Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Dec 18:79:103013.
doi: 10.1016/j.eclinm.2024.103013. eCollection 2025 Jan.

Identifying biomarker-driven subphenotypes of cardiogenic shock: analysis of prospective cohorts and randomized controlled trials

Sabri Soussi  1   2 , Tuukka Tarvasmäki  3 , Antoine Kimmoun  4   5   6 , Mojtaba Ahmadiankalati  7 , Feriel Azibani  2 , Claudia C Dos Santos  8 , Kevin Duarte  9 , Etienne Gayat  2   10 , Jacob C Jentzer  11 , Veli-Pekka Harjola  12 , Benjamin Hibbert  11   13 , Christian Jung  14 , Lassus Johan  3 , Bruno Levy  4   5   6 , Zihang Lu  7 , Patrick R Lawler  15   16 , John C Marshall  8 , Janine Pöss  17 , Malha Sadoune  2 , Alexis Nguyen  2 , Alexandre Raynor  18 , Katell Peoc'h  18   19 , Holger Thiele  17 , Rebecca Mathew  13 , Alexandre Mebazaa  2   10
Affiliations

Identifying biomarker-driven subphenotypes of cardiogenic shock: analysis of prospective cohorts and randomized controlled trials

Sabri Soussi et al. EClinicalMedicine. .

Abstract

Background: Cardiogenic shock (CS) is a heterogeneous clinical syndrome, making it challenging to predict patient trajectory and response to treatment. This study aims to identify biological/molecular CS subphenotypes, evaluate their association with outcome, and explore their impact on heterogeneity of treatment effect (ShockCO-OP, NCT06376318).

Methods: We used unsupervised clustering to integrate plasma biomarker data from two prospective cohorts of CS patients: CardShock (N = 205 [2010-2012, NCT01374867]) and the French and European Outcome reGistry in Intensive Care Units (FROG-ICU) (N = 228 [2011-2013, NCT01367093]) to determine the optimal number of classes. Thereafter, a simplified classifier (Euclidean distances) was used to assign the identified CS subphenotypes in three completed randomized controlled trials (RCTs) (OptimaCC, N = 57 [2011-2016, NCT01367743]; DOREMI, N = 192 [2017-2020, NCT03207165]; and CULPRIT-SHOCK, N = 434 [2013-2017, NCT01927549]) and explore heterogeneity of treatment effect with respect to 28-day mortality (primary outcome).

Findings: Four biomarker-driven CS subphenotypes ('adaptive', 'non-inflammatory', 'cardiopathic', and 'inflammatory') were identified separately in the two cohorts. Patients in the inflammatory and cardiopathic subphenotypes had the highest 28-day mortality (p (log-rank test) = 0.0099 and 0.0055 in the CardShock and FROG-ICU cohorts, respectively). Subphenotype membership significantly improved risk stratification when added to traditional risk factors including the Society for Cardiovascular Angiography and Interventions (SCAI) shock stages (increase in Harrell's C-index by 4% (p = 0.033) and 6% (p = 0.0068) respectively in the CardShock and the FROG-ICU cohorts). The simplified classifier identified CS subphenotypes with similar biological/molecular and outcome characteristics in the three independent RCTs. No significant interaction was observed between treatment effect and subphenotypes.

Interpretation: Subphenotypes with the highest concentration of biomarkers of endothelial dysfunction and inflammation (inflammatory) or myocardial injury/fibrosis (cardiopathic) were associated with mortality independently from the SCAI shock stages.

Funding: Dr Sabri Soussi was awarded the Canadian Institutes of Health Research (CIHR) Doctoral Foreign Study Award (DFSA) and the Merit Awards Program (Department of Anesthesiology and Pain Medicine, University of Toronto, Canada) for the current study.

Keywords: Cardiogenic shock; Critical care; Heterogeneity of treatment effect; Precision medicine; Unsupervised learning.

PubMed Disclaimer

Conflict of interest statement

None of the authors of this paper has a financial or personal relationship with other persons or organizations that could inappropriately influence or bias the content of the paper. Dr. C.J. received honoraria for lectures/presentations from Bristol Myers Squibb (New York, US), Edwards Lifesciences (Irvine, California, US), Daichi Sankyo (Tokyo, Japan) and Boehringer-Ingelheim (Ingelheim, Germany). He received grants or had contracts with the German research foundation, the German ministry for economy and energy, the State of Nordrhine-Westfalia, the German space agency, the European Union and Edwards Lifescience (Irvine, California, US). He had leadership or fiduciary role in the German cardiac society, the German society for internal medicine, the German society for medical intensive care and the European society for intensive care medicine. Dr. H.T. had leadership or fiduciary role in the German cardiac society. Dr. J.P. received travel support to attend meetings from Overcome GmbH (Paris, France). She received grants or had contracts with the German cardiac society, the German society for heart research, Dr. Rolf Schwiete Foundation and Maquet Cardiopulmonary (Baden-Württemberg, Germany). Dr. E.G. received fees as a member of the advisory boards and/or steering committees and/or from research grants from Magnisense (Paris, France), Adrenomed (Hennigsdorf, Germany) and Deltex Medical (Chichester, United Kingdom). Dr. A.M. has received speaker's honoraria from Merck (Darmstadt, Germany), Medtronik (Dublin, Ireland), Viatris (Canonsburg, US), Novartis (Basel, Switzerland), Roche (Basel, Switzerland), Bayer (Leverkusen, Germany) and Astra Zeneca (Cambridge, UK); and fees as a member of the advisory boards and/or steering committees and/or research grants from Adrenomed (Hennigsdorf, Germany), Corteria (Boca Raton, Florida, US), Roche (Basel, Switzerland), Fire1 (Dublin, Ireland), Sphyngotec (Hennigsdorf, Germany), Abbott (Chicago, Illinois), Bayer (Leverkusen, Germany), Implicity (Paris, France), Windtree (Warrington, Pennsylvania, US), S-Form Pharma (Brussels, Belgium), Implicity (Paris, France), 4TEEN4 (Hennigsdorf, Germany) and the French government. He received equipment/material from Sphyngotec (Hennigsdorf, Germany). He is co-inventor of a patent (owned by S-Form pharma) on combination therapy for patients having acute or persistent dyspnea. The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Graphical schematic of study design.
Fig. 2
Fig. 2
(A) Comparison of class-defining variables using latent class analysis at inclusion in the CardShock cohort and the FROG-ICU cohort. Description: continuous variables were plotted after natural log transformation. Every normalized variable was standardized such that all means are scaled to 0 and SDs to 1. Group means of standardized values are shown by subphenotypes. A value of +1 for the standardized variable (y-axis) indicates that the mean value for a given subphenotype was one SD higher than the mean value in a whole cohort. Subphenotype classes sizes at inclusion (n): CardShock cohort: adaptive subphenotype N = 49, non-inflammatory subphenotype N = 56, cardiopathic subphenotype N = 73, inflammatory subphenotype N = 27; FROG-ICU cohort: adaptive subphenotype N = 72, non-inflammatory subphenotype N = 35, cardiopathic subphenotype N = 75, inflammatory subphenotype N = 46. (B) Biomarker heatmap across the different subphenotypes in the analysed studies. The heatmap shows the standardized median values of host response and organ dysfunction circulating markers (i.e., inflammation, endothelial dysfunction, myocardial injury and fibrosis, cardiac congestion and renal injury). Green colour represents lower standardized median values in a given subphenotype compared with the median of the whole study while red colour represents a higher value compared with the standardized median of the whole study. White cells represent biomarkers that were not measured/available. Abbreviations: SD, standard deviations; LCA, latent class analysis; CRP, C-reactive protein; PCT, Procalcitonin; IL6, Interleukin-6; NT-proBNP, N-terminal pro B-type natriuretic peptide; ALT, Alanine transaminase; Hs-cTnT, high-sensitive cardiac troponin T; Hs-cTnI, high-sensitive cardiac troponin I; ST2, soluble form of suppression of tumorigenicity 2; Bio-ADM, Bio-adrenomedullin; DPP3, Dipeptidyl peptidase 3; NGAL, neutrophil gelatinase-associated lipocalin; SP, subphenotype.
Fig. 3
Fig. 3
(A) 28-day survival curves according to subphenotype membership in the CardShock cohort and the FROG-ICU cohort. (B) One-year survival curves according to subphenotype membership in the CardShock and FROG-ICU cohorts. The log-rank tests between the survival curves of the different subphenotypes showed a p < 0.05 in the two cohorts.
Fig. 4
Fig. 4
Comparison of the available class-defining variables across the simplified classifier driven subphenotypes at inclusion in (A) the OptimaCC trial (N = 57), (B) the DOREMI trial (N = 192), and (C) the CULPRIT-SHOCK trial (N = 434). Description: continuous variables were plotted after natural log transformation. Every normalized variable was standardized such that all means are scaled to 0 and SDs to 1. Group means of standardized values are shown by subphenotypes. A value of +1 for the standardized variable (y-axis) indicates that the mean value for a given subphenotype was one SD higher than the mean value in a whole cohort. Sizes of the subphenotype classes determined by the parsimonious classifier (n): OptimaCC trial (A): adaptive subphenotype N = 11, non-inflammatory subphenotype N = 21, cardiopathic subphenotype N = 17, inflammatory subphenotype N = 8; DOREMI trial (B): adaptive subphenotype N = 24, non-inflammatory subphenotype N = 46, cardiopathic subphenotype N = 94, inflammatory subphenotype N = 28; CULPRIT-SHOCK (C): adaptive subphenotype N = 106, non-inflammatory subphenotype N = 141, cardiopathic subphenotype N = 101, inflammatory subphenotype N = 86. The median (IQR) frequencies of assignment of an individual to a given subphenotype across the 20 imputed datasets in the OptimaCC, DOREMI and CULPRIT-SHOCK trials were 0.80 (0.80–0.90), 0.75 (0.75–0.85) and 0.95 (0.90–0.95) respectively. Abbreviations: SD, standard deviations; NT-proBNP, N-terminal pro B-type natriuretic peptide; Bio-ADM, Bio-adrenomedullin; CRP, C-reactive protein; ALT, Alanine transaminase; Hs-cTnT, high-sensitive cardiac troponin T; Hs-cTnI, high-sensitive cardiac troponin I.
Fig. 5
Fig. 5
Unadjusted odds ratios of the association of the four subphenotypes with death within 28 days after ICU admission across the five studies individually (A) and after merging them (B).
See this image and copyright information in PMC

References

    1. Arrigo M., Price S., Baran D.A., et al. Optimising clinical trials in acute myocardial infarction complicated by cardiogenic shock: a statement from the 2020 critical care clinical trialists workshop. Lancet Respir Med. 2021;9(10):1192–1202. - PubMed
    1. Sarma D., Jentzer J.C., Soussi S. Cardiogenic shock: a major challenge for the clinical trialist. Curr Opin Crit Care. 2023;29(4):371–380. - PubMed
    1. Jentzer J.C., Ahmed A.M., Vallabhajosyula S., et al. Shock in the cardiac intensive care unit: changes in epidemiology and prognosis over time. Am Heart J. 2021;232:94–104. - PubMed
    1. Lawler P.R., Mehra M.R. Advancing from a “hemodynamic model” to a “mechanistic disease-modifying model” of cardiogenic shock. J Heart Lung Transplant. 2018;37(11):1285–1288. - PubMed
    1. Iwashyna T.J., Burke J.F., Sussman J.B., Prescott H.C., Hayward R.A., Angus D.C. Implications of heterogeneity of treatment effect for reporting and analysis of randomized trials in critical care. Am J Respir Crit Care Med. 2015;192(9):1045–1051. - PMC - PubMed

Associated data

LinkOut - more resources