| Background:Squamous cell carcinoma of the head and neck is a common clinical malignant tumor,which mainly occurs on the mucous surface of the upper respiratory digestive tract,such as oral mucosa,paranasal sinus mucosa,nasopharyngeal mucosa,pharynx and laryngopharynx mucosa.Radiotherapy has been the previous treatment for advanced squamous cell carcinoma of the head and neck,but because of its limited efficacy,it has developed into concomitant radiotherapy and chemotherapy based on cisplatin(CCRT),which is now the standard scheme for the treatment of advanced squamous cell carcinoma of the head and neck.However,this method not only improves the clinical effect of patients,but also increases a variety of toxic effects,including ototoxicity.Sensory neurological hearing loss may be one of the common adverse events in the treatment of head and neck malignant tumors.Radiotherapy can cause sensorineural deafness in the treatment of head and neck tumors,while concomitant radiotherapy and chemotherapy based on cisplatin has a higher incidence of hearing loss.Sensorineural deafness caused by cisplatin generally starts from the acute stage of treatment,which is bilateral,irreversible and characterized by progressive high frequency loss.Sensorineural hearing loss may occur weeks,months,or even years after treatment,which has a serious impact on the quality of life of patients.As a clinician,it is necessary to assess the risk of hearing loss after treatment and inform the patient.However,due to the lack of corresponding theoretical basis,it is still unable to deal with patient counseling.Therefore,a clinical prediction model is constructed to predict the probability of sensorineural hearing loss after cisplatin radiotherapy and chemotherapy,so as to be used in clinical decision-making.It is a problem that needs to be solved urgently at present.Clinical prediction model,also known as clinical risk assessment,refers to the use of multi-factor model to predict the probability of a certain disease or a certain clinical outcome.The clinical prediction model includes diagnostic model and prognosis model.Diagnostic model refers to the prediction of the probability that the patient currently has a certain disease based on the patient's laboratory data,clinical symptoms and characteristics.Prognostic model refers to the probability of predicting disease recurrence,death,disability or a certain outcome in the future in the current state of disease.This study was based on a prognostic model of sensorineural hearing loss in patients with advanced head and neck squamous cell carcinoma treated with cisplatin combined with radiotherapy and chemotherapy.At present,some of the clinical models of sensorineural deafness have been found to be the dose of radiation to the cochlea and the dose of cisplatin.The baseline hearing level of patients with age is the main risk factor for hearing loss after radiotherapy and chemotherapy.However,because these models are only single-factor clinical correlation models,they have no predictive function for the occurrence of sensorineural deafness.Therefore,combined with various factors of patients,it is a problem to be solved in this study to construct a probabilistic clinical prediction model to predict hearing loss after cisplatin-based chemoradiotherapyThe clinical correlation model had no predictive function for the occurrence of sensorineural deafness.Therefore,combined with various factors of patients,it is a problem to be solved in this study to construct a probabilistic clinical prediction model to predict hearing loss after cisplatin-based chemoradiotherapyBy reviewing and analyzing the laboratory data,clinical characteristics and accompanying mental state of patients,we constructed a prediction model which has important clinical significance and can provide clinical decision-making guidance.Objective:To predict the probability of sensorineural hearing loss in patients with advanced head and neck squamous cell carcinoma treated withcisplatin-based chemoradiotherapy.Methods:The data of patients with advanced head and neck tumors diagnosed in Shandong Provincial Hospital from January 1,2008 to December 31,2018 and treated with cisplatin-based chemoradiotherapy were analyzed retrospectively.A total of 1923 patients were included.After exclusion,1710 patients were included in the study.The follow-up was completed on February 28,2019.Patient inclusion data include general characteristics:age,sex and body mass index(BMI);Laboratory data:d-dimer,platelet,serum total bilirubin count,neutrophil count,lymphocyte count,lipoprotein(a)and coagulation factor VIII;Concomitant diseases:hypertension,diabetes,cerebrovascular disease and cardiovascular disease;mental state after treatment:depression events,insomnia and anxiety events;and dose of cisplatin,whether there is noise exposure.The age data were divided into>60 years old and<60 years old.Among them,depression events,insomnia,anxiety events and noise exposure all refer to the events that occur after treatment to the occurrence of sensorineural hearing loss.Cerebrovascular disease refers to the history of cerebral vascular blockage before treatment,and cardiovascular disease refers to the history of ischemic heart disease or peripheral vascular disease before treatment.The laboratory data were tested in Shandong Provincial Hospital before treatment.The selected patients were randomly divided into training group(n=972)and verification group(n=738).The differences between groups of continuous variables with normal and skewed distribution were analyzed by single factor analysis of variance(one-way ANOVA)and Kruskal-Wallis test.Chi-square test is used to analyze classification variables.The risk factors of hearing loss were analyzed by univariate and multivariate logistic regression analysis combined with generalized estimation equation.In the stage of model development,according to the Akaike information criterion and the results of multivariate regression analysis,a nomogram was constructed to predict the probability of sensorineural hearing loss.The effective threshold was P<0.05.The purpose of this study was to evaluate the probability of sensorineural hearing loss after cisplatin-based chemoradiotherapy.At the same time,the data in the training group,from less to more to establish several different models,evaluate and compare their advantages and disadvantages,and verified by the verification group.According to the prediction model of the training group,finally,the verification group model is used to verify and calibrate it,and its repeatability and credibility are obtained.The performance of the model is evaluated by means of discrimination,verification and calibration.All the statistical analysis was carried out by using the third-party software EmpowerStats based on R language,and the results were tested by double tail test(P<0.05).Results:Through the pure tone hearing threshold test,the bone conduction average hearing loss 10dB and above showed the following outcome events:sensorineural hearing loss at high frequency 1,2 and 4kHz.1.Study population descriptionThe training group and the verification group included 972 and 738 patients,respectively.Training group and verification group in age,sex,platelet,lymphocyte count,lipoprotein(a),hypertension,diabetes,cerebrovascular disease,cardiovascular disease,cisplatin dose,noise exposure,depression events,There is a data similarity in the proportion of insomnia and anxiety events.That is,the comparison of the above 14 indicators between the two groups of data P>0.05.In addition,the outcome events and the incidence of sensorineural deafness were similar in the training group and the verification group,which were 35.4%and 37%,respectively(P=0.495).There were significant differences in BMI,D-dimer,serum total bilirubin,neutrophil count and lipoprotein(a)between the two groups(P<0.05).2.Univariate and multivariate regression analysis of population in training groupUnivariate analysis was carried out on the population in the training group.Age,body mass index,hypertension,diabetes,cardiovascular disease,D-dimer,serum total bilirubin,neutrophil count,lymphocytic count,cisplatin dose,noise exposure,depression events,Insomnia and anxiety events were the risk factors of SNHL after concurrent radiotherapy and chemotherapy with cisplatin.Sex,cerebrovascular disease,platelet,lipoprotein(a)and coagulation factor ? were not the risk factors for hearing loss.Multivariate regression analysis showed that other risk factors were consistent with univariate analysis except age,sex,body mass index and cardiovascular disease(P>0.053.Predictive performance of Model population in training GroupFive models were constructed by integrating different parameters in the training group.Each model contains different parameters and contains the ratio of parameters(Oddsratio,OR).AUC values show the distinguishing performance of the model.Model 1 included only age,hypertension and diabetes,and the performance was poor(AUC=0.795,95%confidence interval[CI],0.764-0.826).However,with the increase of the parameters in the model the AUC value is gradually improved.The AUC value of model 5 was 0.9498and 95%CI was 0.950-0.972,the sensitivity and specificity were 0.878 and 0.919,respectively.There was significant difference between the two groups before and after the establishment of all the models(P<0.05 or P<0.001).4.Verification of the prediction performance of the model for the population in the validation groupThere were 738 people in the validation group.For the five models that have been built by the training group,we verify them one by one in the validation group.It can be seen that each model of the validation group showed a better degree of discrimination,and the lowest AUC was 0.865(model 1),which was larger than that of the training group(0.795).The maximum value of AUC was 0.948(model 5),which was slightly smaller than that of the training group(0.961).However,AUC(0.858 VS 0.865 in model 3 and model 1 was lower than that in model 1,and there was no significant difference between model 3 and model 1(P=0.491).The results showed that D-dimer,platelet and serum total bilirubin did not significantly increase the differentiation of the model.The AUC(0.906)of model 4(0.909)was similar to that of model 2,and there was no significant difference(P=0.706),indicating that the increase of cisplatin dose and noise exposure did not significantly increase the differentiation of the model.However,the validation group model 5,like the training group model 5(AUC=0.961),still had the highest AUC,and the value reached 0.948,which,like the training group,was the most accurate model.Compared with model 4,there was significant statistical difference between model 5 and model 4(P<0.001).The results showed that the performance of model 5 was significantly improved.5.Risk factors of hearing loss in cisplatin-based chemoradiotherapy from nomogramBased on the previous multivariate regression analysis,we constructed a line map of risk factors,including age,hypertension,diabetes,D-dimer,serum total bilirubin,neutrophil count,lymphocytic count,cisplatin dose,noise exposure,depressive events,insomnia and anxiety events.The AUC curve value of this model is 0.961,95%CI,0.955-0.972.It shows excellent discrimination,specificity is 0.925 and sensitivity is 0.895,which shows good reliability.6.Internal validation of the modelWe use the data of the validation group to verify the data of the training group.The validation group model constructs three models,namely,the full parameter(full)model,the reduced parameter(stepwise)model and the sensitivity curve(mfp)model.It can be seen that the AUC of the three models were 0.978(95%IC,0.968-0.988),0.978(95%IC,0.968-0.988)and 0.977(95%IC,0.967-0.987),respectively.7.Calibration of Predictive ModelThe blue calibration curve shows that from 10%to 50%,the predicted value is lower than the actual observed value,from 50%to 85%shows that the predicted value is higher than the actual observed value,and 85%-100%shows that the predicted value is lower than the actual observed value.On the whole,the blue curve is closer to the red curve,which shows the excellent prediction ability of the model.Conclusion:In this model,the general demographic characteristics,laboratory data,concomitant disease,mental state and cisplatin dose were collected.The probability of hearing loss in patients with advanced head and neck squamous cell carcinoma treated with cisplatin can be predicted.The specific manifestation of the model can be applied to the consultation of hearing loss concerns of this kind of patients,and can be used for clinical decision-making.Background:Cisplatin is one of the important chemotherapeutic drugs in the treatment of a variety of cancers,which has been used in the treatment of malignant tumors in adults and children.It has a good therapeutic effect and is more popular in clinic.It can be used for concurrent chemoradiotherapy as a supplement to radiotherapy to improve the survival rate of patients.Concurrent chemoradiotherapy based on cisplatin is often used in the treatment of advanced head and neck malignant tumors.As a platinum drug,it not only brings good clinical efficacy,but also has serious side effects,such as nephrotoxicity,cardiotoxicity,gastrointestinal toxicity,neurotoxicity and ototoxicity.Normal hearing is essential to people's normal life,especially actors,teachers and other professionals.Ototoxicity,as one of the side effects of cisplatin,is mainly reflected in permanent,high-frequency sensorineural deafness,which is reported to be 30%-45%.Once ototoxicity occurs,it will bring greater psychological burden to patients and seriously affect the quality of life.Therefore,the development of drugs with ototoxicity has become particularly urgent.Zingerone is a compound extracted from ginger,which has a high concentration in ginger and can be prepared by thermal degradation of gingerol or zingerenol by reverse aldehyde reaction.Zingerone is a kind of phenolalkanone,which is a non-toxic,low stimulation and cheap compound chemical,which has a wide range of basic pharmacological effects,including anticancer,antioxidant,anti-inflammatory,anti-microbial activity and anti-sugar.A variety of experimental models have been proposed that zingeronecan resist tissue inflammatory damage and toxicity caused by cisplatin or radiation,inhibit oxidative stress caused by radiation,and play a role through its antioxidant stress,anti-inflammation and anti-apoptosis.However,whether zingeronehas a protective effect on cisplatin-induced ototoxicity has not been reported.This study hypothesized that zingeronecould antagonize the ototoxicity induced by cisplatin through antioxidant stress and anti-inflammation.Objectives:The auditory brainstem response audiometry(ABR)test,biochemical test and reverse transcriptase-polymerase chain reaction(reverse transcription-polymerase chain reaction,RT-PCR)were used to determine whether zingeronehas a protective effect on cisplatin-induced ototoxicity in rats.Methods:In this study,36 male Wistar rats were used to test the ABR.After the hearing level was determined,the rats were randomly divided into 6 groups.The blank group was fed with 5 ml/kg normal saline for 7 days,the cisplatin group was injected intraperitoneally with 7.5 mg/kg cisplatin on the 4th day of the experiment,and the zingerone50 group was fed with 50 mg/kg zingeronefor 7 days.Zingerone10+cisplatin group was fed with 10 mg/kg zingeronefor 7 days,and 7.5 mg/kg cisplatin was injected intraperitoneally on the 4th day.Zingerone30+cisplatin group was fed with zingerone30 mg/kg for 7 days,and 7.5 mg/kg cisplatin was injected intraperitoneally on the 4th day.Zingerone50+cisplatin group was fed with 50 mg/kg zingeronefor 7 days,and 7.5 mg/kg cisplatin was injected intraperitoneally on the 4th day.On the seventh day at the beginning of this study,the ABR test was repeated to redefine the hearing level after drug treatment.All rats were then killed and cochlear tissue was used for related biochemical tests,including dipropionaldehyde(Malondialdehyde,MDA),superoxide dismutase(superoxide dismutase,SOD)and glutathione peroxidase(Glutathione peroxidase,GPx).Inflammatory factor TNF-? was tested by RT-PCR.Results:In cisplatin group,the concentration of MDA was significantly higher than that of other groups,while the concentrations of SOD and GPx were significantly lower than those of other groups.After the use of zingerone,the concentration of MDA in the cochlea decreased significantly,which was proportional to the dosage of zingerone,while the SOD and GPx in the cochlea increased significantly,and the increasing proportion was proportional to the dosage of zingerone.ABR test showed that the hearing threshold of cisplatin group was significantly higher than that of other groups,and maintained the same trend at 8kHz,16kHz and 32kHz.After the use of zingerone,the hearing threshold of ABR decreased,and the degree of decrease was proportional to the dosage of zingerone.RT-PCR showed that the expression of TNF-a in cisplatin group was higher than that in other groups.After using zingerone,the expression of TNF-? decreased,which was proportional to the dosage of zingerone.Conclusion:Our data show that zingerone can protect sensorineural deafness and antagonize cisplatin ototoxicity in rats.It is mainly achieved by resisting oxidative stress and inflammatory response. |
PDF Full Text Request