Circulating microRNAs and Cytokines as Prognostic Biomarkers for Doxorubicin-Induced Cardiac Injury and for Evaluating the Effectiveness of an Exercise Intervention

Abstract

Purpose: To define a set of biomarkers that can be used to identify patients at high risk of developing late doxorubicin (DOX)-induced cardiac morbidity with the goal of focused monitoring and early interventions.

Experimental design: Mice received phosphate buffered saline or DOX 2.5 mg/kg 2x/week for 2 weeks. Blood samples were obtained before and after therapy for quantification of miRNAs (6 and 24 hours), cytokines (24 hours), and troponin (24 hours, 4 and 6 weeks). Cardiac function was evaluated using echocardiography before and 24 hours after therapy. To assess the effectiveness of exercise intervention in preventing DOX-induced cardiotoxicity blood samples were collected from mice treated with DOX or DOX + exercise. Plasma samples from 13 DOX-treated patients with sarcoma were also evaluated before and 24 hours after therapy.

Results: Elevations in plasma miRNA-1, miRNA-499 and IL1α, IL1β, and IL6 were seen in DOX-treated mice with decreased ejection fraction and fractional shortening 24 hours after DOX therapy. Troponin levels were not elevated until 4 weeks after therapy. In mice treated with exercise during DOX, there was no elevation in these biomarkers and no change in cardiac function. Elevations in these biomarkers were seen in 12 of 13 patients with sarcoma treated with DOX.

Conclusions: These findings define a potential set of biomarkers to identify and predict patients at risk for developing acute and late cardiovascular diseases with the goal of focused monitoring and early intervention. Further studies are needed to confirm the predictive value of these biomarkers in late cardiotoxicity.

Figure 1.

Effect of DOX therapy on cardiac function and plasma hs-cTnI. Mice received PBS or DOX (2.5 mg/Kg twice a week for 2 weeks). (A) Representative M-mode echocardiograms (parasternal long-axis view) after therapy. Quantification of (B) EF and (C) FS. (D) Quantification of plasma hs-cTnI in control and DOX-treated mice. Greater LV contractility is seen in the control mice than the DOX-treated mice. Data are expressed as the mean ± standard errors. *p < 0.05.

Figure 2.

Effect of DOX therapy on plasma miRNA-1, miRNA-29b and miRNA-499. Mice were treated as in Fig. 1: (A) Blood was collected 6 and 24 h after DOX therapy through retro-orbital puncture and analyzed for levels of miRNA. Values were normalized to a synthetic control (cel-miRNA-39-3p) that was spiked in during RNA extraction. Data are expressed as the mean ± standard error. *p < 0.05. (B) Mouse hearts were collected 24 h after DOX therapy and analyzed for levels of miRNA. The obtained values were normalized to the housekeeping gene U6 The data are expressed as the mean ± standard errors. *p < 0.05.

Figure 3.

ELISA analysis of cytokines in the plasma of control and DOX-treated mice. Mice were treated as in Fig. 1: Blood was collected 24 h after the final dose of DOX (dose 4) through retro-orbital puncture and analyzed for cytokine levels. Data are expressed as the mean ± standard errors. *p < 0.05.

Figure 4.

Effect of DOX and DOX + exercise on cardiac left ventricular (LV) function. Mice were treated with DOX or DOX + exercise (tread mill walking for 45 min, 5 days/week for 2 weeks). (A) Echocardiographic measurements from control, DOX and DOX + exercise-treated mice were taken 24 h after therapy. Representative M-mode echocardiograms (parasternal long-axis view) are displayed. (B) EF and (C) FS percentages in control, DOX and DOX + exercise-treated mice were taken 24 h after therapy. Data are expressed as the mean ± standard errors. *p < 0.05.

Figure 5.

Exercise initiated during DOX treatment prevented DOX-induced increase in plasma miRNAs and cytokines. Mice were treated as in Fig 4. Blood was collected before and 24 h after therapy. (A) miRNA-1, (B) miRNA-499, (C) IL-1α, (D) IL-1β, and (E) IL-6 plasma levels were then quantified. Data are expressed as the mean ± standard errors. *p < 0.05.

Figure 6.

Changes in individual patient miRNA and cytokine levels shown in Table 1. Plasma samples were collected from newly diagnosed AYA sarcoma patients before and 24 h after chemotherapy that included DOX. The obtained baseline values were compared to those obtained after DOX treatment for each patient. Values were normalized to a synthetic control (cel-miRNA-39-3p) which was spiked in during the RNA extraction.