Sik Kwan Chan1, Cheng Lin2, Shao Hui Huang3, Tin Ching Chau1, Qiao Juan Guo2, Brian O'Sullivan3, Ka On Lam4, Sze Chun Chau1, Sum Yin Chan1, Chi Chung Tong1, Varut Vardhanabhuti5, Dora Lai Wan Kwong4, Tsz Him So1, Chor Yi Ng1, To Wai Leung4, Mai Yee Luk1, Anne Wing Mui Lee4, Horace Cheuk Wai Choi1, Jian Ji Pan6, Victor Ho Fun Lee7. 1. Department of Clinical Oncology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong. 2. Department of Radiation Oncology, Fujian Medical University Cancer Hospital, Fujian Cancer Hospital, Fuzhou, China; Fujian Provincial Key Laboratory of Translational Cancer Medicine, Fuzhou, China. 3. Department of Radiation Oncology, Princess Margaret Cancer Centre, University of Toronto, Toronto, Ontario, Canada; Clinical Oncology Center, The University of Hong Kong-Shenzhen Hospital, Shenzhen, China. 4. Department of Clinical Oncology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong; Clinical Oncology Center, The University of Hong Kong-Shenzhen Hospital, Shenzhen, China. 5. Department of Diagnostic Radiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong. 6. Department of Radiation Oncology, Fujian Medical University Cancer Hospital, Fujian Cancer Hospital, Fuzhou, China; Fujian Provincial Key Laboratory of Translational Cancer Medicine, Fuzhou, China. Electronic address: panjianji1956@fjmu.edu.cn. 7. Department of Clinical Oncology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong; Clinical Oncology Center, The University of Hong Kong-Shenzhen Hospital, Shenzhen, China. Electronic address: vhflee@hku.hk.
Abstract
PURPOSE: To propose a refined M1 classification in de novo metastatic nasopharyngeal carcinoma (NPC) based on pooled data from two academic institutions. METHODS: Previously untreated de novo M1 NPC patients prospectively treated at The University of Hong Kong (N = 69) and Fujian Cancer Hospital (N = 114) between 2007 and 2016 were recruited and randomized in a 2:1 ratio to generate training (N = 120) and validation (N = 63) cohorts, respectively. Multivariable analysis (MVA) was performed for the training and validation cohorts to identify anatomic prognostic factors for overall survival (OS). Recursive partitioning analysis (RPA) was performed which incorporated the anatomic prognostic factors identified in the MVA to derive Anatomic-RPA groups which stratified OS in the training cohort, and were then validated in the validation cohort. RESULTS: Median follow-up for the training and validation cohorts was 27.2 and 30.2 months with 3-year OS of 51.6% and 51.1%, respectively. MVA revealed that co-existing liver-bone metastases was the only factor prognostic for OS in both the training and validation cohorts. Anatomic-RPA separated M1 disease into M1a (no co-existing liver-bone metastases) and M1b (co-existing liver-bone metastases) with median OS 39.5 and 23.7 months, respectively (p = 0.004) in the training cohort. RPA for the validation cohort also confirmed good segregation with co-existing liver-bone metastases with median OS 47.7 and 16.0 months, respectively (p = 0.008). CONCLUSION: Our proposal to subdivide de novo M1 NPC into M1a (no co-existing liver-bone metastases) vs. M1b (co-existing liver-bone metastases) provides better OS segregation.
PURPOSE: To propose a refined M1 classification in de novo metastatic nasopharyngeal carcinoma (NPC) based on pooled data from two academic institutions. METHODS: Previously untreated de novo M1 NPC patients prospectively treated at The University of Hong Kong (N = 69) and Fujian Cancer Hospital (N = 114) between 2007 and 2016 were recruited and randomized in a 2:1 ratio to generate training (N = 120) and validation (N = 63) cohorts, respectively. Multivariable analysis (MVA) was performed for the training and validation cohorts to identify anatomic prognostic factors for overall survival (OS). Recursive partitioning analysis (RPA) was performed which incorporated the anatomic prognostic factors identified in the MVA to derive Anatomic-RPA groups which stratified OS in the training cohort, and were then validated in the validation cohort. RESULTS: Median follow-up for the training and validation cohorts was 27.2 and 30.2 months with 3-year OS of 51.6% and 51.1%, respectively. MVA revealed that co-existing liver-bone metastases was the only factor prognostic for OS in both the training and validation cohorts. Anatomic-RPA separated M1 disease into M1a (no co-existing liver-bone metastases) and M1b (co-existing liver-bone metastases) with median OS 39.5 and 23.7 months, respectively (p = 0.004) in the training cohort. RPA for the validation cohort also confirmed good segregation with co-existing liver-bone metastases with median OS 47.7 and 16.0 months, respectively (p = 0.008). CONCLUSION: Our proposal to subdivide de novo M1 NPC into M1a (no co-existing liver-bone metastases) vs. M1b (co-existing liver-bone metastases) provides better OS segregation.