| The first three causes of malignant pleural effusion are lung cancer,breast cancer and malignant lymphoma. Once malignant pleural effusion hasdeveloped, the average survival time of the patient is often less than6months.Malignant ascites is most common in gastrointestinal and gynecologicalcancers, which indicates that the median survival is less than20weeks,particularly in the patients with malignant ascites caused by gastrointestinalcancers, who have the worst prognosis and whose survival times are only12-20weeks. Pericardial effusion is generally observed in dying patients, andits prognosis is worse. The treatment of malignant effusions is often a clinicalchallenge. Currently, the conventional treatments of malignant effusions aremainly composed of diuresis, salt restriction, serous cavity paracentesis,intracavitary chemotherapy, biological response modifiers, traditional chinesemedicine, thermotherapy, etc. However, the effects of the above-mentionedtreatments are all unsatisfactory. After therapy with these methods, there is nosignificant decrease of effusions while it often relapses fast, and almost all ofthe treatments have toxic side effects of various degrees.Studies have shownthat elevated levels of VEGF (vascular endothelial growth factor), tumorangiogenesis, and increased vascular permeability after tumor invading ormetastasizing to the pleuroperitoneum are important formation mechanisms ofserous cavity effusions. The VEGF family includes six types: VEGF-A, B, C,D, E, and PIGF (placental growth factor). These factors are mainly secreted bymacrophages and tumor cells. VEGF-A is the most important regulatory factorin the tumor angiogenesis, and VEGF-C and VEGF-D are the most importantregulatory factors in the tumor lymphangiogenesis. The use of angiogenesisinhibitors, particularly VEGF targeted drugs has become a hot topic of thebasic and clinical researches.Currently, the study on the VEGF in serous cavity effusions is mainly focused on VEGF-A representing the angiogenesis,and there are few studies on the lymphangiogenesis representative factorsVEGF-C and VEGF-D.The domestic first class new drug recombinant humanendostatin (rh-endostatin, endostar) is a broad spectrum humanizedanti-angiogenic drug which targets vascular endothelial cells but not tumorcells and can down-regulate the protecting effect of a variety of proangiogenicfactors on the vascular endothelium, thus playing direct and indirectantiangiogenic effects. In recent years, there are a number of scholars hasexplored the treatment effect of endostar used alone or combined withchemotherapy on malignant serous cavity effusions. They found it efficientand low toxic, not only being able to produce a strong effusions controllingeffect with a high tolerance, but also being able to significantly improve thequality of life of patients. However, there are a lot of work to do in thefollowing areas: the efficacy evaluation system of angiogenesis inhibitors suchas endostar, and exploring new molecular markers to predict and evaluate theefficacy of endostar. How to improve and use molecular markers to guide theclinical diagnosis, evaluate the prognosis, predict the efficacy and optimize thetreatment of malignant serous cavity effusions is the direction of futureworks.In this study, we detected the contents and expression levels ofVEGF-A,C,D in the supernatants and cells from the malignant serous cavityeffusions caused by lung cancer and gastric cancer, to explore the clinicalsignificance of the three factors in the diagnosis and prognosis of malignantserous cavity effusions and in the prediction of efficacy of endostar on theeffusions. We established a stably expressing luciferase gastric NUGC-4-luccell line, and on the basis of this cell line we established gastric cancer ascitestumor model in nude mice for the first time. By combining dynamicmonitoring changes of gastric cancer ascites tumor cells by in vivobioluminescence imaging with detecting the peritoneal nodules MVD(microvessel density), LVD (1ymphatic vessel density), we comprehensivelyevaluated the value of the molecular markers VEGF-A, VEGF-C and VEGF-Din the targeted therapy of malignant serous cavity effusions. Part1Diagnostic and prognostic values of vascular endothelial growthfactor (VEGF)-A,-C and-D in malignant effusion of lung andgastric cancer patientsObjective: To evaluate VEGF-A,C,D the diagnosis, clinical prognosticvalue in malignant effusionMethods: Enzyme-linked immunosorbent assay(ELISA) was used todetect VEGF-A,C,Dprotein content from patients with effusion of the serum79cases (30lung cancer,21gastric cancer,28benign effusions), thesupernatant fluid96cases (38lung cancer,30gastric cancer,28benigneffusion); Immunocytochemistry (ICC) detected VEGF-A,C,D proteinexpression from71cases of effusion cell level (34lung cancer,17gastriccancer,20benign effusions), by SPSS17.0to analyze the diagnosis, prognosis,clinicopathological factors and correlation of clinical significance.Results:1VEGF-A1.1The level of sVEGF-A proteins was not statistically different betweencancer and benign patients (p>0.05). Cancer patients had higher pVEGF-Alevel than those with benign effusion (p <0.05). The upregulated level ofpVEGF-A was not associated with tumor histological types (p>0.05).pVEGF-A level was similar to the corresponding sVEGF-A level in patientswith benign effusions. Nevertheless, pVEGF-A level in malignant effusionwas significantly higher than its corresponding sVEGF-A level (p<0.05).We detected VEGF-A,-C, and-D expressions in20cytological smearsfrom benign effusions and the data showed that there was no positive stainingof anti-VEGF-A,-C, and-D antibodies in these20benign cases. VEGF-Aexpression rate was52.94%and mainly expressed in the cytoplasm ofpositively expressed tumor cells.1.2The data showed that pVEGF-A and cVEGF-A were negativelyassociated with age and pVEGF-A was positively associated with malignantand bloody effusion, pleural or peritoneal metastasis (p<0.05).1.3With the median levels as the cut-off values, patients were divided into high and low pleural effusion parameter groups. One-year survival timeanalysis showed a relatively shorter survival time for patients with pVEGF-Alevels>406.19pg/ml as compared with those presenting with pVEGF-a levels≤406.19pg/ml, although this effect was only of marginal statisticalsignificance (p=0.066). sVEGF-A with the median levels as the cut-off, didnot exhibited statistical significance in predicting survival for patients withmalignant effusions (p>0.05).2VEGF-C, VEGF-D2.1Levels of sVEGF-C， sVEGF-D proteins were not statisticallydifferent between cancer and benign patients (p>0.05). Levels of pVEGF-C，pVEGF-D proteins were not statistically different between cancer and benignpatients (p>0.05).cVEGF-C，cVEGF-D expression rates were70.58%and82.35, respectively, in51patients with lung or gastric cancer. These proteinswere mainly expressed in the cytoplasm of positively expressed tumor cells.Regarding their expression levels of in tumor patients, the expression rate ofcVEGF-D was significantly higher than that of cVEGF-A and cVEGF-C(p<0.05). However, there was no statistically significant difference incVEGF-A,-C and-D expression between lung and gastric cancers (p>0.05).2.2The data showed that pVEGF-C was negatively associated with age.pVEGF-D expression was positively associated with age and negativelyassociated with malignant effusion and pleural metastasis (p<0.05). However,there was no association of sVEGF-C， sVEGF-D proteins with theclinicopathological factors (p>0.05).2.3sVEGF-C,sVEGF-D and pVEGF-C,pVEGF-D with the median levelsas the cut-off, did not exhibited statistical significance in predicting survivalfor patients with malignant effusions (p>0.05).Conclusions:1The supernatant fluid VEGF-A was significantly different betweenbenign and malignant effusion, suggesting that VEGF-A may be used as atumor marker in the clinical setting to discriminate benign from malignanteffusions.VEGF-A levels reached marginal statistical significance as a potential predictor of poor clinical outcome. Given the fact that VEGF-Aexpression might have a survival impact, attention should be paid to thepatients with high VEGF-A expression.2The supernatant fluid and cellular level of VEGF-A were negativelycorrelated with age, probably reflecting with age growing, malignant effusionsand level of VEGF-A are reduced. The supernatant fluid levels of VEGF-A inthe patients with bloody pleural effusion and only pleural metastasis are higherthan that in patients with non-bloody pleural effusion and not only pleuralmetastasis.3The supernatant fluid VEGF-C and VEGF-D had no significantdifference in the clinical diagnosis of benign and malignant, but highlyexpressed at the cellular level. It needs expand samples of benign effusions inclinical practice to further study its diagnostic value after disease stratification.4Serum VEGF-A,C,D levels had no obvious clinical significance in thediagnosis and prognosis of serous cavity effusions, and they were also notsignificantly associated with the clinicalpathological factors. It is suggestedthe serological markers’ diagnosis and prognostic values for serous cavityeffusions may be not as good as that of the local effusion markers.Part2The predictive value of VEGF-A, C, D for the clinical efficacy ofendostar on malignant effusionsObjective: To evaluate the predictive value of VEGF-A,C,D for theclinical efficacy of endostar on malignant effusions.Methods: ELISA was used to detect VEGF-A,C,D protein contents of20pairs of patients with malignant effusions reciving intracavitary endostarbefore and after treatments. The values of the three factors in pridicting theefficacy of endostar were analyzed by SPSS17.0.Results:1Between the endostar monotherapy group, endostar plus cisplatin groupand endostar monotherapy combined with systemic anti-tumor therapy group,there were no significant differences in the objective response rate (RR: CR+PR), clinical benefit rate (DCR:CR+PR+NC) or local progression (PD)(p> 0.05).2The relationships between supernatant fluid VEGF-A,C,D proteincontents with efficacy: Neither the differences between before and aftertreatment （d0-d7） nor the difference ratios before and after treatment（d0-d7/d0）of the three factors reached statistical difference in RR, NC, PDgroups (p>0.05), with the difference ratio of the supernatant VEGF-A(d1-d7/d1) between groups result showed p=0.085.3The relationships between supernatant fluid VEGF-A,C,D proteincontents with treatment: The difference ratios before and after treatment（d0-d7/d0）of supernatant fluid VEGF-A,C,D had no significant differencebetween the three endostar groups (p>0.05).Conclusions:1Neither the differences between before and after treatment（d0-d7） northe difference ratios before and after treatment （d0-d7/d0）of supernatant fluidVEGF-A,C,D had no statistical difference in RR, NC, PD groups, suggestingthe three factors can’t pridict the efficacy of endostar.2The difference ratios before and after treatment (d0-d7/d0) ofVEGF-A,C,D had no statistical difference in different endostar groups,suggesting the three factors can’t guide the clinical treatment.Part3The research of VEGF-A, C, D in gastric cancer ascites tumor modeltreated with endostar given by different waysObjective: To comprehensively evaluate the value of the molecularmarkers VEGF-A,C,D in the gastric cancer ascites model by combiningmonitoring the cancer cells with detecting the vascular and lymphaticendothelial markers.Methods: Cultured NUGC-4-luc cells (RPMI1640medium+10%FBS+1%P/S), suspended a total of5×106NUGC-4-luc cells in50ml PBS. Thesuspension was pipetted with micropipettor and injected into abdominalcavities of28female nude mice. These intraperitoneally inoculated miceunderwent in vivo bioluminescence imaging weekly. One week afterNUGC-4-luc cells injection, mice were randomly divided into four groups of7 mice each: Group1: cisplatin (1mg/kg) d1-3+endostar (8mg/kg) d4-7;Group2: endostar (8mg/kg) d1-4+cisplatin (1mg/kg) d5-7; Group3: endostar(8mg/kg)+cisplatin (1mg/kg) d1,4,7; Group4(control group):50μl ofnormal saline d1-7. All of the treatments were injected intraperitoneally. Everygroup was treated two consecutive weeks, underwent imaging weekly. Oneweek after the final administration, mice were anesthetized and sacrificed.Then the supernatant fluid of ascites was taken and the contents ofVEGF-A,C,D proteins in it were detected by ELISA. VEGF-A,C,D as well asMVD and LVD in the peritoneal nodules were detected by IHC methods afterthe nodules had been embeded into wax blocks. All bioluminescent data werefirst collected and processed using IVIS and then analyzed by SPSS17.0toexplore the differences of imagings, protein expressions and contents ofVEGF-A,C,D, and LVD and MVD between the four treatment groups.Results:1Ascites tumor model5×106NUGC-4-luc cells were injected into abdominal cavities. One weekafter NUGC-4-luc cells injection, mice were randomly divided into fourgroups by the number of cancer cell photons, with no statistical differences inthe four groups (p>0.05). After initiating treatment an accidental deathsoccured in the first18days in Group2.On every other day there were nosignificant differences in the nude mice weights between the four treatmentgroups (p>0.05).After mice were sacrificed,we found bloody ascites wereobvious and peritoneal nodules were small,numerous and widely.2Drug efficacy2.1Volume of ascites in four groups(ml):Group4（4.87±0.45）>Group1（3.1±0.53）>Group2（2.0±0.08）>Group3（1.8±0.16）.2.2The number of peritoneal nodules: Group4（33.75±2.5）> Group2（21.66±5.77）> Group3（18.75±2.5）> Group1（8.75±4.78）.2.3The counts of cancer cell photonsBy using in vivo imaging processing software, tumor growth curves wereplotted according to the number of photons per measurement. On day1and day8, there were no significant differences between the four groups (p>0.05).On day15and day21, group1,2and3were all better than the control group(p<0.01). On day15(after treatment of two weeks), endostar combinationwith cisplatin group (Group3) and cisplatin sequential endostar group (group1)were better than endostar sequential cisplatin group (group2)(p<0.05),however there was no significant difference between the former two groups(p>0.05). On day21(after treatment of two weeks, one week after withdrawn),endostar combination with cisplatin group (Group3) was still better thanendostar sequential cisplatin group (group2)(p<0.05).2.4MVD,LVDThe lumens of blood vessels and lymphatics were determined by CD34and D2-40stained brown to black. There were differences in MVD betweenthe four groups (p<0.01), pairwise comparisons showing group4> group1>group2(p<0.01), and there was no difference between group2and group3(p>0.05). However, there were no statistical differences in LVD numberbetween the four groups (p>0.05).3VEGF-A,C,D3.1There were differences in supernatant VEGF-A between the fourgroups (p<0.05), pairwise comparisons showing there was difference betweeneach two groups of the four groups with the result ofgroup4>group1>group2>group3(p<0.05). The ascites supernatant levels ofVEGF-C and VEGF-D proteins had no significant difference between the fourgroups (p>0.05).3.2VEGF-A,C,D proteins were all cytoplasmic staining. There werestatistical differences in VEGF-A between the four groups (p<0.05), pairwisecomparisons showing Group4>Group2>Group1（p<0.05）， and there was nodifference between group1and group3(p>0.05). However, there were nostatistical differences in VEGF-C and VEGF-D expressions between the fourgroups (p>0.05).4VEGF-A protein content was highly correlated with the volume ofascites (r=0.994, p<0.01). VEGF-A protein expression was highly correlated with the number of peritoneal nodules (r=0.994, p<0.01), but was notcorrelated with the MVD and the counts of cancer cell photons (p>0.05).Conclusions:1On the basis of a stably expressing luciferase gastric NUGC-4-luc celllines, we established gastric cancer ascites tumor model in nude mice for thefirst rime. We provided a stable, reliable, intuitive, convenient, and sensitiveanimal model for exploring the growth and metastasis mechanisms of thegastric cancer peritoneal metastatic tumor and the research and developmentof anticancer drugs by using in vivo imaging techniques.2In this study, the groups of endostar combined and sequentially usedwith cisplatin were all well tolerated, and treatment should be at least twoweeks before observing differences. The present study combined in vivoimaging with the MVD for the first time, and the results suggested:combination group was superior to sequential groups, the former’s killingcapability on cancer cells and inhibiting effect on vascular endothelium wereall the greatest; cisplatin sequential endostar was superior to endostarsequential cisplatin in killing cancer cells, while endostar sequential cisplatinwas superior to cisplatin sequential endostar in inhibiting peritoneal vascularendothelium. It was suggested that we can use molecular markers to forecasttarget changes and use functional imaging to monitor the number of cancercells, to guide the rational choice of the targeted and cytotoxic drugs and theadministration timing.3VEGF levels were highly correlated with the volume of ascites and thenumber of peritoneal nodules, suggesting that the predictive value of VEGFfor the efficacy of malignant effusion may be not only the volume of ascitesbut the size of peritoneal nodules. VEGF-A expressions were not correlatedwith the MVD and the counts of cancer cell photons. These suggested that itcan’t be used to monitor the activity of cancer cells and lymphatic vascularstatus, and it can’t be used to guide the choice of drugs intraperitoneallyinjected as well.4No differences in ascites and peritoneal nodules VEGF-C and VEGF-D proteins and LVD between the four treatment groups were found, so they stillcan’t serve as predictors of efficacy. |
PDF Full Text Request