Paraoxonase-1 activity in patients with cancer_ A systematic review ...

Critical Reviews in Oncology / Hematology 127 (2018) 6–14

Contents lists available at ScienceDirect

Critical Reviews in Oncology / Hematology journal homepage: www.elsevier.com/locate/critrevonc

Paraoxonase-1 activity in patients with cancer: A systematic review and meta-analysis

T

M. Arenasa, E. Rodrígueza,b, A. Sahebkarc,d, S. Sabatere, D. Rizoa, O. Palliséa, M. Hernándeza, ⁎ F. Riuf, J. Campsb, , J. Jovenb a

Department of Radiation Oncology, Hospital Universitari Sant Joan, Institut d’Investigació Sanitària Pere i Virgili, Universitat Rovira i Virgili, Reus, Spain Unitat de Recerca Biomèdica, Hospital Universitari Sant Joan, Institut d’Investigació Sanitària Pere i Virgili, Universitat Rovira i Virgili, Reus, Spain c Biotechnology Research Center, Pharmaceutical Technology Institute, Mashhad University of Medical Sciences, Mashhad, Iran d Neurogenic Inflammation Research Center, Mashhad University of Medical Sciences, Mashhad, Iran e Department of Radiation Oncology, Complejo Hospitalario Universitario de Albacete, Albacete, Spain f Department of Pathology, Hospital Universitari Sant Joan, Institut d’Investigació Sanitària Pere i Virgili, Universitat Rovira i Virgili, Reus, Spain b

A R T I C LE I N FO

A B S T R A C T

Keywords: Meta-analysis Oxidative stress paraoxonase-1

Purpose: Paraoxonase-1 (PON1) is a lipolactonase implicated in the elimination of carcinogenic free radicals and in the scavenging mechanisms to maintain oxidative balance. The objective of the present systematic review and meta-analysis was to evaluate possible alterations in serum PON1 activity in patients with cancer. Methods: A systematic search of the observational studies in humans published in the last 15 years was performed through Medline databases following the PRISMA and STARLITE statements. Further, a keyword-based computerized search with restrictions on publication date, and a meta-analysis of case-control studies was performed. Results: In total, 23 studies were included most of which reported decreased PON1 activity in patients with cancer. This could indicate impaired defense ability against oxidative stress with potential implications in cell proliferation, promotion of genetic instability, and alterations in cellular sensitivity to chemotherapy. Conclusion: This systematic review and meta-analysis confirms a consistent association between cancer and decreased serum PON1 activities. These findings may open fruitful lines of research with clinical relevance, and an understanding of molecular alterations underlying carcinogenesis.

1. Introduction Cancer is often associated with oxidative stress, an outcome of which is disruption in the balance between the systemic effects of toxic reactive species and the body’s capacity to metabolize the excess free radicals and/or to repair the damage that they generate (Betteridge, 2000). Experimental studies have reported increased production of reactive oxygen species in cancerous cells that can cause tumor proliferation, promotion of genetic instability, and alterations in cellular sensitivity to chemotherapy (Michalak et al., 2014). Paraoxonase-1 (PON1) is an antioxidant enzyme found in the membranes of most cells and, as well, in the circulation bound to highdensity lipoproteins (HDL) (Aviram and Rosenblat, 2004; Chander and Kapoor, 1990; Camps et al., 2009). The original function attributed to

PON1 was that of a lactonase (LAC) since lipophilic lactones constitute its primary substrates (Khersonsky and Tawfik, 2005). It is this catalytic capacity that enables PON1 to degrade lipid peroxides within the cell as well as in the lipoproteins in circulation (Ponce-Ruiz et al., 2017). In addition, PON1 has an esterase activity which degrades organophosphate xenobiotics and nerve agents (Camps et al. 2009). PON1 is primarily synthesized by the liver and, in lesser amounts, in the kidney and colon and, subsequently, transported into the blood stream bound to HDL (Mackness and Mackness, 2015). PON1 plays an important role in lipid metabolism as an antioxidant molecule, and in the control of inflammation. PON1 enzyme activity is influenced by inflammation changes and by oxidized low-density lipoprotein (LDL) levels. PON1 has been shown to protect against oxidative stress through the hydrolysis of active oxidized phospholipids, the destruction of lipid hydroperoxides

Abbreviations: ARE, arylesterase; CI, confidence interval; CMA, comprehensive meta-analysis; CRP, C-reactive protein; HDL, high-density lipoproteins; LAC, lactonase; LDL, low-density lipoproteins; PON, paraoxonase; PON1, paraoxonase-1; PRISMA, preferred reporting items for systematic reviews and meta-analyses; SD, standard deviation; SEM, standard error of the mean; SMD, standardized mean difference; STARLITE, sampling strategy, type of study, approaches, range of years, limits, inclusions and exclusions, terms used, electronic sources; TSA, trial sequential analysis ⁎ Corresponding author at: C.Sant Joan s/n, 43201 Reus, Catalonia, Spain. E-mail address: [email protected] (J. Camps). https://doi.org/10.1016/j.critrevonc.2018.04.005 Received 12 February 2018; Received in revised form 12 April 2018; Accepted 23 April 2018 1040-8428/ © 2018 Elsevier B.V. All rights reserved.


M. Arenas et al.

the Newcastle-Ottawa scale are detailed in Table 1.

(carcinogenic lipid-soluble radicals from lipid peroxidation) and H2O2 (via its peroxidase-like activity), the preservation of HDL integrity and function, and the prevention of LDL oxidation (Antognelli et al., 2009). It also exhibits many athero-protective properties such as reducing macrophage oxidative stress while reducing the capacity of macrophages to oxidize LDL. The overall effects are to inhibit cholesterol synthesis and to promote cellular cholesterol efflux. PON1 activity can be measured spectrophotometrically as a function of its catalytic ability to hydrolyze paraoxon (i.e. PON activity), phenylacetate (arylesterase; ARE activity) or synthetic lactones (such as thiobutyl butyrolactone; LAC activity). It is of note that PON1 varies considerably between individuals (both with respect to concentration as well as activity) and this variability has been related to several factors such as genetic mutations (explaining > 60% of variability) and other factors such as smoking, stress, high fat diets, inflammation, and treatment with fibrates (Costa et al., 2005; Eroglu et al., 2013; Précourt et al., 2011). The objective of the present systematic review and meta-analysis was to re-assess the published studies in relation to variations in serum PON1 concentration and activity in patients with cancer.

2.4. Eligibility criteria for quantitative synthesis (meta-analysis) To be included in the meta-analysis, the studies needed to have had a case-control design, with the cases being the cancer patients and the controls being healthy subjects. Also, it must have had a measurement of PON1 activity when the patients had not as-yet received treatment and must have the following data: number of patients in each group, PON or ARE mean activity in each group and standard deviation or confidence interval (CI) for the mean activity. We used the Meta-Essentials workbooks for meta-analysis (Version 1.0), for Microsoft Office Excel 2016, for all of the analyses. The transformed and calculated data consisted mostly of the mean levels of PON and ARE activities pre-treatment in cancer patient group and the mean values of PON and ARE activities in the control group. LAC activity was not considered for meta-analysis because it had been evaluated in only two studies. The weighted mean estimated effect size and 95% CI were the summary statistics and was calculated via Hedges “g” effect.

2. Methods 2.5. Quantitative data synthesis 2.1. Search protocol and eligibility criteria A meta-analysis was conducted using Comprehensive Meta-Analysis (CMA) V2 software (Pierce et al., 2006). Standard deviation (SD) of the mean difference was calculated using the formula: SD = √[(SDcase)2 + (SDcontrol)2 – (2R × SDcase × SDcontrol)], assuming a correlation coefficient (R) = 0.5. Where only the standard error of the mean (SEM) was reported, SD was estimated using the formula: SD = SEM × √n where n is the number of subjects studied. A random-effects model (using DerSimonian-Laird method) and the generic inverse variance method were used to compensate for the heterogeneity of studies with respect to demographic characteristics of studied populations and to differences in study design. Heterogeneity was quantitatively assessed using I2 index. Effect sizes were expressed as standardized mean difference (SMD) and 95% CI. To evaluate the influence of each study on the overall effect size, sensitivity analysis was conducted using leave-one-out method, i.e. removing one study in each successive repeat of the analysis.

Following the PRISMA and STARLITE recommendations, we performed a search involving an expert chain-of-citations approach, followed by a keyword-based computerized search with restrictions on publication date. We searched Medline and SCOPUS databases using an overlapping subsets approach, and applying the following Boolean expression: “paraoxonase” or “paraoxonase-1” or “paraoxonase 1” or “paraoxonase1” or “PON-1” or “PON 1” or “PON1” or “arylesterase” and "activity" or "concentration" and "cancer". We used all interchangeable formats of the search terms. No articles were identified in the Cochrane database of systematic reviews. Hand searching of references of retrieved articles identified additional studies, of which 4 reports met our inclusion criteria. Within the inclusion criteria were observational studies in humans published in English over the past 15 years. The studies addressed: groups of patients with cancer (irrespective of the cancer type or the treatment received), quantitative measurement of serum PON1 concentration and/or activity (either PON, ARE or LAC). Exclusion criteria were studies not involving human subjects i.e. in-vitro studies or those conducted with experimental animals. There were no geographic inclusion/exclusion criteria.

2.6. Publication bias Potential publication bias was explored using visual inspection of Begg’s funnel plot asymmetry, fail-safe N test, Begg’s rank correlation, and Egger’s weighted regression tests. Duval & Tweedie “trim and fill” method was used to adjust the analysis for the effects of publication bias (Duval and Tweedie, 2000).

2.2. Data extraction Data that were extracted included: year of publication, study design, type of cancer, sample size, and clinical and biochemical variables analyzed. Main conclusions were summarized, and missing data were obtained directly from the respective corresponding authors. In some cases, these data were not forthcoming, and the studies were deleted from our analysis.

2.7. Trial sequential analysis We applied trial sequential analysis (TSA) using TSA version 0.9.5.9 Beta (Copenhagen: The Copenhagen Trial Unit, Center for Clinical Intervention Research, 2016) to determine whether the available evidence was robust and conclusive through estimation of the required information (sample) size. Boundaries for concluding the difference in PON and ARE activities between cancer and control groups were calculated based on the O'Brien-Fleming α-spending function. Our assumptions were two-sided testing with type I error and power values of 5% and 80%, respectively. Judgment of robustness and lack of need for further trials was made when the cumulative Z-curve crossed the defined trial sequential monitoring boundary. Judgment of lack of robustness was made when the cumulative Z-curve did not cross the defined trial sequential monitoring boundary, and the required information size was not reached.

2.3. Quality assessment Methodological quality of the included studies was assessed using the Newcastle-Ottawa Scale (Wells et al., 2009) on which three aspects of each eligible study were evaluated: i) the selection of the studied patients (4 items); ii) the comparability of the studied populations (one item); iii) the ascertainment of exposure (3 items) in case-control studies or outcome-of-interest in cohort studies. A study can be awarded a maximum of one point for each item in the selection and exposure categories, whilst the comparability item can receive a maximum two points. Quality of bias assessment of the included studies according to 7


M. Arenas et al.

Table 1 Quality of bias assessment of the included studies according to the Newcastle-Ottawa scale. Study

Faridvand et al. (2016) Korkmaz et al. (2015) Ahmed et al. (2015) Çebi et al. (2015) Afsar et al. (2015) Ellidag et al. (2014) Michalak et al. (2014) Sehitogulları et al. (2014) Malik et al. (2014) Aydin et al. (2013) Eroglu et al. (2013) Bulbuller et al. (2013) Balci et al. (2012) Vecka et al. (2012) Ozturk et al. (2011) Karaman et al. (2010) Arioz et al. (2009) Camuzcuoglu et al. (2009) Krzystek-Korpacka et al. (2008) Elkiran et al. (2007) Akçay et al. (2003a, 2003b) Akçay et al. (2003a, 2003b)

Selection

Comparability†

Exposure

Case definition

Representativeness of the cases

Selection of controls

Definition of controls

Comparability of cases and controls

Ascertainment of exposure

Same method of ascertainment

Nonresponse rate

✰

–

✰

–

✰✰

✰

✰

–

✰

–

✰

✰

✰✰

✰

✰

–

✰ ✰ ✰ ✰ ✰

– – – – –

✰ ✰ ✰ – ✰

– ✰ ✰ – –

✰ ✰✰ ✰✰ ✰ ✰✰

✰ ✰ ✰ ✰ ✰

✰ ✰ ✰ ✰ ✰

– – – – –

✰

–

✰

✰

✰✰

✰

✰

–

✰ ✰ ✰ ✰

– – – –

✰ ✰ ✰ ✰

✰ ✰ ✰ ✰

– ✰✰ ✰✰ ✰✰

✰ ✰ ✰ ✰

✰ ✰ ✰ ✰

– – – –

✰ ✰ ✰ ✰

– ✰ – –

✰ – ✰ ✰

✰ – – ✰

✰✰ ✰✰ ✰✰ ✰✰

✰ ✰ ✰ ✰

✰ ✰ ✰ ✰

– – – –

✰ ✰

– ✰

✰ ✰

– ✰

✰✰ ✰✰

✰ ✰

✰ ✰

– –

✰

–

✰

✰

✰

✰

✰

–

✰ –

✰ –

✰ ✰

✰ –

✰✰ –

✰ ✰

✰ ✰

– –

–

–

✰

–

–

✰

✰

–

3. Results

hydatidiform mole, relative to the healthy subjects (Ozturk et al., 2011).

Of the 63 reports initially identified, 62 were screened in detail. In the primary analysis, 36 studies were excluded. Our selection criteria were fulfilled in 26 reports, and the full-text articles were assessed for eligibility. In the secondary analysis, we selected 23 studies of which 17 were considered for quantitative synthesis (Fig. 1). The majority of reports (22) had a case-control study design, but one consisted of a series of cases. Characteristics of the studies finally included are detailed in Table 2. The overall reviewed data are summarized in Supplementary Figs. 1 and 2.

3.1.2. Gastrointestinal cancer Several studies (Ahmed et al., 2015; Bulbuller et al., 2013; Balci et al. 2012) found that PON and ARE activities and HDL concentrations were significantly lower in patients with colorectal cancer than in the control group. Another important observation in these studies was that PON activity was lower in patients with metastases when compared to patients without metastases. However, there were no differences in these activities when the tumors were segregated with respect to histologic grade. Further, PON/HDL and PON/ARE ratios were significantly lower in colorectal cancer patients relative to healthy control subjects, and both enzyme activities increased significantly post-surgery to reach normal levels after one month (Ahmed et al. 2015). In contrast to previous studies, another report performed the blood analysis post-surgery and observed that the PON and ARE activities of the patients who had colorectal cancer operated-upon were significantly higher, relative to the healthy subjects. In addition, there were no significant differences in PON or ARE activities when segregated with respect to gender, colon or rectal location, or the presence of metastasis (Afsar et al., 2015). Three articles reported, compared to control subjects, lower PON (Sehitogulları et al., 2014; Krzystek-Korpacka et al., 2008; Akçay et al. 2003a, 2003b) and ARE (Sehitogulları et al., 2014; Krzystek-Korpacka et al., 2008) activities in patients with gastric cancer, or with esophageal squamous cell carcinomas, adenocarcinomas of lower esophagus and gastric cardia or adenocarcinomas of the proximal stomach. Further, one of these studies (Akçay et al. 2003a, 2003b) reported a positive correlation between HDL levels and PON activity, and another (Krzystek-Korpacka et al., 2008) found that the

3.1. Types of cancer 3.1.1. Gynecologic cancer In the breast cancer studies (Balci et al., 2012; Bobin-Dubigeon et al., 2012) the patients had significantly lower serum HDL cholesterol concentrations as well as PON and ARE activities, relative to the control group. In addition, low ARE activities were associated with lower survival. However, when patients with and without metastases were compared, no significant differences were found in the enzyme activities. One study conducted in ovarian cancer patients suggested that decreased PON and ARE activities were associated with high stage, high grade tumor, and with pre-operative concentrations of the tumor marker CA-125 (Camuzcuoglu et al., 2009). However, another study in ovarian cancer patients (Michalak et al., 2014) did not find any significant differences in serum PON1 activity. In endometrial cancer, it was found that serum PON and ARE activities were significantly lower in patients relative to controls (Arioz et al., 2009). By contrast, there were not found any significant differences in 26 patients with 8


M. Arenas et al.

Fig. 1. PRISMA flow diagram of studies included in the meta-analysis.

different esophageal tumor locations (upper, middle or lower) nor preand post-adjuvant treatment (chemotherapy or radiotherapy). In patients with pancreatic cancer, two studies found significantly lower ARE, LAC (Vecka et al. 2012), PON (Akçay et al., 2003a, 2003b) activities and serum HDL levels, when compared with healthy controls. Further, investigators performed several subgroup analyses on patients with pancreatic cancer segregated according to their nutritional level and found that those patients with pancreatic cancer and clinically-

decrease in ARE activity reflected the presence of lymph node metastasis, and correlated with tumor extension. ARE activity was also reduced with increasing Glasgow prognostic score (an inflammation and malnutrition index based on individual albumin and C-reactive protein (CRP) concentrations), and with the severity of anemia. However, no significant differences were found, either in PON or ARE activities, between the histology grade of the cancer and the presence or absence of distant metastasis. Neither were there significant differences between Table 2 Characteristics of the studies included. First author

Year

Study design

Cancer analyzed

Patients (n)

Clinical parameter studied

Biochemical parameters studied

Faridvand Bobin-Dubigeon Korkmaz Ahmed Çebi Afsar Ellidag Michalak Sehitogullari Malik Aydin Eroglu Bulbuller Balci Vecka Ozturk Karaman Arioz Camuzcuoglu Krzystek-Korpacka Elkiran Akçay Akçay

2016 2015 2015 2015 2015 2015 2014 2014 2014 2014 2013 2013 2013 2012 2012 2011 2010 2009 2009 2008 2007 2003 2003

Case-control study Series of cases Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study Case-control study

Stage 1 multiple myeloma Breast (recurrence) Papillary thyroid cancer Colorectal cancer Acute myeloid leukemia Colorectal cancer Multiple myeloma Ovarian cancer and benign ovarian tumors Esophageal squamous cell carcinoma Oral squamous cell carcinoma Bladder cancer Prostate cancer Colorectal cancer Lung, breast and colorectal cancer Pancreatic cancer Hydatidiform mole Laryngeal squamous cell carcinoma Endometrial cancer Epithelial ovarian cancer Gastroesophageal malignancies Lung cancer Gastric cancer Pancreatic cancer

69 49 52 130 32 63 40 91 64 75 82 63 79 139 146 80 50 43 53 139 78 40 40

Not considered Short term death Not considered Not considered Not considered Not considered Prognosis Tumor size Cancer progression Not considered Not considered Not considered Not considered Not considered Malnutrition Not considered Not considered Not considered Not considered Clinical pathological Not considered Not considered Not considered

PON, ARE PON, ARE, LAC PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE PON, ARE ARE, LAC PON, ARE PON PON, ARE PON, ARE PON, ARE PON, ARE PON PON

Abbreviations: PON, paraoxonase; ARE, arylesterase; LAC, lactonase. 9


M. Arenas et al.

Fig. 2. Forest plot displaying standardized mean difference and 95% confidence intervals comparing serum paraoxonase (PON) activities between patients with cancer and unaffected controls. Lower plot indicates leave-one-out sensitivity analysis for the meta-analysis of serum PON changes between patients with cancer, and unaffected control subjects.

3.1.4. Head and neck cancer PON and ARE activities have been reported to be significantly decreased in oral, laryngeal, or thyroid cancer patients. In addition, following total thyroidectomy in patients with papillary thyroid cancer, both enzyme activities increased significantly. In one of these studies (Karaman et al. 2010), no statistically significant differences were found in PON activity with respect to tumor differentiation stage, tumor location (glottis, supraglottic or transglottic) or tobacco and alcohol abuse.

relevant malnutrition had significantly lower ARE and LAC activities than those without an important grade of malnutrition. ARE and LAC activities correlated positively with nutritional status (Akçay et al. 2003a, 2003b).

3.1.3. Hematologic cancer ARE activity, but not PON activity, was reported to be decreased in patients with acute myeloid leukemia (Çebi et al., 2015). An evaluation of PON and ARE activities in patients with stage I multiple myeloma found that both activities and PON/HDL and ARE/HDL ratios were significantly lower in patients, compared with the control group (Faridvand et al. 2016) Further, one study investigated prognostic factors in this disease and concluded that myeloma patients without free light chains excretion have a poorer prognosis, and which was associated with decreased PON and ARE activities (Ellidag et al., 2014).

3.1.5. Urologic cancer A study performed in patients with bladder cancer showed lower levels of ARE enzyme activity, compared to controls; no significant differences were recorded with respect to PON enzyme activity (Aydin et al., 2013). Conversely, in patients with prostate cancer, ARE activity remained within the normal range, while PON activity was increased 10


M. Arenas et al.

Fig. 3. Forest plot displaying standardized mean difference and 95% confidence intervals comparing serum arylesterase (ARE) activities between patients with cancer, and unaffected controls. Lower plot indicates leave-one-out sensitivity analysis for the meta-analysis of serum ARE changes between patients with cancer, and unaffected control subjects.

with a 95% CI from 0.73 to 2.08. The weighted mean estimated effect size for ARE activity was 1.25 with a 95% CI from 0.62 to 1.87. Supplementary Table 1 shows the data extracted from every individual study included in the analysis. The global effect can be seen in Supplementary Tables 2 and 3.

(Eroglu et al. 2013). Lung cancer: Two studies (Elkiran et al., 2007; Balci et al., 2012) investigated untreated lung cancer patients and found decreased PON and ARE activities relative to the control group. Nevertheless, there were no statistical differences between PON and ARE activities in relation to tumor histology grade, nor the presence or absence of metastases (Elkiran et al., 2007). Further, HDL cholesterol levels were significantly lower in patients, compared to controls. However, there was no significant correlation between HDL and PON or ARE activities (Balci et al., 2012).

3.3. Comparisons of PON and ARE activities between cancer patients and controls Overall, 16 studies comprising 19 arms compared PON activity between cancer patients and controls. Meta-analysis suggested a lower plasma PON activity in patients with cancer, compared to controls (SMD: −2.23, 95% CI: −2.91, −1.55, p < 0.001; I2 = 95.61%) (Fig. 2). Comparisons of plasma ARE activities were reported in 16

3.2. Results of the statistical analysis The weighted mean estimated effect size for PON activity was 1.40 11


M. Arenas et al.

Fig. 4. Funnel plots detailing publication bias in the meta-analyses of paraoxonase (PON, upper plot) and arylesterase (ARE, lower plot) activities. Open diamonds represent observed effect size; closed diamonds represent imputed effect size.

3.5. Trial sequential analysis

studies comprising 18 arms. Meta-analysis suggested a lower plasma ARE activity in patients with cancer, compared to controls (SMD: −2.16, 95% CI: −2.88, −1.43, p < 0.001; I2 = 96.51%) (Fig. 3). The above-mentioned results on PON and ARE activities were robust in the sensitivity analysis; the statistical significance was not influenced following successive omission of the selected studies included in the metaanalysis.

Considering α and β values of 0.05 and 0.2, respectively, the cumulative Z-curve (blue curve) crossed the conventional boundary of 5% significance (horizontal red line) as well as the monitoring boundaries (red inward sloping curves) for meta-analyses of PON and ARE activities. As such, no further studies were required (Fig. 5). 4. Discussion

3.4. Publication bias Oxidative stress is a primary etiological factor in carcinogenesis. When the oxidative balance deregulates in favor of pro-oxidants, cellular damage can be caused via several different pathways. PON1 is one of the endogenous free-radical scavenging systems in the human body. Despite current studies being unable to identify the exact mechanisms by which PON1 eliminates carcinogenic lipid-soluble radicals, the research to date does provide an insight into the molecular pathways involved in malignant cell transformation and proliferation. Hence, the potential roles of PON1 in various cancer-related aspects, including etiology and prevention, are of increasing interest (Goswami et al., 2009). The perceived order of events impinging on variations in PON1 levels and cancer development remains hotly contested. One hypothesis

There was evidence of publication bias in the meta-analyses of PON and ARE, as indicated by the results of Egger’s linear regression and Begg’s rank correlation tests (Supplementary Table 4). The funnel plots of the study standard error by effect size (SMD) were also asymmetric for both analyses (Fig. 4). These asymmetries were addressed by imputing potentially missing studies using the “trim and fill” method; the differences between the groups with respect to PON and ARE activities remaining statistically significant following imputation of potentially missing studies. The “fail-safe N” test also confirmed the significant difference in PON and ARE activities between cancer and control groups.

12


M. Arenas et al.

Fig. 5. Trial sequential analysis of changes in serum paraoxonase (PON) and arylesterase (ARE) activities in patients with cancer, compared with unaffected control subjects.

theory is reinforced by the results showing (Ahmed et al., 2015) that serum PON and ARE activities were significantly increased one month post-surgery in patients with colorectal cancer. Nevertheless, results from other studies indicate that PON1 activities against different substrates are affected in different ways. A study (Bobin-Dubigeon et al., 2015) showed that only the ARE activity of PON1 decreases in the presence of inflammation. As such, their data do not support the concept of a direct effect of inflammation on the enzyme’s expression. Further, although HDL decreases with inflammation, no changes in oxidized LDL were observed; suggesting that the PON1 anti-oxidative activity was not really affected. That ARE activity would be the only affected hydrolytic function of the enzyme is unclear. The authors raised the hypothesis that some cytokines, or substances generated during inflammation, would decrease the hydrolysis rate by competing with the substrate or by modifying the enzyme’s conformation status

suggests that genetic polymorphisms influencing serum PON1 activity can increase the risk, and contribute to the development of certain tumors. In contrast, another hypothesis suggests that the modifications in serum PON1 levels are not a cause of cancer, but rather a consequence of the systemic inflammatory status in oncological conditions. In the following paragraphs, we compile the evidence that supports these hypotheses The objectives of many of the studies published on PON1 and cancer had been to establish risk groups segregated with respect to PON1 genotype and/or PON1 activities in patients with cancer. However, results have been contradictory. It was suggested that, in order to cope with increasing cellular demand by the growing tumor, cancer cells sequester nutrients from circulation and, due to their utilization, result in decreased antioxidant levels in the circulation. Consequently, lower available PON1 levels would be expected to result in increased lipid peroxidation, and the attendant sequelae (Malik et al., 2014). This

13


M. Arenas et al.

and, as such, adversely affecting its function. This hypothesis would need further in vitro studies for confirmation. Some of the studies we have reviewed here support this hypothesis. For instance, in gastroesophageal cancers, decreases in PON as well as ARE activities were detectable, and the decreases correlated well with the levels of circulating inflammatory markers, including CRP and interleukin-6 (Sehitogulları et al., 2014; Malik et al., 2014). It has been suggested (Korkmaz et al., 2015) that thyroid cancer cells might enhance PON1 internalization into the cells to protect them from cell death induced by increased oxidative stress, and this could explain the findings of reduced serum PON and ARE activities. In addition, serum PON activity was significantly increased in patients with prostate carcinoma, while ARE activity was not (Eroglu et al., 2013). This finding reaffirms the concept that PON1 is a multi-facetted enzyme with different active sites responsible for distinctly separate specialized activities. Some studies assessed the relationship between serum PON1 activities and tumor stage, and reported that local progress of the disease resulted in lowered PON and ARE levels (Camuzcuoglu et al., 2009; Krzystek-Korpacka et al., 2008). Conversely, other studies did not find any significant differences in PON1 activities in relation to the presence of metastases (Afsar et al., 2015; Balci et al., 2012; Krzystek-Korpacka et al., 2008) and, as well, did not find any significant associations between PON or ARE levels and tumor histology stage, or location (Afsar et al., 2015; Sehitogulları et al., 2014; Bulbuller et al., 2013). It is important to note that the methods used in determining the different activities of PON1 differ, not only in substrate employed but also in the selection of methodological criteria including buffers, pH and temperature. This non-standardization is a handicap that preempts comparisons of values obtained by the different studies. Further, studies on PON1 and cancer are scarce, the types of cancer are diverse and heterogeneous, some have small study samples and in others, patients and controls are not matched for genotype. The different polymorphisms influencing serum PON1 activity need to be considered (Mackness and Mackness, 2015) as well as concomitant diseases (hypertension, liver failure, cardiovascular diseases, renal failure, diabetes mellitus) and other factors potentially influencing PON1 activity including medications, tobacco and alcohol abuse. In conclusion, the present systematic review and meta-analysis confirms a consistent association between cancer and decreased serum PON1 activities. Only two studies showed a correlation with prognosis; one in breast cancer recurrence and another in multiple myeloma (Bobin-Dubigeon et al., 2015; Ellidag et al., 2014). This meta-analysis would suggest that more extensive studies need to be performed in larger series of patients with different types of neoplasias in order to confirm, and extend, the current findings. A promising area of research is the investigation of the usefulness of serum PON1 measurement as a biomarker for screening, diagnosis and measurement of treatment efficacy with respect to prognosis and survival in patients with cancer.

Arioz, D.T., Camuzcuoglu, H., Toy, H., Kurt, S., Celik, H., Erel, O., 2009. Assessment of serum paraoxonase and arylesterase activity in patients with endometrial cancer. Eur. J. Gynaecol. Oncol. 30, 679–682. Aviram, M., Rosenblat, M., 2004. Paraoxonases 1, 2, and 3, oxidative stress, and macrophage foam cell formation during atherosclerosis development. Free Radic. Biol. Med. 37, 1304–1316. Aydin, O., Yacinkaya, S., Eren, E., Ergin, M., Eroglu, M., Yilmaz, N., 2013. Diminished arylesterase enzyme activity and total thiol levels in bladder cancer patients. Clin. Lab. 59, 1231–1237. Balci, H., Genc, H., Papila, C., Can, G., Papila, B., Yanardag, H., Uzun, H., 2012. Serum lipid hydroperoxide levels and paraoxonase activity in patients with lung, breast, and colorectal cancer. J. Clin. Lab. Anal. 26, 155–160. Betteridge, D.J., 2000. What is oxidative stress? Metabolism 49 (Suppl (2)), 3–8. Bobin-Dubigeon, C., Jaffré, I., Joalland, M.P., Classe, J.M., Campone, M., Hervé, M., Bard, J.M., 2012. Paraoxonase 1 (PON1) as a marker of short term death in breast cancer recurrence. Clin. Biochem. 45, 1503–1505. Bobin-Dubigeon, C., Lefrançois, A., Classe, J.M., Joalland, M.P., Bard, J.M., 2015. Paired measurement of serum amyloid A (SAA) and paraoxonase 1 (PON1) as useful markers in breast cancer recurrence. Clin. Biochem. 48, 1181–1183. Bulbuller, N., Eren, E., Ellidag, H.Y., Oner, O.Z., Sezer, C., Aydin, O., Yılmaz, N., 2013. Diagnostic value of thiols, paraoxonase 1, arylesterase and oxidative balance in colorectal cancer in human. Neoplasma 60, 419–424. Camps, J., Marsillach, J., Joven, J., 2009. The paraoxonases: role in human diseases and methodological difficulties in measurement. Crit. Rev. Clin. Lab. Sci. 46, 83–106. Camuzcuoglu, H., Arioz, D.T., Toy, H., Kurt, S., Celik, H., Erel, O., 2009. Serum paraoxonase and arylesterase activities in patients with epithelial ovarian cancer. Gynecol. Oncol. 112, 481–485. Çebi, A., Akgun, E., Esen, R., Demir, H., Çifci, A., 2015. The activities of serum paraoxonase and arylesterase and lipid profile in acute myeloid leukemia: preliminary results. Eur. Rev. Med. Pharmacol. Sci. 19, 4590–4594. Chander, R., Kapoor, N.K., 1990. High density lipoprotein is a scavenger of superoxide anions. Biochem. Pharmacol. 40, 1663–1665. Costa, L.G., Cole, T.B., Vitalone, A., Furlong, C.E., 2005. Measurement of paraoxonase (PON1) status as a potential biomarker of susceptibility to organophosphate toxicity. Clin. Chim. Acta 352, 37–47. Duval, S., Tweedie, R., 2000. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56, 455–463. Elkiran, E.T., Mar, N., Aygen, B., Gursu, F., Karaoglu, A., Koca, S., 2007. Serum paraoxonase and arylesterase activities in patients with lung cancer in a Turkish population. BMC Cancer 7, 48. Ellidag, H.Y., Aydin, O., Eren, E., Yilmaz, N., Ergin, M., 2014. Decreased HDL-dependent paraoxonase and arylesterase enzyme activity may indicate a worse prognosis in multiple myeloma. Asian Pac. J. Cancer Prev. 15, 9847–9851. Eroglu, M., Yilmaz, N., Yalcinkaya, S., Ay, N., Aydin, O., Sezer, C., 2013. Enhanced HDL-cholesterol-associated anti-oxidant PON-1 activity in prostate cancer patients. Kaohsiung J. Med. Sci. 29, 368–373. Faridvand, Y., Oskuyi, A.E., Khadem-Ansari, M.H., 2016. Serum 8-isoprostane levels and paraoxonase 1 activity in patients with stage I multiple myeloma. Redox Rep. 21, 204–208. Goswami, B., Tayal, D., Gupta, N., Mallika, V., 2009. Paraoxonase: a multifaceted biomolecule. Clin. Chim. Acta 410, 1–12. Karaman, E., Uzun, H., Papila, I., Balci, H., Ozdilek, A., Genc, H., Yanardag, H., Papila, C., 2010. Serum paraoxonase activity and oxidative DNA damage in patients with laryngeal squamous cell carcinoma. J. Craniofac Surg. 21, 1745–1749. Khersonsky, O., Tawfik, D.S., 2005. Structure-reactivity studies of serum paraoxonase PON1 suggest that its native activity is lactonase. Biochemistry 44, 6371–6382. Korkmaz, H., Tabur, S., Özkaya, M., Aksoy, N., Yildiz, H., Akarsu, E., 2015. Paraoxonase and arylesterase activities in patients with papillary thyroid cancer. Scand. J. Clin. Lab. Invest. 75, 259–264. Krzystek-Korpacka, M., Boehm, D., Matusiewicz, M., Diakowska, D., Grabowski, K., Gamian, A., 2008. Paraoxonase 1 (PON1) status in gastroesophageal malignancies and associated paraneoplastic syndromes - connection with inflammation. Clin. Biochem. 41, 804–811. Mackness, M., Mackness, B., 2015. Human paraoxonase-1 (PON1): Gene structure and expression, promiscuous activities and multiple physiological roles. Gene 567, 12–21. Malik, U.U., Siddiqui, I.A., Hashim, Z., Zarina, S., 2014. Measurement of serum paraoxonase activity and MDA concentrations in patients suffering with oral squamous cell carcinoma. Clin. Chim. Acta 430, 38–42. Michalak, S., Szubert, S., Moszynski, R., Sajdak, S., Szpurek, D., 2014. Serum arylesterase and paraoxonase activities in patients with ovarian tumors. Taiwan J. Obstet. Gynecol. 53, 490–493. Ozturk, E., Balat, O., Dikensoy, E., Ugur, M.G., Ozcan, C., Aydin, A., Erel, O., Kul, S., 2011. No association between serum paraoxonase, arylesterase activities, and hydatidiform mole. Int. J. Gynecol. Cancer 21, 149–151. Pierce, C.A., Software Review: Borenstein, M., Hedges, L.V., Higgins, J.P.T., Rothstein, H.R., 2006. Comprehensive meta-analysis (version 2.2.027) [computer software]. Englewood, NJ: Biostat. Organiz. Res. Meth. 2008 (11), 188–191. Ponce-Ruiz, N., Murillo-González, F.E., Rojas-García, A.E., Mackness, M., Bernal-Hernández, Y.Y., Barrón-Vivanco, B.S., González-Arias, C.A., Medina-Díaz, I.M., 2017. Transcriptional regulation of human paraoxonase 1 by nuclear receptors. Chem. Biol. Interact. 268, 77–84. Précourt, L.P., Amre, D., Denis, M.C., Lavoie, J.C., Delvin, E., Seidman, E., Levy, E., 2011. The three-gene paraoxonase family: physiologic roles, actions and regulation. Atherosclerosis 214, 20–36. Sehitogulları, A., Aslan, M., Sayır, F., Kahraman, A., Demir, H., 2014. Serum paraoxonase-1 enzyme activities and oxidative stress levels in patients with esophageal squamous cell carcinoma. Redox Rep. 19, 199–205. Vecka, M., Jáchymová, M., Vávrová, L., Kodydková, J., Macášek, J., Urbánek, M., Krechler, T., Slabý, A., Dušková, J., Muravská, A., Zák, A., 2012. Paraoxonase-1 (PON1) status in pancreatic cancer: relation to clinical parameters. Folia Biol (Praha) 58, 231–237. Wells, G.A.S.B., O’Connell, D., Peterson, J., Welch, V., Losos, M., Tugwell, P., 2009. The NewcastleOttawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses. Available from:. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.

Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.critrevonc.2018.04.005. References Afsar, C.U., Gunaldı, M., Okuturlar, Y., Gedikbası, A., Tiken, E.E., Kahraman, S., Karaca, F., Ercolak, V., Karabulut, M., 2015. Paraoxonase-1 and arylesterase activities in patients with colorectal cancer. Int. J. Clin. Exp. Med. 8, 21599–21604. Ahmed, N.S., Shafik, N.M., Elraheem, O.A., Abou-Elnoeman, S.E., 2015. Association of paraoxonase-1(Q192R and L55M) gene polymorphisms and activity with colorectal cancer and effect of surgical intervention. Asian Pac. J. Cancer Prev. 16, 803–809. Akçay, M.N., Polat, M.F., Yilmaz, I., Akçay, G., 2003a. Serum paraoxonase levels in pancreatic cancer. Hepatogastroenterology 50 (Suppl 2), ccxxv–ccxxvii. Akçay, M.N., Yilmaz, I., Polat, M.F., Akçay, G., 2003b. Serum paraoxonase levels in gastric cancer. Hepatogastroenterology 50 (Suppl. 2), cclxxiii–cclxxv. Antognelli, C., Del Buono, C., Ludovini, V., Gori, S., Talesa, V.N., Crinò, L., Barberini, F., Rulli, A., 2009. CYP17, GSTP1, PON1 and GLO1 gene polymorphisms as risk factors for breast cancer: an Italian case-control study. BMC Cancer 9, 115.

14