Watson for Oncology在乳腺癌治疗中的应用与思考
|
摘要
目的:本研究旨在分析Watson for Oncology在中国乳腺癌辅助诊治中的应用与医生MDT诊治的一致性,以及差异的可能原因。方法:于2017年1月至2018年1月收治早期乳腺癌患者92例和晚期乳腺癌患者40例。由乳腺癌MDT团队确定的治疗方案为医院方案;根据患者的基本情况,采用Watson for Oncology 17.10版给予的治疗方案为Watson方案,评估两者方案的一致性并解析差异的可能原因。结果:医院方案与Watson方案一致共计101例,占76.51%。其中早期和晚期乳腺癌方案一致分别为73例和28例,分别占79.35%和70.00%。医院方案在Watson方案中的绿色推荐方案49例,占37.12%;橙色考虑方案52例,占39.39%;粉色不推荐方案26例,占19.70%;医院方案另有5例方案不在Watson方案中,占3.79%。Watson方案与医院方案在组织学Ⅲ期的患者中一致率最高,达到95.00%,三阴性乳腺癌无论早期还是晚期都表现出最高的一致率,分别为86.67%和100%。早期Luminal A和晚期HER2阳性型的医院方案在Watson绿色推荐方案占比较低,仅27.27%和0%。早期乳腺癌差异原因包括中低危复发风险的11例患者应用紫杉醇联合环磷酰胺(TC方案)和2例患者应用表柔比星联合环磷酰胺(EC方案),而Watson方案推荐为阿柔比星+环磷酰胺序贯紫杉醇(AC-T方案)和CMF方案(国内很少应用的方案)。HER2阳性不一致的是Watson推荐CMF+曲妥珠单抗方案,而医院方案考虑心脏毒性小而治疗疗效与AC-TH相似的TCb H方案。晚期不一致原因是CDK4/6抑制剂未在国内上市应用等。结论:医院方案和Watson方案在中国乳腺癌患者的辅助治疗和晚期一线治疗中一致性较好,但组织学分期和分子分型影响医院方案与Watson方案的一致率,需要实现本地化优化辅助治疗方案。
Objective: This study aimed to examine the level of agreement(concordance) between treatment recommendations made by Watson for Oncology and a multidisciplinary tumor(MDT) board for therapy in breast cancer. Methods: Treatment recommendations were provided for 132 breast cancers between Jan 2017 and Jan 2018. Watson for Oncology and MDT team provided treatment recommendations for the identical cases, as Watson and Hospital treatment. Treatment recommendations were considered concordant if the tumor board recommendations were designated 'recommended' or 'for consideration' by Watson for Oncology Version 17.10. Results: Treatment concordance between Watson and MDT occurred in 76.51% of breast cancer cases, 73 cases in early breast cancer(79.35%) and 28 cases in metastatic breast cancer(70.00%). Among all the cases, 49 cases were in 'Recommended', 52 cases were in 'for consideration', 26 cases were in disconcordance. Subgroup analysis found that patients with stage Ⅲ disease and triple-negative breast cancer were the most likely to be concordant. Luminal A and HER2-positive cases were less likely to recommended than other types. In early breast cancer, based on low recurrence risk, 11 cases and 2 cases were treated with TC and EC instead of AC-T or CMF recommended by Watson for Oncology. Four patients with HER2-positive were disconcordance with TCb H or AC-TH instead of AC-TH or CMFH. Main reason for disconcordance is that CDK4/6 inhibitors has not been approved in China. Conclusion: Treatment recommendations made by WFO and the tumor board were highly concordant for breast cancer cases examined. Breast cancer stage and subtype had significant influence on concordance, and Watson for Oncology need to be optimized in localization.
Objective: This study aimed to examine the level of agreement(concordance) between treatment recommendations made by Watson for Oncology and a multidisciplinary tumor(MDT) board for therapy in breast cancer. Methods: Treatment recommendations were provided for 132 breast cancers between Jan 2017 and Jan 2018. Watson for Oncology and MDT team provided treatment recommendations for the identical cases, as Watson and Hospital treatment. Treatment recommendations were considered concordant if the tumor board recommendations were designated 'recommended' or 'for consideration' by Watson for Oncology Version 17.10. Results: Treatment concordance between Watson and MDT occurred in 76.51% of breast cancer cases, 73 cases in early breast cancer(79.35%) and 28 cases in metastatic breast cancer(70.00%). Among all the cases, 49 cases were in 'Recommended', 52 cases were in 'for consideration', 26 cases were in disconcordance. Subgroup analysis found that patients with stage Ⅲ disease and triple-negative breast cancer were the most likely to be concordant. Luminal A and HER2-positive cases were less likely to recommended than other types. In early breast cancer, based on low recurrence risk, 11 cases and 2 cases were treated with TC and EC instead of AC-T or CMF recommended by Watson for Oncology. Four patients with HER2-positive were disconcordance with TCb H or AC-TH instead of AC-TH or CMFH. Main reason for disconcordance is that CDK4/6 inhibitors has not been approved in China. Conclusion: Treatment recommendations made by WFO and the tumor board were highly concordant for breast cancer cases examined. Breast cancer stage and subtype had significant influence on concordance, and Watson for Oncology need to be optimized in localization.
引文
[1]Mayo RM,summey JF,Williams Je,et al.Qualitative study of oncologists'views on the cancer Lin Q rapid learning system[J].J oncol Pract,2017,13(3):e176–e184.
[2]于观贞,魏培莲,陈颖,等.人工智能在肿瘤病理诊断和评估中的应用与思考[J].第二军医大学学报,2017,38(11):1349-1354.
[3]徐琰,胡保全.浅谈人工智能在乳腺癌领域的应用进展[J].中华乳腺病杂志,2017,11(5):257-261.
[4]somashekhar s P,sepulveda MJ,Puglielli s,et al.Watson for oncology and breast cancer treatment recommendations:agreement with an expert multidisciplinary tumor board[J].Ann oncol,2018,29(2):418-423.
[2]于观贞,魏培莲,陈颖,等.人工智能在肿瘤病理诊断和评估中的应用与思考[J].第二军医大学学报,2017,38(11):1349-1354.
[3]徐琰,胡保全.浅谈人工智能在乳腺癌领域的应用进展[J].中华乳腺病杂志,2017,11(5):257-261.
[4]somashekhar s P,sepulveda MJ,Puglielli s,et al.Watson for oncology and breast cancer treatment recommendations:agreement with an expert multidisciplinary tumor board[J].Ann oncol,2018,29(2):418-423.