Comparative Study On The Selection Of Treatment Recommendations For Breast Cancer By Watson For Oncology And Clinical Oncologist

Background Due to the advancement of science and technology,great progress has been made in the treatment of breast cancer.The huge increase in the number of breast cancer studies,the continuous updating of evidence-based medical evidence and clinical diagnosis and treatment guidelines,breast cancer oncologists have faced more challenges in the choice of breast cancer treatment recommendations.Clinical decision-support systems(CDSS)appearing under the support of artificial intelligence(AI)technology have the potential to help address this challenge.This study compares the consistency of treatment recommendations between between treatment recommendations made by the AI CDSS Watson for Oncology(WFO)and a multidisciplinary tumor board for breast cancer.Objective To compares the consistency of treatment recommendations between between treatment recommendations made by Watson for Oncology(WFO)and a multidisciplinary tumor board for breast cancer,and to evaluate their clinical application value.Methods Clinical information was collected from 76 postoperative breast cancer patients who attended the A breast surgery department of Shenzhen Second People’s Hospital from May 2016 to June 2018.Criteria for inclusion:(1)Female patients with clinical or pathological stage I-III(AJCC 8th edition breast cancer staging criteria)who underwent preoperative ultrasound and CT examination and no important organ metastasis was found;(2)Patients who did not receive neoadjuvant treatment for breast cancer;(3)Patients with complete clinical information and postoperative immunohistochemical detection of estrogen receptor(ER),progesterone receptor(PR),human epidermal growth factor receptor-2(HER-2)and Ki-67 expression levels in cancer tissue;(4)Patients with breast cancer discussed by MDT.The patients in this study were aggregated,logged into the interface of the Watson tumor system,and entered the corresponding information.The WFO obtained certain treatment recommendations through certain calculations.WFO’s treatment recommendations(which generally include several options for a single case)are categorized into three groups with a corresponding label: green represents ‘recommended treatments’ with a strong base of evidence,amber represents treatments‘for consideration’ that oncologists may consider as suitable alternatives based on their clinical judgment,and red represents treatments that are ‘not recommended’ due to specific contraindications or strong evidence against their use.Compare the treatment recommendations of MDT and WFO: The WFO recommended and considered levels are considered consistent with MDT,and the rest are considered inconsistent.Descriptive statistics of breast cancer case characteristics were calculated using Microsoft Excel and presented as means?±?standard deviation or median(min,max).Row percentages are presented where indicated.Concordance was expressed as percent agreement.Cancer characteristics included patient age,cancer stage,and receptor status.To control for these three determinants of concordance simultaneously,a logistic regression model was estimated with odds ratios and 95% confidence intervals reported by SPSS 24.0.P <0.05 was considered statistically significant.Results Overall,WFO was consistent with MDT treatment regimens in 82.9% of breast cancer cases.A subgroup analysis of the consistency of the two treatment regimens by patient age,pathological stage and tumor molecular typing revealed that the consistency between the two treatment regimens decreases with increasing patient age(P<0.05).Whereas,pathological staging and staging in this study had little effect on protocol consistency.Conclusion Treatment recommendations made by WFO and the tumor board were highly concordant for breast cancer cases examined.Breast cancer patient age had significant influence on concordance,while receptor stage and status alone did not.This study demonstrates that the AI clinical decision-support system WFO may be a helpful tool for breast cancer treatment decision making,especially at centers where expert breast cancer resources are limited.