of Vienna
45.9 to 169.6] ng[Ang1-7]/ml/h vs. 35.9 [IQR: 17.1 to
Währinger Gürtel 18-20
48.2] ng[Ang1-7]/ml/h; p ¼ 0.004), confirming an effi-
1090 Vienna
cient uptake of the drug. Likewise NT-proBNP levels
Austria
decreased significantly (1,933 [IQR: 862 to 3,134] pg/ml
E-mail: [email protected]
vs. 1,309 [IQR: 469 to 2,184] pg/ml; p ¼ 0.041) in line
http://dx.doi.org/10.1016/j.jacc.2016.10.017
with
observations
from
the
clinical
study
(2).
Interestingly there was a trend for higher renin levels after induction of ARNI (425 [IQR: 33 to 944] mIE/ml vs. 1,016 [IQR: 132 to 2,016] m IE/ml; p ¼ 0.071). RAS fingerprints are displayed in Figure 1. For the converted ACE-I group a significant increase in AngII, Ang1-5, and AngIV was apparent. When converting from ARB to ARNI (introduction of NEP inhibition [NEPi]), the proportions of angiotensins remained similar.
However,
patients
displayed
higher
concentrations of all systemic angiotensins on ARNI with significantly higher levels of the downstream metabolites Ang1-5 and AngIV. Although NEP is principally a membrane-bound endopeptidase, the detection of therapy response in plasma suggests that circulating NEP might serve as surrogate for systemic NEP actions and that soluble NEP may be a useful biomarker in HFrEF, as already discussed (3). The introduction of NEPi was not associated
Please note: The authors thank Susanne Weinreder from the Medical University of Vienna for her valuable help with patient recruitment and data collection. This study was supported by an unrestricted grant of the Austrian Heart Foundation (Österreichischer Herzfonds). The study has been partially funded by an unrestricted grant of the Austrian Heart Foundation (Österreichischer Herzfonds). Dr. Wurm has received speaker fees from Novartis. Dr. Poglitsch is an employee of Attoquant Diagnostics (Vienna, Austria). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
REFERENCES 1. Yancy CW, Jessup M, Bozkurt B, et al. 2016 ACC/AHA/HFSA focused update on New Pharmacological Therapy for Heart Failure: an update of the 2013 ACCF/AHA Guideline for the Management of Heart Failure: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. J Am Coll Cardiol 2016;68:1476–88. 2. McMurray JJ, Packer M, Desai AS, et al. Angiotensin-neprilysin inhibition versus enalapril in heart failure. N Engl J Med 2014;371:993–1004. 3. Bayes-Genis A, Barallat J, Richards AM. A test in context: neprilysin: function, inhibition, and biomarker. J Am Coll Cardiol 2016;68:639–53. 4. Gu J, Noe A, Chandra P, et al. Pharmacokinetics and pharmacodynamics of LCZ696, a novel dual-acting angiotensin receptor-neprilysin inhibitor (ARNi). J Clin Pharmacol 2010;50:401–14.
with qualitative alterations in systemic RAS patterns but rather with elevation of all angiotensins. The profound clinical benefits of NEPi may lie in its net effects on the interplay of vasoactive peptides, yet it seems that NEPi also triggers RAS activation. Potential controversial effects of NEPi have long been discussed before the conduction of the PARADIGM-HF (AngiotensinNeprilysin Inhibition versus Enelapril in Heart Failure) trial, and pharmacokinetic data showed that also
Circulating Long Noncoding RNAs in Personalized Medicine Response to Pioglitazone Therapy in Type 2 Diabetes
LCZ696 (ARNI) leads to a significant dose-dependent elevation of AngII but also renin in healthy volunteers (4). RAS activation affects all angiotensin metabolites equally, including AngII and Ang1-7, which in their higher concentration could exert beneficial RAS effects alongside an efficient AT1R blockade. Excess of the downstream angiotensins as AngIV, which is discussed to
exert
vasoprotective
properties,
enhance beneficial ARNI effects. Noemi Pavo, MD Raphael Wurm, MD Georg Goliasch, MD Johannes Franz Novak, Cand Med Guido Strunk, PhD Mariann Gyöngyösi, MD Marko Poglitsch, PhD Marcus D. Säemann, MD *Martin Hülsmann, MD
may
further
Pioglitazone is a thiazolidinedione insulin sensitizer that improves left ventricle (LV) diastolic function in patients with type 2 diabetes mellitus (T2DM) (1). Unfortunately, the clinical use of pioglitazone is limited by the risk of adverse effects (2). Predictive tools
are
essential
to
monitor
therapeutic
effectiveness. Long noncoding RNAs (lncRNAs) are novel
biomarkers
of
cardiac
dysfunction
(3).
Nonetheless, no study has investigated the use of circulating lncRNAs as biomarkers of therapeutic efficiency. Here, we hypothesize that pre-treatment levels of circulating lncRNAs predict the response to pioglitazone therapy. The PIRAMID (Pioglitazone Influence on tRiglyceride Accumulation in the Myocardium In Diabetes) study was a prospective intervention designed to evaluate the effect of pioglitazone on myocardial
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
T A B L E 1 Associations Between Percentage of Change in Parameters of Diastolic Function After 24 Weeks of Pioglitazone Therapy and
Potential Predictors at Baseline
DE-dec peak b –0.014
Age
p Value
0.955
DE-decmean b 0.145
p Value
0.554
DE/A Peak Flow b –0.073
p Value
0.767
DE/E0 b –0.389
p Value
0.110
Body mass index, kg/m2
–0.009
0.971
–0.126
0.608
0.077
0.755
–0.117
0.644
HbA1c, %
–0.153
0.531
–0.132
0.591
0.002
0.994
–0.050
0.844
LDL cholesterol, mmol/l
–0.440
0.078
–0.242
0.350
–0.489
0.047*
–0.049
0.859
HDL cholesterol, mmol/l
0.061
0.810
0.257
0.304
0.093
0.714
0.111
0.670
–0.246
0.325
–0.308
0.213
–0.350
0.155
0.198
0.446
Triglycerides, mmol/l Systolic blood pressure, mm Hg
0.045
0.854
0.133
0.587
0.091
0.710
–0.241
0.336
Diastolic blood pressure, mm Hg
0.392
0.097
0.336
0.160
0.415
0.078
–0.219
0.382
Heart rate, beats/min
0.109
0.657
–0.105
0.670
0.269
0.266
–0.055
0.827
LV end-systolic volume, ml
0.273
0.258
0.099
0.688
0.344
0.150
–0.181
0.472
LV end-diastolic volume, ml
–0.190
0.436
–0.170
0.485
–0.035
0.887
0.178
0.479
LV ejection fraction, %
–0.585
0.008*
–0.291
0.228
–0.537
0.018*
0.441
0.067
E-decpeak, ml/s2 *10–3
–0.273
0.259
–0.336
0.160
–0.480
0.037*
0.245
0.327
E-decmean, ml/s2 *10–3
0.269
0.266
0.476
0.039*
0.328
0.171
–0.075
0.766
–0.336
0.160
–0.480
0.037*
0.245
0.327
0.012
0.956
E/A peak flow E/E0
0.436
0.071
0.556
0.017*
–0.596
0.009*
0.718
–0.214
0.380
–0.045
0.854
–0.144
0.654
–0.440
0.068
–0.467
0.051
–0.380
0.120
0.582
0.009*
0.584
0.492
NT-proBNP, ng/l
–0.089
us-CRP, mg/l SENCR, a.u
0.038*
0.009*
0.604
0.006*
0.351 –0.621
0.167 0.006*
*Statistically significant. a.u ¼ arbitrary units; E-decmean ¼ mean deceleration gradient of the early filling phase; E-decpeak ¼ peak deceleration gradient of the early filling phase; E/A ¼ early filling phase and atrial contraction; E/E0 ¼ left ventricle filling pressures; HbA1c ¼ glycosylated hemoglobin; HDL ¼ high-density lipoprotein; LDL ¼ low-density lipoprotein; LV ¼ left ventricular; NT-proBNP ¼ N-terminal pro–B-type natriuretic peptide; SENCR ¼ smooth muscle and endothelial cell-enriched migration/differentiation-associated long noncoding RNA; us-CRP ¼ ultrasensitive C-reactive protein.
function in patients with well-controlled T2DM (1).
associated with improved LV diastolic function (1).
Seventy-eight T2DM men were assigned to pioglita-
Neither fluid retention nor congestive heart failure
zone (30 mg/day) or metformin (2000 mg/day)
was detected during the study (1).
for 24 weeks (1). A detailed description of the
Baseline levels of circulating SENCR (smooth
characteristics of patients and study procedures
muscle and endothelial cell-enriched migration/
has been previously published (1). The study was
differentiation-associated long noncoding RNA) were
conducted in full compliance with the Declaration
associated with the percentage of change in 4 LV
of Helsinki and with the approval of the ethics
diastolic function parameters, peak deceleration
committees. Written informed consent was obtained
gradient of the early filling phase, mean deceleration
from all participants. A panel of lncRNAs was
gradient of the early filling phase, ratio of the peak
evaluated: LIPCAR (Long intergenic non-coding RNA
filling rates of the early filling phase and atrial
predicting cardiac remodeling), MALAT1 (Metastasis-
contraction, and LV filling pressures (E/E 0 ) after
associated
24 weeks of pioglitazone therapy (Table 1). The
lung
adenocarcinoma
transcript
1),
MIAT (Myocardial infarction-associated transcript),
observed associations remained statistically signifi-
H19, HOTAIR (HOX transcript antisense RNA) and
cant after correction for multiple comparisons using
SENCR (Smooth muscle and endothelial cell-enriched
Benjamini–Hochberg false discovery rate test. The
migration/differentiation-associated long noncoding
association observed was higher than that found for
RNA). LncRNA quantification was performed as
other variables. No association was found in the
previously described by our group (4).
metformin group (p > 0.050 for all associations).
In an initial study cohort (48 randomly selected
Baseline circulating SENCR levels were higher in the
samples, mean 57.5 5.4 years of age, glycosylated
responder group (those patients showing a decrease
hemoglobin 7.2 1.0%), no differences were observed
in E/E 0 >10%; 45%) versus the nonresponder group
between study groups in baseline clinical character-
(responder ¼ 2.13 0.50 arbitrary units; non-
istics or expression levels of selected lncRNAs
responder ¼ 1.59 0.25 arbitrary units; p ¼ 0.020).
(p > 0.050 for all comparison). Pioglitazone was
No differences were observed for other variables
2915
2916
JACC VOL. 68, NO. 25, 2016
Letters
DECEMBER 27, 2016:2911–9
(p > 0.050 for all comparisons). An association between the response to therapy (decrease in E/E
0
>10%) and inclusion in the highest tertile of baseline SENCR was observed (odds ratio [95% confidence interval]: 0.067 [0.005 to 0.823]; p ¼ 0.035; area under the curve [95% CI]: 0.763 [0.558 to 0.967]).
study was supported by Eli Lilly, the Netherlands, which has a partnership with Takeda, the manufacturer of pioglitazone. Metformin tablets and matching placebos were kindly provided by Merck, the Netherlands. Dr. Thum filed patents about the diagnostic use of lncRNAs in cardiovascular diseases. Dr. Thum is founder of Cardior Pharmaceutics GmbH. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Llorente-Cortes and Thum contributed equally to this work.
In a validation cohort of 30 men with wellcontrolled T2DM (mean 55.0 5.7 years of age, glycosylated hemoglobin 7.0 1.0%), SENCR was associated with the percentage change in E/E 0 ( b ¼ –0.708, p ¼ 0.007). Furthermore, pre-treatment SENCR levels were higher in the responder group (responder ¼ 2.10 0.44 arbitrary units; nonresponder ¼ 1.53 0.27 arbitrary units; p ¼ 0.014). We suggest for the first time that circulating lncRNAs are blood-based biomarkers for therapeutic guidance of cardiovascular conditions. We demonstrate that: 1) pre-treatment levels of circulating SENCR are associated with the rate of change in dia-
REFERENCES 1. van der Meer RW, Rijzewijk LJ, de Jong HW, et al. Pioglitazone improves cardiac function and alters myocardial substrate metabolism without affecting cardiac triglyceride accumulation and high-energy phosphate metabolism in patients with well-controlled type 2 diabetes mellitus. Circulation 2009;119:2069–77. 2. Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. Pioglitazone and risk of cardiovascular events in patients with type 2 diabetes mellitus: a metaanalysis of randomized trials. JAMA 2007;298:1180–8. 3. Boon RA, Jae N, Holdt L, Dimmeler S. Long noncoding RNAs: from clinical genetics to therapeutic targets? J Am Coll Cardiol 2016;67:1214–26. 4. Kumarswamy R, Bauters C, Volkmann I, et al. Circulating long noncoding RNA, LIPCAR, predicts survival in patients with heart failure. Circ Res 2014; 114:1569–75.
stolic function; 2) levels of circulating SENCR before therapy allow distinction of responders from nonresponders; and 3) SENCR is a better predictor of the response to therapy than are other traditional clinical
Trimethylamine N-Oxide in Seafood
variables. Thus, assessment of lncRNA levels could constitute a major breakthrough in the development
Rotten or Forgotten?
of personalized medicine to guide medical therapy, with important implications in terms of adverse effects and treatment costs.
We read the paper by Senthong et al. (1) with great interest. The research article was accompanied by
*David de Gonzalo-Calvo, PhD Franziska Kenneweg, MSc Claudia Bang, PhD Rocío Toro, MD, PhD Rutger W. van der Meer, MD, PhD Luuk J. Rijzewijk, MD Johannes W.A. Smit, MD, PhD Hildo J. Lamb, MD, PhD *Vicenta Llorente-Cortes, PhD Thomas Thum, MD, PhD
an editorial comment by Schmid-Schönbein (2).
*Cardiovascular Research Center (CSIC-ICCC)
directly involved in development of atherosclerosis
Biomedical Research Institute Sant Pau (IIB Sant Pau)
therefore seems like a paradox, as intake of seafood
Av. Sant Antoni Maria Claret 167
is generally accepted as cardioprotective.
Pavelló del Convent
During
recent
years,
trimethylamine
N-oxide
(TMAO) has emerged as a putative risk factor linking development of atherosclerosis to intake of socalled unhealthy foods (beef, egg) containing the TMAO precursors carnitine and choline. However, seafood is an important dietary source of free TMAO, in addition to the TMAO precursors carnitine and choline, a fact that appears to be commonly overlooked. To us, the hypothesis that TMAO is
In fish and marine invertebrates, TMAO exerts
08025 Barcelona
important osmoregulatory functions, and the mole-
Spain
cule can be found in high concentrations in these
E-mail: [email protected] OR [email protected]
organisms (3). TMAO from fish may be absorbed
http://dx.doi.org/10.1016/j.jacc.2016.10.014
directly from the gastrointestinal tract and give rise
Please note: The authors are part of the Cardiolinc network. The authors thank Ignasi Gich, MD, PhD (IIB Sant Pau, Barcelona, Spain) and Pablo Martínez Camblor, PhD (Geisel School of Medicine, Hanover, New Hampshire) for their excellent statistical support. This work was supported by the Merck Health Foundation (Merck Serono Research Award in Cardiometabolism 2014), Fundación Pública Andaluza Progreso y Salud (PI-0011/2014), Instituto de Salud Carlos III (FIS PI14/01729), ERC Consolidator grant LongHeart, DFG Th903/10-1, the IFB-Tx (BMBF), the 7th FP-funded EU project HOMAGE and the Foundation Leducq; and cofinanced by the European Fund for Regional Development (E.F.R.D.) and Fundació Marató TV3 (201521 10). The PIRAMID
to plasma levels that are orders of magnitude higher than after beef or egg intake (4). Indeed, TMAO has previously been suggested as a marker of seafood consumption (5). In contrast, TMAO derived from carnitine and choline is produced in the liver, by oxidation of trimethylamine (TMA) produced by gut bacteria. Hepatic oxidation of TMA to TMAO may