Assessment of 8-isoprostane ( 8-isoPGF 2 α ) in Urine of Non-Small Cell Lung Cancer ( NSCLC ) Patients Undergoing Chemotherapy

Non small cell lung cancer (NSCLC) is difficult to treat, being relatively insensitive to chemotherapy (Villanueva , 2011). Traditional platinum based drug regimens such as cisplatin and carboplatin drugs act by crosslinking DNA and interfering with cell division. These drugs also result in reactive oxygen/nitrogen species (ROS/ RNS) generation (e.g. superoxide and hydroxyl radicals, peroxynitrile and H2O2) (Lieberthal et al., 1996; Baek et al., 2003) that enhance the action of the chemotherapy via ROS/RNS induced apoptosis mechanisms (Chirinoa & Pedraza-Chaverri, 2009). Chemotherapy induced apoptosis generally results in ROS/RNS generation in conjunction with cytochrome C, AIPF-1 and caspases-2, -3, and -9 being released into the cytosol, alteration of mitochondrial transmembrane potential and mitochondrial permeability transition (Kannan & Jain, 2000). However generation of ROS/RNS also gives rise to drug-induced toxicity as a result of damage to healthy tissue. The result is a wide range of adverse events, usually including nausea and vomiting, thrombocytopenia, fatigue, cachexia and nephrotoxicity (Fabbro et al., 2006), followed often by the inability to continue cycles of chemotherapy,


Introduction
Non small cell lung cancer (NSCLC) is difficult to treat, being relatively insensitive to chemotherapy (Villanueva , 2011).Traditional platinum based drug regimens such as cisplatin and carboplatin drugs act by crosslinking DNA and interfering with cell division.These drugs also result in reactive oxygen/nitrogen species (ROS/ RNS) generation (e.g.superoxide and hydroxyl radicals, peroxynitrile and H 2 O 2 ) (Lieberthal et al., 1996;Baek et al., 2003) that enhance the action of the chemotherapy via ROS/RNS induced apoptosis mechanisms (Chirinoa & Pedraza-Chaverri, 2009).Chemotherapy induced apoptosis generally results in ROS/RNS generation in conjunction with cytochrome C, AIPF-1 and caspases-2, -3, and -9 being released into the cytosol, alteration of mitochondrial transmembrane potential and mitochondrial permeability transition (Kannan & Jain, 2000).However generation of ROS/RNS also gives rise to drug-induced toxicity as a result of damage to healthy tissue.The result is a wide range of adverse events, usually including nausea and vomiting, thrombocytopenia, fatigue, cachexia and nephrotoxicity (Fabbro et al., 2006), followed often by the inability to continue cycles of chemotherapy,

RESEARCH COMMUNICATION
Assessment of 8-isoprostane (8-isoPGF 2α ) in Urine of Non-Small Cell Lung Cancer (NSCLC) Patients Undergoing Chemotherapy Nutjaree Pratheepawanit Johns, Jeffrey Roy Johns* resulting eventually in death.The level of oxidative stress induced by platinum based chemotherapy has not been studied in detail, and individual responses of human to oxidative stress induced by platinum-based therapy needs investigation in order to better inform dosing and regimen strategies, and affect on healthy tissue.
The F 2 -isoprotanes are stable compounds that are specific products of free-radical-catalyzed lipid peroxidation of arachidonic acid, independent of the cyclooxygenase enzyme.They are present in detectable quantities in normal biological tissues and fluids, are unaffected by lipid content of the diet, and show significant increase in animal models of antioxidant injury (Roberts & Morrow, 2000).As F 2 -isoprotanes levels are modulated by antioxidant status they are ideal markers of oxidative stress.These compounds are formed in situ esterified to phospholipids in vivo and are then cleaved and released into the circulation before excretion in the urine as free isoprostanes (Morrow & Roberts, 1994).
The 8-isoprostane (8-epiPGF 2α , 8-isoPGF 2α , iPF2α-III, 15-F 2t -IsoP) is a reliable marker and recently considered a gold standard for lipid peroxidation (Moore & Roberts, 1998;Kadiiska et al., 2005).Levels of 8-isoprostane in excess of normal levels have also been reported in a large number of pathologies, for risk factors such as obesity, gender, smoking, alcohol abuse, and for toxic xenobiotics as previously reviewed (Montuschi et al. 2004;Montuschi et al., 2007;Basu, 2008).However there are very few reports of isoprostane levels in advanced cancer patients, or in patients undergoing chemotherapy.Oxidative stress is however expected and has been observed in patients with cancer.
As with other cytotoxic xenobiotics, chemotherapy agents may cause lipid peroxidation and increase in oxidative stress.8-isoprostane has been determined in cisplatin-induced renal dysfunction in LLC-PK1 renal epithelial cells (Salahudeen et al., 1998) where a doserelated increase in total (free plus esterified) 8-isoprostane was observed.
This study measured 8-isoprostane levels in urine of patients undergoing chemotherapy for advanced NSCLC as a marker of chemotherapy-induced oxidative stress from ROS/RNS over three cycles (courses) of platinumbased treatment.It also considered creatinine clearance of patient undergoing chemotherapy, and correction of 8-isoprostane levels by creatinine clearance versus urine volume methods.

Study design
Urine and blood samples were collected from patients participating in a randomized, double-blind, placebo controlled trial conducted in advanced cancer patients receiving treatment from a university teaching hospital in Northeast Thailand.The study was approved by the Khon Kaen University Ethics Committee and conducted in accordance with the International Conference on Harmonization on Good Clinical Practice requirements.All patients provided written informed consent.

Patients
Patients with histologically proven advanced NSCLC were recruited.Additional inclusion criteria included age 18 to 70 years, Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤ 2, white blood cell count ≥ 3,000 cell/mm3, serum creatinine ≤ 1.5 mg/ dL, bilirubin ≤ 2 mg/dL, AST ≤ 2.5 times upper limit of normal (ULN) for subjects without metastases or ≤ 4 times ULN for those with liver metastasis and New York Heart Association grade ≤ 2. Patients who had received prior chemotherapy or biotherapy, radiotherapy or surgery within 1 month preceding recruitment, or had more than one type of cancer or brain metastasis were excluded from the trial.Additional exclusion criteria included patients with moderate neuropathy (CTCAE grade ≥ 2), an active infection, or uncontrolled complications.
Most patients received cisplatin (75 mg/m2 intravenously) on day 1 of treatment, and etoposide (50-100 mg/m 2 intravenously) on days 1 to 3, then continuing cycles about every 3 to 4 weeks.Some patients alternatively received paclitaxel and carboplatin regimen.

Data collection
Baseline evaluation included assessment of ECOG PS, standard hematology, chemistry, electrolytes, urinalysis, and physical examination.Collection of blood, saliva, and overnight urine was conducted on the 1 st (prior to chemotherapy), 2 nd and 3 rd chemotherapy cycles (approximately one month apart).Serum and urine creatinine levels were determined by the hospital laboratory as part of normal patient assessment.Overnight urination volumes (19:00-07:00) were carefully measured and recorded, and the volumes combined for analysis.Urine samples were stored at -20 °C until analysis and analyzed within 6 months.Although plasma samples cannot be stored at this temperature because auto-oxidation can occur, this is not problem with urine because it has only trace quantities of lipid, and thus ex vivo formation of F 2 -isoprostanes does not apply in the measurement of urine (Morrow et al., 1990;Morrow et al., 1996).Levels of F 2 -isoprostanes are not increased by incubation of urine at 37 °C for 1 week or during storage of urine for 6 months at -20 °C (Morrow et al., 1990).

Determination of 8-isoprostane in urine samples
Urine 8-isoprostane was measured using an ACE Immunoassay 516315 test kit (Cayman Chemical Inc, MI USA) as per the manufacturer's instructions.Organic free ultrapure water filtered with a 40 um filter was used for all dilutions (Elga DV25 Purewater OptionQ system).Absorbances of the developed plates were read with an Anthos ELISA reader (Labtec Instruments, Sabzburg, Austria) with ADAP 1.6 software at 420 nm after development of the plates, i.e. after the B 0 wells were more than 0.3 A.U. (blank subtracted), 60-90 minutes.Urine samples from the patient's combined overnight urine voids were centrifuged briefly if necessary to remove sediment and diluted 50 fold to guarantee measurements were within the optimum 20-80% B/B 0 range.50 uL was added directly to sample wells of the 96 well plates.The majority of samples were analyzed in duplicate.Samples from different cycles were distributed across plates to reduce and bias from variations in individual plates.Detection limit for the Cayman kit is claimed to be 2.7 pg/ml sample (80% B/B 0 ).
Concentrations were expressed as pg/mL urine, and were also normalized by multiplying by the volume of overnight urine collected to give µg, as this better represents the production of 8-isoprostane under conditions of chemotherapy.The intra-and interassay variations for the enzyme immunoassay were 6 and 10%, respectively.Statistical data analysis was performed with SPSS version 11.

Baseline characteristics
A total of 30 patients were recruited with average age 55 years, but 5 patients withdrew from the study.All 25 remaining patients were advanced non-small cell lung cancer, received platinum-based regimen and reported ECOG performance score of 1.

Urine volumes and creatinine levels
Total volumes of urine collected, number of voids, and urine creatinine levels are shown in Table 1.No significant difference was seen between cycles in average total overnight urine volume or number of urinations, but greater variation in urine volume was seen at cycle 2 and 3. Creatinine levels were significantly different only between cycle 1 and cycle 2 (p=0.002).

Urine 8-isoprostane levels
Average values for urine 8-isoprostane (corrected and uncorrected for urine volume and creatinine) are shown in Table 2. Differences between the mean values of each cycle were analyzed by Wilcoxon signed-rank test.
The average 8-isoprostane levels in urine significantly increased up to 12 fold on chemotherapy treatment, from 532 ± 587 pg/mL at cycle one, 6181 ± 4334 at cycle 2, and 5511 ± 2055 at cycle 3.The uncorrected urine 8-isoprostane levels were significantly different between cycles 1 and 2 (p<0.001), and cycles 1 and 3 (p=0.001),but not between cycles 2 and 3 (p=0.91).Figure . 1 shows the average 8-isoprostane levels (and SD) for the three cycles for the individual subjects.A large variation was observed between individuals.

Discussion
Although it is probably essential to clean up tissue or plasma samples by C18 solid phase extraction cartridges or affinity columns (e.g.Cayman affinity column 416358), particularly if good correlation with GC-MS results is desired (Davies et al, 2006;Yan et al., 2007;Zhang & Saku, 2007), there would seem to be no specific requirement for urine samples with immunoassay methods.Immunoassays have been generally demonstrated to work very well in buffer systems that do not contain large amounts of biological substances.Indeed some ELISA kit manufacturers suggest direct analysis of urine samples (Bioxoytech, Urinary 8-Epi-Prostoglandin F 2α Enzyme Immunoassay for Urinary Isoprostane, Catalog Number 21048, OxisResearch, Portland, OR, USA).
To verify this for the Cayman kit, a simple check was performed.A previously analyzed urine sample with very low isoprostane level (< 3 pg/mL) was selected.To this sample, test kit standards (0-500 pg/ml of 8-isoprostane) were added, assayed in duplicate and compared to the standard curve produced from the same standards in buffer (also in duplicate).Differences less than 5% in concentration were seen, indicating that the urine matrix causes minimal interfere with the AChE competitive enzyme reaction or quantification of the AChE with the Ellman's reagent (5, 5'-dithiobis-(2-nitrobenzoic acid or DTNB) of the Cayman kit.
Study has shown that there is no diurnal variation in the urinary levels of 8-iso-PGF 2α during the day or any statistically significant difference between the 8-iso-PGF 2α levels at any time of the day or in morning urine samples compared to 24-h urine samples.Thus all urine samples collected at any time of the day, including morning urine samples can thus be used to obtain reliable and adequate values of the amount of the 8-isoprostane excretion in urine in healthy individuals (Helmersson & Basu, 1999).
Levels of urine 8-isoprostane at chemotherapy cycle 1 (average 2.2 ± 1.6 ng/mg creatinine) were slightly increased above the normal expected range for healthy people (1.6 ± 0.6 ng/mg creatinine) for almost all the cancer patients.The levels increased dramatically (on average more than 12 fold) at cycle 2, and 3 however, assumable due to oxidative stress and lipid peroxidation caused by the cytotoxic agents used (cisplatin).This is unsurprising considering the extreme cytotoxicity of the chemotherapy drugs.The increase in levels seen (10-200 times) are comparable to that observe with acute poisoning of rats by high doses of CCl4 or diquat (Kadiiska et al., 2005).
All of the patients undergoing chemotherapy showed much lower than normal levels of urine creatinine, and some extremely abnormally low levels (2.2 -9.2 mg/ dL urine), although their serum creatinine was within normal range (Table 1).By comparison, only 7.7% of the healthy US population has urine creatinine values less than the WHO guideline range of 30 -300 mg/dL (Barr et al., 2005), whereas a majority of the patients in this study had levels below this by cycle 2 and 3 of chemotherapy.Adjusting 8-isoprostane concentrations for urine creatinine in these cases produces incredibly high values that are not useful for comparison, even though the unadjusted levels (pg/mL) are within the expected range of the data set.
The question of whether to correct urine 8-isoprostane levels for effects of urine concentration (or volume) arises.Traditionally, levels are reported as pg or ng of 8-isoprostane/mg creatinine.Correcting for concentration with creatinine would seem to be sensible, for example, with drugs that are metabolized and whose appearance in the urine will change with time.In these cases, correction for urine concentration (volume) is therefore essential, particularly for spot samples.Urine 8-isoprostane however is reported to be continuously produced at a constant rate, depending on the level of oxidative stress, and therefore correction may not be necessary.As previously discussed, Liang found that urine concentration did not appear to influence the measurement of 8-isoprostane (Liang et al., 2003).Furthermore there appears to be no diurnal variation in the urinary levels of 8-isoprostane (Helmersson & Basu, 1999).Ideally, the total quantity of the isoprostane of interest produced overnight or per 24 hour period, or concentration per milliliter, uncorrected for volume or creatinine would seem a more useful value.Patients with severe disease, for example cancer, may have renal impairment (renal toxicity of chemotherapy) or other factors that may affect creatinine clearance, but not 8-isoprostane clearance.Furthermore, there was no correlation (Pearson R 2 < 0.01) between urine creatinine and overnight urine volume, indicating that for chemotherapy patients, using creatinine to correct for concentration differences due to urine volume, may not be valid.
Grade et al. ( 2004) also showed that the uncertainty associated with creatinine standardization (19-35%) was higher than the uncertainty related to volume standardization (up to 10%) for 24 hours urine samples, even with average bias of 4% from missed volumes in population studies.Thus correction will contribute to variation and errors in reported levels.
The level of oxidative stress, and therefore 8-isoprostane production expected during chemotherapy would almost certainly be more than that reported for typical pathologies, and might near that reported for severe induced lipo-toxicity e.g. with CCl4 or diquat, i.e. 10-20 times normal healthy levels (Kadiiska et al., 2005 was elevated more significantly (P<0.01) after the chemotherapy and was more evident in higher stage than lower stage patients.
Although several other researchers have reported increased oxidative stress in patients with lung cancer prior to chemotherapy (Zieba et al., 2001;Dalaveris et al., 2009;de Castro et al., 2006), we did not observe increase levels at the beginning of chemotherapy compared with levels previously reported for healthy subjects, although this comparison must be considered carefully due to differences in measurement method as previously discussed.
The role of 8-isoprostane in lung cancer biology and lung pathology is still not clearly understood (Zieba et al., 2001), but there would seem to be important implications for isoprostane levels in human pulmonary pathophysiology, particularly their effect on smooth muscle and their role in mediating free radicals and reactive oxygen species (Janssen, 2001;Janssen, 2008).
In conclusions, the levels of 8-isoprostane as assessed by ELISA in overnight urine collected from lung cancer patients undergoing chemotherapy increased more than 6 to 12 fold from treatment cycle 1 to cycle 2 and 3, indicating significant production of ROS/RNS by chemotherapy drugs, and the resulting lipid peroxidation.Urine creatinine levels were shown to be abnormal for these patients, suggesting that uncorrected 8-isoprostane (pg/mL) or correction by urine volume (µg) is a more suitable measure than correction by urine creatinine (pg 8-isoprostane/mg creatinine).
The effect of cancer chemotherapy on oxidative stress in humans, and 8-isoprostane production, creatinine clearance and urine production has not been previously extensively studied, and future investigation is still warranted.
Further studies should also be conducted to determine whether cancer chemotherapy patients could benefit from antioxidant adjuvant treatment to reduce free radical induced oxidative stress and the associated cellular damage, without compromising the effectiveness of the chemotherapy.