In this study, data from 270 patients with breast cancer or lymphoma in the training cohort were retrospectively analyzed. The results showed that BMI ≥ 25, history of diabetes, BNP ≥ 52pg/ml, LDL ≥ 3.37mmol/L, pathological stage, and cumulative dose of anthracycline ≥ 550mg/㎡ were identified as independent risk factors for CTRCD. We found several studies to demonstrate the reliability of the risk factors in the results. (1) A higher BMI was found to be associated with an increased[9] risk of cardiotoxicity in breast cancer patients in one study, which was confirmed in our study. (2) LDL (p = 0.004) is also an independent risk factor for CTRCD. Treatment of hyperlipidemia and hypercholesterolemia has been shown to significantly reduce mortality and cardiac events in high-risk patients[10, 11]. (3) the QTc extension means that the heart after very abnormal, long QT is the performance of the heart caused by chemotherapy of acute and subacute toxicity, one[12] of the cardiac toxicity caused by chemotherapy can be characterized by the occurrence of arrhythmia and room after the change of the pole, so we take it as one of the factors determining CTRCD, QTc have statistical significance in the training cohort in analysis, But many factor analysis results show that the QTc period (p = 0.121) between the statistical basis of also cannot be used to predict cardiac toxicity caused by chemotherapy, the result may be related to QTc and cardiac toxicity determination under the influence of other risk factors. One study showed that the QTc interval was longer[13] in individuals with higher BMI than in those with normal BMI, which was considered to be related to cardiovascular lipid deposition, microangiopathy, and myocardial interstitial fibrosis caused by high BMI. At the same time, the mechanism of cardiotoxicity caused by QTc prolongation has not been fully eluc[14, 15]idated. (4) BNP is secre[16]ted by the ventricle in response to end-diastolic pressure and volume. During the treatment of cancer with anthracyclines, plasma BNP levels increased[17], and we also obtained the consistent conclusion that BNP was significantly correlated with CTRCE (P < 0.001). However, studies have been contradictory about the correlation between BNP concentration and the standard method of estimating LVEF. The value of BNP does not lie in the prediction of LVEF level, but in its prognostic and predictive value for the occurrence of CHF. BNP played an important role in our nomogram. (5) cTnI as an early diagnosis of cardiac biomarkers of ability is not enough, a systematic review shows that cTnI mainly as to accept radiation, fluorine pyrimidine drugs, platinum antitumor drugs or vascular endothelial growth factor receptor inhibitor markers[18] in patients with acute coronary syndrome, Because our study end points did not include coronary artery injury that was characterized by an acute coronary syndrome, they are not inconsistent with our results. In addition, evaluation of serum troponin can be recommended for suspected immune-checkpoint inhibitor-associated myocarditis. Anthracene ring and HER2 inhibitors (6) the damage to the heart muscle has both direct and indirect effects, this study found that the cumulative dose of 550 mg / ㎡ or anthracene ring are independent risk factors for CTRCD (p < 0.05), consider for anthracene ring CTRCE is caused by a variety of factors, such as: oxidative stress, in the presence of iron to produce reactive oxygen species, It leads to lipid peroxidation of cell membrane, which leads to myocardial cell damage[19]. Inhibition of topoisomerase IIβ, which is active in quiescent nonproliferative cardiomyocytes, leads to activation of cell death pathways and inhibition[20, 21] of mitochondrial biogenesis. Anthracyclines form complexes with intracellular iron, which in turn produce reactive oxygen species that damage DNA, proteins, and lipids, including mitochondrial membranes, and accelerate myocyte death[22]. At the same time, trastuzumab promotes the destructive effect of oxidative stress, leading to DNA fragmentation and induction[23] of the mitochondrial apoptotic pathway. Cardiomyocyte depletion may be the most important mechanism[24] of trastuzumab-associated heart failure over time. (7) The combination of HER-2 targeting classes was not included as an independent risk factor in the final conclusion, which may be related to the inclusion criteria of the study, and the inclusion of lymphoma patients in the study population, resulting in "false negative" results. The national comprehensive cancer network adjustment guide content, from anthracycline-based drugs limitations the preferred scheme[25] of HER2 positive breast cancer, some scholars think that the limitations of its ehrs − 2 positive breast cancer using the standard[26] anthracycline-based drug treatment is a new treatment. Therefore, the comprehensive prediction model should be used with caution when anthracyclines are combined with HER-2 targeted drugs in the treatment of cancers with definite cardiotoxicity. (8) Our study also found no significant association between age (p = 0.719), hypertension (p = 0.181) and CTRCD. Hypertension affects many different systems in the body, and its deleterious effects on the macrovascular system make it a risk factor for coronary heart disease. The main criteria for CTRCD are LVEF change, heart failure signs and myocardial injury markers, or arrhythmia is a secondary criterion. Therefore, myocardial injury is an important mechanism for CTRCD. Hypertension does not directly affect the myocardium, so hypertension cannot be used as an independent risk factor, which is consistent with our conclusion.
Although many studies have reported relevant predictors of anthracycline-induced cardiotoxicity in breast cancer and lymphoma, the establishment of comprehensive cardiotoxicity models based on these risk factors to predict anthracycline-induced cardiotoxicity in combination with other chemotherapeutic agents is still rare. In our study, 22 predictors were first considered to reduce the bias of nomo map, such as demographic information, hypertension, coronary heart disease, type 2 diabetes mellitus, markers of myocardial injury, and types of combined chemotherapy drugs. Firstly, six risk factors were screened out by logistic regression analysis. Considering that the risk factors screened by logistic analysis method over-fit, which reduced the accuracy and actual goodness of fit of the model, we used lasso regression analysis to re-screen the indicators with statistically significant differences between the two groups in the training cohort. Finally, 10 risk factors were selected to construct the prediction model. At the same time, the R software was used to construct the nomo map, and the ROC curve, H-L test, DCA and other methods were used to evaluate the quality of the Nomo map. The C-index value (0.884) and calibration curve showed that the model had a good prediction effect on cardiotoxicity, and had good discrimination and correction ability. In addition, the validation dataset was used for internal validation of the model, and the results showed that there was good agreement between the training cohort and the validation cohort. Therefore, our results are stable and reliable. DCA curves also reflects a net clinical benefit is produced by nomogram of training set and validation set. With the nomogram constructed in this study, clinicians can assess the risk of cardiotoxicity of anthracycline-combined chemotherapy in patients with breast cancer and lymphoma. Clinicians can measure these ready-made after each chemotherapy predictors, and through our nomogram to estimate the risk of cardiac toxicity, and then check the suspected for high-risk patients, according to the results of the assessment for cardiac protective intervention and treatment.