Oxidative stress after anthracycline therapy in patients with solid tumors

Overview

Introduction:
Anthracyclines are regarded as some of the most potent oxidative stress inductors. Despite the fact that oxidative stress induction by anthracyclines is believed to be the key factor in anthracycline-related cardiotoxicity, the precise timeline of oxidative stress changes after anthracycline treatment remains unknown. The aim of the present study is to assess the level of oxidative stress after anthracycline therapy in patients with solid tumors.

Patients:
The study population consists of 128 adult patients (14 males, 114 females, mean age 56 ± 10 years) receiving anthracycline chemotherapy for solid tumors.

The control group consists of 38 patients (4 males, 34 females, mean age 59 ± 11 years) receiving anthracycline-free chemotherapy for solid tumors.

Methods:
The main activities of antioxidant enzymes (catalase, glutathione peroxidase-1, superoxide dismutase, and paraoxonase-1) and concentrations of conjugated dienes, surrogate markers of oxidative stress level, were established at the baseline and after anthracycline therapy (median 45; IQR 27-69 days after the end of anthracycline therapy) in all patients. By comparing the activities of antioxidant enzymes and the concentrations of conjugated dienes, before and after therapy, changes in oxidative stress level within the time period were established for both study groups. Differences between the study groups, with regard to changes in the activities of antioxidant enzymes and the concentrations of conjugated dienes, were also evaluated.

Conclusions:
An increase in oxidative stress was observed after the end of anthracycline therapy in patients with solid tumors. However, our study shows that this persistent elevation of oxidative stress after the end of anthracycline therapy is probably not caused by anthracyclines.

Keywords:
oxidative stress, anthracyclines, solid tumors

Introduction:

Anthracyclines (ANTRA) are cornerstones of many anti-cancer treatment protocols. One of the most serious complications of this chemotherapy is its cardiotoxicity (1-3). The chronic form of ANTRA-induced cardiotoxicity is the most challenging clinical issue. Although the pathogenesis of ANTRA-induced cardiotoxicity is not yet completely understood, oxidative stress (OS) is believed to represent the most important factor in its development (4).

ANTRA have been proven to be potent OS inductors, and this process has been described in detail on the sub-cellular and molecular levels (5). A number of experimental and human studies have shown that ANTRA induces OS. All of these studies refer to ANTRA-induced OS either during or soon after ANTRA therapy (6-8). But ANTRA-induced chronic cardiotoxicity develops later after the end of ANTRA therapy, and there are no reliable echocardiographic markers available after the treatment to predict, with acceptable sensitivity and specificity, the future development of left ventricle dysfunction and heart failure.

The precise pathogenesis of chronic ANTRA-induced cardiotoxicity remains to be elucidated. Is it a negative impulse during ANTRA therapy alone that induces morphological and/or functional changes within myocardium which, despite being undetectable by currently available diagnostic methods, later slowly progress into left ventricle dysfunction? Or does ANTRA therapy induce a long-lasting elevation of OS, that continuously causes myocardial damage?

The precise timeline of changes in OS level during and after ANTRA therapy seems to be a very important issue in understanding the development of ANTRA- induced cardiotoxicity. Since the induction of OS during and soon after ANTRA therapy has already been well described (6-8), the aim of our study was to discover whether an elevation of OS persists longer following the end of ANTRA therapy.

Study population:

The study population consists of 128 adult, ANTRA-naive patients (14 males, 114 females, mean age 56 ± 10 years) suffering from solid tumors that indicated anti-cancer treatment by protocols consisting of ANTRA.

The control group consists of 38 adult, ANTRA-naive patients (4 males, 34 females, mean age 59 ± 11 years) suffering from solid tumors that indicated anti-cancer treatment by protocols free of ANTRA.

The baseline demographic data was the same for both groups (Table 1). All patients enrolled in the study signed an informed consent.

Methods:

Study protocol:

This study was approved by the local ethics committee (Ethics Committee of the General Teaching Hospital in Prague, Czech Republic) and complies with the Declaration of Helsinki.

Blood samples for laboratory tests (activities of main antioxidant enzymes and concentration of conjugated dienes - see section below) were collected from the study population twice during the study. The first examination, for all enrolled patients, took place before the start of the study. Follow-up examinations took place after the end of ANTRA therapy in the study group (45; IQR 27-69 days; which corresponded to 127 ± 35 days after the patients’ enrollment). In the control group, follow-up examinations were performed at a time corresponding to the control group (examinations in the control group took place at 143 ± 47 days after the patients’ enrollment - not a significant difference in comparison with the study group [p=0.07]).

Laboratory methods:

Blood samples were obtained after overnight fasting. The activities of catalase, glutathione reductase-1, and Cu-Zn superoxide dismutase were measured in hemolyzed erythrocytes. The blood samples were collected into tubes with K₂ ethylenediaminetetraacetic acid; erythrocytes were washed three times with an NaCl isotonic solution (9g/l). Serum was used for the determination of all other parameters. The samples were stored at -80°C until the assay. The  hemotological parameters were carried out by routine laboratory techniques using an autoanalyzer (Coulter LH750 - haematological analyser, Beckman Coulter).

Measurement of enzyme activities

Glutathione peroxidase: The activity was measured by the modified method of Paglia and Valentine using tert-butyl hydroperoxide as a substrate (9). Briefly, 580 ml of 172.4 mM tris-HCl buffer containing 0.86 mM of ethylenediaminetetraacetic acid, pH = 8.0; 100 ml of 20 mM of glutathione, 100 ml of 10 U/ml glutathione reductase, 100 ml of 2 mM of reduced nicotinamidadenindinucleotide phosphate and 100 ml of diluted sample were pipetted into the cuvettes. The reaction was started after 10 minutes of incubation at 37 ºC by the addition of 20 ml of 9.99 mM tert-butyl hydroperoxide. The rate of reduced nicotinamidadenindinucleotide phosphate degradation was monitored spectrophotometrically at 340 nm. A blank was run for each sample. The activity of glutathione peroxidase was calculated using the molar extinction coefficient of reduced nicotinamidadenindinucleotide phosphate 6220 M^-1cm ^-1 and expressed as U/g  hemoglobin (U = μmol/min).

Catalase: The activity was determined by the modified method of Aebi (10). The reaction mixture in cuvettes contained 876 ml of 50 mM potassium phosphate buffer, pH = 7.2 and 25 ml of diluted sample. The reaction was started after 10 minutes of incubation at 30 ºC by the addition of 99 ml of 10 mM hydroxide peroxide. The rate of hydroxide peroxide degradation was monitored spectrophotometrically at 240 nm. A blank was run for each sample. Catalase activity was calculated using the molar extinction coefficient of hydroxide peroxide 43.6 M^-1cm ^-1 and expressed as kU/g  hemoglobin (U = μmol/min).

Cu-Zn superoxide dismutase: The activity was determined according to the modified method of Štípek et al. (11). The reaction mixture in cuvettes contained 700 ml of 50 mM potassium phosphate buffer, pH = 7.2; 50 ml of xanthine oxidase; 100ml of nitroblue tetrazolium - formazan and 50 ml of diluted sample. The reaction was started after 10 minutes of incubation at 25 ºC by the addition of 100 ml of 1 mM xanthine. The rate of NBT-formazan generation was monitored spectrophotometrically at 540 nm. A blank was run for each sample. Superoxide dismutase activity was calculated by means of a calibration curve and expressed as U/g  hemoglobin (U = μmol/min).

Paraoxonase 1: The arylesterase activity of paraoxonase-1 was measured according to the method of Eckerson et al. using phenylacetate as a substrate (12). Briefly, 900 ml of 20 mM tris-HCl buffer containing 1 mM CaCl₂, pH=8.0 was added to cuvettes followed by 50 ml of diluted serum sample. The reaction was started by the addition of 50 ml of 100 mM phenylacetate. The rate of phenol generation was monitored spectrophotometrically at 270 nm. A blank was run for each sample. The arylesterase activity of paraoxonase-1 was calculated using the molar extinction coefficient of the produced phenol, 1310 M^-1cm^-1 and expressed as U/ml serum (U = μmol/min).

Measurement of concentration of conjugated dienes in serum low density lipoproteins:

Serum low density lipoproteins (LDL) were isolated by the precipitation method of Ahotupa et al. (13). Concentrations of conjugated dienes in the precipitated LDL were measured by the modified method of Wieland et al. (14). The precipitation buffer consisted of 0.064 M trisodium citrate adjusted to pH 5.05 with 5 M HCl, and contained 50.000 IU/l heparin. Sample - 110 ml of serum with ethylenediaminetetraacetic acid (10:1 v/v) was added to 1 ml of the heparin-citrate buffer. After mixing, the suspension was incubated for 10 min at room temperature. The precipitated lipoproteins were then separated by centrifugation at 2800 rpm for 10 min. The supernatant was removed and the pellet was resuspended in 100 ml of NaCl isotonic solution (9g/l). Lipids were extracted by chloroform – methanol (2:1) and the mixture was incubated for 10 min with intermittent mixing; 250 ml redistilled water was used for phase separation. The mixture was centrifuged at 3000 rpm for 5 min. The 800 ml of lower layer (infranatant) was dried under nitrogen, redissolved in 300 ml of cyclohexane, and analyzed spectrophotometrically at 234 nm. The concentration of conjugated dienes was calculated using the molar extinction coefficient 2.95 x 10⁴ M^-1cm^-1 and expressed as mmol/l serum.

Statistical methods:

Normally distributed continuous variables (data) are expressed as mean value ± standard deviation (SD). Continuous variables (data) that do not follow a normal distribution are expressed as median and interquartile range (IQR; 25.-75. percentile).

The Student’s t-test or Mann-Whitney for continuous variables and the chi-square test (with Yates correction when applicable) for categorical variables were performed in order to assess the differences between the study and control groups.

Pearson’s correlation coefficient was used to test the correlation between two sets of continuous data.

All tests performed were two-tailed, and a p-value ≤ 0.05 was regarded as statistically significant. 

Microsoft EXCEL software was used to perform all statistical analysis

Results:

Activities of antioxidant systems before ANTRA therapy

No statistically significant differences in the baseline activities of catalase, glutathione peroxidase, superoxide dismutase and paraoxonase-1 were found between the study and control groups (Table 2).

Concentrations of conjugated dienes in LDL before ANTRA therapy

No statistically significant difference was found in baseline concentrations of conjugated dienes in LDL between the study and control groups (Table 3).

Changes in the activities of antioxidant enzymes after ANTRA therapy

At the follow-up, statistically significant changes in the activities of catalase, glutathione peroxidase and paraoxonase-1 were observed in the study group, in comparison to baseline values (Table 3).

Statistically significant changes in the activities of catalase, glutathione peroxidase and superoxide dismutase were found in the control group at the follow-up, in comparison to the baseline (Table 4).

However, we failed to detect any significant differences, with regard to changes in the activities of the above-mentioned antioxidant enzymes, between the study and control groups after the therapy, in comparison to the baseline examination (Δcatalase_{study vs. control groups}: p=0.18; Δglutathione peroxidase_{study vs.control groups}: p=0.68; Δsuperoxide dismutase_{study vs. control groups}: p=0.07; Δparaoxonase-1_{study vs. control groups}: p=0.74)

Changes in concentrations of conjugated dienes after ANTRA therapy

A statistically significant change in the concentrations of conjugated dienes was observed in the study group at the follow-up, in comparison to the baseline (Table 3); this was not the case for the control group (Table 4). 

However, we failed to detect any significant differences, with regard to changes in the concentrations of conjugated dienes, between the study and control groups after the therapy, in comparison to the baseline examination (Δconjugated dienes_{study vs. control groups}: p=0.91)

Discussion:

The main finding of our study is that there is a long-lasting (45; IQR 27-69 days after the end of ANTRA therapy in the case of our study) increase in the activities of major antioxidant enzymes (catalase, glutathione peroxidase) and in the concentrations of products of OS (conjugated dienes in precipitated LDL), that are probably not a consequence of ANTRA therapy. The results of our study prove the presence of long-lasting OS, in patients with solid tumors treated by ANTRA, which is probably not caused by the ANTRA therapy.

ANTRA have been proven to be potent OS inductors (4), and this process has been described in detail on the sub-cellular and molecular levels (5). Unfortunately, there is limited knowledge on the level and the dynamics of OS induced by ANTRA in humans. Some studies have shown that there is an induction of OS during ANTRA therapy in humans (6-8). However, these studies did not also evaluate changes in the activities of antioxidant enzymes in a control group consisting of patients with solid tumors treated by a therapy other than ANTRA. Therefore we cannot conclude that induction of OS in patients undergoing ANTRA therapy is caused only by ANTRA, and that no other factors play a role in this process. Only limited data are available.

Only limited data are available on the dynamics of late OS in patients treated with ANTRA. To the best of our knowledge, there is only one previously published study addressing the issue of late OS following ANTRA therapy. This study failed to demonstrate any long-lasting (3-12 months after the end of therapy) increase in OS (15).

The results of our study provide us with additional data to supplement our current knowledge. The increased activity of antioxidant enzymes and concentrations of OS products after ANTRA therapy, in comparison to the baseline, were observed in our study, with the follow-up median of 45 days after the end of ANTRA therapy. Since similar changes were also found in the control group, these changes were not caused by ANTRA therapy. The increased levels of OS in patients following ANTRA therapy are probably explained by reasons other than ANTRA therapy alone.

Although follow-up examinations were performed after the end of ANTRA therapy, this did not correspond to the end of anti-cancer therapy. A number of patients were receiving other types of chemotherapy or radiotherapy at the time of follow-up examinations (other chemotherapy in the study group in 59% of cases [taxanes in 59% of these cases], other chemotherapy in the control group in 26% of cases [p=0.0007]; radiotherapy in 51% of cases in the study group, radiotherapy in 37% of cases in the control group [p=0.14]). Even though one could speculate on the possible influence of ongoing anti-cancer therapy on OS levels, we failed to find any statistically significant difference, in either activities of antioxidant enzymes or concentrations of conjugated dienes, between the groups of patients, with or without ongoing anti-cancer therapy (neither chemotherapy nor radiotherapy).

There was a statistically significant decrease in hemoglobin (p<0.00001) and albumin (p=0.00004) concentrations after ANTRA therapy, in comparison to the baseline, in the study patients, facts that may represent other factors linked to OS induction (aenemia, malnutrition). Surprisingly, our study found no significant correlation between changes in  hemoglobin/albumin concentrations on the one hand, and changes in the activities of selected antioxidant enzymes and/or concentrations of conjugated dienes on the other, except for paraoxonase-1. Weak, but statistically significant correlations were found between changes in hemoglobin/albumin concentrations and changes in paraoxonase-1 activity (Δalbumin/Δparaoxonase-1: r=0.25; p=0.002, Δhemoglobin/Δparaoxonase-1: r=0.25; p<0.0001) after the therapy, in comparison to the baseline.

Based on available data and the results of our study, we can conclude that ANTRA therapy probably causes OS induction during the therapy (6). There is an ongoing increase in OS levels in patients after the end of ANTRA therapy, but factors other than ANTRA are probably responsible for this persistent increase: the tumor’s influence, the impact of the activity of antioxidant enzymes, other anti-cancer therapies (apart from ANTRA) and complications arising from the disease. No long-term increase in OS levels can be observed later (3-12 months) after the end of ANTRA therapy (13).

With the results of our study, a number of questions have arisen concerning ANTRA-induced cardiomyopathy. Since ANTRA-induced OS is only present during ANTRA therapy, the pathogenesis of ANTRA-induced cardiomyopathy remains unclear.

Is it possible that myocardial changes caused by OS during ANTRA therapy are not detectable at the time, but are later able to manifest as cardiomyopathy? Despite the fact that morphological and functional myocardial changes are detectable during ANTRA therapy, none of them have been able to predict the development of ANTRA-induced cardiomyopathy so far. It therefore remains questionable whether these changes are not in fact more likely to reflect acute cardiotoxicity. If we believe OS to be an important factor in ANTRA-induced cardiomyopathy, what is the role of OS, unrelated to ANTRA therapy (OS linked to cancer, OS connected to other treatments, OS connected to disease/treatment complications), in the development of ANTRA-induced cardiomyopathy?

Study limitations:

Our study has one significant limitation with regard to ANTRA-induced cardiotoxicity, which is the primary object of our program. Since we do not have any data on the correlation between systemic and myocardial levels of OS, the major limitation of our study is that OS was studied on a systemic level and not on a myocardial level. But the aim of our study was to evaluate the levels of OS induced by ANTRA therapy. Therefore we believe that the above-mentioned limitation does not influence the actual results of our study. On the other hand, the above-mentioned facts could limit the application of the study’s results concerning ANTRA-induced cardiomyopathy pathophysiology.

Conclusion:

Increased levels of OS are present in patients with solid tumors treated with an anti-cancer protocol consisting of ANTRA, even after the end of ANTRA treatment. However, our study demonstrates that the increase in OS in this period is not caused by the ANTRA therapy. Despite the fact that the reasons for a persistent increase in OS levels remain unknown, one can speculate that the ongoing impact of an oncologic disease or of complications plays an important role in this process. Since OS (induced by ANTRA therapy) is believed to be an important factor in ANTRA-induced cardiotoxicity, the role of OS, induced by factors other than ANTRA, in the pathogenesis of ANTRA-induced cardiotoxicity remains to be elucidated. 

List of abbreviations:

• ANTRA ‒ anthracyclines
• OS ‒ oxidative stress
• LDL ‒ low density lipoproteins
• SD ‒ standard deviation
• IQR ‒ interquartile range

The study was supported by grant IGA MZ ČR NS 9774-4

Address for correspondence:

MUDr. Kocík Miroslav Ph.D

IV. interní klinika

Všeobecná fakultní nemocnice v Praze

U Nemocnice 2

Praha 2

128 08

Česká republika

telefon: +420 2 24 96 25 04

fax: +420 2 24 96 28 79

miroslav.kocik@seznam.cz