Invest Clin 65(3): 308 - 320, 2024 https://doi.org/10.54817/IC.v65n3a04

Corresponding author: Ji Zhang, Department of Cardiology, Wujin Hospital Affiliated with Jiangsu University, the

Wujin Clinical College of Xuzhou Medical University, No. 2, Changzhou City, Jiangsu Province, China. Telephone

number: +86 519 8557 9192. E-mail: js_zhangji@126.com

Elevated CA125 values predict adverse

outcomes in acute heart failure.

Ji Zhang

, Wenhua Li

, Jie Hui

and Jianqiang Xiao

Department of Cardiology, Wujin Hospital Affiliated with Jiangsu University, the Wujin

Clinical College of Xuzhou Medical University, Changzhou City, Jiangsu Province,

China.

Department of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou,

China.

Keywords: carbohydrate antigen 125; risk prediction; N-terminal pro-B-type natriuretic

peptide; acute heart failure.

Abstract. In acute heart failure (AHF), elevated carbohydrate antigen 125

(CA125) and N-terminal pro-B-type natriuretic peptide (NTproBNP) have been

shown to correlate with adverse events. We sought to quantify their prognostic

usefulness in predicting the six-month combined death/heart failure readmis-

sion endpoint. The study included 352 patients admitted for AHF. The primary

endpoint was the six-month combined endpoint of death/AHF rehospitaliza-

tion. CA125 and NTproBNP were dichotomized according to the best cut-offs

to predict the six-month primary endpoint. The independent association of

CA125 and NTproBNP with the primary endpoint was assessed by multivari-

ate Cox regression analysis, and their incremental prognostic utility was evalu-

ated by net reclassification improvement (NRI) and integrated discrimination

improvement (IDI) index. Forty-seven (13.4%) deaths and 113 (32.1%) AHF

rehospitalizations were identified at the six-month follow-up. The subjects with

CA125≥39.7 U/mL and NTproBNP≥3900 pg/mL had significantly higher cu-

mulative event rates (56.1% vs. 33.3% and 53.3% vs. 33.8%, both p<0.001).

Elevated CA125 (HR 1.93; 95% CI [1.32-2.83]; p=0.001) was associated with

a higher HR (hazard ratio) than NTproBNP≥3900 pg/mL (HR 1.71; 95% CI

[1.19-2.48]; p=0.004) after adjusting for established risk factors. Elevated

CA125 still independently predicted adverse events when CA125 and NTproB-

NP entered the same multivariate model. Furthermore, risk reclassification

analyses demonstrated significant improvements in NRI of 22.3% (p=0.014)

and IDI of 2.7% (p=0.012) when adding CA125 to the base model + NTproBNP.

Elevated CA125 and NTproBNP predicted adverse outcomes in AHF patients.

CA125 added prognostic value to NTproBNP; thus, their combination conferred

greater predictive capacity.

Predictive value of CA125 in acute heart failure 309

Vol. 65(3): 308 - 320, 2024

Valores elevados de CA125 predicen resultados adversos

en la insuficiencia cardíaca aguda.

Invest Clin 2024; 65 (3): 308 – 320

Palabras clave: antígeno carbohidrato 125; predicción del riesgo; péptido natriurético

tipo pro-B N-terminal; insuficiencia cardíaca aguda.

Resumen. En la insuficiencia cardíaca aguda, se ha demostrado que el an-

tígeno de carbohidratos 125 (CA125) elevado y el péptido natriurético tipo B

N-terminal (NTproBNP) se correlacionan con eventos adversos. Intentamos cuan-

tificar su utilidad pronóstica al predecir el punto final combinado de 6 meses de

readmisión por muerte/insuficiencia cardíaca. El estudio incluyó a 352 pacien-

tes ingresados por insuficiencia cardíaca aguda. El punto final principal fue el

punto final combinado de 6 meses de muerte/rehospitalización aguda. CA125 y

NTPROBNP se dicotomizaron de acuerdo con los mejores límites para predecir

el punto final primario de 6 meses. La asociación independiente de CA125 y

NTPROBNP con el punto final primario se evaluó mediante análisis multivariado

de regresión de Cox, y su utilidad pronóstica incremental se evaluó mediante la

mejora de la reclasificación neta (NRI) y el índice de mejora de la discriminación

integrada (IDI). En el seguimiento a los 6 meses se identificaron un total de

47 (13,4%) muertes y 113 (32,1%) rehospitalizaciones por insuficiencia cardíaca

aguda. Los sujetos con CA125≥39,7 U/mL y NTproBNP≥3900 pg/mL presenta-

ron tasas de acontecimientos acumulativos significativamente más altas (56,1%

frente a 33,3% y 53,3% frente a 33,8%, p<0,001 en ambos casos). CA125 elevado

(HR: 1,93; IC del 95% [1,32-2,83]; p = 0,001) se asoció con un HR superior al

NTproBNP ≥3900 pg/mL (HR 1,71; IC del 95% [1,19-2,48]; p = 0,004) después

del ajuste por los factores de riesgo establecidos. CA125 elevado aún predijo de

forma independiente los acontecimientos adversos cuando tanto CA125 como

NTproBNP se introdujeron juntos en el mismo modelo multivariante. Además,

los análisis de reclasificación del riesgo demostraron mejoras significativas en el

NRI del 22,3% (p = 0,014) y en el IDI del 2,7% (p = 0,012) al añadir CA125 al

modelo base + NTproBNP. Los niveles elevados de CA125 y NTproBNP predijeron

los resultados adversos en los pacientes con insuficiencia cardíaca aguda. CA125

añadió valor pronóstico al NTproBNP y, por lo tanto, su combinación confirió una

mayor capacidad predictiva.

Received: 30-10-2023 Accepted: 18-06-2024

INTRODUCTION

Given the variations in clinical pre-

sentation and the impact of comorbidities

in acute heart failure (AHF) patients, risk

prediction remains challenging. Identifying

high-risk subjects will help in further man-

agement by optimizing diuretic therapy, in-

creasing the frequency of monitoring visits,

and other therapeutic measures.

Published studies indicate that high

levels of several biomarkers

, including na-

triuretic peptides

2,3

, sST-2

4-6

, cardiac tro-

ponins

7,8

, and carbohydrate antigen 125

(CA125)

9,10

, correlate with AHF severity and

adverse outcomes. Based on different patho-

310 Zhang et al.

Investigación Clínica 65(3): 2024

physiological pathways involving heart fail-

ure progression and response patterns for

modification over time, we speculate that in-

tegrating multiple biomarkers will improve

prognostic power in subjects admitted for

AHF. As a widely used biomarker for moni-

toring ovarian cancer

, CA125 has been

studied in heart disease patients

10,12-14

and

especially in heart failure

10,12,15

, emerging as

a surrogate for fluid overload and/or cyto-

kine production in AHF

Our study was performed to evaluate

the prognostic utility of CA125 in predicting

the six-month combined endpoint of death/

AHF rehospitalization among AHF patients.

METHODS

Study population and design

This prospective, observational cohort

study from a single centre included 352 pa-

tients consecutively admitted to the cardiol-

ogy ward from December 2019 to September

2020 due to AHF following current guide-

lines

17,18

. AHF was the primary diagnosis of

hospitalization for our study. Patients with

a diagnosis of severe hepatic disease, sepsis,

ongoing dialysis treatment for end-stage re-

nal disease, pulmonary embolism, or acute

rheumatic and autoimmune diseases were

excluded by design. Demographic informa-

tion, vital signs, medications, and medical

history were collected, along with standard

echocardiographic evaluation, laboratory re-

sults and 12-lead electrocardiogram during

index admission. Intravenous furosemide or

torasemide was used in all patients at least

24 h after admission. 11.4% and 4.5% of the

patients received intravenous treatment

with vasodilators and vasopressors, respec-

tively. The established treatment guidelines

were followed

17-19

. Time to death/AHF read-

mission, whichever occurred first, was the

primary endpoint at the six-month follow-up.

Subjects were followed up through out-

patient service or by telephone. Three pa-

tients were lost to follow-up during the study

period. Patients were censored free of events

or lost to follow-up at last contact within this

period or at six months. The local ethics com-

mittee approved this study, and all patients

provided informed consent for their partici-

pation following the Declaration of Helsinki.

Biomarker measurement

CA125 serum levels were obtained be-

tween 5:30 and 8:00 h on the second day of

admission. In contrast, N-terminal pro-B-

type natriuretic peptide (NTproBNP) serum

levels were immediately determined after ad-

mission using commercially available immu-

noassay kits (Elecsys CA125 II assay, Roche

Diagnostics and Vitros Immunodiagnostic

Products NT-proBNP Reagent Pack, Ortho-

Clinical Diagnostics, respectively). A techni-

cian blinded to the clinical information per-

formed the biomarker assay.

Statistical analysis

Categorical variables are presented as

frequencies and percentages, and continuous

variables are summarized as the mean ± stan-

dard deviation or median (interquartile range).

We dichotomized both biomarkers according

to the best predictive cut-offs and compared

between-group baseline characteristics using

the t-test, Mann‒Whitney test, chi-square or

Fisher exact test, as appropriate. The result-

ing cut-off values were 39.7 U/mL for CA125

and 3900 pg/mL for NTproBNP. The cumula-

tive rate of events (death or AHF readmission)

among CA125 or NTproBNP categories was

estimated and compared using the Kaplan‒

Meier method and log-rank test. Univariate

and multivariate Cox analyses determined the

relationship of CA125 and NTproBNP with the

primary endpoint. Candidate variables in the

initial multivariate model included clinical

characteristics such as age, sex, weight, history

of atrial fibrillation, diabetes, hypertension and

acute myocardial infarction on admission. The

biochemical variables included were serum

creatinine, blood hemoglobin, and serum so-

dium. We also included left ventricular ejection

fraction (LVEF>50% [reference], 36%-50%,

and ≤35%), admission heart rate, admission

Predictive value of CA125 in acute heart failure 311

Vol. 65(3): 308 - 320, 2024

systolic blood pressure, evidence of pleural ef-

fusion, peripheral oedema, and AHF category

(worsening heart failure [WHF] or new-onset

heart failure) in our analyses. For multivariate

Cox regression analyses, we retained factors

with p<0.15 in univariate Cox analysis and

those clinically relevant. Given the number of

events available, the included variables were

carefully chosen, and a parsimonious multivari-

ate Cox model was derived. CA125, NTproBNP,

or both biomarkers were first entered individu-

ally in the multivariate model. The Schoenfeld

residuals were used to test the assumption of

proportional hazards over time.

Harrell’s C-statistics measured the dis-

criminative ability of the models. The incre-

mental prognostic utility of CA125 for NT-

proBNP and baseline variables was evaluated

using integrated discrimination improve-

ment (IDI) and net reclassification improve-

ment (NRI) with the corresponding P values.

We performed two multiple linear analyses

to examine the relations of log-transformed

CA125 and NTproBNP to clinical variables.

A 2-sided p value of <0.05 was consid-

ered statistically significant in all analyses.

The principal analysis was performed using

SPSS 26.0. Risk reclassification was calcu-

lated in R 4.0.3.

RESULTS

Baseline characteristics

Of 352 subjects, 49.4% had LVEF>50%.

Heart failure with preserved ejection frac-

tion predominated in our population, with

only 17.0% and 21.2% exhibiting LVEF≤35%

for patients with CA125<39.7 U/mL and pa-

tients with CA125≥39.7 U/ mL, respective-

ly. The sample consisted of 46.9% females,

and the mean age was 76±11 years. The me-

dian baseline levels of CA125 and NTproB-

NP in the entire population were 43.2 U/

mL (21.6-102.7) and 5170 pg/mL (2748-

10000), respectively. The baseline charac-

teristics of the study participants by CA125

categories are shown in Tables 1, 2 and 3. Pa-

tients with elevated CA125 (CA125≥39.7 U/

mL) exhibited a worse clinical profile. Lower

LVEF and pleural effusion were more preva-

lent when CA125 was elevated. Subjects with

CA125 ≥39.7 U/mL had a higher proportion

of treatment with digitalis at discharge and

a lower proportion with sodium-glucose co-

transporter 2 inhibitor. Lower LVEF, pleural

effusion, peripheral edema and paroxysmal

nocturnal dyspnea were more prevalent

when CA125 was elevated. No differences

were detected in the presence of orthopnea

and moist rales in lung fields.

Clinical predictors of CA125

and NTproBNP

Table 4 lists those variables indepen-

dently correlated with log-transformed

CA125 and NTproBNP. We identified different

clinical predictors of these two biomarkers

in the AHF setting. For lnCA125, the most

important predictors were pleural effusion

and WHF (standardized β coefficients 0.392

and 0.231, respectively). The most important

predictors of lnNTproBNP were serum creati-

nine, weight and LVEF (standardized β coeffi-

cients 0.382, -0.306 and -0.286, respectively).

Moreover, we found differential associa-

tions of CA125 and NTproBNP with clinical

presentations of AHF. A presentation as WHF

was associated with higher CA125 levels;

conversely, admission for new-onset heart

failure was independently and positively re-

lated to NTproBNP values.

CA125 levels, NTproBNP levels,

and the primary endpoint

In total, 47 patients (13.4%) died (12

deaths occurred during the index admission

and 35 post-discharge), and 113 (32.1%)

AHF rehospitalizations were identified at the

six-month follow-up. CA125 and NTproBNP

values in subjects experiencing death/AHF

rehospitalization were significantly higher

when compared with those free of events

(56.3 U/mL [27.2-135.6] vs. 33.9 U/mL

[18.4-79.8] and 6255 pg/mL [3425-6255]

vs. 4085 pg/mL [2390-8015], respectively,

p<0.001 for both).

312 Zhang et al.

Investigación Clínica 65(3): 2024

Table 1

Demographic and medical characteristics stratified by CA125 categories.

CA125<39.7U/mL

(n=159)

CA125≥39.7U/mL

(n=193)

Age, years 77±9 75±12 0.085

Female, n (%) 83 (52.2) 82 (42.5) 0.069

Weight, kg 60.4±12.3 59.4±12.1 0.425

Hypertension, n (%) 108 (67.9) 112 (58.0) 0.056

Diabetes mellitus, n (%) 44 (27.7) 55 (28.5) 0.864

Atrial fibrillation, n (%) 76 (47.8) 107 (55.4) 0.153

Previous coronary artery disease, n (%) 38 (23.9) 33 (17.1) 0.114

Previous myocardial infarction, n (%) 19 (11.9) 15 (7.8) 0.187

Acute myocardial infarction, n (%) 25 (15.7) 11 (5.7) 0.002

Previous PCI, n (%) 15 (9.4) 9 (4.7) 0.077

Valvular heart disease, n (%) 17 (10.7) 24 (12.4) 0.612

WHF, n (%) 131 (82.4) 182 (94.3) <0.001

Previous pacemaker, n (%) 4 (2.5) 5 (2.6) 1.000

Anemia, n (%) 43 (27.0) 51 (26.4) 0.896

Previous stroke, n (%) 17 (10.7) 22 (11.4) 0.833

COPD, n (%) 35 (22.0) 45 (23.3) 0.771

Previous malignancy, n (%) 2 (1.3) 12 (6.2) 0.018

Pleural effusion, n (%) 59 (37.1) 140 (72.5) <0.001

Peripheral oedema, n (%) 38 (23.9) 96 (49.7) <0.001

CA125, carbohydrate antigen 125; PCI, percutaneous coronary intervention; WHF, worsening heart

failure; COPD, chronic obstructive pulmonary disease. Anemia, defined as a hemoglobin level <120g/L

in men and <110g/L in women.

By the Kaplan–Meier method, subjects

with CA125≥39.7 U/mL and NTproBNP≥3900

pg/mL exhibited significantly higher cumula-

tive event rates (56.1% vs. 33.3% and 53.3% vs.

33.8%, both p<0.001, Fig. 1A, B). When com-

bined (Fig. 1C), patients with both biomark-

ers elevated had the highest cumulative event

rate (61.5%); intermediate when only one of

them was elevated: 44.2% for those with only

CA125 elevated and 40.5% for subjects with

only NTproBNP elevated, respectively, and low-

er (25.3%) for patients with values below the

chosen biomarker cutpoints, p trend <0.001.

Table 5 displays the results of univariate

and multivariate modelling. In the multivariate

Cox analysis, elevated CA125 (HR 1.93; 95% CI

[1.32-2.83]; p=0.001) was associated with a

higher adjusted HR than NTproBNP≥3900 pm/

mL (HR 1.71; 95% CI [1.19-2.48]; p=0.004).

Elevated CA125 still independently predicted

adverse events when CA125 and NTproBNP en-

tered in the same multivariate model. In the

final Cox model, serum creatinine and NTproB-

NP≥3900 pm/mL were other independent pre-

dictors. No interactions were found when these

two biomarkers were included in the final Cox

model (p=0.508).

We compared the performance of each

regression model by using Harrell’s C-sta-

tistic as a discrimination measure. Com-

pared with the model including NTproBNP

alone (0.623), CA125 alone (0.635) or none

(0.606), the Cox model including CA125 and

NTproBNP had a higher C-statistic (0.648).

Predictive value of CA125 in acute heart failure 313

Vol. 65(3): 308 - 320, 2024

Table 2

Vital signs, laboratory and echocardiography data stratified by CA125 categories.

CA125<39.7U/mL

(n=159)

CA125≥39.7U/mL

(n=193)

Heart rate, b.p.m. 87±22 93±24 0.033

Systolic blood pressure, mmHg 136±24 135±23 0.564

Diastolic blood pressure, mmHg 80±16 83±15 0.150

Haemoglobin, g/L 123.6±24.0 126.3±25.6 0.314

Serum creatinine, umol/L 86 (69–117) 83 (66–114) 0.340

Sodium, mmol/L 140.1±5.3 139.0±5.3 0.057

NTproBNP, pg/mL 4200 (2510–7940) 5990 (3245–11400) 0.002

CA125, U/mL 21 (13-27) 91 (56-173) <0.001

LVEF, % 50±13 47±13 0.020

LVEF ≤ 35%, n (%) 27 (17.0) 41 (21.2) 0.313

LVEF ≤ 50%, n (%) 43 (27.0) 67 (34.7) 0.122

LVDD, mm 53±9 53±10 0.802

LVSD, mm 39±10 40±11 0.317

LAD, mm 46±8 48±9 0.035

CA125, carbohydrate antigen 125; NTproBNP, N-terminal pro-B-type natriuretic peptide; LVEF, left ventricular ejection

fraction; LVDD, left ventricular diastolic diameter; LVSD, left ventricular systolic diameter; LAD, left atrial diameter.

Table 3

Medical treatment stratified by CA125 categories

CA125<39.7U/mL CA125≥39.7U/mL p

Intravenous administration of vasopresors , n (%) 8(5.0) 8(4.1) 0.691

Intravenous administration of vasodilators, n (%) 21(13.2) 19(9.8) 0.322

Medication before admission

Loop diuretic, n (%) 61(38.4) 64(33.2) 0.310

Spironolactone, n (%) 55(34.6) 60(31.1) 0.486

ACEI/ARB/ARNI, n (%) 48(30.2) 50(25.9) 0.372

Beta-blocker, n (%) 46(28.9) 48(24.9) 0.391

Digitalis, n (%) 17(10.7) 25(13.0) 0.515

SGLT2 inhibitor, n (%) 20(12.6) 17(8.8) 0.251

Medication at discharge (340 cases discharged after improvement)

Loop diuretic, n (%) 138(89.0) 172(93.0) 0.202

Spironolactone, n (%) 122(78.7) 159(85.9) 0.079

ACEI/ARB/ARNI, n (%) 105(67.7) 112(60.5) 0.169

Beta-blocker, n (%) 106(68.4) 130(70.3) 0.649

Digitalis, n (%) 30(19.4) 56(30.3) 0.021

SGLT2 inhibitor, n (%) 61(39.4) 47(25.4) 0.006

CA125, carbohydrate antigen 125; ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor

blockers; ARNI, angiotensin receptor neprilysin inhibitor; SGLT2, sodium-glucose co-transporter 2.

314 Zhang et al.

Investigación Clínica 65(3): 2024

IDI and NRI values were significantly

higher when adding each biomarker or both

to the baseline variables model. Further-

more, a significant improvement in NRI of

22.3% (p=0.014) and IDI of 2.7% (p=0.012)

was observed when adding CA125 to the

base model + NTproBNP, supporting the

incremental prognostic effect on top of NT-

proBNP (Table 6).

DISCUSSION

Our study compared the risk prediction

capacity of NTproBNP and CA125 in the set-

ting of AHF. After multivariate adjustment,

the elevation of CA125 and NTproBNP had a

positive prognostic effect on adverse events.

Not only elevated NTproBNP but also CA125

remained independent predictors of poor

outcomes by combining both biomarkers.

Table 4

Clinical predictors of lnCA125 and lnNTproBNP.

Standardized

β regression

coefficient

Ln (CA125)

Pleural effusion, n (%) 0.392 <0.001

WHF, n (%) 0.231 <0.001

Peripheral oedema, n (%) 0.173 <0.001

Weight, kg -0.154 0.002

Age, years -0.151 0.003

LVEF, % -0.132 0.006

Sodium, mmol/L -0.106 0.018

Ln (NTproBNP)

Serum creatinine, umol/L 0.382 <0.001

Weight, kg -0.306 <0.001

LVEF, % -0.286 <0.001

Pleural effusion, n (%) 0.154 <0.001

WHF, n (%) -0.141 0.001

LVDD, n (%) 0.139 0.019

Ln(CA125), antigen carbohydrate 125 natural loga-

rithm; Ln(NTproBNP), N-terminal pro–B-type natriu-

retic peptide natural logarithm; WHF, worsening heart

failure; LVEF, left ventricular ejection fraction; LVDD,

left ventricular diastolic diameter.

Additionally, adding CA125 to the model,

including NTproBNP, significantly improved

the predictive power.

Congestion, as a strong predictor of heart

failure-related readmission and death

, is re-

sponsible for most heart failure decompensa-

tion and is an essential therapeutic target in

AHF

17,18

; however, evaluation of congestion

remains a challenge in the routine manage-

ment of AHF

. Perhaps due to the limited

accuracy of signs and symptoms for quantify-

ing fluid overload severity

22,23

, signs of con-

gestion (peripheral edema, pleural effusion,

and other signs) are not routinely used for

risk stratification. Suitable biomarkers would

optimize risk prediction. CA125 levels cor-

relate well with signs of fluid congestion

9,10,16

and pulmonary artery wedge pressure

10,16

. In

this study, the most important clinical predic-

tor of serum CA125 levels was the presence of

pleural effusion. As a marker of congestion,

CA125 is related to adverse events in heart

failure patients

9,10

and indicates heart fail-

ure severity. However, in the BIOSTAT-CHF

study, CA125 levels were highly predictive of

adverse outcomes, beyond and independently

of surrogates of congestion

. We also con-

firmed the predictive value of CA125 in stage

D heart failure independently of pleural effu-

sion or ascites. Therefore, we think CA125

could provide added prognostic value over

surrogates of congestion

. Elevated CA125

is an independent predictor with incremental

prognostic value over traditional prognostica-

tors

and natriuretic peptides

, and thus, com-

bining both biomarkers improved risk stratifi-

cation in AHF

Interestingly, although CA125 has been

shown to be a potential tool for treatment

guidance in AHF

12,25

, little support is avail-

able regarding the benefits of NP-guided

therapy over usual care

. In the CHANCE-

HF trial, compared to the standard of care,

a CA125-guided therapy characterized by

a higher frequency of furosemide equiva-

lent dose adjustments and ambulatory in-

travenous furosemide administrations ac-

cording to CA125 response and clinical

Predictive value of CA125 in acute heart failure 315

Vol. 65(3): 308 - 320, 2024

Table 5

CA125 and NTproBNP hazard ratios for 6-month combined endpoint of death/AHF readmission.

Variables Univariate Multivariate

HR (95% CI) p-value HR (95% CI) p-value

Age, /10 years increase 1.10 (0.96-1.29) 0.167 1.06 (0.90-1.24) 0.457

Atrial fibrillation, n (%) 1.38 (1.01-1.89) 0.046 1.21 (0.85-1.71) 0.291

Serum creatinine, /SD increase 1.20 (1.00-1.01) 0.004 1.20 (1.04-1.38) 0.014

LVEF≤35%, n (%) 0.76 (0.48-1.21) 0.251 0.75 (0.45-1.27) 0.285

LVEF≤50%, n (%) 1.37 (0.98-1.92) 0.069 1.30 (0.92-1.86) 0.142

Systolic blood pressure, /10 mmHg increase 0.93(0.87-1.00) 0.042 0.96(0.90-1.03) 0.222

Sodium, /SD increase 0.90 (0.79-1.03) 0.112 0.92 (0.79-1.06) 0.243

WHF, n (%) 1.39 (0.80-2.40) 0.244 1.15 (0.63-2.08) 0.647

Pleural effusion, n (%) 1.32 (0.96-1.82) 0.086 0.99 (0.69-1.42) 0.955

Peripheral oedema, n (%) 1.40 (1.04-1.91) 0.034 1.14 (0.82-1.60) 0.430

CA125≥39.7U/mL, n (%) 2.00 (1.44-2.79) <0.001 1.78 (1.22-2.61) 0.003

NTproBNP≥3900pg/mL, n (%) 1.78 (1.26-2.50) 0.001 1.57 (1.08-2.27) 0.018

HR from Cox regression analysis. Multivariate HR from the model containing CA125 + NTproBNP + baseline varia-

bles. HR, hazard ratio; CI, confidence intervals; SD, standard deviation.

Fig. 1. Kaplan‒Meier estimates for the six-month combined endpoint of death/AHF rehospitalization stratified by

CA125 (A), NTproBNP (B) and the combination of CA125 and NTproBNP (C). AHF: acute heart failure.

316 Zhang et al.

Investigación Clínica 65(3): 2024

profile indicated a significantly reduced risk

of 1-year mortality or AHF readmission

In a recent multicentre randomized study

of 160 AHF subjects with renal dysfunction,

a CA125-guided diuretic strategy with an

admission loop diuretics dose determined

based on CA125 levels significantly im-

proved 72-h eGFR

. Briefly, in subjects with

high CA125 levels, high-intensity diuretic

treatment and/or closer follow-up were ad-

vocated. When CA125 was low or decreased,

a down-titration was recommended in both

trials, which endorsed the role of CA125-

guided decongestion treatment in AHF.

This study included a non-selected

hospitalized population of patients with

AHF. Based on this, we think many patients

hospitalized for AHF have preserved ejec-

tion fraction in the real world. Although

we pay more attention to heart failure with

reduced ejection fraction, Dunlay et al. re-

ported that two-thirds of advanced heart

failure subjects had LVEF>40%

. The pre-

dominance of preserved ejection fraction in

our population may explain a slightly lower

proportion of treatment with renin–angio-

tensin system inhibitors and beta-blockers.

Advanced heart failure occurs primarily in

the elderly, and the cardiorenal syndrome

is common. Sodium-glucose cotransporter

2 inhibitors may carry a higher risk of hy-

potension in older adults, in patients with

renal dysfunction and taking loop diuretics.

We know that patients with elevated CA125

had a worse clinical profile, which may be

one possible explanation for a lower propor-

tion of treatment with sodium-glucose co-

transporter 2 inhibitor and a higher propor-

tion of treatment with digitalis in patients

with elevated CA125 values.

In this study, we used a cut-off value

of 39.7 U/mL for CA125, but our previ-

ous paper confirmed that CA125≥65.7 U/

mL was highly predictive of adverse out-

comes in stage D heart failure patients

We noticed that some researchers divided

patients based on the normal CA125 lev-

els (<35 U/mL) derived primarily from

cancer studies

. The optimal cut-off for

defining normal vs. elevated values in dif-

ferent heart failure scenarios has not been

established. We think the value of CA125

we obtained could provide a particular

reference value in the setting of AHF and

stage D heart failure.

Given the long half-life of CA125 (ap-

proximately 5-12 days)

and the shorter

mean half-life of NTproBNP (60-120 min)

CA125 potentially provides pathophysiologi-

cal information several weeks prior, and NT-

proBNP could provide acute hemodynamic

information, similar to glycated hemoglobin

and serum glucose in diabetes. One study re-

ported that levels of CA125 and NTproBNP

represent distinct pathophysiological states

related to heart failure severity

. The com-

bined use of CA125 and NTproBNP improved

risk stratification, and this multimarker ap-

proach holds promise in guiding depletion

therapy, showing the need to incorporate

CA125 into daily clinical practice. In addi-

tion, conversely to natriuretic peptides, age,

sex, body weight, and renal function did not

significantly influence CA125 levels

12,21

. In

the current study, we found that NTproBNP

strongly depended on serum creatinine,

weight, and LVEF, while CA125 appeared

not to be significantly influenced by these

factors. Beyond these considerations, ad-

ditional benefits for implementing CA125

Table 6

Reclassification results for 6-month combined

endpoint of death/AHF rehospitalization.

NRI (%)

(p-value)

IDI (%)

(p-value)

Model 2 vs. 1 16.2(0.014) 2.6(0.010)

Model 3 vs. 1 23.8(0.008) 3.5(0.002)

Model 4 vs. 1 27.0(0.002) 5.3(<0.001)

Model 4 vs. 2 22.3(0.024) 2.7(0.020)

NRI, net reclassifification improvement; IDI, integrated

discrimination improvement. Model 1 = base model.

Model 2 = base model + NTproBNP categories. Model

3 = base model + CA125 categories. Model 4 = base

model + NTproBNP categories + CA125 categories.

Predictive value of CA125 in acute heart failure 317

Vol. 65(3): 308 - 320, 2024

testing in daily clinical practice arise from

its standardized measurement, low cost, and

wide availability.

Our study had some limitations. First,

its observational design makes it suscep-

tible to confounding factors and bias. Sec-

ond, it is a single-centre study that pre-

cludes the extrapolation of results. Third,

it is impossible to extrapolate findings to

patients undergoing renal dialysis because

this study included patients with baseline

serum creatinine values ≤360 µmol/L. Fi-

nally, we measured CA125 levels at the

one-time point after an overnight fast on

the second day of admission; however, peak

CA125 levels might better reflect fluid

overload in patients with AHF.

In conclusion, in AHF patients, elevat-

ed CA125 levels were highly predictive of

six-month death/AHF readmission, adding

prognostic value to NTproBNP and clinical

risk factors. Measuring simultaneously these

two biomarkers conferred greater predic-

tive capacity when compared with either of

them alone. Hence, this glycoprotein should

be considered a complement for optimal risk

prediction. The underlying mechanisms of

CA125 in AHF syndromes remain unclear,

and more research is needed.

Conflict of interest

The authors have no conflicts of inter-

est to declare.

Funding

Not applicable.

ORCID numbers authors

• Ji Zhang (JZ):

0000-0001-6099-9336

• Wenhua Li (WHL):

0000-0001-5496-0435

• Jie Hui (JH):

0000-0001-6517-5977

• Jianqiang Xiao (JZ):

0000-0002-7513-2230

Participation of the authors

JZ and WHL wrote and designed the

study. JZ and JH wrote and revised the man-

uscript. JQX collected data. All authors read

and approved the final manuscript.

REFERENCES

1. Levy WC, Mozaffarian D, Linker DT, Su-

tradhar SC, Anker SD, Cropp AB, Anand

I, Maggioni A, Burton P, Sullivan MD,

Pitt B, Poole-Wilson PA, Mann DL, Pac

ker M. The Seattle Heart Failure Model:

prediction of survival in heart failure. Cir

culation 2006; 113:1424-33. doi:10.1161/

CIRCULATIONAHA.105.584102.

2. Stienen S, Salah K, Eurlings LW, Betten

court P, Pimenta JM, Metra M, Bayes-

Genis A, Verdiani V, Bettari L, Lazzarini

V, Tijssen JP, Pinto YM, Kok WE. Cha

llenging the two concepts in determining

the appropriate pre-discharge N-terminal

pro-brain natriuretic peptide treatment

target in acute decompensated heart failu

re patients: absolute or relative discharge

levels? Eur J Heart Fail 2015; 17:936-44.

doi:10.1002/ejhf.320.

3. Kociol RD, Horton JR, Fonarow GC, Re

yes EM, Shaw LK, O’Connor CM, Felker

GM, Hernandez AF. Admission, dischar

ge, or change in B-type natriuretic pep-

tide and long-term outcomes: data from

Organized Program to Initiate Lifesaving

Treatment in Hospitalized Patients with

Heart Failure (OPTIMIZE-HF) linked to

Medicare claims. Circ Heart Fail 2011;

4:628-36. doi:10.1161/CIRCHEARTFAILU

RE.111.962290.

4. Aleksova A, Paldino A, Beltrami AP, Pa

doan L, Iacoviello M, Sinagra G, Emdin

M, Maisel AS. Cardiac biomarkers in the

emergency department: the role of solu

ble ST2 (sST2) in acute heart failure and

acute coronary syndrome-There is meat

on the bone. J Clin Med 2019; 8:270.

doi:10.3390/jcm8020270.

5. van Vark LC, Lesman-Leegte I, Baart SJ,

Postmus D, Pinto YM, Orsel JG, Wes

tenbrink BD, Brunner-la Rocca HP, van

318 Zhang et al.

Investigación Clínica 65(3): 2024

Miltenburg AJM, Boersma E, Hillege HL,

Akkerhuis KM; TRIUMPH Investigators.

Prognostic value of Serial ST2 measure

ments in patients with acute heart failu-

re. J Am Coll Cardiol 2017; 70:2378-2388.

doi:10.1016/j.jacc.2017.09.026.

6. Morrow DA, Velazquez EJ, DeVore AD,

Prescott MF, Duffy CI, Gurmu Y, McCa

gue K, Rocha R, Braunwald E. Cardiovas-

cular biomarkers in patients with acute de-

compensated heart failure randomized to

sacubitril-valsartan or enalapril in the PIO

NEER-HF trial. Eur Heart J 2019; 40:3345-

3352. doi:10.1093/eurheartj/ehz240.

7. Peacock WF 4th, De Marco T, Fonarow

GC, Diercks D, Wynne J, Apple FS, Wu

AH; ADHERE Investigators. Cardiac tro

ponin and outcome in acute heart failu-

re. N Engl J Med 2008; 358:2117-1126.

doi:10.1056/NEJMoa0706824.

8. Felker GM, Mentz RJ, Teerlink JR, Voors

AA, Pang PS, Ponikowski P, Greenberg

BH, Filippatos G, Davison BA, Cotter G,

Prescott MF, Hua TA, Lopez-Pintado S,

Severin T, Metra M. Serial high sensitivity

cardiac troponin T measurement in acute

heart failure: insights from the RELAX-

AHF study. Eur J Heart Fail 2015; 17:1262-

1670. doi:10.1002/ejhf.341.

9. Núñez J, Bayés-Genís A, Revuelta-López

E, Ter Maaten JM, Miñana G, Barallat J,

Cserkóová A, Bodi V, Fernández-Cisnal A,

Núñez E, Sanchis J, Lang C, Ng LL, Metra

M, Voors AA. Clinical role of CA125 in wor

sening heart failure: A BIOSTAT-CHF Study

Subanalysis. JACC Heart Fail 2020; 8:386-

397. doi:10.1016/j.jchf.2019.12.005.

10. Núñez J, Sanchis J, Bodí V, Fonarow GC,

Núñez E, Bertomeu-González V, Miñana

G, Consuegra L, Bosch MJ, Carratalá A,

Chorro FJ, Llàcer A. Improvement in risk

stratification with the combination of the

tumour marker antigen carbohydrate 125

and brain natriuretic peptide in patients

with acute heart failure. Eur Heart J 2010;

31:1752-1763. doi:10.1093/eurheartj/ehq

142.

11. Marcus CS, Maxwell GL, Darcy KM, Ha

milton CA, McGuire WP. Current ap-

proaches and challenges in managing

and monitoring treatment response in

ovarian cancer. J Cancer 2014; 5:25-30.

doi:10.7150/jca.7810.

12. Núñez J, Llàcer P, Bertomeu-González V,

Bosch MJ, Merlos P, García-Blas S, Mon

tagud V, Bodí V, Bertomeu-Martínez V, Pe-

drosa V, Mendizábal A, Cordero A, Gallego

J, Palau P, Miñana G, Santas E, Morell S,

Llàcer A, Chorro FJ, Sanchis J, Fácila L;

CHANCE-HF Investigators. Carbohydra

te Antigen-125-Guided Therapy in Acute

Heart Failure: CHANCE-HF: A Randomized

Study. JACC Heart Fail 2016; 4:833-843.

doi:10.1016/j.jchf.2016.06.007.

13. Rheude T, Pellegrini C, Schmid H,

Trenkwalder T, Mayr NP, Joner M, Kasel

AM, Holdenrieder S, Nunez J, Sanchis J,

Bodi V, Schunkert H, Kastrati A, Hengs

tenberg C, Husser O. Comparison of car-

bohydrate antigen 125 and N-Terminal

pro-brain natriuretic peptide for risk pre

diction after transcatheter aortic valve im-

plantation. Am J Cardiol 2018; 121:461-

468. doi:10.1016/j.amjcard.2017.11.020.

14. Falcão F, Oliveira F, Cantarelli F, Canta

relli R, Brito Júnior P, Lemos H, Silva

P, Camboim I, Freire MC, Carvalho O,

Sobral Filho DC. Carbohydrate antigen

125 for mortality risk prediction following

acute myocardial infarction. Sci Rep

2020; 10:11016. doi:10.1038/s41598-020-

67548-8.

15. Falcão F, de Oliveira FRA, da Silva MCFC,

Sobral Filho DC. Carbohydrate antigen

125: a promising tool for risk stratifica

tion in heart diseases. Biomark Med 2018;

12:367-381. doi:10.2217/bmm-2017-0452.

16. Núñez J, Miñana G, Núñez E, Chorro

FJ, Bodí V, Sanchis J. Clinical utility of

antigen carbohydrate 125 in heart fai

lure. Heart Fail Rev 2014; 19:575-584.

doi:10.1007/s10741-013-9402-y.

17. Ponikowski P, Voors AA, Anker SD, Bue

no H, Cleland JGF, Coats AJS, Falk V,

González-Juanatey JR, Harjola VP,

Jankowska EA, Jessup M, Linde C, Ni

hoyannopoulos P, Parissis JT, Pieske B,

Riley JP, Rosano GMC, Ruilope LM, Rus

chitzka F, Rutten FH, van der Meer P;

ESC Scientific Document Group. 2016

Predictive value of CA125 in acute heart failure 319

Vol. 65(3): 308 - 320, 2024

ESC Guidelines for the diagnosis and

treatment of acute and chronic heart

failure: The Task Force for the diagnosis

and treatment of acute and chronic heart

failure of the European Society of Car

diology (ESC)Developed with the special

contribution of the Heart Failure Associa

tion (HFA) of the ESC. Eur Heart J 2016;

37:2129-2200. doi:10.1093/eurheartj/eh

w128.

18. Yancy CW, Jessup M, Bozkurt B, Butler

J, Casey DE Jr, Colvin MM, Drazner MH,

Filippatos GS, Fonarow GC, Givertz MM,

Hollenberg SM, Lindenfeld J, Masoudi

FA, McBride PE, Peterson PN, Stevenson

LW, Westlake C. 2017 ACC/AHA/HFSA

Focused 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 Guideli

nes and the Heart Failure Society of Ame-

rica. Circulation 2017; 136:e137-e161.

doi:10.1161/CIR.0000000000000509.

19. Ibanez B, James S, Agewall S, Antunes

MJ, Bucciarelli-Ducci C, Bueno H, Cafo

rio ALP, Crea F, Goudevenos JA, Halvor-

sen S, Hindricks G, Kastrati A, Lenzen

MJ, Prescott E, Roffi M, Valgimigli M, Va

renhorst C, Vranckx P, Widimský P; ESC

Scientific Document Group. 2017 ESC

Guidelines for the management of acute

myocardial infarction in patients presen

ting with ST-segment elevation: The Task

Force for the management of acute myo

cardial infarction in patients presenting

with ST-segment elevation of the European

Society of Cardiology (ESC). Eur Heart J

2018; 39:119-177. doi:10.1093/eurheartj/

ehx393.

20. Rubio-Gracia J, Demissei BG, Ter Maa

ten JM, Cleland JG, O’Connor CM, Me-

tra M, Ponikowski P, Teerlink JR, Cotter

G, Davison BA, Givertz MM, Bloomfield

DM, Dittrich H, Damman K, Pérez-Calvo

JI, Voors AA. Prevalence, predictors and

clinical outcome of residual congestion in

acute decompensated heart failure. Int J

Cardiol 2018; 258:185-191. doi:10.1016/j.

ijcard.2018.01.067.

21. Mullens W, Damman K, Harjola VP, Me

bazaa A, Brunner-La Rocca HP, Martens

P, Testani JM, Tang WHW, Orso F, Ros

signol P, Metra M, Filippatos G, Sefe-

rovic PM, Ruschitzka F, Coats AJ. The

use of diuretics in heart failure with con

gestion - a position statement from the

Heart Failure Association of the Euro

pean Society of Cardiology. Eur J Heart

Fail. 2019;21(2):137-155. doi:10.1002/

ejhf.1369.

22. Parrinello G, Torres D, Paterna S, di Pas

quale P, Licata G. The pathophysiology of

acute heart failure: the key role of fluid

accumulation. Am Heart J 2008; 156:e19.

doi:10.1016/j.ahj.2008.04.031.

23. Gheorghiade M, Follath F, Ponikows

ki P, Barsuk JH, Blair JE, Cleland JG,

Dickstein K, Drazner MH, Fonarow GC,

Jaarsma T, Jondeau G, Sendon JL, Me

bazaa A, Metra M, Nieminen M, Pang

PS, Seferovic P, Stevenson LW, van Vel

dhuisen DJ, Zannad F, Anker SD, Rho-

des A, McMurray JJ, Filippatos G; Eu-

ropean Society of Cardiology; European

Society of Intensive Care Medicine. As

sessing and grading congestion in acute

heart failure: a scientific statement from

the acute heart failure committee of the

heart failure association of the European

Society of Cardiology and endorsed by the

European Society of Intensive Care Medi

cine. Eur J Heart Fail 2010; 12:423-433.

doi:10.1093/eurjhf/hfq045.

24. Zhang J, Li W, Xiao J, Hui J, Li Y. Prog

nostic significance of carbohydrate antigen

125 in stage D heart failure. BMC Cardio

vasc Disord 2023; 23:108. Published 2023

Feb 25. doi:10.1186/s12872-023-03139-5.

25. Núñez J, Llàcer P, García-Blas S, Bona

nad C, Ventura S, Núñez JM, Sánchez R,

Fácila L, de la Espriella R, Vaquer JM,

Cordero A, Roqué M, Chamorro C, Bodi

V, Valero E, Santas E, Moreno MDC, Miña

na G, Carratalá A, Rodríguez E, Mollar A,

Palau P, Bosch MJ, Bertomeu-González V,

Lupón J, Navarro J, Chorro FJ, Górriz

JL, Sanchis J, Voors AA, Bayés-Genís A.

CA125-guided diuretic treatment versus

usual care in patients with acute heart

320 Zhang et al.

Investigación Clínica 65(3): 2024

failure and renal dysfunction. Am J Med

2020; 133:370-380.e4. doi:10.1016/j.amj

med.2019.07.041.

26. Felker GM, Anstrom KJ, Adams KF, Eze

kowitz JA, Fiuzat M, Houston-Miller

N, Januzzi JL Jr, Mark DB, Piña IL,

Passmore G, Whellan DJ, Yang H, Coo

per LS, Leifer ES, Desvigne-Nickens P,

O’Connor CM. Effect of natriuretic pep

tide-guided therapy on hospitalization or

cardiovascular mortality in high-risk pa

tients with heart failure and reduced ejec-

tion fraction: A randomized clinical trial.

JAMA 2017; 318:713-720. doi:10.1001/

jama.2017.10565.

27. Dunlay SM, Roger VL, Killian JM, Weston

SA, Schulte PJ, Subramaniam AV, Blecker

SB, Redfield MM. Advanced heart failure

epidemiology and outcomes: a population-

based study. JACC Heart Fail 2021; 9:722-

732. doi: 10.1016/j.jchf.2021.05.009.

28. Lorenzo M, Palau P, Llàcer P, Domín

guez E, Ventura B, Núñez G, Miñana G,

Solsona J, Santas E, De La Espriella R,

Bodí V, Núñez E, Sanchis J, Bayés-Genís

A, Núñez J. Clinical utility of antigen

carbohydrate 125 for planning the optimal

length of stay in acute heart failure. Eur J

Intern Med 2021; 92:94-99. doi: 10.1016/j.

ejim.2021.05.037.

29. Clerico A, Carlo Zucchelli G, Pilo A, Pas

sino C, Emdin M. Clinical relevance of

biological variation: the lesson of brain na

triuretic peptide (BNP) and NT-proBNP as-

say. Clin Chem Lab Med 2006; 44:366-378.

doi:10.1515/CCLM.2006.063.